static void update_send_head(struct sock *sk, struct tcp_sock *tp,
struct sk_buff *skb)
{
- sk->sk_send_head = skb->next;
- if (sk->sk_send_head == (struct sk_buff *)&sk->sk_write_queue)
- sk->sk_send_head = NULL;
+ tcp_advance_send_head(sk, skb);
tp->snd_nxt = TCP_SKB_CB(skb)->end_seq;
tcp_packets_out_inc(sk, tp, skb);
}
(*rcv_wscale) = 0;
if (wscale_ok) {
/* Set window scaling on max possible window
- * See RFC1323 for an explanation of the limit to 14
+ * See RFC1323 for an explanation of the limit to 14
*/
space = max_t(u32, sysctl_tcp_rmem[2], sysctl_rmem_max);
space = min_t(u32, space, *window_clamp);
u32 new_win = __tcp_select_window(sk);
/* Never shrink the offered window */
- if(new_win < cur_win) {
+ if (new_win < cur_win) {
/* Danger Will Robinson!
* Don't update rcv_wup/rcv_wnd here or else
* we will not be able to advertise a zero
return new_win;
}
-static void tcp_build_and_update_options(__u32 *ptr, struct tcp_sock *tp,
- __u32 tstamp)
+static void tcp_build_and_update_options(__be32 *ptr, struct tcp_sock *tp,
+ __u32 tstamp, __u8 **md5_hash)
{
if (tp->rx_opt.tstamp_ok) {
- *ptr++ = __constant_htonl((TCPOPT_NOP << 24) |
- (TCPOPT_NOP << 16) |
- (TCPOPT_TIMESTAMP << 8) |
- TCPOLEN_TIMESTAMP);
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
+ TCPOLEN_TIMESTAMP);
*ptr++ = htonl(tstamp);
*ptr++ = htonl(tp->rx_opt.ts_recent);
}
(TCPOPT_SACK << 8) |
(TCPOLEN_SACK_BASE + (tp->rx_opt.eff_sacks *
TCPOLEN_SACK_PERBLOCK)));
- for(this_sack = 0; this_sack < tp->rx_opt.eff_sacks; this_sack++) {
+
+ for (this_sack = 0; this_sack < tp->rx_opt.eff_sacks; this_sack++) {
*ptr++ = htonl(sp[this_sack].start_seq);
*ptr++ = htonl(sp[this_sack].end_seq);
}
+
if (tp->rx_opt.dsack) {
tp->rx_opt.dsack = 0;
tp->rx_opt.eff_sacks--;
}
}
+#ifdef CONFIG_TCP_MD5SIG
+ if (md5_hash) {
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_MD5SIG << 8) |
+ TCPOLEN_MD5SIG);
+ *md5_hash = (__u8 *)ptr;
+ }
+#endif
}
/* Construct a tcp options header for a SYN or SYN_ACK packet.
* If this is every changed make sure to change the definition of
* MAX_SYN_SIZE to match the new maximum number of options that you
* can generate.
+ *
+ * Note - that with the RFC2385 TCP option, we make room for the
+ * 16 byte MD5 hash. This will be filled in later, so the pointer for the
+ * location to be filled is passed back up.
*/
-static void tcp_syn_build_options(__u32 *ptr, int mss, int ts, int sack,
+static void tcp_syn_build_options(__be32 *ptr, int mss, int ts, int sack,
int offer_wscale, int wscale, __u32 tstamp,
- __u32 ts_recent)
+ __u32 ts_recent, __u8 **md5_hash)
{
/* We always get an MSS option.
* The option bytes which will be seen in normal data
*/
*ptr++ = htonl((TCPOPT_MSS << 24) | (TCPOLEN_MSS << 16) | mss);
if (ts) {
- if(sack)
- *ptr++ = __constant_htonl((TCPOPT_SACK_PERM << 24) | (TCPOLEN_SACK_PERM << 16) |
- (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP);
+ if (sack)
+ *ptr++ = htonl((TCPOPT_SACK_PERM << 24) |
+ (TCPOLEN_SACK_PERM << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
+ TCPOLEN_TIMESTAMP);
else
- *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
- (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP);
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_TIMESTAMP << 8) |
+ TCPOLEN_TIMESTAMP);
*ptr++ = htonl(tstamp); /* TSVAL */
*ptr++ = htonl(ts_recent); /* TSECR */
- } else if(sack)
- *ptr++ = __constant_htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16) |
- (TCPOPT_SACK_PERM << 8) | TCPOLEN_SACK_PERM);
+ } else if (sack)
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_SACK_PERM << 8) |
+ TCPOLEN_SACK_PERM);
if (offer_wscale)
- *ptr++ = htonl((TCPOPT_NOP << 24) | (TCPOPT_WINDOW << 16) | (TCPOLEN_WINDOW << 8) | (wscale));
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_WINDOW << 16) |
+ (TCPOLEN_WINDOW << 8) |
+ (wscale));
+#ifdef CONFIG_TCP_MD5SIG
+ /*
+ * If MD5 is enabled, then we set the option, and include the size
+ * (always 18). The actual MD5 hash is added just before the
+ * packet is sent.
+ */
+ if (md5_hash) {
+ *ptr++ = htonl((TCPOPT_NOP << 24) |
+ (TCPOPT_NOP << 16) |
+ (TCPOPT_MD5SIG << 8) |
+ TCPOLEN_MD5SIG);
+ *md5_hash = (__u8 *) ptr;
+ }
+#endif
}
/* This routine actually transmits TCP packets queued in by
struct tcp_sock *tp;
struct tcp_skb_cb *tcb;
int tcp_header_size;
+#ifdef CONFIG_TCP_MD5SIG
+ struct tcp_md5sig_key *md5;
+ __u8 *md5_hash_location;
+#endif
struct tcphdr *th;
int sysctl_flags;
int err;
sysctl_flags = 0;
if (unlikely(tcb->flags & TCPCB_FLAG_SYN)) {
tcp_header_size = sizeof(struct tcphdr) + TCPOLEN_MSS;
- if(sysctl_tcp_timestamps) {
+ if (sysctl_tcp_timestamps) {
tcp_header_size += TCPOLEN_TSTAMP_ALIGNED;
sysctl_flags |= SYSCTL_FLAG_TSTAMPS;
}
(tp->rx_opt.eff_sacks *
TCPOLEN_SACK_PERBLOCK));
}
-
+
if (tcp_packets_in_flight(tp) == 0)
tcp_ca_event(sk, CA_EVENT_TX_START);
- th = (struct tcphdr *) skb_push(skb, tcp_header_size);
- skb->h.th = th;
+#ifdef CONFIG_TCP_MD5SIG
+ /*
+ * Are we doing MD5 on this segment? If so - make
+ * room for it.
+ */
+ md5 = tp->af_specific->md5_lookup(sk, sk);
+ if (md5)
+ tcp_header_size += TCPOLEN_MD5SIG_ALIGNED;
+#endif
+
+ skb_push(skb, tcp_header_size);
+ skb_reset_transport_header(skb);
skb_set_owner_w(skb, sk);
/* Build TCP header and checksum it. */
+ th = tcp_hdr(skb);
th->source = inet->sport;
th->dest = inet->dport;
th->seq = htonl(tcb->seq);
th->ack_seq = htonl(tp->rcv_nxt);
- *(((__u16 *)th) + 6) = htons(((tcp_header_size >> 2) << 12) |
+ *(((__be16 *)th) + 6) = htons(((tcp_header_size >> 2) << 12) |
tcb->flags);
if (unlikely(tcb->flags & TCPCB_FLAG_SYN)) {
/* RFC1323: The window in SYN & SYN/ACK segments
* is never scaled.
*/
- th->window = htons(tp->rcv_wnd);
+ th->window = htons(min(tp->rcv_wnd, 65535U));
} else {
th->window = htons(tcp_select_window(sk));
}
}
if (unlikely(tcb->flags & TCPCB_FLAG_SYN)) {
- tcp_syn_build_options((__u32 *)(th + 1),
+ tcp_syn_build_options((__be32 *)(th + 1),
tcp_advertise_mss(sk),
(sysctl_flags & SYSCTL_FLAG_TSTAMPS),
(sysctl_flags & SYSCTL_FLAG_SACK),
(sysctl_flags & SYSCTL_FLAG_WSCALE),
tp->rx_opt.rcv_wscale,
tcb->when,
- tp->rx_opt.ts_recent);
+ tp->rx_opt.ts_recent,
+
+#ifdef CONFIG_TCP_MD5SIG
+ md5 ? &md5_hash_location :
+#endif
+ NULL);
} else {
- tcp_build_and_update_options((__u32 *)(th + 1),
- tp, tcb->when);
+ tcp_build_and_update_options((__be32 *)(th + 1),
+ tp, tcb->when,
+#ifdef CONFIG_TCP_MD5SIG
+ md5 ? &md5_hash_location :
+#endif
+ NULL);
TCP_ECN_send(sk, tp, skb, tcp_header_size);
}
+#ifdef CONFIG_TCP_MD5SIG
+ /* Calculate the MD5 hash, as we have all we need now */
+ if (md5) {
+ tp->af_specific->calc_md5_hash(md5_hash_location,
+ md5,
+ sk, NULL, NULL,
+ tcp_hdr(skb),
+ sk->sk_protocol,
+ skb->len);
+ }
+#endif
+
icsk->icsk_af_ops->send_check(sk, skb->len, skb);
if (likely(tcb->flags & TCPCB_FLAG_ACK))
if (likely(err <= 0))
return err;
- tcp_enter_cwr(sk);
+ tcp_enter_cwr(sk, 1);
- /* NET_XMIT_CN is special. It does not guarantee,
- * that this packet is lost. It tells that device
- * is about to start to drop packets or already
- * drops some packets of the same priority and
- * invokes us to send less aggressively.
- */
- return err == NET_XMIT_CN ? 0 : err;
+ return net_xmit_eval(err);
#undef SYSCTL_FLAG_TSTAMPS
#undef SYSCTL_FLAG_WSCALE
}
-/* This routine just queue's the buffer
+/* This routine just queue's the buffer
*
* NOTE: probe0 timer is not checked, do not forget tcp_push_pending_frames,
* otherwise socket can stall.
/* Advance write_seq and place onto the write_queue. */
tp->write_seq = TCP_SKB_CB(skb)->end_seq;
skb_header_release(skb);
- __skb_queue_tail(&sk->sk_write_queue, skb);
+ tcp_add_write_queue_tail(sk, skb);
sk_charge_skb(sk, skb);
-
- /* Queue it, remembering where we must start sending. */
- if (sk->sk_send_head == NULL)
- sk->sk_send_head = skb;
}
static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int mss_now)
/* Function to create two new TCP segments. Shrinks the given segment
* to the specified size and appends a new segment with the rest of the
- * packet to the list. This won't be called frequently, I hope.
+ * packet to the list. This won't be called frequently, I hope.
* Remember, these are still headerless SKBs at this point.
*/
int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, unsigned int mss_now)
BUG_ON(len > skb->len);
- clear_all_retrans_hints(tp);
+ clear_all_retrans_hints(tp);
nsize = skb_headlen(skb) - len;
if (nsize < 0)
nsize = 0;
/* Link BUFF into the send queue. */
skb_header_release(buff);
- __skb_append(skb, buff, &sk->sk_write_queue);
+ tcp_insert_write_queue_after(skb, buff, sk);
return 0;
}
}
skb_shinfo(skb)->nr_frags = k;
- skb->tail = skb->data;
+ skb_reset_tail_pointer(skb);
skb->data_len -= len;
skb->len = skb->data_len;
}
icsk->icsk_mtup.enabled = sysctl_tcp_mtu_probing > 1;
icsk->icsk_mtup.search_high = tp->rx_opt.mss_clamp + sizeof(struct tcphdr) +
- icsk->icsk_af_ops->net_header_len;
+ icsk->icsk_af_ops->net_header_len;
icsk->icsk_mtup.search_low = tcp_mss_to_mtu(sk, sysctl_tcp_base_mss);
icsk->icsk_mtup.probe_size = 0;
}
mss_now -= (TCPOLEN_SACK_BASE_ALIGNED +
(tp->rx_opt.eff_sacks * TCPOLEN_SACK_PERBLOCK));
+#ifdef CONFIG_TCP_MD5SIG
+ if (tp->af_specific->md5_lookup(sk, sk))
+ mss_now -= TCPOLEN_MD5SIG_ALIGNED;
+#endif
+
xmit_size_goal = mss_now;
if (doing_tso) {
if (tp->packets_out > tp->snd_cwnd_used)
tp->snd_cwnd_used = tp->packets_out;
- if ((s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto)
+ if (sysctl_tcp_slow_start_after_idle &&
+ (s32)(tcp_time_stamp - tp->snd_cwnd_stamp) >= inet_csk(sk)->icsk_rto)
tcp_cwnd_application_limited(sk);
}
}
u32 in_flight, cwnd;
/* Don't be strict about the congestion window for the final FIN. */
- if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)
+ if ((TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) &&
+ tcp_skb_pcount(skb) == 1)
return 1;
in_flight = tcp_packets_in_flight(tp);
*/
static inline int tcp_nagle_check(const struct tcp_sock *tp,
- const struct sk_buff *skb,
+ const struct sk_buff *skb,
unsigned mss_now, int nonagle)
{
return (skb->len < mss_now &&
return !after(end_seq, tp->snd_una + tp->snd_wnd);
}
-/* This checks if the data bearing packet SKB (usually sk->sk_send_head)
+/* This checks if the data bearing packet SKB (usually tcp_send_head(sk))
* should be put on the wire right now. If so, it returns the number of
* packets allowed by the congestion window.
*/
return cwnd_quota;
}
-static inline int tcp_skb_is_last(const struct sock *sk,
- const struct sk_buff *skb)
-{
- return skb->next == (struct sk_buff *)&sk->sk_write_queue;
-}
-
int tcp_may_send_now(struct sock *sk, struct tcp_sock *tp)
{
- struct sk_buff *skb = sk->sk_send_head;
+ struct sk_buff *skb = tcp_send_head(sk);
return (skb &&
tcp_snd_test(sk, skb, tcp_current_mss(sk, 1),
/* Link BUFF into the send queue. */
skb_header_release(buff);
- __skb_append(skb, buff, &sk->sk_write_queue);
+ tcp_insert_write_queue_after(skb, buff, sk);
return 0;
}
u32 send_win, cong_win, limit, in_flight;
if (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN)
- return 0;
+ goto send_now;
if (icsk->icsk_ca_state != TCP_CA_Open)
- return 0;
+ goto send_now;
+
+ /* Defer for less than two clock ticks. */
+ if (!tp->tso_deferred && ((jiffies<<1)>>1) - (tp->tso_deferred>>1) > 1)
+ goto send_now;
in_flight = tcp_packets_in_flight(tp);
/* If a full-sized TSO skb can be sent, do it. */
if (limit >= 65536)
- return 0;
+ goto send_now;
if (sysctl_tcp_tso_win_divisor) {
u32 chunk = min(tp->snd_wnd, tp->snd_cwnd * tp->mss_cache);
*/
chunk /= sysctl_tcp_tso_win_divisor;
if (limit >= chunk)
- return 0;
+ goto send_now;
} else {
/* Different approach, try not to defer past a single
* ACK. Receiver should ACK every other full sized
* then send now.
*/
if (limit > tcp_max_burst(tp) * tp->mss_cache)
- return 0;
+ goto send_now;
}
/* Ok, it looks like it is advisable to defer. */
+ tp->tso_deferred = 1 | (jiffies<<1);
+
return 1;
+
+send_now:
+ tp->tso_deferred = 0;
+ return 0;
}
/* Create a new MTU probe if we are ready.
/* Have enough data in the send queue to probe? */
len = 0;
- if ((skb = sk->sk_send_head) == NULL)
+ if ((skb = tcp_send_head(sk)) == NULL)
return -1;
while ((len += skb->len) < probe_size && !tcp_skb_is_last(sk, skb))
- skb = skb->next;
+ skb = tcp_write_queue_next(sk, skb);
if (len < probe_size)
return -1;
return -1;
sk_charge_skb(sk, nskb);
- skb = sk->sk_send_head;
- __skb_insert(nskb, skb->prev, skb, &sk->sk_write_queue);
- sk->sk_send_head = nskb;
+ skb = tcp_send_head(sk);
+ tcp_insert_write_queue_before(nskb, skb, sk);
+ tcp_advance_send_head(sk, skb);
TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(skb)->seq;
TCP_SKB_CB(nskb)->end_seq = TCP_SKB_CB(skb)->seq + probe_size;
len = 0;
while (len < probe_size) {
- next = skb->next;
+ next = tcp_write_queue_next(sk, skb);
copy = min_t(int, skb->len, probe_size - len);
if (nskb->ip_summed)
skb_copy_bits(skb, 0, skb_put(nskb, copy), copy);
else
nskb->csum = skb_copy_and_csum_bits(skb, 0,
- skb_put(nskb, copy), copy, nskb->csum);
+ skb_put(nskb, copy), copy, nskb->csum);
if (skb->len <= copy) {
/* We've eaten all the data from this skb.
* Throw it away. */
TCP_SKB_CB(nskb)->flags |= TCP_SKB_CB(skb)->flags;
- __skb_unlink(skb, &sk->sk_write_queue);
+ tcp_unlink_write_queue(skb, sk);
sk_stream_free_skb(sk, skb);
} else {
TCP_SKB_CB(nskb)->flags |= TCP_SKB_CB(skb)->flags &
- ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
+ ~(TCPCB_FLAG_FIN|TCPCB_FLAG_PSH);
if (!skb_shinfo(skb)->nr_frags) {
skb_pull(skb, copy);
if (skb->ip_summed != CHECKSUM_PARTIAL)
sent_pkts = 1;
}
- while ((skb = sk->sk_send_head)) {
+ while ((skb = tcp_send_head(sk))) {
unsigned int limit;
tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
tcp_cwnd_validate(sk, tp);
return 0;
}
- return !tp->packets_out && sk->sk_send_head;
+ return !tp->packets_out && tcp_send_head(sk);
}
/* Push out any pending frames which were held back due to
void __tcp_push_pending_frames(struct sock *sk, struct tcp_sock *tp,
unsigned int cur_mss, int nonagle)
{
- struct sk_buff *skb = sk->sk_send_head;
+ struct sk_buff *skb = tcp_send_head(sk);
if (skb) {
if (tcp_write_xmit(sk, cur_mss, nonagle))
void tcp_push_one(struct sock *sk, unsigned int mss_now)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct sk_buff *skb = sk->sk_send_head;
+ struct sk_buff *skb = tcp_send_head(sk);
unsigned int tso_segs, cwnd_quota;
BUG_ON(!skb || skb->len < mss_now);
/* This function returns the amount that we can raise the
* usable window based on the following constraints
- *
+ *
* 1. The window can never be shrunk once it is offered (RFC 793)
* 2. We limit memory per socket
*
* side SWS prevention criteria. The problem is that under this rule
* a stream of single byte packets will cause the right side of the
* window to always advance by a single byte.
- *
+ *
* Of course, if the sender implements sender side SWS prevention
* then this will not be a problem.
- *
+ *
* BSD seems to make the following compromise:
- *
+ *
* If the free space is less than the 1/4 of the maximum
* space available and the free space is less than 1/2 mss,
* then set the window to 0.
int window;
if (mss > full_space)
- mss = full_space;
+ mss = full_space;
if (free_space < full_space/2) {
icsk->icsk_ack.quick = 0;
*/
if (window <= free_space - mss || window > free_space)
window = (free_space/mss)*mss;
+ else if (mss == full_space &&
+ free_space > window + full_space/2)
+ window = free_space;
}
return window;
static void tcp_retrans_try_collapse(struct sock *sk, struct sk_buff *skb, int mss_now)
{
struct tcp_sock *tp = tcp_sk(sk);
- struct sk_buff *next_skb = skb->next;
+ struct sk_buff *next_skb = tcp_write_queue_next(sk, skb);
/* The first test we must make is that neither of these two
* SKB's are still referenced by someone else.
u16 flags = TCP_SKB_CB(skb)->flags;
/* Also punt if next skb has been SACK'd. */
- if(TCP_SKB_CB(next_skb)->sacked & TCPCB_SACKED_ACKED)
+ if (TCP_SKB_CB(next_skb)->sacked & TCPCB_SACKED_ACKED)
return;
/* Next skb is out of window. */
clear_all_retrans_hints(tp);
/* Ok. We will be able to collapse the packet. */
- __skb_unlink(next_skb, &sk->sk_write_queue);
+ tcp_unlink_write_queue(next_skb, sk);
- memcpy(skb_put(skb, next_skb_size), next_skb->data, next_skb_size);
+ skb_copy_from_linear_data(next_skb,
+ skb_put(skb, next_skb_size),
+ next_skb_size);
- skb->ip_summed = next_skb->ip_summed;
+ if (next_skb->ip_summed == CHECKSUM_PARTIAL)
+ skb->ip_summed = CHECKSUM_PARTIAL;
if (skb->ip_summed != CHECKSUM_PARTIAL)
skb->csum = csum_block_add(skb->csum, next_skb->csum, skb_size);
}
/* Do a simple retransmit without using the backoff mechanisms in
- * tcp_timer. This is used for path mtu discovery.
+ * tcp_timer. This is used for path mtu discovery.
* The socket is already locked here.
- */
+ */
void tcp_simple_retransmit(struct sock *sk)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
unsigned int mss = tcp_current_mss(sk, 0);
int lost = 0;
- sk_stream_for_retrans_queue(skb, sk) {
- if (skb->len > mss &&
+ tcp_for_write_queue(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
+ if (skb->len > mss &&
!(TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_ACKED)) {
if (TCP_SKB_CB(skb)->sacked&TCPCB_SACKED_RETRANS) {
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
tcp_sync_left_out(tp);
- /* Don't muck with the congestion window here.
+ /* Don't muck with the congestion window here.
* Reason is that we do not increase amount of _data_
* in network, but units changed and effective
* cwnd/ssthresh really reduced now.
{
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
- unsigned int cur_mss = tcp_current_mss(sk, 0);
+ unsigned int cur_mss = tcp_current_mss(sk, 0);
int err;
/* Inconslusive MTU probe */
}
/* Collapse two adjacent packets if worthwhile and we can. */
- if(!(TCP_SKB_CB(skb)->flags & TCPCB_FLAG_SYN) &&
- (skb->len < (cur_mss >> 1)) &&
- (skb->next != sk->sk_send_head) &&
- (skb->next != (struct sk_buff *)&sk->sk_write_queue) &&
- (skb_shinfo(skb)->nr_frags == 0 && skb_shinfo(skb->next)->nr_frags == 0) &&
- (tcp_skb_pcount(skb) == 1 && tcp_skb_pcount(skb->next) == 1) &&
- (sysctl_tcp_retrans_collapse != 0))
+ if (!(TCP_SKB_CB(skb)->flags & TCPCB_FLAG_SYN) &&
+ (skb->len < (cur_mss >> 1)) &&
+ (tcp_write_queue_next(sk, skb) != tcp_send_head(sk)) &&
+ (!tcp_skb_is_last(sk, skb)) &&
+ (skb_shinfo(skb)->nr_frags == 0 && skb_shinfo(tcp_write_queue_next(sk, skb))->nr_frags == 0) &&
+ (tcp_skb_pcount(skb) == 1 && tcp_skb_pcount(tcp_write_queue_next(sk, skb)) == 1) &&
+ (sysctl_tcp_retrans_collapse != 0))
tcp_retrans_try_collapse(sk, skb, cur_mss);
if (inet_csk(sk)->icsk_af_ops->rebuild_header(sk))
* retransmit when old data is attached. So strip it off
* since it is cheap to do so and saves bytes on the network.
*/
- if(skb->len > 0 &&
- (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) &&
- tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) {
+ if (skb->len > 0 &&
+ (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) &&
+ tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) {
if (!pskb_trim(skb, 0)) {
TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->end_seq - 1;
skb_shinfo(skb)->gso_segs = 1;
skb = tp->retransmit_skb_hint;
packet_cnt = tp->retransmit_cnt_hint;
}else{
- skb = sk->sk_write_queue.next;
+ skb = tcp_write_queue_head(sk);
packet_cnt = 0;
}
/* First pass: retransmit lost packets. */
if (tp->lost_out) {
- sk_stream_for_retrans_queue_from(skb, sk) {
+ tcp_for_write_queue_from(skb, sk) {
__u8 sacked = TCP_SKB_CB(skb)->sacked;
+ if (skb == tcp_send_head(sk))
+ break;
/* we could do better than to assign each time */
tp->retransmit_skb_hint = skb;
tp->retransmit_cnt_hint = packet_cnt;
else
NET_INC_STATS_BH(LINUX_MIB_TCPSLOWSTARTRETRANS);
- if (skb ==
- skb_peek(&sk->sk_write_queue))
+ if (skb == tcp_write_queue_head(sk))
inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
inet_csk(sk)->icsk_rto,
TCP_RTO_MAX);
skb = tp->forward_skb_hint;
packet_cnt = tp->forward_cnt_hint;
} else{
- skb = sk->sk_write_queue.next;
+ skb = tcp_write_queue_head(sk);
packet_cnt = 0;
}
- sk_stream_for_retrans_queue_from(skb, sk) {
+ tcp_for_write_queue_from(skb, sk) {
+ if (skb == tcp_send_head(sk))
+ break;
tp->forward_cnt_hint = packet_cnt;
tp->forward_skb_hint = skb;
break;
}
- if (skb == skb_peek(&sk->sk_write_queue))
+ if (skb == tcp_write_queue_head(sk))
inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
inet_csk(sk)->icsk_rto,
TCP_RTO_MAX);
*/
void tcp_send_fin(struct sock *sk)
{
- struct tcp_sock *tp = tcp_sk(sk);
- struct sk_buff *skb = skb_peek_tail(&sk->sk_write_queue);
+ struct tcp_sock *tp = tcp_sk(sk);
+ struct sk_buff *skb = tcp_write_queue_tail(sk);
int mss_now;
-
+
/* Optimization, tack on the FIN if we have a queue of
* unsent frames. But be careful about outgoing SACKS
* and IP options.
*/
mss_now = tcp_current_mss(sk, 1);
- if (sk->sk_send_head != NULL) {
+ if (tcp_send_head(sk) != NULL) {
TCP_SKB_CB(skb)->flags |= TCPCB_FLAG_FIN;
TCP_SKB_CB(skb)->end_seq++;
tp->write_seq++;
{
struct sk_buff* skb;
- skb = skb_peek(&sk->sk_write_queue);
+ skb = tcp_write_queue_head(sk);
if (skb == NULL || !(TCP_SKB_CB(skb)->flags&TCPCB_FLAG_SYN)) {
printk(KERN_DEBUG "tcp_send_synack: wrong queue state\n");
return -EFAULT;
struct sk_buff *nskb = skb_copy(skb, GFP_ATOMIC);
if (nskb == NULL)
return -ENOMEM;
- __skb_unlink(skb, &sk->sk_write_queue);
+ tcp_unlink_write_queue(skb, sk);
skb_header_release(nskb);
- __skb_queue_head(&sk->sk_write_queue, nskb);
+ __tcp_add_write_queue_head(sk, nskb);
sk_stream_free_skb(sk, skb);
sk_charge_skb(sk, nskb);
skb = nskb;
struct tcphdr *th;
int tcp_header_size;
struct sk_buff *skb;
+#ifdef CONFIG_TCP_MD5SIG
+ struct tcp_md5sig_key *md5;
+ __u8 *md5_hash_location;
+#endif
skb = sock_wmalloc(sk, MAX_TCP_HEADER + 15, 1, GFP_ATOMIC);
if (skb == NULL)
(ireq->wscale_ok ? TCPOLEN_WSCALE_ALIGNED : 0) +
/* SACK_PERM is in the place of NOP NOP of TS */
((ireq->sack_ok && !ireq->tstamp_ok) ? TCPOLEN_SACKPERM_ALIGNED : 0));
- skb->h.th = th = (struct tcphdr *) skb_push(skb, tcp_header_size);
+#ifdef CONFIG_TCP_MD5SIG
+ /* Are we doing MD5 on this segment? If so - make room for it */
+ md5 = tcp_rsk(req)->af_specific->md5_lookup(sk, req);
+ if (md5)
+ tcp_header_size += TCPOLEN_MD5SIG_ALIGNED;
+#endif
+ skb_push(skb, tcp_header_size);
+ skb_reset_transport_header(skb);
+
+ th = tcp_hdr(skb);
memset(th, 0, sizeof(struct tcphdr));
th->syn = 1;
th->ack = 1;
th->seq = htonl(TCP_SKB_CB(skb)->seq);
th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1);
if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */
- __u8 rcv_wscale;
+ __u8 rcv_wscale;
/* Set this up on the first call only */
req->window_clamp = tp->window_clamp ? : dst_metric(dst, RTAX_WINDOW);
/* tcp_full_space because it is guaranteed to be the first packet */
- tcp_select_initial_window(tcp_full_space(sk),
+ tcp_select_initial_window(tcp_full_space(sk),
dst_metric(dst, RTAX_ADVMSS) - (ireq->tstamp_ok ? TCPOLEN_TSTAMP_ALIGNED : 0),
&req->rcv_wnd,
&req->window_clamp,
ireq->wscale_ok,
&rcv_wscale);
- ireq->rcv_wscale = rcv_wscale;
+ ireq->rcv_wscale = rcv_wscale;
}
/* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */
- th->window = htons(req->rcv_wnd);
+ th->window = htons(min(req->rcv_wnd, 65535U));
TCP_SKB_CB(skb)->when = tcp_time_stamp;
- tcp_syn_build_options((__u32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok,
+ tcp_syn_build_options((__be32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok,
ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale,
TCP_SKB_CB(skb)->when,
- req->ts_recent);
+ req->ts_recent,
+ (
+#ifdef CONFIG_TCP_MD5SIG
+ md5 ? &md5_hash_location :
+#endif
+ NULL)
+ );
skb->csum = 0;
th->doff = (tcp_header_size >> 2);
TCP_INC_STATS(TCP_MIB_OUTSEGS);
+
+#ifdef CONFIG_TCP_MD5SIG
+ /* Okay, we have all we need - do the md5 hash if needed */
+ if (md5) {
+ tp->af_specific->calc_md5_hash(md5_hash_location,
+ md5,
+ NULL, dst, req,
+ tcp_hdr(skb), sk->sk_protocol,
+ skb->len);
+ }
+#endif
+
return skb;
}
-/*
+/*
* Do all connect socket setups that can be done AF independent.
- */
+ */
static void tcp_connect_init(struct sock *sk)
{
struct dst_entry *dst = __sk_dst_get(sk);
tp->tcp_header_len = sizeof(struct tcphdr) +
(sysctl_tcp_timestamps ? TCPOLEN_TSTAMP_ALIGNED : 0);
+#ifdef CONFIG_TCP_MD5SIG
+ if (tp->af_specific->md5_lookup(sk, sk) != NULL)
+ tp->tcp_header_len += TCPOLEN_MD5SIG_ALIGNED;
+#endif
+
/* If user gave his TCP_MAXSEG, record it to clamp */
if (tp->rx_opt.user_mss)
tp->rx_opt.mss_clamp = tp->rx_opt.user_mss;
/*
* Build a SYN and send it off.
- */
+ */
int tcp_connect(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
TCP_SKB_CB(buff)->when = tcp_time_stamp;
tp->retrans_stamp = TCP_SKB_CB(buff)->when;
skb_header_release(buff);
- __skb_queue_tail(&sk->sk_write_queue, buff);
+ __tcp_add_write_queue_tail(sk, buff);
sk_charge_skb(sk, buff);
tp->packets_out += tcp_skb_pcount(buff);
tcp_transmit_skb(sk, buff, 1, GFP_KERNEL);
/* We don't queue it, tcp_transmit_skb() sets ownership. */
skb = alloc_skb(MAX_TCP_HEADER, GFP_ATOMIC);
- if (skb == NULL)
+ if (skb == NULL)
return -1;
/* Reserve space for headers and set control bits. */
struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *skb;
- if ((skb = sk->sk_send_head) != NULL &&
+ if ((skb = tcp_send_head(sk)) != NULL &&
before(TCP_SKB_CB(skb)->seq, tp->snd_una+tp->snd_wnd)) {
int err;
unsigned int mss = tcp_current_mss(sk, 0);
err = tcp_write_wakeup(sk);
- if (tp->packets_out || !sk->sk_send_head) {
+ if (tp->packets_out || !tcp_send_head(sk)) {
/* Cancel probe timer, if it is not required. */
icsk->icsk_probes_out = 0;
icsk->icsk_backoff = 0;
if (icsk->icsk_backoff < sysctl_tcp_retries2)
icsk->icsk_backoff++;
icsk->icsk_probes_out++;
- inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
min(icsk->icsk_rto << icsk->icsk_backoff, TCP_RTO_MAX),
TCP_RTO_MAX);
} else {
*/
if (!icsk->icsk_probes_out)
icsk->icsk_probes_out = 1;
- inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
+ inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0,
min(icsk->icsk_rto << icsk->icsk_backoff,
TCP_RESOURCE_PROBE_INTERVAL),
TCP_RTO_MAX);
projects / linux-2.6 / blobdiff