* 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_frto_response __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 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_ACKED (FLAG_DATA_ACKED|FLAG_SYN_ACKED)
#define FLAG_NOT_DUP (FLAG_DATA|FLAG_WIN_UPDATE|FLAG_ACKED)
#define IsFack(tp) ((tp)->rx_opt.sack_ok & 2)
#define IsDSack(tp) ((tp)->rx_opt.sack_ok & 4)
+#define IsSackFrto() (sysctl_tcp_frto == 0x2)
+
#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 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 {
*
* "len" is invariant segment length, including TCP header.
*/
- len += skb->data - skb->h.raw;
+ len += skb->data - skb_transport_header(skb);
if (len >= TCP_MIN_RCVMSS + sizeof(struct tcphdr) ||
/* If PSH is not set, packet should be
* full sized, provided peer TCP is not badly broken.
* to handle super-low mtu links fairly.
*/
(len >= TCP_MIN_MSS + sizeof(struct tcphdr) &&
- !(tcp_flag_word(skb->h.th)&TCP_REMNANT))) {
+ !(tcp_flag_word(tcp_hdr(skb)) & TCP_REMNANT))) {
/* Subtract also invariant (if peer is RFC compliant),
* tcp header plus fixed timestamp option length.
* Resulting "len" is MSS free of SACK jitter.
return;
}
}
+ if (icsk->icsk_ack.pending & ICSK_ACK_PUSHED)
+ icsk->icsk_ack.pending |= ICSK_ACK_PUSHED2;
icsk->icsk_ack.pending |= ICSK_ACK_PUSHED;
}
}
*/
/* Slow part of check#2. */
-static int __tcp_grow_window(const struct sock *sk, struct tcp_sock *tp,
- const struct sk_buff *skb)
+static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb)
{
+ struct tcp_sock *tp = tcp_sk(sk);
/* Optimize this! */
int truesize = tcp_win_from_space(skb->truesize)/2;
int window = tcp_win_from_space(sysctl_tcp_rmem[2])/2;
return 0;
}
-static void tcp_grow_window(struct sock *sk, struct tcp_sock *tp,
+static void tcp_grow_window(struct sock *sk,
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) &&
if (tcp_win_from_space(skb->truesize) <= skb->len)
incr = 2*tp->advmss;
else
- incr = __tcp_grow_window(sk, tp, skb);
+ incr = __tcp_grow_window(sk, skb);
if (incr) {
tp->rcv_ssthresh = min(tp->rcv_ssthresh + incr, tp->window_clamp);
}
/* 5. Recalculate window clamp after socket hit its memory bounds. */
-static void tcp_clamp_window(struct sock *sk, struct tcp_sock *tp)
+static void tcp_clamp_window(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
icsk->icsk_ack.quick = 0;
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);
}
}
}
-
+
new_measure:
tp->rcvq_space.seq = tp->copied_seq;
tp->rcvq_space.time = tcp_time_stamp;
* each ACK we send, he increments snd_cwnd and transmits more of his
* queue. -DaveM
*/
-static void tcp_event_data_recv(struct sock *sk, struct tcp_sock *tp, struct sk_buff *skb)
+static void tcp_event_data_recv(struct sock *sk, struct sk_buff *skb)
{
+ struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
u32 now;
tcp_measure_rcv_mss(sk, skb);
tcp_rcv_rtt_measure(tp);
-
+
now = tcp_time_stamp;
if (!icsk->icsk_ack.ato) {
TCP_ECN_check_ce(tp, skb);
if (skb->len >= 128)
- tcp_grow_window(sk, tp, skb);
+ tcp_grow_window(sk, skb);
}
/* Called to compute a smoothed rtt estimate. The data fed to this
/* 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:
* does not matter how to _calculate_ it. Seems, it was trap
* that VJ failed to avoid. 8)
*/
- if(m == 0)
+ if (m == 0)
m = 1;
if (tp->srtt != 0) {
m -= (tp->srtt >> 3); /* m is now error in rtt est */
}
/* Set slow start threshold and cwnd not falling to slow start */
-void tcp_enter_cwr(struct sock *sk)
+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 (inet_csk(sk)->icsk_ca_state < TCP_CA_CWR) {
+ if (icsk->icsk_ca_state < TCP_CA_CWR) {
tp->undo_marker = 0;
- tp->snd_ssthresh = inet_csk(sk)->icsk_ca_ops->ssthresh(sk);
+ 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;
{
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);
+ unsigned char *ptr = (skb_transport_header(ack_skb) +
+ TCP_SKB_CB(ack_skb)->sacked);
+ struct tcp_sack_block_wire *sp = (struct tcp_sack_block_wire *)(ptr+2);
+ struct sk_buff *cached_skb;
int num_sacks = (ptr[1] - TCPOLEN_SACK_BASE)>>3;
int reord = tp->packets_out;
int prior_fackets;
u32 lost_retrans = 0;
int flag = 0;
- int dup_sack = 0;
+ int found_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)) {
+ found_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))) {
+ found_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 (found_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 = tcp_write_queue_head(sk);
+ 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;
+ int dup_sack = (found_dup_sack && (i == first_sack_index));
- /* 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))
flag |= FLAG_DATA_LOST;
- sk_stream_for_retrans_queue_from(skb, sk) {
+ tcp_for_write_queue_from(skb, sk) {
int in_sack, pcount;
u8 sacked;
- tp->fastpath_skb_hint = skb;
- tp->fastpath_cnt_hint = fack_count;
+ if (skb == tcp_send_head(sk))
+ break;
+
+ cached_skb = skb;
+ cached_fack_count = fack_count;
+ if (i == first_sack_index) {
+ tp->fastpath_skb_hint = skb;
+ tp->fastpath_cnt_hint = fack_count;
+ }
/* 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);
}
/* clear lost hint */
tp->retransmit_skb_hint = NULL;
}
+ /* SACK enhanced F-RTO detection.
+ * Set flag if and only if non-rexmitted
+ * segments below frto_highmark are
+ * SACKed (RFC4138; Appendix B).
+ * Clearing correct due to in-order walk
+ */
+ if (after(end_seq, tp->frto_highmark)) {
+ flag &= ~FLAG_ONLY_ORIG_SACKED;
+ } else {
+ if (!(sacked & TCPCB_RETRANS))
+ flag |= FLAG_ONLY_ORIG_SACKED;
+ }
}
TCP_SKB_CB(skb)->sacked |= TCPCB_SACKED_ACKED;
if (lost_retrans && icsk->icsk_ca_state == TCP_CA_Recovery) {
struct sk_buff *skb;
- sk_stream_for_retrans_queue(skb, sk) {
+ tcp_for_write_queue(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
if (after(TCP_SKB_CB(skb)->seq, lost_retrans))
break;
if (!after(TCP_SKB_CB(skb)->end_seq, tp->snd_una))
tp->left_out = tp->sacked_out + tp->lost_out;
- if ((reord < tp->fackets_out) && icsk->icsk_ca_state != TCP_CA_Loss)
+ if ((reord < tp->fackets_out) && icsk->icsk_ca_state != TCP_CA_Loss &&
+ (!tp->frto_highmark || after(tp->snd_una, tp->frto_highmark)))
tcp_update_reordering(sk, ((tp->fackets_out + 1) - reord), 0);
#if FASTRETRANS_DEBUG > 0
return flag;
}
-/* RTO occurred, but do not yet enter loss state. Instead, transmit two new
- * segments to see from the next ACKs whether any data was really missing.
- * If the RTO was spurious, new ACKs should arrive.
+/* 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);
+ struct sk_buff *skb;
+
+ if (!sysctl_tcp_frto)
+ return 0;
+
+ if (IsSackFrto())
+ return 1;
+
+ /* Avoid expensive walking of rexmit queue if possible */
+ if (tp->retrans_out > 1)
+ return 0;
+
+ skb = tcp_write_queue_head(sk);
+ 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)
{
struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *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)) {
+ 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);
- tp->snd_ssthresh = icsk->icsk_ca_ops->ssthresh(sk);
+ /* Our state is too optimistic in ssthresh() call because cwnd
+ * is not reduced until tcp_enter_frto_loss() when previous FRTO
+ * 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);
}
- /* Have to clear retransmission markers here to keep the bookkeeping
- * in shape, even though we are not yet in Loss state.
- * If something was really lost, it is eventually caught up
- * in tcp_enter_frto_loss.
- */
- tp->retrans_out = 0;
tp->undo_marker = tp->snd_una;
tp->undo_retrans = 0;
- sk_stream_for_retrans_queue(skb, sk) {
- TCP_SKB_CB(skb)->sacked &= ~TCPCB_RETRANS;
+ skb = tcp_write_queue_head(sk);
+ 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_sync_left_out(tp);
- tcp_set_ca_state(sk, TCP_CA_Open);
- tp->frto_highmark = tp->snd_nxt;
+ /* Earlier loss recovery underway (see RFC4138; Appendix B).
+ * The last condition is necessary at least in tp->frto_counter case.
+ */
+ if (IsSackFrto() && (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)
+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->sacked_out = 0;
tp->lost_out = 0;
tp->fackets_out = 0;
+ tp->retrans_out = 0;
- sk_stream_for_retrans_queue(skb, sk) {
+ tcp_for_write_queue(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
cnt += tcp_skb_pcount(skb);
- 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 {
+ TCP_SKB_CB(skb)->sacked &= ~(TCPCB_LOST|TCPCB_SACKED_RETRANS);
+ }
if (!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED)) {
/* Do not mark those segments lost that were
}
tcp_sync_left_out(tp);
- tp->snd_cwnd = tp->frto_counter + tcp_packets_in_flight(tp)+1;
+ tp->snd_cwnd = tcp_packets_in_flight(tp) + allowed_segments;
tp->snd_cwnd_cnt = 0;
tp->snd_cwnd_stamp = tcp_time_stamp;
tp->undo_marker = 0;
if (!how)
tp->undo_marker = tp->snd_una;
- sk_stream_for_retrans_queue(skb, sk) {
+ tcp_for_write_queue(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
cnt += tcp_skb_pcount(skb);
if (TCP_SKB_CB(skb)->sacked&TCPCB_RETRANS)
tp->undo_marker = 0;
tcp_set_ca_state(sk, TCP_CA_Loss);
tp->high_seq = tp->snd_nxt;
TCP_ECN_queue_cwr(tp);
+ /* Abort FRTO algorithm if one is in progress */
+ tp->frto_counter = 0;
clear_all_retrans_hints(tp);
}
* receiver _host_ is heavily congested (or buggy).
* Do processing similar to RTO timeout.
*/
- if ((skb = skb_peek(&sk->sk_write_queue)) != NULL &&
+ if ((skb = tcp_write_queue_head(sk)) != NULL &&
(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) {
struct inet_connection_sock *icsk = inet_csk(sk);
NET_INC_STATS_BH(LINUX_MIB_TCPSACKRENEGING);
tcp_enter_loss(sk, 1);
icsk->icsk_retransmits++;
- tcp_retransmit_skb(sk, skb_peek(&sk->sk_write_queue));
+ 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 (tcp_time_stamp - TCP_SKB_CB(skb)->when > inet_csk(sk)->icsk_rto);
}
-static inline int tcp_head_timedout(struct sock *sk, struct tcp_sock *tp)
+static inline int tcp_head_timedout(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
return tp->packets_out &&
- tcp_skb_timedout(sk, skb_peek(&sk->sk_write_queue));
+ tcp_skb_timedout(sk, tcp_write_queue_head(sk));
}
/* Linux NewReno/SACK/FACK/ECN state machine.
* Main question: may we further continue forward transmission
* with the same cwnd?
*/
-static int tcp_time_to_recover(struct sock *sk, struct tcp_sock *tp)
+static int tcp_time_to_recover(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
__u32 packets_out;
+ /* Do not perform any recovery during FRTO algorithm */
+ if (tp->frto_counter)
+ return 0;
+
/* Trick#1: The loss is proven. */
if (tp->lost_out)
return 1;
/* Trick#3 : when we use RFC2988 timer restart, fast
* retransmit can be triggered by timeout of queue head.
*/
- if (tcp_head_timedout(sk, tp))
+ if (tcp_head_timedout(sk))
return 1;
/* Trick#4: It is still not OK... But will it be useful to delay
packets_out = tp->packets_out;
if (packets_out <= tp->reordering &&
tp->sacked_out >= max_t(__u32, packets_out/2, sysctl_tcp_reordering) &&
- !tcp_may_send_now(sk, tp)) {
+ !tcp_may_send_now(sk)) {
/* We have nothing to send. This connection is limited
* either by receiver window or by application.
*/
/* Account for ACK, ACKing some data in Reno Recovery phase. */
-static void tcp_remove_reno_sacks(struct sock *sk, struct tcp_sock *tp, int acked)
+static void tcp_remove_reno_sacks(struct sock *sk, int acked)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
if (acked > 0) {
/* One ACK acked hole. The rest eat duplicate ACKs. */
if (acked-1 >= tp->sacked_out)
}
/* Mark head of queue up as lost. */
-static void tcp_mark_head_lost(struct sock *sk, struct tcp_sock *tp,
+static void tcp_mark_head_lost(struct sock *sk,
int packets, u32 high_seq)
{
+ struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *skb;
int cnt;
skb = tp->lost_skb_hint;
cnt = tp->lost_cnt_hint;
} else {
- skb = sk->sk_write_queue.next;
+ skb = tcp_write_queue_head(sk);
cnt = 0;
}
- sk_stream_for_retrans_queue_from(skb, sk) {
+ tcp_for_write_queue_from(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
/* TODO: do this better */
/* this is not the most efficient way to do this... */
tp->lost_skb_hint = skb;
/* clear xmit_retransmit_queue hints
* if this is beyond hint */
- if(tp->retransmit_skb_hint != NULL &&
- before(TCP_SKB_CB(skb)->seq,
- TCP_SKB_CB(tp->retransmit_skb_hint)->seq)) {
-
+ if (tp->retransmit_skb_hint != NULL &&
+ before(TCP_SKB_CB(skb)->seq,
+ TCP_SKB_CB(tp->retransmit_skb_hint)->seq))
tp->retransmit_skb_hint = NULL;
- }
+
}
}
tcp_sync_left_out(tp);
/* Account newly detected lost packet(s) */
-static void tcp_update_scoreboard(struct sock *sk, struct tcp_sock *tp)
+static void tcp_update_scoreboard(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
if (IsFack(tp)) {
int lost = tp->fackets_out - tp->reordering;
if (lost <= 0)
lost = 1;
- tcp_mark_head_lost(sk, tp, lost, tp->high_seq);
+ tcp_mark_head_lost(sk, lost, tp->high_seq);
} else {
- tcp_mark_head_lost(sk, tp, 1, tp->high_seq);
+ tcp_mark_head_lost(sk, 1, tp->high_seq);
}
/* New heuristics: it is possible only after we switched
* 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)) {
struct sk_buff *skb;
skb = tp->scoreboard_skb_hint ? tp->scoreboard_skb_hint
- : sk->sk_write_queue.next;
+ : tcp_write_queue_head(sk);
- sk_stream_for_retrans_queue_from(skb, sk) {
+ tcp_for_write_queue_from(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
if (!tcp_skb_timedout(sk, skb))
break;
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);
/* Undo procedures. */
#if FASTRETRANS_DEBUG > 1
-static void DBGUNDO(struct sock *sk, struct tcp_sock *tp, const char *msg)
+static void DBGUNDO(struct sock *sk, const char *msg)
{
+ struct tcp_sock *tp = tcp_sk(sk);
struct inet_sock *inet = inet_sk(sk);
+
printk(KERN_DEBUG "Undo %s %u.%u.%u.%u/%u c%u l%u ss%u/%u p%u\n",
msg,
NIPQUAD(inet->daddr), ntohs(inet->dport),
}
/* People celebrate: "We love our President!" */
-static int tcp_try_undo_recovery(struct sock *sk, struct tcp_sock *tp)
+static int tcp_try_undo_recovery(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
if (tcp_may_undo(tp)) {
/* Happy end! We did not retransmit anything
* or our original transmission succeeded.
*/
- DBGUNDO(sk, tp, inet_csk(sk)->icsk_ca_state == TCP_CA_Loss ? "loss" : "retrans");
+ DBGUNDO(sk, inet_csk(sk)->icsk_ca_state == TCP_CA_Loss ? "loss" : "retrans");
tcp_undo_cwr(sk, 1);
if (inet_csk(sk)->icsk_ca_state == TCP_CA_Loss)
NET_INC_STATS_BH(LINUX_MIB_TCPLOSSUNDO);
}
/* Try to undo cwnd reduction, because D-SACKs acked all retransmitted data */
-static void tcp_try_undo_dsack(struct sock *sk, struct tcp_sock *tp)
+static void tcp_try_undo_dsack(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
if (tp->undo_marker && !tp->undo_retrans) {
- DBGUNDO(sk, tp, "D-SACK");
+ DBGUNDO(sk, "D-SACK");
tcp_undo_cwr(sk, 1);
tp->undo_marker = 0;
NET_INC_STATS_BH(LINUX_MIB_TCPDSACKUNDO);
/* Undo during fast recovery after partial ACK. */
-static int tcp_try_undo_partial(struct sock *sk, struct tcp_sock *tp,
- int acked)
+static int tcp_try_undo_partial(struct sock *sk, int acked)
{
+ struct tcp_sock *tp = tcp_sk(sk);
/* Partial ACK arrived. Force Hoe's retransmit. */
int failed = IsReno(tp) || tp->fackets_out>tp->reordering;
tcp_update_reordering(sk, tcp_fackets_out(tp) + acked, 1);
- DBGUNDO(sk, tp, "Hoe");
+ DBGUNDO(sk, "Hoe");
tcp_undo_cwr(sk, 0);
NET_INC_STATS_BH(LINUX_MIB_TCPPARTIALUNDO);
}
/* Undo during loss recovery after partial ACK. */
-static int tcp_try_undo_loss(struct sock *sk, struct tcp_sock *tp)
+static int tcp_try_undo_loss(struct sock *sk)
{
+ struct tcp_sock *tp = tcp_sk(sk);
+
if (tcp_may_undo(tp)) {
struct sk_buff *skb;
- sk_stream_for_retrans_queue(skb, sk) {
+ tcp_for_write_queue(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
}
clear_all_retrans_hints(tp);
- DBGUNDO(sk, tp, "partial loss");
+ DBGUNDO(sk, "partial loss");
tp->lost_out = 0;
tp->left_out = tp->sacked_out;
tcp_undo_cwr(sk, 1);
tcp_ca_event(sk, CA_EVENT_COMPLETE_CWR);
}
-static void tcp_try_to_open(struct sock *sk, struct tcp_sock *tp, int flag)
+static void tcp_try_to_open(struct sock *sk, int flag)
{
- tp->left_out = tp->sacked_out;
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ tcp_sync_left_out(tp);
if (tp->retrans_out == 0)
tp->retrans_stamp = 0;
if (flag&FLAG_ECE)
- tcp_enter_cwr(sk);
+ tcp_enter_cwr(sk, 1);
if (inet_csk(sk)->icsk_ca_state != TCP_CA_CWR) {
int state = TCP_CA_Open;
* 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;
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, tp->fackets_out-tp->reordering, tp->high_seq);
+ tcp_mark_head_lost(sk, tp->fackets_out-tp->reordering, tp->high_seq);
NET_INC_STATS_BH(LINUX_MIB_TCPLOSS);
}
/* E. Check state exit conditions. State can be terminated
* when high_seq is ACKed. */
if (icsk->icsk_ca_state == TCP_CA_Open) {
- if (!sysctl_tcp_frto)
- BUG_TRAP(tp->retrans_out == 0);
+ BUG_TRAP(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, tp))
+ if (tcp_try_undo_recovery(sk))
return;
break;
break;
case TCP_CA_Disorder:
- tcp_try_undo_dsack(sk, tp);
+ tcp_try_undo_dsack(sk);
if (!tp->undo_marker ||
/* For SACK case do not Open to allow to undo
* catching for all duplicate ACKs. */
case TCP_CA_Recovery:
if (IsReno(tp))
tcp_reset_reno_sack(tp);
- if (tcp_try_undo_recovery(sk, tp))
+ if (tcp_try_undo_recovery(sk))
return;
tcp_complete_cwr(sk);
break;
} else {
int acked = prior_packets - tp->packets_out;
if (IsReno(tp))
- tcp_remove_reno_sacks(sk, tp, acked);
- is_dupack = tcp_try_undo_partial(sk, tp, acked);
+ tcp_remove_reno_sacks(sk, acked);
+ is_dupack = tcp_try_undo_partial(sk, acked);
}
break;
case TCP_CA_Loss:
if (flag&FLAG_DATA_ACKED)
icsk->icsk_retransmits = 0;
- if (!tcp_try_undo_loss(sk, tp)) {
+ if (!tcp_try_undo_loss(sk)) {
tcp_moderate_cwnd(tp);
tcp_xmit_retransmit_queue(sk);
return;
}
if (icsk->icsk_ca_state == TCP_CA_Disorder)
- tcp_try_undo_dsack(sk, tp);
+ tcp_try_undo_dsack(sk);
- if (!tcp_time_to_recover(sk, tp)) {
- tcp_try_to_open(sk, tp, flag);
+ if (!tcp_time_to_recover(sk)) {
+ tcp_try_to_open(sk, flag);
return;
}
tcp_set_ca_state(sk, TCP_CA_Recovery);
}
- if (is_dupack || tcp_head_timedout(sk, tp))
- tcp_update_scoreboard(sk, tp);
+ if (is_dupack || tcp_head_timedout(sk))
+ tcp_update_scoreboard(sk);
tcp_cwnd_down(sk);
tcp_xmit_retransmit_queue(sk);
}
tcp_ack_no_tstamp(sk, seq_rtt, flag);
}
-static void tcp_cong_avoid(struct sock *sk, u32 ack, u32 rtt,
+static void tcp_cong_avoid(struct sock *sk, u32 ack,
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);
+ icsk->icsk_ca_ops->cong_avoid(sk, ack, in_flight, good);
tcp_sk(sk)->snd_cwnd_stamp = tcp_time_stamp;
}
* RFC2988 recommends to restart timer to now+rto.
*/
-static 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 = tcp_sk(sk);
+
if (!tp->packets_out) {
inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS);
} else {
__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 u32 tcp_usrtt(const struct sk_buff *skb)
-{
- struct timeval tv, now;
-
- do_gettimeofday(&now);
- skb_get_timestamp(skb, &tv);
- return (now.tv_sec - tv.tv_sec) * 1000000 + (now.tv_usec - tv.tv_usec);
-}
-
/* Remove acknowledged frames from the retransmission queue. */
static int tcp_clean_rtx_queue(struct sock *sk, __s32 *seq_rtt_p)
{
struct sk_buff *skb;
__u32 now = tcp_time_stamp;
int acked = 0;
+ int prior_packets = tp->packets_out;
__s32 seq_rtt = -1;
- u32 pkts_acked = 0;
- void (*rtt_sample)(struct sock *sk, u32 usrtt)
- = icsk->icsk_ca_ops->rtt_sample;
+ ktime_t last_ackt = net_invalid_timestamp();
- while ((skb = skb_peek(&sk->sk_write_queue)) &&
- skb != sk->sk_send_head) {
- struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
+ while ((skb = tcp_write_queue_head(sk)) &&
+ skb != tcp_send_head(sk)) {
+ struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
__u8 sacked = scb->sacked;
/* If our packet is before the ack sequence we can
*/
if (!(scb->flags & TCPCB_FLAG_SYN)) {
acked |= FLAG_DATA_ACKED;
- ++pkts_acked;
} else {
acked |= FLAG_SYN_ACKED;
tp->retrans_stamp = 0;
if (sacked) {
if (sacked & TCPCB_RETRANS) {
- if(sacked & TCPCB_SACKED_RETRANS)
+ if (sacked & TCPCB_SACKED_RETRANS)
tp->retrans_out -= tcp_skb_pcount(skb);
acked |= FLAG_RETRANS_DATA_ACKED;
seq_rtt = -1;
} else if (seq_rtt < 0) {
seq_rtt = now - scb->when;
- if (rtt_sample)
- (*rtt_sample)(sk, tcp_usrtt(skb));
+ last_ackt = skb->tstamp;
}
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));
+ last_ackt = skb->tstamp;
}
tcp_dec_pcount_approx(&tp->fackets_out, skb);
tcp_packets_out_dec(tp, skb);
- __skb_unlink(skb, &sk->sk_write_queue);
+ tcp_unlink_write_queue(skb, sk);
sk_stream_free_skb(sk, skb);
clear_all_retrans_hints(tp);
}
if (acked&FLAG_ACKED) {
+ u32 pkts_acked = prior_packets - tp->packets_out;
+ const struct tcp_congestion_ops *ca_ops
+ = inet_csk(sk)->icsk_ca_ops;
+
tcp_ack_update_rtt(sk, acked, seq_rtt);
- tcp_ack_packets_out(sk, tp);
+ tcp_ack_packets_out(sk);
- if (icsk->icsk_ca_ops->pkts_acked)
- icsk->icsk_ca_ops->pkts_acked(sk, pkts_acked);
+ /* Is the ACK triggering packet unambiguous? */
+ if (acked & FLAG_RETRANS_DATA_ACKED)
+ last_ackt = net_invalid_timestamp();
+
+ if (ca_ops->pkts_acked)
+ ca_ops->pkts_acked(sk, pkts_acked, last_ackt);
}
#if FASTRETRANS_DEBUG > 0
/* Was it a usable window open? */
- if (!after(TCP_SKB_CB(sk->sk_send_head)->end_seq,
+ if (!after(TCP_SKB_CB(tcp_send_head(sk))->end_seq,
tp->snd_una + tp->snd_wnd)) {
icsk->icsk_backoff = 0;
inet_csk_clear_xmit_timer(sk, ICSK_TIME_PROBE0);
* 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 tcp_sock *tp,
- struct sk_buff *skb, u32 ack, u32 ack_seq)
+static int tcp_ack_update_window(struct sock *sk, struct sk_buff *skb, u32 ack,
+ u32 ack_seq)
{
+ struct tcp_sock *tp = tcp_sk(sk);
int flag = 0;
- u32 nwin = ntohs(skb->h.th->window);
+ u32 nwin = ntohs(tcp_hdr(skb)->window);
- if (likely(!skb->h.th->syn))
+ if (likely(!tcp_hdr(skb)->syn))
nwin <<= tp->rx_opt.snd_wscale;
if (tcp_may_update_window(tp, ack, ack_seq, nwin)) {
* fast path is recovered for sending TCP.
*/
tp->pred_flags = 0;
- tcp_fast_path_check(sk, tp);
+ tcp_fast_path_check(sk);
if (nwin > tp->max_window) {
tp->max_window = nwin;
return flag;
}
-static void tcp_process_frto(struct sock *sk, u32 prior_snd_una)
+/* 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)
+{
+ tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
+ tp->snd_cwnd_cnt = 0;
+ TCP_ECN_queue_cwr(tp);
+ tcp_moderate_cwnd(tp);
+}
+
+/* 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)
+{
+ tcp_enter_cwr(sk, 0);
+}
+
+static void tcp_undo_spur_to_response(struct sock *sk, int flag)
+{
+ if (flag&FLAG_ECE)
+ tcp_ratehalving_spur_to_response(sk);
+ else
+ tcp_undo_cwr(sk, 1);
+}
+
+/* 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
+ */
+static int tcp_process_frto(struct sock *sk, u32 prior_snd_una, int flag)
{
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
- * go-back-N slow start
+
+ /* Duplicate the behavior from Loss state (fastretrans_alert) */
+ if (flag&FLAG_DATA_ACKED)
+ inet_csk(sk)->icsk_retransmits = 0;
+
+ if (!before(tp->snd_una, tp->frto_highmark)) {
+ tcp_enter_frto_loss(sk, (tp->frto_counter == 1 ? 2 : 3), flag);
+ return 1;
+ }
+
+ if (!IsSackFrto() || IsReno(tp)) {
+ /* RFC4138 shortcoming in step 2; should also have case c):
+ * ACK isn't duplicate nor advances window, e.g., opposite dir
+ * data, winupdate
*/
- tcp_enter_frto_loss(sk);
- return;
+ if ((tp->snd_una == prior_snd_una) && (flag&FLAG_NOT_DUP) &&
+ !(flag&FLAG_FORWARD_PROGRESS))
+ 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;
+ }
}
if (tp->frto_counter == 1) {
- /* First ACK after RTO advances the window: allow two new
- * segments out.
- */
+ /* Sending of the next skb must be allowed or no FRTO */
+ if (!tcp_send_head(sk) ||
+ after(TCP_SKB_CB(tcp_send_head(sk))->end_seq,
+ tp->snd_una + tp->snd_wnd)) {
+ tcp_enter_frto_loss(sk, (tp->frto_counter == 1 ? 2 : 3),
+ flag);
+ return 1;
+ }
+
tp->snd_cwnd = tcp_packets_in_flight(tp) + 2;
+ tp->frto_counter = 2;
+ return 1;
} else {
- /* Also the second ACK after RTO advances the window.
- * The RTO was likely spurious. Reduce cwnd and continue
- * in congestion avoidance
- */
- tp->snd_cwnd = min(tp->snd_cwnd, tp->snd_ssthresh);
- tcp_moderate_cwnd(tp);
+ 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;
+ }
+ tp->frto_counter = 0;
}
-
- /* F-RTO affects on two new ACKs following RTO.
- * At latest on third ACK the TCP behavior is back to normal.
- */
- tp->frto_counter = (tp->frto_counter + 1) % 3;
+ return 0;
}
/* This routine deals with incoming acks, but not outgoing ones. */
u32 prior_in_flight;
s32 seq_rtt;
int prior_packets;
+ int frto_cwnd = 0;
/* If the ack is newer than sent or older than previous acks
* then we can probably ignore it.
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.
else
NET_INC_STATS_BH(LINUX_MIB_TCPPUREACKS);
- flag |= tcp_ack_update_window(sk, tp, skb, ack, ack_seq);
+ 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);
- if (TCP_ECN_rcv_ecn_echo(tp, skb->h.th))
+ if (TCP_ECN_rcv_ecn_echo(tp, tcp_hdr(skb)))
flag |= FLAG_ECE;
tcp_ca_event(sk, CA_EVENT_SLOW_ACK);
flag |= tcp_clean_rtx_queue(sk, &seq_rtt);
if (tp->frto_counter)
- tcp_process_frto(sk, prior_snd_una);
+ frto_cwnd = tcp_process_frto(sk, prior_snd_una, flag);
if (tcp_ack_is_dubious(sk, flag)) {
/* Advance CWND, if state allows this. */
- if ((flag & FLAG_DATA_ACKED) && tcp_may_raise_cwnd(sk, flag))
- tcp_cong_avoid(sk, ack, seq_rtt, prior_in_flight, 0);
+ if ((flag & FLAG_DATA_ACKED) && !frto_cwnd &&
+ tcp_may_raise_cwnd(sk, flag))
+ tcp_cong_avoid(sk, ack, prior_in_flight, 0);
tcp_fastretrans_alert(sk, prior_snd_una, prior_packets, flag);
} else {
- if ((flag & FLAG_DATA_ACKED))
- tcp_cong_avoid(sk, ack, seq_rtt, prior_in_flight, 1);
+ if ((flag & FLAG_DATA_ACKED) && !frto_cwnd)
+ tcp_cong_avoid(sk, ack, prior_in_flight, 1);
}
if ((flag & FLAG_FORWARD_PROGRESS) || !(flag&FLAG_NOT_DUP))
* being used to time the probes, and is probably far higher than
* it needs to be for normal retransmission.
*/
- if (sk->sk_send_head)
+ if (tcp_send_head(sk))
tcp_ack_probe(sk);
return 1;
void tcp_parse_options(struct sk_buff *skb, struct tcp_options_received *opt_rx, int estab)
{
unsigned char *ptr;
- struct tcphdr *th = skb->h.th;
+ struct tcphdr *th = tcp_hdr(skb);
int length=(th->doff*4)-sizeof(struct tcphdr);
ptr = (unsigned char *)(th + 1);
opt_rx->saw_tstamp = 0;
- while(length>0) {
- int opcode=*ptr++;
+ while (length > 0) {
+ 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));
+ if (opsize==TCPOLEN_MSS && th->syn && !estab) {
+ 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;
}
break;
case TCPOPT_WINDOW:
- if(opsize==TCPOLEN_WINDOW && th->syn && !estab)
+ if (opsize==TCPOLEN_WINDOW && th->syn && !estab)
if (sysctl_tcp_window_scaling) {
__u8 snd_wscale = *(__u8 *) ptr;
opt_rx->wscale_ok = 1;
if (snd_wscale > 14) {
- if(net_ratelimit())
+ if (net_ratelimit())
printk(KERN_INFO "tcp_parse_options: Illegal window "
"scaling value %d >14 received.\n",
snd_wscale);
}
break;
case TCPOPT_TIMESTAMP:
- if(opsize==TCPOLEN_TIMESTAMP) {
+ if (opsize==TCPOLEN_TIMESTAMP) {
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;
case TCPOPT_SACK_PERM:
- if(opsize==TCPOLEN_SACK_PERM && th->syn && !estab) {
+ if (opsize==TCPOLEN_SACK_PERM && th->syn && !estab) {
if (sysctl_tcp_sack) {
opt_rx->sack_ok = 1;
tcp_sack_reset(opt_rx);
break;
case TCPOPT_SACK:
- if((opsize >= (TCPOLEN_SACK_BASE + TCPOLEN_SACK_PERBLOCK)) &&
+ if ((opsize >= (TCPOLEN_SACK_BASE + TCPOLEN_SACK_PERBLOCK)) &&
!((opsize - TCPOLEN_SACK_BASE) % TCPOLEN_SACK_PERBLOCK) &&
opt_rx->sack_ok) {
TCP_SKB_CB(skb)->sacked = (ptr - 2) - (unsigned char *)th;
}
- };
- ptr+=opsize-2;
- length-=opsize;
- };
+ break;
+#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;
+ }
}
}
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 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 = xtime.tv_sec;
+ tp->rx_opt.ts_recent_stamp = get_seconds();
}
static inline void tcp_replace_ts_recent(struct tcp_sock *tp, u32 seq)
* Not only, also it occurs for expired timestamps.
*/
- if((s32)(tp->rx_opt.rcv_tsval - tp->rx_opt.ts_recent) >= 0 ||
- xtime.tv_sec >= tp->rx_opt.ts_recent_stamp + TCP_PAWS_24DAYS)
+ if ((s32)(tp->rx_opt.rcv_tsval - tp->rx_opt.ts_recent) >= 0 ||
+ get_seconds() >= tp->rx_opt.ts_recent_stamp + TCP_PAWS_24DAYS)
tcp_store_ts_recent(tp);
}
}
static int tcp_disordered_ack(const struct sock *sk, const struct sk_buff *skb)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct tcphdr *th = skb->h.th;
+ struct tcphdr *th = tcp_hdr(skb);
u32 seq = TCP_SKB_CB(skb)->seq;
u32 ack = TCP_SKB_CB(skb)->ack_seq;
{
const struct tcp_sock *tp = tcp_sk(sk);
return ((s32)(tp->rx_opt.ts_recent - tp->rx_opt.rcv_tsval) > TCP_PAWS_WINDOW &&
- xtime.tv_sec < tp->rx_opt.ts_recent_stamp + TCP_PAWS_24DAYS &&
+ get_seconds() < tp->rx_opt.ts_recent_stamp + TCP_PAWS_24DAYS &&
!tcp_disordered_ack(sk, skb));
}
printk(KERN_ERR "%s: Impossible, sk->sk_state=%d\n",
__FUNCTION__, sk->sk_state);
break;
- };
+ }
/* It _is_ possible, that we have something out-of-order _after_ FIN.
* Probably, we should reset in this case. For now drop them.
*/
tp->rx_opt.num_sacks--;
tp->rx_opt.eff_sacks = min(tp->rx_opt.num_sacks + tp->rx_opt.dsack, 4 - tp->rx_opt.tstamp_ok);
- for(i=this_sack; i < tp->rx_opt.num_sacks; i++)
+ for (i=this_sack; i < tp->rx_opt.num_sacks; i++)
sp[i] = sp[i+1];
continue;
}
tp->rx_opt.num_sacks--;
sp--;
}
- for(; this_sack > 0; this_sack--, sp--)
+ for (; this_sack > 0; this_sack--, sp--)
*sp = *(sp-1);
new_sack:
return;
}
- for(this_sack = 0; this_sack < num_sacks; ) {
+ for (this_sack = 0; this_sack < num_sacks; ) {
/* Check if the start of the sack is covered by RCV.NXT. */
if (!before(tp->rcv_nxt, sp->start_seq)) {
int i;
__skb_unlink(skb, &tp->out_of_order_queue);
__skb_queue_tail(&sk->sk_receive_queue, skb);
tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
- if(skb->h.th->fin)
- tcp_fin(skb, sk, skb->h.th);
+ if (tcp_hdr(skb)->fin)
+ tcp_fin(skb, sk, tcp_hdr(skb));
}
}
static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
{
- struct tcphdr *th = skb->h.th;
+ struct tcphdr *th = tcp_hdr(skb);
struct tcp_sock *tp = tcp_sk(sk);
int eaten = -1;
__skb_queue_tail(&sk->sk_receive_queue, skb);
}
tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
- if(skb->len)
- tcp_event_data_recv(sk, tp, skb);
- if(th->fin)
+ if (skb->len)
+ tcp_event_data_recv(sk, skb);
+ if (th->fin)
tcp_fin(skb, sk, th);
if (!skb_queue_empty(&tp->out_of_order_queue)) {
if (tp->rx_opt.num_sacks)
tcp_sack_remove(tp);
- tcp_fast_path_check(sk, tp);
+ tcp_fast_path_check(sk);
if (eaten > 0)
__kfree_skb(skb);
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 &&
* - bloated or contains data before "start" or
* overlaps to the next one.
*/
- if (!skb->h.th->syn && !skb->h.th->fin &&
+ if (!tcp_hdr(skb)->syn && !tcp_hdr(skb)->fin &&
(tcp_win_from_space(skb->truesize) > skb->len ||
before(TCP_SKB_CB(skb)->seq, start) ||
(skb->next != tail &&
start = TCP_SKB_CB(skb)->end_seq;
skb = skb->next;
}
- if (skb == tail || skb->h.th->syn || skb->h.th->fin)
+ if (skb == tail || tcp_hdr(skb)->syn || tcp_hdr(skb)->fin)
return;
while (before(start, end)) {
nskb = alloc_skb(copy+header, GFP_ATOMIC);
if (!nskb)
return;
+
+ skb_set_mac_header(nskb, skb_mac_header(skb) - skb->head);
+ skb_set_network_header(nskb, (skb_network_header(skb) -
+ skb->head));
+ skb_set_transport_header(nskb, (skb_transport_header(skb) -
+ skb->head));
skb_reserve(nskb, header);
memcpy(nskb->head, skb->head, header);
- nskb->nh.raw = nskb->head + (skb->nh.raw-skb->head);
- nskb->h.raw = nskb->head + (skb->h.raw-skb->head);
- nskb->mac.raw = nskb->head + (skb->mac.raw-skb->head);
memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start;
__skb_insert(nskb, skb->prev, skb, list);
__kfree_skb(skb);
NET_INC_STATS_BH(LINUX_MIB_TCPRCVCOLLAPSED);
skb = next;
- if (skb == tail || skb->h.th->syn || skb->h.th->fin)
+ if (skb == tail ||
+ tcp_hdr(skb)->syn ||
+ tcp_hdr(skb)->fin)
return;
}
}
*/
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);
NET_INC_STATS_BH(LINUX_MIB_PRUNECALLED);
if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)
- tcp_clamp_window(sk, tp);
+ tcp_clamp_window(sk);
else if (tcp_memory_pressure)
tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U * tp->advmss);
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 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 = tcp_sk(sk);
+
/* 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) +
+ 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),
demanded = max_t(unsigned int, tp->snd_cwnd,
tp->reordering + 1);
}
}
-static inline void tcp_data_snd_check(struct sock *sk, struct tcp_sock *tp)
+static inline void tcp_data_snd_check(struct sock *sk)
{
- tcp_push_pending_frames(sk, tp);
+ tcp_push_pending_frames(sk);
tcp_check_space(sk);
}
* 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))
int err;
local_bh_enable();
- if (skb->ip_summed==CHECKSUM_UNNECESSARY)
+ if (skb_csum_unnecessary(skb))
err = skb_copy_datagram_iovec(skb, hlen, tp->ucopy.iov, chunk);
else
err = skb_copy_and_csum_datagram_iovec(skb, hlen,
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();
static inline int tcp_checksum_complete_user(struct sock *sk, struct sk_buff *skb)
{
- return skb->ip_summed != CHECKSUM_UNNECESSARY &&
+ return !skb_csum_unnecessary(skb) &&
__tcp_checksum_complete_user(sk, skb);
}
+#ifdef CONFIG_NET_DMA
+static int tcp_dma_try_early_copy(struct sock *sk, struct sk_buff *skb, int hlen)
+{
+ 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_csum_unnecessary(skb)) {
+
+ 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(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);
+ }
+ } 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);
- tcp_data_snd_check(sk, tp);
+ __kfree_skb(skb);
+ tcp_data_snd_check(sk);
return 0;
} else { /* Header too small */
TCP_INC_STATS_BH(TCP_MIB_INERRS);
}
} 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))
tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq;
}
- tcp_event_data_recv(sk, tp, skb);
+ tcp_event_data_recv(sk, skb);
if (TCP_SKB_CB(skb)->ack_seq != tp->snd_una) {
/* Well, only one small jumplet in fast path... */
tcp_ack(sk, skb, FLAG_DATA);
- tcp_data_snd_check(sk, tp);
+ tcp_data_snd_check(sk);
if (!inet_csk_ack_scheduled(sk))
goto no_ack;
}
__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
goto discard;
}
- if(th->rst) {
+ if (th->rst) {
tcp_reset(sk);
goto discard;
}
}
step5:
- if(th->ack)
+ if (th->ack)
tcp_ack(sk, skb, FLAG_SLOWPATH);
tcp_rcv_rtt_measure_ts(sk, skb);
/* step 7: process the segment text */
tcp_data_queue(sk, skb);
- tcp_data_snd_check(sk, tp);
+ tcp_data_snd_check(sk);
tcp_ack_snd_check(sk);
return 0;
*/
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);
* 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. */
icsk->icsk_af_ops->rebuild_header(sk);
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_mtup_init(sk);
tcp_sync_mss(sk, icsk->icsk_pmtu_cookie);
/*
* 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)
{
goto discard;
case TCP_LISTEN:
- if(th->ack)
+ if (th->ack)
return 1;
- if(th->rst)
+ if (th->rst)
goto discard;
- if(th->syn) {
+ if (th->syn) {
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;
/* Do step6 onward by hand. */
tcp_urg(sk, skb, th);
__kfree_skb(skb);
- tcp_data_snd_check(sk, tp);
+ tcp_data_snd_check(sk);
return 0;
}
}
/* step 2: check RST bit */
- if(th->rst) {
+ if (th->rst) {
tcp_reset(sk);
goto discard;
}
if (th->ack) {
int acceptable = tcp_ack(sk, skb, FLAG_SLOWPATH);
- switch(sk->sk_state) {
+ switch (sk->sk_state) {
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);
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_data could move socket to TIME-WAIT */
if (sk->sk_state != TCP_CLOSE) {
- tcp_data_snd_check(sk, tp);
+ tcp_data_snd_check(sk);
tcp_ack_snd_check(sk);
}
- if (!queued) {
+ if (!queued) {
discard:
__kfree_skb(skb);
}
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff