Boot Linux faster!

Check our new training course

Boot Linux faster!

Check our new training course
and Creative Commons CC-BY-SA
lecture and lab materials

Bootlin logo

Elixir Cross Referencer

   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
 812
 813
 814
 815
 816
 817
 818
 819
 820
 821
 822
 823
 824
 825
 826
 827
 828
 829
 830
 831
 832
 833
 834
 835
 836
 837
 838
 839
 840
 841
 842
 843
 844
 845
 846
 847
 848
 849
 850
 851
 852
 853
 854
 855
 856
 857
 858
 859
 860
 861
 862
 863
 864
 865
 866
 867
 868
 869
 870
 871
 872
 873
 874
 875
 876
 877
 878
 879
 880
 881
 882
 883
 884
 885
 886
 887
 888
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
/*
 * Copyright (c) 2017 Linaro Limited
 *
 * SPDX-License-Identifier: Apache-2.0
 */

/* libc headers */
#include <fcntl.h>

/* Zephyr headers */
#include <logging/log.h>
LOG_MODULE_REGISTER(net_sock, CONFIG_NET_SOCKETS_LOG_LEVEL);

#include <kernel.h>
#include <net/net_context.h>
#include <net/net_pkt.h>
#include <net/socket.h>
#include <syscall_handler.h>
#include <misc/fdtable.h>

#include "sockets_internal.h"

#define SET_ERRNO(x) \
	{ int _err = x; if (_err < 0) { errno = -_err; return -1; } }

#define VTABLE_CALL(fn, sock, ...) \
	do { \
		const struct socket_op_vtable *vtable; \
		void *ctx = get_sock_vtable(sock, &vtable); \
		if (ctx == NULL || vtable->fn == NULL) { \
			return -1; \
		} \
		return vtable->fn(ctx, __VA_ARGS__); \
	} while (0)

const struct socket_op_vtable sock_fd_op_vtable;

static inline void *get_sock_vtable(
			int sock, const struct socket_op_vtable **vtable)
{
	return z_get_fd_obj_and_vtable(sock,
				       (const struct fd_op_vtable **)vtable);
}

static void zsock_received_cb(struct net_context *ctx,
			      struct net_pkt *pkt,
			      union net_ip_header *ip_hdr,
			      union net_proto_header *proto_hdr,
			      int status,
			      void *user_data);

static inline int k_fifo_wait_non_empty(struct k_fifo *fifo, int32_t timeout)
{
	struct k_poll_event events[] = {
		K_POLL_EVENT_INITIALIZER(K_POLL_TYPE_FIFO_DATA_AVAILABLE,
					 K_POLL_MODE_NOTIFY_ONLY, fifo),
	};

	return k_poll(events, ARRAY_SIZE(events), timeout);
}

static void zsock_flush_queue(struct net_context *ctx)
{
	bool is_listen = net_context_get_state(ctx) == NET_CONTEXT_LISTENING;
	void *p;

	/* recv_q and accept_q are shared via a union */
	while ((p = k_fifo_get(&ctx->recv_q, K_NO_WAIT)) != NULL) {
		if (is_listen) {
			NET_DBG("discarding ctx %p", p);
			net_context_put(p);
		} else {
			NET_DBG("discarding pkt %p", p);
			net_pkt_unref(p);
		}
	}

	/* Some threads might be waiting on recv, cancel the wait */
	k_fifo_cancel_wait(&ctx->recv_q);
}

int zsock_socket_internal(int family, int type, int proto)
{
	int fd = z_reserve_fd();
	struct net_context *ctx;
	int res;

	if (fd < 0) {
		return -1;
	}

	res = net_context_get(family, type, proto, &ctx);
	if (res < 0) {
		z_free_fd(fd);
		errno = -res;
		return -1;
	}

	/* Initialize user_data, all other calls will preserve it */
	ctx->user_data = NULL;

	/* recv_q and accept_q are in union */
	k_fifo_init(&ctx->recv_q);

#ifdef CONFIG_USERSPACE
	/* Set net context object as initialized and grant access to the
	 * calling thread (and only the calling thread)
	 */
	z_object_recycle(ctx);
#endif

	/* TCP context is effectively owned by both application
	 * and the stack: stack may detect that peer closed/aborted
	 * connection, but it must not dispose of the context behind
	 * the application back. Likewise, when application "closes"
	 * context, it's not disposed of immediately - there's yet
	 * closing handshake for stack to perform.
	 */
	if (proto == IPPROTO_TCP) {
		net_context_ref(ctx);
	}

	z_finalize_fd(fd, ctx, (const struct fd_op_vtable *)&sock_fd_op_vtable);

	NET_DBG("socket: ctx=%p, fd=%d", ctx, fd);

	return fd;
}

int z_impl_zsock_socket(int family, int type, int proto)
{
#if defined(CONFIG_NET_SOCKETS_SOCKOPT_TLS)
	if (((proto >= IPPROTO_TLS_1_0) && (proto <= IPPROTO_TLS_1_2)) ||
	    (proto >= IPPROTO_DTLS_1_0 && proto <= IPPROTO_DTLS_1_2)) {
		return ztls_socket(family, type, proto);
	}
#endif

#if defined(CONFIG_NET_SOCKETS_PACKET)
	if (family == AF_PACKET) {
		return zpacket_socket(family, type, proto);
	}
#endif

#if defined(CONFIG_NET_SOCKETS_CAN)
	if (family == AF_CAN) {
		return zcan_socket(family, type, proto);
	}
#endif

	return zsock_socket_internal(family, type, proto);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_socket, family, type, proto)
{
	/* implementation call to net_context_get() should do all necessary
	 * checking
	 */
	return z_impl_zsock_socket(family, type, proto);
}
#endif /* CONFIG_USERSPACE */

int zsock_close_ctx(struct net_context *ctx)
{
#ifdef CONFIG_USERSPACE
	z_object_uninit(ctx);
#endif
	/* Reset callbacks to avoid any race conditions while
	 * flushing queues. No need to check return values here,
	 * as these are fail-free operations and we're closing
	 * socket anyway.
	 */
	if (net_context_get_state(ctx) == NET_CONTEXT_LISTENING) {
		(void)net_context_accept(ctx, NULL, K_NO_WAIT, NULL);
	} else {
		(void)net_context_recv(ctx, NULL, K_NO_WAIT, NULL);
	}

	zsock_flush_queue(ctx);

	SET_ERRNO(net_context_put(ctx));

	return 0;
}

int z_impl_zsock_close(int sock)
{
	const struct fd_op_vtable *vtable;
	void *ctx = z_get_fd_obj_and_vtable(sock, &vtable);

	if (ctx == NULL) {
		return -1;
	}

	z_free_fd(sock);

	NET_DBG("close: ctx=%p, fd=%d", ctx, sock);

	return z_fdtable_call_ioctl(vtable, ctx, ZFD_IOCTL_CLOSE);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_close, sock)
{
	return z_impl_zsock_close(sock);
}
#endif /* CONFIG_USERSPACE */

int z_impl_zsock_shutdown(int sock, int how)
{
	/* shutdown() is described by POSIX as just disabling recv() and/or
	 * send() operations on socket. Of course, real-world software mostly
	 * calls it for side effects. We treat it as null operation so far.
	 */
	ARG_UNUSED(sock);
	ARG_UNUSED(how);

	LOG_WRN("shutdown() not implemented");

	return 0;
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_shutdown, sock, how)
{
	return z_impl_zsock_shutdown(sock, how);
}
#endif /* CONFIG_USERSPACE */

static void zsock_accepted_cb(struct net_context *new_ctx,
			      struct sockaddr *addr, socklen_t addrlen,
			      int status, void *user_data) {
	struct net_context *parent = user_data;

	NET_DBG("parent=%p, ctx=%p, st=%d", parent, new_ctx, status);

	if (status == 0) {
		/* This just installs a callback, so cannot fail. */
		(void)net_context_recv(new_ctx, zsock_received_cb, K_NO_WAIT,
				       NULL);
		k_fifo_init(&new_ctx->recv_q);

		k_fifo_put(&parent->accept_q, new_ctx);
	}
}

static void zsock_received_cb(struct net_context *ctx,
			      struct net_pkt *pkt,
			      union net_ip_header *ip_hdr,
			      union net_proto_header *proto_hdr,
			      int status,
			      void *user_data)
{
	NET_DBG("ctx=%p, pkt=%p, st=%d, user_data=%p", ctx, pkt, status,
		user_data);

	/* if pkt is NULL, EOF */
	if (!pkt) {
		struct net_pkt *last_pkt = k_fifo_peek_tail(&ctx->recv_q);

		if (!last_pkt) {
			/* If there're no packets in the queue, recv() may
			 * be blocked waiting on it to become non-empty,
			 * so cancel that wait.
			 */
			sock_set_eof(ctx);
			k_fifo_cancel_wait(&ctx->recv_q);
			NET_DBG("Marked socket %p as peer-closed", ctx);
		} else {
			net_pkt_set_eof(last_pkt, true);
			NET_DBG("Set EOF flag on pkt %p", last_pkt);
		}
		return;
	}

	/* Normal packet */
	net_pkt_set_eof(pkt, false);

	if (net_context_get_type(ctx) == SOCK_STREAM) {
		net_context_update_recv_wnd(ctx, -net_pkt_remaining_data(pkt));
	}

	k_fifo_put(&ctx->recv_q, pkt);
}

int zsock_bind_ctx(struct net_context *ctx, const struct sockaddr *addr,
		   socklen_t addrlen)
{
	SET_ERRNO(net_context_bind(ctx, addr, addrlen));
	/* For DGRAM socket, we expect to receive packets after call to
	 * bind(), but for STREAM socket, next expected operation is
	 * listen(), which doesn't work if recv callback is set.
	 */
	if (net_context_get_type(ctx) == SOCK_DGRAM) {
		SET_ERRNO(net_context_recv(ctx, zsock_received_cb, K_NO_WAIT,
					   ctx->user_data));
	}

	return 0;
}

int z_impl_zsock_bind(int sock, const struct sockaddr *addr, socklen_t addrlen)
{
	VTABLE_CALL(bind, sock, addr, addrlen);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_bind, sock, addr, addrlen)
{
	struct sockaddr_storage dest_addr_copy;

	Z_OOPS(Z_SYSCALL_VERIFY(addrlen <= sizeof(dest_addr_copy)));
	Z_OOPS(z_user_from_copy(&dest_addr_copy, (void *)addr, addrlen));

	return z_impl_zsock_bind(sock, (struct sockaddr *)&dest_addr_copy,
				addrlen);
}
#endif /* CONFIG_USERSPACE */

int zsock_connect_ctx(struct net_context *ctx, const struct sockaddr *addr,
		      socklen_t addrlen)
{
	SET_ERRNO(net_context_connect(ctx, addr, addrlen, NULL,
			      K_MSEC(CONFIG_NET_SOCKETS_CONNECT_TIMEOUT),
			      NULL));
	SET_ERRNO(net_context_recv(ctx, zsock_received_cb, K_NO_WAIT,
				   ctx->user_data));

	return 0;
}

int z_impl_zsock_connect(int sock, const struct sockaddr *addr,
			socklen_t addrlen)
{
	VTABLE_CALL(connect, sock, addr, addrlen);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_connect, sock, addr, addrlen)
{
	struct sockaddr_storage dest_addr_copy;

	Z_OOPS(Z_SYSCALL_VERIFY(addrlen <= sizeof(dest_addr_copy)));
	Z_OOPS(z_user_from_copy(&dest_addr_copy, (void *)addr, addrlen));

	return z_impl_zsock_connect(sock, (struct sockaddr *)&dest_addr_copy,
				   addrlen);
}
#endif /* CONFIG_USERSPACE */

int zsock_listen_ctx(struct net_context *ctx, int backlog)
{
	SET_ERRNO(net_context_listen(ctx, backlog));
	SET_ERRNO(net_context_accept(ctx, zsock_accepted_cb, K_NO_WAIT, ctx));

	return 0;
}

int z_impl_zsock_listen(int sock, int backlog)
{
	VTABLE_CALL(listen, sock, backlog);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_listen, sock, backlog)
{
	return z_impl_zsock_listen(sock, backlog);
}
#endif /* CONFIG_USERSPACE */

int zsock_accept_ctx(struct net_context *parent, struct sockaddr *addr,
		     socklen_t *addrlen)
{
	int fd;

	fd = z_reserve_fd();
	if (fd < 0) {
		return -1;
	}

	struct net_context *ctx = k_fifo_get(&parent->accept_q, K_FOREVER);

#ifdef CONFIG_USERSPACE
	z_object_recycle(ctx);
#endif

	if (addr != NULL && addrlen != NULL) {
		int len = MIN(*addrlen, sizeof(ctx->remote));

		memcpy(addr, &ctx->remote, len);
		/* addrlen is a value-result argument, set to actual
		 * size of source address
		 */
		if (ctx->remote.sa_family == AF_INET) {
			*addrlen = sizeof(struct sockaddr_in);
		} else if (ctx->remote.sa_family == AF_INET6) {
			*addrlen = sizeof(struct sockaddr_in6);
		} else {
			errno = ENOTSUP;
			return -1;
		}
	}

	/* TCP context is effectively owned by both application
	 * and the stack: stack may detect that peer closed/aborted
	 * connection, but it must not dispose of the context behind
	 * the application back. Likewise, when application "closes"
	 * context, it's not disposed of immediately - there's yet
	 * closing handshake for stack to perform.
	 */
	net_context_ref(ctx);

	NET_DBG("accept: ctx=%p, fd=%d", ctx, fd);

	z_finalize_fd(fd, ctx, (const struct fd_op_vtable *)&sock_fd_op_vtable);

	return fd;
}

int z_impl_zsock_accept(int sock, struct sockaddr *addr, socklen_t *addrlen)
{
	VTABLE_CALL(accept, sock, addr, addrlen);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_accept, sock, addr, addrlen)
{
	socklen_t addrlen_copy;
	int ret;

	Z_OOPS(z_user_from_copy(&addrlen_copy, (void *)addrlen,
			     sizeof(socklen_t)));

	if (Z_SYSCALL_MEMORY_WRITE(addr, addrlen_copy)) {
		errno = EFAULT;
		return -1;
	}

	ret = z_impl_zsock_accept(sock, (struct sockaddr *)addr, &addrlen_copy);

	if (ret >= 0 &&
	    z_user_to_copy((void *)addrlen, &addrlen_copy,
			   sizeof(socklen_t))) {
		errno = EINVAL;
		return -1;
	}

	return ret;
}
#endif /* CONFIG_USERSPACE */

ssize_t zsock_sendto_ctx(struct net_context *ctx, const void *buf, size_t len,
			 int flags,
			 const struct sockaddr *dest_addr, socklen_t addrlen)
{
	s32_t timeout = K_FOREVER;
	int status;

	if ((flags & ZSOCK_MSG_DONTWAIT) || sock_is_nonblock(ctx)) {
		timeout = K_NO_WAIT;
	}

	/* Register the callback before sending in order to receive the response
	 * from the peer.
	 */
	status = net_context_recv(ctx, zsock_received_cb,
				  K_NO_WAIT, ctx->user_data);
	if (status < 0) {
		errno = -status;
		return -1;
	}

	if (dest_addr) {
		status = net_context_sendto(ctx, buf, len, dest_addr,
					    addrlen, NULL, timeout,
					    ctx->user_data);
	} else {
		status = net_context_send(ctx, buf, len, NULL, timeout,
					  ctx->user_data);
	}

	if (status < 0) {
		errno = -status;
		return -1;
	}

	return status;
}

ssize_t z_impl_zsock_sendto(int sock, const void *buf, size_t len, int flags,
			   const struct sockaddr *dest_addr, socklen_t addrlen)
{
	VTABLE_CALL(sendto, sock, buf, len, flags, dest_addr, addrlen);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_sendto, sock, buf, len, flags, dest_addr, addrlen)
{
	struct sockaddr_storage dest_addr_copy;

	Z_OOPS(Z_SYSCALL_MEMORY_READ(buf, len));
	if (dest_addr) {
		Z_OOPS(Z_SYSCALL_VERIFY(addrlen <= sizeof(dest_addr_copy)));
		Z_OOPS(z_user_from_copy(&dest_addr_copy, (void *)dest_addr,
					addrlen));
	}

	return z_impl_zsock_sendto(sock, (const void *)buf, len, flags,
			dest_addr ? (struct sockaddr *)&dest_addr_copy : NULL,
			addrlen);
}
#endif /* CONFIG_USERSPACE */

static int sock_get_pkt_src_addr(struct net_pkt *pkt,
				 enum net_ip_protocol proto,
				 struct sockaddr *addr,
				 socklen_t addrlen)
{
	int ret = 0;
	struct net_pkt_cursor backup;
	u16_t *port;

	if (!addr || !pkt) {
		return -EINVAL;
	}

	net_pkt_cursor_backup(pkt, &backup);
	net_pkt_cursor_init(pkt);

	addr->sa_family = net_pkt_family(pkt);

	if (IS_ENABLED(CONFIG_NET_IPV4) &&
	    net_pkt_family(pkt) == AF_INET) {
		NET_PKT_DATA_ACCESS_CONTIGUOUS_DEFINE(ipv4_access,
						      struct net_ipv4_hdr);
		struct sockaddr_in *addr4 = net_sin(addr);
		struct net_ipv4_hdr *ipv4_hdr;

		if (addrlen < sizeof(struct sockaddr_in)) {
			ret = -EINVAL;
			goto error;
		}

		ipv4_hdr = (struct net_ipv4_hdr *)net_pkt_get_data(
							pkt, &ipv4_access);
		if (!ipv4_hdr || net_pkt_acknowledge_data(pkt, &ipv4_access)) {
			ret = -ENOBUFS;
			goto error;
		}

		net_ipaddr_copy(&addr4->sin_addr, &ipv4_hdr->src);
		port = &addr4->sin_port;
	} else if (IS_ENABLED(CONFIG_NET_IPV6) &&
		   net_pkt_family(pkt) == AF_INET6) {
		NET_PKT_DATA_ACCESS_CONTIGUOUS_DEFINE(ipv6_access,
						      struct net_ipv6_hdr);
		struct sockaddr_in6 *addr6 = net_sin6(addr);
		struct net_ipv6_hdr *ipv6_hdr;

		if (addrlen < sizeof(struct sockaddr_in6)) {
			ret = -EINVAL;
			goto error;
		}

		ipv6_hdr = (struct net_ipv6_hdr *)net_pkt_get_data(
							pkt, &ipv6_access);
		if (!ipv6_hdr ||
		    net_pkt_acknowledge_data(pkt, &ipv6_access) ||
		    net_pkt_skip(pkt, net_pkt_ipv6_ext_len(pkt))) {
			ret = -ENOBUFS;
			goto error;
		}

		net_ipaddr_copy(&addr6->sin6_addr, &ipv6_hdr->src);
		port = &addr6->sin6_port;
	} else {
		ret = -ENOTSUP;
		goto error;
	}

	if (IS_ENABLED(CONFIG_NET_UDP) && proto == IPPROTO_UDP) {
		NET_PKT_DATA_ACCESS_DEFINE(udp_access, struct net_udp_hdr);
		struct net_udp_hdr *udp_hdr;

		udp_hdr = (struct net_udp_hdr *)net_pkt_get_data(pkt,
								 &udp_access);
		if (!udp_hdr) {
			ret = -ENOBUFS;
			goto error;
		}

		*port = udp_hdr->src_port;
	} else if (IS_ENABLED(CONFIG_NET_TCP) && proto == IPPROTO_TCP) {
		NET_PKT_DATA_ACCESS_DEFINE(tcp_access, struct net_tcp_hdr);
		struct net_tcp_hdr *tcp_hdr;

		tcp_hdr = (struct net_tcp_hdr *)net_pkt_get_data(pkt,
								 &tcp_access);
		if (!tcp_hdr) {
			ret = -ENOBUFS;
			goto error;
		}

		*port = tcp_hdr->src_port;
	} else {
		ret = -ENOTSUP;
	}

error:
	net_pkt_cursor_restore(pkt, &backup);

	return ret;
}

static inline ssize_t zsock_recv_dgram(struct net_context *ctx,
				       void *buf,
				       size_t max_len,
				       int flags,
				       struct sockaddr *src_addr,
				       socklen_t *addrlen)
{
	s32_t timeout = K_FOREVER;
	size_t recv_len = 0;
	struct net_pkt_cursor backup;
	struct net_pkt *pkt;

	if ((flags & ZSOCK_MSG_DONTWAIT) || sock_is_nonblock(ctx)) {
		timeout = K_NO_WAIT;
	}

	if (flags & ZSOCK_MSG_PEEK) {
		int res;

		res = k_fifo_wait_non_empty(&ctx->recv_q, timeout);
		/* EAGAIN when timeout expired, EINTR when cancelled */
		if (res && res != -EAGAIN && res != -EINTR) {
			errno = -res;
			return -1;
		}

		pkt = k_fifo_peek_head(&ctx->recv_q);
	} else {
		pkt = k_fifo_get(&ctx->recv_q, timeout);
	}

	if (!pkt) {
		errno = EAGAIN;
		return -1;
	}

	net_pkt_cursor_backup(pkt, &backup);

	if (src_addr && addrlen) {
		int rv;

		rv = sock_get_pkt_src_addr(pkt, net_context_get_ip_proto(ctx),
					   src_addr, *addrlen);
		if (rv < 0) {
			errno = -rv;
			return -1;
		}

		/* addrlen is a value-result argument, set to actual
		 * size of source address
		 */
		if (src_addr->sa_family == AF_INET) {
			*addrlen = sizeof(struct sockaddr_in);
		} else if (src_addr->sa_family == AF_INET6) {
			*addrlen = sizeof(struct sockaddr_in6);
		} else {
			errno = ENOTSUP;
			return -1;
		}
	}

	recv_len = net_pkt_remaining_data(pkt);
	if (recv_len > max_len) {
		recv_len = max_len;
	}

	if (net_pkt_read(pkt, buf, recv_len)) {
		errno = ENOBUFS;
		return -1;
	}

	if (!(flags & ZSOCK_MSG_PEEK)) {
		net_pkt_unref(pkt);
	} else {
		net_pkt_cursor_restore(pkt, &backup);
	}

	return recv_len;
}

static inline ssize_t zsock_recv_stream(struct net_context *ctx,
					void *buf,
					size_t max_len,
					int flags)
{
	s32_t timeout = K_FOREVER;
	size_t recv_len = 0;
	struct net_pkt_cursor backup;
	int res;

	if (!net_context_is_used(ctx)) {
		errno = EBADF;
		return -1;
	}

	if ((flags & ZSOCK_MSG_DONTWAIT) || sock_is_nonblock(ctx)) {
		timeout = K_NO_WAIT;
	}

	do {
		struct net_pkt *pkt;
		size_t data_len;

		if (sock_is_eof(ctx)) {
			return 0;
		}

		res = k_fifo_wait_non_empty(&ctx->recv_q, timeout);
		/* EAGAIN when timeout expired, EINTR when cancelled */
		if (res && res != -EAGAIN && res != -EINTR) {
			errno = -res;
			return -1;
		}

		pkt = k_fifo_peek_head(&ctx->recv_q);
		if (!pkt) {
			/* Either timeout expired, or wait was cancelled
			 * due to connection closure by peer.
			 */
			NET_DBG("NULL return from fifo");
			if (sock_is_eof(ctx)) {
				return 0;
			} else {
				errno = EAGAIN;
				return -1;
			}
		}

		net_pkt_cursor_backup(pkt, &backup);

		data_len = net_pkt_remaining_data(pkt);
		recv_len = data_len;
		if (recv_len > max_len) {
			recv_len = max_len;
		}

		/* Actually copy data to application buffer */
		if (net_pkt_read(pkt, buf, recv_len)) {
			errno = ENOBUFS;
			return -1;
		}

		if (!(flags & ZSOCK_MSG_PEEK)) {
			if (recv_len == data_len) {
				/* Finished processing head pkt in
				 * the fifo. Drop it from there.
				 */
				k_fifo_get(&ctx->recv_q, K_NO_WAIT);
				if (net_pkt_eof(pkt)) {
					sock_set_eof(ctx);
				}

				net_pkt_unref(pkt);
			}
		} else {
			net_pkt_cursor_restore(pkt, &backup);
		}
	} while (recv_len == 0);

	if (!(flags & ZSOCK_MSG_PEEK)) {
		net_context_update_recv_wnd(ctx, recv_len);
	}

	return recv_len;
}

ssize_t zsock_recvfrom_ctx(struct net_context *ctx, void *buf, size_t max_len,
			   int flags,
			   struct sockaddr *src_addr, socklen_t *addrlen)
{
	enum net_sock_type sock_type = net_context_get_type(ctx);

	if (sock_type == SOCK_DGRAM) {
		return zsock_recv_dgram(ctx, buf, max_len, flags, src_addr, addrlen);
	} else if (sock_type == SOCK_STREAM) {
		return zsock_recv_stream(ctx, buf, max_len, flags);
	} else {
		__ASSERT(0, "Unknown socket type");
	}

	return 0;
}

ssize_t z_impl_zsock_recvfrom(int sock, void *buf, size_t max_len, int flags,
			     struct sockaddr *src_addr, socklen_t *addrlen)
{
	VTABLE_CALL(recvfrom, sock, buf, max_len, flags, src_addr, addrlen);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_recvfrom, sock, buf, max_len, flags, src_addr,
		  addrlen_param)
{
	socklen_t addrlen_copy;
	socklen_t *addrlen_ptr = (socklen_t *)addrlen_param;
	ssize_t ret;

	if (Z_SYSCALL_MEMORY_WRITE(buf, max_len)) {
		errno = EFAULT;
		return -1;
	}

	if (addrlen_param) {
		Z_OOPS(z_user_from_copy(&addrlen_copy,
					(socklen_t *)addrlen_param,
					sizeof(socklen_t)));
	}
	Z_OOPS(src_addr && Z_SYSCALL_MEMORY_WRITE(src_addr, addrlen_copy));

	ret = z_impl_zsock_recvfrom(sock, (void *)buf, max_len, flags,
				   (struct sockaddr *)src_addr,
				   addrlen_param ? &addrlen_copy : NULL);

	if (addrlen_param) {
		Z_OOPS(z_user_to_copy(addrlen_ptr, &addrlen_copy,
				      sizeof(socklen_t)));
	}

	return ret;
}
#endif /* CONFIG_USERSPACE */

/* As this is limited function, we don't follow POSIX signature, with
 * "..." instead of last arg.
 */
int z_impl_zsock_fcntl(int sock, int cmd, int flags)
{
	const struct fd_op_vtable *vtable;
	void *obj;

	obj = z_get_fd_obj_and_vtable(sock, &vtable);
	if (obj == NULL) {
		return -1;
	}

	return z_fdtable_call_ioctl(vtable, obj, cmd, flags);
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_fcntl, sock, cmd, flags)
{
	return z_impl_zsock_fcntl(sock, cmd, flags);
}
#endif

static int zsock_poll_prepare_ctx(struct net_context *ctx,
				  struct zsock_pollfd *pfd,
				  struct k_poll_event **pev,
				  struct k_poll_event *pev_end)
{
	if (pfd->events & ZSOCK_POLLIN) {
		if (*pev == pev_end) {
			errno = ENOMEM;
			return -1;
		}

		(*pev)->obj = &ctx->recv_q;
		(*pev)->type = K_POLL_TYPE_FIFO_DATA_AVAILABLE;
		(*pev)->mode = K_POLL_MODE_NOTIFY_ONLY;
		(*pev)->state = K_POLL_STATE_NOT_READY;
		(*pev)++;
	}

	/* If socket is already in EOF, it can be reported
	 * immediately, so we tell poll() to short-circuit wait.
	 */
	if (sock_is_eof(ctx)) {
		errno = EALREADY;
		return -1;
	}

	return 0;
}

static int zsock_poll_update_ctx(struct net_context *ctx,
				 struct zsock_pollfd *pfd,
				 struct k_poll_event **pev)
{
	ARG_UNUSED(ctx);

	/* For now, assume that socket is always writable */
	if (pfd->events & ZSOCK_POLLOUT) {
		pfd->revents |= ZSOCK_POLLOUT;
	}

	if (pfd->events & ZSOCK_POLLIN) {
		if ((*pev)->state != K_POLL_STATE_NOT_READY || sock_is_eof(ctx)) {
			pfd->revents |= ZSOCK_POLLIN;
		}
		(*pev)++;
	}

	return 0;
}

static inline int time_left(u32_t start, u32_t timeout)
{
	u32_t elapsed = k_uptime_get_32() - start;

	return timeout - elapsed;
}

int z_impl_zsock_poll(struct zsock_pollfd *fds, int nfds, int timeout)
{
	bool retry;
	int ret = 0;
	int i, remaining_time;
	struct zsock_pollfd *pfd;
	struct k_poll_event poll_events[CONFIG_NET_SOCKETS_POLL_MAX];
	struct k_poll_event *pev;
	struct k_poll_event *pev_end = poll_events + ARRAY_SIZE(poll_events);
	const struct fd_op_vtable *vtable;
	u32_t entry_time = k_uptime_get_32();

	if (timeout < 0) {
		timeout = K_FOREVER;
	}

	pev = poll_events;
	for (pfd = fds, i = nfds; i--; pfd++) {
		struct net_context *ctx;

		/* Per POSIX, negative fd's are just ignored */
		if (pfd->fd < 0) {
			continue;
		}

		ctx = z_get_fd_obj_and_vtable(pfd->fd, &vtable);
		if (ctx == NULL) {
			/* Will set POLLNVAL in return loop */
			continue;
		}

		if (z_fdtable_call_ioctl(vtable, ctx, ZFD_IOCTL_POLL_PREPARE,
					 pfd, &pev, pev_end) < 0) {
			/* If POLL_PREPARE returned with EALREADY, it means
			 * it already detected that some socket is ready. In
			 * this case, we still perform a k_poll to pick up
			 * as many events as possible, but without any wait.
			 * TODO: optimize, use ret value, instead of setting
			 * errno.
			 */
			if (errno == EALREADY) {
				timeout = K_NO_WAIT;
				continue;
			}

			return -1;
		}
	}

	remaining_time = timeout;

	do {
		ret = k_poll(poll_events, pev - poll_events, remaining_time);
		/* EAGAIN when timeout expired, EINTR when cancelled (i.e. EOF) */
		if (ret != 0 && ret != -EAGAIN && ret != -EINTR) {
			errno = -ret;
			return -1;
		}

		retry = false;
		ret = 0;

		pev = poll_events;
		for (pfd = fds, i = nfds; i--; pfd++) {
			struct net_context *ctx;

			pfd->revents = 0;

			if (pfd->fd < 0) {
				continue;
			}

			ctx = z_get_fd_obj_and_vtable(pfd->fd, &vtable);
			if (ctx == NULL) {
				pfd->revents = ZSOCK_POLLNVAL;
				ret++;
				continue;
			}

			if (z_fdtable_call_ioctl(vtable, ctx, ZFD_IOCTL_POLL_UPDATE,
						 pfd, &pev) < 0) {
				if (errno == EAGAIN) {
					retry = true;
					continue;
				}

				return -1;
			}

			if (pfd->revents != 0) {
				ret++;
			}
		}

		if (retry) {
			if (ret > 0) {
				break;
			}

			if (timeout == K_NO_WAIT) {
				break;
			}

			if (timeout != K_FOREVER) {
				/* Recalculate the timeout value. */
				remaining_time = time_left(entry_time, timeout);
				if (remaining_time <= 0) {
					break;
				}
			}
		}
	} while (retry);

	return ret;
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_poll, fds, nfds, timeout)
{
	struct zsock_pollfd *fds_copy;
	unsigned int fds_size;
	int ret;

	/* Copy fds array from user mode */
	if (__builtin_umul_overflow(nfds, sizeof(struct zsock_pollfd),
				    &fds_size)) {
		errno = EFAULT;
		return -1;
	}
	fds_copy = z_user_alloc_from_copy((void *)fds, fds_size);
	if (!fds_copy) {
		errno = ENOMEM;
		return -1;
	}

	ret = z_impl_zsock_poll(fds_copy, nfds, timeout);

	if (ret >= 0) {
		z_user_to_copy((void *)fds, fds_copy, fds_size);
	}
	k_free(fds_copy);

	return ret;
}
#endif

int z_impl_zsock_inet_pton(sa_family_t family, const char *src, void *dst)
{
	if (net_addr_pton(family, src, dst) == 0) {
		return 1;
	} else {
		return 0;
	}
}

#ifdef CONFIG_USERSPACE
Z_SYSCALL_HANDLER(zsock_inet_pton, family, src, dst)
{
	int dst_size;
	char src_copy[NET_IPV6_ADDR_LEN];
	char dst_copy[sizeof(struct in6_addr)];
	int ret;

	switch (family) {
	case AF_INET:
		dst_size = sizeof(struct in_addr);
		break;
	case AF_INET6:
		dst_size = sizeof(struct in6_addr);
		break;
	default:
		errno = EAFNOSUPPORT;
		return -1;
	}

	Z_OOPS(z_user_string_copy(src_copy, (char *)src, sizeof(src_copy)));
	ret = z_impl_zsock_inet_pton(family, src_copy, dst_copy);
	Z_OOPS(z_user_to_copy((void *)dst, dst_copy, dst_size));

	return ret;
}
#endif

int zsock_getsockopt_ctx(struct net_context *ctx, int level, int optname,
			 void *optval, socklen_t *optlen)
{
	errno = ENOPROTOOPT;
	return -1;
}

int zsock_getsockopt(int sock, int level, int optname,
		     void *optval, socklen_t *optlen)
{
	VTABLE_CALL(getsockopt, sock, level, optname, optval, optlen);
}

int zsock_setsockopt_ctx(struct net_context *ctx, int level, int optname,
			 const void *optval, socklen_t optlen)
{
	switch (level) {
	case SOL_SOCKET:
		switch (optname) {
		case SO_REUSEADDR:
			/* Ignore for now. Provided to let port
			 * existing apps.
			 */
			return 0;
		}
		break;

	case IPPROTO_TCP:
		switch (optname) {
		case TCP_NODELAY:
			/* Ignore for now. Provided to let port
			 * existing apps.
			 */
			return 0;
		}
		break;

	case IPPROTO_IPV6:
		switch (optname) {
		case IPV6_V6ONLY:
			/* Ignore for now. Provided to let port
			 * existing apps.
			 */
			return 0;
		}
		break;
	}

	errno = ENOPROTOOPT;
	return -1;
}

int zsock_setsockopt(int sock, int level, int optname,
		     const void *optval, socklen_t optlen)
{
	VTABLE_CALL(setsockopt, sock, level, optname, optval, optlen);
}

static ssize_t sock_read_vmeth(void *obj, void *buffer, size_t count)
{
	return zsock_recvfrom_ctx(obj, buffer, count, 0, NULL, 0);
}

static ssize_t sock_write_vmeth(void *obj, const void *buffer, size_t count)
{
	return zsock_sendto_ctx(obj, buffer, count, 0, NULL, 0);
}

static int sock_ioctl_vmeth(void *obj, unsigned int request, va_list args)
{
	switch (request) {

	/* In Zephyr, fcntl() is just an alias of ioctl(). */
	case F_GETFL:
		if (sock_is_nonblock(obj)) {
		    return O_NONBLOCK;
		}

		return 0;

	case F_SETFL: {
		int flags;

		flags = va_arg(args, int);

		if (flags & O_NONBLOCK) {
			sock_set_flag(obj, SOCK_NONBLOCK, SOCK_NONBLOCK);
		} else {
			sock_set_flag(obj, SOCK_NONBLOCK, 0);
		}

		return 0;
	}

	case ZFD_IOCTL_CLOSE:
		return zsock_close_ctx(obj);

	case ZFD_IOCTL_POLL_PREPARE: {
		struct zsock_pollfd *pfd;
		struct k_poll_event **pev;
		struct k_poll_event *pev_end;

		pfd = va_arg(args, struct zsock_pollfd *);
		pev = va_arg(args, struct k_poll_event **);
		pev_end = va_arg(args, struct k_poll_event *);

		return zsock_poll_prepare_ctx(obj, pfd, pev, pev_end);
	}

	case ZFD_IOCTL_POLL_UPDATE: {
		struct zsock_pollfd *pfd;
		struct k_poll_event **pev;

		pfd = va_arg(args, struct zsock_pollfd *);
		pev = va_arg(args, struct k_poll_event **);

		return zsock_poll_update_ctx(obj, pfd, pev);
	}

	default:
		errno = EOPNOTSUPP;
		return -1;
	}
}

static int sock_bind_vmeth(void *obj, const struct sockaddr *addr,
			   socklen_t addrlen)
{
	return zsock_bind_ctx(obj, addr, addrlen);
}

static int sock_connect_vmeth(void *obj, const struct sockaddr *addr,
			      socklen_t addrlen)
{
	return zsock_connect_ctx(obj, addr, addrlen);
}

static int sock_listen_vmeth(void *obj, int backlog)
{
	return zsock_listen_ctx(obj, backlog);
}

static int sock_accept_vmeth(void *obj, struct sockaddr *addr,
			     socklen_t *addrlen)
{
	return zsock_accept_ctx(obj, addr, addrlen);
}

static ssize_t sock_sendto_vmeth(void *obj, const void *buf, size_t len,
				 int flags, const struct sockaddr *dest_addr,
				 socklen_t addrlen)
{
	return zsock_sendto_ctx(obj, buf, len, flags, dest_addr, addrlen);
}

static ssize_t sock_recvfrom_vmeth(void *obj, void *buf, size_t max_len,
				   int flags, struct sockaddr *src_addr,
				   socklen_t *addrlen)
{
	return zsock_recvfrom_ctx(obj, buf, max_len, flags,
				  src_addr, addrlen);
}

static int sock_getsockopt_vmeth(void *obj, int level, int optname,
				 void *optval, socklen_t *optlen)
{
	return zsock_getsockopt_ctx(obj, level, optname, optval, optlen);
}

static int sock_setsockopt_vmeth(void *obj, int level, int optname,
				 const void *optval, socklen_t optlen)
{
	return zsock_setsockopt_ctx(obj, level, optname, optval, optlen);
}


const struct socket_op_vtable sock_fd_op_vtable = {
	.fd_vtable = {
		.read = sock_read_vmeth,
		.write = sock_write_vmeth,
		.ioctl = sock_ioctl_vmeth,
	},
	.bind = sock_bind_vmeth,
	.connect = sock_connect_vmeth,
	.listen = sock_listen_vmeth,
	.accept = sock_accept_vmeth,
	.sendto = sock_sendto_vmeth,
	.recvfrom = sock_recvfrom_vmeth,
	.getsockopt = sock_getsockopt_vmeth,
	.setsockopt = sock_setsockopt_vmeth,
};