1 This is a light-weight, user-space implementation of RFC 793 (TCP), without any
2 reliance on an IP layer. It can be used to provide multiple in-order, reliable
3 streams on top of any datagram layer.
5 UTCP does not rely on a specific event system. Instead, the application feeds
6 it with incoming packets using utcp_recv(), and outgoing data for the streams
7 using utcp_send(). Most of the rest is handled by callbacks. The application
8 must however call utcp_timeout() regularly to have UTCP handle packet loss.
10 The application should run utcp_init() for every peer it wants to communicate
13 DIFFERENCES FROM RFC 793:
15 * No checksum. UTCP requires the application to handle packet integrity.
16 * 32-bit window size. Big window sizes are the default.
21 * Implement send buffer
23 * Handle retransmission
24 - Proper timeout handling
28 * Nagle (add PSH back to signal receiver that now we want an immediate ACK?)
30 * Congestion window scaling
42 SYN|FIN + request data ->
43 <- SYN|ACK|FIN + response data
46 Does this need special care or can we rely on higher level MACs?
51 793 Transmission Control Protocol (Functional Specification)
52 2581 TCP Congestion Control
53 2988 Computing TCP's Retransmission Timer
60 - snd.una: the sequence number of the first byte we did not receive an ACK for
61 - snd.nxt: the sequence number of the first byte after the last packet we sent (due to retransmission, this may go backwards)
62 - snd.wnd: the number of bytes we have left in our (UTCP/application?) input buffer
63 - snd.last: the sequence number of the last byte that was enqueued in the TCP stream (increases only monotonically)
65 - rcv.nxt: the sequence number of the first byte after the last one we passed up to the application
66 - rcv.wnd: the number of bytes the receiver has left in its input buffer (may be more/less than our send buffer size)
68 - The only packets that do not have ACK set must either have SYN or RST set
69 - Only packets received with rcv.nxt <= hdr.seq <= rcv.nxt + rcv.wnd are valid, drop others.
70 - If it has ACK set, and it's higher than snd.una, update snd.una.
71 But don't update it past c->snd.next. (RST in that case?)
73 - SYN and FIN each count as one byte for the sequence numbering, but no actual byte is transferred in the payload.
78 This timer is intended to catch the case when we are waiting very long for a response but nothing happens.
79 The timeout is in the order of minutes.
81 - The conn timeout is set whenever there is unacknowledged data, or when we are in the TIME_WAIT status.
82 - If snd.una is advanced while the timeout is set, we re-set the timeout.
83 - If the conn timeout expires, close the connection immediately.
90 This timer is intended to catch the case where we didn't get an ACK from the peer.
91 In principle, the timeout should be slightly longer than the maximum latency along the path.
93 - The rtrx timer is set whenever we send a packet that must be ACKed by the peer:
94 - when it contains data
95 - when SYN or FIN is set
96 - The rtrx timer is reset when we receive a packet that advances snd.una.
97 - it is cleared when snd.una == snd.last
98 - otherwise the timeout is set to the value of utcp->rto
99 - If the rtrx timer expires, retransmit at least one packet, multiply the timeout by two, and rearm the timeout.
101 The value of RTO is calculated according to the RFC. At the moment, no
102 timestamps are added to packets. When the RTT timer is not set, start it when
103 sending a packet. When the ACK arrives, stop the timer and use the time
104 difference as a measured RTT value. Use the algorithm from RFC 6298 to update
110 CLOSED: this connection is closed, all packets received will result in RST.
116 LISTEN: (= no connection yet): only allow SYN packets, it application does not accept, return RST|ACK, else SYN|ACK.
117 RX: on accept, send SYNACK, go to SYN_RECEIVED
122 SYN_SENT: we sent a SYN, now expecting SYN|ACK
123 RX: must be valid SYNACK, send ACK, go to ESTABLISHED
124 TX: put in send buffer (TODO: send SYN again with data?)
127 SYN_RECEIVED: we received a SYN, sent back a SYN|ACK, now expecting an ACK
128 RX: must be valid ACK, go to ESTABLISHED
129 TX: put in send buffer (TODO: send SYNACK again with data?)
130 RT: send SYNACK again
132 ESTABLISHED: SYN is acked, we can now send/receive normal data.
133 RX: process data, return ACK. If FIN set, go to CLOSE_WAIT
134 TX: put in send buffer, segmentize and send
135 RT: send unACKed data again
137 FIN_WAIT_1: we want to close the connection, and just sent a FIN, waiting for it to be ACKed.
138 RX: process data, return ACK. If our FIN is acked, go to FIN_WAIT_2, if a FIN was also received, go to CLOSING
140 RT: send unACKed data or else FIN again
142 FIN_WAIT_2: our FIN is ACKed, just waiting for more data or FIN from the peer.
143 RX: process data, return ACK. If a FIN was also received, go to CLOSING
145 RT: should not happen, clear timeouts
147 CLOSE_WAIT: we received a FIN, we sent back an ACK
148 RX: only return an ACK.
149 TX: put in send buffer, segmentize and send
150 RT: send unACKed data again
152 CLOSING: we had already sent a FIN, and we received a FIN back, now waiting for it to be ACKed.
153 RX: if it's ACKed, set conn timeout, go to TIME_WAIT
155 RT: send unACKed data or else FIN again
157 LAST_ACK: we are waiting for the last ACK before we can CLOSE
158 RX: if it's ACKed, go to CLOSED
162 TIME_WAIT: connection is in princple closed, but our last ACK might not have been received, so just wait a while to see if a FIN gets retransmitted so we can resend the ACK.
163 RX: if we receive anything, reset conn timeout.
165 RT: should not happen, clear rtrx timeout
170 - Put the packet in the send buffer.
171 - Decide how much to send:
172 - Not more than receive window allows
173 - Not more that congestion window allows
174 - Segmentize and send the packets
175 - At the end, snd.nxt is advanced with the number of bytes sent
176 - Set the rtrx and conn timers if they have not been set
181 - Decide how much to send:
182 - Not more than we have in the send buffer
183 - Not more than receive window allows
184 - Not more that congestion window allows
185 - Segmentize and send packets
186 - No advancement of sequence numbers happen
187 - Reset the rtrx timers
192 1 Drop invalid packets:
193 a Invalid flags or state
195 c hdr.seq not within our receive window
196 d hdr.ack ahead of snd.nxt or behind snd.una
199 a reset conn timer if so
200 b check if our SYN or FIN has been acked
201 c check if any data been acked
202 - remove ACKed data from send buffer
204 d no advance? NewReno
205 4 If snd.una == snd.nxt, clear rtrx and conn timer
206 5 Process state changes due to SYN
207 6 Send new data to application
208 7 Process state changes due to FIN
213 We want to send as much packets as possible that won't cause any packets to be
214 dropped. So we should not send more than the available bandwidth, and not more
215 in one go than buffers along the path can handle.
217 To start, we use "self-clocking". We send one packet, and wait for an ACK
218 before sending another packet. On a network with a finite bandwidth but zero
219 delay (latency), this will send packets as efficiently as possible. We don't
220 need any timers to control the outgoing packet rate, that's why we call this
221 self-clocked. However, latency is non-zero, and this means a number of packets
222 is always on the way between the sender and receiver. The amount of packets
223 "inbetween" is in principle the bandwidth times the delay (bandwidth-delay
226 Delay is fairly easy to measure (equal to half the round-trip time of a packet,
227 which in TCP is easily obtained from the SYN and SYNACK pair, or the ACK in
228 response of a segment), however bandwidth is more difficult and might change
229 more rapidly than the latency.
231 Back to the "inbetween" packets: ideally we would like to fill the available
232 inbetween space completely. It should be easy to see that in that case,
233 self-clocking will still work as intended. Our estimate of the amount of
234 packets in the inbetween space is called the congestion window (CWND). If we
235 know the BDP, we can set the CWND to it, however if we don't know it, we can
236 start with a small CWND and gradually increase it (for example, every time we
237 receive an ACK, send the next 2 segments). At some point, we will start sending
238 at a higher rate than the available bandwidth, in which case packets will
239 inevitably be lost. We detect that because we do not receive an ACK for our
240 data, and then we have to reduce the CWND (for example, by half).
242 The trick is to choose an algorithm that best keeps the CWND to the effective
245 A nice introduction is RFC 2001.
247 snd.cwnd: size of the congestion window.
248 snd.nxt - snd.una: number of unacknowledged bytes, = number of bytes in flight.
249 snd.cwnd - (snd.nxt - snd.una): unused size of congestion window