/**
* Called by tcp_close() to send a segment including FIN flag but not data.
*
* @param pcb the tcp_pcb over which to send a segment
* @return ERR_OK if sent, another err_t otherwise
*/
err_t
tcp_send_fin(struct tcp_pcb *pcb)
{
/* first, try to add the fin to the last unsent segment */
if (pcb->unsent != NULL) {
struct tcp_seg *last_unsent;
for (last_unsent = pcb->unsent; last_unsent->next != NULL;
last_unsent = last_unsent->next);
if ((TCPH_FLAGS(last_unsent->tcphdr) & (TCP_SYN | TCP_FIN | TCP_RST)) == 0) {
/* no SYN/FIN/RST flag in the header, we can add the FIN flag */
TCPH_SET_FLAG(last_unsent->tcphdr, TCP_FIN);
return ERR_OK;
}
}
/* no data, no length, flags, copy=1, no optdata */
return tcp_enqueue_flags(pcb, TCP_FIN);
}
/*-----------------------------------------------------------------------------------*/
void
tcpdump(struct pbuf *p)
{
struct ip_hdr *iphdr;
#if LWIP_UDP
struct udp_hdr *udphdr;
#endif
#if LWIP_TCP
struct tcp_hdr *tcphdr;
char flags[5];
int i;
int len;
int offset;
#endif
if (file == NULL) {
return;
}
iphdr = (struct ip_hdr *)p->payload;
switch (IPH_PROTO(iphdr)) {
#if LWIP_TCP
case IP_PROTO_TCP:
tcphdr = (struct tcp_hdr *)((char *)iphdr + IP_HLEN);
pbuf_header(p, -IP_HLEN);
if (inet_chksum_pseudo(p, IP_PROTO_TCP, p->tot_len,
(ip4_addr_t *)&(iphdr->src),
(ip4_addr_t *)&(iphdr->dest)) != 0) {
LWIP_DEBUGF(TCPDUMP_DEBUG, ("tcpdump: IP checksum failed!\n"));
/*
fprintf(file, "chksum 0x%lx ", tcphdr->chksum);
tcphdr->chksum = 0;
fprintf(file, "should be 0x%lx ", inet_chksum_pseudo(p, (ip_addr_t *)&(iphdr->src),
(ip_addr_t *)&(iphdr->dest), IP_PROTO_TCP, p->tot_len));*/
fprintf(file, "!chksum ");
}
i = 0;
if (TCPH_FLAGS(tcphdr) & TCP_SYN) {
flags[i++] = 'S';
}
if (TCPH_FLAGS(tcphdr) & TCP_PSH) {
flags[i++] = 'P';
}
if (TCPH_FLAGS(tcphdr) & TCP_FIN) {
flags[i++] = 'F';
}
if (TCPH_FLAGS(tcphdr) & TCP_RST) {
flags[i++] = 'R';
}
if (i == 0) {
flags[i++] = '.';
}
flags[i++] = 0;
fprintf(file, "%d.%d.%d.%d.%u > %d.%d.%d.%d.%u: ",
(int)(ntohl(iphdr->src.addr) >> 24) & 0xff,
(int)(ntohl(iphdr->src.addr) >> 16) & 0xff,
(int)(ntohl(iphdr->src.addr) >> 8) & 0xff,
(int)(ntohl(iphdr->src.addr) >> 0) & 0xff,
ntohs(tcphdr->src),
(int)(ntohl(iphdr->dest.addr) >> 24) & 0xff,
(int)(ntohl(iphdr->dest.addr) >> 16) & 0xff,
(int)(ntohl(iphdr->dest.addr) >> 8) & 0xff,
(int)(ntohl(iphdr->dest.addr) >> 0) & 0xff,
ntohs(tcphdr->dest));
offset = TCPH_HDRLEN(tcphdr);
len = ntohs(IPH_LEN(iphdr)) - offset * 4 - IP_HLEN;
if (len != 0 || flags[0] != '.') {
fprintf(file, "%s %u:%u(%u) ", flags, ntohl(tcphdr->seqno),
ntohl(tcphdr->seqno) + len, len);
}
if (TCPH_FLAGS(tcphdr) & TCP_ACK) {
fprintf(file, "ack %u ", ntohl(tcphdr->ackno));
}
fprintf(file, "wnd %u\n", ntohs(tcphdr->wnd));
fflush(file);
pbuf_header(p, IP_HLEN);
break;
#endif /* LWIP_TCP */
#if LWIP_UDP
case IP_PROTO_UDP:
udphdr = (struct udp_hdr *)((char *)iphdr + IP_HLEN);
pbuf_header(p, -IP_HLEN);
if (inet_chksum_pseudo(p, IP_PROTO_UDP, p->tot_len,
(ip4_addr_t *)&(iphdr->src),
(ip4_addr_t *)&(iphdr->dest)) != 0) {
LWIP_DEBUGF(TCPDUMP_DEBUG, ("tcpdump: IP checksum failed!\n"));
/*
fprintf(file, "chksum 0x%lx ", tcphdr->chksum);
tcphdr->chksum = 0;
fprintf(file, "should be 0x%lx ", inet_chksum_pseudo(p, (ip_addr_t *)&(iphdr->src),
(ip_addr_t *)&(iphdr->dest), IP_PROTO_TCP, p->tot_len));*/
fprintf(file, "!chksum ");
}
fprintf(file, "%d.%d.%d.%d.%u > %d.%d.%d.%d.%u: ",
(int)(ntohl(iphdr->src.addr) >> 24) & 0xff,
(int)(ntohl(iphdr->src.addr) >> 16) & 0xff,
(int)(ntohl(iphdr->src.addr) >> 8) & 0xff,
(int)(ntohl(iphdr->src.addr) >> 0) & 0xff,
(int)(ntohl(iphdr->dest.addr) >> 24) & 0xff,
ntohs(udphdr->src),
(int)(ntohl(iphdr->dest.addr) >> 16) & 0xff,
(int)(ntohl(iphdr->dest.addr) >> 8) & 0xff,
(int)(ntohl(iphdr->dest.addr) >> 0) & 0xff,
ntohs(udphdr->dest));
fprintf(file, "U ");
len = ntohs(IPH_LEN(iphdr)) - sizeof(struct udp_hdr) - IP_HLEN;
fprintf(file, " %d\n", len);
fflush(file);
pbuf_header(p, IP_HLEN);
break;
#endif /* LWIP_UDP */
default:
LWIP_DEBUGF(TCPDUMP_DEBUG, ("unhandled IP protocol: %d\n", (int)IPH_PROTO(iphdr)));
break;
}
}
/**
* Enqueue either data or TCP options (but not both) for tranmission
*
*
*
* @arg pcb Protocol control block for the TCP connection to enqueue data for.
* @arg arg Pointer to the data to be enqueued for sending.
* @arg len Data length in bytes
* @arg flags
* @arg copy 1 if data must be copied, 0 if data is non-volatile and can be
* referenced.
* @arg optdata
* @arg optlen
*/
err_t
tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
u8_t flags, u8_t copy,
u8_t *optdata, u8_t optlen)
{
struct pbuf *p;
struct tcp_seg *seg, *useg, *queue;
u32_t left, seqno;
u16_t seglen;
void *ptr;
u8_t queuelen;
LWIP_DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue(pcb=%p, arg=%p, len=%"U16_F", flags=%"X16_F", copy=%"U16_F")\n",
(void *)pcb, arg, len, (u16_t)flags, (u16_t)copy));
LWIP_ASSERT("tcp_enqueue: len == 0 || optlen == 0 (programmer violates API)",
len == 0 || optlen == 0);
LWIP_ASSERT("tcp_enqueue: arg == NULL || optdata == NULL (programmer violates API)",
arg == NULL || optdata == NULL);
/* fail on too much data */
if (len > pcb->snd_buf) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 3, ("tcp_enqueue: too much data (len=%"U16_F" > snd_buf=%"U16_F")\n", len, pcb->snd_buf));
return ERR_MEM;
}
left = len;
ptr = arg;
/* seqno will be the sequence number of the first segment enqueued
* by the call to this function. */
seqno = pcb->snd_lbb;
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_enqueue: queuelen: %"U16_F"\n", (u16_t)pcb->snd_queuelen));
/* If total number of pbufs on the unsent/unacked queues exceeds the
* configured maximum, return an error */
queuelen = pcb->snd_queuelen;
if (queuelen >= TCP_SND_QUEUELEN) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 3, ("tcp_enqueue: too long queue %"U16_F" (max %"U16_F")\n", queuelen, TCP_SND_QUEUELEN));
TCP_STATS_INC(tcp.memerr);
return ERR_MEM;
}
if (queuelen != 0) {
LWIP_ASSERT("tcp_enqueue: pbufs on queue => at least one queue non-empty",
pcb->unacked != NULL || pcb->unsent != NULL);
} else {
LWIP_ASSERT("tcp_enqueue: no pbufs on queue => both queues empty",
pcb->unacked == NULL && pcb->unsent == NULL);
}
/* First, break up the data into segments and tuck them together in
* the local "queue" variable. */
useg = queue = seg = NULL;
seglen = 0;
while (queue == NULL || left > 0) {
/* The segment length should be the MSS if the data to be enqueued
* is larger than the MSS. */
seglen = left > pcb->mss? pcb->mss: left;
/* Allocate memory for tcp_seg, and fill in fields. */
seg = memp_malloc(MEMP_TCP_SEG);
if (seg == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: could not allocate memory for tcp_seg\n"));
goto memerr;
}
seg->next = NULL;
seg->p = NULL;
/* first segment of to-be-queued data? */
if (queue == NULL) {
queue = seg;
}
/* subsequent segments of to-be-queued data */
else {
/* Attach the segment to the end of the queued segments */
LWIP_ASSERT("useg != NULL", useg != NULL);
useg->next = seg;
}
/* remember last segment of to-be-queued data for next iteration */
useg = seg;
/* If copy is set, memory should be allocated
* and data copied into pbuf, otherwise data comes from
* ROM or other static memory, and need not be copied. If
* optdata is != NULL, we have options instead of data. */
/* options? */
if (optdata != NULL) {
if ((seg->p = pbuf_alloc(PBUF_TRANSPORT, optlen, PBUF_RAM)) == NULL) {
goto memerr;
}
++queuelen;
seg->dataptr = seg->p->payload;
}
/* copy from volatile memory? */
else if (copy) {
if ((seg->p = pbuf_alloc(PBUF_TRANSPORT, seglen, PBUF_RAM)) == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue : could not allocate memory for pbuf copy size %"U16_F"\n", seglen));
goto memerr;
}
++queuelen;
if (arg != NULL) {
memcpy(seg->p->payload, ptr, seglen);
}
seg->dataptr = seg->p->payload;
}
/* do not copy data */
else {
/* First, allocate a pbuf for holding the data.
* since the referenced data is available at least until it is sent out on the
* link (as it has to be ACKed by the remote party) we can safely use PBUF_ROM
* instead of PBUF_REF here.
*/
if ((p = pbuf_alloc(PBUF_TRANSPORT, seglen, PBUF_ROM)) == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: could not allocate memory for zero-copy pbuf\n"));
goto memerr;
}
++queuelen;
/* reference the non-volatile payload data */
p->payload = ptr;
seg->dataptr = ptr;
/* Second, allocate a pbuf for the headers. */
if ((seg->p = pbuf_alloc(PBUF_TRANSPORT, 0, PBUF_RAM)) == NULL) {
/* If allocation fails, we have to deallocate the data pbuf as
* well. */
pbuf_free(p);
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: could not allocate memory for header pbuf\n"));
goto memerr;
}
++queuelen;
/* Concatenate the headers and data pbufs together. */
pbuf_cat(seg->p/*header*/, p/*data*/);
p = NULL;
}
/* Now that there are more segments queued, we check again if the
length of the queue exceeds the configured maximum. */
if (queuelen > TCP_SND_QUEUELEN) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: queue too long %"U16_F" (%"U16_F")\n", queuelen, TCP_SND_QUEUELEN));
goto memerr;
}
seg->len = seglen;
/* build TCP header */
if (pbuf_header(seg->p, TCP_HLEN)) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: no room for TCP header in pbuf.\n"));
TCP_STATS_INC(tcp.err);
goto memerr;
}
seg->tcphdr = seg->p->payload;
seg->tcphdr->src = htons(pcb->local_port);
seg->tcphdr->dest = htons(pcb->remote_port);
seg->tcphdr->seqno = htonl(seqno);
seg->tcphdr->urgp = 0;
TCPH_FLAGS_SET(seg->tcphdr, flags);
/* don't fill in tcphdr->ackno and tcphdr->wnd until later */
/* Copy the options into the header, if they are present. */
if (optdata == NULL) {
TCPH_HDRLEN_SET(seg->tcphdr, 5);
}
else {
TCPH_HDRLEN_SET(seg->tcphdr, (5 + optlen / 4));
/* Copy options into data portion of segment.
Options can thus only be sent in non data carrying
segments such as SYN|ACK. */
memcpy(seg->dataptr, optdata, optlen);
}
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | DBG_TRACE, ("tcp_enqueue: queueing %"U32_F":%"U32_F" (0x%"X16_F")\n",
ntohl(seg->tcphdr->seqno),
ntohl(seg->tcphdr->seqno) + TCP_TCPLEN(seg),
(u16_t)flags));
left -= seglen;
seqno += seglen;
ptr = (void *)((u8_t *)ptr + seglen);
}
/* Now that the data to be enqueued has been broken up into TCP
segments in the queue variable, we add them to the end of the
pcb->unsent queue. */
if (pcb->unsent == NULL) {
useg = NULL;
}
else {
for (useg = pcb->unsent; useg->next != NULL; useg = useg->next);
}
/* { useg is last segment on the unsent queue, NULL if list is empty } */
/* If there is room in the last pbuf on the unsent queue,
chain the first pbuf on the queue together with that. */
if (useg != NULL &&
TCP_TCPLEN(useg) != 0 &&
!(TCPH_FLAGS(useg->tcphdr) & (TCP_SYN | TCP_FIN)) &&
!(flags & (TCP_SYN | TCP_FIN)) &&
/* fit within max seg size */
useg->len + queue->len <= pcb->mss) {
/* Remove TCP header from first segment of our to-be-queued list */
pbuf_header(queue->p, -TCP_HLEN);
pbuf_cat(useg->p, queue->p);
useg->len += queue->len;
useg->next = queue->next;
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | DBG_TRACE | DBG_STATE, ("tcp_enqueue: chaining segments, new len %"U16_F"\n", useg->len));
if (seg == queue) {
seg = NULL;
}
memp_free(MEMP_TCP_SEG, queue);
}
else {
/* empty list */
if (useg == NULL) {
/* initialize list with this segment */
pcb->unsent = queue;
}
/* enqueue segment */
else {
useg->next = queue;
}
}
if ((flags & TCP_SYN) || (flags & TCP_FIN)) {
++len;
}
pcb->snd_lbb += len;
pcb->snd_buf -= len;
/* update number of segments on the queues */
pcb->snd_queuelen = queuelen;
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_enqueue: %"S16_F" (after enqueued)\n", pcb->snd_queuelen));
if (pcb->snd_queuelen != 0) {
LWIP_ASSERT("tcp_enqueue: valid queue length",
pcb->unacked != NULL || pcb->unsent != NULL);
}
/* Set the PSH flag in the last segment that we enqueued, but only
if the segment has data (indicated by seglen > 0). */
if (seg != NULL && seglen > 0 && seg->tcphdr != NULL) {
TCPH_SET_FLAG(seg->tcphdr, TCP_PSH);
}
return ERR_OK;
memerr:
TCP_STATS_INC(tcp.memerr);
if (queue != NULL) {
tcp_segs_free(queue);
}
if (pcb->snd_queuelen != 0) {
LWIP_ASSERT("tcp_enqueue: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
LWIP_DEBUGF(TCP_QLEN_DEBUG | DBG_STATE, ("tcp_enqueue: %"S16_F" (with mem err)\n", pcb->snd_queuelen));
return ERR_MEM;
}
/**
* This function does the actual transmission of a packet. The packet is
* contained in the pbuf that is passed to the function. This pbuf
* can be chained.
*
* @param netif
* the lwip network interface structure for this netfrontif
* @param p
* the packet to send (e.g. IP packet including MAC addresses and type)
* @return
* ERR_OK when the packet could be enqueued for sending; an err_t value otherwise
*/
static inline err_t netfrontif_transmit(struct netif *netif, struct pbuf *p)
{
struct netfrontif *nfi = netif->state;
#if LWIP_CHECKSUM_PARTIAL || defined CONFIG_LWIP_BATCHTX
s16_t ip_hdr_offset;
const struct eth_hdr *ethhdr;
const struct ip_hdr *iphdr;
#endif /* LWIP_CHECKSUM_PARTIAL || defined CONFIG_LWIP_BATCHTX */
#ifdef CONFIG_LWIP_BATCHTX
const struct tcp_hdr *tcphdr;
#endif /* CONFIG_LWIP_BATCHTX */
int tso = 0;
int push = 1;
err_t err;
LWIP_DEBUGF(NETIF_DEBUG, ("netfrontif_transmit: %c%c: "
"Transmitting %u bytes\n",
netif->name[0], netif->name[1],
p->tot_len));
#if LWIP_CHECKSUM_PARTIAL || defined CONFIG_LWIP_BATCHTX
/* detect if payload contains a TCP packet */
/* NOTE: We assume here that all protocol headers are in the first pbuf of a pbuf chain! */
ip_hdr_offset = SIZEOF_ETH_HDR;
ethhdr = (struct eth_hdr *) p->payload;
#if ETHARP_SUPPORT_VLAN
if (type == PP_HTONS(ETHTYPE_VLAN)) {
type = ((struct eth_vlan_hdr*)(((uintptr_t)ethhdr) + SIZEOF_ETH_HDR))->tpid;
ip_hdr_offset = SIZEOF_ETH_HDR + SIZEOF_VLAN_HDR;
}
#endif /* ETHARP_SUPPORT_VLAN */
/* TODO: PPP support? */
switch (ethhdr->type) {
case PP_HTONS(ETHTYPE_IP):
iphdr = (struct ip_hdr *)((uintptr_t) p->payload + ip_hdr_offset);
if (IPH_PROTO(iphdr) != IP_PROTO_TCP) {
goto xmit; /* IPv4 but not TCP */
}
#if LWIP_CHECKSUM_PARTIAL
tso = XEN_NETIF_GSO_TYPE_TCPV4; /* TCPv4 segmentation and checksum offloading */
#endif /* LWIP_CHECKSUM_PARTIAL */
#ifdef CONFIG_LWIP_BATCHTX
/* push only when FIN, RST, PSH, or URG flag is set */
tcphdr = (struct tcp_hdr *)((uintptr_t) p->payload + ip_hdr_offset + (IPH_HL(iphdr) * 4));
push = (TCPH_FLAGS(tcphdr) & (TCP_FIN | TCP_RST | TCP_PSH | TCP_URG));
#endif /* CONFIG_LWIP_BATCHTX */
break;
#if IPV6_SUPPORT
case PP_HTONS(ETHTYPE_IPV6):
if (IP6H_NEXTH((struct ip6_hdr *)((uintptr_t) p->payload + ip_hdr_offset)) != IP6_NEXTH_TCP)
goto xmit; /* IPv6 but not TCP */
#if LWIP_CHECKSUM_PARTIAL
tso = XEN_NETIF_GSO_TYPE_TCPV6; /* TCPv6 segmentation and checksum offloading */
#endif /* LWIP_CHECKSUM_PARTIAL */
#ifdef CONFIG_LWIP_BATCHTX
/* push only when FIN, RST, PSH, or URG flag is set */
#error "TSOv6 is not yet supported. Please add it"
tcphdr = NULL;
push = (TCPH_FLAGS(tcphdr) & (TCP_FIN | TCP_RST | TCP_PSH | TCP_URG));
#endif /* CONFIG_LWIP_BATCHTX */
break;
#endif /* IPV6_SUPPORT */
default:
break; /* non-IP packet */
}
#endif /* LWIP_CHECKSUM_PARTIAL || defined CONFIG_LWIP_BATCHTX */
xmit:
#if ETH_PAD_SIZE
pbuf_header(p, -ETH_PAD_SIZE); /* drop the padding word */
#endif
err = netfront_xmit_pbuf(nfi->dev, p, tso, push);
if (likely(err == ERR_OK)) {
LINK_STATS_INC(link.xmit);
} else {
LWIP_DEBUGF(NETIF_DEBUG, ("netfrontif_transmit: transmission failed, dropping packet: %d\n", err));
LINK_STATS_INC(link.drop);
}
#if ETH_PAD_SIZE
pbuf_header(p, ETH_PAD_SIZE); /* reclaim the padding word */
#endif
return err;
}
/**
* The initial input processing of TCP. It verifies the TCP header, demultiplexes
* the segment between the PCBs and passes it on to tcp_process(), which implements
* the TCP finite state machine. This function is called by the IP layer (in
* ip_input()).
*
* @param p received TCP segment to process (p->payload pointing to the IP header)
* @param inp network interface on which this segment was received
*/
void
tcp_input(struct pbuf *p, struct netif *inp)
{
struct tcp_pcb *pcb, *prev;
struct tcp_pcb_listen *lpcb;
u8_t hdrlen;
err_t err;
PERF_START;
TCP_STATS_INC(tcp.recv);
snmp_inc_tcpinsegs();
iphdr = p->payload;
tcphdr = (struct tcp_hdr *)((u8_t *)p->payload + IPH_HL(iphdr) * 4);
#if TCP_INPUT_DEBUG
tcp_debug_print(tcphdr);
#endif
/* remove header from payload */
if (pbuf_header(p, -((s16_t)(IPH_HL(iphdr) * 4))) || (p->tot_len < sizeof(struct tcp_hdr))) {
/* drop short packets */
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: short packet (%"U16_F" bytes) discarded\n", p->tot_len));
TCP_STATS_INC(tcp.lenerr);
TCP_STATS_INC(tcp.drop);
snmp_inc_tcpinerrs();
pbuf_free(p);
return;
}
/* Don't even process incoming broadcasts/multicasts. */
if (ip_addr_isbroadcast(&(iphdr->dest), inp) ||
ip_addr_ismulticast(&(iphdr->dest))) {
TCP_STATS_INC(tcp.proterr);
TCP_STATS_INC(tcp.drop);
snmp_inc_tcpinerrs();
pbuf_free(p);
return;
}
#if CHECKSUM_CHECK_TCP
/* Verify TCP checksum. */
if (inet_chksum_pseudo(p, (struct ip_addr *)&(iphdr->src),
(struct ip_addr *)&(iphdr->dest),
IP_PROTO_TCP, p->tot_len) != 0) {
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packet discarded due to failing checksum 0x%04"X16_F"\n",
inet_chksum_pseudo(p, (struct ip_addr *)&(iphdr->src), (struct ip_addr *)&(iphdr->dest),
IP_PROTO_TCP, p->tot_len)));
#if TCP_DEBUG
tcp_debug_print(tcphdr);
#endif /* TCP_DEBUG */
TCP_STATS_INC(tcp.chkerr);
TCP_STATS_INC(tcp.drop);
snmp_inc_tcpinerrs();
pbuf_free(p);
return;
}
#endif
/* Move the payload pointer in the pbuf so that it points to the
TCP data instead of the TCP header. */
hdrlen = TCPH_HDRLEN(tcphdr);
if(pbuf_header(p, -(hdrlen * 4))){
/* drop short packets */
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: short packet\n"));
TCP_STATS_INC(tcp.lenerr);
TCP_STATS_INC(tcp.drop);
snmp_inc_tcpinerrs();
pbuf_free(p);
return;
}
/* Convert fields in TCP header to host byte order. */
tcphdr->src = ntohs(tcphdr->src);
tcphdr->dest = ntohs(tcphdr->dest);
seqno = tcphdr->seqno = ntohl(tcphdr->seqno);
ackno = tcphdr->ackno = ntohl(tcphdr->ackno);
tcphdr->wnd = ntohs(tcphdr->wnd);
flags = TCPH_FLAGS(tcphdr) & TCP_FLAGS;
tcplen = p->tot_len + ((flags & TCP_FIN || flags & TCP_SYN)? 1: 0);
/* Demultiplex an incoming segment. First, we check if it is destined
for an active connection. */
prev = NULL;
for(pcb = tcp_active_pcbs; pcb != NULL; pcb = pcb->next) {
LWIP_ASSERT("tcp_input: active pcb->state != CLOSED", pcb->state != CLOSED);
LWIP_ASSERT("tcp_input: active pcb->state != TIME-WAIT", pcb->state != TIME_WAIT);
LWIP_ASSERT("tcp_input: active pcb->state != LISTEN", pcb->state != LISTEN);
if (pcb->remote_port == tcphdr->src &&
pcb->local_port == tcphdr->dest &&
ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src)) &&
ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest))) {
/* Move this PCB to the front of the list so that subsequent
lookups will be faster (we exploit locality in TCP segment
arrivals). */
LWIP_ASSERT("tcp_input: pcb->next != pcb (before cache)", pcb->next != pcb);
if (prev != NULL) {
prev->next = pcb->next;
pcb->next = tcp_active_pcbs;
tcp_active_pcbs = pcb;
}
LWIP_ASSERT("tcp_input: pcb->next != pcb (after cache)", pcb->next != pcb);
break;
}
prev = pcb;
}
if (pcb == NULL) {
/* If it did not go to an active connection, we check the connections
in the TIME-WAIT state. */
for(pcb = tcp_tw_pcbs; pcb != NULL; pcb = pcb->next) {
LWIP_ASSERT("tcp_input: TIME-WAIT pcb->state == TIME-WAIT", pcb->state == TIME_WAIT);
if (pcb->remote_port == tcphdr->src &&
pcb->local_port == tcphdr->dest &&
ip_addr_cmp(&(pcb->remote_ip), &(iphdr->src)) &&
ip_addr_cmp(&(pcb->local_ip), &(iphdr->dest))) {
/* We don't really care enough to move this PCB to the front
of the list since we are not very likely to receive that
many segments for connections in TIME-WAIT. */
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for TIME_WAITing connection.\n"));
tcp_timewait_input(pcb);
pbuf_free(p);
return;
}
}
/* Finally, if we still did not get a match, we check all PCBs that
are LISTENing for incoming connections. */
prev = NULL;
for(lpcb = tcp_listen_pcbs.listen_pcbs; lpcb != NULL; lpcb = lpcb->next) {
if ((ip_addr_isany(&(lpcb->local_ip)) ||
ip_addr_cmp(&(lpcb->local_ip), &(iphdr->dest))) &&
lpcb->local_port == tcphdr->dest) {
/* Move this PCB to the front of the list so that subsequent
lookups will be faster (we exploit locality in TCP segment
arrivals). */
if (prev != NULL) {
((struct tcp_pcb_listen *)prev)->next = lpcb->next;
/* our successor is the remainder of the listening list */
lpcb->next = tcp_listen_pcbs.listen_pcbs;
/* put this listening pcb at the head of the listening list */
tcp_listen_pcbs.listen_pcbs = lpcb;
}
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: packed for LISTENing connection.\n"));
tcp_listen_input(lpcb);
pbuf_free(p);
return;
}
prev = (struct tcp_pcb *)lpcb;
}
}
#if TCP_INPUT_DEBUG
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("+-+-+-+-+-+-+-+-+-+-+-+-+-+- tcp_input: flags "));
tcp_debug_print_flags(TCPH_FLAGS(tcphdr));
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("-+-+-+-+-+-+-+-+-+-+-+-+-+-+\n"));
#endif /* TCP_INPUT_DEBUG */
if (pcb != NULL) {
/* The incoming segment belongs to a connection. */
#if TCP_INPUT_DEBUG
#if TCP_DEBUG
tcp_debug_print_state(pcb->state);
#endif /* TCP_DEBUG */
#endif /* TCP_INPUT_DEBUG */
/* Set up a tcp_seg structure. */
inseg.next = NULL;
inseg.len = p->tot_len;
inseg.dataptr = p->payload;
inseg.p = p;
inseg.tcphdr = tcphdr;
recv_data = NULL;
recv_flags = 0;
/* If there is data which was previously "refused" by upper layer */
if (pcb->refused_data != NULL) {
/* Notify again application with data previously received. */
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: notify kept packet\n"));
TCP_EVENT_RECV(pcb, pcb->refused_data, ERR_OK, err);
if (err == ERR_OK) {
pcb->refused_data = NULL;
} else {
/* drop incoming packets, because pcb is "full" */
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: drop incoming packets, because pcb is \"full\"\n"));
TCP_STATS_INC(tcp.drop);
snmp_inc_tcpinerrs();
pbuf_free(p);
return;
}
}
tcp_input_pcb = pcb;
err = tcp_process(pcb);
tcp_input_pcb = NULL;
/* A return value of ERR_ABRT means that tcp_abort() was called
and that the pcb has been freed. If so, we don't do anything. */
if (err != ERR_ABRT) {
if (recv_flags & TF_RESET) {
/* TF_RESET means that the connection was reset by the other
end. We then call the error callback to inform the
application that the connection is dead before we
deallocate the PCB. */
TCP_EVENT_ERR(pcb->errf, pcb->callback_arg, ERR_RST);
tcp_pcb_remove(&tcp_active_pcbs, pcb);
memp_free(MEMP_TCP_PCB, pcb);
} else if (recv_flags & TF_CLOSED) {
/* The connection has been closed and we will deallocate the
PCB. */
tcp_pcb_remove(&tcp_active_pcbs, pcb);
memp_free(MEMP_TCP_PCB, pcb);
} else {
err = ERR_OK;
/* If the application has registered a "sent" function to be
called when new send buffer space is available, we call it
now. */
if (pcb->acked > 0) {
TCP_EVENT_SENT(pcb, pcb->acked, err);
}
if (recv_data != NULL) {
if(flags & TCP_PSH) {
recv_data->flags |= PBUF_FLAG_PUSH;
}
/* Notify application that data has been received. */
TCP_EVENT_RECV(pcb, recv_data, ERR_OK, err);
/* If the upper layer can't receive this data, store it */
if (err != ERR_OK) {
pcb->refused_data = recv_data;
LWIP_DEBUGF(TCP_INPUT_DEBUG, ("tcp_input: keep incoming packet, because pcb is \"full\"\n"));
}
}
/* If a FIN segment was received, we call the callback
function with a NULL buffer to indicate EOF. */
if (recv_flags & TF_GOT_FIN) {
TCP_EVENT_RECV(pcb, NULL, ERR_OK, err);
}
/* If there were no errors, we try to send something out. */
if (err == ERR_OK) {
tcp_output(pcb);
}
}
}
/* give up our reference to inseg.p */
if (inseg.p != NULL)
{
pbuf_free(inseg.p);
inseg.p = NULL;
}
#if TCP_INPUT_DEBUG
#if TCP_DEBUG
tcp_debug_print_state(pcb->state);
#endif /* TCP_DEBUG */
#endif /* TCP_INPUT_DEBUG */
} else {
/* If no matching PCB was found, send a TCP RST (reset) to the
sender. */
LWIP_DEBUGF(TCP_RST_DEBUG, ("tcp_input: no PCB match found, resetting.\n"));
if (!(TCPH_FLAGS(tcphdr) & TCP_RST)) {
TCP_STATS_INC(tcp.proterr);
TCP_STATS_INC(tcp.drop);
tcp_rst(ackno, seqno + tcplen,
&(iphdr->dest), &(iphdr->src),
tcphdr->dest, tcphdr->src);
}
pbuf_free(p);
}
LWIP_ASSERT("tcp_input: tcp_pcbs_sane()", tcp_pcbs_sane());
PERF_STOP("tcp_input");
}
err_t tcp_enqueue (
struct tcp_pcb* pcb, void* arg, u16_t len,
u8_t flags, u8_t copy,
u8_t* optdata, u8_t optlen
) {
struct pbuf* p;
struct tcp_seg* seg, *useg, *queue;
u32_t left, seqno;
u16_t seglen;
void* ptr;
u8_t queuelen;
flags_t lPCBFlags = pcb -> flags;
int iSegCNT = 0;
left = len;
ptr = arg;
if ( len > pcb -> snd_buf ) return ERR_MEM;
seqno = pcb -> snd_lbb;
queue = NULL;
queuelen = pcb -> snd_queuelen;
if ( queuelen >= TCP_SND_QUEUELEN ) goto memerr;
seg = useg = NULL;
seglen = 0;
while ( !queue || left > 0 ) {
if ( lPCBFlags & TF_EVENSEG ) {
++iSegCNT;
seglen = left > pcb -> mss ? pcb -> mss
: (((iSegCNT%2) == 1)? ((left + 1) / 2): left);
} else seglen = left > pcb -> mss ? pcb -> mss : left;
seg = memp_malloc ( MEMP_TCP_SEG );
if ( !seg ) goto memerr;
seg -> next = NULL;
seg -> p = NULL;
if ( !queue )
useg = queue = seg;
else {
useg -> next = seg;
useg = seg;
} /* end else */
if (optdata != NULL) {
if ((seg->p = pbuf_alloc(PBUF_TRANSPORT, optlen, PBUF_RAM)) == NULL) {
goto memerr;
}
++queuelen;
seg->dataptr = seg->p->payload;
}
else if (copy) {
if ((seg->p = pbuf_alloc(PBUF_TRANSPORT, seglen, PBUF_RAM)) == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue : could not allocate memory for pbuf copy size %u\n", seglen));
goto memerr;
}
++queuelen;
if (arg != NULL) {
mips_memcpy(seg->p->payload, ptr, seglen);
}
seg->dataptr = seg->p->payload;
}
/* do not copy data */
else {
/* first, allocate a pbuf for holding the data.
* since the referenced data is available at least until it is sent out on the
* link (as it has to be ACKed by the remote party) we can safely use PBUF_ROM
* instead of PBUF_REF here.
*/
if ((p = pbuf_alloc(PBUF_TRANSPORT, seglen, PBUF_ROM)) == NULL) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: could not allocate memory for zero-copy pbuf\n"));
goto memerr;
}
++queuelen;
p->payload = ptr;
seg->dataptr = ptr;
/* Second, allocate a pbuf for the headers. */
if ((seg->p = pbuf_alloc(PBUF_TRANSPORT, 0, PBUF_RAM)) == NULL) {
/* If allocation fails, we have to deallocate the data pbuf as
* well. */
pbuf_free(p);
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: could not allocate memory for header pbuf\n"));
goto memerr;
}
++queuelen;
/* Concatenate the headers and data pbufs together. */
pbuf_cat(seg->p, p);
p = NULL;
}
/* Now that there are more segments queued, we check again if the
length of the queue exceeds the configured maximum. */
if (queuelen > TCP_SND_QUEUELEN) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: queue too long %u (%u)\n", queuelen, TCP_SND_QUEUELEN));
goto memerr;
}
seg->len = seglen;
if (pbuf_header(seg->p, TCP_HLEN)) {
LWIP_DEBUGF(TCP_OUTPUT_DEBUG | 2, ("tcp_enqueue: no room for TCP header in pbuf.\n"));
TCP_STATS_INC(tcp.err);
goto memerr;
}
seg->tcphdr = seg->p->payload;
seg->tcphdr->src = htons(pcb->local_port);
seg->tcphdr->dest = htons(pcb->remote_port);
seg->tcphdr->seqno = htonl(seqno);
seg->tcphdr->urgp = 0;
TCPH_FLAGS_SET(seg->tcphdr, flags);
/* don't fill in tcphdr->ackno and tcphdr->wnd until later */
/* Copy the options into the header, if they are present. */
if (optdata == NULL) {
TCPH_HDRLEN_SET(seg->tcphdr, 5);
}
else {
TCPH_HDRLEN_SET(seg->tcphdr, (5 + optlen / 4));
/* Copy options into data portion of segment.
Options can thus only be sent in non data carrying
segments such as SYN|ACK. */
mips_memcpy(seg->dataptr, optdata, optlen);
}
left -= seglen;
seqno += seglen;
ptr = (void *)((char *)ptr + seglen);
}
/* Now that the data to be enqueued has been broken up into TCP
segments in the queue variable, we add them to the end of the
pcb->unsent queue. */
if (pcb->unsent == NULL) {
useg = NULL;
}
else {
for (useg = pcb->unsent; useg->next != NULL; useg = useg->next);
}
/* If there is room in the last pbuf on the unsent queue,
chain the first pbuf on the queue together with that. */
if (useg != NULL &&
TCP_TCPLEN(useg) != 0 &&
!(TCPH_FLAGS(useg->tcphdr) & (TCP_SYN | TCP_FIN)) &&
!(flags & (TCP_SYN | TCP_FIN)) &&
useg->len + queue->len <= pcb->mss) {
/* Remove TCP header from first segment. */
pbuf_header(queue->p, -TCP_HLEN);
pbuf_cat(useg->p, queue->p);
useg->len += queue->len;
useg->next = queue->next;
if (seg == queue) seg = NULL;
memp_free(MEMP_TCP_SEG, queue);
}
else {
if (useg == NULL) {
pcb->unsent = queue;
}
else {
useg->next = queue;
}
}
if ((flags & TCP_SYN) || (flags & TCP_FIN)) {
++len;
}
pcb->snd_lbb += len;
pcb->snd_buf -= len;
pcb->snd_queuelen = queuelen;
LWIP_DEBUGF(TCP_QLEN_DEBUG, ("tcp_enqueue: %d (after enqueued)\n", pcb->snd_queuelen));
if (pcb->snd_queuelen != 0) {
LWIP_ASSERT("tcp_enqueue: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
/* Set the PSH flag in the last segment that we enqueued, but only
if the segment has data (indicated by seglen > 0). */
if (seg != NULL && seglen > 0 && seg->tcphdr != NULL) {
TCPH_SET_FLAG(seg->tcphdr, TCP_PSH);
}
return ERR_OK;
memerr:
TCP_STATS_INC(tcp.memerr);
if (queue != NULL) {
tcp_segs_free(queue);
}
if (pcb->snd_queuelen != 0) {
LWIP_ASSERT("tcp_enqueue: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
LWIP_DEBUGF(TCP_QLEN_DEBUG | DBG_STATE, ("tcp_enqueue: %d (with mem err)\n", pcb->snd_queuelen));
return ERR_MEM;
}
/**
* 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;
}
/**
* Send persist timer zero-window probes to keep a connection active
* when a window update is lost.
*
* Called by tcp_slowtmr()
*
* @param pcb the tcp_pcb for which to send a zero-window probe packet
*/
void
tcp_zero_window_probe(struct tcp_pcb *pcb)
{
struct pbuf *p;
struct tcp_hdr *tcphdr;
struct tcp_seg *seg;
u16_t len;
u8_t is_fin;
LWIP_DEBUGF(TCP_DEBUG,
("tcp_zero_window_probe: sending ZERO WINDOW probe to %"
U16_F".%"U16_F".%"U16_F".%"U16_F"\n",
ip4_addr1_16(&pcb->remote_ip), ip4_addr2_16(&pcb->remote_ip),
ip4_addr3_16(&pcb->remote_ip), ip4_addr4_16(&pcb->remote_ip)));
LWIP_DEBUGF(TCP_DEBUG,
("tcp_zero_window_probe: tcp_ticks %"U32_F
" pcb->tmr %"U32_F" pcb->keep_cnt_sent %"U16_F"\n",
tcp_ticks, pcb->tmr, pcb->keep_cnt_sent));
seg = pcb->unacked;
if(seg == NULL) {
seg = pcb->unsent;
}
if(seg == NULL) {
return;
}
is_fin = ((TCPH_FLAGS(seg->tcphdr) & TCP_FIN) != 0) && (seg->len == 0);
/* we want to send one seqno: either FIN or data (no options) */
len = is_fin ? 0 : 1;
p = tcp_output_alloc_header(pcb, 0, len, seg->tcphdr->seqno);
if(p == NULL) {
LWIP_DEBUGF(TCP_DEBUG, ("tcp_zero_window_probe: no memory for pbuf\n"));
return;
}
tcphdr = (struct tcp_hdr *)p->payload;
if (is_fin) {
/* FIN segment, no data */
TCPH_FLAGS_SET(tcphdr, TCP_ACK | TCP_FIN);
} else {
/* Data segment, copy in one byte from the head of the unacked queue */
*((char *)p->payload + TCP_HLEN) = *(char *)seg->dataptr;
}
#if CHECKSUM_GEN_TCP
tcphdr->chksum = inet_chksum_pseudo(p, &pcb->local_ip, &pcb->remote_ip,
IP_PROTO_TCP, (u16_t)p->tot_len);
#endif
TCP_STATS_INC(tcp.xmit);
/* Send output to IP */
#if LWIP_NETIF_HWADDRHINT
ip_output_hinted(p, &pcb->local_ip, &pcb->remote_ip, pcb->ttl, 0, IP_PROTO_TCP,
&(pcb->addr_hint));
#elif LWIP_3RD_PARTY_L3
pcb->ip_output(p, pcb, 0);
#else /* LWIP_NETIF_HWADDRHINT*/
ip_output(p, &pcb->local_ip, &pcb->remote_ip, pcb->ttl, 0, IP_PROTO_TCP);
#endif /* LWIP_NETIF_HWADDRHINT*/
tcp_tx_pbuf_free(pcb, p);
LWIP_DEBUGF(TCP_DEBUG, ("tcp_zero_window_probe: seqno %"U32_F
" ackno %"U32_F".\n",
pcb->snd_nxt - 1, pcb->rcv_nxt));
}
/*-----------------------------------------------------------------------------------*/
err_t
tcp_enqueue(struct tcp_pcb *pcb, void *arg, u16_t len,
u8_t flags, u8_t copy,
u8_t *optdata, u8_t optlen)
{
struct pbuf *p;
struct tcp_seg *seg, *useg, *queue;
u32_t left, seqno;
u16_t seglen;
void *ptr;
u8_t queuelen;
left = len;
ptr = arg;
if(len > pcb->snd_buf) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: too much data %d\n", len));
return ERR_MEM;
}
seqno = pcb->snd_lbb;
queue = NULL;
DEBUGF(TCP_QLEN_DEBUG, ("tcp_enqueue: %d\n", pcb->snd_queuelen));
queuelen = pcb->snd_queuelen;
if(queuelen >= TCP_SND_QUEUELEN) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: too long queue %d (max %d)\n", queuelen, TCP_SND_QUEUELEN));
goto memerr;
}
#ifdef LWIP_DEBUG
if(pcb->snd_queuelen != 0) {
ASSERT("tcp_enqueue: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
#endif /* LWIP_DEBUG */
seg = NULL;
seglen = 0;
while(queue == NULL || left > 0) {
seglen = left > pcb->mss? pcb->mss: left;
/* allocate memory for tcp_seg, and fill in fields */
seg = memp_malloc(MEMP_TCP_SEG);
if(seg == NULL) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: could not allocate memory for tcp_seg\n"));
goto memerr;
}
seg->next = NULL;
seg->p = NULL;
if(queue == NULL) {
queue = seg;
} else {
for(useg = queue; useg->next != NULL; useg = useg->next);
useg->next = seg;
}
/* If copy is set, memory should be allocated
and data copied into pbuf, otherwise data comes from
ROM or other static memory, and need not be copied. If
optdata is != NULL, we have options instead of data. */
if(optdata != NULL) {
if((seg->p = pbuf_alloc(PBUF_TRANSPORT, optlen, PBUF_RAM)) == NULL) {
goto memerr;
}
++queuelen;
seg->dataptr = seg->p->payload;
} else if(copy) {
if((seg->p = pbuf_alloc(PBUF_TRANSPORT, seglen, PBUF_RAM)) == NULL) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: could not allocate memory for pbuf copy\n"));
goto memerr;
}
++queuelen;
if(arg != NULL) {
memcpy(seg->p->payload, ptr, seglen);
}
seg->dataptr = seg->p->payload;
} else {
/* Do not copy the data. */
if((p = pbuf_alloc(PBUF_TRANSPORT, seglen, PBUF_ROM)) == NULL) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: could not allocate memory for pbuf non-copy\n"));
goto memerr;
}
++queuelen;
p->payload = ptr;
seg->dataptr = ptr;
if((seg->p = pbuf_alloc(PBUF_TRANSPORT, 0, PBUF_RAM)) == NULL) {
pbuf_free(p);
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: could not allocate memory for header pbuf\n"));
goto memerr;
}
++queuelen;
pbuf_chain(seg->p, p);
}
if(queuelen > TCP_SND_QUEUELEN) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: queue too long %d (%d)\n", queuelen, TCP_SND_QUEUELEN));
goto memerr;
}
seg->len = seglen;
/* if((flags & TCP_SYN) || (flags & TCP_FIN)) {
++seg->len;
}*/
/* build TCP header */
if(pbuf_header(seg->p, TCP_HLEN)) {
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: no room for TCP header in pbuf.\n"));
#ifdef TCP_STATS
++stats.tcp.err;
#endif /* TCP_STATS */
goto memerr;
}
seg->tcphdr = seg->p->payload;
seg->tcphdr->src = htons(pcb->local_port);
seg->tcphdr->dest = htons(pcb->remote_port);
seg->tcphdr->seqno = htonl(seqno);
seg->tcphdr->urgp = 0;
TCPH_FLAGS_SET(seg->tcphdr, flags);
/* don't fill in tcphdr->ackno and tcphdr->wnd until later */
if(optdata == NULL) {
TCPH_OFFSET_SET(seg->tcphdr, 5 << 4);
} else {
TCPH_OFFSET_SET(seg->tcphdr, (5 + optlen / 4) << 4);
/* Copy options into data portion of segment.
Options can thus only be sent in non data carrying
segments such as SYN|ACK. */
memcpy(seg->dataptr, optdata, optlen);
}
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_enqueue: queueing %lu:%lu (0x%x)\n",
ntohl(seg->tcphdr->seqno),
ntohl(seg->tcphdr->seqno) + TCP_TCPLEN(seg),
flags));
left -= seglen;
seqno += seglen;
ptr = (void *)((char *)ptr + seglen);
}
/* Go to the last segment on the ->unsent queue. */
if(pcb->unsent == NULL) {
useg = NULL;
} else {
for(useg = pcb->unsent; useg->next != NULL; useg = useg->next);
}
/* If there is room in the last pbuf on the unsent queue,
chain the first pbuf on the queue together with that. */
if(useg != NULL &&
TCP_TCPLEN(useg) != 0 &&
!(TCPH_FLAGS(useg->tcphdr) & (TCP_SYN | TCP_FIN)) &&
!(flags & (TCP_SYN | TCP_FIN)) &&
useg->len + queue->len <= pcb->mss) {
/* Remove TCP header from first segment. */
pbuf_header(queue->p, -TCP_HLEN);
pbuf_chain(useg->p, queue->p);
useg->len += queue->len;
useg->next = queue->next;
DEBUGF(TCP_OUTPUT_DEBUG, ("tcp_output: chaining, new len %u\n", useg->len));
if(seg == queue) {
seg = NULL;
}
memp_free(MEMP_TCP_SEG, queue);
} else {
if(useg == NULL) {
pcb->unsent = queue;
} else {
useg->next = queue;
}
}
if((flags & TCP_SYN) || (flags & TCP_FIN)) {
++len;
}
pcb->snd_lbb += len;
pcb->snd_buf -= len;
pcb->snd_queuelen = queuelen;
DEBUGF(TCP_QLEN_DEBUG, ("tcp_enqueue: %d (after enqueued)\n", pcb->snd_queuelen));
#ifdef LWIP_DEBUG
if(pcb->snd_queuelen != 0) {
ASSERT("tcp_enqueue: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
#endif /* LWIP_DEBUG */
/* Set the PSH flag in the last segment that we enqueued, but only
if the segment has data (indicated by seglen > 0). */
if(seg != NULL && seglen > 0 && seg->tcphdr != NULL) {
TCPH_FLAGS_SET(seg->tcphdr, TCPH_FLAGS(seg->tcphdr) | TCP_PSH);
}
return ERR_OK;
memerr:
#ifdef TCP_STATS
++stats.tcp.memerr;
#endif /* TCP_STATS */
if(queue != NULL) {
tcp_segs_free(queue);
}
#ifdef LWIP_DEBUG
if(pcb->snd_queuelen != 0) {
ASSERT("tcp_enqueue: valid queue length", pcb->unacked != NULL ||
pcb->unsent != NULL);
}
#endif /* LWIP_DEBUG */
DEBUGF(TCP_QLEN_DEBUG, ("tcp_enqueue: %d (with mem err)\n", pcb->snd_queuelen));
return ERR_MEM;
}
/* 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 {
