* Andi Kleen: Make sure we never ack data there is not
* enough room for. Also make this condition
* a fatal error if it might still happen.
- * Andi Kleen: Add tcp_measure_rcv_mss to make
+ * Andi Kleen: Add tcp_measure_rcv_mss to make
* connections with MSS<min(MTU,ann. MSS)
- * work without delayed acks.
+ * work without delayed acks.
* Andi Kleen: Process packets with PSH set in the
* fast path.
* J Hadi Salim: ECN support
* Pasi Sarolahti: F-RTO for dealing with spurious RTOs
*/
-#include <linux/config.h>
#include <linux/mm.h>
#include <linux/module.h>
#include <linux/sysctl.h>
#include <net/inet_common.h>
#include <linux/ipsec.h>
#include <asm/unaligned.h>
-
-int sysctl_tcp_timestamps = 1;
-int sysctl_tcp_window_scaling = 1;
-int sysctl_tcp_sack = 1;
-int sysctl_tcp_fack = 1;
-int sysctl_tcp_reordering = TCP_FASTRETRANS_THRESH;
-int sysctl_tcp_ecn;
-int sysctl_tcp_dsack = 1;
-int sysctl_tcp_app_win = 31;
-int sysctl_tcp_adv_win_scale = 2;
-
-int sysctl_tcp_stdurg;
-int sysctl_tcp_rfc1337;
-int sysctl_tcp_max_orphans = NR_FILE;
-int sysctl_tcp_frto;
-int sysctl_tcp_nometrics_save;
-
-int sysctl_tcp_moderate_rcvbuf = 1;
-int sysctl_tcp_abc = 1;
+#include <net/netdma.h>
+
+int sysctl_tcp_timestamps __read_mostly = 1;
+int sysctl_tcp_window_scaling __read_mostly = 1;
+int sysctl_tcp_sack __read_mostly = 1;
+int sysctl_tcp_fack __read_mostly = 1;
+int sysctl_tcp_reordering __read_mostly = TCP_FASTRETRANS_THRESH;
+int sysctl_tcp_ecn __read_mostly;
+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_stdurg __read_mostly;
+int sysctl_tcp_rfc1337 __read_mostly;
+int sysctl_tcp_max_orphans __read_mostly = NR_FILE;
+int sysctl_tcp_frto __read_mostly;
+int sysctl_tcp_nometrics_save __read_mostly;
+
+int sysctl_tcp_moderate_rcvbuf __read_mostly = 1;
+int sysctl_tcp_abc __read_mostly;
#define FLAG_DATA 0x01 /* Incoming frame contained data. */
#define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */
#define TCP_REMNANT (TCP_FLAG_FIN|TCP_FLAG_URG|TCP_FLAG_SYN|TCP_FLAG_PSH)
-/* Adapt the MSS value used to make delayed ack decision to the
+/* Adapt the MSS value used to make delayed ack decision to the
* real world.
- */
-static inline void tcp_measure_rcv_mss(struct sock *sk,
- const struct sk_buff *skb)
+ */
+static void tcp_measure_rcv_mss(struct sock *sk,
+ const struct sk_buff *skb)
{
struct inet_connection_sock *icsk = inet_csk(sk);
- const unsigned int lss = icsk->icsk_ack.last_seg_size;
+ const unsigned int lss = icsk->icsk_ack.last_seg_size;
unsigned int len;
- icsk->icsk_ack.last_seg_size = 0;
+ icsk->icsk_ack.last_seg_size = 0;
/* skb->len may jitter because of SACKs, even if peer
* sends good full-sized frames.
*/
- len = skb->len;
+ len = skb_shinfo(skb)->gso_size ?: skb->len;
if (len >= icsk->icsk_ack.rcv_mss) {
icsk->icsk_ack.rcv_mss = len;
} else {
return;
}
}
+ if (icsk->icsk_ack.pending & ICSK_ACK_PUSHED)
+ icsk->icsk_ack.pending |= ICSK_ACK_PUSHED2;
icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
}
}
return 0;
}
-static inline void tcp_grow_window(struct sock *sk, struct tcp_sock *tp,
- struct sk_buff *skb)
+static void tcp_grow_window(struct sock *sk, struct tcp_sock *tp,
+ struct sk_buff *skb)
{
/* Check #1 */
if (tp->rcv_ssthresh < tp->window_clamp &&
tp->rcv_ssthresh = min(tp->window_clamp, 2U*tp->advmss);
}
+
+/* Initialize RCV_MSS value.
+ * RCV_MSS is an our guess about MSS used by the peer.
+ * We haven't any direct information about the MSS.
+ * It's better to underestimate the RCV_MSS rather than overestimate.
+ * Overestimations make us ACKing less frequently than needed.
+ * Underestimations are more easy to detect and fix by tcp_measure_rcv_mss().
+ */
+void tcp_initialize_rcv_mss(struct sock *sk)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ unsigned int hint = min_t(unsigned int, tp->advmss, tp->mss_cache);
+
+ hint = min(hint, tp->rcv_wnd/2);
+ hint = min(hint, TCP_MIN_RCVMSS);
+ hint = max(hint, TCP_MIN_MSS);
+
+ inet_csk(sk)->icsk_ack.rcv_mss = hint;
+}
+
/* Receiver "autotuning" code.
*
* The algorithm for RTT estimation w/o timestamps is based on
* are stalled on filesystem I/O.
*
* Also, since we are only going for a minimum in the
- * non-timestamp case, we do not smoother things out
+ * non-timestamp case, we do not smooth things out
* else with timestamps disabled convergence takes too
* long.
*/
struct tcp_sock *tp = tcp_sk(sk);
int time;
int space;
-
+
if (tp->rcvq_space.time == 0)
goto new_measure;
-
+
time = tcp_time_stamp - tp->rcvq_space.time;
if (time < (tp->rcv_rtt_est.rtt >> 3) ||
tp->rcv_rtt_est.rtt == 0)
return;
-
+
space = 2 * (tp->copied_seq - tp->rcvq_space.seq);
space = max(tp->rcvq_space.space, space);
tp->rcvq_space.space = space;
- if (sysctl_tcp_moderate_rcvbuf) {
+ if (sysctl_tcp_moderate_rcvbuf &&
+ !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
int new_clamp = space;
/* Receive space grows, normalize in order to
}
}
}
-
+
new_measure:
tp->rcvq_space.seq = tp->copied_seq;
tp->rcvq_space.time = tcp_time_stamp;
tcp_measure_rcv_mss(sk, skb);
tcp_rcv_rtt_measure(tp);
-
+
now = tcp_time_stamp;
if (!icsk->icsk_ack.ato) {
/* The following amusing code comes from Jacobson's
* article in SIGCOMM '88. Note that rtt and mdev
* are scaled versions of rtt and mean deviation.
- * This is designed to be as fast as possible
+ * This is designed to be as fast as possible
* m stands for "measurement".
*
* On a 1990 paper the rto value is changed to:
*
* Funny. This algorithm seems to be very broken.
* These formulae increase RTO, when it should be decreased, increase
- * too slowly, when it should be increased fastly, decrease too fastly
+ * too slowly, when it should be increased quickly, decrease too quickly
* etc. I guess in BSD RTO takes ONE value, so that it is absolutely
* does not matter how to _calculate_ it. Seems, it was trap
* that VJ failed to avoid. 8)
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)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ tp->prior_ssthresh = 0;
+ tp->bytes_acked = 0;
+ if (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) {
+ tp->undo_marker = 0;
+ tp->snd_ssthresh = inet_csk(sk)->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);
+ }
+}
+
/* Initialize metrics on socket. */
static void tcp_init_metrics(struct sock *sk)
const struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
unsigned char *ptr = ack_skb->h.raw + TCP_SKB_CB(ack_skb)->sacked;
- struct tcp_sack_block *sp = (struct tcp_sack_block *)(ptr+2);
+ struct tcp_sack_block_wire *sp = (struct tcp_sack_block_wire *)(ptr+2);
+ struct sk_buff *cached_skb;
int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
int reord = tp->packets_out;
int prior_fackets;
u32 lost_retrans = 0;
int flag = 0;
int dup_sack = 0;
+ int cached_fack_count;
int i;
+ int first_sack_index;
if (!tp->sacked_out)
tp->fackets_out = 0;
prior_fackets = tp->fackets_out;
+ /* Check for D-SACK. */
+ if (before(ntohl(sp[0].start_seq), TCP_SKB_CB(ack_skb)->ack_seq)) {
+ dup_sack = 1;
+ tp->rx_opt.sack_ok |= 4;
+ NET_INC_STATS_BH(LINUX_MIB_TCPDSACKRECV);
+ } else if (num_sacks > 1 &&
+ !after(ntohl(sp[0].end_seq), ntohl(sp[1].end_seq)) &&
+ !before(ntohl(sp[0].start_seq), ntohl(sp[1].start_seq))) {
+ dup_sack = 1;
+ tp->rx_opt.sack_ok |= 4;
+ NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOFORECV);
+ }
+
+ /* D-SACK for already forgotten data...
+ * Do dumb counting. */
+ if (dup_sack &&
+ !after(ntohl(sp[0].end_seq), prior_snd_una) &&
+ after(ntohl(sp[0].end_seq), tp->undo_marker))
+ tp->undo_retrans--;
+
+ /* Eliminate too old ACKs, but take into
+ * account more or less fresh ones, they can
+ * contain valid SACK info.
+ */
+ if (before(TCP_SKB_CB(ack_skb)->ack_seq, prior_snd_una - tp->max_window))
+ return 0;
+
/* SACK fastpath:
* if the only SACK change is the increase of the end_seq of
* the first block then only apply that SACK block
* and use retrans queue hinting otherwise slowpath */
flag = 1;
- for (i = 0; i< num_sacks; i++) {
- __u32 start_seq = ntohl(sp[i].start_seq);
- __u32 end_seq = ntohl(sp[i].end_seq);
+ for (i = 0; i < num_sacks; i++) {
+ __be32 start_seq = sp[i].start_seq;
+ __be32 end_seq = sp[i].end_seq;
- if (i == 0){
+ if (i == 0) {
if (tp->recv_sack_cache[i].start_seq != start_seq)
flag = 0;
} else {
}
tp->recv_sack_cache[i].start_seq = start_seq;
tp->recv_sack_cache[i].end_seq = end_seq;
-
- /* Check for D-SACK. */
- if (i == 0) {
- u32 ack = TCP_SKB_CB(ack_skb)->ack_seq;
-
- if (before(start_seq, ack)) {
- dup_sack = 1;
- tp->rx_opt.sack_ok |= 4;
- NET_INC_STATS_BH(LINUX_MIB_TCPDSACKRECV);
- } else if (num_sacks > 1 &&
- !after(end_seq, ntohl(sp[1].end_seq)) &&
- !before(start_seq, ntohl(sp[1].start_seq))) {
- dup_sack = 1;
- tp->rx_opt.sack_ok |= 4;
- NET_INC_STATS_BH(LINUX_MIB_TCPDSACKOFORECV);
- }
-
- /* D-SACK for already forgotten data...
- * Do dumb counting. */
- if (dup_sack &&
- !after(end_seq, prior_snd_una) &&
- after(end_seq, tp->undo_marker))
- tp->undo_retrans--;
-
- /* Eliminate too old ACKs, but take into
- * account more or less fresh ones, they can
- * contain valid SACK info.
- */
- if (before(ack, prior_snd_una - tp->max_window))
- return 0;
- }
+ }
+ /* Clear the rest of the cache sack blocks so they won't match mistakenly. */
+ for (; i < ARRAY_SIZE(tp->recv_sack_cache); i++) {
+ tp->recv_sack_cache[i].start_seq = 0;
+ tp->recv_sack_cache[i].end_seq = 0;
}
+ first_sack_index = 0;
if (flag)
num_sacks = 1;
else {
for (j = 0; j < i; j++){
if (after(ntohl(sp[j].start_seq),
ntohl(sp[j+1].start_seq))){
- sp[j].start_seq = htonl(tp->recv_sack_cache[j+1].start_seq);
- sp[j].end_seq = htonl(tp->recv_sack_cache[j+1].end_seq);
- sp[j+1].start_seq = htonl(tp->recv_sack_cache[j].start_seq);
- sp[j+1].end_seq = htonl(tp->recv_sack_cache[j].end_seq);
+ struct tcp_sack_block_wire tmp;
+
+ tmp = sp[j];
+ sp[j] = sp[j+1];
+ sp[j+1] = tmp;
+
+ /* Track where the first SACK block goes to */
+ if (j == first_sack_index)
+ first_sack_index = j+1;
}
}
/* clear flag as used for different purpose in following code */
flag = 0;
+ /* Use SACK fastpath hint if valid */
+ cached_skb = tp->fastpath_skb_hint;
+ cached_fack_count = tp->fastpath_cnt_hint;
+ if (!cached_skb) {
+ cached_skb = sk->sk_write_queue.next;
+ cached_fack_count = 0;
+ }
+
for (i=0; i<num_sacks; i++, sp++) {
struct sk_buff *skb;
__u32 start_seq = ntohl(sp->start_seq);
__u32 end_seq = ntohl(sp->end_seq);
int fack_count;
- /* Use SACK fastpath hint if valid */
- if (tp->fastpath_skb_hint) {
- skb = tp->fastpath_skb_hint;
- fack_count = tp->fastpath_cnt_hint;
- } else {
- skb = sk->sk_write_queue.next;
- fack_count = 0;
- }
+ skb = cached_skb;
+ fack_count = cached_fack_count;
/* Event "B" in the comment above. */
if (after(end_seq, tp->high_seq))
int in_sack, pcount;
u8 sacked;
- tp->fastpath_skb_hint = skb;
- tp->fastpath_cnt_hint = fack_count;
+ cached_skb = skb;
+ cached_fack_count = fack_count;
+ if (i == first_sack_index) {
+ tp->fastpath_skb_hint = skb;
+ tp->fastpath_cnt_hint = fack_count;
+ }
/* The retransmission queue is always in order, so
* we can short-circuit the walk early.
else
pkt_len = (end_seq -
TCP_SKB_CB(skb)->seq);
- if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->tso_size))
+ if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->gso_size))
break;
pcount = tcp_skb_pcount(skb);
}
tp->frto_counter = 1;
if (icsk->icsk_ca_state <= TCP_CA_Disorder ||
- tp->snd_una == tp->high_seq ||
- (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) {
+ tp->snd_una == tp->high_seq ||
+ (icsk->icsk_ca_state == TCP_CA_Loss && !icsk->icsk_retransmits)) {
tp->prior_ssthresh = tcp_current_ssthresh(sk);
tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
tcp_ca_event(sk, CA_EVENT_FRTO);
* Hence, we can detect timed out packets during fast
* retransmit without falling to slow start.
*/
- if (tcp_head_timedout(sk, tp)) {
+ if (!IsReno(tp) && tcp_head_timedout(sk, tp)) {
struct sk_buff *skb;
skb = tp->scoreboard_skb_hint ? tp->scoreboard_skb_hint
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)
{
- const struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
int decr = tp->snd_cwnd_cnt + 1;
tp->snd_cwnd_cnt = decr&1;
decr >>= 1;
- if (decr && tp->snd_cwnd > icsk->icsk_ca_ops->min_cwnd(sk))
+ 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);
}
}
+static void tcp_mtup_probe_failed(struct sock *sk)
+{
+ struct inet_connection_sock *icsk = inet_csk(sk);
+
+ icsk->icsk_mtup.search_high = icsk->icsk_mtup.probe_size - 1;
+ icsk->icsk_mtup.probe_size = 0;
+}
+
+static void tcp_mtup_probe_success(struct sock *sk, struct sk_buff *skb)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
+
+ /* FIXME: breaks with very large cwnd */
+ tp->prior_ssthresh = tcp_current_ssthresh(sk);
+ tp->snd_cwnd = tp->snd_cwnd *
+ tcp_mss_to_mtu(sk, tp->mss_cache) /
+ icsk->icsk_mtup.probe_size;
+ tp->snd_cwnd_cnt = 0;
+ tp->snd_cwnd_stamp = tcp_time_stamp;
+ tp->rcv_ssthresh = tcp_current_ssthresh(sk);
+
+ icsk->icsk_mtup.search_low = icsk->icsk_mtup.probe_size;
+ icsk->icsk_mtup.probe_size = 0;
+ tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
+}
+
+
/* Process an event, which can update packets-in-flight not trivially.
* Main goal of this function is to calculate new estimate for left_out,
* taking into account both packets sitting in receiver's buffer and
* 1. Reno does not count dupacks (sacked_out) automatically. */
if (!tp->packets_out)
tp->sacked_out = 0;
- /* 2. SACK counts snd_fack in packets inaccurately. */
+ /* 2. SACK counts snd_fack in packets inaccurately. */
if (tp->sacked_out == 0)
tp->fackets_out = 0;
- /* Now state machine starts.
+ /* Now state machine starts.
* A. ECE, hence prohibit cwnd undoing, the reduction is required. */
if (flag&FLAG_ECE)
tp->prior_ssthresh = 0;
return;
}
+ /* MTU probe failure: don't reduce cwnd */
+ if (icsk->icsk_ca_state < TCP_CA_CWR &&
+ icsk->icsk_mtup.probe_size &&
+ tp->snd_una == tp->mtu_probe.probe_seq_start) {
+ tcp_mtup_probe_failed(sk);
+ /* Restores the reduction we did in tcp_mtup_probe() */
+ tp->snd_cwnd++;
+ tcp_simple_retransmit(sk);
+ return;
+ }
+
/* Otherwise enter Recovery state */
if (IsReno(tp))
tcp_ack_no_tstamp(sk, seq_rtt, flag);
}
-static inline void tcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt,
- u32 in_flight, int good)
+static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt,
+ u32 in_flight, int good)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
icsk->icsk_ca_ops->cong_avoid(sk, ack, rtt, in_flight, good);
* RFC2988 recommends to restart timer to now+rto.
*/
-static inline void tcp_ack_packets_out(struct sock *sk, struct tcp_sock *tp)
+static void tcp_ack_packets_out(struct sock *sk, struct tcp_sock *tp)
{
if (!tp->packets_out) {
inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS);
__u32 now, __s32 *seq_rtt)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
+ struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
__u32 seq = tp->snd_una;
__u32 packets_acked;
int acked = 0;
return acked;
}
-static inline u32 tcp_usrtt(const struct sk_buff *skb)
+static u32 tcp_usrtt(struct timeval *tv)
{
- struct timeval tv, now;
+ struct timeval now;
do_gettimeofday(&now);
- skb_get_timestamp(skb, &tv);
- return (now.tv_sec - tv.tv_sec) * 1000000 + (now.tv_usec - tv.tv_usec);
+ return (now.tv_sec - tv->tv_sec) * 1000000 + (now.tv_usec - tv->tv_usec);
}
/* Remove acknowledged frames from the retransmission queue. */
u32 pkts_acked = 0;
void (*rtt_sample)(struct sock *sk, u32 usrtt)
= icsk->icsk_ca_ops->rtt_sample;
+ struct timeval tv = { .tv_sec = 0, .tv_usec = 0 };
while ((skb = skb_peek(&sk->sk_write_queue)) &&
skb != sk->sk_send_head) {
- struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
+ struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
__u8 sacked = scb->sacked;
/* If our packet is before the ack sequence we can
tp->retrans_stamp = 0;
}
+ /* MTU probing checks */
+ if (icsk->icsk_mtup.probe_size) {
+ if (!after(tp->mtu_probe.probe_seq_end, TCP_SKB_CB(skb)->end_seq)) {
+ tcp_mtup_probe_success(sk, skb);
+ }
+ }
+
if (sacked) {
if (sacked & TCPCB_RETRANS) {
if(sacked & TCPCB_SACKED_RETRANS)
seq_rtt = -1;
} else if (seq_rtt < 0) {
seq_rtt = now - scb->when;
- if (rtt_sample)
- (*rtt_sample)(sk, tcp_usrtt(skb));
+ skb_get_timestamp(skb, &tv);
}
if (sacked & TCPCB_SACKED_ACKED)
tp->sacked_out -= tcp_skb_pcount(skb);
}
} else if (seq_rtt < 0) {
seq_rtt = now - scb->when;
- if (rtt_sample)
- (*rtt_sample)(sk, tcp_usrtt(skb));
+ skb_get_timestamp(skb, &tv);
}
tcp_dec_pcount_approx(&tp->fackets_out, skb);
tcp_packets_out_dec(tp, skb);
if (acked&FLAG_ACKED) {
tcp_ack_update_rtt(sk, acked, seq_rtt);
tcp_ack_packets_out(sk, tp);
+ if (rtt_sample && !(acked & FLAG_RETRANS_DATA_ACKED))
+ (*rtt_sample)(sk, tcp_usrtt(&tv));
if (icsk->icsk_ca_ops->pkts_acked)
icsk->icsk_ca_ops->pkts_acked(sk, pkts_acked);
if (nwin > tp->max_window) {
tp->max_window = nwin;
- tcp_sync_mss(sk, tp->pmtu_cookie);
+ tcp_sync_mss(sk, inet_csk(sk)->icsk_pmtu_cookie);
}
}
}
static void tcp_process_frto(struct sock *sk, u32 prior_snd_una)
{
struct tcp_sock *tp = tcp_sk(sk);
-
+
tcp_sync_left_out(tp);
-
+
if (tp->snd_una == prior_snd_una ||
!before(tp->snd_una, tp->frto_highmark)) {
/* RTO was caused by loss, start retransmitting in
if (before(ack, prior_snd_una))
goto old_ack;
- if (sysctl_tcp_abc && icsk->icsk_ca_state < TCP_CA_CWR)
- tp->bytes_acked += ack - prior_snd_una;
+ 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);
+ }
if (!(flag&FLAG_SLOWPATH) && after(ack, prior_snd_una)) {
/* Window is constant, pure forward advance.
opt_rx->saw_tstamp = 0;
while(length>0) {
- int opcode=*ptr++;
+ int opcode=*ptr++;
int opsize;
switch (opcode) {
return;
if (opsize > length)
return; /* don't parse partial options */
- switch(opcode) {
+ switch(opcode) {
case TCPOPT_MSS:
if(opsize==TCPOLEN_MSS && th->syn && !estab) {
- u16 in_mss = ntohs(get_unaligned((__u16 *)ptr));
+ u16 in_mss = ntohs(get_unaligned((__be16 *)ptr));
if (in_mss) {
if (opt_rx->user_mss && opt_rx->user_mss < in_mss)
in_mss = opt_rx->user_mss;
if ((estab && opt_rx->tstamp_ok) ||
(!estab && sysctl_tcp_timestamps)) {
opt_rx->saw_tstamp = 1;
- opt_rx->rcv_tsval = ntohl(get_unaligned((__u32 *)ptr));
- opt_rx->rcv_tsecr = ntohl(get_unaligned((__u32 *)(ptr+4)));
+ opt_rx->rcv_tsval = ntohl(get_unaligned((__be32 *)ptr));
+ opt_rx->rcv_tsecr = ntohl(get_unaligned((__be32 *)(ptr+4)));
}
}
break;
opt_rx->sack_ok) {
TCP_SKB_CB(skb)->sacked = (ptr - 2) - (unsigned char *)th;
}
- };
- ptr+=opsize-2;
- length-=opsize;
- };
+#ifdef CONFIG_TCP_MD5SIG
+ case TCPOPT_MD5SIG:
+ /*
+ * The MD5 Hash has already been
+ * checked (see tcp_v{4,6}_do_rcv()).
+ */
+ break;
+#endif
+ };
+ ptr+=opsize-2;
+ length-=opsize;
+ };
}
}
/* Fast parse options. This hopes to only see timestamps.
* If it is wrong it falls back on tcp_parse_options().
*/
-static inline int tcp_fast_parse_options(struct sk_buff *skb, struct tcphdr *th,
- struct tcp_sock *tp)
+static int tcp_fast_parse_options(struct sk_buff *skb, struct tcphdr *th,
+ struct tcp_sock *tp)
{
if (th->doff == sizeof(struct tcphdr)>>2) {
tp->rx_opt.saw_tstamp = 0;
return 0;
} else if (tp->rx_opt.tstamp_ok &&
th->doff == (sizeof(struct tcphdr)>>2)+(TCPOLEN_TSTAMP_ALIGNED>>2)) {
- __u32 *ptr = (__u32 *)(th + 1);
- if (*ptr == ntohl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
+ __be32 *ptr = (__be32 *)(th + 1);
+ if (*ptr == htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
| (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP)) {
tp->rx_opt.saw_tstamp = 1;
++ptr;
}
}
-static __inline__ int
-tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, u32 end_seq)
+static inline int tcp_sack_extend(struct tcp_sack_block *sp, u32 seq, u32 end_seq)
{
if (!after(seq, sp->end_seq) && !after(sp->start_seq, end_seq)) {
if (before(seq, sp->start_seq))
return 0;
}
-static inline void tcp_dsack_set(struct tcp_sock *tp, u32 seq, u32 end_seq)
+static void tcp_dsack_set(struct tcp_sock *tp, u32 seq, u32 end_seq)
{
if (tp->rx_opt.sack_ok && sysctl_tcp_dsack) {
if (before(seq, tp->rcv_nxt))
}
}
-static inline void tcp_dsack_extend(struct tcp_sock *tp, u32 seq, u32 end_seq)
+static void tcp_dsack_extend(struct tcp_sock *tp, u32 seq, u32 end_seq)
{
if (!tp->rx_opt.dsack)
tcp_dsack_set(tp, seq, end_seq);
}
}
-static __inline__ void tcp_sack_swap(struct tcp_sack_block *sack1, struct tcp_sack_block *sack2)
+static inline void tcp_sack_swap(struct tcp_sack_block *sack1, struct tcp_sack_block *sack2)
{
__u32 tmp;
TCP_SKB_CB(skb)->end_seq);
tcp_dsack_set(tp, TCP_SKB_CB(skb)->seq, tp->rcv_nxt);
-
+
/* If window is closed, drop tail of packet. But after
* remembering D-SACK for its head made in previous line.
*/
}
}
__skb_insert(skb, skb1, skb1->next, &tp->out_of_order_queue);
-
+
/* And clean segments covered by new one as whole. */
while ((skb1 = skb->next) !=
(struct sk_buff*)&tp->out_of_order_queue &&
int offset = start - TCP_SKB_CB(skb)->seq;
int size = TCP_SKB_CB(skb)->end_seq - start;
- if (offset < 0) BUG();
+ BUG_ON(offset < 0);
if (size > 0) {
size = min(copy, size);
if (skb_copy_bits(skb, offset, skb_put(nskb, size), size))
*/
static int tcp_prune_queue(struct sock *sk)
{
- struct tcp_sock *tp = tcp_sk(sk);
+ struct tcp_sock *tp = tcp_sk(sk);
SOCK_DEBUG(sk, "prune_queue: c=%x\n", tp->copied_seq);
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 win_used = max(tp->snd_cwnd_used, 2U);
+ 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_stamp = tcp_time_stamp;
}
-static inline int tcp_should_expand_sndbuf(struct sock *sk, struct tcp_sock *tp)
+static int tcp_should_expand_sndbuf(struct sock *sk, struct tcp_sock *tp)
{
/* If the user specified a specific send buffer setting, do
* not modify it.
struct tcp_sock *tp = tcp_sk(sk);
if (tcp_should_expand_sndbuf(sk, tp)) {
- int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) +
+ int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) +
MAX_TCP_HEADER + 16 + sizeof(struct sk_buff),
demanded = max_t(unsigned int, tp->snd_cwnd,
tp->reordering + 1);
sk->sk_write_space(sk);
}
-static inline void tcp_check_space(struct sock *sk)
+static void tcp_check_space(struct sock *sk)
{
if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) {
sock_reset_flag(sk, SOCK_QUEUE_SHRUNK);
}
}
-static __inline__ void tcp_data_snd_check(struct sock *sk, struct tcp_sock *tp)
+static inline void tcp_data_snd_check(struct sock *sk, struct tcp_sock *tp)
{
tcp_push_pending_frames(sk, tp);
tcp_check_space(sk);
}
}
-static __inline__ void tcp_ack_snd_check(struct sock *sk)
+static inline void tcp_ack_snd_check(struct sock *sk)
{
if (!inet_csk_ack_scheduled(sk)) {
/* We sent a data segment already. */
* For 1003.1g we should support a new option TCP_STDURG to permit
* either form (or just set the sysctl tcp_stdurg).
*/
-
+
static void tcp_check_urg(struct sock * sk, struct tcphdr * th)
{
struct tcp_sock *tp = tcp_sk(sk);
u32 ptr = tp->urg_seq - ntohl(th->seq) + (th->doff * 4) -
th->syn;
- /* Is the urgent pointer pointing into this packet? */
+ /* Is the urgent pointer pointing into this packet? */
if (ptr < skb->len) {
u8 tmp;
if (skb_copy_bits(skb, ptr, &tmp, 1))
return err;
}
-static int __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
+static __sum16 __tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
{
- int result;
+ __sum16 result;
if (sock_owned_by_user(sk)) {
local_bh_enable();
return result;
}
-static __inline__ int
-tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
+static inline int tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
{
return skb->ip_summed != CHECKSUM_UNNECESSARY &&
__tcp_checksum_complete_user(sk, skb);
}
+#ifdef CONFIG_NET_DMA
+static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, int hlen)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ int chunk = skb->len - hlen;
+ int dma_cookie;
+ int copied_early = 0;
+
+ if (tp->ucopy.wakeup)
+ return 0;
+
+ if (!tp->ucopy.dma_chan && tp->ucopy.pinned_list)
+ tp->ucopy.dma_chan = get_softnet_dma();
+
+ if (tp->ucopy.dma_chan && skb->ip_summed == CHECKSUM_UNNECESSARY) {
+
+ dma_cookie = dma_skb_copy_datagram_iovec(tp->ucopy.dma_chan,
+ skb, hlen, tp->ucopy.iov, chunk, tp->ucopy.pinned_list);
+
+ if (dma_cookie < 0)
+ goto out;
+
+ tp->ucopy.dma_cookie = dma_cookie;
+ copied_early = 1;
+
+ tp->ucopy.len -= chunk;
+ tp->copied_seq += chunk;
+ tcp_rcv_space_adjust(sk);
+
+ if ((tp->ucopy.len == 0) ||
+ (tcp_flag_word(skb->h.th) & TCP_FLAG_PSH) ||
+ (atomic_read(&sk->sk_rmem_alloc) > (sk->sk_rcvbuf >> 1))) {
+ tp->ucopy.wakeup = 1;
+ sk->sk_data_ready(sk, 0);
+ }
+ } else if (chunk > 0) {
+ tp->ucopy.wakeup = 1;
+ sk->sk_data_ready(sk, 0);
+ }
+out:
+ return copied_early;
+}
+#endif /* CONFIG_NET_DMA */
+
/*
- * TCP receive function for the ESTABLISHED state.
+ * TCP receive function for the ESTABLISHED state.
*
- * It is split into a fast path and a slow path. The fast path is
+ * It is split into a fast path and a slow path. The fast path is
* disabled when:
* - A zero window was announced from us - zero window probing
- * is only handled properly in the slow path.
+ * is only handled properly in the slow path.
* - Out of order segments arrived.
* - Urgent data is expected.
* - There is no buffer space left
* - Unexpected TCP flags/window values/header lengths are received
- * (detected by checking the TCP header against pred_flags)
+ * (detected by checking the TCP header against pred_flags)
* - Data is sent in both directions. Fast path only supports pure senders
* or pure receivers (this means either the sequence number or the ack
* value must stay constant)
* - Unexpected TCP option.
*
- * When these conditions are not satisfied it drops into a standard
+ * When these conditions are not satisfied it drops into a standard
* receive procedure patterned after RFC793 to handle all cases.
* The first three cases are guaranteed by proper pred_flags setting,
- * the rest is checked inline. Fast processing is turned on in
+ * 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,
/*
* Header prediction.
- * The code loosely follows the one in the famous
+ * The code loosely follows the one in the famous
* "30 instruction TCP receive" Van Jacobson mail.
- *
- * Van's trick is to deposit buffers into socket queue
+ *
+ * Van's trick is to deposit buffers into socket queue
* on a device interrupt, to call tcp_recv function
* on the receive process context and checksum and copy
* the buffer to user space. smart...
*
- * Our current scheme is not silly either but we take the
+ * Our current scheme is not silly either but we take the
* extra cost of the net_bh soft interrupt processing...
* We do checksum and copy also but from device to kernel.
*/
* if header_prediction is to be made
* 'S' will always be tp->tcp_header_len >> 2
* '?' will be 0 for the fast path, otherwise pred_flags is 0 to
- * turn it off (when there are holes in the receive
+ * turn it off (when there are holes in the receive
* space for instance)
* PSH flag is ignored.
*/
/* Check timestamp */
if (tcp_header_len == sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) {
- __u32 *ptr = (__u32 *)(th + 1);
+ __be32 *ptr = (__be32 *)(th + 1);
/* No? Slow path! */
- if (*ptr != ntohl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
+ if (*ptr != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
| (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP))
goto slow_path;
tp->rx_opt.saw_tstamp = 1;
- ++ptr;
+ ++ptr;
tp->rx_opt.rcv_tsval = ntohl(*ptr);
++ptr;
tp->rx_opt.rcv_tsecr = ntohl(*ptr);
tp->rcv_nxt == tp->rcv_wup)
tcp_store_ts_recent(tp);
- tcp_rcv_rtt_measure_ts(sk, skb);
-
/* We know that such packets are checksummed
* on entry.
*/
tcp_ack(sk, skb, 0);
- __kfree_skb(skb);
+ __kfree_skb(skb);
tcp_data_snd_check(sk, tp);
return 0;
} else { /* Header too small */
}
} else {
int eaten = 0;
+ int copied_early = 0;
- if (tp->ucopy.task == current &&
- tp->copied_seq == tp->rcv_nxt &&
- len - tcp_header_len <= tp->ucopy.len &&
- sock_owned_by_user(sk)) {
- __set_current_state(TASK_RUNNING);
+ 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)) {
+ copied_early = 1;
+ eaten = 1;
+ }
+#endif
+ if (tp->ucopy.task == current && sock_owned_by_user(sk) && !copied_early) {
+ __set_current_state(TASK_RUNNING);
- if (!tcp_copy_to_iovec(sk, skb, tcp_header_len)) {
+ if (!tcp_copy_to_iovec(sk, skb, tcp_header_len))
+ eaten = 1;
+ }
+ if (eaten) {
/* Predicted packet is in window by definition.
* seq == rcv_nxt and rcv_wup <= rcv_nxt.
* Hence, check seq<=rcv_wup reduces to:
__skb_pull(skb, tcp_header_len);
tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
NET_INC_STATS_BH(LINUX_MIB_TCPHPHITSTOUSER);
- eaten = 1;
}
+ if (copied_early)
+ tcp_cleanup_rbuf(sk, skb->len);
}
if (!eaten) {
if (tcp_checksum_complete_user(sk, skb))
__tcp_ack_snd_check(sk, 0);
no_ack:
+#ifdef CONFIG_NET_DMA
+ if (copied_early)
+ __skb_queue_tail(&sk->sk_async_wait_queue, skb);
+ else
+#endif
if (eaten)
__kfree_skb(skb);
else
struct tcphdr *th, unsigned len)
{
struct tcp_sock *tp = tcp_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
int saved_clamp = tp->rx_opt.mss_clamp;
tcp_parse_options(skb, &tp->rx_opt, 0);
if (th->ack) {
- struct inet_connection_sock *icsk;
/* rfc793:
* "If the state is SYN-SENT then
* first check the ACK bit
*/
TCP_ECN_rcv_synack(tp, th);
- if (tp->ecn_flags&TCP_ECN_OK)
- sock_set_flag(sk, SOCK_NO_LARGESEND);
tp->snd_wl1 = TCP_SKB_CB(skb)->seq;
tcp_ack(sk, skb, FLAG_SLOWPATH);
if (tp->rx_opt.sack_ok && sysctl_tcp_fack)
tp->rx_opt.sack_ok |= 2;
- tcp_sync_mss(sk, tp->pmtu_cookie);
+ tcp_mtup_init(sk);
+ tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
tcp_initialize_rcv_mss(sk);
/* Remember, tcp_poll() does not lock socket!
* Change state from SYN-SENT only after copied_seq
* is initialized. */
tp->copied_seq = tp->rcv_nxt;
- mb();
+ smp_mb();
tcp_set_state(sk, TCP_ESTABLISHED);
+ security_inet_conn_established(sk, skb);
+
/* Make sure socket is routed, for correct metrics. */
- tp->af_specific->rebuild_header(sk);
+ icsk->icsk_af_ops->rebuild_header(sk);
tcp_init_metrics(sk);
sk_wake_async(sk, 0, POLL_OUT);
}
- icsk = inet_csk(sk);
-
if (sk->sk_write_pending ||
icsk->icsk_accept_queue.rskq_defer_accept ||
icsk->icsk_ack.pingpong) {
tp->max_window = tp->snd_wnd;
TCP_ECN_rcv_syn(tp, th);
- if (tp->ecn_flags&TCP_ECN_OK)
- sock_set_flag(sk, SOCK_NO_LARGESEND);
- tcp_sync_mss(sk, tp->pmtu_cookie);
+ tcp_mtup_init(sk);
+ tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
tcp_initialize_rcv_mss(sk);
/*
* This function implements the receiving procedure of RFC 793 for
- * all states except ESTABLISHED and TIME_WAIT.
+ * all states except ESTABLISHED and TIME_WAIT.
* It's called from both tcp_v4_rcv and tcp_v6_rcv and should be
* address independent.
*/
-
+
int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
struct tcphdr *th, unsigned len)
{
struct tcp_sock *tp = tcp_sk(sk);
+ struct inet_connection_sock *icsk = inet_csk(sk);
int queued = 0;
tp->rx_opt.saw_tstamp = 0;
goto discard;
if(th->syn) {
- if(tp->af_specific->conn_request(sk, skb) < 0)
+ if (icsk->icsk_af_ops->conn_request(sk, skb) < 0)
return 1;
- /* Now we have several options: In theory there is
- * nothing else in the frame. KA9Q has an option to
+ /* Now we have several options: In theory there is
+ * nothing else in the frame. KA9Q has an option to
* send data with the syn, BSD accepts data with the
- * syn up to the [to be] advertised window and
- * Solaris 2.1 gives you a protocol error. For now
- * we just ignore it, that fits the spec precisely
+ * syn up to the [to be] advertised window and
+ * Solaris 2.1 gives you a protocol error. For now
+ * we just ignore it, that fits the spec precisely
* and avoids incompatibilities. It would be nice in
* future to drop through and process the data.
*
- * Now that TTCP is starting to be used we ought to
+ * Now that TTCP is starting to be used we ought to
* queue this data.
* But, this leaves one open to an easy denial of
- * service attack, and SYN cookies can't defend
+ * service attack, and SYN cookies can't defend
* against this problem. So, we drop the data
- * in the interest of security over speed.
+ * in the interest of security over speed unless
+ * it's still in use.
*/
- goto discard;
+ kfree_skb(skb);
+ return 0;
}
goto discard;
case TCP_SYN_RECV:
if (acceptable) {
tp->copied_seq = tp->rcv_nxt;
- mb();
+ smp_mb();
tcp_set_state(sk, TCP_ESTABLISHED);
sk->sk_state_change(sk);
/* Make sure socket is routed, for
* correct metrics.
*/
- tp->af_specific->rebuild_header(sk);
+ icsk->icsk_af_ops->rebuild_header(sk);
tcp_init_metrics(sk);
*/
tp->lsndtime = tcp_time_stamp;
+ tcp_mtup_init(sk);
tcp_initialize_rcv_mss(sk);
tcp_init_buffer_space(sk);
tcp_fast_path_on(tp);
case TCP_FIN_WAIT1:
case TCP_FIN_WAIT2:
/* RFC 793 says to queue data in these states,
- * RFC 1122 says we MUST send a reset.
+ * RFC 1122 says we MUST send a reset.
* BSD 4.4 also does reset.
*/
if (sk->sk_shutdown & RCV_SHUTDOWN) {
}
}
/* Fall through */
- case TCP_ESTABLISHED:
+ case TCP_ESTABLISHED:
tcp_data_queue(sk, skb);
queued = 1;
break;
tcp_ack_snd_check(sk);
}
- if (!queued) {
+ if (!queued) {
discard:
__kfree_skb(skb);
}
EXPORT_SYMBOL(sysctl_tcp_ecn);
EXPORT_SYMBOL(sysctl_tcp_reordering);
-EXPORT_SYMBOL(sysctl_tcp_abc);
EXPORT_SYMBOL(tcp_parse_options);
EXPORT_SYMBOL(tcp_rcv_established);
EXPORT_SYMBOL(tcp_rcv_state_process);
+EXPORT_SYMBOL(tcp_initialize_rcv_mss);