aboutsummaryrefslogtreecommitdiff
path: root/slirp/socket.c
blob: 8f73e906e28f0750f19faceb6d715a20b2c92667 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
/*
 * Copyright (c) 1995 Danny Gasparovski.
 *
 * Please read the file COPYRIGHT for the
 * terms and conditions of the copyright.
 */

#include "qemu-common.h"
#include <slirp.h>
#include "ip_icmp.h"
#ifdef __sun__
#include <sys/filio.h>
#endif

static void sofcantrcvmore(struct socket *so);
static void sofcantsendmore(struct socket *so);

struct socket *
solookup(struct socket **last, struct socket *head,
         struct in_addr laddr, u_int lport,
         struct in_addr faddr, u_int fport)
{
    struct socket *so = *last;

    /* Optimisation */
    if (so != head &&
            so->so_lport == lport &&
            so->so_laddr.s_addr == laddr.s_addr &&
            (!faddr.s_addr ||
                (so->so_faddr.s_addr == faddr.s_addr &&
                 so->so_fport == fport))) {
        return so;
    }

    for (so = head->so_next; so != head; so = so->so_next) {
        if (so->so_lport == lport &&
            so->so_laddr.s_addr == laddr.s_addr &&
            (!faddr.s_addr ||
                (so->so_faddr.s_addr == faddr.s_addr &&
                 so->so_fport == fport))) {
            *last = so;
            return so;
        }
    }

    return (struct socket *)NULL;
}

/*
 * Create a new socket, initialise the fields
 * It is the responsibility of the caller to
 * insque() it into the correct linked-list
 */
struct socket *
socreate(Slirp *slirp)
{
  struct socket *so;

  so = (struct socket *)malloc(sizeof(struct socket));
  if(so) {
    memset(so, 0, sizeof(struct socket));
    so->so_state = SS_NOFDREF;
    so->s = -1;
    so->slirp = slirp;
    so->pollfds_idx = -1;
  }
  return(so);
}

/*
 * remque and free a socket, clobber cache
 */
void
sofree(struct socket *so)
{
  Slirp *slirp = so->slirp;

  if (so->so_emu==EMU_RSH && so->extra) {
	sofree(so->extra);
	so->extra=NULL;
  }
  if (so == slirp->tcp_last_so) {
      slirp->tcp_last_so = &slirp->tcb;
  } else if (so == slirp->udp_last_so) {
      slirp->udp_last_so = &slirp->udb;
  } else if (so == slirp->icmp_last_so) {
      slirp->icmp_last_so = &slirp->icmp;
  }
  m_free(so->so_m);

  if(so->so_next && so->so_prev)
    remque(so);  /* crashes if so is not in a queue */

  free(so);
}

size_t sopreprbuf(struct socket *so, struct iovec *iov, int *np)
{
	int n, lss, total;
	struct sbuf *sb = &so->so_snd;
	int len = sb->sb_datalen - sb->sb_cc;
	int mss = so->so_tcpcb->t_maxseg;

	DEBUG_CALL("sopreprbuf");
	DEBUG_ARG("so = %p", so);

	if (len <= 0)
		return 0;

	iov[0].iov_base = sb->sb_wptr;
        iov[1].iov_base = NULL;
        iov[1].iov_len = 0;
	if (sb->sb_wptr < sb->sb_rptr) {
		iov[0].iov_len = sb->sb_rptr - sb->sb_wptr;
		/* Should never succeed, but... */
		if (iov[0].iov_len > len)
		   iov[0].iov_len = len;
		if (iov[0].iov_len > mss)
		   iov[0].iov_len -= iov[0].iov_len%mss;
		n = 1;
	} else {
		iov[0].iov_len = (sb->sb_data + sb->sb_datalen) - sb->sb_wptr;
		/* Should never succeed, but... */
		if (iov[0].iov_len > len) iov[0].iov_len = len;
		len -= iov[0].iov_len;
		if (len) {
			iov[1].iov_base = sb->sb_data;
			iov[1].iov_len = sb->sb_rptr - sb->sb_data;
			if(iov[1].iov_len > len)
			   iov[1].iov_len = len;
			total = iov[0].iov_len + iov[1].iov_len;
			if (total > mss) {
				lss = total%mss;
				if (iov[1].iov_len > lss) {
					iov[1].iov_len -= lss;
					n = 2;
				} else {
					lss -= iov[1].iov_len;
					iov[0].iov_len -= lss;
					n = 1;
				}
			} else
				n = 2;
		} else {
			if (iov[0].iov_len > mss)
			   iov[0].iov_len -= iov[0].iov_len%mss;
			n = 1;
		}
	}
	if (np)
		*np = n;

	return iov[0].iov_len + (n - 1) * iov[1].iov_len;
}

/*
 * Read from so's socket into sb_snd, updating all relevant sbuf fields
 * NOTE: This will only be called if it is select()ed for reading, so
 * a read() of 0 (or less) means it's disconnected
 */
int
soread(struct socket *so)
{
	int n, nn;
	struct sbuf *sb = &so->so_snd;
	struct iovec iov[2];

	DEBUG_CALL("soread");
	DEBUG_ARG("so = %p", so);

	/*
	 * No need to check if there's enough room to read.
	 * soread wouldn't have been called if there weren't
	 */
	sopreprbuf(so, iov, &n);

#ifdef HAVE_READV
	nn = readv(so->s, (struct iovec *)iov, n);
	DEBUG_MISC((dfd, " ... read nn = %d bytes\n", nn));
#else
	nn = qemu_recv(so->s, iov[0].iov_base, iov[0].iov_len,0);
#endif
	if (nn <= 0) {
		if (nn < 0 && (errno == EINTR || errno == EAGAIN))
			return 0;
		else {
			DEBUG_MISC((dfd, " --- soread() disconnected, nn = %d, errno = %d-%s\n", nn, errno,strerror(errno)));
			sofcantrcvmore(so);
			tcp_sockclosed(sototcpcb(so));
			return -1;
		}
	}

#ifndef HAVE_READV
	/*
	 * If there was no error, try and read the second time round
	 * We read again if n = 2 (ie, there's another part of the buffer)
	 * and we read as much as we could in the first read
	 * We don't test for <= 0 this time, because there legitimately
	 * might not be any more data (since the socket is non-blocking),
	 * a close will be detected on next iteration.
	 * A return of -1 wont (shouldn't) happen, since it didn't happen above
	 */
	if (n == 2 && nn == iov[0].iov_len) {
            int ret;
            ret = qemu_recv(so->s, iov[1].iov_base, iov[1].iov_len,0);
            if (ret > 0)
                nn += ret;
        }

	DEBUG_MISC((dfd, " ... read nn = %d bytes\n", nn));
#endif

	/* Update fields */
	sb->sb_cc += nn;
	sb->sb_wptr += nn;
	if (sb->sb_wptr >= (sb->sb_data + sb->sb_datalen))
		sb->sb_wptr -= sb->sb_datalen;
	return nn;
}

int soreadbuf(struct socket *so, const char *buf, int size)
{
    int n, nn, copy = size;
	struct sbuf *sb = &so->so_snd;
	struct iovec iov[2];

	DEBUG_CALL("soreadbuf");
	DEBUG_ARG("so = %p", so);

	/*
	 * No need to check if there's enough room to read.
	 * soread wouldn't have been called if there weren't
	 */
	if (sopreprbuf(so, iov, &n) < size)
        goto err;

    nn = MIN(iov[0].iov_len, copy);
    memcpy(iov[0].iov_base, buf, nn);

    copy -= nn;
    buf += nn;

    if (copy == 0)
        goto done;

    memcpy(iov[1].iov_base, buf, copy);

done:
    /* Update fields */
	sb->sb_cc += size;
	sb->sb_wptr += size;
	if (sb->sb_wptr >= (sb->sb_data + sb->sb_datalen))
		sb->sb_wptr -= sb->sb_datalen;
    return size;
err:

    sofcantrcvmore(so);
    tcp_sockclosed(sototcpcb(so));
    fprintf(stderr, "soreadbuf buffer to small");
    return -1;
}

/*
 * Get urgent data
 *
 * When the socket is created, we set it SO_OOBINLINE,
 * so when OOB data arrives, we soread() it and everything
 * in the send buffer is sent as urgent data
 */
void
sorecvoob(struct socket *so)
{
	struct tcpcb *tp = sototcpcb(so);

	DEBUG_CALL("sorecvoob");
	DEBUG_ARG("so = %p", so);

	/*
	 * We take a guess at how much urgent data has arrived.
	 * In most situations, when urgent data arrives, the next
	 * read() should get all the urgent data.  This guess will
	 * be wrong however if more data arrives just after the
	 * urgent data, or the read() doesn't return all the
	 * urgent data.
	 */
	soread(so);
	tp->snd_up = tp->snd_una + so->so_snd.sb_cc;
	tp->t_force = 1;
	tcp_output(tp);
	tp->t_force = 0;
}

/*
 * Send urgent data
 * There's a lot duplicated code here, but...
 */
int
sosendoob(struct socket *so)
{
	struct sbuf *sb = &so->so_rcv;
	char buff[2048]; /* XXX Shouldn't be sending more oob data than this */

	int n, len;

	DEBUG_CALL("sosendoob");
	DEBUG_ARG("so = %p", so);
	DEBUG_ARG("sb->sb_cc = %d", sb->sb_cc);

	if (so->so_urgc > 2048)
	   so->so_urgc = 2048; /* XXXX */

	if (sb->sb_rptr < sb->sb_wptr) {
		/* We can send it directly */
		n = slirp_send(so, sb->sb_rptr, so->so_urgc, (MSG_OOB)); /* |MSG_DONTWAIT)); */
		so->so_urgc -= n;

		DEBUG_MISC((dfd, " --- sent %d bytes urgent data, %d urgent bytes left\n", n, so->so_urgc));
	} else {
		/*
		 * Since there's no sendv or sendtov like writev,
		 * we must copy all data to a linear buffer then
		 * send it all
		 */
		len = (sb->sb_data + sb->sb_datalen) - sb->sb_rptr;
		if (len > so->so_urgc) len = so->so_urgc;
		memcpy(buff, sb->sb_rptr, len);
		so->so_urgc -= len;
		if (so->so_urgc) {
			n = sb->sb_wptr - sb->sb_data;
			if (n > so->so_urgc) n = so->so_urgc;
			memcpy((buff + len), sb->sb_data, n);
			so->so_urgc -= n;
			len += n;
		}
		n = slirp_send(so, buff, len, (MSG_OOB)); /* |MSG_DONTWAIT)); */
#ifdef DEBUG
		if (n != len)
		   DEBUG_ERROR((dfd, "Didn't send all data urgently XXXXX\n"));
#endif
		DEBUG_MISC((dfd, " ---2 sent %d bytes urgent data, %d urgent bytes left\n", n, so->so_urgc));
	}

	sb->sb_cc -= n;
	sb->sb_rptr += n;
	if (sb->sb_rptr >= (sb->sb_data + sb->sb_datalen))
		sb->sb_rptr -= sb->sb_datalen;

	return n;
}

/*
 * Write data from so_rcv to so's socket,
 * updating all sbuf field as necessary
 */
int
sowrite(struct socket *so)
{
	int  n,nn;
	struct sbuf *sb = &so->so_rcv;
	int len = sb->sb_cc;
	struct iovec iov[2];

	DEBUG_CALL("sowrite");
	DEBUG_ARG("so = %p", so);

	if (so->so_urgc) {
		sosendoob(so);
		if (sb->sb_cc == 0)
			return 0;
	}

	/*
	 * No need to check if there's something to write,
	 * sowrite wouldn't have been called otherwise
	 */

	iov[0].iov_base = sb->sb_rptr;
        iov[1].iov_base = NULL;
        iov[1].iov_len = 0;
	if (sb->sb_rptr < sb->sb_wptr) {
		iov[0].iov_len = sb->sb_wptr - sb->sb_rptr;
		/* Should never succeed, but... */
		if (iov[0].iov_len > len) iov[0].iov_len = len;
		n = 1;
	} else {
		iov[0].iov_len = (sb->sb_data + sb->sb_datalen) - sb->sb_rptr;
		if (iov[0].iov_len > len) iov[0].iov_len = len;
		len -= iov[0].iov_len;
		if (len) {
			iov[1].iov_base = sb->sb_data;
			iov[1].iov_len = sb->sb_wptr - sb->sb_data;
			if (iov[1].iov_len > len) iov[1].iov_len = len;
			n = 2;
		} else
			n = 1;
	}
	/* Check if there's urgent data to send, and if so, send it */

#ifdef HAVE_READV
	nn = writev(so->s, (const struct iovec *)iov, n);

	DEBUG_MISC((dfd, "  ... wrote nn = %d bytes\n", nn));
#else
	nn = slirp_send(so, iov[0].iov_base, iov[0].iov_len,0);
#endif
	/* This should never happen, but people tell me it does *shrug* */
	if (nn < 0 && (errno == EAGAIN || errno == EINTR))
		return 0;

	if (nn <= 0) {
		DEBUG_MISC((dfd, " --- sowrite disconnected, so->so_state = %x, errno = %d\n",
			so->so_state, errno));
		sofcantsendmore(so);
		tcp_sockclosed(sototcpcb(so));
		return -1;
	}

#ifndef HAVE_READV
	if (n == 2 && nn == iov[0].iov_len) {
            int ret;
            ret = slirp_send(so, iov[1].iov_base, iov[1].iov_len,0);
            if (ret > 0)
                nn += ret;
        }
        DEBUG_MISC((dfd, "  ... wrote nn = %d bytes\n", nn));
#endif

	/* Update sbuf */
	sb->sb_cc -= nn;
	sb->sb_rptr += nn;
	if (sb->sb_rptr >= (sb->sb_data + sb->sb_datalen))
		sb->sb_rptr -= sb->sb_datalen;

	/*
	 * If in DRAIN mode, and there's no more data, set
	 * it CANTSENDMORE
	 */
	if ((so->so_state & SS_FWDRAIN) && sb->sb_cc == 0)
		sofcantsendmore(so);

	return nn;
}

/*
 * recvfrom() a UDP socket
 */
void
sorecvfrom(struct socket *so)
{
	struct sockaddr_storage addr;
	struct sockaddr_storage saddr, daddr;
	socklen_t addrlen = sizeof(struct sockaddr_storage);

	DEBUG_CALL("sorecvfrom");
	DEBUG_ARG("so = %p", so);

	if (so->so_type == IPPROTO_ICMP) {   /* This is a "ping" reply */
	  char buff[256];
	  int len;

	  len = recvfrom(so->s, buff, 256, 0,
			 (struct sockaddr *)&addr, &addrlen);
	  /* XXX Check if reply is "correct"? */

	  if(len == -1 || len == 0) {
	    u_char code=ICMP_UNREACH_PORT;

	    if(errno == EHOSTUNREACH) code=ICMP_UNREACH_HOST;
	    else if(errno == ENETUNREACH) code=ICMP_UNREACH_NET;

	    DEBUG_MISC((dfd," udp icmp rx errno = %d-%s\n",
			errno,strerror(errno)));
	    icmp_error(so->so_m, ICMP_UNREACH,code, 0,strerror(errno));
	  } else {
	    icmp_reflect(so->so_m);
            so->so_m = NULL; /* Don't m_free() it again! */
	  }
	  /* No need for this socket anymore, udp_detach it */
	  udp_detach(so);
	} else {                            	/* A "normal" UDP packet */
	  struct mbuf *m;
          int len;
#ifdef _WIN32
          unsigned long n;
#else
          int n;
#endif

	  m = m_get(so->slirp);
	  if (!m) {
	      return;
	  }
	  m->m_data += IF_MAXLINKHDR;

	  /*
	   * XXX Shouldn't FIONREAD packets destined for port 53,
	   * but I don't know the max packet size for DNS lookups
	   */
	  len = M_FREEROOM(m);
	  /* if (so->so_fport != htons(53)) { */
	  ioctlsocket(so->s, FIONREAD, &n);

	  if (n > len) {
	    n = (m->m_data - m->m_dat) + m->m_len + n + 1;
	    m_inc(m, n);
	    len = M_FREEROOM(m);
	  }
	  /* } */

	  m->m_len = recvfrom(so->s, m->m_data, len, 0,
			      (struct sockaddr *)&addr, &addrlen);
	  DEBUG_MISC((dfd, " did recvfrom %d, errno = %d-%s\n",
		      m->m_len, errno,strerror(errno)));
	  if(m->m_len<0) {
	    u_char code=ICMP_UNREACH_PORT;

	    if(errno == EHOSTUNREACH) code=ICMP_UNREACH_HOST;
	    else if(errno == ENETUNREACH) code=ICMP_UNREACH_NET;

	    DEBUG_MISC((dfd," rx error, tx icmp ICMP_UNREACH:%i\n", code));
	    icmp_error(so->so_m, ICMP_UNREACH,code, 0,strerror(errno));
	    m_free(m);
	  } else {
	  /*
	   * Hack: domain name lookup will be used the most for UDP,
	   * and since they'll only be used once there's no need
	   * for the 4 minute (or whatever) timeout... So we time them
	   * out much quicker (10 seconds  for now...)
	   */
	    if (so->so_expire) {
	      if (so->so_fport == htons(53))
		so->so_expire = curtime + SO_EXPIREFAST;
	      else
		so->so_expire = curtime + SO_EXPIRE;
	    }

	    /*
	     * If this packet was destined for CTL_ADDR,
	     * make it look like that's where it came from
	     */
	    saddr = addr;
	    sotranslate_in(so, &saddr);
	    daddr = so->lhost.ss;

	    switch (so->so_ffamily) {
	    case AF_INET:
	        udp_output(so, m, (struct sockaddr_in *) &saddr,
	                   (struct sockaddr_in *) &daddr,
	                   so->so_iptos);
	        break;
	    default:
	        break;
	    }
	  } /* rx error */
	} /* if ping packet */
}

/*
 * sendto() a socket
 */
int
sosendto(struct socket *so, struct mbuf *m)
{
	int ret;
	struct sockaddr_storage addr;

	DEBUG_CALL("sosendto");
	DEBUG_ARG("so = %p", so);
	DEBUG_ARG("m = %p", m);

	addr = so->fhost.ss;
	DEBUG_CALL(" sendto()ing)");
	sotranslate_out(so, &addr);

	/* Don't care what port we get */
	ret = sendto(so->s, m->m_data, m->m_len, 0,
		     (struct sockaddr *)&addr, sizeof(addr));
	if (ret < 0)
		return -1;

	/*
	 * Kill the socket if there's no reply in 4 minutes,
	 * but only if it's an expirable socket
	 */
	if (so->so_expire)
		so->so_expire = curtime + SO_EXPIRE;
	so->so_state &= SS_PERSISTENT_MASK;
	so->so_state |= SS_ISFCONNECTED; /* So that it gets select()ed */
	return 0;
}

/*
 * Listen for incoming TCP connections
 */
struct socket *
tcp_listen(Slirp *slirp, uint32_t haddr, u_int hport, uint32_t laddr,
           u_int lport, int flags)
{
	struct sockaddr_in addr;
	struct socket *so;
	int s, opt = 1;
	socklen_t addrlen = sizeof(addr);
	memset(&addr, 0, addrlen);

	DEBUG_CALL("tcp_listen");
	DEBUG_ARG("haddr = %x", haddr);
	DEBUG_ARG("hport = %d", hport);
	DEBUG_ARG("laddr = %x", laddr);
	DEBUG_ARG("lport = %d", lport);
	DEBUG_ARG("flags = %x", flags);

	so = socreate(slirp);
	if (!so) {
	  return NULL;
	}

	/* Don't tcp_attach... we don't need so_snd nor so_rcv */
	if ((so->so_tcpcb = tcp_newtcpcb(so)) == NULL) {
		free(so);
		return NULL;
	}
	insque(so, &slirp->tcb);

	/*
	 * SS_FACCEPTONCE sockets must time out.
	 */
	if (flags & SS_FACCEPTONCE)
	   so->so_tcpcb->t_timer[TCPT_KEEP] = TCPTV_KEEP_INIT*2;

	so->so_state &= SS_PERSISTENT_MASK;
	so->so_state |= (SS_FACCEPTCONN | flags);
	so->so_lfamily = AF_INET;
	so->so_lport = lport; /* Kept in network format */
	so->so_laddr.s_addr = laddr; /* Ditto */

	addr.sin_family = AF_INET;
	addr.sin_addr.s_addr = haddr;
	addr.sin_port = hport;

	if (((s = qemu_socket(AF_INET,SOCK_STREAM,0)) < 0) ||
	    (socket_set_fast_reuse(s) < 0) ||
	    (bind(s,(struct sockaddr *)&addr, sizeof(addr)) < 0) ||
	    (listen(s,1) < 0)) {
		int tmperrno = errno; /* Don't clobber the real reason we failed */

		close(s);
		sofree(so);
		/* Restore the real errno */
#ifdef _WIN32
		WSASetLastError(tmperrno);
#else
		errno = tmperrno;
#endif
		return NULL;
	}
	qemu_setsockopt(s, SOL_SOCKET, SO_OOBINLINE, &opt, sizeof(int));

	getsockname(s,(struct sockaddr *)&addr,&addrlen);
	so->so_ffamily = AF_INET;
	so->so_fport = addr.sin_port;
	if (addr.sin_addr.s_addr == 0 || addr.sin_addr.s_addr == loopback_addr.s_addr)
	   so->so_faddr = slirp->vhost_addr;
	else
	   so->so_faddr = addr.sin_addr;

	so->s = s;
	return so;
}

/*
 * Various session state calls
 * XXX Should be #define's
 * The socket state stuff needs work, these often get call 2 or 3
 * times each when only 1 was needed
 */
void
soisfconnecting(struct socket *so)
{
	so->so_state &= ~(SS_NOFDREF|SS_ISFCONNECTED|SS_FCANTRCVMORE|
			  SS_FCANTSENDMORE|SS_FWDRAIN);
	so->so_state |= SS_ISFCONNECTING; /* Clobber other states */
}

void
soisfconnected(struct socket *so)
{
	so->so_state &= ~(SS_ISFCONNECTING|SS_FWDRAIN|SS_NOFDREF);
	so->so_state |= SS_ISFCONNECTED; /* Clobber other states */
}

static void
sofcantrcvmore(struct socket *so)
{
	if ((so->so_state & SS_NOFDREF) == 0) {
		shutdown(so->s,0);
	}
	so->so_state &= ~(SS_ISFCONNECTING);
	if (so->so_state & SS_FCANTSENDMORE) {
	   so->so_state &= SS_PERSISTENT_MASK;
	   so->so_state |= SS_NOFDREF; /* Don't select it */
	} else {
	   so->so_state |= SS_FCANTRCVMORE;
	}
}

static void
sofcantsendmore(struct socket *so)
{
	if ((so->so_state & SS_NOFDREF) == 0) {
            shutdown(so->s,1);           /* send FIN to fhost */
	}
	so->so_state &= ~(SS_ISFCONNECTING);
	if (so->so_state & SS_FCANTRCVMORE) {
	   so->so_state &= SS_PERSISTENT_MASK;
	   so->so_state |= SS_NOFDREF; /* as above */
	} else {
	   so->so_state |= SS_FCANTSENDMORE;
	}
}

/*
 * Set write drain mode
 * Set CANTSENDMORE once all data has been write()n
 */
void
sofwdrain(struct socket *so)
{
	if (so->so_rcv.sb_cc)
		so->so_state |= SS_FWDRAIN;
	else
		sofcantsendmore(so);
}

/*
 * Translate addr in host addr when it is a virtual address
 */
void sotranslate_out(struct socket *so, struct sockaddr_storage *addr)
{
    Slirp *slirp = so->slirp;
    struct sockaddr_in *sin = (struct sockaddr_in *)addr;

    switch (addr->ss_family) {
    case AF_INET:
        if ((so->so_faddr.s_addr & slirp->vnetwork_mask.s_addr) ==
                slirp->vnetwork_addr.s_addr) {
            /* It's an alias */
            if (so->so_faddr.s_addr == slirp->vnameserver_addr.s_addr) {
                if (get_dns_addr(&sin->sin_addr) < 0) {
                    sin->sin_addr = loopback_addr;
                }
            } else {
                sin->sin_addr = loopback_addr;
            }
        }

        DEBUG_MISC((dfd, " addr.sin_port=%d, "
            "addr.sin_addr.s_addr=%.16s\n",
            ntohs(sin->sin_port), inet_ntoa(sin->sin_addr)));
        break;

    default:
        break;
    }
}

void sotranslate_in(struct socket *so, struct sockaddr_storage *addr)
{
    Slirp *slirp = so->slirp;
    struct sockaddr_in *sin = (struct sockaddr_in *)addr;

    switch (addr->ss_family) {
    case AF_INET:
        if ((so->so_faddr.s_addr & slirp->vnetwork_mask.s_addr) ==
            slirp->vnetwork_addr.s_addr) {
            uint32_t inv_mask = ~slirp->vnetwork_mask.s_addr;

            if ((so->so_faddr.s_addr & inv_mask) == inv_mask) {
                sin->sin_addr = slirp->vhost_addr;
            } else if (sin->sin_addr.s_addr == loopback_addr.s_addr ||
                       so->so_faddr.s_addr != slirp->vhost_addr.s_addr) {
                sin->sin_addr = so->so_faddr;
            }
        }
        break;

    default:
        break;
    }
}

/*
 * Translate connections from localhost to the real hostname
 */
void sotranslate_accept(struct socket *so)
{
    Slirp *slirp = so->slirp;

    switch (so->so_ffamily) {
    case AF_INET:
        if (so->so_faddr.s_addr == INADDR_ANY ||
            (so->so_faddr.s_addr & loopback_mask) ==
            (loopback_addr.s_addr & loopback_mask)) {
           so->so_faddr = slirp->vhost_addr;
        }
        break;

    default:
        break;
    }
}