aboutsummaryrefslogtreecommitdiff
path: root/src/net/ipv4.c
blob: 9c5cf2eb49d5d65d04dbd48842ffb93645af8268 (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
#include <string.h>
#include <stdint.h>
#include <stdlib.h>
#include <stdio.h>
#include <errno.h>
#include <byteswap.h>
#include <ipxe/list.h>
#include <ipxe/in.h>
#include <ipxe/arp.h>
#include <ipxe/if_ether.h>
#include <ipxe/iobuf.h>
#include <ipxe/netdevice.h>
#include <ipxe/ip.h>
#include <ipxe/tcpip.h>
#include <ipxe/dhcp.h>
#include <ipxe/settings.h>
#include <ipxe/fragment.h>
#include <ipxe/ipstat.h>
#include <ipxe/profile.h>

/** @file
 *
 * IPv4 protocol
 *
 */

FILE_LICENCE ( GPL2_OR_LATER );

/* Unique IP datagram identification number (high byte) */
static uint8_t next_ident_high = 0;

/** List of IPv4 miniroutes */
struct list_head ipv4_miniroutes = LIST_HEAD_INIT ( ipv4_miniroutes );

/** IPv4 statistics */
static struct ip_statistics ipv4_stats;

/** IPv4 statistics family */
struct ip_statistics_family
ipv4_stats_family __ip_statistics_family ( IP_STATISTICS_IPV4 ) = {
	.version = 4,
	.stats = &ipv4_stats,
};

/** Transmit profiler */
static struct profiler ipv4_tx_profiler __profiler = { .name = "ipv4.tx" };

/** Receive profiler */
static struct profiler ipv4_rx_profiler __profiler = { .name = "ipv4.rx" };

/**
 * Add IPv4 minirouting table entry
 *
 * @v netdev		Network device
 * @v address		IPv4 address
 * @v netmask		Subnet mask
 * @v gateway		Gateway address (if any)
 * @ret miniroute	Routing table entry, or NULL
 */
static struct ipv4_miniroute * __malloc
add_ipv4_miniroute ( struct net_device *netdev, struct in_addr address,
		     struct in_addr netmask, struct in_addr gateway ) {
	struct ipv4_miniroute *miniroute;

	DBGC ( netdev, "IPv4 add %s", inet_ntoa ( address ) );
	DBGC ( netdev, "/%s ", inet_ntoa ( netmask ) );
	if ( gateway.s_addr )
		DBGC ( netdev, "gw %s ", inet_ntoa ( gateway ) );
	DBGC ( netdev, "via %s\n", netdev->name );

	/* Allocate and populate miniroute structure */
	miniroute = malloc ( sizeof ( *miniroute ) );
	if ( ! miniroute ) {
		DBGC ( netdev, "IPv4 could not add miniroute\n" );
		return NULL;
	}

	/* Record routing information */
	miniroute->netdev = netdev_get ( netdev );
	miniroute->address = address;
	miniroute->netmask = netmask;
	miniroute->gateway = gateway;
		
	/* Add to end of list if we have a gateway, otherwise
	 * to start of list.
	 */
	if ( gateway.s_addr ) {
		list_add_tail ( &miniroute->list, &ipv4_miniroutes );
	} else {
		list_add ( &miniroute->list, &ipv4_miniroutes );
	}

	return miniroute;
}

/**
 * Delete IPv4 minirouting table entry
 *
 * @v miniroute		Routing table entry
 */
static void del_ipv4_miniroute ( struct ipv4_miniroute *miniroute ) {
	struct net_device *netdev = miniroute->netdev;

	DBGC ( netdev, "IPv4 del %s", inet_ntoa ( miniroute->address ) );
	DBGC ( netdev, "/%s ", inet_ntoa ( miniroute->netmask ) );
	if ( miniroute->gateway.s_addr )
		DBGC ( netdev, "gw %s ", inet_ntoa ( miniroute->gateway ) );
	DBGC ( netdev, "via %s\n", miniroute->netdev->name );

	netdev_put ( miniroute->netdev );
	list_del ( &miniroute->list );
	free ( miniroute );
}

/**
 * Perform IPv4 routing
 *
 * @v dest		Final destination address
 * @ret dest		Next hop destination address
 * @ret miniroute	Routing table entry to use, or NULL if no route
 *
 * If the route requires use of a gateway, the next hop destination
 * address will be overwritten with the gateway address.
 */
static struct ipv4_miniroute * ipv4_route ( struct in_addr *dest ) {
	struct ipv4_miniroute *miniroute;
	int local;
	int has_gw;

	/* Find first usable route in routing table */
	list_for_each_entry ( miniroute, &ipv4_miniroutes, list ) {
		if ( ! netdev_is_open ( miniroute->netdev ) )
			continue;
		local = ( ( ( dest->s_addr ^ miniroute->address.s_addr )
			    & miniroute->netmask.s_addr ) == 0 );
		has_gw = ( miniroute->gateway.s_addr );
		if ( local || has_gw ) {
			if ( ! local )
				*dest = miniroute->gateway;
			return miniroute;
		}
	}

	return NULL;
}

/**
 * Determine transmitting network device
 *
 * @v st_dest		Destination network-layer address
 * @ret netdev		Transmitting network device, or NULL
 */
static struct net_device * ipv4_netdev ( struct sockaddr_tcpip *st_dest ) {
	struct sockaddr_in *sin_dest = ( ( struct sockaddr_in * ) st_dest );
	struct in_addr dest = sin_dest->sin_addr;
	struct ipv4_miniroute *miniroute;

	/* Find routing table entry */
	miniroute = ipv4_route ( &dest );
	if ( ! miniroute )
		return NULL;

	return miniroute->netdev;
}

/**
 * Check if IPv4 fragment matches fragment reassembly buffer
 *
 * @v fragment		Fragment reassembly buffer
 * @v iobuf		I/O buffer
 * @v hdrlen		Length of non-fragmentable potion of I/O buffer
 * @ret is_fragment	Fragment matches this reassembly buffer
 */
static int ipv4_is_fragment ( struct fragment *fragment,
			      struct io_buffer *iobuf,
			      size_t hdrlen __unused ) {
	struct iphdr *frag_iphdr = fragment->iobuf->data;
	struct iphdr *iphdr = iobuf->data;

	return ( ( iphdr->src.s_addr == frag_iphdr->src.s_addr ) &&
		 ( iphdr->ident == frag_iphdr->ident ) );
}

/**
 * Get IPv4 fragment offset
 *
 * @v iobuf		I/O buffer
 * @v hdrlen		Length of non-fragmentable potion of I/O buffer
 * @ret offset		Offset
 */
static size_t ipv4_fragment_offset ( struct io_buffer *iobuf,
				     size_t hdrlen __unused ) {
	struct iphdr *iphdr = iobuf->data;

	return ( ( ntohs ( iphdr->frags ) & IP_MASK_OFFSET ) << 3 );
}

/**
 * Check if more fragments exist
 *
 * @v iobuf		I/O buffer
 * @v hdrlen		Length of non-fragmentable potion of I/O buffer
 * @ret more_frags	More fragments exist
 */
static int ipv4_more_fragments ( struct io_buffer *iobuf,
				 size_t hdrlen __unused ) {
	struct iphdr *iphdr = iobuf->data;

	return ( iphdr->frags & htons ( IP_MASK_MOREFRAGS ) );
}

/** IPv4 fragment reassembler */
static struct fragment_reassembler ipv4_reassembler = {
	.list = LIST_HEAD_INIT ( ipv4_reassembler.list ),
	.is_fragment = ipv4_is_fragment,
	.fragment_offset = ipv4_fragment_offset,
	.more_fragments = ipv4_more_fragments,
	.stats = &ipv4_stats,
};

/**
 * Add IPv4 pseudo-header checksum to existing checksum
 *
 * @v iobuf		I/O buffer
 * @v csum		Existing checksum
 * @ret csum		Updated checksum
 */
static uint16_t ipv4_pshdr_chksum ( struct io_buffer *iobuf, uint16_t csum ) {
	struct ipv4_pseudo_header pshdr;
	struct iphdr *iphdr = iobuf->data;
	size_t hdrlen = ( ( iphdr->verhdrlen & IP_MASK_HLEN ) * 4 );

	/* Build pseudo-header */
	pshdr.src = iphdr->src;
	pshdr.dest = iphdr->dest;
	pshdr.zero_padding = 0x00;
	pshdr.protocol = iphdr->protocol;
	pshdr.len = htons ( iob_len ( iobuf ) - hdrlen );

	/* Update the checksum value */
	return tcpip_continue_chksum ( csum, &pshdr, sizeof ( pshdr ) );
}

/**
 * Transmit IP packet
 *
 * @v iobuf		I/O buffer
 * @v tcpip		Transport-layer protocol
 * @v st_src		Source network-layer address
 * @v st_dest		Destination network-layer address
 * @v netdev		Network device to use if no route found, or NULL
 * @v trans_csum	Transport-layer checksum to complete, or NULL
 * @ret rc		Status
 *
 * This function expects a transport-layer segment and prepends the IP header
 */
static int ipv4_tx ( struct io_buffer *iobuf,
		     struct tcpip_protocol *tcpip_protocol,
		     struct sockaddr_tcpip *st_src,
		     struct sockaddr_tcpip *st_dest,
		     struct net_device *netdev,
		     uint16_t *trans_csum ) {
	struct iphdr *iphdr = iob_push ( iobuf, sizeof ( *iphdr ) );
	struct sockaddr_in *sin_src = ( ( struct sockaddr_in * ) st_src );
	struct sockaddr_in *sin_dest = ( ( struct sockaddr_in * ) st_dest );
	struct ipv4_miniroute *miniroute;
	struct in_addr next_hop;
	struct in_addr netmask = { .s_addr = 0 };
	uint8_t ll_dest_buf[MAX_LL_ADDR_LEN];
	const void *ll_dest;
	int rc;

	/* Start profiling */
	profile_start ( &ipv4_tx_profiler );

	/* Update statistics */
	ipv4_stats.out_requests++;

	/* Fill up the IP header, except source address */
	memset ( iphdr, 0, sizeof ( *iphdr ) );
	iphdr->verhdrlen = ( IP_VER | ( sizeof ( *iphdr ) / 4 ) );
	iphdr->service = IP_TOS;
	iphdr->len = htons ( iob_len ( iobuf ) );	
	iphdr->ttl = IP_TTL;
	iphdr->protocol = tcpip_protocol->tcpip_proto;
	iphdr->dest = sin_dest->sin_addr;

	/* Use routing table to identify next hop and transmitting netdev */
	next_hop = iphdr->dest;
	if ( sin_src )
		iphdr->src = sin_src->sin_addr;
	if ( ( next_hop.s_addr != INADDR_BROADCAST ) &&
	     ( ! IN_MULTICAST ( ntohl ( next_hop.s_addr ) ) ) &&
	     ( ( miniroute = ipv4_route ( &next_hop ) ) != NULL ) ) {
		iphdr->src = miniroute->address;
		netmask = miniroute->netmask;
		netdev = miniroute->netdev;
	}
	if ( ! netdev ) {
		DBGC ( sin_dest->sin_addr, "IPv4 has no route to %s\n",
		       inet_ntoa ( iphdr->dest ) );
		ipv4_stats.out_no_routes++;
		rc = -ENETUNREACH;
		goto err;
	}

	/* (Ab)use the "ident" field to convey metadata about the
	 * network device statistics into packet traces.  Useful for
	 * extracting debug information from non-debug builds.
	 */
	iphdr->ident = htons ( ( (++next_ident_high) << 8 ) |
			       ( ( netdev->rx_stats.bad & 0xf ) << 4 ) |
			       ( ( netdev->rx_stats.good & 0xf ) << 0 ) );

	/* Fix up checksums */
	if ( trans_csum )
		*trans_csum = ipv4_pshdr_chksum ( iobuf, *trans_csum );
	iphdr->chksum = tcpip_chksum ( iphdr, sizeof ( *iphdr ) );

	/* Print IP4 header for debugging */
	DBGC2 ( sin_dest->sin_addr, "IPv4 TX %s->", inet_ntoa ( iphdr->src ) );
	DBGC2 ( sin_dest->sin_addr, "%s len %d proto %d id %04x csum %04x\n",
		inet_ntoa ( iphdr->dest ), ntohs ( iphdr->len ),
		iphdr->protocol, ntohs ( iphdr->ident ),
		ntohs ( iphdr->chksum ) );

	/* Calculate link-layer destination address, if possible */
	if ( ( ( next_hop.s_addr ^ INADDR_BROADCAST ) & ~netmask.s_addr ) == 0){
		/* Broadcast address */
		ipv4_stats.out_bcast_pkts++;
		ll_dest = netdev->ll_broadcast;
	} else if ( IN_MULTICAST ( ntohl ( next_hop.s_addr ) ) ) {
		/* Multicast address */
		ipv4_stats.out_mcast_pkts++;
		if ( ( rc = netdev->ll_protocol->mc_hash ( AF_INET, &next_hop,
							   ll_dest_buf ) ) !=0){
			DBGC ( sin_dest->sin_addr, "IPv4 could not hash "
			       "multicast %s: %s\n",
			       inet_ntoa ( next_hop ), strerror ( rc ) );
			goto err;
		}
		ll_dest = ll_dest_buf;
	} else {
		/* Unicast address */
		ll_dest = NULL;
	}

	/* Update statistics */
	ipv4_stats.out_transmits++;
	ipv4_stats.out_octets += iob_len ( iobuf );

	/* Hand off to link layer (via ARP if applicable) */
	if ( ll_dest ) {
		if ( ( rc = net_tx ( iobuf, netdev, &ipv4_protocol, ll_dest,
				     netdev->ll_addr ) ) != 0 ) {
			DBGC ( sin_dest->sin_addr, "IPv4 could not transmit "
			       "packet via %s: %s\n",
			       netdev->name, strerror ( rc ) );
			return rc;
		}
	} else {
		if ( ( rc = arp_tx ( iobuf, netdev, &ipv4_protocol, &next_hop,
				     &iphdr->src, netdev->ll_addr ) ) != 0 ) {
			DBGC ( sin_dest->sin_addr, "IPv4 could not transmit "
			       "packet via %s: %s\n",
			       netdev->name, strerror ( rc ) );
			return rc;
		}
	}

	profile_stop ( &ipv4_tx_profiler );
	return 0;

 err:
	free_iob ( iobuf );
	return rc;
}

/**
 * Check if network device has any IPv4 address
 *
 * @v netdev		Network device
 * @ret has_any_addr	Network device has any IPv4 address
 */
int ipv4_has_any_addr ( struct net_device *netdev ) {
	struct ipv4_miniroute *miniroute;

	list_for_each_entry ( miniroute, &ipv4_miniroutes, list ) {
		if ( miniroute->netdev == netdev )
			return 1;
	}
	return 0;
}

/**
 * Check if network device has a specific IPv4 address
 *
 * @v netdev		Network device
 * @v addr		IPv4 address
 * @ret has_addr	Network device has this IPv4 address
 */
static int ipv4_has_addr ( struct net_device *netdev, struct in_addr addr ) {
	struct ipv4_miniroute *miniroute;

	list_for_each_entry ( miniroute, &ipv4_miniroutes, list ) {
		if ( ( miniroute->netdev == netdev ) &&
		     ( miniroute->address.s_addr == addr.s_addr ) ) {
			/* Found matching address */
			return 1;
		}
	}
	return 0;
}

/**
 * Process incoming packets
 *
 * @v iobuf		I/O buffer
 * @v netdev		Network device
 * @v ll_dest		Link-layer destination address
 * @v ll_source		Link-layer destination source
 * @v flags		Packet flags
 * @ret rc		Return status code
 *
 * This function expects an IP4 network datagram. It processes the headers 
 * and sends it to the transport layer.
 */
static int ipv4_rx ( struct io_buffer *iobuf,
		     struct net_device *netdev,
		     const void *ll_dest __unused,
		     const void *ll_source __unused,
		     unsigned int flags ) {
	struct iphdr *iphdr = iobuf->data;
	size_t hdrlen;
	size_t len;
	union {
		struct sockaddr_in sin;
		struct sockaddr_tcpip st;
	} src, dest;
	uint16_t csum;
	uint16_t pshdr_csum;
	int rc;

	/* Start profiling */
	profile_start ( &ipv4_rx_profiler );

	/* Update statistics */
	ipv4_stats.in_receives++;
	ipv4_stats.in_octets += iob_len ( iobuf );
	if ( flags & LL_BROADCAST ) {
		ipv4_stats.in_bcast_pkts++;
	} else if ( flags & LL_MULTICAST ) {
		ipv4_stats.in_mcast_pkts++;
	}

	/* Sanity check the IPv4 header */
	if ( iob_len ( iobuf ) < sizeof ( *iphdr ) ) {
		DBGC ( iphdr->src, "IPv4 packet too short at %zd bytes (min "
		       "%zd bytes)\n", iob_len ( iobuf ), sizeof ( *iphdr ) );
		goto err_header;
	}
	if ( ( iphdr->verhdrlen & IP_MASK_VER ) != IP_VER ) {
		DBGC ( iphdr->src, "IPv4 version %#02x not supported\n",
		       iphdr->verhdrlen );
		goto err_header;
	}
	hdrlen = ( ( iphdr->verhdrlen & IP_MASK_HLEN ) * 4 );
	if ( hdrlen < sizeof ( *iphdr ) ) {
		DBGC ( iphdr->src, "IPv4 header too short at %zd bytes (min "
		       "%zd bytes)\n", hdrlen, sizeof ( *iphdr ) );
		goto err_header;
	}
	if ( hdrlen > iob_len ( iobuf ) ) {
		DBGC ( iphdr->src, "IPv4 header too long at %zd bytes "
		       "(packet is %zd bytes)\n", hdrlen, iob_len ( iobuf ) );
		goto err_header;
	}
	if ( ( csum = tcpip_chksum ( iphdr, hdrlen ) ) != 0 ) {
		DBGC ( iphdr->src, "IPv4 checksum incorrect (is %04x "
		       "including checksum field, should be 0000)\n", csum );
		goto err_header;
	}
	len = ntohs ( iphdr->len );
	if ( len < hdrlen ) {
		DBGC ( iphdr->src, "IPv4 length too short at %zd bytes "
		       "(header is %zd bytes)\n", len, hdrlen );
		goto err_header;
	}
	if ( len > iob_len ( iobuf ) ) {
		DBGC ( iphdr->src, "IPv4 length too long at %zd bytes "
		       "(packet is %zd bytes)\n", len, iob_len ( iobuf ) );
		ipv4_stats.in_truncated_pkts++;
		goto err_other;
	}

	/* Truncate packet to correct length */
	iob_unput ( iobuf, ( iob_len ( iobuf ) - len ) );

	/* Print IPv4 header for debugging */
	DBGC2 ( iphdr->src, "IPv4 RX %s<-", inet_ntoa ( iphdr->dest ) );
	DBGC2 ( iphdr->src, "%s len %d proto %d id %04x csum %04x\n",
		inet_ntoa ( iphdr->src ), ntohs ( iphdr->len ), iphdr->protocol,
		ntohs ( iphdr->ident ), ntohs ( iphdr->chksum ) );

	/* Discard unicast packets not destined for us */
	if ( ( ! ( flags & LL_MULTICAST ) ) &&
	     ipv4_has_any_addr ( netdev ) &&
	     ( ! ipv4_has_addr ( netdev, iphdr->dest ) ) ) {
		DBGC ( iphdr->src, "IPv4 discarding non-local unicast packet "
		       "for %s\n", inet_ntoa ( iphdr->dest ) );
		ipv4_stats.in_addr_errors++;
		goto err_other;
	}

	/* Perform fragment reassembly if applicable */
	if ( iphdr->frags & htons ( IP_MASK_OFFSET | IP_MASK_MOREFRAGS ) ) {
		/* Pass the fragment to fragment_reassemble() which returns
		 * either a fully reassembled I/O buffer or NULL.
		 */
		iobuf = fragment_reassemble ( &ipv4_reassembler, iobuf,
					      &hdrlen );
		if ( ! iobuf )
			return 0;
		iphdr = iobuf->data;
	}

	/* Construct socket addresses, calculate pseudo-header
	 * checksum, and hand off to transport layer
	 */
	memset ( &src, 0, sizeof ( src ) );
	src.sin.sin_family = AF_INET;
	src.sin.sin_addr = iphdr->src;
	memset ( &dest, 0, sizeof ( dest ) );
	dest.sin.sin_family = AF_INET;
	dest.sin.sin_addr = iphdr->dest;
	pshdr_csum = ipv4_pshdr_chksum ( iobuf, TCPIP_EMPTY_CSUM );
	iob_pull ( iobuf, hdrlen );
	if ( ( rc = tcpip_rx ( iobuf, netdev, iphdr->protocol, &src.st,
			       &dest.st, pshdr_csum, &ipv4_stats ) ) != 0 ) {
		DBGC ( src.sin.sin_addr, "IPv4 received packet rejected by "
		       "stack: %s\n", strerror ( rc ) );
		return rc;
	}

	profile_stop ( &ipv4_rx_profiler );
	return 0;

 err_header:
	ipv4_stats.in_hdr_errors++;
 err_other:
	free_iob ( iobuf );
	return -EINVAL;
}

/** 
 * Check existence of IPv4 address for ARP
 *
 * @v netdev		Network device
 * @v net_addr		Network-layer address
 * @ret rc		Return status code
 */
static int ipv4_arp_check ( struct net_device *netdev, const void *net_addr ) {
	const struct in_addr *address = net_addr;

	if ( ipv4_has_addr ( netdev, *address ) )
		return 0;

	return -ENOENT;
}

/**
 * Convert IPv4 address to dotted-quad notation
 *
 * @v in	IP address
 * @ret string	IP address in dotted-quad notation
 */
char * inet_ntoa ( struct in_addr in ) {
	static char buf[16]; /* "xxx.xxx.xxx.xxx" */
	uint8_t *bytes = ( uint8_t * ) &in;
	
	sprintf ( buf, "%d.%d.%d.%d", bytes[0], bytes[1], bytes[2], bytes[3] );
	return buf;
}

/**
 * Transcribe IP address
 *
 * @v net_addr	IP address
 * @ret string	IP address in dotted-quad notation
 *
 */
static const char * ipv4_ntoa ( const void *net_addr ) {
	return inet_ntoa ( * ( ( struct in_addr * ) net_addr ) );
}

/**
 * Transcribe IPv4 socket address
 *
 * @v sa		Socket address
 * @ret string		Socket address in standard notation
 */
static const char * ipv4_sock_ntoa ( struct sockaddr *sa ) {
	struct sockaddr_in *sin = ( ( struct sockaddr_in * ) sa );

	return inet_ntoa ( sin->sin_addr );
}

/**
 * Parse IPv4 socket address
 *
 * @v string		Socket address string
 * @v sa		Socket address to fill in
 * @ret rc		Return status code
 */
static int ipv4_sock_aton ( const char *string, struct sockaddr *sa ) {
	struct sockaddr_in *sin = ( ( struct sockaddr_in * ) sa );
	struct in_addr in;

	if ( inet_aton ( string, &in ) ) {
		sin->sin_addr = in;
		return 0;
	}
	return -EINVAL;
}

/** IPv4 protocol */
struct net_protocol ipv4_protocol __net_protocol = {
	.name = "IP",
	.net_proto = htons ( ETH_P_IP ),
	.net_addr_len = sizeof ( struct in_addr ),
	.rx = ipv4_rx,
	.ntoa = ipv4_ntoa,
};

/** IPv4 TCPIP net protocol */
struct tcpip_net_protocol ipv4_tcpip_protocol __tcpip_net_protocol = {
	.name = "IPv4",
	.sa_family = AF_INET,
	.header_len = sizeof ( struct iphdr ),
	.tx = ipv4_tx,
	.netdev = ipv4_netdev,
};

/** IPv4 ARP protocol */
struct arp_net_protocol ipv4_arp_protocol __arp_net_protocol = {
	.net_protocol = &ipv4_protocol,
	.check = ipv4_arp_check,
};

/** IPv4 socket address converter */
struct sockaddr_converter ipv4_sockaddr_converter __sockaddr_converter = {
	.family = AF_INET,
	.ntoa = ipv4_sock_ntoa,
	.aton = ipv4_sock_aton,
};

/******************************************************************************
 *
 * Settings
 *
 ******************************************************************************
 */

/**
 * Parse IPv4 address setting value
 *
 * @v type		Setting type
 * @v value		Formatted setting value
 * @v buf		Buffer to contain raw value
 * @v len		Length of buffer
 * @ret len		Length of raw value, or negative error
 */
int parse_ipv4_setting ( const struct setting_type *type __unused,
			 const char *value, void *buf, size_t len ) {
	struct in_addr ipv4;

	/* Parse IPv4 address */
	if ( inet_aton ( value, &ipv4 ) == 0 )
		return -EINVAL;

	/* Copy to buffer */
	if ( len > sizeof ( ipv4 ) )
		len = sizeof ( ipv4 );
	memcpy ( buf, &ipv4, len );

	return ( sizeof ( ipv4 ) );
}

/**
 * Format IPv4 address setting value
 *
 * @v type		Setting type
 * @v raw		Raw setting value
 * @v raw_len		Length of raw setting value
 * @v buf		Buffer to contain formatted value
 * @v len		Length of buffer
 * @ret len		Length of formatted value, or negative error
 */
int format_ipv4_setting ( const struct setting_type *type __unused,
			  const void *raw, size_t raw_len, char *buf,
			  size_t len ) {
	const struct in_addr *ipv4 = raw;

	if ( raw_len < sizeof ( *ipv4 ) )
		return -EINVAL;
	return snprintf ( buf, len, "%s", inet_ntoa ( *ipv4 ) );
}

/** IPv4 address setting */
const struct setting ip_setting __setting ( SETTING_IP, ip ) = {
	.name = "ip",
	.description = "IP address",
	.tag = DHCP_EB_YIADDR,
	.type = &setting_type_ipv4,
};

/** IPv4 subnet mask setting */
const struct setting netmask_setting __setting ( SETTING_IP, netmask ) = {
	.name = "netmask",
	.description = "Subnet mask",
	.tag = DHCP_SUBNET_MASK,
	.type = &setting_type_ipv4,
};

/** Default gateway setting */
const struct setting gateway_setting __setting ( SETTING_IP, gateway ) = {
	.name = "gateway",
	.description = "Default gateway",
	.tag = DHCP_ROUTERS,
	.type = &setting_type_ipv4,
};

/**
 * Create IPv4 routing table based on configured settings
 *
 * @ret rc		Return status code
 */
static int ipv4_create_routes ( void ) {
	struct ipv4_miniroute *miniroute;
	struct ipv4_miniroute *tmp;
	struct net_device *netdev;
	struct settings *settings;
	struct in_addr address = { 0 };
	struct in_addr netmask = { 0 };
	struct in_addr gateway = { 0 };

	/* Delete all existing routes */
	list_for_each_entry_safe ( miniroute, tmp, &ipv4_miniroutes, list )
		del_ipv4_miniroute ( miniroute );

	/* Create a route for each configured network device */
	for_each_netdev ( netdev ) {
		settings = netdev_settings ( netdev );
		/* Get IPv4 address */
		address.s_addr = 0;
		fetch_ipv4_setting ( settings, &ip_setting, &address );
		if ( ! address.s_addr )
			continue;
		/* Get subnet mask */
		fetch_ipv4_setting ( settings, &netmask_setting, &netmask );
		/* Calculate default netmask, if necessary */
		if ( ! netmask.s_addr ) {
			if ( IN_CLASSA ( ntohl ( address.s_addr ) ) ) {
				netmask.s_addr = htonl ( IN_CLASSA_NET );
			} else if ( IN_CLASSB ( ntohl ( address.s_addr ) ) ) {
				netmask.s_addr = htonl ( IN_CLASSB_NET );
			} else if ( IN_CLASSC ( ntohl ( address.s_addr ) ) ) {
				netmask.s_addr = htonl ( IN_CLASSC_NET );
			}
		}
		/* Get default gateway, if present */
		fetch_ipv4_setting ( settings, &gateway_setting, &gateway );
		/* Configure route */
		miniroute = add_ipv4_miniroute ( netdev, address,
						 netmask, gateway );
		if ( ! miniroute )
			return -ENOMEM;
	}

	return 0;
}

/** IPv4 settings applicator */
struct settings_applicator ipv4_settings_applicator __settings_applicator = {
	.apply = ipv4_create_routes,
};

/* Drag in ICMPv4 */
REQUIRE_OBJECT ( icmpv4 );