diff options
Diffstat (limited to 'net/ipv4/tcp_input.c')
| -rw-r--r-- | net/ipv4/tcp_input.c | 3120 |
1 files changed, 1530 insertions, 1590 deletions
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6d8ab1c4efc..40639c288dc 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -61,6 +61,8 @@ * Pasi Sarolahti: F-RTO for dealing with spurious RTOs */ +#define pr_fmt(fmt) "TCP: " fmt + #include <linux/mm.h> #include <linux/slab.h> #include <linux/module.h> @@ -79,24 +81,23 @@ int sysctl_tcp_sack __read_mostly = 1; int sysctl_tcp_fack __read_mostly = 1; int sysctl_tcp_reordering __read_mostly = TCP_FASTRETRANS_THRESH; EXPORT_SYMBOL(sysctl_tcp_reordering); -int sysctl_tcp_ecn __read_mostly = 2; -EXPORT_SYMBOL(sysctl_tcp_ecn); int sysctl_tcp_dsack __read_mostly = 1; int sysctl_tcp_app_win __read_mostly = 31; -int sysctl_tcp_adv_win_scale __read_mostly = 2; +int sysctl_tcp_adv_win_scale __read_mostly = 1; EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); +/* rfc5961 challenge ack rate limiting */ +int sysctl_tcp_challenge_ack_limit = 100; + int sysctl_tcp_stdurg __read_mostly; int sysctl_tcp_rfc1337 __read_mostly; int sysctl_tcp_max_orphans __read_mostly = NR_FILE; int sysctl_tcp_frto __read_mostly = 2; -int sysctl_tcp_frto_response __read_mostly; -int sysctl_tcp_nometrics_save __read_mostly; int sysctl_tcp_thin_dupack __read_mostly; int sysctl_tcp_moderate_rcvbuf __read_mostly = 1; -int sysctl_tcp_abc __read_mostly; +int sysctl_tcp_early_retrans __read_mostly = 3; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ #define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */ @@ -105,19 +106,17 @@ int sysctl_tcp_abc __read_mostly; #define FLAG_SYN_ACKED 0x10 /* This ACK acknowledged SYN. */ #define FLAG_DATA_SACKED 0x20 /* New SACK. */ #define FLAG_ECE 0x40 /* ECE in this ACK */ -#define FLAG_DATA_LOST 0x80 /* SACK detected data lossage. */ #define FLAG_SLOWPATH 0x100 /* Do not skip RFC checks for window update.*/ -#define FLAG_ONLY_ORIG_SACKED 0x200 /* SACKs only non-rexmit sent before RTO */ +#define FLAG_ORIG_SACK_ACKED 0x200 /* Never retransmitted data are (s)acked */ #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_UPDATE_TS_RECENT 0x4000 /* tcp_replace_ts_recent() */ #define FLAG_ACKED (FLAG_DATA_ACKED|FLAG_SYN_ACKED) #define FLAG_NOT_DUP (FLAG_DATA|FLAG_WIN_UPDATE|FLAG_ACKED) #define FLAG_CA_ALERT (FLAG_DATA_SACKED|FLAG_ECE) #define FLAG_FORWARD_PROGRESS (FLAG_ACKED|FLAG_DATA_SACKED) -#define FLAG_ANY_PROGRESS (FLAG_FORWARD_PROGRESS|FLAG_SND_UNA_ADVANCED) #define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH) #define TCP_HP_BITS (~(TCP_RESERVED_BITS|TCP_FLAG_PSH)) @@ -174,7 +173,7 @@ static void tcp_measure_rcv_mss(struct sock *sk, const struct sk_buff *skb) 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); + unsigned int quickacks = tcp_sk(sk)->rcv_wnd / (2 * icsk->icsk_ack.rcv_mss); if (quickacks == 0) quickacks = 2; @@ -194,9 +193,10 @@ static void tcp_enter_quickack_mode(struct sock *sk) * and the session is not interactive. */ -static inline int tcp_in_quickack_mode(const struct sock *sk) +static inline bool tcp_in_quickack_mode(const struct sock *sk) { const struct inet_connection_sock *icsk = inet_csk(sk); + return icsk->icsk_ack.quick && !icsk->icsk_ack.pingpong; } @@ -206,7 +206,7 @@ static inline void TCP_ECN_queue_cwr(struct tcp_sock *tp) tp->ecn_flags |= TCP_ECN_QUEUE_CWR; } -static inline void TCP_ECN_accept_cwr(struct tcp_sock *tp, struct sk_buff *skb) +static inline void TCP_ECN_accept_cwr(struct tcp_sock *tp, const struct sk_buff *skb) { if (tcp_hdr(skb)->cwr) tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR; @@ -217,36 +217,49 @@ static inline void TCP_ECN_withdraw_cwr(struct tcp_sock *tp) tp->ecn_flags &= ~TCP_ECN_DEMAND_CWR; } -static inline void TCP_ECN_check_ce(struct tcp_sock *tp, struct sk_buff *skb) +static inline void TCP_ECN_check_ce(struct tcp_sock *tp, const struct sk_buff *skb) { - if (tp->ecn_flags & TCP_ECN_OK) { - if (INET_ECN_is_ce(TCP_SKB_CB(skb)->flags)) - tp->ecn_flags |= TCP_ECN_DEMAND_CWR; + if (!(tp->ecn_flags & TCP_ECN_OK)) + return; + + switch (TCP_SKB_CB(skb)->ip_dsfield & INET_ECN_MASK) { + case INET_ECN_NOT_ECT: /* Funny extension: if ECT is not set on a segment, - * it is surely retransmit. It is not in ECN RFC, - * but Linux follows this rule. */ - else if (INET_ECN_is_not_ect((TCP_SKB_CB(skb)->flags))) + * and we already seen ECT on a previous segment, + * it is probably a retransmit. + */ + if (tp->ecn_flags & TCP_ECN_SEEN) + tcp_enter_quickack_mode((struct sock *)tp); + break; + case INET_ECN_CE: + if (!(tp->ecn_flags & TCP_ECN_DEMAND_CWR)) { + /* Better not delay acks, sender can have a very low cwnd */ tcp_enter_quickack_mode((struct sock *)tp); + tp->ecn_flags |= TCP_ECN_DEMAND_CWR; + } + /* fallinto */ + default: + tp->ecn_flags |= TCP_ECN_SEEN; } } -static inline void TCP_ECN_rcv_synack(struct tcp_sock *tp, struct tcphdr *th) +static inline void TCP_ECN_rcv_synack(struct tcp_sock *tp, const struct tcphdr *th) { 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) +static inline void TCP_ECN_rcv_syn(struct tcp_sock *tp, const struct tcphdr *th) { 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) +static bool TCP_ECN_rcv_ecn_echo(const struct tcp_sock *tp, const struct tcphdr *th) { if (th->ece && !th->syn && (tp->ecn_flags & TCP_ECN_OK)) - return 1; - return 0; + return true; + return false; } /* Buffer size and advertised window tuning. @@ -254,16 +267,33 @@ static inline int TCP_ECN_rcv_ecn_echo(struct tcp_sock *tp, struct tcphdr *th) * 1. Tuning sk->sk_sndbuf, when connection enters established state. */ -static void tcp_fixup_sndbuf(struct sock *sk) +static void tcp_sndbuf_expand(struct sock *sk) { - int sndmem = tcp_sk(sk)->rx_opt.mss_clamp + MAX_TCP_HEADER + 16 + - sizeof(struct sk_buff); + const struct tcp_sock *tp = tcp_sk(sk); + int sndmem, per_mss; + u32 nr_segs; - if (sk->sk_sndbuf < 3 * sndmem) { - sk->sk_sndbuf = 3 * sndmem; - if (sk->sk_sndbuf > sysctl_tcp_wmem[2]) - sk->sk_sndbuf = sysctl_tcp_wmem[2]; - } + /* Worst case is non GSO/TSO : each frame consumes one skb + * and skb->head is kmalloced using power of two area of memory + */ + per_mss = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) + + MAX_TCP_HEADER + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); + + per_mss = roundup_pow_of_two(per_mss) + + SKB_DATA_ALIGN(sizeof(struct sk_buff)); + + nr_segs = max_t(u32, TCP_INIT_CWND, tp->snd_cwnd); + nr_segs = max_t(u32, nr_segs, tp->reordering + 1); + + /* Fast Recovery (RFC 5681 3.2) : + * Cubic needs 1.7 factor, rounded to 2 to include + * extra cushion (application might react slowly to POLLOUT) + */ + sndmem = 2 * nr_segs * per_mss; + + if (sk->sk_sndbuf < sndmem) + sk->sk_sndbuf = min(sndmem, sysctl_tcp_wmem[2]); } /* 2. Tuning advertised window (window_clamp, rcv_ssthresh) @@ -309,14 +339,14 @@ 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, const struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); /* Check #1 */ if (tp->rcv_ssthresh < tp->window_clamp && (int)tp->rcv_ssthresh < tcp_space(sk) && - !tcp_memory_pressure) { + !sk_under_memory_pressure(sk)) { int incr; /* Check #2. Increase window, if skb with such overhead @@ -328,6 +358,7 @@ static void tcp_grow_window(struct sock *sk, struct sk_buff *skb) incr = __tcp_grow_window(sk, skb); if (incr) { + incr = max_t(int, incr, 2 * skb->len); tp->rcv_ssthresh = min(tp->rcv_ssthresh + incr, tp->window_clamp); inet_csk(sk)->icsk_ack.quick |= 1; @@ -336,26 +367,28 @@ static void tcp_grow_window(struct sock *sk, struct sk_buff *skb) } /* 3. Tuning rcvbuf, when connection enters established state. */ - static void tcp_fixup_rcvbuf(struct sock *sk) { - struct tcp_sock *tp = tcp_sk(sk); - int rcvmem = tp->advmss + MAX_TCP_HEADER + 16 + sizeof(struct sk_buff); + u32 mss = tcp_sk(sk)->advmss; + int rcvmem; + + rcvmem = 2 * SKB_TRUESIZE(mss + MAX_TCP_HEADER) * + tcp_default_init_rwnd(mss); - /* Try to select rcvbuf so that 4 mss-sized segments - * will fit to window and corresponding skbs will fit to our rcvbuf. - * (was 3; 4 is minimum to allow fast retransmit to work.) + /* Dynamic Right Sizing (DRS) has 2 to 3 RTT latency + * Allow enough cushion so that sender is not limited by our window */ - while (tcp_win_from_space(rcvmem) < tp->advmss) - rcvmem += 128; - if (sk->sk_rcvbuf < 4 * rcvmem) - sk->sk_rcvbuf = min(4 * rcvmem, sysctl_tcp_rmem[2]); + if (sysctl_tcp_moderate_rcvbuf) + rcvmem <<= 2; + + if (sk->sk_rcvbuf < rcvmem) + sk->sk_rcvbuf = min(rcvmem, sysctl_tcp_rmem[2]); } /* 4. Try to fixup all. It is made immediately after connection enters * established state. */ -static void tcp_init_buffer_space(struct sock *sk) +void tcp_init_buffer_space(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); int maxwin; @@ -363,9 +396,11 @@ static void tcp_init_buffer_space(struct sock *sk) if (!(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) tcp_fixup_rcvbuf(sk); if (!(sk->sk_userlocks & SOCK_SNDBUF_LOCK)) - tcp_fixup_sndbuf(sk); + tcp_sndbuf_expand(sk); tp->rcvq_space.space = tp->rcv_wnd; + tp->rcvq_space.time = tcp_time_stamp; + tp->rcvq_space.seq = tp->copied_seq; maxwin = tcp_full_space(sk); @@ -398,8 +433,8 @@ static void tcp_clamp_window(struct sock *sk) if (sk->sk_rcvbuf < sysctl_tcp_rmem[2] && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && - !tcp_memory_pressure && - atomic_long_read(&tcp_memory_allocated) < sysctl_tcp_mem[0]) { + !sk_under_memory_pressure(sk) && + sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) { sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc), sysctl_tcp_rmem[2]); } @@ -416,7 +451,7 @@ static void tcp_clamp_window(struct sock *sk) */ void tcp_initialize_rcv_mss(struct sock *sk) { - struct tcp_sock *tp = tcp_sk(sk); + const 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); @@ -460,8 +495,11 @@ static void tcp_rcv_rtt_update(struct tcp_sock *tp, u32 sample, int win_dep) if (!win_dep) { m -= (new_sample >> 3); new_sample += m; - } else if (m < new_sample) - new_sample = m << 3; + } else { + m <<= 3; + if (m < new_sample) + new_sample = m; + } } else { /* No previous measure. */ new_sample = m << 3; @@ -477,7 +515,7 @@ 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, tcp_time_stamp - tp->rcv_rtt_est.time, 1); new_measure: tp->rcv_rtt_est.seq = tp->rcv_nxt + tp->rcv_wnd; @@ -502,49 +540,62 @@ void tcp_rcv_space_adjust(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); int time; - int space; - - if (tp->rcvq_space.time == 0) - goto new_measure; + int copied; time = tcp_time_stamp - tp->rcvq_space.time; if (time < (tp->rcv_rtt_est.rtt >> 3) || tp->rcv_rtt_est.rtt == 0) return; - space = 2 * (tp->copied_seq - tp->rcvq_space.seq); + /* Number of bytes copied to user in last RTT */ + copied = tp->copied_seq - tp->rcvq_space.seq; + if (copied <= tp->rcvq_space.space) + goto new_measure; + + /* A bit of theory : + * copied = bytes received in previous RTT, our base window + * To cope with packet losses, we need a 2x factor + * To cope with slow start, and sender growing its cwin by 100 % + * every RTT, we need a 4x factor, because the ACK we are sending + * now is for the next RTT, not the current one : + * <prev RTT . ><current RTT .. ><next RTT .... > + */ + + if (sysctl_tcp_moderate_rcvbuf && + !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) { + int rcvwin, rcvmem, rcvbuf; - space = max(tp->rcvq_space.space, space); + /* minimal window to cope with packet losses, assuming + * steady state. Add some cushion because of small variations. + */ + rcvwin = (copied << 1) + 16 * tp->advmss; - if (tp->rcvq_space.space != space) { - int rcvmem; + /* If rate increased by 25%, + * assume slow start, rcvwin = 3 * copied + * If rate increased by 50%, + * assume sender can use 2x growth, rcvwin = 4 * copied + */ + if (copied >= + tp->rcvq_space.space + (tp->rcvq_space.space >> 2)) { + if (copied >= + tp->rcvq_space.space + (tp->rcvq_space.space >> 1)) + rcvwin <<= 1; + else + rcvwin += (rcvwin >> 1); + } - tp->rcvq_space.space = space; + rcvmem = SKB_TRUESIZE(tp->advmss + MAX_TCP_HEADER); + while (tcp_win_from_space(rcvmem) < tp->advmss) + rcvmem += 128; - if (sysctl_tcp_moderate_rcvbuf && - !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) { - int new_clamp = space; + rcvbuf = min(rcvwin / tp->advmss * rcvmem, sysctl_tcp_rmem[2]); + if (rcvbuf > sk->sk_rcvbuf) { + sk->sk_rcvbuf = rcvbuf; - /* Receive space grows, normalize in order to - * take into account packet headers and sk_buff - * structure overhead. - */ - space /= tp->advmss; - if (!space) - space = 1; - rcvmem = (tp->advmss + MAX_TCP_HEADER + - 16 + sizeof(struct sk_buff)); - while (tcp_win_from_space(rcvmem) < tp->advmss) - rcvmem += 128; - space *= rcvmem; - space = min(space, sysctl_tcp_rmem[2]); - if (space > sk->sk_rcvbuf) { - sk->sk_rcvbuf = space; - - /* Make the window clamp follow along. */ - tp->window_clamp = new_clamp; - } + /* Make the window clamp follow along. */ + tp->window_clamp = rcvwin; } } + tp->rcvq_space.space = copied; new_measure: tp->rcvq_space.seq = tp->copied_seq; @@ -616,10 +667,11 @@ static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb) * To save cycles in the RFC 1323 implementation it was better to break * it up into three procedures. -- erics */ -static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) +static void tcp_rtt_estimator(struct sock *sk, long mrtt_us) { struct tcp_sock *tp = tcp_sk(sk); - long m = mrtt; /* RTT */ + long m = mrtt_us; /* RTT */ + u32 srtt = tp->srtt_us; /* The following amusing code comes from Jacobson's * article in SIGCOMM '88. Note that rtt and mdev @@ -637,14 +689,12 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) * does not matter how to _calculate_ it. Seems, it was trap * that VJ failed to avoid. 8) */ - if (m == 0) - m = 1; - if (tp->srtt != 0) { - m -= (tp->srtt >> 3); /* m is now error in rtt est */ - tp->srtt += m; /* rtt = 7/8 rtt + 1/8 new */ + if (srtt != 0) { + m -= (srtt >> 3); /* m is now error in rtt est */ + srtt += m; /* rtt = 7/8 rtt + 1/8 new */ if (m < 0) { m = -m; /* m is now abs(error) */ - m -= (tp->mdev >> 2); /* similar update on mdev */ + m -= (tp->mdev_us >> 2); /* similar update on mdev */ /* This is similar to one of Eifel findings. * Eifel blocks mdev updates when rtt decreases. * This solution is a bit different: we use finer gain @@ -656,33 +706,62 @@ static void tcp_rtt_estimator(struct sock *sk, const __u32 mrtt) if (m > 0) m >>= 3; } else { - m -= (tp->mdev >> 2); /* similar update on mdev */ + m -= (tp->mdev_us >> 2); /* similar update on mdev */ } - tp->mdev += m; /* mdev = 3/4 mdev + 1/4 new */ - if (tp->mdev > tp->mdev_max) { - tp->mdev_max = tp->mdev; - if (tp->mdev_max > tp->rttvar) - tp->rttvar = tp->mdev_max; + tp->mdev_us += m; /* mdev = 3/4 mdev + 1/4 new */ + if (tp->mdev_us > tp->mdev_max_us) { + tp->mdev_max_us = tp->mdev_us; + if (tp->mdev_max_us > tp->rttvar_us) + tp->rttvar_us = tp->mdev_max_us; } if (after(tp->snd_una, tp->rtt_seq)) { - if (tp->mdev_max < tp->rttvar) - tp->rttvar -= (tp->rttvar - tp->mdev_max) >> 2; + if (tp->mdev_max_us < tp->rttvar_us) + tp->rttvar_us -= (tp->rttvar_us - tp->mdev_max_us) >> 2; tp->rtt_seq = tp->snd_nxt; - tp->mdev_max = tcp_rto_min(sk); + tp->mdev_max_us = tcp_rto_min_us(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->mdev_max = tp->rttvar = max(tp->mdev, tcp_rto_min(sk)); + srtt = m << 3; /* take the measured time to be rtt */ + tp->mdev_us = m << 1; /* make sure rto = 3*rtt */ + tp->rttvar_us = max(tp->mdev_us, tcp_rto_min_us(sk)); + tp->mdev_max_us = tp->rttvar_us; tp->rtt_seq = tp->snd_nxt; } + tp->srtt_us = max(1U, srtt); +} + +/* Set the sk_pacing_rate to allow proper sizing of TSO packets. + * Note: TCP stack does not yet implement pacing. + * FQ packet scheduler can be used to implement cheap but effective + * TCP pacing, to smooth the burst on large writes when packets + * in flight is significantly lower than cwnd (or rwin) + */ +static void tcp_update_pacing_rate(struct sock *sk) +{ + const struct tcp_sock *tp = tcp_sk(sk); + u64 rate; + + /* set sk_pacing_rate to 200 % of current rate (mss * cwnd / srtt) */ + rate = (u64)tp->mss_cache * 2 * (USEC_PER_SEC << 3); + + rate *= max(tp->snd_cwnd, tp->packets_out); + + if (likely(tp->srtt_us)) + do_div(rate, tp->srtt_us); + + /* ACCESS_ONCE() is needed because sch_fq fetches sk_pacing_rate + * without any lock. We want to make sure compiler wont store + * intermediate values in this location. + */ + ACCESS_ONCE(sk->sk_pacing_rate) = min_t(u64, rate, + sk->sk_max_pacing_rate); } /* Calculate rto without backoff. This is the second half of Van Jacobson's * routine referred to above. */ -static inline void tcp_set_rto(struct sock *sk) +static void tcp_set_rto(struct sock *sk) { const struct tcp_sock *tp = tcp_sk(sk); /* Old crap is replaced with new one. 8) @@ -709,228 +788,31 @@ static inline void tcp_set_rto(struct sock *sk) tcp_bound_rto(sk); } -/* Save metrics learned by this TCP session. - This function is called only, when TCP finishes successfully - i.e. when it enters TIME-WAIT or goes from LAST-ACK to CLOSE. - */ -void tcp_update_metrics(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - struct dst_entry *dst = __sk_dst_get(sk); - - if (sysctl_tcp_nometrics_save) - return; - - dst_confirm(dst); - - if (dst && (dst->flags & DST_HOST)) { - const struct inet_connection_sock *icsk = inet_csk(sk); - int m; - unsigned long rtt; - - if (icsk->icsk_backoff || !tp->srtt) { - /* This session failed to estimate rtt. Why? - * Probably, no packets returned in time. - * Reset our results. - */ - if (!(dst_metric_locked(dst, RTAX_RTT))) - dst->metrics[RTAX_RTT - 1] = 0; - return; - } - - rtt = dst_metric_rtt(dst, RTAX_RTT); - m = rtt - tp->srtt; - - /* If newly calculated rtt larger than stored one, - * store new one. Otherwise, use EWMA. Remember, - * rtt overestimation is always better than underestimation. - */ - if (!(dst_metric_locked(dst, RTAX_RTT))) { - if (m <= 0) - set_dst_metric_rtt(dst, RTAX_RTT, tp->srtt); - else - set_dst_metric_rtt(dst, RTAX_RTT, rtt - (m >> 3)); - } - - if (!(dst_metric_locked(dst, RTAX_RTTVAR))) { - unsigned long var; - if (m < 0) - m = -m; - - /* Scale deviation to rttvar fixed point */ - m >>= 1; - if (m < tp->mdev) - m = tp->mdev; - - var = dst_metric_rtt(dst, RTAX_RTTVAR); - if (m >= var) - var = m; - else - var -= (var - m) >> 2; - - set_dst_metric_rtt(dst, RTAX_RTTVAR, var); - } - - if (tcp_in_initial_slowstart(tp)) { - /* Slow start still did not finish. */ - if (dst_metric(dst, RTAX_SSTHRESH) && - !dst_metric_locked(dst, RTAX_SSTHRESH) && - (tp->snd_cwnd >> 1) > dst_metric(dst, RTAX_SSTHRESH)) - 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; - } else if (tp->snd_cwnd > tp->snd_ssthresh && - icsk->icsk_ca_state == TCP_CA_Open) { - /* Cong. avoidance phase, cwnd is reliable. */ - if (!dst_metric_locked(dst, RTAX_SSTHRESH)) - dst->metrics[RTAX_SSTHRESH-1] = - max(tp->snd_cwnd >> 1, tp->snd_ssthresh); - if (!dst_metric_locked(dst, RTAX_CWND)) - dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_cwnd) >> 1; - } else { - /* Else slow start did not finish, cwnd is non-sense, - ssthresh may be also invalid. - */ - if (!dst_metric_locked(dst, RTAX_CWND)) - dst->metrics[RTAX_CWND-1] = (dst_metric(dst, RTAX_CWND) + tp->snd_ssthresh) >> 1; - if (dst_metric(dst, RTAX_SSTHRESH) && - !dst_metric_locked(dst, RTAX_SSTHRESH) && - tp->snd_ssthresh > dst_metric(dst, RTAX_SSTHRESH)) - dst->metrics[RTAX_SSTHRESH-1] = tp->snd_ssthresh; - } - - if (!dst_metric_locked(dst, RTAX_REORDERING)) { - if (dst_metric(dst, RTAX_REORDERING) < tp->reordering && - tp->reordering != sysctl_tcp_reordering) - dst->metrics[RTAX_REORDERING-1] = tp->reordering; - } - } -} - -__u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst) +__u32 tcp_init_cwnd(const struct tcp_sock *tp, const struct dst_entry *dst) { __u32 cwnd = (dst ? dst_metric(dst, RTAX_INITCWND) : 0); if (!cwnd) - cwnd = rfc3390_bytes_to_packets(tp->mss_cache); + cwnd = TCP_INIT_CWND; return min_t(__u32, cwnd, tp->snd_cwnd_clamp); } -/* Set slow start threshold and cwnd not falling to slow start */ -void tcp_enter_cwr(struct sock *sk, const int set_ssthresh) -{ - struct tcp_sock *tp = tcp_sk(sk); - const struct inet_connection_sock *icsk = inet_csk(sk); - - tp->prior_ssthresh = 0; - tp->bytes_acked = 0; - if (icsk->icsk_ca_state < TCP_CA_CWR) { - tp->undo_marker = 0; - if (set_ssthresh) - tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); - tp->snd_cwnd = min(tp->snd_cwnd, - tcp_packets_in_flight(tp) + 1U); - tp->snd_cwnd_cnt = 0; - tp->high_seq = tp->snd_nxt; - tp->snd_cwnd_stamp = tcp_time_stamp; - TCP_ECN_queue_cwr(tp); - - tcp_set_ca_state(sk, TCP_CA_CWR); - } -} - /* * Packet counting of FACK is based on in-order assumptions, therefore TCP * disables it when reordering is detected */ -static void tcp_disable_fack(struct tcp_sock *tp) +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; + tp->rx_opt.sack_ok &= ~TCP_FACK_ENABLED; } /* Take a notice that peer is sending D-SACKs */ static void tcp_dsack_seen(struct tcp_sock *tp) { - tp->rx_opt.sack_ok |= 4; -} - -/* Initialize metrics on socket. */ - -static void tcp_init_metrics(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - struct dst_entry *dst = __sk_dst_get(sk); - - if (dst == NULL) - goto reset; - - dst_confirm(dst); - - if (dst_metric_locked(dst, RTAX_CWND)) - tp->snd_cwnd_clamp = dst_metric(dst, RTAX_CWND); - if (dst_metric(dst, RTAX_SSTHRESH)) { - tp->snd_ssthresh = dst_metric(dst, RTAX_SSTHRESH); - if (tp->snd_ssthresh > tp->snd_cwnd_clamp) - tp->snd_ssthresh = tp->snd_cwnd_clamp; - } - if (dst_metric(dst, RTAX_REORDERING) && - tp->reordering != dst_metric(dst, RTAX_REORDERING)) { - tcp_disable_fack(tp); - tp->reordering = dst_metric(dst, RTAX_REORDERING); - } - - if (dst_metric(dst, RTAX_RTT) == 0) - goto reset; - - if (!tp->srtt && dst_metric_rtt(dst, RTAX_RTT) < (TCP_TIMEOUT_INIT << 3)) - goto reset; - - /* Initial rtt is determined from SYN,SYN-ACK. - * The segment is small and rtt may appear much - * less than real one. Use per-dst memory - * to make it more realistic. - * - * A bit of theory. RTT is time passed after "normal" sized packet - * is sent until it is ACKed. In normal circumstances sending small - * packets force peer to delay ACKs and calculation is correct too. - * The algorithm is adaptive and, provided we follow specs, it - * NEVER underestimate RTT. BUT! If peer tries to make some clever - * tricks sort of "quick acks" for time long enough to decrease RTT - * to low value, and then abruptly stops to do it and starts to delay - * ACKs, wait for troubles. - */ - if (dst_metric_rtt(dst, RTAX_RTT) > tp->srtt) { - tp->srtt = dst_metric_rtt(dst, RTAX_RTT); - tp->rtt_seq = tp->snd_nxt; - } - if (dst_metric_rtt(dst, RTAX_RTTVAR) > tp->mdev) { - tp->mdev = dst_metric_rtt(dst, RTAX_RTTVAR); - tp->mdev_max = tp->rttvar = max(tp->mdev, tcp_rto_min(sk)); - } - tcp_set_rto(sk); - if (inet_csk(sk)->icsk_rto < TCP_TIMEOUT_INIT && !tp->rx_opt.saw_tstamp) - goto reset; - -cwnd: - tp->snd_cwnd = tcp_init_cwnd(tp, dst); - tp->snd_cwnd_stamp = tcp_time_stamp; - return; - -reset: - /* Play conservative. If timestamps are not - * supported, TCP will fail to recalculate correct - * rtt, if initial rto is too small. FORGET ALL AND RESET! - */ - if (!tp->rx_opt.saw_tstamp && tp->srtt) { - tp->srtt = 0; - tp->mdev = tp->mdev_max = tp->rttvar = TCP_TIMEOUT_INIT; - inet_csk(sk)->icsk_rto = TCP_TIMEOUT_INIT; - } - goto cwnd; + tp->rx_opt.sack_ok |= TCP_DSACK_SEEN; } static void tcp_update_reordering(struct sock *sk, const int metric, @@ -954,15 +836,18 @@ static void tcp_update_reordering(struct sock *sk, const int metric, NET_INC_STATS_BH(sock_net(sk), mib_idx); #if FASTRETRANS_DEBUG > 1 - printk(KERN_DEBUG "Disorder%d %d %u f%u s%u rr%d\n", - tp->rx_opt.sack_ok, inet_csk(sk)->icsk_ca_state, - tp->reordering, - tp->fackets_out, - tp->sacked_out, - tp->undo_marker ? tp->undo_retrans : 0); + pr_debug("Disorder%d %d %u f%u s%u rr%d\n", + tp->rx_opt.sack_ok, inet_csk(sk)->icsk_ca_state, + tp->reordering, + tp->fackets_out, + tp->sacked_out, + tp->undo_marker ? tp->undo_retrans : 0); #endif tcp_disable_fack(tp); } + + if (metric > 0) + tcp_disable_early_retrans(tp); } /* This must be called before lost_out is incremented */ @@ -1020,13 +905,11 @@ static void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, * These 6 states form finite state machine, controlled by the following events: * 1. New ACK (+SACK) arrives. (tcp_sacktag_write_queue()) * 2. Retransmission. (tcp_retransmit_skb(), tcp_xmit_retransmit_queue()) - * 3. Loss detection event of one of three flavors: + * 3. Loss detection event of two flavors: * A. Scoreboard estimator decided the packet is lost. * A'. Reno "three dupacks" marks head of queue lost. - * A''. Its FACK modfication, head until snd.fack is lost. - * B. SACK arrives sacking data transmitted after never retransmitted - * hole was sent out. - * C. SACK arrives sacking SND.NXT at the moment, when the + * A''. Its FACK modification, head until snd.fack is lost. + * B. SACK arrives sacking SND.NXT at the moment, when the * segment was retransmitted. * 4. D-SACK added new rule: D-SACK changes any tag to S. * @@ -1095,36 +978,36 @@ static void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, * the exact amount is rather hard to quantify. However, tp->max_window can * be used as an exaggerated estimate. */ -static int tcp_is_sackblock_valid(struct tcp_sock *tp, int is_dsack, - u32 start_seq, u32 end_seq) +static bool tcp_is_sackblock_valid(struct tcp_sock *tp, bool is_dsack, + u32 start_seq, u32 end_seq) { /* Too far in future, or reversed (interpretation is ambiguous) */ if (after(end_seq, tp->snd_nxt) || !before(start_seq, end_seq)) - return 0; + return false; /* Nasty start_seq wrap-around check (see comments above) */ if (!before(start_seq, tp->snd_nxt)) - return 0; + return false; /* In outstanding window? ...This is valid exit for D-SACKs too. * start_seq == snd_una is non-sensical (see comments above) */ if (after(start_seq, tp->snd_una)) - return 1; + return true; if (!is_dsack || !tp->undo_marker) - return 0; + return false; /* ...Then it's D-SACK, and must reside below snd_una completely */ - if (!after(end_seq, tp->snd_una)) - return 0; + if (after(end_seq, tp->snd_una)) + return false; if (!before(start_seq, tp->undo_marker)) - return 1; + return true; /* Too old */ if (!after(end_seq, tp->undo_marker)) - return 0; + return false; /* Undo_marker boundary crossing (overestimates a lot). Known already: * start_seq < undo_marker and end_seq >= undo_marker. @@ -1133,7 +1016,7 @@ static int tcp_is_sackblock_valid(struct tcp_sock *tp, int is_dsack, } /* Check for lost retransmit. This superb idea is borrowed from "ratehalving". - * Event "C". Later note: FACK people cheated me again 8), we have to account + * Event "B". Later note: FACK people cheated me again 8), we have to account * for reordering! Ugly, but should help. * * Search retransmitted skbs from write_queue that were sent when snd_nxt was @@ -1196,17 +1079,17 @@ static void tcp_mark_lost_retrans(struct sock *sk) tp->lost_retrans_low = new_low_seq; } -static int tcp_check_dsack(struct sock *sk, struct sk_buff *ack_skb, - struct tcp_sack_block_wire *sp, int num_sacks, - u32 prior_snd_una) +static bool tcp_check_dsack(struct sock *sk, const struct sk_buff *ack_skb, + struct tcp_sack_block_wire *sp, int num_sacks, + u32 prior_snd_una) { struct tcp_sock *tp = tcp_sk(sk); u32 start_seq_0 = get_unaligned_be32(&sp[0].start_seq); u32 end_seq_0 = get_unaligned_be32(&sp[0].end_seq); - int dup_sack = 0; + bool dup_sack = false; if (before(start_seq_0, TCP_SKB_CB(ack_skb)->ack_seq)) { - dup_sack = 1; + dup_sack = true; tcp_dsack_seen(tp); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDSACKRECV); } else if (num_sacks > 1) { @@ -1215,7 +1098,7 @@ static int tcp_check_dsack(struct sock *sk, struct sk_buff *ack_skb, if (!after(end_seq_0, end_seq_1) && !before(start_seq_0, start_seq_1)) { - dup_sack = 1; + dup_sack = true; tcp_dsack_seen(tp); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDSACKOFORECV); @@ -1223,7 +1106,7 @@ static int tcp_check_dsack(struct sock *sk, struct sk_buff *ack_skb, } /* D-SACK for already forgotten data... Do dumb counting. */ - if (dup_sack && + if (dup_sack && tp->undo_marker && tp->undo_retrans > 0 && !after(end_seq_0, prior_snd_una) && after(end_seq_0, tp->undo_marker)) tp->undo_retrans--; @@ -1232,9 +1115,10 @@ static int tcp_check_dsack(struct sock *sk, struct sk_buff *ack_skb, } struct tcp_sacktag_state { - int reord; - int fack_count; - int flag; + int reord; + int fack_count; + long rtt_us; /* RTT measured by SACKing never-retransmitted data */ + int flag; }; /* Check if skb is fully within the SACK block. In presence of GSO skbs, @@ -1246,9 +1130,10 @@ struct tcp_sacktag_state { * FIXME: this could be merged to shift decision code */ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb, - u32 start_seq, u32 end_seq) + u32 start_seq, u32 end_seq) { - int in_sack, err; + int err; + bool in_sack; unsigned int pkt_len; unsigned int mss; @@ -1277,12 +1162,12 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb, unsigned int new_len = (pkt_len / mss) * mss; if (!in_sack && new_len < pkt_len) { new_len += mss; - if (new_len > skb->len) + if (new_len >= skb->len) return 0; } pkt_len = new_len; } - err = tcp_fragment(sk, skb, pkt_len, mss); + err = tcp_fragment(sk, skb, pkt_len, mss, GFP_ATOMIC); if (err < 0) return err; } @@ -1290,24 +1175,27 @@ static int tcp_match_skb_to_sack(struct sock *sk, struct sk_buff *skb, return in_sack; } -static u8 tcp_sacktag_one(struct sk_buff *skb, struct sock *sk, - struct tcp_sacktag_state *state, - int dup_sack, int pcount) +/* Mark the given newly-SACKed range as such, adjusting counters and hints. */ +static u8 tcp_sacktag_one(struct sock *sk, + struct tcp_sacktag_state *state, u8 sacked, + u32 start_seq, u32 end_seq, + int dup_sack, int pcount, + const struct skb_mstamp *xmit_time) { struct tcp_sock *tp = tcp_sk(sk); - u8 sacked = TCP_SKB_CB(skb)->sacked; int fack_count = state->fack_count; /* Account D-SACK for retransmitted packet. */ if (dup_sack && (sacked & TCPCB_RETRANS)) { - if (after(TCP_SKB_CB(skb)->end_seq, tp->undo_marker)) + if (tp->undo_marker && tp->undo_retrans > 0 && + after(end_seq, tp->undo_marker)) tp->undo_retrans--; if (sacked & TCPCB_SACKED_ACKED) state->reord = min(fack_count, state->reord); } /* Nothing to do; acked frame is about to be dropped (was ACKed). */ - if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una)) + if (!after(end_seq, tp->snd_una)) return sacked; if (!(sacked & TCPCB_SACKED_ACKED)) { @@ -1326,14 +1214,20 @@ static u8 tcp_sacktag_one(struct sk_buff *skb, struct sock *sk, /* New sack for not retransmitted frame, * which was in hole. It is reordering. */ - if (before(TCP_SKB_CB(skb)->seq, + if (before(start_seq, tcp_highest_sack_seq(tp))) state->reord = min(fack_count, state->reord); - - /* SACK enhanced F-RTO (RFC4138; Appendix B) */ - if (!after(TCP_SKB_CB(skb)->end_seq, tp->frto_highmark)) - state->flag |= FLAG_ONLY_ORIG_SACKED; + if (!after(end_seq, tp->high_seq)) + state->flag |= FLAG_ORIG_SACK_ACKED; + /* Pick the earliest sequence sacked for RTT */ + if (state->rtt_us < 0) { + struct skb_mstamp now; + + skb_mstamp_get(&now); + state->rtt_us = skb_mstamp_us_delta(&now, + xmit_time); + } } if (sacked & TCPCB_LOST) { @@ -1350,8 +1244,7 @@ static u8 tcp_sacktag_one(struct sk_buff *skb, struct sock *sk, /* 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)) + before(start_seq, TCP_SKB_CB(tp->lost_skb_hint)->seq)) tp->lost_cnt_hint += pcount; if (fack_count > tp->fackets_out) @@ -1370,19 +1263,32 @@ static u8 tcp_sacktag_one(struct sk_buff *skb, struct sock *sk, return sacked; } -static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, - struct tcp_sacktag_state *state, - unsigned int pcount, int shifted, int mss, - int dup_sack) +/* Shift newly-SACKed bytes from this skb to the immediately previous + * already-SACKed sk_buff. Mark the newly-SACKed bytes as such. + */ +static bool tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, + struct tcp_sacktag_state *state, + unsigned int pcount, int shifted, int mss, + bool dup_sack) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *prev = tcp_write_queue_prev(sk, skb); + u32 start_seq = TCP_SKB_CB(skb)->seq; /* start of newly-SACKed */ + u32 end_seq = start_seq + shifted; /* end of newly-SACKed */ BUG_ON(!pcount); - /* Tweak before seqno plays */ - if (!tcp_is_fack(tp) && tcp_is_sack(tp) && tp->lost_skb_hint && - !before(TCP_SKB_CB(tp->lost_skb_hint)->seq, TCP_SKB_CB(skb)->seq)) + /* Adjust counters and hints for the newly sacked sequence + * range but discard the return value since prev is already + * marked. We must tag the range first because the seq + * advancement below implicitly advances + * tcp_highest_sack_seq() when skb is highest_sack. + */ + tcp_sacktag_one(sk, state, TCP_SKB_CB(skb)->sacked, + start_seq, end_seq, dup_sack, pcount, + &skb->skb_mstamp); + + if (skb == tp->lost_skb_hint) tp->lost_cnt_hint += pcount; TCP_SKB_CB(prev)->end_seq += shifted; @@ -1408,30 +1314,28 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, skb_shinfo(skb)->gso_type = 0; } - /* We discard results */ - tcp_sacktag_one(skb, sk, state, dup_sack, pcount); - /* Difference in this won't matter, both ACKed by the same cumul. ACK */ TCP_SKB_CB(prev)->sacked |= (TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS); if (skb->len > 0) { BUG_ON(!tcp_skb_pcount(skb)); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SACKSHIFTED); - return 0; + return false; } /* Whole SKB was eaten :-) */ if (skb == tp->retransmit_skb_hint) tp->retransmit_skb_hint = prev; - if (skb == tp->scoreboard_skb_hint) - tp->scoreboard_skb_hint = prev; if (skb == tp->lost_skb_hint) { tp->lost_skb_hint = prev; tp->lost_cnt_hint -= tcp_skb_pcount(prev); } - TCP_SKB_CB(skb)->flags |= TCP_SKB_CB(prev)->flags; + TCP_SKB_CB(prev)->tcp_flags |= TCP_SKB_CB(skb)->tcp_flags; + if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) + TCP_SKB_CB(prev)->end_seq++; + if (skb == tcp_highest_sack(sk)) tcp_advance_highest_sack(sk, skb); @@ -1440,19 +1344,19 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_SACKMERGED); - return 1; + return true; } /* I wish gso_size would have a bit more sane initialization than * something-or-zero which complicates things */ -static int tcp_skb_seglen(struct sk_buff *skb) +static int tcp_skb_seglen(const struct sk_buff *skb) { return tcp_skb_pcount(skb) == 1 ? skb->len : tcp_skb_mss(skb); } /* Shifting pages past head area doesn't work */ -static int skb_can_shift(struct sk_buff *skb) +static int skb_can_shift(const struct sk_buff *skb) { return !skb_headlen(skb) && skb_is_nonlinear(skb); } @@ -1463,7 +1367,7 @@ static int skb_can_shift(struct sk_buff *skb) static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb, struct tcp_sacktag_state *state, u32 start_seq, u32 end_seq, - int dup_sack) + bool dup_sack) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *prev; @@ -1558,6 +1462,10 @@ static struct sk_buff *tcp_shift_skb_data(struct sock *sk, struct sk_buff *skb, } } + /* tcp_sacktag_one() won't SACK-tag ranges below snd_una */ + if (!after(TCP_SKB_CB(skb)->seq + len, tp->snd_una)) + goto fallback; + if (!skb_shift(prev, skb, len)) goto fallback; if (!tcp_shifted_skb(sk, skb, state, pcount, len, mss, dup_sack)) @@ -1598,14 +1506,14 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, struct tcp_sack_block *next_dup, struct tcp_sacktag_state *state, u32 start_seq, u32 end_seq, - int dup_sack_in) + bool dup_sack_in) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *tmp; tcp_for_write_queue_from(skb, sk) { int in_sack = 0; - int dup_sack = dup_sack_in; + bool dup_sack = dup_sack_in; if (skb == tcp_send_head(sk)) break; @@ -1620,7 +1528,7 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, next_dup->start_seq, next_dup->end_seq); if (in_sack > 0) - dup_sack = 1; + dup_sack = true; } /* skb reference here is a bit tricky to get right, since @@ -1648,10 +1556,15 @@ static struct sk_buff *tcp_sacktag_walk(struct sk_buff *skb, struct sock *sk, break; if (in_sack) { - TCP_SKB_CB(skb)->sacked = tcp_sacktag_one(skb, sk, - state, - dup_sack, - tcp_skb_pcount(skb)); + TCP_SKB_CB(skb)->sacked = + tcp_sacktag_one(sk, + state, + TCP_SKB_CB(skb)->sacked, + TCP_SKB_CB(skb)->seq, + TCP_SKB_CB(skb)->end_seq, + dup_sack, + tcp_skb_pcount(skb), + &skb->skb_mstamp); if (!before(TCP_SKB_CB(skb)->seq, tcp_highest_sack_seq(tp))) @@ -1701,19 +1614,18 @@ static struct sk_buff *tcp_maybe_skipping_dsack(struct sk_buff *skb, return skb; } -static int tcp_sack_cache_ok(struct tcp_sock *tp, struct tcp_sack_block *cache) +static int tcp_sack_cache_ok(const struct tcp_sock *tp, const 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, const struct sk_buff *ack_skb, + u32 prior_snd_una, long *sack_rtt_us) { - 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); + const unsigned char *ptr = (skb_transport_header(ack_skb) + + TCP_SKB_CB(ack_skb)->sacked); struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2); struct tcp_sack_block sp[TCP_NUM_SACKS]; struct tcp_sack_block *cache; @@ -1721,12 +1633,13 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, struct sk_buff *skb; int num_sacks = min(TCP_NUM_SACKS, (ptr[1] - TCPOLEN_SACK_BASE) >> 3); int used_sacks; - int found_dup_sack = 0; + bool found_dup_sack = false; int i, j; int first_sack_index; state.flag = 0; state.reord = tp->packets_out; + state.rtt_us = -1L; if (!tp->sacked_out) { if (WARN_ON(tp->fackets_out)) @@ -1752,7 +1665,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, used_sacks = 0; first_sack_index = 0; for (i = 0; i < num_sacks; i++) { - int dup_sack = !i && found_dup_sack; + bool dup_sack = !i && found_dup_sack; sp[used_sacks].start_seq = get_unaligned_be32(&sp_wire[i].start_seq); sp[used_sacks].end_seq = get_unaligned_be32(&sp_wire[i].end_seq); @@ -1819,16 +1732,12 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, 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)); + bool dup_sack = (found_dup_sack && (i == first_sack_index)); struct tcp_sack_block *next_dup = NULL; if (found_dup_sack && ((i + 1) == first_sack_index)) next_dup = &sp[i + 1]; - /* Event "B" in the comment above. */ - if (after(end_seq, tp->high_seq)) - state.flag |= FLAG_DATA_LOST; - /* Skip too early cached blocks */ while (tcp_sack_cache_ok(tp, cache) && !before(start_seq, cache->end_seq)) @@ -1887,12 +1796,6 @@ walk: start_seq, end_seq, dup_sack); advance_sp: - /* SACK enhanced FRTO (RFC4138, Appendix B): Clearing correct - * due to in-order walk - */ - if (after(end_seq, tp->frto_highmark)) - state.flag &= ~FLAG_ONLY_ORIG_SACKED; - i++; } @@ -1909,8 +1812,7 @@ advance_sp: tcp_verify_left_out(tp); if ((state.reord < tp->fackets_out) && - ((icsk->icsk_ca_state != TCP_CA_Loss) || tp->undo_marker) && - (!tp->frto_highmark || after(tp->snd_una, tp->frto_highmark))) + ((inet_csk(sk)->icsk_ca_state != TCP_CA_Loss) || tp->undo_marker)) tcp_update_reordering(sk, tp->fackets_out - state.reord, 0); out: @@ -1921,13 +1823,14 @@ out: WARN_ON((int)tp->retrans_out < 0); WARN_ON((int)tcp_packets_in_flight(tp) < 0); #endif + *sack_rtt_us = state.rtt_us; return state.flag; } /* Limits sacked_out so that sum with lost_out isn't ever larger than - * packets_out. Returns zero if sacked_out adjustement wasn't necessary. + * packets_out. Returns false if sacked_out adjustement wasn't necessary. */ -static int tcp_limit_reno_sacked(struct tcp_sock *tp) +static bool tcp_limit_reno_sacked(struct tcp_sock *tp) { u32 holes; @@ -1936,9 +1839,9 @@ static int tcp_limit_reno_sacked(struct tcp_sock *tp) if ((tp->sacked_out + holes) > tp->packets_out) { tp->sacked_out = tp->packets_out - holes; - return 1; + return true; } - return 0; + return false; } /* If we receive more dupacks than we expected counting segments @@ -1984,205 +1887,13 @@ static inline void tcp_reset_reno_sack(struct tcp_sock *tp) tp->sacked_out = 0; } -static int tcp_is_sackfrto(const struct tcp_sock *tp) -{ - return (sysctl_tcp_frto == 0x2) && !tcp_is_reno(tp); -} - -/* F-RTO can only be used if TCP has never retransmitted anything other than - * head (SACK enhanced variant from Appendix B of RFC4138 is more robust here) - */ -int tcp_use_frto(struct sock *sk) -{ - const struct tcp_sock *tp = tcp_sk(sk); - const struct inet_connection_sock *icsk = inet_csk(sk); - struct sk_buff *skb; - - if (!sysctl_tcp_frto) - return 0; - - /* MTU probe and F-RTO won't really play nicely along currently */ - if (icsk->icsk_mtup.probe_size) - return 0; - - if (tcp_is_sackfrto(tp)) - return 1; - - /* Avoid expensive walking of rexmit queue if possible */ - if (tp->retrans_out > 1) - return 0; - - skb = tcp_write_queue_head(sk); - if (tcp_skb_is_last(sk, skb)) - return 1; - skb = tcp_write_queue_next(sk, skb); /* Skips head */ - tcp_for_write_queue_from(skb, sk) { - if (skb == tcp_send_head(sk)) - break; - 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)) - break; - } - return 1; -} - -/* RTO occurred, but do not yet enter Loss state. Instead, defer RTO - * recovery a bit and use heuristics in tcp_process_frto() to detect if - * the RTO was spurious. Only clear SACKED_RETRANS of the head here to - * keep retrans_out counting accurate (with SACK F-RTO, other than head - * may still have that bit set); TCPCB_LOST and remaining SACKED_RETRANS - * bits are handled if the Loss state is really to be entered (in - * tcp_enter_frto_loss). - * - * Do like tcp_enter_loss() would; when RTO expires the second time it - * does: - * "Reduce ssthresh if it has not yet been made inside this window." - */ -void tcp_enter_frto(struct sock *sk) -{ - const struct inet_connection_sock *icsk = inet_csk(sk); - struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *skb; - - if ((!tp->frto_counter && icsk->icsk_ca_state <= TCP_CA_Disorder) || - tp->snd_una == tp->high_seq || - ((icsk->icsk_ca_state == TCP_CA_Loss || tp->frto_counter) && - !icsk->icsk_retransmits)) { - tp->prior_ssthresh = tcp_current_ssthresh(sk); - /* Our state is too optimistic in ssthresh() call because cwnd - * is not reduced until tcp_enter_frto_loss() when previous F-RTO - * recovery has not yet completed. Pattern would be this: RTO, - * Cumulative ACK, RTO (2xRTO for the same segment does not end - * up here twice). - * RFC4138 should be more specific on what to do, even though - * RTO is quite unlikely to occur after the first Cumulative ACK - * due to back-off and complexity of triggering events ... - */ - if (tp->frto_counter) { - u32 stored_cwnd; - stored_cwnd = tp->snd_cwnd; - tp->snd_cwnd = 2; - tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); - tp->snd_cwnd = stored_cwnd; - } else { - tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); - } - /* ... in theory, cong.control module could do "any tricks" in - * ssthresh(), which means that ca_state, lost bits and lost_out - * counter would have to be faked before the call occurs. We - * consider that too expensive, unlikely and hacky, so modules - * using these in ssthresh() must deal these incompatibility - * issues if they receives CA_EVENT_FRTO and frto_counter != 0 - */ - tcp_ca_event(sk, CA_EVENT_FRTO); - } - - tp->undo_marker = tp->snd_una; - tp->undo_retrans = 0; - - skb = tcp_write_queue_head(sk); - if (TCP_SKB_CB(skb)->sacked & TCPCB_RETRANS) - tp->undo_marker = 0; - if (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS) { - TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; - tp->retrans_out -= tcp_skb_pcount(skb); - } - 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. - */ - if (tcp_is_sackfrto(tp) && (tp->frto_counter || - ((1 << icsk->icsk_ca_state) & (TCPF_CA_Recovery|TCPF_CA_Loss))) && - after(tp->high_seq, tp->snd_una)) { - tp->frto_highmark = tp->high_seq; - } else { - tp->frto_highmark = tp->snd_nxt; - } - tcp_set_ca_state(sk, TCP_CA_Disorder); - tp->high_seq = tp->snd_nxt; - tp->frto_counter = 1; -} - -/* Enter Loss state after F-RTO was applied. Dupack arrived after RTO, - * which indicates that we should follow the traditional RTO recovery, - * i.e. mark everything lost and do go-back-N retransmission. - */ -static void tcp_enter_frto_loss(struct sock *sk, int allowed_segments, int flag) -{ - struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *skb; - - tp->lost_out = 0; - tp->retrans_out = 0; - if (tcp_is_reno(tp)) - tcp_reset_reno_sack(tp); - - 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)) { - /* 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); - /* ...enter this if branch just for the first segment */ - flag |= FLAG_DATA_ACKED; - } else { - if (TCP_SKB_CB(skb)->sacked & TCPCB_RETRANS) - tp->undo_marker = 0; - TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS; - } - - /* Marking forward transmissions that were made after RTO lost - * can cause unnecessary retransmissions in some scenarios, - * SACK blocks will mitigate that in some but not in all cases. - * We used to not mark them but it was causing break-ups with - * receivers that do only in-order receival. - * - * TODO: we could detect presence of such receiver and select - * different behavior per flow. - */ - if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) { - TCP_SKB_CB(skb)->sacked |= TCPCB_LOST; - tp->lost_out += tcp_skb_pcount(skb); - tp->retransmit_high = TCP_SKB_CB(skb)->end_seq; - } - } - tcp_verify_left_out(tp); - - tp->snd_cwnd = tcp_packets_in_flight(tp) + allowed_segments; - tp->snd_cwnd_cnt = 0; - tp->snd_cwnd_stamp = tcp_time_stamp; - tp->frto_counter = 0; - tp->bytes_acked = 0; - - tp->reordering = min_t(unsigned int, tp->reordering, - sysctl_tcp_reordering); - tcp_set_ca_state(sk, TCP_CA_Loss); - tp->high_seq = tp->snd_nxt; - TCP_ECN_queue_cwr(tp); - - tcp_clear_all_retrans_hints(tp); -} - static void tcp_clear_retrans_partial(struct tcp_sock *tp) { tp->retrans_out = 0; tp->lost_out = 0; tp->undo_marker = 0; - tp->undo_retrans = 0; + tp->undo_retrans = -1; } void tcp_clear_retrans(struct tcp_sock *tp) @@ -2202,10 +1913,13 @@ void tcp_enter_loss(struct sock *sk, int how) const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; + bool new_recovery = false; /* Reduce ssthresh if it has not yet been made inside this window. */ - if (icsk->icsk_ca_state <= TCP_CA_Disorder || tp->snd_una == tp->high_seq || + if (icsk->icsk_ca_state <= TCP_CA_Disorder || + !after(tp->high_seq, tp->snd_una) || (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) { + new_recovery = true; tp->prior_ssthresh = tcp_current_ssthresh(sk); tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk); tcp_ca_event(sk, CA_EVENT_LOSS); @@ -2214,17 +1928,13 @@ void tcp_enter_loss(struct sock *sk, int how) tp->snd_cwnd_cnt = 0; tp->snd_cwnd_stamp = tcp_time_stamp; - tp->bytes_acked = 0; tcp_clear_retrans_partial(tp); if (tcp_is_reno(tp)) tcp_reset_reno_sack(tp); - if (!how) { - /* Push undo marker, if it was plain RTO and nothing - * was retransmitted. */ - tp->undo_marker = tp->snd_una; - } else { + tp->undo_marker = tp->snd_una; + if (how) { tp->sacked_out = 0; tp->fackets_out = 0; } @@ -2234,8 +1944,9 @@ 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_SACKED_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) { TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_ACKED; @@ -2246,13 +1957,24 @@ 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); + /* Timeout in disordered state after receiving substantial DUPACKs + * suggests that the degree of reordering is over-estimated. + */ + if (icsk->icsk_ca_state <= TCP_CA_Disorder && + tp->sacked_out >= sysctl_tcp_reordering) + tp->reordering = min_t(unsigned int, tp->reordering, + sysctl_tcp_reordering); tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->snd_nxt; TCP_ECN_queue_cwr(tp); - /* Abort F-RTO algorithm if one is in progress */ - tp->frto_counter = 0; + + /* F-RTO RFC5682 sec 3.1 step 1: retransmit SND.UNA if no previous + * loss recovery is underway except recurring timeout(s) on + * the same SND.UNA (sec 3.2). Disable F-RTO on path MTU probing + */ + tp->frto = sysctl_tcp_frto && + (new_recovery || icsk->icsk_retransmits) && + !inet_csk(sk)->icsk_mtup.probe_size; } /* If ACK arrived pointing to a remembered SACK, it means that our @@ -2261,7 +1983,7 @@ void tcp_enter_loss(struct sock *sk, int how) * * Do processing similar to RTO timeout. */ -static int tcp_check_sack_reneging(struct sock *sk, int flag) +static bool tcp_check_sack_reneging(struct sock *sk, int flag) { if (flag & FLAG_SACK_RENEGING) { struct inet_connection_sock *icsk = inet_csk(sk); @@ -2272,12 +1994,12 @@ static int tcp_check_sack_reneging(struct sock *sk, int flag) tcp_retransmit_skb(sk, tcp_write_queue_head(sk)); inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, icsk->icsk_rto, TCP_RTO_MAX); - return 1; + return true; } - return 0; + return false; } -static inline int tcp_fackets_out(struct tcp_sock *tp) +static inline int tcp_fackets_out(const struct tcp_sock *tp) { return tcp_is_reno(tp) ? tp->sacked_out + 1 : tp->fackets_out; } @@ -2297,22 +2019,33 @@ static inline int tcp_fackets_out(struct tcp_sock *tp) * they differ. Since neither occurs due to loss, TCP should really * ignore them. */ -static inline int tcp_dupack_heuristics(struct tcp_sock *tp) +static inline int tcp_dupack_heuristics(const 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) -{ - return tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto; -} - -static inline int tcp_head_timedout(struct sock *sk) +static bool tcp_pause_early_retransmit(struct sock *sk, int flag) { struct tcp_sock *tp = tcp_sk(sk); + unsigned long delay; + + /* Delay early retransmit and entering fast recovery for + * max(RTT/4, 2msec) unless ack has ECE mark, no RTT samples + * available, or RTO is scheduled to fire first. + */ + if (sysctl_tcp_early_retrans < 2 || sysctl_tcp_early_retrans > 3 || + (flag & FLAG_ECE) || !tp->srtt_us) + return false; + + delay = max(usecs_to_jiffies(tp->srtt_us >> 5), + msecs_to_jiffies(2)); - return tp->packets_out && - tcp_skb_timedout(sk, tcp_write_queue_head(sk)); + if (!time_after(inet_csk(sk)->icsk_timeout, (jiffies + delay))) + return false; + + inet_csk_reset_xmit_timer(sk, ICSK_TIME_EARLY_RETRANS, delay, + TCP_RTO_MAX); + return true; } /* Linux NewReno/SACK/FACK/ECN state machine. @@ -2408,28 +2141,18 @@ static inline int tcp_head_timedout(struct sock *sk) * Main question: may we further continue forward transmission * with the same cwnd? */ -static int tcp_time_to_recover(struct sock *sk) +static bool tcp_time_to_recover(struct sock *sk, int flag) { struct tcp_sock *tp = tcp_sk(sk); __u32 packets_out; - /* Do not perform any recovery during F-RTO algorithm */ - if (tp->frto_counter) - return 0; - /* Trick#1: The loss is proven. */ if (tp->lost_out) - return 1; + return true; /* Not-A-Trick#2 : Classic rule... */ if (tcp_dupack_heuristics(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_is_fack(tp) && tcp_head_timedout(sk)) - return 1; + return true; /* Trick#4: It is still not OK... But will it be useful to delay * recovery more? @@ -2441,7 +2164,7 @@ static int tcp_time_to_recover(struct sock *sk) /* We have nothing to send. This connection is limited * either by receiver window or by application. */ - return 1; + return true; } /* If a thin stream is detected, retransmit after first @@ -2452,51 +2175,26 @@ static int tcp_time_to_recover(struct sock *sk) if ((tp->thin_dupack || sysctl_tcp_thin_dupack) && tcp_stream_is_thin(tp) && tcp_dupack_heuristics(tp) > 1 && tcp_is_sack(tp) && !tcp_send_head(sk)) - return 1; - - return 0; -} - -/* New heuristics: it is possible only after we switched to restart timer - * each time when something is ACKed. Hence, we can detect timed out packets - * during fast retransmit without falling to slow start. - * - * Usefulness of this as is very questionable, since we should know which of - * the segments is the next to timeout which is relatively expensive to find - * in general case unless we add some data structure just for that. The - * current approach certainly won't find the right one too often and when it - * finally does find _something_ it usually marks large part of the window - * right away (because a retransmission with a larger timestamp blocks the - * loop from advancing). -ij - */ -static void tcp_timeout_skbs(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *skb; - - if (!tcp_is_fack(tp) || !tcp_head_timedout(sk)) - return; - - skb = tp->scoreboard_skb_hint; - if (tp->scoreboard_skb_hint == NULL) - skb = tcp_write_queue_head(sk); - - tcp_for_write_queue_from(skb, sk) { - if (skb == tcp_send_head(sk)) - break; - if (!tcp_skb_timedout(sk, skb)) - break; - - tcp_skb_mark_lost(tp, skb); - } + return true; - tp->scoreboard_skb_hint = skb; + /* Trick#6: TCP early retransmit, per RFC5827. To avoid spurious + * retransmissions due to small network reorderings, we implement + * Mitigation A.3 in the RFC and delay the retransmission for a short + * interval if appropriate. + */ + if (tp->do_early_retrans && !tp->retrans_out && tp->sacked_out && + (tp->packets_out >= (tp->sacked_out + 1) && tp->packets_out < 4) && + !tcp_may_send_now(sk)) + return !tcp_pause_early_retransmit(sk, flag); - tcp_verify_left_out(tp); + return false; } -/* Mark head of queue up as lost. With RFC3517 SACK, the packets is - * is against sacked "cnt", otherwise it's against facked "cnt" +/* Detect loss in event "A" above by marking head of queue up as lost. + * For FACK or non-SACK(Reno) senders, the first "packets" number of segments + * are considered lost. For RFC3517 SACK, a segment is considered lost if it + * has at least tp->reordering SACKed seqments above it; "packets" refers to + * the maximum SACKed segments to pass before reaching this limit. */ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) { @@ -2505,6 +2203,8 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) int cnt, oldcnt; int err; unsigned int mss; + /* Use SACK to deduce losses of new sequences sent during recovery */ + const u32 loss_high = tcp_is_sack(tp) ? tp->snd_nxt : tp->high_seq; WARN_ON(packets > tp->packets_out); if (tp->lost_skb_hint) { @@ -2526,7 +2226,7 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) tp->lost_skb_hint = skb; tp->lost_cnt_hint = cnt; - if (after(TCP_SKB_CB(skb)->end_seq, tp->high_seq)) + if (after(TCP_SKB_CB(skb)->end_seq, loss_high)) break; oldcnt = cnt; @@ -2536,11 +2236,13 @@ static void tcp_mark_head_lost(struct sock *sk, int packets, int mark_head) if (cnt > packets) { if ((tcp_is_sack(tp) && !tcp_is_fack(tp)) || + (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED) || (oldcnt >= packets)) break; mss = skb_shinfo(skb)->gso_size; - err = tcp_fragment(sk, skb, (packets - oldcnt) * mss, mss); + err = tcp_fragment(sk, skb, (packets - oldcnt) * mss, + mss, GFP_ATOMIC); if (err < 0) break; cnt = packets; @@ -2574,8 +2276,6 @@ static void tcp_update_scoreboard(struct sock *sk, int fast_rexmit) else if (fast_rexmit) tcp_mark_head_lost(sk, 1, 1); } - - tcp_timeout_skbs(sk); } /* CWND moderation, preventing bursts due to too big ACKs @@ -2588,39 +2288,10 @@ static inline void tcp_moderate_cwnd(struct tcp_sock *tp) tp->snd_cwnd_stamp = tcp_time_stamp; } -/* Lower bound on congestion window is slow start threshold - * unless congestion avoidance choice decides to overide it. - */ -static inline u32 tcp_cwnd_min(const struct sock *sk) -{ - const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops; - - return ca_ops->min_cwnd ? ca_ops->min_cwnd(sk) : tcp_sk(sk)->snd_ssthresh; -} - -/* Decrease cwnd each second ack. */ -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; - 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_stamp = tcp_time_stamp; - } -} - /* Nothing was retransmitted or returned timestamp is less * than timestamp of the first retransmission. */ -static inline int tcp_packet_delayed(struct tcp_sock *tp) +static inline bool tcp_packet_delayed(const struct tcp_sock *tp) { return !tp->retrans_stamp || (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && @@ -2636,22 +2307,22 @@ static void DBGUNDO(struct sock *sk, const char *msg) struct inet_sock *inet = inet_sk(sk); if (sk->sk_family == AF_INET) { - printk(KERN_DEBUG "Undo %s %pI4/%u c%u l%u ss%u/%u p%u\n", - msg, - &inet->inet_daddr, ntohs(inet->inet_dport), - tp->snd_cwnd, tcp_left_out(tp), - tp->snd_ssthresh, tp->prior_ssthresh, - tp->packets_out); - } -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + pr_debug("Undo %s %pI4/%u c%u l%u ss%u/%u p%u\n", + msg, + &inet->inet_daddr, ntohs(inet->inet_dport), + tp->snd_cwnd, tcp_left_out(tp), + tp->snd_ssthresh, tp->prior_ssthresh, + tp->packets_out); + } +#if IS_ENABLED(CONFIG_IPV6) else if (sk->sk_family == AF_INET6) { struct ipv6_pinfo *np = inet6_sk(sk); - printk(KERN_DEBUG "Undo %s %pI6/%u c%u l%u ss%u/%u p%u\n", - msg, - &np->daddr, ntohs(inet->inet_dport), - tp->snd_cwnd, tcp_left_out(tp), - tp->snd_ssthresh, tp->prior_ssthresh, - tp->packets_out); + pr_debug("Undo %s %pI6/%u c%u l%u ss%u/%u p%u\n", + msg, + &np->daddr, ntohs(inet->inet_dport), + tp->snd_cwnd, tcp_left_out(tp), + tp->snd_ssthresh, tp->prior_ssthresh, + tp->packets_out); } #endif } @@ -2659,10 +2330,22 @@ static void DBGUNDO(struct sock *sk, const char *msg) #define DBGUNDO(x...) do { } while (0) #endif -static void tcp_undo_cwr(struct sock *sk, const int undo) +static void tcp_undo_cwnd_reduction(struct sock *sk, bool unmark_loss) { struct tcp_sock *tp = tcp_sk(sk); + if (unmark_loss) { + struct sk_buff *skb; + + tcp_for_write_queue(skb, sk) { + if (skb == tcp_send_head(sk)) + break; + TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST; + } + tp->lost_out = 0; + tcp_clear_all_retrans_hints(tp); + } + if (tp->prior_ssthresh) { const struct inet_connection_sock *icsk = inet_csk(sk); @@ -2671,24 +2354,24 @@ static void tcp_undo_cwr(struct sock *sk, const int undo) else tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh << 1); - if (undo && tp->prior_ssthresh > tp->snd_ssthresh) { + if (tp->prior_ssthresh > tp->snd_ssthresh) { tp->snd_ssthresh = tp->prior_ssthresh; TCP_ECN_withdraw_cwr(tp); } } else { tp->snd_cwnd = max(tp->snd_cwnd, tp->snd_ssthresh); } - tcp_moderate_cwnd(tp); tp->snd_cwnd_stamp = tcp_time_stamp; + tp->undo_marker = 0; } -static inline int tcp_may_undo(struct tcp_sock *tp) +static inline bool tcp_may_undo(const struct tcp_sock *tp) { return tp->undo_marker && (!tp->undo_retrans || tcp_packet_delayed(tp)); } /* People celebrate: "We love our President!" */ -static int tcp_try_undo_recovery(struct sock *sk) +static bool tcp_try_undo_recovery(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); @@ -2699,37 +2382,37 @@ static int tcp_try_undo_recovery(struct sock *sk) * or our original transmission succeeded. */ DBGUNDO(sk, inet_csk(sk)->icsk_ca_state == TCP_CA_Loss ? "loss" : "retrans"); - tcp_undo_cwr(sk, 1); + tcp_undo_cwnd_reduction(sk, false); if (inet_csk(sk)->icsk_ca_state == TCP_CA_Loss) mib_idx = LINUX_MIB_TCPLOSSUNDO; else mib_idx = LINUX_MIB_TCPFULLUNDO; NET_INC_STATS_BH(sock_net(sk), mib_idx); - tp->undo_marker = 0; } if (tp->snd_una == tp->high_seq && tcp_is_reno(tp)) { /* Hold old state until something *above* high_seq * is ACKed. For Reno it is MUST to prevent false * fast retransmits (RFC2582). SACK TCP is safe. */ tcp_moderate_cwnd(tp); - return 1; + return true; } tcp_set_ca_state(sk, TCP_CA_Open); - return 0; + return false; } /* Try to undo cwnd reduction, because D-SACKs acked all retransmitted data */ -static void tcp_try_undo_dsack(struct sock *sk) +static bool tcp_try_undo_dsack(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); if (tp->undo_marker && !tp->undo_retrans) { DBGUNDO(sk, "D-SACK"); - tcp_undo_cwr(sk, 1); - tp->undo_marker = 0; + tcp_undo_cwnd_reduction(sk, false); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPDSACKUNDO); + return true; } + return false; } /* We can clear retrans_stamp when there are no retransmissions in the @@ -2746,85 +2429,115 @@ static void tcp_try_undo_dsack(struct sock *sk) * that successive retransmissions of a segment must not advance * retrans_stamp under any conditions. */ -static int tcp_any_retrans_done(struct sock *sk) +static bool tcp_any_retrans_done(const struct sock *sk) { - struct tcp_sock *tp = tcp_sk(sk); + const struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; if (tp->retrans_out) - return 1; + return true; skb = tcp_write_queue_head(sk); if (unlikely(skb && TCP_SKB_CB(skb)->sacked & TCPCB_EVER_RETRANS)) - return 1; + return true; - return 0; + return false; } -/* Undo during fast recovery after partial ACK. */ - -static int tcp_try_undo_partial(struct sock *sk, int acked) +/* Undo during loss recovery after partial ACK or using F-RTO. */ +static bool tcp_try_undo_loss(struct sock *sk, bool frto_undo) { struct tcp_sock *tp = tcp_sk(sk); - /* Partial ACK arrived. Force Hoe's retransmit. */ - int failed = tcp_is_reno(tp) || (tcp_fackets_out(tp) > tp->reordering); - if (tcp_may_undo(tp)) { - /* Plain luck! Hole if filled with delayed - * packet, rather than with a retransmit. - */ - if (!tcp_any_retrans_done(sk)) - tp->retrans_stamp = 0; + if (frto_undo || tcp_may_undo(tp)) { + tcp_undo_cwnd_reduction(sk, true); - tcp_update_reordering(sk, tcp_fackets_out(tp) + acked, 1); + DBGUNDO(sk, "partial loss"); + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSSUNDO); + if (frto_undo) + NET_INC_STATS_BH(sock_net(sk), + LINUX_MIB_TCPSPURIOUSRTOS); + inet_csk(sk)->icsk_retransmits = 0; + if (frto_undo || tcp_is_sack(tp)) + tcp_set_ca_state(sk, TCP_CA_Open); + return true; + } + return false; +} - DBGUNDO(sk, "Hoe"); - tcp_undo_cwr(sk, 0); - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPPARTIALUNDO); +/* The cwnd reduction in CWR and Recovery use the PRR algorithm + * https://datatracker.ietf.org/doc/draft-ietf-tcpm-proportional-rate-reduction/ + * It computes the number of packets to send (sndcnt) based on packets newly + * delivered: + * 1) If the packets in flight is larger than ssthresh, PRR spreads the + * cwnd reductions across a full RTT. + * 2) If packets in flight is lower than ssthresh (such as due to excess + * losses and/or application stalls), do not perform any further cwnd + * reductions, but instead slow start up to ssthresh. + */ +static void tcp_init_cwnd_reduction(struct sock *sk, const bool set_ssthresh) +{ + struct tcp_sock *tp = tcp_sk(sk); - /* So... Do not make Hoe's retransmit yet. - * If the first packet was delayed, the rest - * ones are most probably delayed as well. - */ - failed = 0; - } - return failed; + tp->high_seq = tp->snd_nxt; + tp->tlp_high_seq = 0; + tp->snd_cwnd_cnt = 0; + tp->prior_cwnd = tp->snd_cwnd; + tp->prr_delivered = 0; + tp->prr_out = 0; + if (set_ssthresh) + tp->snd_ssthresh = inet_csk(sk)->icsk_ca_ops->ssthresh(sk); + TCP_ECN_queue_cwr(tp); } -/* Undo during loss recovery after partial ACK. */ -static int tcp_try_undo_loss(struct sock *sk) +static void tcp_cwnd_reduction(struct sock *sk, const int prior_unsacked, + int fast_rexmit) { struct tcp_sock *tp = tcp_sk(sk); + int sndcnt = 0; + int delta = tp->snd_ssthresh - tcp_packets_in_flight(tp); + int newly_acked_sacked = prior_unsacked - + (tp->packets_out - tp->sacked_out); + + tp->prr_delivered += newly_acked_sacked; + if (tcp_packets_in_flight(tp) > tp->snd_ssthresh) { + u64 dividend = (u64)tp->snd_ssthresh * tp->prr_delivered + + tp->prior_cwnd - 1; + sndcnt = div_u64(dividend, tp->prior_cwnd) - tp->prr_out; + } else { + sndcnt = min_t(int, delta, + max_t(int, tp->prr_delivered - tp->prr_out, + newly_acked_sacked) + 1); + } - if (tcp_may_undo(tp)) { - struct sk_buff *skb; - tcp_for_write_queue(skb, sk) { - if (skb == tcp_send_head(sk)) - break; - TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST; - } + sndcnt = max(sndcnt, (fast_rexmit ? 1 : 0)); + tp->snd_cwnd = tcp_packets_in_flight(tp) + sndcnt; +} - tcp_clear_all_retrans_hints(tp); +static inline void tcp_end_cwnd_reduction(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); - DBGUNDO(sk, "partial loss"); - tp->lost_out = 0; - tcp_undo_cwr(sk, 1); - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSSUNDO); - inet_csk(sk)->icsk_retransmits = 0; - tp->undo_marker = 0; - if (tcp_is_sack(tp)) - tcp_set_ca_state(sk, TCP_CA_Open); - return 1; + /* Reset cwnd to ssthresh in CWR or Recovery (unless it's undone) */ + if (inet_csk(sk)->icsk_ca_state == TCP_CA_CWR || + (tp->undo_marker && tp->snd_ssthresh < TCP_INFINITE_SSTHRESH)) { + tp->snd_cwnd = tp->snd_ssthresh; + tp->snd_cwnd_stamp = tcp_time_stamp; } - return 0; + tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR); } -static inline void tcp_complete_cwr(struct sock *sk) +/* Enter CWR state. Disable cwnd undo since congestion is proven with ECN */ +void tcp_enter_cwr(struct sock *sk, const int set_ssthresh) { struct tcp_sock *tp = tcp_sk(sk); - tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); - tp->snd_cwnd_stamp = tcp_time_stamp; - tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR); + + tp->prior_ssthresh = 0; + if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) { + tp->undo_marker = 0; + tcp_init_cwnd_reduction(sk, set_ssthresh); + tcp_set_ca_state(sk, TCP_CA_CWR); + } } static void tcp_try_keep_open(struct sock *sk) @@ -2832,7 +2545,7 @@ static void tcp_try_keep_open(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); int state = TCP_CA_Open; - if (tcp_left_out(tp) || tcp_any_retrans_done(sk) || tp->undo_marker) + if (tcp_left_out(tp) || tcp_any_retrans_done(sk)) state = TCP_CA_Disorder; if (inet_csk(sk)->icsk_ca_state != state) { @@ -2841,13 +2554,13 @@ static void tcp_try_keep_open(struct sock *sk) } } -static void tcp_try_to_open(struct sock *sk, int flag) +static void tcp_try_to_open(struct sock *sk, int flag, const int prior_unsacked) { struct tcp_sock *tp = tcp_sk(sk); tcp_verify_left_out(tp); - if (!tp->frto_counter && !tcp_any_retrans_done(sk)) + if (!tcp_any_retrans_done(sk)) tp->retrans_stamp = 0; if (flag & FLAG_ECE) @@ -2855,9 +2568,8 @@ static void tcp_try_to_open(struct sock *sk, int flag) if (inet_csk(sk)->icsk_ca_state != TCP_CA_CWR) { tcp_try_keep_open(sk); - tcp_moderate_cwnd(tp); } else { - tcp_cwnd_down(sk, flag); + tcp_cwnd_reduction(sk, prior_unsacked, 0); } } @@ -2939,6 +2651,115 @@ void tcp_simple_retransmit(struct sock *sk) } EXPORT_SYMBOL(tcp_simple_retransmit); +static void tcp_enter_recovery(struct sock *sk, bool ece_ack) +{ + struct tcp_sock *tp = tcp_sk(sk); + int mib_idx; + + if (tcp_is_reno(tp)) + mib_idx = LINUX_MIB_TCPRENORECOVERY; + else + mib_idx = LINUX_MIB_TCPSACKRECOVERY; + + NET_INC_STATS_BH(sock_net(sk), mib_idx); + + tp->prior_ssthresh = 0; + tp->undo_marker = tp->snd_una; + tp->undo_retrans = tp->retrans_out ? : -1; + + if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) { + if (!ece_ack) + tp->prior_ssthresh = tcp_current_ssthresh(sk); + tcp_init_cwnd_reduction(sk, true); + } + tcp_set_ca_state(sk, TCP_CA_Recovery); +} + +/* Process an ACK in CA_Loss state. Move to CA_Open if lost data are + * recovered or spurious. Otherwise retransmits more on partial ACKs. + */ +static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + struct tcp_sock *tp = tcp_sk(sk); + bool recovered = !before(tp->snd_una, tp->high_seq); + + if (tp->frto) { /* F-RTO RFC5682 sec 3.1 (sack enhanced version). */ + /* Step 3.b. A timeout is spurious if not all data are + * lost, i.e., never-retransmitted data are (s)acked. + */ + if (tcp_try_undo_loss(sk, flag & FLAG_ORIG_SACK_ACKED)) + return; + + if (after(tp->snd_nxt, tp->high_seq) && + (flag & FLAG_DATA_SACKED || is_dupack)) { + tp->frto = 0; /* Loss was real: 2nd part of step 3.a */ + } else if (flag & FLAG_SND_UNA_ADVANCED && !recovered) { + tp->high_seq = tp->snd_nxt; + __tcp_push_pending_frames(sk, tcp_current_mss(sk), + TCP_NAGLE_OFF); + if (after(tp->snd_nxt, tp->high_seq)) + return; /* Step 2.b */ + tp->frto = 0; + } + } + + if (recovered) { + /* F-RTO RFC5682 sec 3.1 step 2.a and 1st part of step 3.a */ + icsk->icsk_retransmits = 0; + tcp_try_undo_recovery(sk); + return; + } + if (flag & FLAG_DATA_ACKED) + icsk->icsk_retransmits = 0; + if (tcp_is_reno(tp)) { + /* A Reno DUPACK means new data in F-RTO step 2.b above are + * delivered. Lower inflight to clock out (re)tranmissions. + */ + if (after(tp->snd_nxt, tp->high_seq) && is_dupack) + tcp_add_reno_sack(sk); + else if (flag & FLAG_SND_UNA_ADVANCED) + tcp_reset_reno_sack(tp); + } + if (tcp_try_undo_loss(sk, false)) + return; + tcp_xmit_retransmit_queue(sk); +} + +/* Undo during fast recovery after partial ACK. */ +static bool tcp_try_undo_partial(struct sock *sk, const int acked, + const int prior_unsacked) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (tp->undo_marker && tcp_packet_delayed(tp)) { + /* Plain luck! Hole if filled with delayed + * packet, rather than with a retransmit. + */ + tcp_update_reordering(sk, tcp_fackets_out(tp) + acked, 1); + + /* We are getting evidence that the reordering degree is higher + * than we realized. If there are no retransmits out then we + * can undo. Otherwise we clock out new packets but do not + * mark more packets lost or retransmit more. + */ + if (tp->retrans_out) { + tcp_cwnd_reduction(sk, prior_unsacked, 0); + return true; + } + + if (!tcp_any_retrans_done(sk)) + tp->retrans_stamp = 0; + + DBGUNDO(sk, "partial recovery"); + tcp_undo_cwnd_reduction(sk, true); + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPPARTIALUNDO); + tcp_try_keep_open(sk); + return true; + } + return false; +} + /* 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 @@ -2950,14 +2771,15 @@ EXPORT_SYMBOL(tcp_simple_retransmit); * 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, const int acked, + const int prior_unsacked, + bool is_dupack, 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) && + bool do_lost = is_dupack || ((flag & FLAG_DATA_SACKED) && (tcp_fackets_out(tp) > tp->reordering)); - int fast_rexmit = 0, mib_idx; + int fast_rexmit = 0; if (WARN_ON(!tp->packets_out && tp->sacked_out)) tp->sacked_out = 0; @@ -2973,47 +2795,21 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) if (tcp_check_sack_reneging(sk, flag)) return; - /* C. Process data loss notification, provided it is valid. */ - 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, 0); - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSS); - } - - /* D. Check consistency of the current state. */ + /* C. Check consistency of the current state. */ tcp_verify_left_out(tp); - /* E. Check state exit conditions. State can be terminated + /* D. Check state exit conditions. State can be terminated * when high_seq is ACKed. */ if (icsk->icsk_ca_state == TCP_CA_Open) { WARN_ON(tp->retrans_out != 0); tp->retrans_stamp = 0; } else if (!before(tp->snd_una, tp->high_seq)) { switch (icsk->icsk_ca_state) { - case TCP_CA_Loss: - icsk->icsk_retransmits = 0; - if (tcp_try_undo_recovery(sk)) - return; - break; - case TCP_CA_CWR: /* CWR is to be held something *above* high_seq * is ACKed for CWR bit to reach receiver. */ if (tp->snd_una != tp->high_seq) { - tcp_complete_cwr(sk); - tcp_set_ca_state(sk, TCP_CA_Open); - } - break; - - case TCP_CA_Disorder: - tcp_try_undo_dsack(sk); - if (!tp->undo_marker || - /* For SACK case do not Open to allow to undo - * catching for all duplicate ACKs. */ - tcp_is_reno(tp) || tp->snd_una != tp->high_seq) { - tp->undo_marker = 0; + tcp_end_cwnd_reduction(sk); tcp_set_ca_state(sk, TCP_CA_Open); } break; @@ -3023,33 +2819,34 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) tcp_reset_reno_sack(tp); if (tcp_try_undo_recovery(sk)) return; - tcp_complete_cwr(sk); + tcp_end_cwnd_reduction(sk); break; } } - /* F. Process state. */ + /* E. Process state. */ switch (icsk->icsk_ca_state) { case TCP_CA_Recovery: if (!(flag & FLAG_SND_UNA_ADVANCED)) { if (tcp_is_reno(tp) && is_dupack) tcp_add_reno_sack(sk); - } else - do_lost = tcp_try_undo_partial(sk, pkts_acked); - break; - case TCP_CA_Loss: - if (flag & FLAG_DATA_ACKED) - icsk->icsk_retransmits = 0; - if (tcp_is_reno(tp) && flag & FLAG_SND_UNA_ADVANCED) - tcp_reset_reno_sack(tp); - if (!tcp_try_undo_loss(sk)) { - tcp_moderate_cwnd(tp); - tcp_xmit_retransmit_queue(sk); + } else { + if (tcp_try_undo_partial(sk, acked, prior_unsacked)) + return; + /* Partial ACK arrived. Force fast retransmit. */ + do_lost = tcp_is_reno(tp) || + tcp_fackets_out(tp) > tp->reordering; + } + if (tcp_try_undo_dsack(sk)) { + tcp_try_keep_open(sk); return; } + break; + case TCP_CA_Loss: + tcp_process_loss(sk, flag, is_dupack); if (icsk->icsk_ca_state != TCP_CA_Open) return; - /* Loss is undone; fall through to processing in Open state. */ + /* Fall through to processing in Open state. */ default: if (tcp_is_reno(tp)) { if (flag & FLAG_SND_UNA_ADVANCED) @@ -3058,11 +2855,11 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) tcp_add_reno_sack(sk); } - if (icsk->icsk_ca_state == TCP_CA_Disorder) + if (icsk->icsk_ca_state <= TCP_CA_Disorder) tcp_try_undo_dsack(sk); - if (!tcp_time_to_recover(sk)) { - tcp_try_to_open(sk, flag); + if (!tcp_time_to_recover(sk, flag)) { + tcp_try_to_open(sk, flag, prior_unsacked); return; } @@ -3078,120 +2875,130 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag) } /* Otherwise enter Recovery state */ - - if (tcp_is_reno(tp)) - mib_idx = LINUX_MIB_TCPRENORECOVERY; - else - mib_idx = LINUX_MIB_TCPSACKRECOVERY; - - NET_INC_STATS_BH(sock_net(sk), mib_idx); - - tp->high_seq = tp->snd_nxt; - tp->prior_ssthresh = 0; - tp->undo_marker = tp->snd_una; - tp->undo_retrans = tp->retrans_out; - - if (icsk->icsk_ca_state < TCP_CA_CWR) { - 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); - } - - tp->bytes_acked = 0; - tp->snd_cwnd_cnt = 0; - tcp_set_ca_state(sk, TCP_CA_Recovery); + tcp_enter_recovery(sk, (flag & FLAG_ECE)); fast_rexmit = 1; } - if (do_lost || (tcp_is_fack(tp) && tcp_head_timedout(sk))) + if (do_lost) tcp_update_scoreboard(sk, fast_rexmit); - tcp_cwnd_down(sk, flag); + tcp_cwnd_reduction(sk, prior_unsacked, fast_rexmit); tcp_xmit_retransmit_queue(sk); } -static void tcp_valid_rtt_meas(struct sock *sk, u32 seq_rtt) +static inline bool tcp_ack_update_rtt(struct sock *sk, const int flag, + long seq_rtt_us, long sack_rtt_us) { - tcp_rtt_estimator(sk, seq_rtt); - tcp_set_rto(sk); - inet_csk(sk)->icsk_backoff = 0; -} + const struct tcp_sock *tp = tcp_sk(sk); + + /* Prefer RTT measured from ACK's timing to TS-ECR. This is because + * broken middle-boxes or peers may corrupt TS-ECR fields. But + * Karn's algorithm forbids taking RTT if some retransmitted data + * is acked (RFC6298). + */ + if (flag & FLAG_RETRANS_DATA_ACKED) + seq_rtt_us = -1L; + + if (seq_rtt_us < 0) + seq_rtt_us = sack_rtt_us; -/* Read draft-ietf-tcplw-high-performance before mucking - * with this code. (Supersedes RFC1323) - */ -static void tcp_ack_saw_tstamp(struct sock *sk, int flag) -{ /* RTTM Rule: A TSecr value received in a segment is used to * update the averaged RTT measurement only if the segment * acknowledges some new data, i.e., only if it advances the * left edge of the send window. - * * See draft-ietf-tcplw-high-performance-00, section 3.3. - * 1998/04/10 Andrey V. Savochkin <saw@msu.ru> - * - * Changed: reset backoff as soon as we see the first valid sample. - * If we do not, we get strongly overestimated rto. With timestamps - * samples are accepted even from very old segments: f.e., when rtt=1 - * increases to 8, we retransmit 5 times and after 8 seconds delayed - * answer arrives rto becomes 120 seconds! If at least one of segments - * in window is lost... Voila. --ANK (010210) */ - struct tcp_sock *tp = tcp_sk(sk); - - tcp_valid_rtt_meas(sk, tcp_time_stamp - tp->rx_opt.rcv_tsecr); -} + if (seq_rtt_us < 0 && tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && + flag & FLAG_ACKED) + seq_rtt_us = jiffies_to_usecs(tcp_time_stamp - tp->rx_opt.rcv_tsecr); -static void tcp_ack_no_tstamp(struct sock *sk, u32 seq_rtt, int flag) -{ - /* We don't have a timestamp. Can only use - * packets that are not retransmitted to determine - * rtt estimates. Also, we must not reset the - * backoff for rto until we get a non-retransmitted - * packet. This allows us to deal with a situation - * where the network delay has increased suddenly. - * I.e. Karn's algorithm. (SIGCOMM '87, p5.) - */ + if (seq_rtt_us < 0) + return false; - if (flag & FLAG_RETRANS_DATA_ACKED) - return; + tcp_rtt_estimator(sk, seq_rtt_us); + tcp_set_rto(sk); - tcp_valid_rtt_meas(sk, seq_rtt); + /* RFC6298: only reset backoff on valid RTT measurement. */ + inet_csk(sk)->icsk_backoff = 0; + return true; } -static inline void tcp_ack_update_rtt(struct sock *sk, const int flag, - const s32 seq_rtt) +/* Compute time elapsed between (last) SYNACK and the ACK completing 3WHS. */ +static void tcp_synack_rtt_meas(struct sock *sk, const u32 synack_stamp) { - const struct tcp_sock *tp = tcp_sk(sk); - /* Note that peer MAY send zero echo. In this case it is ignored. (rfc1323) */ - if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) - tcp_ack_saw_tstamp(sk, flag); - else if (seq_rtt >= 0) - tcp_ack_no_tstamp(sk, seq_rtt, flag); + struct tcp_sock *tp = tcp_sk(sk); + long seq_rtt_us = -1L; + + if (synack_stamp && !tp->total_retrans) + seq_rtt_us = jiffies_to_usecs(tcp_time_stamp - synack_stamp); + + /* If the ACK acks both the SYNACK and the (Fast Open'd) data packets + * sent in SYN_RECV, SYNACK RTT is the smooth RTT computed in tcp_ack() + */ + if (!tp->srtt_us) + tcp_ack_update_rtt(sk, FLAG_SYN_ACKED, seq_rtt_us, -1L); } -static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 in_flight) +static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 acked) { const struct inet_connection_sock *icsk = inet_csk(sk); - icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight); + + icsk->icsk_ca_ops->cong_avoid(sk, ack, acked); tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp; } /* Restart timer after forward progress on connection. * RFC2988 recommends to restart timer to now+rto. */ -static void tcp_rearm_rto(struct sock *sk) +void tcp_rearm_rto(struct sock *sk) { + const struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); + /* If the retrans timer is currently being used by Fast Open + * for SYN-ACK retrans purpose, stay put. + */ + if (tp->fastopen_rsk) + return; + 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); + u32 rto = inet_csk(sk)->icsk_rto; + /* Offset the time elapsed after installing regular RTO */ + if (icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS || + icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) { + struct sk_buff *skb = tcp_write_queue_head(sk); + const u32 rto_time_stamp = TCP_SKB_CB(skb)->when + rto; + s32 delta = (s32)(rto_time_stamp - tcp_time_stamp); + /* delta may not be positive if the socket is locked + * when the retrans timer fires and is rescheduled. + */ + if (delta > 0) + rto = delta; + } + inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS, rto, + TCP_RTO_MAX); } } +/* This function is called when the delayed ER timer fires. TCP enters + * fast recovery and performs fast-retransmit. + */ +void tcp_resume_early_retransmit(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + tcp_rearm_rto(sk); + + /* Stop if ER is disabled after the delayed ER timer is scheduled */ + if (!tp->do_early_retrans) + return; + + tcp_enter_recovery(sk, false); + tcp_update_scoreboard(sk, 1); + tcp_xmit_retransmit_queue(sk); +} + /* If we get here, the whole TSO packet has not been acked. */ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb) { @@ -3218,25 +3025,27 @@ static u32 tcp_tso_acked(struct sock *sk, struct sk_buff *skb) * arrived at the other end. */ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, - u32 prior_snd_una) + u32 prior_snd_una, long sack_rtt_us) { - struct tcp_sock *tp = tcp_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); + struct skb_mstamp first_ackt, last_ackt, now; + struct tcp_sock *tp = tcp_sk(sk); + u32 prior_sacked = tp->sacked_out; + u32 reord = tp->packets_out; + bool fully_acked = true; + long ca_seq_rtt_us = -1L; + long seq_rtt_us = -1L; struct sk_buff *skb; - u32 now = tcp_time_stamp; - int fully_acked = 1; - int flag = 0; u32 pkts_acked = 0; - u32 reord = tp->packets_out; - u32 prior_sacked = tp->sacked_out; - s32 seq_rtt = -1; - s32 ca_seq_rtt = -1; - ktime_t last_ackt = net_invalid_timestamp(); + bool rtt_update; + int flag = 0; + + first_ackt.v64 = 0; while ((skb = tcp_write_queue_head(sk)) && skb != tcp_send_head(sk)) { struct tcp_skb_cb *scb = TCP_SKB_CB(skb); - u32 acked_pcount; u8 sacked = scb->sacked; + u32 acked_pcount; /* Determine how many packets and what bytes were acked, tso and else */ if (after(scb->end_seq, tp->snd_una)) { @@ -3248,7 +3057,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (!acked_pcount) break; - fully_acked = 0; + fully_acked = false; } else { acked_pcount = tcp_skb_pcount(skb); } @@ -3257,18 +3066,16 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, 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 = ca_seq_rtt; - } + last_ackt = skb->skb_mstamp; + WARN_ON_ONCE(last_ackt.v64 == 0); + if (!first_ackt.v64) + first_ackt = last_ackt; + if (!(sacked & TCPCB_SACKED_ACKED)) reord = min(pkts_acked, reord); + if (!after(scb->end_seq, tp->high_seq)) + flag |= FLAG_ORIG_SACK_ACKED; } if (sacked & TCPCB_SACKED_ACKED) @@ -3286,7 +3093,7 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, * connection startup slow start one packet too * quickly. This is severely frowned upon behavior. */ - if (!(scb->flags & TCPHDR_SYN)) { + if (!(scb->tcp_flags & TCPHDR_SYN)) { flag |= FLAG_DATA_ACKED; } else { flag |= FLAG_SYN_ACKED; @@ -3298,7 +3105,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, tcp_unlink_write_queue(skb, sk); sk_wmem_free_skb(sk, skb); - tp->scoreboard_skb_hint = NULL; if (skb == tp->retransmit_skb_hint) tp->retransmit_skb_hint = NULL; if (skb == tp->lost_skb_hint) @@ -3311,18 +3117,24 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) flag |= FLAG_SACK_RENEGING; + skb_mstamp_get(&now); + if (first_ackt.v64) { + seq_rtt_us = skb_mstamp_us_delta(&now, &first_ackt); + ca_seq_rtt_us = skb_mstamp_us_delta(&now, &last_ackt); + } + + rtt_update = tcp_ack_update_rtt(sk, flag, seq_rtt_us, sack_rtt_us); + if (flag & FLAG_ACKED) { const struct tcp_congestion_ops *ca_ops = inet_csk(sk)->icsk_ca_ops; + tcp_rearm_rto(sk); if (unlikely(icsk->icsk_mtup.probe_size && !after(tp->mtu_probe.probe_seq_end, tp->snd_una))) { tcp_mtup_probe_success(sk); } - 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 { @@ -3339,23 +3151,16 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, tp->fackets_out -= min(pkts_acked, tp->fackets_out); - if (ca_ops->pkts_acked) { - s32 rtt_us = -1; - - /* Is the ACK triggering packet unambiguous? */ - if (!(flag & FLAG_RETRANS_DATA_ACKED)) { - /* High resolution needed and available? */ - if (ca_ops->flags & TCP_CONG_RTT_STAMP && - !ktime_equal(last_ackt, - net_invalid_timestamp())) - rtt_us = ktime_us_delta(ktime_get_real(), - last_ackt); - else if (ca_seq_rtt > 0) - rtt_us = jiffies_to_usecs(ca_seq_rtt); - } + if (ca_ops->pkts_acked) + ca_ops->pkts_acked(sk, pkts_acked, ca_seq_rtt_us); - ca_ops->pkts_acked(sk, pkts_acked, rtt_us); - } + } else if (skb && rtt_update && sack_rtt_us >= 0 && + sack_rtt_us > skb_mstamp_us_delta(&now, &skb->skb_mstamp)) { + /* Do not re-arm RTO if the sack RTT is measured from data sent + * after when the head was last (re)transmitted. Otherwise the + * timeout may continue to extend in loss recovery. + */ + tcp_rearm_rto(sk); } #if FASTRETRANS_DEBUG > 0 @@ -3365,18 +3170,18 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, if (!tp->packets_out && tcp_is_sack(tp)) { icsk = inet_csk(sk); if (tp->lost_out) { - printk(KERN_DEBUG "Leak l=%u %d\n", - tp->lost_out, icsk->icsk_ca_state); + pr_debug("Leak l=%u %d\n", + tp->lost_out, icsk->icsk_ca_state); tp->lost_out = 0; } if (tp->sacked_out) { - printk(KERN_DEBUG "Leak s=%u %d\n", - tp->sacked_out, icsk->icsk_ca_state); + pr_debug("Leak s=%u %d\n", + tp->sacked_out, icsk->icsk_ca_state); tp->sacked_out = 0; } if (tp->retrans_out) { - printk(KERN_DEBUG "Leak r=%u %d\n", - tp->retrans_out, icsk->icsk_ca_state); + pr_debug("Leak r=%u %d\n", + tp->retrans_out, icsk->icsk_ca_state); tp->retrans_out = 0; } } @@ -3404,23 +3209,34 @@ static void tcp_ack_probe(struct sock *sk) } } -static inline int tcp_ack_is_dubious(const struct sock *sk, const int flag) +static inline bool tcp_ack_is_dubious(const struct sock *sk, const int flag) { return !(flag & FLAG_NOT_DUP) || (flag & FLAG_CA_ALERT) || inet_csk(sk)->icsk_ca_state != TCP_CA_Open; } -static inline int tcp_may_raise_cwnd(const struct sock *sk, const int flag) +/* Decide wheather to run the increase function of congestion control. */ +static inline bool tcp_may_raise_cwnd(const struct sock *sk, const int flag) { - const struct tcp_sock *tp = tcp_sk(sk); - return (!(flag & FLAG_ECE) || tp->snd_cwnd < tp->snd_ssthresh) && - !((1 << inet_csk(sk)->icsk_ca_state) & (TCPF_CA_Recovery | TCPF_CA_CWR)); + if (tcp_in_cwnd_reduction(sk)) + return false; + + /* If reordering is high then always grow cwnd whenever data is + * delivered regardless of its ordering. Otherwise stay conservative + * and only grow cwnd on in-order delivery (RFC5681). A stretched ACK w/ + * new SACK or ECE mark may first advance cwnd here and later reduce + * cwnd in tcp_fastretrans_alert() based on more states. + */ + if (tcp_sk(sk)->reordering > sysctl_tcp_reordering) + return flag & FLAG_FORWARD_PROGRESS; + + return flag & FLAG_DATA_ACKED; } /* 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, +static inline bool tcp_may_update_window(const struct tcp_sock *tp, const u32 ack, const u32 ack_seq, const u32 nwin) { @@ -3434,7 +3250,7 @@ static inline int tcp_may_update_window(const struct tcp_sock *tp, * Window update algorithm, described in RFC793/RFC1122 (used in linux-2.2 * and in FreeBSD. NetBSD's one is even worse.) is wrong. */ -static int tcp_ack_update_window(struct sock *sk, struct sk_buff *skb, u32 ack, +static int tcp_ack_update_window(struct sock *sk, const struct sk_buff *skb, u32 ack, u32 ack_seq) { struct tcp_sock *tp = tcp_sk(sk); @@ -3469,164 +3285,103 @@ static int tcp_ack_update_window(struct sock *sk, struct sk_buff *skb, u32 ack, return flag; } -/* A very conservative spurious RTO response algorithm: reduce cwnd and - * continue in congestion avoidance. - */ -static void tcp_conservative_spur_to_response(struct tcp_sock *tp) +/* RFC 5961 7 [ACK Throttling] */ +static void tcp_send_challenge_ack(struct sock *sk) { - tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh); - tp->snd_cwnd_cnt = 0; - tp->bytes_acked = 0; - TCP_ECN_queue_cwr(tp); - tcp_moderate_cwnd(tp); + /* unprotected vars, we dont care of overwrites */ + static u32 challenge_timestamp; + static unsigned int challenge_count; + u32 now = jiffies / HZ; + + if (now != challenge_timestamp) { + challenge_timestamp = now; + challenge_count = 0; + } + if (++challenge_count <= sysctl_tcp_challenge_ack_limit) { + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPCHALLENGEACK); + tcp_send_ack(sk); + } } -/* A conservative spurious RTO response algorithm: reduce cwnd using - * rate halving and continue in congestion avoidance. - */ -static void tcp_ratehalving_spur_to_response(struct sock *sk) +static void tcp_store_ts_recent(struct tcp_sock *tp) { - tcp_enter_cwr(sk, 0); + tp->rx_opt.ts_recent = tp->rx_opt.rcv_tsval; + tp->rx_opt.ts_recent_stamp = get_seconds(); } -static void tcp_undo_spur_to_response(struct sock *sk, int flag) +static void tcp_replace_ts_recent(struct tcp_sock *tp, u32 seq) { - if (flag & FLAG_ECE) - tcp_ratehalving_spur_to_response(sk); - else - tcp_undo_cwr(sk, 1); + if (tp->rx_opt.saw_tstamp && !after(seq, tp->rcv_wup)) { + /* PAWS bug workaround wrt. ACK frames, the PAWS discard + * extra check below makes sure this can only happen + * for pure ACK frames. -DaveM + * + * Not only, also it occurs for expired timestamps. + */ + + if (tcp_paws_check(&tp->rx_opt, 0)) + tcp_store_ts_recent(tp); + } } -/* F-RTO spurious RTO detection algorithm (RFC4138) - * - * F-RTO affects during two new ACKs following RTO (well, almost, see inline - * comments). State (ACK number) is kept in frto_counter. When ACK advances - * window (but not to or beyond highest sequence sent before RTO): - * On First ACK, send two new segments out. - * On Second ACK, RTO was likely spurious. Do spurious response (response - * algorithm is not part of the F-RTO detection algorithm - * given in RFC4138 but can be selected separately). - * Otherwise (basically on duplicate ACK), RTO was (likely) caused by a loss - * and TCP falls back to conventional RTO recovery. F-RTO allows overriding - * of Nagle, this is done using frto_counter states 2 and 3, when a new data - * segment of any size sent during F-RTO, state 2 is upgraded to 3. - * - * Rationale: if the RTO was spurious, new ACKs should arrive from the - * original window even after we transmit two new data segments. - * - * SACK version: - * on first step, wait until first cumulative ACK arrives, then move to - * the second step. In second step, the next ACK decides. - * - * F-RTO is implemented (mainly) in four functions: - * - tcp_use_frto() is used to determine if TCP is can use F-RTO - * - tcp_enter_frto() prepares TCP state on RTO if F-RTO is used, it is - * called when tcp_use_frto() showed green light - * - tcp_process_frto() handles incoming ACKs during F-RTO algorithm - * - tcp_enter_frto_loss() is called if there is not enough evidence - * to prove that the RTO is indeed spurious. It transfers the control - * from F-RTO to the conventional RTO recovery +/* This routine deals with acks during a TLP episode. + * Ref: loss detection algorithm in draft-dukkipati-tcpm-tcp-loss-probe. */ -static int tcp_process_frto(struct sock *sk, int flag) +static void tcp_process_tlp_ack(struct sock *sk, u32 ack, int flag) { struct tcp_sock *tp = tcp_sk(sk); + bool is_tlp_dupack = (ack == tp->tlp_high_seq) && + !(flag & (FLAG_SND_UNA_ADVANCED | + FLAG_NOT_DUP | FLAG_DATA_SACKED)); - tcp_verify_left_out(tp); - - /* Duplicate the behavior from Loss state (fastretrans_alert) */ - if (flag & FLAG_DATA_ACKED) - inet_csk(sk)->icsk_retransmits = 0; - - if ((flag & FLAG_NONHEAD_RETRANS_ACKED) || - ((tp->frto_counter >= 2) && (flag & FLAG_RETRANS_DATA_ACKED))) - tp->undo_marker = 0; - - if (!before(tp->snd_una, tp->frto_highmark)) { - tcp_enter_frto_loss(sk, (tp->frto_counter == 1 ? 2 : 3), flag); - return 1; - } - - if (!tcp_is_sackfrto(tp)) { - /* RFC4138 shortcoming in step 2; should also have case c): - * ACK isn't duplicate nor advances window, e.g., opposite dir - * data, winupdate - */ - if (!(flag & FLAG_ANY_PROGRESS) && (flag & FLAG_NOT_DUP)) - return 1; - - 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)) { - /* Prevent sending of new data. */ - tp->snd_cwnd = min(tp->snd_cwnd, - tcp_packets_in_flight(tp)); - return 1; - } - - if ((tp->frto_counter >= 2) && - (!(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)) - return 1; - - tcp_enter_frto_loss(sk, 3, flag); - return 1; - } + /* Mark the end of TLP episode on receiving TLP dupack or when + * ack is after tlp_high_seq. + */ + if (is_tlp_dupack) { + tp->tlp_high_seq = 0; + return; } - if (tp->frto_counter == 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) { - case 2: - tcp_undo_spur_to_response(sk, flag); - break; - case 1: - tcp_conservative_spur_to_response(tp); - break; - default: - tcp_ratehalving_spur_to_response(sk); - break; + if (after(ack, tp->tlp_high_seq)) { + tp->tlp_high_seq = 0; + /* Don't reduce cwnd if DSACK arrives for TLP retrans. */ + if (!(flag & FLAG_DSACKING_ACK)) { + tcp_init_cwnd_reduction(sk, true); + tcp_set_ca_state(sk, TCP_CA_CWR); + tcp_end_cwnd_reduction(sk); + tcp_try_keep_open(sk); + NET_INC_STATS_BH(sock_net(sk), + LINUX_MIB_TCPLOSSPROBERECOVERY); } - tp->frto_counter = 0; - tp->undo_marker = 0; - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSPURIOUSRTOS); } - return 0; } /* This routine deals with incoming acks, but not outgoing ones. */ -static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) +static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); u32 prior_snd_una = tp->snd_una; u32 ack_seq = TCP_SKB_CB(skb)->seq; u32 ack = TCP_SKB_CB(skb)->ack_seq; - u32 prior_in_flight; + bool is_dupack = false; u32 prior_fackets; - int prior_packets; - int frto_cwnd = 0; + int prior_packets = tp->packets_out; + const int prior_unsacked = tp->packets_out - tp->sacked_out; + int acked = 0; /* Number of packets newly acked */ + long sack_rtt_us = -1L; /* If the ack is older than previous acks * then we can probably ignore it. */ - if (before(ack, prior_snd_una)) + if (before(ack, prior_snd_una)) { + /* RFC 5961 5.2 [Blind Data Injection Attack].[Mitigation] */ + if (before(ack, prior_snd_una - tp->max_window)) { + tcp_send_challenge_ack(sk); + return -1; + } goto old_ack; + } /* If the ack includes data we haven't sent yet, discard * this segment (RFC793 Section 3.9). @@ -3634,20 +3389,20 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) if (after(ack, tp->snd_nxt)) goto invalid_ack; + if (icsk->icsk_pending == ICSK_TIME_EARLY_RETRANS || + icsk->icsk_pending == ICSK_TIME_LOSS_PROBE) + tcp_rearm_rto(sk); + if (after(ack, prior_snd_una)) flag |= FLAG_SND_UNA_ADVANCED; - if (sysctl_tcp_abc) { - if (icsk->icsk_ca_state < TCP_CA_CWR) - 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); - } - prior_fackets = tp->fackets_out; - prior_in_flight = tcp_packets_in_flight(tp); + + /* ts_recent update must be made after we are sure that the packet + * is in window. + */ + if (flag & FLAG_UPDATE_TS_RECENT) + tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq); if (!(flag & FLAG_SLOWPATH) && after(ack, prior_snd_una)) { /* Window is constant, pure forward advance. @@ -3670,7 +3425,8 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) flag |= tcp_ack_update_window(sk, skb, ack, ack_seq); if (TCP_SKB_CB(skb)->sacked) - flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una); + flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una, + &sack_rtt_us); if (TCP_ECN_rcv_ecn_echo(tp, tcp_hdr(skb))) flag |= FLAG_ECE; @@ -3684,43 +3440,52 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag) sk->sk_err_soft = 0; icsk->icsk_probes_out = 0; tp->rcv_tstamp = tcp_time_stamp; - prior_packets = tp->packets_out; if (!prior_packets) goto no_queue; /* See if we can take anything off of the retransmit queue. */ - flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una); + acked = tp->packets_out; + flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una, + sack_rtt_us); + acked -= tp->packets_out; - 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; + /* Advance cwnd if state allows */ + if (tcp_may_raise_cwnd(sk, flag)) + tcp_cong_avoid(sk, ack, acked); 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); - 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); + is_dupack = !(flag & (FLAG_SND_UNA_ADVANCED | FLAG_NOT_DUP)); + tcp_fastretrans_alert(sk, acked, prior_unsacked, + is_dupack, flag); } + if (tp->tlp_high_seq) + tcp_process_tlp_ack(sk, ack, flag); - if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) - dst_confirm(__sk_dst_get(sk)); + if ((flag & FLAG_FORWARD_PROGRESS) || !(flag & FLAG_NOT_DUP)) { + struct dst_entry *dst = __sk_dst_get(sk); + if (dst) + dst_confirm(dst); + } + if (icsk->icsk_pending == ICSK_TIME_RETRANS) + tcp_schedule_loss_probe(sk); + tcp_update_pacing_rate(sk); return 1; no_queue: + /* If data was DSACKed, see if we can undo a cwnd reduction. */ + if (flag & FLAG_DSACKING_ACK) + tcp_fastretrans_alert(sk, acked, prior_unsacked, + is_dupack, flag); /* If this ack opens up a zero window, clear backoff. It was * being used to time the probes, and is probably far higher than * it needs to be for normal retransmission. */ if (tcp_send_head(sk)) tcp_ack_probe(sk); + + if (tp->tlp_high_seq) + tcp_process_tlp_ack(sk, ack, flag); return 1; invalid_ack: @@ -3728,10 +3493,14 @@ invalid_ack: return -1; old_ack: + /* If data was SACKed, tag it and see if we should send more data. + * If data was DSACKed, see if we can undo a cwnd reduction. + */ if (TCP_SKB_CB(skb)->sacked) { - tcp_sacktag_write_queue(sk, skb, prior_snd_una); - if (icsk->icsk_ca_state == TCP_CA_Open) - tcp_try_keep_open(sk); + flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una, + &sack_rtt_us); + tcp_fastretrans_alert(sk, acked, prior_unsacked, + is_dupack, flag); } SOCK_DEBUG(sk, "Ack %u before %u:%u\n", ack, tp->snd_una, tp->snd_nxt); @@ -3742,14 +3511,15 @@ old_ack: * 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, - u8 **hvpp, int estab) +void tcp_parse_options(const struct sk_buff *skb, + struct tcp_options_received *opt_rx, int estab, + struct tcp_fastopen_cookie *foc) { - unsigned char *ptr; - struct tcphdr *th = tcp_hdr(skb); + const unsigned char *ptr; + const struct tcphdr *th = tcp_hdr(skb); int length = (th->doff * 4) - sizeof(struct tcphdr); - ptr = (unsigned char *)(th + 1); + ptr = (const unsigned char *)(th + 1); opt_rx->saw_tstamp = 0; while (length > 0) { @@ -3786,10 +3556,9 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, __u8 snd_wscale = *(__u8 *)ptr; opt_rx->wscale_ok = 1; if (snd_wscale > 14) { - if (net_ratelimit()) - printk(KERN_INFO "tcp_parse_options: Illegal window " - "scaling value %d >14 received.\n", - snd_wscale); + net_info_ratelimited("%s: Illegal window scaling value %d >14 received\n", + __func__, + snd_wscale); snd_wscale = 14; } opt_rx->snd_wscale = snd_wscale; @@ -3807,7 +3576,7 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, case TCPOPT_SACK_PERM: if (opsize == TCPOLEN_SACK_PERM && th->syn && !estab && sysctl_tcp_sack) { - opt_rx->sack_ok = 1; + opt_rx->sack_ok = TCP_SACK_SEEN; tcp_sack_reset(opt_rx); } break; @@ -3827,32 +3596,24 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, */ break; #endif - case TCPOPT_COOKIE: - /* This option is variable length. + case TCPOPT_EXP: + /* Fast Open option shares code 254 using a + * 16 bits magic number. It's valid only in + * SYN or SYN-ACK with an even size. */ - switch (opsize) { - case TCPOLEN_COOKIE_BASE: - /* not yet implemented */ - break; - case TCPOLEN_COOKIE_PAIR: - /* not yet implemented */ - break; - case TCPOLEN_COOKIE_MIN+0: - case TCPOLEN_COOKIE_MIN+2: - case TCPOLEN_COOKIE_MIN+4: - case TCPOLEN_COOKIE_MIN+6: - case TCPOLEN_COOKIE_MAX: - /* 16-bit multiple */ - opt_rx->cookie_plus = opsize; - *hvpp = ptr; - break; - default: - /* ignore option */ + if (opsize < TCPOLEN_EXP_FASTOPEN_BASE || + get_unaligned_be16(ptr) != TCPOPT_FASTOPEN_MAGIC || + foc == NULL || !th->syn || (opsize & 1)) break; - } + foc->len = opsize - TCPOLEN_EXP_FASTOPEN_BASE; + if (foc->len >= TCP_FASTOPEN_COOKIE_MIN && + foc->len <= TCP_FASTOPEN_COOKIE_MAX) + memcpy(foc->val, ptr + 2, foc->len); + else if (foc->len != 0) + foc->len = -1; break; - } + } ptr += opsize-2; length -= opsize; } @@ -3860,9 +3621,9 @@ void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, } EXPORT_SYMBOL(tcp_parse_options); -static int tcp_parse_aligned_timestamp(struct tcp_sock *tp, struct tcphdr *th) +static bool tcp_parse_aligned_timestamp(struct tcp_sock *tp, const struct tcphdr *th) { - __be32 *ptr = (__be32 *)(th + 1); + const __be32 *ptr = (const __be32 *)(th + 1); if (*ptr == htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP)) { @@ -3870,41 +3631,48 @@ static int tcp_parse_aligned_timestamp(struct tcp_sock *tp, struct tcphdr *th) ++ptr; tp->rx_opt.rcv_tsval = ntohl(*ptr); ++ptr; - tp->rx_opt.rcv_tsecr = ntohl(*ptr); - return 1; + if (*ptr) + tp->rx_opt.rcv_tsecr = ntohl(*ptr) - tp->tsoffset; + else + tp->rx_opt.rcv_tsecr = 0; + return true; } - return 0; + return false; } /* Fast parse options. This hopes to only see timestamps. * If it is wrong it falls back on tcp_parse_options(). */ -static int tcp_fast_parse_options(struct sk_buff *skb, struct tcphdr *th, - struct tcp_sock *tp, u8 **hvpp) +static bool tcp_fast_parse_options(const struct sk_buff *skb, + const struct tcphdr *th, struct tcp_sock *tp) { /* In the spirit of fast parsing, compare doff directly to constant * values. Because equality is used, short doff can be ignored here. */ if (th->doff == (sizeof(*th) / 4)) { tp->rx_opt.saw_tstamp = 0; - return 0; + return false; } else if (tp->rx_opt.tstamp_ok && th->doff == ((sizeof(*th) + TCPOLEN_TSTAMP_ALIGNED) / 4)) { if (tcp_parse_aligned_timestamp(tp, th)) - return 1; + return true; } - tcp_parse_options(skb, &tp->rx_opt, hvpp, 1); - return 1; + + tcp_parse_options(skb, &tp->rx_opt, 1, NULL); + if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) + tp->rx_opt.rcv_tsecr -= tp->tsoffset; + + return true; } #ifdef CONFIG_TCP_MD5SIG /* * Parse MD5 Signature option */ -u8 *tcp_parse_md5sig_option(struct tcphdr *th) +const u8 *tcp_parse_md5sig_option(const struct tcphdr *th) { - int length = (th->doff << 2) - sizeof (*th); - u8 *ptr = (u8*)(th + 1); + int length = (th->doff << 2) - sizeof(*th); + const u8 *ptr = (const u8 *)(th + 1); /* If the TCP option is too short, we can short cut */ if (length < TCPOLEN_MD5SIG) @@ -3914,7 +3682,7 @@ u8 *tcp_parse_md5sig_option(struct tcphdr *th) int opcode = *ptr++; int opsize; - switch(opcode) { + switch (opcode) { case TCPOPT_EOL: return NULL; case TCPOPT_NOP: @@ -3935,27 +3703,6 @@ u8 *tcp_parse_md5sig_option(struct tcphdr *th) EXPORT_SYMBOL(tcp_parse_md5sig_option); #endif -static inline void tcp_store_ts_recent(struct tcp_sock *tp) -{ - tp->rx_opt.ts_recent = tp->rx_opt.rcv_tsval; - tp->rx_opt.ts_recent_stamp = get_seconds(); -} - -static inline void tcp_replace_ts_recent(struct tcp_sock *tp, u32 seq) -{ - if (tp->rx_opt.saw_tstamp && !after(seq, tp->rcv_wup)) { - /* PAWS bug workaround wrt. ACK frames, the PAWS discard - * extra check below makes sure this can only happen - * for pure ACK frames. -DaveM - * - * Not only, also it occurs for expired timestamps. - */ - - if (tcp_paws_check(&tp->rx_opt, 0)) - tcp_store_ts_recent(tp); - } -} - /* Sorry, PAWS as specified is broken wrt. pure-ACKs -DaveM * * It is not fatal. If this ACK does _not_ change critical state (seqs, window) @@ -3981,8 +3728,8 @@ static inline void tcp_replace_ts_recent(struct tcp_sock *tp, u32 seq) static int tcp_disordered_ack(const struct sock *sk, const struct sk_buff *skb) { - struct tcp_sock *tp = tcp_sk(sk); - struct tcphdr *th = tcp_hdr(skb); + const struct tcp_sock *tp = tcp_sk(sk); + const struct tcphdr *th = tcp_hdr(skb); u32 seq = TCP_SKB_CB(skb)->seq; u32 ack = TCP_SKB_CB(skb)->ack_seq; @@ -3999,7 +3746,7 @@ 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, +static inline bool tcp_paws_discard(const struct sock *sk, const struct sk_buff *skb) { const struct tcp_sock *tp = tcp_sk(sk); @@ -4021,14 +3768,14 @@ static inline int tcp_paws_discard(const struct sock *sk, * (borrowed from freebsd) */ -static inline int tcp_sequence(struct tcp_sock *tp, u32 seq, u32 end_seq) +static inline bool tcp_sequence(const struct tcp_sock *tp, u32 seq, u32 end_seq) { return !before(end_seq, tp->rcv_wup) && !after(seq, tp->rcv_nxt + tcp_receive_window(tp)); } /* When we get a reset we do this. */ -static void tcp_reset(struct sock *sk) +void tcp_reset(struct sock *sk) { /* We want the right error as BSD sees it (and indeed as we do). */ switch (sk->sk_state) { @@ -4066,9 +3813,10 @@ static void tcp_reset(struct sock *sk) * * If we are in FINWAIT-2, a received FIN moves us to TIME-WAIT. */ -static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) +static void tcp_fin(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); + const struct dst_entry *dst; inet_csk_schedule_ack(sk); @@ -4080,7 +3828,9 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) case TCP_ESTABLISHED: /* Move to CLOSE_WAIT */ tcp_set_state(sk, TCP_CLOSE_WAIT); - inet_csk(sk)->icsk_ack.pingpong = 1; + dst = __sk_dst_get(sk); + if (!dst || !dst_metric(dst, RTAX_QUICKACK)) + inet_csk(sk)->icsk_ack.pingpong = 1; break; case TCP_CLOSE_WAIT: @@ -4110,7 +3860,7 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) /* 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", + pr_err("%s: Impossible, sk->sk_state=%d\n", __func__, sk->sk_state); break; } @@ -4135,7 +3885,7 @@ static void tcp_fin(struct sk_buff *skb, struct sock *sk, struct tcphdr *th) } } -static inline int tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, +static inline bool 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)) { @@ -4143,9 +3893,9 @@ static inline int tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, sp->start_seq = seq; if (after(end_seq, sp->end_seq)) sp->end_seq = end_seq; - return 1; + return true; } - return 0; + return false; } static void tcp_dsack_set(struct sock *sk, u32 seq, u32 end_seq) @@ -4178,7 +3928,7 @@ static void tcp_dsack_extend(struct sock *sk, u32 seq, u32 end_seq) tcp_sack_extend(tp->duplicate_sack, seq, end_seq); } -static void tcp_send_dupack(struct sock *sk, struct sk_buff *skb) +static void tcp_send_dupack(struct sock *sk, const struct sk_buff *skb) { struct tcp_sock *tp = tcp_sk(sk); @@ -4292,7 +4042,7 @@ static void tcp_sack_remove(struct tcp_sock *tp) WARN_ON(before(tp->rcv_nxt, sp->end_seq)); /* Zap this SACK, by moving forward any other SACKS. */ - for (i=this_sack+1; i < num_sacks; i++) + for (i = this_sack+1; i < num_sacks; i++) tp->selective_acks[i-1] = tp->selective_acks[i]; num_sacks--; continue; @@ -4337,37 +4087,261 @@ static void tcp_ofo_queue(struct sock *sk) __skb_queue_tail(&sk->sk_receive_queue, skb); tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; if (tcp_hdr(skb)->fin) - tcp_fin(skb, sk, tcp_hdr(skb)); + tcp_fin(sk); } } -static int tcp_prune_ofo_queue(struct sock *sk); +static bool tcp_prune_ofo_queue(struct sock *sk); static int tcp_prune_queue(struct sock *sk); -static inline int tcp_try_rmem_schedule(struct sock *sk, unsigned int size) +static int tcp_try_rmem_schedule(struct sock *sk, struct sk_buff *skb, + unsigned int size) { if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf || - !sk_rmem_schedule(sk, size)) { + !sk_rmem_schedule(sk, skb, size)) { if (tcp_prune_queue(sk) < 0) return -1; - if (!sk_rmem_schedule(sk, size)) { + if (!sk_rmem_schedule(sk, skb, size)) { if (!tcp_prune_ofo_queue(sk)) return -1; - if (!sk_rmem_schedule(sk, size)) + if (!sk_rmem_schedule(sk, skb, size)) return -1; } } return 0; } +/** + * tcp_try_coalesce - try to merge skb to prior one + * @sk: socket + * @to: prior buffer + * @from: buffer to add in queue + * @fragstolen: pointer to boolean + * + * Before queueing skb @from after @to, try to merge them + * to reduce overall memory use and queue lengths, if cost is small. + * Packets in ofo or receive queues can stay a long time. + * Better try to coalesce them right now to avoid future collapses. + * Returns true if caller should free @from instead of queueing it + */ +static bool tcp_try_coalesce(struct sock *sk, + struct sk_buff *to, + struct sk_buff *from, + bool *fragstolen) +{ + int delta; + + *fragstolen = false; + + if (tcp_hdr(from)->fin) + return false; + + /* Its possible this segment overlaps with prior segment in queue */ + if (TCP_SKB_CB(from)->seq != TCP_SKB_CB(to)->end_seq) + return false; + + if (!skb_try_coalesce(to, from, fragstolen, &delta)) + return false; + + atomic_add(delta, &sk->sk_rmem_alloc); + sk_mem_charge(sk, delta); + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOALESCE); + TCP_SKB_CB(to)->end_seq = TCP_SKB_CB(from)->end_seq; + TCP_SKB_CB(to)->ack_seq = TCP_SKB_CB(from)->ack_seq; + return true; +} + +static void tcp_data_queue_ofo(struct sock *sk, struct sk_buff *skb) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct sk_buff *skb1; + u32 seq, end_seq; + + TCP_ECN_check_ce(tp, skb); + + if (unlikely(tcp_try_rmem_schedule(sk, skb, skb->truesize))) { + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPOFODROP); + __kfree_skb(skb); + return; + } + + /* Disable header prediction. */ + tp->pred_flags = 0; + inet_csk_schedule_ack(sk); + + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPOFOQUEUE); + 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); + + skb1 = skb_peek_tail(&tp->out_of_order_queue); + if (!skb1) { + /* Initial out of order segment, build 1 SACK. */ + if (tcp_is_sack(tp)) { + tp->rx_opt.num_sacks = 1; + tp->selective_acks[0].start_seq = TCP_SKB_CB(skb)->seq; + tp->selective_acks[0].end_seq = + TCP_SKB_CB(skb)->end_seq; + } + __skb_queue_head(&tp->out_of_order_queue, skb); + goto end; + } + + seq = TCP_SKB_CB(skb)->seq; + end_seq = TCP_SKB_CB(skb)->end_seq; + + if (seq == TCP_SKB_CB(skb1)->end_seq) { + bool fragstolen; + + if (!tcp_try_coalesce(sk, skb1, skb, &fragstolen)) { + __skb_queue_after(&tp->out_of_order_queue, skb1, skb); + } else { + tcp_grow_window(sk, skb); + kfree_skb_partial(skb, fragstolen); + skb = NULL; + } + + if (!tp->rx_opt.num_sacks || + tp->selective_acks[0].end_seq != seq) + goto add_sack; + + /* Common case: data arrive in order after hole. */ + tp->selective_acks[0].end_seq = end_seq; + goto end; + } + + /* Find place to insert this segment. */ + while (1) { + if (!after(TCP_SKB_CB(skb1)->seq, seq)) + break; + if (skb_queue_is_first(&tp->out_of_order_queue, skb1)) { + skb1 = NULL; + break; + } + skb1 = skb_queue_prev(&tp->out_of_order_queue, skb1); + } + + /* Do skb overlap to previous one? */ + if (skb1 && before(seq, TCP_SKB_CB(skb1)->end_seq)) { + if (!after(end_seq, TCP_SKB_CB(skb1)->end_seq)) { + /* All the bits are present. Drop. */ + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPOFOMERGE); + __kfree_skb(skb); + skb = NULL; + tcp_dsack_set(sk, seq, end_seq); + goto add_sack; + } + if (after(seq, TCP_SKB_CB(skb1)->seq)) { + /* Partial overlap. */ + tcp_dsack_set(sk, seq, + TCP_SKB_CB(skb1)->end_seq); + } else { + if (skb_queue_is_first(&tp->out_of_order_queue, + skb1)) + skb1 = NULL; + else + skb1 = skb_queue_prev( + &tp->out_of_order_queue, + skb1); + } + } + if (!skb1) + __skb_queue_head(&tp->out_of_order_queue, skb); + else + __skb_queue_after(&tp->out_of_order_queue, skb1, skb); + + /* And clean segments covered by new one as whole. */ + while (!skb_queue_is_last(&tp->out_of_order_queue, skb)) { + skb1 = skb_queue_next(&tp->out_of_order_queue, skb); + + if (!after(end_seq, TCP_SKB_CB(skb1)->seq)) + break; + if (before(end_seq, TCP_SKB_CB(skb1)->end_seq)) { + tcp_dsack_extend(sk, TCP_SKB_CB(skb1)->seq, + end_seq); + break; + } + __skb_unlink(skb1, &tp->out_of_order_queue); + tcp_dsack_extend(sk, TCP_SKB_CB(skb1)->seq, + TCP_SKB_CB(skb1)->end_seq); + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPOFOMERGE); + __kfree_skb(skb1); + } + +add_sack: + if (tcp_is_sack(tp)) + tcp_sack_new_ofo_skb(sk, seq, end_seq); +end: + if (skb) { + tcp_grow_window(sk, skb); + skb_set_owner_r(skb, sk); + } +} + +static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, int hdrlen, + bool *fragstolen) +{ + int eaten; + struct sk_buff *tail = skb_peek_tail(&sk->sk_receive_queue); + + __skb_pull(skb, hdrlen); + eaten = (tail && + tcp_try_coalesce(sk, tail, skb, fragstolen)) ? 1 : 0; + tcp_sk(sk)->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + if (!eaten) { + __skb_queue_tail(&sk->sk_receive_queue, skb); + skb_set_owner_r(skb, sk); + } + return eaten; +} + +int tcp_send_rcvq(struct sock *sk, struct msghdr *msg, size_t size) +{ + struct sk_buff *skb = NULL; + struct tcphdr *th; + bool fragstolen; + + if (size == 0) + return 0; + + skb = alloc_skb(size + sizeof(*th), sk->sk_allocation); + if (!skb) + goto err; + + if (tcp_try_rmem_schedule(sk, skb, size + sizeof(*th))) + goto err_free; + + th = (struct tcphdr *)skb_put(skb, sizeof(*th)); + skb_reset_transport_header(skb); + memset(th, 0, sizeof(*th)); + + if (memcpy_fromiovec(skb_put(skb, size), msg->msg_iov, size)) + goto err_free; + + TCP_SKB_CB(skb)->seq = tcp_sk(sk)->rcv_nxt; + TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + size; + TCP_SKB_CB(skb)->ack_seq = tcp_sk(sk)->snd_una - 1; + + if (tcp_queue_rcv(sk, skb, sizeof(*th), &fragstolen)) { + WARN_ON_ONCE(fragstolen); /* should not happen */ + __kfree_skb(skb); + } + return size; + +err_free: + kfree_skb(skb); +err: + return -ENOMEM; +} + static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) { - struct tcphdr *th = tcp_hdr(skb); + const struct tcphdr *th = tcp_hdr(skb); struct tcp_sock *tp = tcp_sk(sk); int eaten = -1; + bool fragstolen = false; if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) goto drop; @@ -4400,7 +4374,7 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) if (!skb_copy_datagram_iovec(skb, 0, tp->ucopy.iov, chunk)) { tp->ucopy.len -= chunk; tp->copied_seq += chunk; - eaten = (chunk == skb->len && !th->fin); + eaten = (chunk == skb->len); tcp_rcv_space_adjust(sk); } local_bh_disable(); @@ -4409,17 +4383,16 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) if (eaten <= 0) { queue_and_out: if (eaten < 0 && - tcp_try_rmem_schedule(sk, skb->truesize)) + tcp_try_rmem_schedule(sk, skb, skb->truesize)) goto drop; - skb_set_owner_r(skb, sk); - __skb_queue_tail(&sk->sk_receive_queue, skb); + eaten = tcp_queue_rcv(sk, skb, 0, &fragstolen); } tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; if (skb->len) tcp_event_data_recv(sk, skb); if (th->fin) - tcp_fin(skb, sk, th); + tcp_fin(sk); if (!skb_queue_empty(&tp->out_of_order_queue)) { tcp_ofo_queue(sk); @@ -4437,9 +4410,9 @@ queue_and_out: tcp_fast_path_check(sk); if (eaten > 0) - __kfree_skb(skb); - else if (!sock_flag(sk, SOCK_DEAD)) - sk->sk_data_ready(sk, 0); + kfree_skb_partial(skb, fragstolen); + if (!sock_flag(sk, SOCK_DEAD)) + sk->sk_data_ready(sk); return; } @@ -4478,105 +4451,7 @@ drop: goto queue_and_out; } - TCP_ECN_check_ce(tp, skb); - - if (tcp_try_rmem_schedule(sk, skb->truesize)) - goto drop; - - /* Disable header prediction. */ - tp->pred_flags = 0; - inet_csk_schedule_ack(sk); - - 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); - - skb_set_owner_r(skb, sk); - - if (!skb_peek(&tp->out_of_order_queue)) { - /* Initial out of order segment, build 1 SACK. */ - if (tcp_is_sack(tp)) { - tp->rx_opt.num_sacks = 1; - tp->selective_acks[0].start_seq = TCP_SKB_CB(skb)->seq; - tp->selective_acks[0].end_seq = - TCP_SKB_CB(skb)->end_seq; - } - __skb_queue_head(&tp->out_of_order_queue, skb); - } else { - struct sk_buff *skb1 = skb_peek_tail(&tp->out_of_order_queue); - u32 seq = TCP_SKB_CB(skb)->seq; - u32 end_seq = TCP_SKB_CB(skb)->end_seq; - - if (seq == TCP_SKB_CB(skb1)->end_seq) { - __skb_queue_after(&tp->out_of_order_queue, skb1, skb); - - if (!tp->rx_opt.num_sacks || - tp->selective_acks[0].end_seq != seq) - goto add_sack; - - /* Common case: data arrive in order after hole. */ - tp->selective_acks[0].end_seq = end_seq; - return; - } - - /* Find place to insert this segment. */ - while (1) { - if (!after(TCP_SKB_CB(skb1)->seq, seq)) - break; - if (skb_queue_is_first(&tp->out_of_order_queue, skb1)) { - skb1 = NULL; - break; - } - skb1 = skb_queue_prev(&tp->out_of_order_queue, skb1); - } - - /* Do skb overlap to previous one? */ - if (skb1 && before(seq, TCP_SKB_CB(skb1)->end_seq)) { - if (!after(end_seq, TCP_SKB_CB(skb1)->end_seq)) { - /* All the bits are present. Drop. */ - __kfree_skb(skb); - tcp_dsack_set(sk, seq, end_seq); - goto add_sack; - } - if (after(seq, TCP_SKB_CB(skb1)->seq)) { - /* Partial overlap. */ - tcp_dsack_set(sk, seq, - TCP_SKB_CB(skb1)->end_seq); - } else { - if (skb_queue_is_first(&tp->out_of_order_queue, - skb1)) - skb1 = NULL; - else - skb1 = skb_queue_prev( - &tp->out_of_order_queue, - skb1); - } - } - if (!skb1) - __skb_queue_head(&tp->out_of_order_queue, skb); - else - __skb_queue_after(&tp->out_of_order_queue, skb1, skb); - - /* And clean segments covered by new one as whole. */ - while (!skb_queue_is_last(&tp->out_of_order_queue, skb)) { - skb1 = skb_queue_next(&tp->out_of_order_queue, skb); - - if (!after(end_seq, TCP_SKB_CB(skb1)->seq)) - break; - if (before(end_seq, TCP_SKB_CB(skb1)->end_seq)) { - tcp_dsack_extend(sk, TCP_SKB_CB(skb1)->seq, - end_seq); - break; - } - __skb_unlink(skb1, &tp->out_of_order_queue); - tcp_dsack_extend(sk, TCP_SKB_CB(skb1)->seq, - TCP_SKB_CB(skb1)->end_seq); - __kfree_skb(skb1); - } - -add_sack: - if (tcp_is_sack(tp)) - tcp_sack_new_ofo_skb(sk, seq, end_seq); - } + tcp_data_queue_ofo(sk, skb); } static struct sk_buff *tcp_collapse_one(struct sock *sk, struct sk_buff *skb, @@ -4755,10 +4630,10 @@ static void tcp_collapse_ofo_queue(struct sock *sk) * Purge the out-of-order queue. * Return true if queue was pruned. */ -static int tcp_prune_ofo_queue(struct sock *sk) +static bool tcp_prune_ofo_queue(struct sock *sk) { struct tcp_sock *tp = tcp_sk(sk); - int res = 0; + bool res = false; if (!skb_queue_empty(&tp->out_of_order_queue)) { NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_OFOPRUNED); @@ -4772,7 +4647,7 @@ static int tcp_prune_ofo_queue(struct sock *sk) if (tp->rx_opt.sack_ok) tcp_sack_reset(&tp->rx_opt); sk_mem_reclaim(sk); - res = 1; + res = true; } return res; } @@ -4794,7 +4669,7 @@ static int tcp_prune_queue(struct sock *sk) if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf) tcp_clamp_window(sk); - else if (tcp_memory_pressure) + else if (sk_under_memory_pressure(sk)) tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U * tp->advmss); tcp_collapse_ofo_queue(sk); @@ -4827,51 +4702,29 @@ 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. - */ -void tcp_cwnd_application_limited(struct sock *sk) +static bool tcp_should_expand_sndbuf(const struct sock *sk) { - struct tcp_sock *tp = tcp_sk(sk); - - if (inet_csk(sk)->icsk_ca_state == TCP_CA_Open && - sk->sk_socket && !test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - /* Limited by application or receiver window. */ - u32 init_win = tcp_init_cwnd(tp, __sk_dst_get(sk)); - u32 win_used = max(tp->snd_cwnd_used, init_win); - if (win_used < tp->snd_cwnd) { - tp->snd_ssthresh = tcp_current_ssthresh(sk); - tp->snd_cwnd = (tp->snd_cwnd + win_used) >> 1; - } - tp->snd_cwnd_used = 0; - } - tp->snd_cwnd_stamp = tcp_time_stamp; -} - -static int tcp_should_expand_sndbuf(struct sock *sk) -{ - struct tcp_sock *tp = tcp_sk(sk); + const struct tcp_sock *tp = tcp_sk(sk); /* If the user specified a specific send buffer setting, do * not modify it. */ if (sk->sk_userlocks & SOCK_SNDBUF_LOCK) - return 0; + return false; /* If we are under global TCP memory pressure, do not expand. */ - if (tcp_memory_pressure) - return 0; + if (sk_under_memory_pressure(sk)) + return false; /* If we are under soft global TCP memory pressure, do not expand. */ - if (atomic_long_read(&tcp_memory_allocated) >= sysctl_tcp_mem[0]) - return 0; + if (sk_memory_allocated(sk) >= sk_prot_mem_limits(sk, 0)) + return false; /* If we filled the congestion window, do not expand. */ if (tp->packets_out >= tp->snd_cwnd) - return 0; + return false; - return 1; + return true; } /* When incoming ACK allowed to free some skb from write_queue, @@ -4885,13 +4738,7 @@ static void tcp_new_space(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); if (tcp_should_expand_sndbuf(sk)) { - int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) + - MAX_TCP_HEADER + 16 + sizeof(struct sk_buff); - int demanded = max_t(unsigned int, tp->snd_cwnd, - tp->reordering + 1); - sndmem *= 2 * demanded; - if (sndmem > sk->sk_sndbuf) - sk->sk_sndbuf = min(sndmem, sysctl_tcp_wmem[2]); + tcp_sndbuf_expand(sk); tp->snd_cwnd_stamp = tcp_time_stamp; } @@ -4958,7 +4805,7 @@ 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, const struct tcphdr *th) { struct tcp_sock *tp = tcp_sk(sk); u32 ptr = ntohs(th->urg_ptr); @@ -5024,7 +4871,7 @@ static void tcp_check_urg(struct sock *sk, struct tcphdr *th) } /* This is the 'fast' part of urgent handling. */ -static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th) +static void tcp_urg(struct sock *sk, struct sk_buff *skb, const struct tcphdr *th) { struct tcp_sock *tp = tcp_sk(sk); @@ -5044,7 +4891,7 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th) BUG(); tp->urg_data = TCP_URG_VALID | tmp; if (!sock_flag(sk, SOCK_DEAD)) - sk->sk_data_ready(sk, 0); + sk->sk_data_ready(sk); } } } @@ -5087,7 +4934,7 @@ static __sum16 __tcp_checksum_complete_user(struct sock *sk, return result; } -static inline int tcp_checksum_complete_user(struct sock *sk, +static inline bool tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb) { return !skb_csum_unnecessary(skb) && @@ -5095,19 +4942,19 @@ static inline int tcp_checksum_complete_user(struct sock *sk, } #ifdef CONFIG_NET_DMA -static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, +static bool 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; int dma_cookie; - int copied_early = 0; + bool copied_early = false; if (tp->ucopy.wakeup) - return 0; + return false; if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list) - tp->ucopy.dma_chan = dma_find_channel(DMA_MEMCPY); + tp->ucopy.dma_chan = net_dma_find_channel(); if (tp->ucopy.dma_chan && skb_csum_unnecessary(skb)) { @@ -5120,7 +4967,7 @@ static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, goto out; tp->ucopy.dma_cookie = dma_cookie; - copied_early = 1; + copied_early = true; tp->ucopy.len -= chunk; tp->copied_seq += chunk; @@ -5130,11 +4977,11 @@ static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, (tcp_flag_word(tcp_hdr(skb)) & TCP_FLAG_PSH) || (atomic_read(&sk->sk_rmem_alloc) > (sk->sk_rcvbuf >> 1))) { tp->ucopy.wakeup = 1; - sk->sk_data_ready(sk, 0); + sk->sk_data_ready(sk); } } else if (chunk > 0) { tp->ucopy.wakeup = 1; - sk->sk_data_ready(sk, 0); + sk->sk_data_ready(sk); } out: return copied_early; @@ -5144,15 +4991,13 @@ out: /* Does PAWS and seqno based validation of an incoming segment, flags will * play significant role here. */ -static int tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, - struct tcphdr *th, int syn_inerr) +static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, + const struct tcphdr *th, int syn_inerr) { - u8 *hash_location; struct tcp_sock *tp = tcp_sk(sk); /* RFC1323: H1. Apply PAWS check first. */ - if (tcp_fast_parse_options(skb, th, tp, &hash_location) && - tp->rx_opt.saw_tstamp && + if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp && tcp_paws_discard(sk, skb)) { if (!th->rst) { NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED); @@ -5170,38 +5015,48 @@ static int tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, * an acknowledgment should be sent in reply (unless the RST * bit is set, if so drop the segment and return)". */ - if (!th->rst) + if (!th->rst) { + if (th->syn) + goto syn_challenge; tcp_send_dupack(sk, skb); + } goto discard; } /* Step 2: check RST bit */ if (th->rst) { - tcp_reset(sk); + /* RFC 5961 3.2 : + * If sequence number exactly matches RCV.NXT, then + * RESET the connection + * else + * Send a challenge ACK + */ + if (TCP_SKB_CB(skb)->seq == tp->rcv_nxt) + tcp_reset(sk); + else + tcp_send_challenge_ack(sk); goto discard; } - /* ts_recent update must be made after we are sure that the packet - * is in window. - */ - tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq); - /* step 3: check security and precedence [ignored] */ - /* step 4: Check for a SYN in window. */ - if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) { + /* step 4: Check for a SYN + * RFC 5691 4.2 : Send a challenge ack + */ + if (th->syn) { +syn_challenge: if (syn_inerr) TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS); - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN); - tcp_reset(sk); - return -1; + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPSYNCHALLENGE); + tcp_send_challenge_ack(sk); + goto discard; } - return 1; + return true; discard: __kfree_skb(skb); - return 0; + return false; } /* @@ -5227,12 +5082,13 @@ discard: * the rest is checked inline. Fast processing is turned on in * tcp_data_queue when everything is OK. */ -int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, - struct tcphdr *th, unsigned len) +void tcp_rcv_established(struct sock *sk, struct sk_buff *skb, + const struct tcphdr *th, unsigned int len) { struct tcp_sock *tp = tcp_sk(sk); - int res; + if (unlikely(sk->sk_rx_dst == NULL)) + inet_csk(sk)->icsk_af_ops->sk_rx_dst_set(sk, skb); /* * Header prediction. * The code loosely follows the one in the famous @@ -5304,7 +5160,7 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, tcp_ack(sk, skb, 0); __kfree_skb(skb); tcp_data_snd_check(sk); - return 0; + return; } else { /* Header too small */ TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS); goto discard; @@ -5312,11 +5168,14 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, } else { int eaten = 0; int copied_early = 0; + bool fragstolen = false; if (tp->copied_seq == tp->rcv_nxt && len - tcp_header_len <= tp->ucopy.len) { #ifdef CONFIG_NET_DMA - if (tcp_dma_try_early_copy(sk, skb, tcp_header_len)) { + if (tp->ucopy.task == current && + sock_owned_by_user(sk) && + tcp_dma_try_early_copy(sk, skb, tcp_header_len)) { copied_early = 1; eaten = 1; } @@ -5352,6 +5211,9 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, if (tcp_checksum_complete_user(sk, skb)) goto csum_error; + if ((int)skb->truesize > sk->sk_forward_alloc) + goto step5; + /* Predicted packet is in window by definition. * seq == rcv_nxt and rcv_wup <= rcv_nxt. * Hence, check seq<=rcv_wup reduces to: @@ -5363,16 +5225,11 @@ int tcp_rcv_established(struct sock *sk, struct sk_buff *skb, tcp_rcv_rtt_measure_ts(sk, skb); - if ((int)skb->truesize > sk->sk_forward_alloc) - goto step5; - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPHITS); /* Bulk data transfer: receiver */ - __skb_pull(skb, tcp_header_len); - __skb_queue_tail(&sk->sk_receive_queue, skb); - skb_set_owner_r(skb, sk); - tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + eaten = tcp_queue_rcv(sk, skb, tcp_header_len, + &fragstolen); } tcp_event_data_recv(sk, skb); @@ -5394,10 +5251,9 @@ no_ack: else #endif if (eaten) - __kfree_skb(skb); - else - sk->sk_data_ready(sk, 0); - return 0; + kfree_skb_partial(skb, fragstolen); + sk->sk_data_ready(sk); + return; } } @@ -5405,16 +5261,18 @@ slow_path: if (len < (th->doff << 2) || tcp_checksum_complete_user(sk, skb)) goto csum_error; + if (!th->ack && !th->rst) + goto discard; + /* * Standard slow path. */ - res = tcp_validate_incoming(sk, skb, th, 1); - if (res <= 0) - return -res; + if (!tcp_validate_incoming(sk, skb, th, 1)) + return; step5: - if (th->ack && tcp_ack(sk, skb, FLAG_SLOWPATH) < 0) + if (tcp_ack(sk, skb, FLAG_SLOWPATH | FLAG_UPDATE_TS_RECENT) < 0) goto discard; tcp_rcv_rtt_measure_ts(sk, skb); @@ -5427,27 +5285,113 @@ step5: tcp_data_snd_check(sk); tcp_ack_snd_check(sk); - return 0; + return; csum_error: + TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_CSUMERRORS); TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS); discard: __kfree_skb(skb); - return 0; } EXPORT_SYMBOL(tcp_rcv_established); +void tcp_finish_connect(struct sock *sk, struct sk_buff *skb) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct inet_connection_sock *icsk = inet_csk(sk); + + tcp_set_state(sk, TCP_ESTABLISHED); + + if (skb != NULL) { + icsk->icsk_af_ops->sk_rx_dst_set(sk, skb); + security_inet_conn_established(sk, skb); + } + + /* Make sure socket is routed, for correct metrics. */ + icsk->icsk_af_ops->rebuild_header(sk); + + tcp_init_metrics(sk); + + tcp_init_congestion_control(sk); + + /* Prevent spurious tcp_cwnd_restart() on first data + * packet. + */ + tp->lsndtime = tcp_time_stamp; + + tcp_init_buffer_space(sk); + + if (sock_flag(sk, SOCK_KEEPOPEN)) + inet_csk_reset_keepalive_timer(sk, keepalive_time_when(tp)); + + if (!tp->rx_opt.snd_wscale) + __tcp_fast_path_on(tp, tp->snd_wnd); + else + tp->pred_flags = 0; + + if (!sock_flag(sk, SOCK_DEAD)) { + sk->sk_state_change(sk); + sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT); + } +} + +static bool tcp_rcv_fastopen_synack(struct sock *sk, struct sk_buff *synack, + struct tcp_fastopen_cookie *cookie) +{ + struct tcp_sock *tp = tcp_sk(sk); + struct sk_buff *data = tp->syn_data ? tcp_write_queue_head(sk) : NULL; + u16 mss = tp->rx_opt.mss_clamp; + bool syn_drop; + + if (mss == tp->rx_opt.user_mss) { + struct tcp_options_received opt; + + /* Get original SYNACK MSS value if user MSS sets mss_clamp */ + tcp_clear_options(&opt); + opt.user_mss = opt.mss_clamp = 0; + tcp_parse_options(synack, &opt, 0, NULL); + mss = opt.mss_clamp; + } + + if (!tp->syn_fastopen) /* Ignore an unsolicited cookie */ + cookie->len = -1; + + /* The SYN-ACK neither has cookie nor acknowledges the data. Presumably + * the remote receives only the retransmitted (regular) SYNs: either + * the original SYN-data or the corresponding SYN-ACK is lost. + */ + syn_drop = (cookie->len <= 0 && data && tp->total_retrans); + + tcp_fastopen_cache_set(sk, mss, cookie, syn_drop); + + if (data) { /* Retransmit unacked data in SYN */ + tcp_for_write_queue_from(data, sk) { + if (data == tcp_send_head(sk) || + __tcp_retransmit_skb(sk, data)) + break; + } + tcp_rearm_rto(sk); + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENACTIVEFAIL); + return true; + } + tp->syn_data_acked = tp->syn_data; + if (tp->syn_data_acked) + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPFASTOPENACTIVE); + return false; +} + static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, - struct tcphdr *th, unsigned len) + const struct tcphdr *th, unsigned int len) { - u8 *hash_location; struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); - struct tcp_cookie_values *cvp = tp->cookie_values; + struct tcp_fastopen_cookie foc = { .len = -1 }; int saved_clamp = tp->rx_opt.mss_clamp; - tcp_parse_options(skb, &tp->rx_opt, &hash_location, 0); + tcp_parse_options(skb, &tp->rx_opt, 0, &foc); + if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr) + tp->rx_opt.rcv_tsecr -= tp->tsoffset; if (th->ack) { /* rfc793: @@ -5457,11 +5401,9 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, * If SEG.ACK =< ISS, or SEG.ACK > SND.NXT, send * a reset (unless the RST bit is set, if so drop * the segment and return)" - * - * We do not send data with SYN, so that RFC-correct - * test reduces to: */ - if (TCP_SKB_CB(skb)->ack_seq != tp->snd_nxt) + if (!after(TCP_SKB_CB(skb)->ack_seq, tp->snd_una) || + after(TCP_SKB_CB(skb)->ack_seq, tp->snd_nxt)) goto reset_and_undo; if (tp->rx_opt.saw_tstamp && tp->rx_opt.rcv_tsecr && @@ -5503,7 +5445,7 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, TCP_ECN_rcv_synack(tp, th); - tp->snd_wl1 = TCP_SKB_CB(skb)->seq; + tcp_init_wl(tp, TCP_SKB_CB(skb)->seq); tcp_ack(sk, skb, FLAG_SLOWPATH); /* Ok.. it's good. Set up sequence numbers and @@ -5516,7 +5458,6 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, * never scaled. */ tp->snd_wnd = ntohs(th->window); - tcp_init_wl(tp, TCP_SKB_CB(skb)->seq); if (!tp->rx_opt.wscale_ok) { tp->rx_opt.snd_wscale = tp->rx_opt.rcv_wscale = 0; @@ -5545,61 +5486,13 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, * is initialized. */ tp->copied_seq = tp->rcv_nxt; - if (cvp != NULL && - cvp->cookie_pair_size > 0 && - tp->rx_opt.cookie_plus > 0) { - int cookie_size = tp->rx_opt.cookie_plus - - TCPOLEN_COOKIE_BASE; - int cookie_pair_size = cookie_size - + cvp->cookie_desired; - - /* A cookie extension option was sent and returned. - * Note that each incoming SYNACK replaces the - * Responder cookie. The initial exchange is most - * fragile, as protection against spoofing relies - * entirely upon the sequence and timestamp (above). - * This replacement strategy allows the correct pair to - * pass through, while any others will be filtered via - * Responder verification later. - */ - if (sizeof(cvp->cookie_pair) >= cookie_pair_size) { - memcpy(&cvp->cookie_pair[cvp->cookie_desired], - hash_location, cookie_size); - cvp->cookie_pair_size = cookie_pair_size; - } - } - smp_mb(); - tcp_set_state(sk, TCP_ESTABLISHED); - security_inet_conn_established(sk, skb); - - /* Make sure socket is routed, for correct metrics. */ - icsk->icsk_af_ops->rebuild_header(sk); - - tcp_init_metrics(sk); - - tcp_init_congestion_control(sk); - - /* Prevent spurious tcp_cwnd_restart() on first data - * packet. - */ - tp->lsndtime = tcp_time_stamp; - - tcp_init_buffer_space(sk); + tcp_finish_connect(sk, skb); - if (sock_flag(sk, SOCK_KEEPOPEN)) - inet_csk_reset_keepalive_timer(sk, keepalive_time_when(tp)); - - if (!tp->rx_opt.snd_wscale) - __tcp_fast_path_on(tp, tp->snd_wnd); - else - tp->pred_flags = 0; - - if (!sock_flag(sk, SOCK_DEAD)) { - sk->sk_state_change(sk); - sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT); - } + if ((tp->syn_fastopen || tp->syn_data) && + tcp_rcv_fastopen_synack(sk, skb, &foc)) + return -1; if (sk->sk_write_pending || icsk->icsk_accept_queue.rskq_defer_accept || @@ -5613,8 +5506,6 @@ static int tcp_rcv_synsent_state_process(struct sock *sk, struct sk_buff *skb, */ inet_csk_schedule_ack(sk); icsk->icsk_ack.lrcvtime = tcp_time_stamp; - icsk->icsk_ack.ato = TCP_ATO_MIN; - tcp_incr_quickack(sk); tcp_enter_quickack_mode(sk); inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK, TCP_DELACK_MAX, TCP_RTO_MAX); @@ -5680,7 +5571,9 @@ discard: tcp_send_synack(sk); #if 0 /* Note, we could accept data and URG from this segment. - * There are no obstacles to make this. + * There are no obstacles to make this (except that we must + * either change tcp_recvmsg() to prevent it from returning data + * before 3WHS completes per RFC793, or employ TCP Fast Open). * * However, if we ignore data in ACKless segments sometimes, * we have no reasons to accept it sometimes. @@ -5716,12 +5609,14 @@ reset_and_undo: */ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, - struct tcphdr *th, unsigned len) + const struct tcphdr *th, unsigned int len) { struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); + struct request_sock *req; int queued = 0; - int res; + bool acceptable; + u32 synack_stamp; tp->rx_opt.saw_tstamp = 0; @@ -5737,6 +5632,8 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, goto discard; if (th->syn) { + if (th->fin) + goto discard; if (icsk->icsk_af_ops->conn_request(sk, skb) < 0) return 1; @@ -5774,124 +5671,167 @@ int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb, return 0; } - res = tcp_validate_incoming(sk, skb, th, 0); - if (res <= 0) - return -res; + req = tp->fastopen_rsk; + if (req != NULL) { + WARN_ON_ONCE(sk->sk_state != TCP_SYN_RECV && + sk->sk_state != TCP_FIN_WAIT1); + + if (tcp_check_req(sk, skb, req, NULL, true) == NULL) + goto discard; + } + + if (!th->ack && !th->rst) + goto discard; + + if (!tcp_validate_incoming(sk, skb, th, 0)) + return 0; /* step 5: check the ACK field */ - if (th->ack) { - int acceptable = tcp_ack(sk, skb, FLAG_SLOWPATH) > 0; - - switch (sk->sk_state) { - case TCP_SYN_RECV: - if (acceptable) { - tp->copied_seq = tp->rcv_nxt; - smp_mb(); - tcp_set_state(sk, TCP_ESTABLISHED); - sk->sk_state_change(sk); - - /* Note, that this wakeup is only for marginal - * crossed SYN case. Passively open sockets - * are not waked up, because sk->sk_sleep == - * NULL and sk->sk_socket == NULL. - */ - 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) << - tp->rx_opt.snd_wscale; - tcp_init_wl(tp, TCP_SKB_CB(skb)->seq); - - /* tcp_ack considers this ACK as duplicate - * and does not calculate rtt. - * Force it here. - */ - tcp_ack_update_rtt(sk, 0, 0); + acceptable = tcp_ack(sk, skb, FLAG_SLOWPATH | + FLAG_UPDATE_TS_RECENT) > 0; - if (tp->rx_opt.tstamp_ok) - tp->advmss -= TCPOLEN_TSTAMP_ALIGNED; + switch (sk->sk_state) { + case TCP_SYN_RECV: + if (!acceptable) + return 1; - /* Make sure socket is routed, for - * correct metrics. - */ - icsk->icsk_af_ops->rebuild_header(sk); + /* Once we leave TCP_SYN_RECV, we no longer need req + * so release it. + */ + if (req) { + synack_stamp = tcp_rsk(req)->snt_synack; + tp->total_retrans = req->num_retrans; + reqsk_fastopen_remove(sk, req, false); + } else { + synack_stamp = tp->lsndtime; + /* Make sure socket is routed, for correct metrics. */ + icsk->icsk_af_ops->rebuild_header(sk); + tcp_init_congestion_control(sk); + + tcp_mtup_init(sk); + tp->copied_seq = tp->rcv_nxt; + tcp_init_buffer_space(sk); + } + smp_mb(); + tcp_set_state(sk, TCP_ESTABLISHED); + sk->sk_state_change(sk); - tcp_init_metrics(sk); + /* Note, that this wakeup is only for marginal crossed SYN case. + * Passively open sockets are not waked up, because + * sk->sk_sleep == NULL and sk->sk_socket == NULL. + */ + if (sk->sk_socket) + sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT); - tcp_init_congestion_control(sk); + tp->snd_una = TCP_SKB_CB(skb)->ack_seq; + tp->snd_wnd = ntohs(th->window) << tp->rx_opt.snd_wscale; + tcp_init_wl(tp, TCP_SKB_CB(skb)->seq); + tcp_synack_rtt_meas(sk, synack_stamp); - /* Prevent spurious tcp_cwnd_restart() on - * first data packet. - */ - tp->lsndtime = tcp_time_stamp; + if (tp->rx_opt.tstamp_ok) + tp->advmss -= TCPOLEN_TSTAMP_ALIGNED; - tcp_mtup_init(sk); - tcp_initialize_rcv_mss(sk); - tcp_init_buffer_space(sk); - tcp_fast_path_on(tp); - } else { + if (req) { + /* Re-arm the timer because data may have been sent out. + * This is similar to the regular data transmission case + * when new data has just been ack'ed. + * + * (TFO) - we could try to be more aggressive and + * retransmitting any data sooner based on when they + * are sent out. + */ + tcp_rearm_rto(sk); + } else + tcp_init_metrics(sk); + + tcp_update_pacing_rate(sk); + + /* Prevent spurious tcp_cwnd_restart() on first data packet */ + tp->lsndtime = tcp_time_stamp; + + tcp_initialize_rcv_mss(sk); + tcp_fast_path_on(tp); + break; + + case TCP_FIN_WAIT1: { + struct dst_entry *dst; + int tmo; + + /* If we enter the TCP_FIN_WAIT1 state and we are a + * Fast Open socket and this is the first acceptable + * ACK we have received, this would have acknowledged + * our SYNACK so stop the SYNACK timer. + */ + if (req != NULL) { + /* Return RST if ack_seq is invalid. + * Note that RFC793 only says to generate a + * DUPACK for it but for TCP Fast Open it seems + * better to treat this case like TCP_SYN_RECV + * above. + */ + if (!acceptable) return 1; - } + /* We no longer need the request sock. */ + reqsk_fastopen_remove(sk, req, false); + tcp_rearm_rto(sk); + } + if (tp->snd_una != tp->write_seq) break; - case TCP_FIN_WAIT1: - if (tp->snd_una == tp->write_seq) { - tcp_set_state(sk, TCP_FIN_WAIT2); - sk->sk_shutdown |= SEND_SHUTDOWN; - dst_confirm(__sk_dst_get(sk)); - - if (!sock_flag(sk, SOCK_DEAD)) - /* Wake up lingering close() */ - sk->sk_state_change(sk); - else { - int tmo; - - if (tp->linger2 < 0 || - (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq && - after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt))) { - tcp_done(sk); - NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONDATA); - return 1; - } + tcp_set_state(sk, TCP_FIN_WAIT2); + sk->sk_shutdown |= SEND_SHUTDOWN; - tmo = tcp_fin_time(sk); - if (tmo > TCP_TIMEWAIT_LEN) { - inet_csk_reset_keepalive_timer(sk, tmo - TCP_TIMEWAIT_LEN); - } else if (th->fin || sock_owned_by_user(sk)) { - /* Bad case. We could lose such FIN otherwise. - * It is not a big problem, but it looks confusing - * and not so rare event. We still can lose it now, - * if it spins in bh_lock_sock(), but it is really - * marginal case. - */ - inet_csk_reset_keepalive_timer(sk, tmo); - } else { - tcp_time_wait(sk, TCP_FIN_WAIT2, tmo); - goto discard; - } - } - } - break; + dst = __sk_dst_get(sk); + if (dst) + dst_confirm(dst); - case TCP_CLOSING: - if (tp->snd_una == tp->write_seq) { - tcp_time_wait(sk, TCP_TIME_WAIT, 0); - goto discard; - } + if (!sock_flag(sk, SOCK_DEAD)) { + /* Wake up lingering close() */ + sk->sk_state_change(sk); break; + } - case TCP_LAST_ACK: - if (tp->snd_una == tp->write_seq) { - tcp_update_metrics(sk); - tcp_done(sk); - goto discard; - } - break; + if (tp->linger2 < 0 || + (TCP_SKB_CB(skb)->end_seq != TCP_SKB_CB(skb)->seq && + after(TCP_SKB_CB(skb)->end_seq - th->fin, tp->rcv_nxt))) { + tcp_done(sk); + NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONDATA); + return 1; } - } else - goto discard; + + tmo = tcp_fin_time(sk); + if (tmo > TCP_TIMEWAIT_LEN) { + inet_csk_reset_keepalive_timer(sk, tmo - TCP_TIMEWAIT_LEN); + } else if (th->fin || sock_owned_by_user(sk)) { + /* Bad case. We could lose such FIN otherwise. + * It is not a big problem, but it looks confusing + * and not so rare event. We still can lose it now, + * if it spins in bh_lock_sock(), but it is really + * marginal case. + */ + inet_csk_reset_keepalive_timer(sk, tmo); + } else { + tcp_time_wait(sk, TCP_FIN_WAIT2, tmo); + goto discard; + } + break; + } + + case TCP_CLOSING: + if (tp->snd_una == tp->write_seq) { + tcp_time_wait(sk, TCP_TIME_WAIT, 0); + goto discard; + } + break; + + case TCP_LAST_ACK: + if (tp->snd_una == tp->write_seq) { + tcp_update_metrics(sk); + tcp_done(sk); + goto discard; + } + break; + } /* step 6: check the URG bit */ tcp_urg(sk, skb, th); |