1 | /*
|
---|
2 | * Copyright (c) 1982, 1986, 1993, 1994
|
---|
3 | * The Regents of the University of California. All rights reserved.
|
---|
4 | *
|
---|
5 | * Redistribution and use in source and binary forms, with or without
|
---|
6 | * modification, are permitted provided that the following conditions
|
---|
7 | * are met:
|
---|
8 | * 1. Redistributions of source code must retain the above copyright
|
---|
9 | * notice, this list of conditions and the following disclaimer.
|
---|
10 | * 2. Redistributions in binary form must reproduce the above copyright
|
---|
11 | * notice, this list of conditions and the following disclaimer in the
|
---|
12 | * documentation and/or other materials provided with the distribution.
|
---|
13 | * 3. All advertising materials mentioning features or use of this software
|
---|
14 | * must display the following acknowledgement:
|
---|
15 | * This product includes software developed by the University of
|
---|
16 | * California, Berkeley and its contributors.
|
---|
17 | * 4. Neither the name of the University nor the names of its contributors
|
---|
18 | * may be used to endorse or promote products derived from this software
|
---|
19 | * without specific prior written permission.
|
---|
20 | *
|
---|
21 | * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
---|
22 | * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
---|
23 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
---|
24 | * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
---|
25 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
---|
26 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
---|
27 | * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
---|
28 | * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
---|
29 | * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
---|
30 | * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
---|
31 | * SUCH DAMAGE.
|
---|
32 | *
|
---|
33 | * @(#)tcp_var.h 8.3 (Berkeley) 4/10/94
|
---|
34 | * tcp_var.h,v 1.3 1994/08/21 05:27:39 paul Exp
|
---|
35 | */
|
---|
36 |
|
---|
37 | #ifndef _TCP_VAR_H_
|
---|
38 | #define _TCP_VAR_H_
|
---|
39 |
|
---|
40 | #ifdef VBOX_WITH_BSD_REASS
|
---|
41 | #include "queue.h"
|
---|
42 | #endif /* VBOX_WITH_BSD_REASS */
|
---|
43 |
|
---|
44 | #include "tcpip.h"
|
---|
45 | #include "tcp_timer.h"
|
---|
46 |
|
---|
47 | #if SIZEOF_CHAR_P == 4
|
---|
48 | typedef struct tcpiphdr *tcpiphdrp_32;
|
---|
49 | # define u32ptr_done(pData, u32, ptr) do {} while (0)
|
---|
50 | # define ptr_to_u32(pData, ptr) (ptr)
|
---|
51 | # define u32_to_ptr(pData, u32, type) ((type)(u32))
|
---|
52 | #else
|
---|
53 | typedef u_int32_t tcpiphdrp_32;
|
---|
54 | # include <iprt/types.h>
|
---|
55 | # include <iprt/assert.h>
|
---|
56 |
|
---|
57 | #if defined(VBOX_WITH_BSD_REASS)
|
---|
58 | # define u32ptr_done(pData, u32, ptr) do {} while (0)
|
---|
59 | # define ptr_to_u32(pData, ptr) (ptr)
|
---|
60 | # define u32_to_ptr(pData, u32, type) ((type)(u32))
|
---|
61 | #else /* !VBOX_WITH_BSD_REASS */
|
---|
62 | # define u32ptr_done(pData, u32, ptr) VBoxU32PtrDone((pData), (ptr), (u32))
|
---|
63 | # define ptr_to_u32(pData, ptr) VBoxU32PtrHash((pData), (ptr))
|
---|
64 | # define u32_to_ptr(pData, u32, type) ((type)VBoxU32PtrLookup((pData), (u32)))
|
---|
65 | #endif /* !VBOX_WITH_BSD_REASS */
|
---|
66 |
|
---|
67 | #endif
|
---|
68 |
|
---|
69 | #ifdef VBOX_WITH_BSD_REASS
|
---|
70 | /* TCP segment queue entry */
|
---|
71 | struct tseg_qent {
|
---|
72 | LIST_ENTRY(tseg_qent) tqe_q;
|
---|
73 | int tqe_len; /* TCP segment data length */
|
---|
74 | struct tcphdr *tqe_th; /* a pointer to tcp header */
|
---|
75 | struct mbuf *tqe_m; /* mbuf contains packet */
|
---|
76 | };
|
---|
77 | LIST_HEAD(tsegqe_head, tseg_qent);
|
---|
78 | #endif
|
---|
79 |
|
---|
80 | /*
|
---|
81 | * Tcp control block, one per tcp; fields:
|
---|
82 | */
|
---|
83 | struct tcpcb {
|
---|
84 | #ifndef VBOX_WITH_BSD_REASS
|
---|
85 | tcpiphdrp_32 seg_next; /* sequencing queue */
|
---|
86 | tcpiphdrp_32 seg_prev;
|
---|
87 | #else /* VBOX_WITH_BSD_REASS */
|
---|
88 | LIST_ENTRY(tcpcb) t_list;
|
---|
89 | struct tsegqe_head t_segq; /* segment reassembly queue */
|
---|
90 | int t_segqlen; /* segment reassembly queue length */
|
---|
91 | #endif /* VBOX_WITH_BSD_REASS */
|
---|
92 | short t_state; /* state of this connection */
|
---|
93 | short t_timer[TCPT_NTIMERS]; /* tcp timers */
|
---|
94 | short t_rxtshift; /* log(2) of rexmt exp. backoff */
|
---|
95 | short t_rxtcur; /* current retransmit value */
|
---|
96 | short t_dupacks; /* consecutive dup acks recd */
|
---|
97 | u_short t_maxseg; /* maximum segment size */
|
---|
98 | char t_force; /* 1 if forcing out a byte */
|
---|
99 | u_short t_flags;
|
---|
100 | #define TF_ACKNOW 0x0001 /* ack peer immediately */
|
---|
101 | #define TF_DELACK 0x0002 /* ack, but try to delay it */
|
---|
102 | #define TF_NODELAY 0x0004 /* don't delay packets to coalesce */
|
---|
103 | #define TF_NOOPT 0x0008 /* don't use tcp options */
|
---|
104 | #define TF_SENTFIN 0x0010 /* have sent FIN */
|
---|
105 | #define TF_REQ_SCALE 0x0020 /* have/will request window scaling */
|
---|
106 | #define TF_RCVD_SCALE 0x0040 /* other side has requested scaling */
|
---|
107 | #define TF_REQ_TSTMP 0x0080 /* have/will request timestamps */
|
---|
108 | #define TF_RCVD_TSTMP 0x0100 /* a timestamp was received in SYN */
|
---|
109 | #define TF_SACK_PERMIT 0x0200 /* other side said I could SACK */
|
---|
110 |
|
---|
111 | /* Make it static for now */
|
---|
112 | /* struct tcpiphdr *t_template; / * skeletal packet for transmit */
|
---|
113 | struct tcpiphdr t_template;
|
---|
114 |
|
---|
115 | struct socket *t_socket; /* back pointer to socket */
|
---|
116 | /*
|
---|
117 | * The following fields are used as in the protocol specification.
|
---|
118 | * See RFC783, Dec. 1981, page 21.
|
---|
119 | */
|
---|
120 | /* send sequence variables */
|
---|
121 | tcp_seq snd_una; /* send unacknowledged */
|
---|
122 | tcp_seq snd_nxt; /* send next */
|
---|
123 | tcp_seq snd_up; /* send urgent pointer */
|
---|
124 | tcp_seq snd_wl1; /* window update seg seq number */
|
---|
125 | tcp_seq snd_wl2; /* window update seg ack number */
|
---|
126 | tcp_seq iss; /* initial send sequence number */
|
---|
127 | u_int32_t snd_wnd; /* send window */
|
---|
128 | /* receive sequence variables */
|
---|
129 | u_int32_t rcv_wnd; /* receive window */
|
---|
130 | tcp_seq rcv_nxt; /* receive next */
|
---|
131 | tcp_seq rcv_up; /* receive urgent pointer */
|
---|
132 | tcp_seq irs; /* initial receive sequence number */
|
---|
133 | /*
|
---|
134 | * Additional variables for this implementation.
|
---|
135 | */
|
---|
136 | /* receive variables */
|
---|
137 | tcp_seq rcv_adv; /* advertised window */
|
---|
138 | /* retransmit variables */
|
---|
139 | tcp_seq snd_max; /* highest sequence number sent;
|
---|
140 | * used to recognize retransmits
|
---|
141 | */
|
---|
142 | /* congestion control (for slow start, source quench, retransmit after loss) */
|
---|
143 | u_int32_t snd_cwnd; /* congestion-controlled window */
|
---|
144 | u_int32_t snd_ssthresh; /* snd_cwnd size threshold for
|
---|
145 | * for slow start exponential to
|
---|
146 | * linear switch
|
---|
147 | */
|
---|
148 | /*
|
---|
149 | * transmit timing stuff. See below for scale of srtt and rttvar.
|
---|
150 | * "Variance" is actually smoothed difference.
|
---|
151 | */
|
---|
152 | short t_idle; /* inactivity time */
|
---|
153 | short t_rtt; /* round trip time */
|
---|
154 | tcp_seq t_rtseq; /* sequence number being timed */
|
---|
155 | short t_srtt; /* smoothed round-trip time */
|
---|
156 | short t_rttvar; /* variance in round-trip time */
|
---|
157 | u_short t_rttmin; /* minimum rtt allowed */
|
---|
158 | u_int32_t max_sndwnd; /* largest window peer has offered */
|
---|
159 |
|
---|
160 | /* out-of-band data */
|
---|
161 | char t_oobflags; /* have some */
|
---|
162 | char t_iobc; /* input character */
|
---|
163 | #define TCPOOB_HAVEDATA 0x01
|
---|
164 | #define TCPOOB_HADDATA 0x02
|
---|
165 | short t_softerror; /* possible error not yet reported */
|
---|
166 |
|
---|
167 | /* RFC 1323 variables */
|
---|
168 | u_char snd_scale; /* window scaling for send window */
|
---|
169 | u_char rcv_scale; /* window scaling for recv window */
|
---|
170 | u_char request_r_scale; /* pending window scaling */
|
---|
171 | u_char requested_s_scale;
|
---|
172 | u_int32_t ts_recent; /* timestamp echo data */
|
---|
173 | u_int32_t ts_recent_age; /* when last updated */
|
---|
174 | tcp_seq last_ack_sent;
|
---|
175 |
|
---|
176 | };
|
---|
177 |
|
---|
178 | #ifdef VBOX_WITH_BSD_REASS
|
---|
179 | LIST_HEAD(tcpcbhead, tcpcb);
|
---|
180 | #endif /* VBOX_WITH_BSD_REASS */
|
---|
181 |
|
---|
182 | #define sototcpcb(so) ((so)->so_tcpcb)
|
---|
183 |
|
---|
184 | /*
|
---|
185 | * The smoothed round-trip time and estimated variance
|
---|
186 | * are stored as fixed point numbers scaled by the values below.
|
---|
187 | * For convenience, these scales are also used in smoothing the average
|
---|
188 | * (smoothed = (1/scale)sample + ((scale-1)/scale)smoothed).
|
---|
189 | * With these scales, srtt has 3 bits to the right of the binary point,
|
---|
190 | * and thus an "ALPHA" of 0.875. rttvar has 2 bits to the right of the
|
---|
191 | * binary point, and is smoothed with an ALPHA of 0.75.
|
---|
192 | */
|
---|
193 | #define TCP_RTT_SCALE 8 /* multiplier for srtt; 3 bits frac. */
|
---|
194 | #define TCP_RTT_SHIFT 3 /* shift for srtt; 3 bits frac. */
|
---|
195 | #define TCP_RTTVAR_SCALE 4 /* multiplier for rttvar; 2 bits */
|
---|
196 | #define TCP_RTTVAR_SHIFT 2 /* multiplier for rttvar; 2 bits */
|
---|
197 |
|
---|
198 | /*
|
---|
199 | * The initial retransmission should happen at rtt + 4 * rttvar.
|
---|
200 | * Because of the way we do the smoothing, srtt and rttvar
|
---|
201 | * will each average +1/2 tick of bias. When we compute
|
---|
202 | * the retransmit timer, we want 1/2 tick of rounding and
|
---|
203 | * 1 extra tick because of +-1/2 tick uncertainty in the
|
---|
204 | * firing of the timer. The bias will give us exactly the
|
---|
205 | * 1.5 tick we need. But, because the bias is
|
---|
206 | * statistical, we have to test that we don't drop below
|
---|
207 | * the minimum feasible timer (which is 2 ticks).
|
---|
208 | * This macro assumes that the value of TCP_RTTVAR_SCALE
|
---|
209 | * is the same as the multiplier for rttvar.
|
---|
210 | */
|
---|
211 | #define TCP_REXMTVAL(tp) \
|
---|
212 | (((tp)->t_srtt >> TCP_RTT_SHIFT) + (tp)->t_rttvar)
|
---|
213 |
|
---|
214 | /* XXX
|
---|
215 | * We want to avoid doing m_pullup on incoming packets but that
|
---|
216 | * means avoiding dtom on the tcp reassembly code. That in turn means
|
---|
217 | * keeping an mbuf pointer in the reassembly queue (since we might
|
---|
218 | * have a cluster). As a quick hack, the source & destination
|
---|
219 | * port numbers (which are no longer needed once we've located the
|
---|
220 | * tcpcb) are overlayed with an mbuf pointer.
|
---|
221 | */
|
---|
222 | #if SIZEOF_CHAR_P == 4
|
---|
223 | typedef struct mbuf *mbufp_32;
|
---|
224 | /* VBox change which is too much bother to #ifdef */
|
---|
225 | # define REASS_MBUF_SET(ti, p) (*(mbufp_32 *)&((ti)->ti_t)) = (p)
|
---|
226 | # define REASS_MBUF_GET(ti) ((struct mbuf *)(*(mbufp_32 *)&((ti)->ti_t)))
|
---|
227 | #else
|
---|
228 | typedef u_int32_t mbufp_32;
|
---|
229 | /* VBox change which is too much bother to #ifdef */
|
---|
230 | # define REASS_MBUF_SET(ti, p) (*(mbufp_32 *)&((ti)->ti_t)) = ptr_to_u32(pData, p)
|
---|
231 | # define REASS_MBUF_GET(ti) u32_to_ptr(pData, (*(mbufp_32 *)&((ti)->ti_t)), struct mbuf *)
|
---|
232 | #endif
|
---|
233 | /*#define REASS_MBUF(ti) (*(mbufp_32 *)&((ti)->ti_t)) - replaced by REASS_MBUF_GET/SET */
|
---|
234 |
|
---|
235 | /*
|
---|
236 | * TCP statistics.
|
---|
237 | * Many of these should be kept per connection,
|
---|
238 | * but that's inconvenient at the moment.
|
---|
239 | */
|
---|
240 | struct tcpstat_t {
|
---|
241 | u_long tcps_connattempt; /* connections initiated */
|
---|
242 | u_long tcps_accepts; /* connections accepted */
|
---|
243 | u_long tcps_connects; /* connections established */
|
---|
244 | u_long tcps_drops; /* connections dropped */
|
---|
245 | u_long tcps_conndrops; /* embryonic connections dropped */
|
---|
246 | u_long tcps_closed; /* conn. closed (includes drops) */
|
---|
247 | u_long tcps_segstimed; /* segs where we tried to get rtt */
|
---|
248 | u_long tcps_rttupdated; /* times we succeeded */
|
---|
249 | u_long tcps_delack; /* delayed acks sent */
|
---|
250 | u_long tcps_timeoutdrop; /* conn. dropped in rxmt timeout */
|
---|
251 | u_long tcps_rexmttimeo; /* retransmit timeouts */
|
---|
252 | u_long tcps_persisttimeo; /* persist timeouts */
|
---|
253 | u_long tcps_keeptimeo; /* keepalive timeouts */
|
---|
254 | u_long tcps_keepprobe; /* keepalive probes sent */
|
---|
255 | u_long tcps_keepdrops; /* connections dropped in keepalive */
|
---|
256 |
|
---|
257 | u_long tcps_sndtotal; /* total packets sent */
|
---|
258 | u_long tcps_sndpack; /* data packets sent */
|
---|
259 | u_long tcps_sndbyte; /* data bytes sent */
|
---|
260 | u_long tcps_sndrexmitpack; /* data packets retransmitted */
|
---|
261 | u_long tcps_sndrexmitbyte; /* data bytes retransmitted */
|
---|
262 | u_long tcps_sndacks; /* ack-only packets sent */
|
---|
263 | u_long tcps_sndprobe; /* window probes sent */
|
---|
264 | u_long tcps_sndurg; /* packets sent with URG only */
|
---|
265 | u_long tcps_sndwinup; /* window update-only packets sent */
|
---|
266 | u_long tcps_sndctrl; /* control (SYN|FIN|RST) packets sent */
|
---|
267 |
|
---|
268 | u_long tcps_rcvtotal; /* total packets received */
|
---|
269 | u_long tcps_rcvpack; /* packets received in sequence */
|
---|
270 | u_long tcps_rcvbyte; /* bytes received in sequence */
|
---|
271 | u_long tcps_rcvbadsum; /* packets received with ccksum errs */
|
---|
272 | u_long tcps_rcvbadoff; /* packets received with bad offset */
|
---|
273 | /* u_long tcps_rcvshort; */ /* packets received too short */
|
---|
274 | u_long tcps_rcvduppack; /* duplicate-only packets received */
|
---|
275 | u_long tcps_rcvdupbyte; /* duplicate-only bytes received */
|
---|
276 | u_long tcps_rcvpartduppack; /* packets with some duplicate data */
|
---|
277 | u_long tcps_rcvpartdupbyte; /* dup. bytes in part-dup. packets */
|
---|
278 | u_long tcps_rcvoopack; /* out-of-order packets received */
|
---|
279 | u_long tcps_rcvoobyte; /* out-of-order bytes received */
|
---|
280 | u_long tcps_rcvpackafterwin; /* packets with data after window */
|
---|
281 | u_long tcps_rcvbyteafterwin; /* bytes rcvd after window */
|
---|
282 | u_long tcps_rcvafterclose; /* packets rcvd after "close" */
|
---|
283 | u_long tcps_rcvwinprobe; /* rcvd window probe packets */
|
---|
284 | u_long tcps_rcvdupack; /* rcvd duplicate acks */
|
---|
285 | u_long tcps_rcvacktoomuch; /* rcvd acks for unsent data */
|
---|
286 | u_long tcps_rcvackpack; /* rcvd ack packets */
|
---|
287 | u_long tcps_rcvackbyte; /* bytes acked by rcvd acks */
|
---|
288 | u_long tcps_rcvwinupd; /* rcvd window update packets */
|
---|
289 | /* u_long tcps_pawsdrop; */ /* segments dropped due to PAWS */
|
---|
290 | u_long tcps_predack; /* times hdr predict ok for acks */
|
---|
291 | u_long tcps_preddat; /* times hdr predict ok for data pkts */
|
---|
292 | u_long tcps_socachemiss; /* tcp_last_so misses */
|
---|
293 | u_long tcps_didnuttin; /* Times tcp_output didn't do anything XXX */
|
---|
294 | #ifdef VBOX_WITH_BSD_REASS
|
---|
295 | u_long tcps_rcvmemdrop;
|
---|
296 | #endif /* VBOX_WITH_BSD_REASS */
|
---|
297 | };
|
---|
298 |
|
---|
299 |
|
---|
300 | #endif
|
---|