}
}
+/* This must be called before lost_out is incremented */
+static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb)
+{
+ if ((tp->retransmit_skb_hint == NULL) ||
+ before(TCP_SKB_CB(skb)->seq,
+ TCP_SKB_CB(tp->retransmit_skb_hint)->seq))
+ tp->retransmit_skb_hint = skb;
+
+ if (!tp->lost_out ||
+ after(TCP_SKB_CB(skb)->end_seq, tp->retransmit_high))
+ tp->retransmit_high = TCP_SKB_CB(skb)->end_seq;
+}
+
+static void tcp_skb_mark_lost(struct tcp_sock *tp, struct sk_buff *skb)
+{
+ if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) {
+ tcp_verify_retransmit_hint(tp, skb);
+
+ tp->lost_out += tcp_skb_pcount(skb);
+ TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
+ }
+}
+
+void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, struct sk_buff *skb)
+{
+ tcp_verify_retransmit_hint(tp, skb);
+
+ if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) {
+ tp->lost_out += tcp_skb_pcount(skb);
+ TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
+ }
+}
+
/* This procedure tags the retransmission queue when SACKs arrive.
*
* We have three tag bits: SACKED(S), RETRANS(R) and LOST(L).
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
tp->retrans_out -= tcp_skb_pcount(skb);
- /* clear lost hint */
- tp->retransmit_skb_hint = NULL;
-
- if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) {
- tp->lost_out += tcp_skb_pcount(skb);
- TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
- }
+ tcp_skb_mark_lost_uncond_verify(tp, skb);
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT);
} else {
if (before(ack_seq, new_low_seq))
~(TCPCB_LOST|TCPCB_SACKED_RETRANS);
tp->lost_out -= tcp_skb_pcount(skb);
tp->retrans_out -= tcp_skb_pcount(skb);
-
- /* clear lost hint */
- tp->retransmit_skb_hint = NULL;
}
} else {
if (!(sacked & TCPCB_RETRANS)) {
if (sacked & TCPCB_LOST) {
TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
tp->lost_out -= tcp_skb_pcount(skb);
-
- /* clear lost hint */
- tp->retransmit_skb_hint = NULL;
}
}
if (dup_sack && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS)) {
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
tp->retrans_out -= tcp_skb_pcount(skb);
- tp->retransmit_skb_hint = NULL;
}
return flag;
out:
#if FASTRETRANS_DEBUG > 0
- BUG_TRAP((int)tp->sacked_out >= 0);
- BUG_TRAP((int)tp->lost_out >= 0);
- BUG_TRAP((int)tp->retrans_out >= 0);
- BUG_TRAP((int)tcp_packets_in_flight(tp) >= 0);
+ WARN_ON((int)tp->sacked_out < 0);
+ WARN_ON((int)tp->lost_out < 0);
+ WARN_ON((int)tp->retrans_out < 0);
+ WARN_ON((int)tcp_packets_in_flight(tp) < 0);
#endif
return flag;
}
if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) {
TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
tp->lost_out += tcp_skb_pcount(skb);
+ tp->retransmit_high = TCP_SKB_CB(skb)->end_seq;
}
}
tcp_verify_left_out(tp);
tp->high_seq = tp->snd_nxt;
TCP_ECN_queue_cwr(tp);
- tcp_clear_retrans_hints_partial(tp);
+ tcp_clear_all_retrans_hints(tp);
}
static void tcp_clear_retrans_partial(struct tcp_sock *tp)
/* Push undo marker, if it was plain RTO and nothing
* was retransmitted. */
tp->undo_marker = tp->snd_una;
- tcp_clear_retrans_hints_partial(tp);
} else {
tp->sacked_out = 0;
tp->fackets_out = 0;
- tcp_clear_all_retrans_hints(tp);
}
+ tcp_clear_all_retrans_hints(tp);
tcp_for_write_queue(skb, sk) {
if (skb == tcp_send_head(sk))
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_ACKED;
TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
tp->lost_out += tcp_skb_pcount(skb);
+ tp->retransmit_high = TCP_SKB_CB(skb)->end_seq;
}
}
tcp_verify_left_out(tp);
return 0;
}
-/* RFC: This is from the original, I doubt that this is necessary at all:
- * clear xmit_retrans hint if seq of this skb is beyond hint. How could we
- * retransmitted past LOST markings in the first place? I'm not fully sure
- * about undo and end of connection cases, which can cause R without L?
- */
-static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb)
-{
- 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;
-}
-
/* Mark head of queue up as lost. With RFC3517 SACK, the packets is
* is against sacked "cnt", otherwise it's against facked "cnt"
*/
int err;
unsigned int mss;
- BUG_TRAP(packets <= tp->packets_out);
+ WARN_ON(packets > tp->packets_out);
if (tp->lost_skb_hint) {
skb = tp->lost_skb_hint;
cnt = tp->lost_cnt_hint;
cnt = packets;
}
- if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_SACKED_ACKED|TCPCB_LOST))) {
- TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
- tp->lost_out += tcp_skb_pcount(skb);
- tcp_verify_retransmit_hint(tp, skb);
- }
+ tcp_skb_mark_lost(tp, skb);
}
tcp_verify_left_out(tp);
}
if (!tcp_skb_timedout(sk, skb))
break;
- if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_SACKED_ACKED|TCPCB_LOST))) {
- TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
- tp->lost_out += tcp_skb_pcount(skb);
- tcp_verify_retransmit_hint(tp, skb);
- }
+ tcp_skb_mark_lost(tp, skb);
}
tp->scoreboard_skb_hint = skb;
}
tcp_moderate_cwnd(tp);
tp->snd_cwnd_stamp = tcp_time_stamp;
-
- /* There is something screwy going on with the retrans hints after
- an undo */
- tcp_clear_all_retrans_hints(tp);
}
static inline int tcp_may_undo(struct tcp_sock *tp)
/* E. Check state exit conditions. State can be terminated
* when high_seq is ACKed. */
if (icsk->icsk_ca_state == TCP_CA_Open) {
- BUG_TRAP(tp->retrans_out == 0);
+ WARN_ON(tp->retrans_out != 0);
tp->retrans_stamp = 0;
} else if (!before(tp->snd_una, tp->high_seq)) {
switch (icsk->icsk_ca_state) {
}
#if FASTRETRANS_DEBUG > 0
- BUG_TRAP((int)tp->sacked_out >= 0);
- BUG_TRAP((int)tp->lost_out >= 0);
- BUG_TRAP((int)tp->retrans_out >= 0);
+ WARN_ON((int)tp->sacked_out < 0);
+ WARN_ON((int)tp->lost_out < 0);
+ WARN_ON((int)tp->retrans_out < 0);
if (!tp->packets_out && tcp_is_sack(tp)) {
icsk = inet_csk(sk);
if (tp->lost_out) {
* log. Something worked...
*/
sk->sk_err_soft = 0;
+ icsk->icsk_probes_out = 0;
tp->rcv_tstamp = tcp_time_stamp;
prior_packets = tp->packets_out;
if (!prior_packets)
return 1;
no_queue:
- icsk->icsk_probes_out = 0;
-
/* If this ack opens up a zero window, clear backoff. It was
* being used to time the probes, and is probably far higher than
* it needs to be for normal retransmission.
}
}
+static int tcp_parse_aligned_timestamp(struct tcp_sock *tp, struct tcphdr *th)
+{
+ __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;
+ tp->rx_opt.rcv_tsval = ntohl(*ptr);
+ ++ptr;
+ tp->rx_opt.rcv_tsecr = ntohl(*ptr);
+ return 1;
+ }
+ return 0;
+}
+
/* Fast parse options. This hopes to only see timestamps.
* If it is wrong it falls back on tcp_parse_options().
*/
return 0;
} else if (tp->rx_opt.tstamp_ok &&
th->doff == (sizeof(struct tcphdr)>>2)+(TCPOLEN_TSTAMP_ALIGNED>>2)) {
- __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;
- tp->rx_opt.rcv_tsval = ntohl(*ptr);
- ++ptr;
- tp->rx_opt.rcv_tsecr = ntohl(*ptr);
+ if (tcp_parse_aligned_timestamp(tp, th))
return 1;
- }
}
tcp_parse_options(skb, &tp->rx_opt, 1);
return 1;
int i;
/* RCV.NXT must cover all the block! */
- BUG_TRAP(!before(tp->rcv_nxt, sp->end_seq));
+ WARN_ON(before(tp->rcv_nxt, sp->end_seq));
/* Zap this SACK, by moving forward any other SACKS. */
for (i=this_sack+1; i < num_sacks; i++)
}
}
+static struct sk_buff *tcp_collapse_one(struct sock *sk, struct sk_buff *skb,
+ struct sk_buff_head *list)
+{
+ struct sk_buff *next = skb->next;
+
+ __skb_unlink(skb, list);
+ __kfree_skb(skb);
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
+
+ return next;
+}
+
/* Collapse contiguous sequence of skbs head..tail with
* sequence numbers start..end.
* Segments with FIN/SYN are not collapsed (only because this
for (skb = head; skb != tail;) {
/* No new bits? It is possible on ofo queue. */
if (!before(start, TCP_SKB_CB(skb)->end_seq)) {
- struct sk_buff *next = skb->next;
- __skb_unlink(skb, list);
- __kfree_skb(skb);
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
- skb = next;
+ skb = tcp_collapse_one(sk, skb, list);
continue;
}
start += size;
}
if (!before(start, TCP_SKB_CB(skb)->end_seq)) {
- struct sk_buff *next = skb->next;
- __skb_unlink(skb, list);
- __kfree_skb(skb);
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
- skb = next;
+ skb = tcp_collapse_one(sk, skb, list);
if (skb == tail ||
tcp_hdr(skb)->syn ||
tcp_hdr(skb)->fin)
}
#endif /* CONFIG_NET_DMA */
+/* Does PAWS and seqno based validation of an incoming segment, flags will
+ * play significant role here.
+ */
+static int tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
+ struct tcphdr *th, int syn_inerr)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ /* RFC1323: H1. Apply PAWS check first. */
+ if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
+ tcp_paws_discard(sk, skb)) {
+ if (!th->rst) {
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
+ tcp_send_dupack(sk, skb);
+ goto discard;
+ }
+ /* Reset is accepted even if it did not pass PAWS. */
+ }
+
+ /* Step 1: check sequence number */
+ if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
+ /* RFC793, page 37: "In all states except SYN-SENT, all reset
+ * (RST) segments are validated by checking their SEQ-fields."
+ * And page 69: "If an incoming segment is not acceptable,
+ * an acknowledgment should be sent in reply (unless the RST
+ * bit is set, if so drop the segment and return)".
+ */
+ if (!th->rst)
+ tcp_send_dupack(sk, skb);
+ goto discard;
+ }
+
+ /* Step 2: check RST bit */
+ if (th->rst) {
+ tcp_reset(sk);
+ goto discard;
+ }
+
+ /* ts_recent update must be made after we are sure that the packet
+ * is in window.
+ */
+ tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
+
+ /* step 3: check security and precedence [ignored] */
+
+ /* step 4: Check for a SYN in window. */
+ if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
+ if (syn_inerr)
+ TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
+ tcp_reset(sk);
+ return -1;
+ }
+
+ return 1;
+
+discard:
+ __kfree_skb(skb);
+ return 0;
+}
+
/*
* TCP receive function for the ESTABLISHED state.
*
struct tcphdr *th, unsigned len)
{
struct tcp_sock *tp = tcp_sk(sk);
+ int res;
/*
* Header prediction.
/* Check timestamp */
if (tcp_header_len == sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) {
- __be32 *ptr = (__be32 *)(th + 1);
-
/* No? Slow path! */
- if (*ptr != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
- | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP))
+ if (!tcp_parse_aligned_timestamp(tp, th))
goto slow_path;
- tp->rx_opt.saw_tstamp = 1;
- ++ptr;
- tp->rx_opt.rcv_tsval = ntohl(*ptr);
- ++ptr;
- tp->rx_opt.rcv_tsecr = ntohl(*ptr);
-
/* If PAWS failed, check it more carefully in slow path */
if ((s32)(tp->rx_opt.rcv_tsval - tp->rx_opt.ts_recent) < 0)
goto slow_path;
if (len < (th->doff << 2) || tcp_checksum_complete_user(sk, skb))
goto csum_error;
- /*
- * RFC1323: H1. Apply PAWS check first.
- */
- if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
- tcp_paws_discard(sk, skb)) {
- if (!th->rst) {
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
- tcp_send_dupack(sk, skb);
- goto discard;
- }
- /* Resets are accepted even if PAWS failed.
-
- ts_recent update must be made after we are sure
- that the packet is in window.
- */
- }
-
/*
* Standard slow path.
*/
- if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
- /* RFC793, page 37: "In all states except SYN-SENT, all reset
- * (RST) segments are validated by checking their SEQ-fields."
- * And page 69: "If an incoming segment is not acceptable,
- * an acknowledgment should be sent in reply (unless the RST bit
- * is set, if so drop the segment and return)".
- */
- if (!th->rst)
- tcp_send_dupack(sk, skb);
- goto discard;
- }
-
- if (th->rst) {
- tcp_reset(sk);
- goto discard;
- }
-
- tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
-
- if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
- TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
- tcp_reset(sk);
- return 1;
- }
+ res = tcp_validate_incoming(sk, skb, th, 1);
+ if (res <= 0)
+ return -res;
step5:
if (th->ack)
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
int queued = 0;
+ int res;
tp->rx_opt.saw_tstamp = 0;
return 0;
}
- if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
- tcp_paws_discard(sk, skb)) {
- if (!th->rst) {
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
- tcp_send_dupack(sk, skb);
- goto discard;
- }
- /* Reset is accepted even if it did not pass PAWS. */
- }
-
- /* step 1: check sequence number */
- if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
- if (!th->rst)
- tcp_send_dupack(sk, skb);
- goto discard;
- }
-
- /* step 2: check RST bit */
- if (th->rst) {
- tcp_reset(sk);
- goto discard;
- }
-
- tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
-
- /* step 3: check security and precedence [ignored] */
-
- /* step 4:
- *
- * Check for a SYN in window.
- */
- if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
- tcp_reset(sk);
- return 1;
- }
+ res = tcp_validate_incoming(sk, skb, th, 0);
+ if (res <= 0)
+ return -res;
/* step 5: check the ACK field */
if (th->ack) {
git.openpandora.org / pandora-kernel.git / blobdiff