/**
* Requeue the first unacked segment for retransmission
*
* Called by tcp_receive() for fast retramsmit.
*
* @param pcb the tcp_pcb for which to retransmit the first unacked segment
*/
void
tcp_rexmit(struct tcp_pcb *pcb)
{
struct tcp_seg *seg;
struct tcp_seg **cur_seg;
if (pcb->unacked == NULL) {
return;
}
/* Move the first unacked segment to the unsent queue */
/* Keep the unsent queue sorted. */
seg = pcb->unacked;
pcb->unacked = seg->next;
cur_seg = &(pcb->unsent);
while (*cur_seg &&
TCP_SEQ_LT((*cur_seg)->seqno, seg->seqno)) {
cur_seg = &((*cur_seg)->next );
}
seg->next = *cur_seg;
*cur_seg = seg;
#if TCP_OVERSIZE
if (seg->next == NULL) {
/* the retransmitted segment is last in unsent, so reset unsent_oversize */
pcb->unsent_oversize = 0;
}
#endif /* TCP_OVERSIZE */
++pcb->nrtx;
/* Don't take any rtt measurements after retransmitting. */
pcb->rttest = 0;
}
/**
Check whether the sequence number of the incoming segment is acceptable.
@param Tcb Pointer to the TCP_CB of this TCP instance.
@param Seg Pointer to the incoming segment.
@retval 1 The sequence number is acceptable.
@retval 0 The sequence number is not acceptable.
**/
INTN
TcpSeqAcceptable (
IN TCP_CB *Tcb,
IN TCP_SEG *Seg
)
{
return (TCP_SEQ_LEQ (Tcb->RcvWl2, Seg->End) &&
TCP_SEQ_LT (Seg->Seq, Tcb->RcvWl2 + Tcb->RcvWnd));
}
/*----------------------------------------------------------------------------*/
static inline int
ValidateSequence(mtcp_manager_t mtcp, tcp_stream *cur_stream, uint32_t cur_ts,
struct tcphdr *tcph, uint32_t seq, uint32_t ack_seq, int payloadlen)
{
/* Protect Against Wrapped Sequence number (PAWS) */
if (!tcph->rst && cur_stream->saw_timestamp) {
struct tcp_timestamp ts;
if (!ParseTCPTimestamp(cur_stream, &ts,
(uint8_t *)tcph + TCP_HEADER_LEN,
(tcph->doff << 2) - TCP_HEADER_LEN)) {
/* if there is no timestamp */
/* TODO: implement here */
TRACE_DBG("No timestamp found.\n");
return FALSE;
}
/* RFC1323: if SEG.TSval < TS.Recent, drop and send ack */
if (TCP_SEQ_LT(ts.ts_val, cur_stream->rcvvar->ts_recent)) {
/* TODO: ts_recent should be invalidated
before timestamp wraparound for long idle flow */
TRACE_DBG("PAWS Detect wrong timestamp. "
"seq: %u, ts_val: %u, prev: %u\n",
seq, ts.ts_val, cur_stream->rcvvar->ts_recent);
EnqueueACK(mtcp, cur_stream, cur_ts, ACK_OPT_NOW);
return FALSE;
} else {
/* valid timestamp */
if (TCP_SEQ_GT(ts.ts_val, cur_stream->rcvvar->ts_recent)) {
TRACE_TSTAMP("Timestamp update. cur: %u, prior: %u "
"(time diff: %uus)\n",
ts.ts_val, cur_stream->rcvvar->ts_recent,
TS_TO_USEC(cur_ts - cur_stream->rcvvar->ts_last_ts_upd));
cur_stream->rcvvar->ts_last_ts_upd = cur_ts;
}
cur_stream->rcvvar->ts_recent = ts.ts_val;
cur_stream->rcvvar->ts_lastack_rcvd = ts.ts_ref;
}
}
/* TCP sequence validation */
if (!TCP_SEQ_BETWEEN(seq + payloadlen, cur_stream->rcv_nxt,
cur_stream->rcv_nxt + cur_stream->rcvvar->rcv_wnd)) {
/* if RST bit is set, ignore the segment */
if (tcph->rst)
return FALSE;
if (cur_stream->state == TCP_ST_ESTABLISHED) {
/* check if it is to get window advertisement */
if (seq + 1 == cur_stream->rcv_nxt) {
#if 0
TRACE_DBG("Window update request. (seq: %u, rcv_wnd: %u)\n",
seq, cur_stream->rcvvar->rcv_wnd);
#endif
EnqueueACK(mtcp, cur_stream, cur_ts, ACK_OPT_AGGREGATE);
return FALSE;
}
if (TCP_SEQ_LEQ(seq, cur_stream->rcv_nxt)) {
EnqueueACK(mtcp, cur_stream, cur_ts, ACK_OPT_AGGREGATE);
} else {
EnqueueACK(mtcp, cur_stream, cur_ts, ACK_OPT_NOW);
}
} else {
if (cur_stream->state == TCP_ST_TIME_WAIT) {
TRACE_DBG("Stream %d: tw expire update to %u\n",
cur_stream->id, cur_stream->rcvvar->ts_tw_expire);
AddtoTimewaitList(mtcp, cur_stream, cur_ts);
}
AddtoControlList(mtcp, cur_stream, cur_ts);
}
return FALSE;
}
return TRUE;
}
/* find out what we can send and send it */
err_t
tcp_output(struct tcp_pcb *pcb)
{
struct pbuf *p;
struct tcp_hdr *tcphdr;
struct tcp_seg *seg, *useg;
u32_t wnd;
#if TCP_CWND_DEBUG
s16_t i = 0;
#endif /* TCP_CWND_DEBUG */
/* First, check if we are invoked by the TCP input processing
code. If so, we do not output anything. Instead, we rely on the
input processing code to call us when input processing is done
with. */
if (tcp_input_pcb == pcb) {
return ERR_OK;
}
wnd = LWIP_MIN(pcb->snd_wnd, pcb->cwnd);
seg = pcb->unsent;
/* useg should point to last segment on unacked queue */
useg = pcb->unacked;
if (useg != NULL) {
for (; useg->next != NULL; useg = useg->next);
}
/* If the TF_ACK_NOW flag is set and no data will be sent (either
* because the ->unsent queue is empty or because the window does
* not allow it), construct an empty ACK segment and send it.
*
* If data is to be sent, we will just piggyback the ACK (see below).
*/
if (pcb->flags & TF_ACK_NOW &&
(seg == NULL ||
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len > wnd)) {
p = pbuf_alloc(PBUF_IP, TCP_HLEN, PBUF_RAM);
if (p == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: (ACK) could not allocate pbuf\n"));
return ERR_BUF;
}
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: sending ACK for %"U32_F"\n", pcb->rcv_nxt));
/* remove ACK flags from the PCB, as we send an empty ACK now */
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
tcphdr = p->payload;
tcphdr->src = htons(pcb->local_port);
tcphdr->dest = htons(pcb->remote_port);
tcphdr->seqno = htonl(pcb->snd_nxt);
tcphdr->ackno = htonl(pcb->rcv_nxt);
TCPH_FLAGS_SET(tcphdr, TCP_ACK);
tcphdr->wnd = htons(pcb->rcv_wnd);
tcphdr->urgp = 0;
TCPH_HDRLEN_SET(tcphdr, 5);
tcphdr->chksum = 0;
#if CHECKSUM_GEN_TCP
tcphdr->chksum = inet_chksum_pseudo(p, &(pcb->local_ip), &(pcb->remote_ip),
IP_PROTO_TCP, p->tot_len);
#endif
ip_output(p, &(pcb->local_ip), &(pcb->remote_ip), pcb->ttl, pcb->tos,
IP_PROTO_TCP);
pbuf_free(p);
return ERR_OK;
}
#if TCP_OUTPUT_DEBUG
if (seg == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: nothing to send (%p)\n", (void*)pcb->unsent));
}
#endif /* TCP_OUTPUT_DEBUG */
#if TCP_CWND_DEBUG
if (seg == NULL) {
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %"U32_F", cwnd %"U16_F", wnd %"U32_F", seg == NULL, ack %"U32_F"\n",
pcb->snd_wnd, pcb->cwnd, wnd,
pcb->lastack));
} else {
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %"U32_F", cwnd %"U16_F", wnd %"U32_F", effwnd %"U32_F", seq %"U32_F", ack %"U32_F"\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len,
ntohl(seg->tcphdr->seqno), pcb->lastack));
}
#endif /* TCP_CWND_DEBUG */
/* data available and window allows it to be sent? */
while (seg != NULL &&
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len <= wnd) {
#if TCP_CWND_DEBUG
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %"U32_F", cwnd %"U16_F", wnd %"U32_F", effwnd %"U32_F", seq %"U32_F", ack %"U32_F", i %"S16_F"\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) + seg->len -
pcb->lastack,
ntohl(seg->tcphdr->seqno), pcb->lastack, i));
++i;
#endif /* TCP_CWND_DEBUG */
pcb->unsent = seg->next;
if (pcb->state != SYN_SENT) {
TCPH_SET_FLAG(seg->tcphdr, TCP_ACK);
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
}
tcp_output_segment(seg, pcb);
pcb->snd_nxt = ntohl(seg->tcphdr->seqno) + TCP_TCPLEN(seg);
if (TCP_SEQ_LT(pcb->snd_max, pcb->snd_nxt)) {
pcb->snd_max = pcb->snd_nxt;
}
/* put segment on unacknowledged list if length > 0 */
if (TCP_TCPLEN(seg) > 0) {
seg->next = NULL;
/* unacked list is empty? */
if (pcb->unacked == NULL) {
pcb->unacked = seg;
useg = seg;
/* unacked list is not empty? */
} else {
/* In the case of fast retransmit, the packet should not go to the tail
* of the unacked queue, but rather at the head. We need to check for
* this case. -STJ Jul 27, 2004 */
if (TCP_SEQ_LT(ntohl(seg->tcphdr->seqno), ntohl(useg->tcphdr->seqno))){
/* add segment to head of unacked list */
seg->next = pcb->unacked;
pcb->unacked = seg;
} else {
/* add segment to tail of unacked list */
useg->next = seg;
useg = useg->next;
}
}
/* do not queue empty segments on the unacked list */
} else {
tcp_seg_free(seg);
}
seg = pcb->unsent;
}
return ERR_OK;
}
/* find out what we can send and send it */
err_t
tcp_output(struct tcp_pcb *pcb)
{
struct pbuf *p;
struct tcp_hdr *tcphdr;
struct tcp_seg *seg, *useg;
u32_t wnd;
#if TCP_CWND_DEBUG
int i = 0;
#endif /* TCP_CWND_DEBUG */
/* First, check if we are invoked by the TCP input processing
code. If so, we do not output anything. Instead, we rely on the
input processing code to call us when input processing is done
with. */
if (tcp_input_pcb == pcb) {
return ERR_OK;
}
wnd = LWIP_MIN(pcb->snd_wnd, pcb->cwnd);
seg = pcb->unsent;
/* useg should point to last segment on unacked queue */
useg = pcb->unacked;
if (useg != NULL) {
for (; useg->next != NULL; useg = useg->next);
}
/* If the TF_ACK_NOW flag is set, we check if there is data that is
to be sent. If data is to be sent out, we'll just piggyback our
acknowledgement with the outgoing segment. If no data will be
sent (either because the ->unsent queue is empty or because the
window doesn't allow it) we'll have to construct an empty ACK
segment and send it. */
if (pcb->flags & TF_ACK_NOW &&
(seg == NULL ||
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len > wnd)) {
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
p = pbuf_alloc(PBUF_IP, TCP_HLEN, PBUF_RAM);
if (p == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: (ACK) could not allocate pbuf\n"));
return ERR_BUF;
}
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: sending ACK for %lu\n", pcb->rcv_nxt));
tcphdr = p->payload;
tcphdr->src = htons(pcb->local_port);
tcphdr->dest = htons(pcb->remote_port);
tcphdr->seqno = htonl(pcb->snd_nxt);
tcphdr->ackno = htonl(pcb->rcv_nxt);
TCPH_FLAGS_SET(tcphdr, TCP_ACK);
tcphdr->wnd = htons(pcb->rcv_wnd);
tcphdr->urgp = 0;
TCPH_HDRLEN_SET(tcphdr, 5);
tcphdr->chksum = 0;
tcphdr->chksum = inet_chksum_pseudo(p, &(pcb->local_ip), &(pcb->remote_ip),
IP_PROTO_TCP, p->tot_len);
ip_output(p, &(pcb->local_ip), &(pcb->remote_ip), pcb->ttl, pcb->tos,
IP_PROTO_TCP);
pbuf_free(p);
return ERR_OK;
}
#if TCP_OUTPUT_DEBUG
if (seg == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: nothing to send (%p)\n", pcb->unsent));
}
#endif /* TCP_OUTPUT_DEBUG */
#if TCP_CWND_DEBUG
if (seg == NULL) {
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %lu, cwnd %lu, wnd %lu, seg == NULL, ack %lu\n",
pcb->snd_wnd, pcb->cwnd, wnd,
pcb->lastack));
} else {
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %lu, cwnd %lu, wnd %lu, effwnd %lu, seq %lu, ack %lu\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len,
ntohl(seg->tcphdr->seqno), pcb->lastack));
}
#endif /* TCP_CWND_DEBUG */
while (seg != NULL &&
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len <= wnd) {
#if TCP_CWND_DEBUG
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %lu, cwnd %lu, wnd %lu, effwnd %lu, seq %lu, ack %lu, i%d\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) + seg->len -
pcb->lastack,
ntohl(seg->tcphdr->seqno), pcb->lastack, i));
++i;
#endif /* TCP_CWND_DEBUG */
pcb->unsent = seg->next;
if (pcb->state != SYN_SENT) {
TCPH_SET_FLAG(seg->tcphdr, TCP_ACK);
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
}
tcp_output_segment(seg, pcb);
pcb->snd_nxt = ntohl(seg->tcphdr->seqno) + TCP_TCPLEN(seg);
if (TCP_SEQ_LT(pcb->snd_max, pcb->snd_nxt)) {
pcb->snd_max = pcb->snd_nxt;
}
/* put segment on unacknowledged list if length > 0 */
if (TCP_TCPLEN(seg) > 0) {
seg->next = NULL;
if (pcb->unacked == NULL) {
pcb->unacked = seg;
useg = seg;
} else {
useg->next = seg;
useg = useg->next;
}
} else {
tcp_seg_free(seg);
}
seg = pcb->unsent;
}
return ERR_OK;
}
/**
* Called by tcp_process. Checks if the given segment is an ACK for outstanding
* data, and if so frees the memory of the buffered data. Next, is places the
* segment on any of the receive queues (pcb->recved or pcb->ooseq). If the segment
* is buffered, the pbuf is referenced by pbuf_ref so that it will not be freed until
* i it has been removed from the buffer.
*
* If the incoming segment constitutes an ACK for a segment that was used for RTT
* estimation, the RTT is estimated here as well.
*
* Called from tcp_process().
*
* @return 1 if the incoming segment is the next in sequence, 0 if not
*/
static u8_t
tcp_receive(struct tcp_pcb *pcb)
{
struct tcp_seg *next;
#if TCP_QUEUE_OOSEQ
struct tcp_seg *prev, *cseg;
#endif
struct pbuf *p;
s32_t off;
s16_t m;
u32_t right_wnd_edge;
u16_t new_tot_len;
u8_t accepted_inseq = 0;
if (flags & TCP_ACK) {
right_wnd_edge = pcb->snd_wnd + pcb->snd_wl1;
/* Update window. */
if (TCP_SEQ_LT(pcb->snd_wl1, seqno) ||
(pcb->snd_wl1 == seqno && TCP_SEQ_LT(pcb->snd_wl2, ackno)) ||
(pcb->snd_wl2 == ackno && tcphdr->wnd > pcb->snd_wnd)) {
pcb->snd_wnd = tcphdr->wnd;
pcb->snd_wl1 = seqno;
pcb->snd_wl2 = ackno;
if (pcb->snd_wnd > 0 && pcb->persist_backoff > 0) {
pcb->persist_backoff = 0;
}
LWIP_DEBUGF(TCP_WND_DEBUG, ("tcp_receive: window update %"U16_F"\n", pcb->snd_wnd));
#if TCP_WND_DEBUG
} else {
if (pcb->snd_wnd != tcphdr->wnd) {
LWIP_DEBUGF(TCP_WND_DEBUG, ("tcp_receive: no window update lastack %"U32_F" snd_max %"U32_F" ackno %"U32_F" wl1 %"U32_F" seqno %"U32_F" wl2 %"U32_F"\n",
pcb->lastack, pcb->snd_max, ackno, pcb->snd_wl1, seqno, pcb->snd_wl2));
}
#endif /* TCP_WND_DEBUG */
}
if (pcb->lastack == ackno) {
pcb->acked = 0;
if (pcb->snd_wl1 + pcb->snd_wnd == right_wnd_edge){
++pcb->dupacks;
if (pcb->dupacks >= 3 && pcb->unacked != NULL) {
if (!(pcb->flags & TF_INFR)) {
/* This is fast retransmit. Retransmit the first unacked segment. */
LWIP_DEBUGF(TCP_FR_DEBUG, ("tcp_receive: dupacks %"U16_F" (%"U32_F"), fast retransmit %"U32_F"\n",
(u16_t)pcb->dupacks, pcb->lastack,
ntohl(pcb->unacked->tcphdr->seqno)));
tcp_rexmit(pcb);
/* Set ssthresh to max (FlightSize / 2, 2*SMSS) */
/*pcb->ssthresh = LWIP_MAX((pcb->snd_max -
pcb->lastack) / 2,
2 * pcb->mss);*/
/* Set ssthresh to half of the minimum of the current cwnd and the advertised window */
if (pcb->cwnd > pcb->snd_wnd)
pcb->ssthresh = pcb->snd_wnd / 2;
else
pcb->ssthresh = pcb->cwnd / 2;
/* The minimum value for ssthresh should be 2 MSS */
if (pcb->ssthresh < 2*pcb->mss) {
LWIP_DEBUGF(TCP_FR_DEBUG, ("tcp_receive: The minimum value for ssthresh %"U16_F" should be min 2 mss %"U16_F"...\n", pcb->ssthresh, 2*pcb->mss));
pcb->ssthresh = 2*pcb->mss;
}
pcb->cwnd = pcb->ssthresh + 3 * pcb->mss;
pcb->flags |= TF_INFR;
} else {
/* Inflate the congestion window, but not if it means that
the value overflows. */
if ((u16_t)(pcb->cwnd + pcb->mss) > pcb->cwnd) {
pcb->cwnd += pcb->mss;
}
}
}
} else {
LWIP_DEBUGF(TCP_FR_DEBUG, ("tcp_receive: dupack averted %"U32_F" %"U32_F"\n",
pcb->snd_wl1 + pcb->snd_wnd, right_wnd_edge));
}
} else if (TCP_SEQ_BETWEEN(ackno, pcb->lastack+1, pcb->snd_max)){
/* We come here when the ACK acknowledges new data. */
/* Reset the "IN Fast Retransmit" flag, since we are no longer
in fast retransmit. Also reset the congestion window to the
slow start threshold. */
if (pcb->flags & TF_INFR) {
pcb->flags &= ~TF_INFR;
pcb->cwnd = pcb->ssthresh;
}
/* Reset the number of retransmissions. */
pcb->nrtx = 0;
/* Reset the retransmission time-out. */
pcb->rto = (pcb->sa >> 3) + pcb->sv;
/* Update the send buffer space. Diff between the two can never exceed 64K? */
pcb->acked = (u16_t)(ackno - pcb->lastack);
pcb->snd_buf += pcb->acked;
/* Reset the fast retransmit variables. */
pcb->dupacks = 0;
pcb->lastack = ackno;
/* Update the congestion control variables (cwnd and
ssthresh). */
if (pcb->state >= ESTABLISHED) {
if (pcb->cwnd < pcb->ssthresh) {
if ((u16_t)(pcb->cwnd + pcb->mss) > pcb->cwnd) {
pcb->cwnd += pcb->mss;
}
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: slow start cwnd %"U16_F"\n", pcb->cwnd));
} else {
u16_t new_cwnd = (pcb->cwnd + pcb->mss * pcb->mss / pcb->cwnd);
if (new_cwnd > pcb->cwnd) {
pcb->cwnd = new_cwnd;
}
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: congestion avoidance cwnd %"U16_F"\n", pcb->cwnd));
}
}
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: ACK for %"U32_F", unacked->seqno %"U32_F":%"U32_F"\n",
ackno,
pcb->unacked != NULL?
ntohl(pcb->unacked->tcphdr->seqno): 0,
pcb->unacked != NULL?
ntohl(pcb->unacked->tcphdr->seqno) + TCP_TCPLEN(pcb->unacked): 0));
/* Remove segment from the unacknowledged list if the incoming
ACK acknowlegdes them. */
while (pcb->unacked != NULL &&
TCP_SEQ_LEQ(ntohl(pcb->unacked->tcphdr->seqno) +
TCP_TCPLEN(pcb->unacked), ackno)) {
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: removing %"U32_F":%"U32_F" from pcb->unacked\n",
ntohl(pcb->unacked->tcphdr->seqno),
ntohl(pcb->unacked->tcphdr->seqno) +
TCP_TCPLEN(pcb->unacked)));
next = pcb->unacked;
pcb->unacked = pcb->unacked->next;
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_receive: queuelen %"U16_F" ... ", (u16_t)pcb->snd_queuelen));
LWIP_ASSERT("pcb->snd_queuelen >= pbuf_clen(next->p)", (pcb->snd_queuelen >= pbuf_clen(next->p)));
pcb->snd_queuelen -= pbuf_clen(next->p);
tcp_seg_free(next);
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("%"U16_F" (after freeing unacked)\n", (u16_t)pcb->snd_queuelen));
if (pcb->snd_queuelen != 0) {
LWIP_ASSERT("tcp_receive: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
}
/* If there's nothing left to acknowledge, stop the retransmit
timer, otherwise reset it to start again */
if(pcb->unacked == NULL)
pcb->rtime = -1;
else
pcb->rtime = 0;
pcb->polltmr = 0;
} else {
/**
* Find out what we can send and send it
*
* @param pcb Protocol control block for the TCP connection to send data
* @return ERR_OK if data has been sent or nothing to send
* another err_t on error
*/
err_t
tcp_output(struct tcp_pcb *pcb)
{
struct tcp_seg *seg, *useg;
u32_t wnd, snd_nxt;
#if TCP_CWND_DEBUG
s16_t i = 0;
#endif /* TCP_CWND_DEBUG */
/* First, check if we are invoked by the TCP input processing
code. If so, we do not output anything. Instead, we rely on the
input processing code to call us when input processing is done
with. */
if (tcp_input_pcb == pcb) {
return ERR_OK;
}
wnd = LWIP_MIN(pcb->snd_wnd, pcb->cwnd);
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %"U32_F", cwnd %"U32_F
", wnd %"U32_F"\n",pcb->snd_wnd, pcb->cwnd, wnd ));
seg = pcb->unsent;
/* If the TF_ACK_NOW flag is set and no data will be sent (either
* because the ->unsent queue is empty or because the window does
* not allow it), construct an empty ACK segment and send it.
*
* If data is to be sent, we will just piggyback the ACK (see below).
*/
if (pcb->flags & TF_ACK_NOW &&
(seg == NULL ||
seg->seqno - pcb->lastack + seg->len > wnd)) {
return tcp_send_empty_ack(pcb);
}
#if TCP_OUTPUT_DEBUG
if (seg == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: nothing to send (%p)\n",
(void*)pcb->unsent));
}
#endif /* TCP_OUTPUT_DEBUG */
#if TCP_CWND_DEBUG
if (seg == NULL) {
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %"U32_F
", cwnd %"U32_F", wnd %"U32_F
", seg == NULL, ack %"U32_F"\n",
pcb->snd_wnd, pcb->cwnd, wnd, pcb->lastack));
} else {
LWIP_DEBUGF(TCP_CWND_DEBUG,
("tcp_output: snd_wnd %"U32_F", cwnd %"U32_F", wnd %"U32_F
", effwnd %"U32_F", seq %"U32_F", ack %"U32_F"\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len,
ntohl(seg->tcphdr->seqno), pcb->lastack));
}
#endif /* TCP_CWND_DEBUG */
/* data available and window allows it to be sent? */
while (seg != NULL &&
seg->seqno - pcb->lastack + seg->len <= wnd) {
LWIP_ASSERT("RST not expected here!",
(TCPH_FLAGS(seg->tcphdr) & TCP_RST) == 0);
/* Stop sending if the nagle algorithm would prevent it
* Don't stop:
* - if tcp_write had a memory error before (prevent delayed ACK timeout) or
* - if FIN was already enqueued for this PCB (SYN is always alone in a segment -
* either seg->next != NULL or pcb->unacked == NULL;
* RST is no sent using tcp_write/tcp_output.
*/
if((tcp_do_output_nagle(pcb) == 0) &&
((pcb->flags & (TF_NAGLEMEMERR | TF_FIN)) == 0)){
if ( pcb->snd_sml_snt > (pcb->unacked ? pcb->unacked->len : 0) ) {
break;
}
else {
if ( (u32_t)((seg->next ? seg->next->len : 0) + seg->len) <= pcb->snd_sml_add ) {
pcb->snd_sml_snt = pcb->snd_sml_add;
}
}
}
#if TCP_CWND_DEBUG
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %"U32_F", cwnd %"U16_F", wnd %"U32_F", effwnd %"U32_F", seq %"U32_F", ack %"U32_F", i %"S16_F"\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) + seg->len -
pcb->lastack,
ntohl(seg->tcphdr->seqno), pcb->lastack, i));
++i;
#endif /* TCP_CWND_DEBUG */
pcb->unsent = seg->next;
if (get_tcp_state(pcb) != SYN_SENT) {
TCPH_SET_FLAG(seg->tcphdr, TCP_ACK);
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
}
#if TCP_OVERSIZE_DBGCHECK
seg->oversize_left = 0;
#endif /* TCP_OVERSIZE_DBGCHECK */
tcp_output_segment(seg, pcb);
snd_nxt = seg->seqno + TCP_TCPLEN(seg);
if (TCP_SEQ_LT(pcb->snd_nxt, snd_nxt)) {
pcb->snd_nxt = snd_nxt;
}
/* put segment on unacknowledged list if length > 0 */
if (TCP_TCPLEN(seg) > 0) {
seg->next = NULL;
/* unacked list is empty? */
if (pcb->unacked == NULL) {
pcb->unacked = seg;
pcb->last_unacked = seg;
/* unacked list is not empty? */
} else {
/* In the case of fast retransmit, the packet should not go to the tail
* of the unacked queue, but rather somewhere before it. We need to check for
* this case. -STJ Jul 27, 2004 */
useg = pcb->last_unacked;
if (TCP_SEQ_LT(seg->seqno, useg->seqno)) {
/* add segment to before tail of unacked list, keeping the list sorted */
struct tcp_seg **cur_seg = &(pcb->unacked);
while (*cur_seg &&
TCP_SEQ_LT((*cur_seg)->seqno, seg->seqno)) {
cur_seg = &((*cur_seg)->next );
}
LWIP_ASSERT("Value of last_unacked is invalid",
*cur_seg != pcb->last_unacked->next);
seg->next = (*cur_seg);
(*cur_seg) = seg;
} else {
/* add segment to tail of unacked list */
useg->next = seg;
pcb->last_unacked = seg;
}
}
/* do not queue empty segments on the unacked list */
} else {
tcp_tx_seg_free(pcb, seg);
}
seg = pcb->unsent;
}
#if TCP_OVERSIZE
if (pcb->unsent == NULL) {
/* last unsent has been removed, reset unsent_oversize */
pcb->unsent_oversize = 0;
}
#endif /* TCP_OVERSIZE */
pcb->flags &= ~TF_NAGLEMEMERR;
return ERR_OK;
}
/* find out what we can send and send it */
err_t
tcp_output(struct tcp_pcb *pcb)
{
struct pbuf *p;
struct tcp_hdr *tcphdr;
struct tcp_seg *seg, *useg;
u32_t wnd;
#if TCP_CWND_DEBUG
int i = 0;
#endif /* TCP_CWND_DEBUG */
wnd = MIN(pcb->snd_wnd, pcb->cwnd);
seg = pcb->unsent;
if(((seg == NULL)
|| (ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len > wnd))
&& (pcb->flags & TF_ACK_NOW)) {
/* If no segments are enqueued but we should send an ACK, we
construct the ACK and send it. */
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
pcb->rcv_adv = pcb->rcv_nxt + pcb->rcv_wnd;
p = pbuf_alloc(PBUF_TRANSPORT, 0, PBUF_RAM);
if(p == NULL) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: (ACK) could not allocate pbuf\n"));
return ERR_BUF;
}
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: sending ACK for %lu\n", pcb->rcv_nxt));
if(pbuf_header(p, TCP_HLEN)) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: (ACK) no room for TCP header in pbuf.\n"));
#ifdef TCP_STATS
++stats.tcp.err;
#endif /* TCP_STATS */
pbuf_free(p);
return ERR_BUF;
}
tcphdr = p->payload;
tcphdr->src = htons(pcb->local_port);
tcphdr->dest = htons(pcb->remote_port);
tcphdr->seqno = htonl(pcb->snd_nxt);
tcphdr->ackno = htonl(pcb->rcv_nxt);
TCPH_FLAGS_SET(tcphdr, TCP_ACK);
tcphdr->wnd = htons(pcb->rcv_wnd);
tcphdr->urgp = 0;
TCPH_OFFSET_SET(tcphdr, 5 << 4);
tcphdr->chksum = 0;
#if CHECKSUM_GEN_TCP
tcphdr->chksum = inet_chksum_pseudo(p, &(pcb->local_ip), &(pcb->remote_ip),
IP_PROTO_TCP, p->tot_len);
#endif
ip_output(p, &(pcb->local_ip), &(pcb->remote_ip), TCP_TTL,
IP_PROTO_TCP);
pbuf_free(p);
return ERR_OK;
}
#if TCP_OUTPUT_DEBUG
if(seg == NULL) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: nothing to send\n"));
}
#endif /* TCP_OUTPUT_DEBUG */
#if TCP_CWND_DEBUG
if(seg == NULL) {
DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %lu, cwnd %lu, wnd %lu, seg == NULL, ack %lu\n",
pcb->snd_wnd, pcb->cwnd, wnd,
pcb->lastack));
} else {
DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %lu, cwnd %lu, wnd %lu, effwnd %lu, seq %lu, ack %lu\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len,
ntohl(seg->tcphdr->seqno), pcb->lastack));
}
#endif /* TCP_CWND_DEBUG */
while(seg != NULL &&
ntohl(seg->tcphdr->seqno) - pcb->lastack + seg->len <= wnd) {
pcb->rtime = 0;
#if TCP_CWND_DEBUG
DEBUGF(TCP_CWND_DEBUG, ("tcp_output: snd_wnd %lu, cwnd %lu, wnd %lu, effwnd %lu, seq %lu, ack %lu, i%d\n",
pcb->snd_wnd, pcb->cwnd, wnd,
ntohl(seg->tcphdr->seqno) + seg->len -
pcb->lastack,
ntohl(seg->tcphdr->seqno), pcb->lastack, i));
++i;
#endif /* TCP_CWND_DEBUG */
pcb->unsent = seg->next;
if(pcb->state != SYN_SENT) {
TCPH_FLAGS_SET(seg->tcphdr, TCPH_FLAGS(seg->tcphdr) | TCP_ACK);
pcb->flags &= ~(TF_ACK_DELAY | TF_ACK_NOW);
pcb->rcv_adv = pcb->rcv_nxt + pcb->rcv_wnd;
}
tcp_output_segment(seg, pcb);
pcb->snd_nxt = ntohl(seg->tcphdr->seqno) + TCP_TCPLEN(seg);
if(TCP_SEQ_LT(pcb->snd_max, pcb->snd_nxt)) {
pcb->snd_max = pcb->snd_nxt;
}
/* put segment on unacknowledged list if length > 0 */
if(TCP_TCPLEN(seg) > 0) {
seg->next = NULL;
if(pcb->unacked == NULL) {
pcb->unacked = seg;
} else {
for(useg = pcb->unacked; useg->next != NULL; useg = useg->next);
useg->next = seg;
}
/* seg->rtime = 0;*/
} else {
tcp_seg_free(seg);
}
seg = pcb->unsent;
}
return ERR_OK;
}
/**
* Called by tcp_process. Checks if the given segment is an ACK for outstanding
* data, and if so frees the memory of the buffered data. Next, is places the
* segment on any of the receive queues (pcb->recved or pcb->ooseq). If the segment
* is buffered, the pbuf is referenced by pbuf_ref so that it will not be freed until
* i it has been removed from the buffer.
*
* If the incoming segment constitutes an ACK for a segment that was used for RTT
* estimation, the RTT is estimated here as well.
*
* Called from tcp_process().
*/
static void
tcp_receive(struct tcp_pcb *pcb)
{
struct tcp_seg *next;
#if TCP_QUEUE_OOSEQ
struct tcp_seg *prev, *cseg;
#endif
struct pbuf *p;
s32_t off;
s16_t m;
u32_t right_wnd_edge;
u16_t new_tot_len;
int found_dupack = 0;
if (flags & TCP_ACK) {
right_wnd_edge = pcb->snd_wnd + pcb->snd_wl2;
/* Update window. */
if (TCP_SEQ_LT(pcb->snd_wl1, seqno) ||
(pcb->snd_wl1 == seqno && TCP_SEQ_LT(pcb->snd_wl2, ackno)) ||
(pcb->snd_wl2 == ackno && tcphdr->wnd > pcb->snd_wnd)) {
pcb->snd_wnd = tcphdr->wnd;
pcb->snd_wl1 = seqno;
pcb->snd_wl2 = ackno;
if (pcb->snd_wnd > 0 && pcb->persist_backoff > 0) {
pcb->persist_backoff = 0;
}
LWIP_DEBUGF(TCP_WND_DEBUG, ("tcp_receive: window update %"U16_F"\n", pcb->snd_wnd));
#if TCP_WND_DEBUG
} else {
if (pcb->snd_wnd != tcphdr->wnd) {
LWIP_DEBUGF(TCP_WND_DEBUG,
("tcp_receive: no window update lastack %"U32_F" ackno %"
U32_F" wl1 %"U32_F" seqno %"U32_F" wl2 %"U32_F"\n",
pcb->lastack, ackno, pcb->snd_wl1, seqno, pcb->snd_wl2));
}
#endif /* TCP_WND_DEBUG */
}
/* (From Stevens TCP/IP Illustrated Vol II, p970.) Its only a
* duplicate ack if:
* 1) It doesn't ACK new data
* 2) length of received packet is zero (i.e. no payload)
* 3) the advertised window hasn't changed
* 4) There is outstanding unacknowledged data (retransmission timer running)
* 5) The ACK is == biggest ACK sequence number so far seen (snd_una)
*
* If it passes all five, should process as a dupack:
* a) dupacks < 3: do nothing
* b) dupacks == 3: fast retransmit
* c) dupacks > 3: increase cwnd
*
* If it only passes 1-3, should reset dupack counter (and add to
* stats, which we don't do in lwIP)
*
* If it only passes 1, should reset dupack counter
*
*/
/* Clause 1 */
if (TCP_SEQ_LEQ(ackno, pcb->lastack)) {
pcb->acked = 0;
/* Clause 2 */
if (tcplen == 0) {
/* Clause 3 */
if (pcb->snd_wl2 + pcb->snd_wnd == right_wnd_edge) {
/* Clause 4 */
if (pcb->rtime >= 0) {
/* Clause 5 */
if (pcb->lastack == ackno) {
found_dupack = 1;
if (pcb->dupacks + 1 > pcb->dupacks)
++pcb->dupacks;
if (pcb->dupacks > 3) {
/* Inflate the congestion window, but not if it means that
the value overflows. */
if ((u16_t)(pcb->cwnd + pcb->mss) > pcb->cwnd) {
pcb->cwnd += pcb->mss;
}
} else if (pcb->dupacks == 3) {
/* Do fast retransmit */
tcp_rexmit_fast(pcb);
}
}
}
}
}
/* If Clause (1) or more is true, but not a duplicate ack, reset
* count of consecutive duplicate acks */
if (!found_dupack) {
pcb->dupacks = 0;
}
} else if (TCP_SEQ_BETWEEN(ackno, pcb->lastack+1, pcb->snd_nxt)) {
/* We come here when the ACK acknowledges new data. */
/* Reset the "IN Fast Retransmit" flag, since we are no longer
in fast retransmit. Also reset the congestion window to the
slow start threshold. */
if (pcb->flags & TF_INFR) {
pcb->flags &= ~TF_INFR;
pcb->cwnd = pcb->ssthresh;
}
/* Reset the number of retransmissions. */
pcb->nrtx = 0;
/* Reset the retransmission time-out. */
pcb->rto = (pcb->sa >> 3) + pcb->sv;
/* Update the send buffer space. Diff between the two can never exceed 64K? */
pcb->acked = (u16_t)(ackno - pcb->lastack);
pcb->snd_buf += pcb->acked;
/* Reset the fast retransmit variables. */
pcb->dupacks = 0;
pcb->lastack = ackno;
/* Update the congestion control variables (cwnd and
ssthresh). */
if (pcb->state >= ESTABLISHED) {
if (pcb->cwnd < pcb->ssthresh) {
if ((u16_t)(pcb->cwnd + pcb->mss) > pcb->cwnd) {
pcb->cwnd += pcb->mss;
}
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: slow start cwnd %"U16_F"\n", pcb->cwnd));
} else {
u16_t new_cwnd = (pcb->cwnd + pcb->mss * pcb->mss / pcb->cwnd);
if (new_cwnd > pcb->cwnd) {
pcb->cwnd = new_cwnd;
}
LWIP_DEBUGF(TCP_CWND_DEBUG, ("tcp_receive: congestion avoidance cwnd %"U16_F"\n", pcb->cwnd));
}
}
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: ACK for %"U32_F", unacked->seqno %"U32_F":%"U32_F"\n",
ackno,
pcb->unacked != NULL?
ntohl(pcb->unacked->tcphdr->seqno): 0,
pcb->unacked != NULL?
ntohl(pcb->unacked->tcphdr->seqno) + TCP_TCPLEN(pcb->unacked): 0));
/* Remove segment from the unacknowledged list if the incoming
ACK acknowlegdes them. */
while (pcb->unacked != NULL &&
TCP_SEQ_LEQ(ntohl(pcb->unacked->tcphdr->seqno) +
TCP_TCPLEN(pcb->unacked), ackno)) {
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_receive: removing %"U32_F":%"U32_F" from pcb->unacked\n",
ntohl(pcb->unacked->tcphdr->seqno),
ntohl(pcb->unacked->tcphdr->seqno) +
TCP_TCPLEN(pcb->unacked)));
next = pcb->unacked;
pcb->unacked = pcb->unacked->next;
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_receive: queuelen %"U16_F" ... ", (u16_t)pcb->snd_queuelen));
LWIP_ASSERT("pcb->snd_queuelen >= pbuf_clen(next->p)", (pcb->snd_queuelen >= pbuf_clen(next->p)));
/* Prevent ACK for FIN to generate a sent event */
if ((pcb->acked != 0) && ((TCPH_FLAGS(next->tcphdr) & TCP_FIN) != 0)) {
pcb->acked--;
}
pcb->snd_queuelen -= pbuf_clen(next->p);
tcp_seg_free(next);
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("%"U16_F" (after freeing unacked)\n", (u16_t)pcb->snd_queuelen));
if (pcb->snd_queuelen != 0) {
LWIP_ASSERT("tcp_receive: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
}
/* If there's nothing left to acknowledge, stop the retransmit
timer, otherwise reset it to start again */
if(pcb->unacked == NULL)
pcb->rtime = -1;
else
pcb->rtime = 0;
pcb->polltmr = 0;
} else {
