flow_dissector.c 37.8 KB
Newer Older
1
#include <linux/kernel.h>
E
Eric Dumazet 已提交
2
#include <linux/skbuff.h>
3
#include <linux/export.h>
E
Eric Dumazet 已提交
4 5 6
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/if_vlan.h>
7
#include <net/dsa.h>
S
Simon Horman 已提交
8
#include <net/dst_metadata.h>
E
Eric Dumazet 已提交
9
#include <net/ip.h>
E
Eric Dumazet 已提交
10
#include <net/ipv6.h>
11 12
#include <net/gre.h>
#include <net/pptp.h>
13
#include <net/tipc.h>
14 15 16 17
#include <linux/igmp.h>
#include <linux/icmp.h>
#include <linux/sctp.h>
#include <linux/dccp.h>
E
Eric Dumazet 已提交
18 19 20
#include <linux/if_tunnel.h>
#include <linux/if_pppox.h>
#include <linux/ppp_defs.h>
21
#include <linux/stddef.h>
22
#include <linux/if_ether.h>
23
#include <linux/mpls.h>
24
#include <linux/tcp.h>
25
#include <net/flow_dissector.h>
26
#include <scsi/fc/fc_fcoe.h>
27
#include <uapi/linux/batadv_packet.h>
E
Eric Dumazet 已提交
28

29 30
static void dissector_set_key(struct flow_dissector *flow_dissector,
			      enum flow_dissector_key_id key_id)
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
{
	flow_dissector->used_keys |= (1 << key_id);
}

void skb_flow_dissector_init(struct flow_dissector *flow_dissector,
			     const struct flow_dissector_key *key,
			     unsigned int key_count)
{
	unsigned int i;

	memset(flow_dissector, 0, sizeof(*flow_dissector));

	for (i = 0; i < key_count; i++, key++) {
		/* User should make sure that every key target offset is withing
		 * boundaries of unsigned short.
		 */
		BUG_ON(key->offset > USHRT_MAX);
48 49
		BUG_ON(dissector_uses_key(flow_dissector,
					  key->key_id));
50

51
		dissector_set_key(flow_dissector, key->key_id);
52 53 54
		flow_dissector->offset[key->key_id] = key->offset;
	}

55 56
	/* Ensure that the dissector always includes control and basic key.
	 * That way we are able to avoid handling lack of these in fast path.
57
	 */
58 59 60 61
	BUG_ON(!dissector_uses_key(flow_dissector,
				   FLOW_DISSECTOR_KEY_CONTROL));
	BUG_ON(!dissector_uses_key(flow_dissector,
				   FLOW_DISSECTOR_KEY_BASIC));
62 63 64
}
EXPORT_SYMBOL(skb_flow_dissector_init);

S
Simon Horman 已提交
65 66 67 68 69 70 71 72 73 74
/**
 * skb_flow_get_be16 - extract be16 entity
 * @skb: sk_buff to extract from
 * @poff: offset to extract at
 * @data: raw buffer pointer to the packet
 * @hlen: packet header length
 *
 * The function will try to retrieve a be32 entity at
 * offset poff
 */
75 76
static __be16 skb_flow_get_be16(const struct sk_buff *skb, int poff,
				void *data, int hlen)
S
Simon Horman 已提交
77 78 79 80 81 82 83 84 85 86
{
	__be16 *u, _u;

	u = __skb_header_pointer(skb, poff, sizeof(_u), data, hlen, &_u);
	if (u)
		return *u;

	return 0;
}

87
/**
88 89
 * __skb_flow_get_ports - extract the upper layer ports and return them
 * @skb: sk_buff to extract the ports from
90 91
 * @thoff: transport header offset
 * @ip_proto: protocol for which to get port offset
92 93
 * @data: raw buffer pointer to the packet, if NULL use skb->data
 * @hlen: packet header length, if @data is NULL use skb_headlen(skb)
94 95 96 97
 *
 * The function will try to retrieve the ports at offset thoff + poff where poff
 * is the protocol port offset returned from proto_ports_offset
 */
98 99
__be32 __skb_flow_get_ports(const struct sk_buff *skb, int thoff, u8 ip_proto,
			    void *data, int hlen)
100 101 102
{
	int poff = proto_ports_offset(ip_proto);

103 104 105 106 107
	if (!data) {
		data = skb->data;
		hlen = skb_headlen(skb);
	}

108 109 110
	if (poff >= 0) {
		__be32 *ports, _ports;

111 112
		ports = __skb_header_pointer(skb, thoff + poff,
					     sizeof(_ports), data, hlen, &_ports);
113 114 115 116 117 118
		if (ports)
			return *ports;
	}

	return 0;
}
119
EXPORT_SYMBOL(__skb_flow_get_ports);
120

S
Simon Horman 已提交
121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
static void
skb_flow_dissect_set_enc_addr_type(enum flow_dissector_key_id type,
				   struct flow_dissector *flow_dissector,
				   void *target_container)
{
	struct flow_dissector_key_control *ctrl;

	if (!dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_CONTROL))
		return;

	ctrl = skb_flow_dissector_target(flow_dissector,
					 FLOW_DISSECTOR_KEY_ENC_CONTROL,
					 target_container);
	ctrl->addr_type = type;
}

137 138 139 140
void
skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
			     struct flow_dissector *flow_dissector,
			     void *target_container)
S
Simon Horman 已提交
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
{
	struct ip_tunnel_info *info;
	struct ip_tunnel_key *key;

	/* A quick check to see if there might be something to do. */
	if (!dissector_uses_key(flow_dissector,
				FLOW_DISSECTOR_KEY_ENC_KEYID) &&
	    !dissector_uses_key(flow_dissector,
				FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS) &&
	    !dissector_uses_key(flow_dissector,
				FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS) &&
	    !dissector_uses_key(flow_dissector,
				FLOW_DISSECTOR_KEY_ENC_CONTROL) &&
	    !dissector_uses_key(flow_dissector,
				FLOW_DISSECTOR_KEY_ENC_PORTS))
		return;

	info = skb_tunnel_info(skb);
	if (!info)
		return;

	key = &info->key;

	switch (ip_tunnel_info_af(info)) {
	case AF_INET:
		skb_flow_dissect_set_enc_addr_type(FLOW_DISSECTOR_KEY_IPV4_ADDRS,
						   flow_dissector,
						   target_container);
		if (dissector_uses_key(flow_dissector,
				       FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS)) {
			struct flow_dissector_key_ipv4_addrs *ipv4;

			ipv4 = skb_flow_dissector_target(flow_dissector,
							 FLOW_DISSECTOR_KEY_ENC_IPV4_ADDRS,
							 target_container);
			ipv4->src = key->u.ipv4.src;
			ipv4->dst = key->u.ipv4.dst;
		}
		break;
	case AF_INET6:
		skb_flow_dissect_set_enc_addr_type(FLOW_DISSECTOR_KEY_IPV6_ADDRS,
						   flow_dissector,
						   target_container);
		if (dissector_uses_key(flow_dissector,
				       FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS)) {
			struct flow_dissector_key_ipv6_addrs *ipv6;

			ipv6 = skb_flow_dissector_target(flow_dissector,
							 FLOW_DISSECTOR_KEY_ENC_IPV6_ADDRS,
							 target_container);
			ipv6->src = key->u.ipv6.src;
			ipv6->dst = key->u.ipv6.dst;
		}
		break;
	}

	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_KEYID)) {
		struct flow_dissector_key_keyid *keyid;

		keyid = skb_flow_dissector_target(flow_dissector,
						  FLOW_DISSECTOR_KEY_ENC_KEYID,
						  target_container);
		keyid->keyid = tunnel_id_to_key32(key->tun_id);
	}

	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ENC_PORTS)) {
		struct flow_dissector_key_ports *tp;

		tp = skb_flow_dissector_target(flow_dissector,
					       FLOW_DISSECTOR_KEY_ENC_PORTS,
					       target_container);
		tp->src = key->tp_src;
		tp->dst = key->tp_dst;
	}
}
216
EXPORT_SYMBOL(skb_flow_dissect_tunnel_info);
S
Simon Horman 已提交
217

218 219 220 221 222 223 224
static enum flow_dissect_ret
__skb_flow_dissect_mpls(const struct sk_buff *skb,
			struct flow_dissector *flow_dissector,
			void *target_container, void *data, int nhoff, int hlen)
{
	struct flow_dissector_key_keyid *key_keyid;
	struct mpls_label *hdr, _hdr[2];
225
	u32 entry, label;
226 227

	if (!dissector_uses_key(flow_dissector,
228 229
				FLOW_DISSECTOR_KEY_MPLS_ENTROPY) &&
	    !dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_MPLS))
230 231 232 233 234 235 236
		return FLOW_DISSECT_RET_OUT_GOOD;

	hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data,
				   hlen, &_hdr);
	if (!hdr)
		return FLOW_DISSECT_RET_OUT_BAD;

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
	entry = ntohl(hdr[0].entry);
	label = (entry & MPLS_LS_LABEL_MASK) >> MPLS_LS_LABEL_SHIFT;

	if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_MPLS)) {
		struct flow_dissector_key_mpls *key_mpls;

		key_mpls = skb_flow_dissector_target(flow_dissector,
						     FLOW_DISSECTOR_KEY_MPLS,
						     target_container);
		key_mpls->mpls_label = label;
		key_mpls->mpls_ttl = (entry & MPLS_LS_TTL_MASK)
					>> MPLS_LS_TTL_SHIFT;
		key_mpls->mpls_tc = (entry & MPLS_LS_TC_MASK)
					>> MPLS_LS_TC_SHIFT;
		key_mpls->mpls_bos = (entry & MPLS_LS_S_MASK)
					>> MPLS_LS_S_SHIFT;
	}

	if (label == MPLS_LABEL_ENTROPY) {
256 257 258 259 260 261 262 263
		key_keyid = skb_flow_dissector_target(flow_dissector,
						      FLOW_DISSECTOR_KEY_MPLS_ENTROPY,
						      target_container);
		key_keyid->keyid = hdr[1].entry & htonl(MPLS_LS_LABEL_MASK);
	}
	return FLOW_DISSECT_RET_OUT_GOOD;
}

264 265 266 267 268 269 270 271 272 273 274 275 276
static enum flow_dissect_ret
__skb_flow_dissect_arp(const struct sk_buff *skb,
		       struct flow_dissector *flow_dissector,
		       void *target_container, void *data, int nhoff, int hlen)
{
	struct flow_dissector_key_arp *key_arp;
	struct {
		unsigned char ar_sha[ETH_ALEN];
		unsigned char ar_sip[4];
		unsigned char ar_tha[ETH_ALEN];
		unsigned char ar_tip[4];
	} *arp_eth, _arp_eth;
	const struct arphdr *arp;
277
	struct arphdr _arp;
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

	if (!dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_ARP))
		return FLOW_DISSECT_RET_OUT_GOOD;

	arp = __skb_header_pointer(skb, nhoff, sizeof(_arp), data,
				   hlen, &_arp);
	if (!arp)
		return FLOW_DISSECT_RET_OUT_BAD;

	if (arp->ar_hrd != htons(ARPHRD_ETHER) ||
	    arp->ar_pro != htons(ETH_P_IP) ||
	    arp->ar_hln != ETH_ALEN ||
	    arp->ar_pln != 4 ||
	    (arp->ar_op != htons(ARPOP_REPLY) &&
	     arp->ar_op != htons(ARPOP_REQUEST)))
		return FLOW_DISSECT_RET_OUT_BAD;

	arp_eth = __skb_header_pointer(skb, nhoff + sizeof(_arp),
				       sizeof(_arp_eth), data,
				       hlen, &_arp_eth);
	if (!arp_eth)
		return FLOW_DISSECT_RET_OUT_BAD;

	key_arp = skb_flow_dissector_target(flow_dissector,
					    FLOW_DISSECTOR_KEY_ARP,
					    target_container);

	memcpy(&key_arp->sip, arp_eth->ar_sip, sizeof(key_arp->sip));
	memcpy(&key_arp->tip, arp_eth->ar_tip, sizeof(key_arp->tip));

	/* Only store the lower byte of the opcode;
	 * this covers ARPOP_REPLY and ARPOP_REQUEST.
	 */
	key_arp->op = ntohs(arp->ar_op) & 0xff;

	ether_addr_copy(key_arp->sha, arp_eth->ar_sha);
	ether_addr_copy(key_arp->tha, arp_eth->ar_tha);

	return FLOW_DISSECT_RET_OUT_GOOD;
}

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
static enum flow_dissect_ret
__skb_flow_dissect_gre(const struct sk_buff *skb,
		       struct flow_dissector_key_control *key_control,
		       struct flow_dissector *flow_dissector,
		       void *target_container, void *data,
		       __be16 *p_proto, int *p_nhoff, int *p_hlen,
		       unsigned int flags)
{
	struct flow_dissector_key_keyid *key_keyid;
	struct gre_base_hdr *hdr, _hdr;
	int offset = 0;
	u16 gre_ver;

	hdr = __skb_header_pointer(skb, *p_nhoff, sizeof(_hdr),
				   data, *p_hlen, &_hdr);
	if (!hdr)
		return FLOW_DISSECT_RET_OUT_BAD;

	/* Only look inside GRE without routing */
	if (hdr->flags & GRE_ROUTING)
		return FLOW_DISSECT_RET_OUT_GOOD;

	/* Only look inside GRE for version 0 and 1 */
	gre_ver = ntohs(hdr->flags & GRE_VERSION);
	if (gre_ver > 1)
		return FLOW_DISSECT_RET_OUT_GOOD;

	*p_proto = hdr->protocol;
	if (gre_ver) {
		/* Version1 must be PPTP, and check the flags */
		if (!(*p_proto == GRE_PROTO_PPP && (hdr->flags & GRE_KEY)))
			return FLOW_DISSECT_RET_OUT_GOOD;
	}

	offset += sizeof(struct gre_base_hdr);

	if (hdr->flags & GRE_CSUM)
		offset += sizeof(((struct gre_full_hdr *) 0)->csum) +
			  sizeof(((struct gre_full_hdr *) 0)->reserved1);

	if (hdr->flags & GRE_KEY) {
		const __be32 *keyid;
		__be32 _keyid;

		keyid = __skb_header_pointer(skb, *p_nhoff + offset,
					     sizeof(_keyid),
					     data, *p_hlen, &_keyid);
		if (!keyid)
			return FLOW_DISSECT_RET_OUT_BAD;

		if (dissector_uses_key(flow_dissector,
				       FLOW_DISSECTOR_KEY_GRE_KEYID)) {
			key_keyid = skb_flow_dissector_target(flow_dissector,
							      FLOW_DISSECTOR_KEY_GRE_KEYID,
							      target_container);
			if (gre_ver == 0)
				key_keyid->keyid = *keyid;
			else
				key_keyid->keyid = *keyid & GRE_PPTP_KEY_MASK;
		}
		offset += sizeof(((struct gre_full_hdr *) 0)->key);
	}

	if (hdr->flags & GRE_SEQ)
		offset += sizeof(((struct pptp_gre_header *) 0)->seq);

	if (gre_ver == 0) {
		if (*p_proto == htons(ETH_P_TEB)) {
			const struct ethhdr *eth;
			struct ethhdr _eth;

			eth = __skb_header_pointer(skb, *p_nhoff + offset,
						   sizeof(_eth),
						   data, *p_hlen, &_eth);
			if (!eth)
				return FLOW_DISSECT_RET_OUT_BAD;
			*p_proto = eth->h_proto;
			offset += sizeof(*eth);

			/* Cap headers that we access via pointers at the
			 * end of the Ethernet header as our maximum alignment
			 * at that point is only 2 bytes.
			 */
			if (NET_IP_ALIGN)
				*p_hlen = *p_nhoff + offset;
		}
	} else { /* version 1, must be PPTP */
		u8 _ppp_hdr[PPP_HDRLEN];
		u8 *ppp_hdr;

		if (hdr->flags & GRE_ACK)
			offset += sizeof(((struct pptp_gre_header *) 0)->ack);

		ppp_hdr = __skb_header_pointer(skb, *p_nhoff + offset,
					       sizeof(_ppp_hdr),
					       data, *p_hlen, _ppp_hdr);
		if (!ppp_hdr)
			return FLOW_DISSECT_RET_OUT_BAD;

		switch (PPP_PROTOCOL(ppp_hdr)) {
		case PPP_IP:
			*p_proto = htons(ETH_P_IP);
			break;
		case PPP_IPV6:
			*p_proto = htons(ETH_P_IPV6);
			break;
		default:
			/* Could probably catch some more like MPLS */
			break;
		}

		offset += PPP_HDRLEN;
	}

	*p_nhoff += offset;
	key_control->flags |= FLOW_DIS_ENCAPSULATION;
	if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP)
		return FLOW_DISSECT_RET_OUT_GOOD;

438
	return FLOW_DISSECT_RET_PROTO_AGAIN;
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
/**
 * __skb_flow_dissect_batadv() - dissect batman-adv header
 * @skb: sk_buff to with the batman-adv header
 * @key_control: flow dissectors control key
 * @data: raw buffer pointer to the packet, if NULL use skb->data
 * @p_proto: pointer used to update the protocol to process next
 * @p_nhoff: pointer used to update inner network header offset
 * @hlen: packet header length
 * @flags: any combination of FLOW_DISSECTOR_F_*
 *
 * ETH_P_BATMAN packets are tried to be dissected. Only
 * &struct batadv_unicast packets are actually processed because they contain an
 * inner ethernet header and are usually followed by actual network header. This
 * allows the flow dissector to continue processing the packet.
 *
 * Return: FLOW_DISSECT_RET_PROTO_AGAIN when &struct batadv_unicast was found,
 *  FLOW_DISSECT_RET_OUT_GOOD when dissector should stop after encapsulation,
 *  otherwise FLOW_DISSECT_RET_OUT_BAD
 */
static enum flow_dissect_ret
__skb_flow_dissect_batadv(const struct sk_buff *skb,
			  struct flow_dissector_key_control *key_control,
			  void *data, __be16 *p_proto, int *p_nhoff, int hlen,
			  unsigned int flags)
{
	struct {
		struct batadv_unicast_packet batadv_unicast;
		struct ethhdr eth;
	} *hdr, _hdr;

	hdr = __skb_header_pointer(skb, *p_nhoff, sizeof(_hdr), data, hlen,
				   &_hdr);
	if (!hdr)
		return FLOW_DISSECT_RET_OUT_BAD;

	if (hdr->batadv_unicast.version != BATADV_COMPAT_VERSION)
		return FLOW_DISSECT_RET_OUT_BAD;

	if (hdr->batadv_unicast.packet_type != BATADV_UNICAST)
		return FLOW_DISSECT_RET_OUT_BAD;

	*p_proto = hdr->eth.h_proto;
	*p_nhoff += sizeof(*hdr);

	key_control->flags |= FLOW_DIS_ENCAPSULATION;
	if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP)
		return FLOW_DISSECT_RET_OUT_GOOD;

	return FLOW_DISSECT_RET_PROTO_AGAIN;
}

492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
static void
__skb_flow_dissect_tcp(const struct sk_buff *skb,
		       struct flow_dissector *flow_dissector,
		       void *target_container, void *data, int thoff, int hlen)
{
	struct flow_dissector_key_tcp *key_tcp;
	struct tcphdr *th, _th;

	if (!dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_TCP))
		return;

	th = __skb_header_pointer(skb, thoff, sizeof(_th), data, hlen, &_th);
	if (!th)
		return;

	if (unlikely(__tcp_hdrlen(th) < sizeof(_th)))
		return;

	key_tcp = skb_flow_dissector_target(flow_dissector,
					    FLOW_DISSECTOR_KEY_TCP,
					    target_container);
	key_tcp->flags = (*(__be16 *) &tcp_flag_word(th) & htons(0x0FFF));
}

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
static void
__skb_flow_dissect_ipv4(const struct sk_buff *skb,
			struct flow_dissector *flow_dissector,
			void *target_container, void *data, const struct iphdr *iph)
{
	struct flow_dissector_key_ip *key_ip;

	if (!dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_IP))
		return;

	key_ip = skb_flow_dissector_target(flow_dissector,
					   FLOW_DISSECTOR_KEY_IP,
					   target_container);
	key_ip->tos = iph->tos;
	key_ip->ttl = iph->ttl;
}

static void
__skb_flow_dissect_ipv6(const struct sk_buff *skb,
			struct flow_dissector *flow_dissector,
			void *target_container, void *data, const struct ipv6hdr *iph)
{
	struct flow_dissector_key_ip *key_ip;

	if (!dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_IP))
		return;

	key_ip = skb_flow_dissector_target(flow_dissector,
					   FLOW_DISSECTOR_KEY_IP,
					   target_container);
	key_ip->tos = ipv6_get_dsfield(iph);
	key_ip->ttl = iph->hop_limit;
}

550 551 552 553 554 555 556 557 558 559 560 561
/* Maximum number of protocol headers that can be parsed in
 * __skb_flow_dissect
 */
#define MAX_FLOW_DISSECT_HDRS	15

static bool skb_flow_dissect_allowed(int *num_hdrs)
{
	++*num_hdrs;

	return (*num_hdrs <= MAX_FLOW_DISSECT_HDRS);
}

562 563 564
/**
 * __skb_flow_dissect - extract the flow_keys struct and return it
 * @skb: sk_buff to extract the flow from, can be NULL if the rest are specified
565 566
 * @flow_dissector: list of keys to dissect
 * @target_container: target structure to put dissected values into
567 568 569 570 571
 * @data: raw buffer pointer to the packet, if NULL use skb->data
 * @proto: protocol for which to get the flow, if @data is NULL use skb->protocol
 * @nhoff: network header offset, if @data is NULL use skb_network_offset(skb)
 * @hlen: packet header length, if @data is NULL use skb_headlen(skb)
 *
572 573 574 575 576
 * The function will try to retrieve individual keys into target specified
 * by flow_dissector from either the skbuff or a raw buffer specified by the
 * rest parameters.
 *
 * Caller must take care of zeroing target container memory.
577
 */
578 579 580
bool __skb_flow_dissect(const struct sk_buff *skb,
			struct flow_dissector *flow_dissector,
			void *target_container,
581 582
			void *data, __be16 proto, int nhoff, int hlen,
			unsigned int flags)
E
Eric Dumazet 已提交
583
{
584
	struct flow_dissector_key_control *key_control;
585 586 587
	struct flow_dissector_key_basic *key_basic;
	struct flow_dissector_key_addrs *key_addrs;
	struct flow_dissector_key_ports *key_ports;
S
Simon Horman 已提交
588
	struct flow_dissector_key_icmp *key_icmp;
T
Tom Herbert 已提交
589
	struct flow_dissector_key_tags *key_tags;
590
	struct flow_dissector_key_vlan *key_vlan;
591
	enum flow_dissect_ret fdret;
592
	bool skip_vlan = false;
593
	int num_hdrs = 0;
594
	u8 ip_proto = 0;
595
	bool ret;
E
Eric Dumazet 已提交
596

597 598
	if (!data) {
		data = skb->data;
599 600
		proto = skb_vlan_tag_present(skb) ?
			 skb->vlan_proto : skb->protocol;
601
		nhoff = skb_network_offset(skb);
602
		hlen = skb_headlen(skb);
603
#if IS_ENABLED(CONFIG_NET_DSA)
604
		if (unlikely(skb->dev && netdev_uses_dsa(skb->dev))) {
605 606 607 608 609 610 611 612 613 614
			const struct dsa_device_ops *ops;
			int offset;

			ops = skb->dev->dsa_ptr->tag_ops;
			if (ops->flow_dissect &&
			    !ops->flow_dissect(skb, &proto, &offset)) {
				hlen -= offset;
				nhoff += offset;
			}
		}
615
#endif
616 617
	}

618 619 620 621 622 623 624
	/* It is ensured by skb_flow_dissector_init() that control key will
	 * be always present.
	 */
	key_control = skb_flow_dissector_target(flow_dissector,
						FLOW_DISSECTOR_KEY_CONTROL,
						target_container);

625 626 627 628 629 630
	/* It is ensured by skb_flow_dissector_init() that basic key will
	 * be always present.
	 */
	key_basic = skb_flow_dissector_target(flow_dissector,
					      FLOW_DISSECTOR_KEY_BASIC,
					      target_container);
E
Eric Dumazet 已提交
631

632 633
	if (dissector_uses_key(flow_dissector,
			       FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
634 635 636 637 638 639 640 641 642
		struct ethhdr *eth = eth_hdr(skb);
		struct flow_dissector_key_eth_addrs *key_eth_addrs;

		key_eth_addrs = skb_flow_dissector_target(flow_dissector,
							  FLOW_DISSECTOR_KEY_ETH_ADDRS,
							  target_container);
		memcpy(key_eth_addrs, &eth->h_dest, sizeof(*key_eth_addrs));
	}

643
proto_again:
644 645
	fdret = FLOW_DISSECT_RET_CONTINUE;

E
Eric Dumazet 已提交
646
	switch (proto) {
647
	case htons(ETH_P_IP): {
E
Eric Dumazet 已提交
648 649
		const struct iphdr *iph;
		struct iphdr _iph;
650

651
		iph = __skb_header_pointer(skb, nhoff, sizeof(_iph), data, hlen, &_iph);
652 653 654 655 656
		if (!iph || iph->ihl < 5) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}

657
		nhoff += iph->ihl * 4;
E
Eric Dumazet 已提交
658

659 660
		ip_proto = iph->protocol;

661 662 663 664 665 666 667 668 669 670
		if (dissector_uses_key(flow_dissector,
				       FLOW_DISSECTOR_KEY_IPV4_ADDRS)) {
			key_addrs = skb_flow_dissector_target(flow_dissector,
							      FLOW_DISSECTOR_KEY_IPV4_ADDRS,
							      target_container);

			memcpy(&key_addrs->v4addrs, &iph->saddr,
			       sizeof(key_addrs->v4addrs));
			key_control->addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS;
		}
671 672

		if (ip_is_fragment(iph)) {
673
			key_control->flags |= FLOW_DIS_IS_FRAGMENT;
674 675

			if (iph->frag_off & htons(IP_OFFSET)) {
676 677
				fdret = FLOW_DISSECT_RET_OUT_GOOD;
				break;
678
			} else {
679
				key_control->flags |= FLOW_DIS_FIRST_FRAG;
680 681 682 683 684
				if (!(flags &
				      FLOW_DISSECTOR_F_PARSE_1ST_FRAG)) {
					fdret = FLOW_DISSECT_RET_OUT_GOOD;
					break;
				}
685 686 687
			}
		}

688 689 690
		__skb_flow_dissect_ipv4(skb, flow_dissector,
					target_container, data, iph);

691 692 693 694
		if (flags & FLOW_DISSECTOR_F_STOP_AT_L3) {
			fdret = FLOW_DISSECT_RET_OUT_GOOD;
			break;
		}
695

E
Eric Dumazet 已提交
696 697
		break;
	}
698
	case htons(ETH_P_IPV6): {
E
Eric Dumazet 已提交
699 700
		const struct ipv6hdr *iph;
		struct ipv6hdr _iph;
701

702
		iph = __skb_header_pointer(skb, nhoff, sizeof(_iph), data, hlen, &_iph);
703 704 705 706
		if (!iph) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}
E
Eric Dumazet 已提交
707 708 709

		ip_proto = iph->nexthdr;
		nhoff += sizeof(struct ipv6hdr);
710

711 712
		if (dissector_uses_key(flow_dissector,
				       FLOW_DISSECTOR_KEY_IPV6_ADDRS)) {
713 714 715
			key_addrs = skb_flow_dissector_target(flow_dissector,
							      FLOW_DISSECTOR_KEY_IPV6_ADDRS,
							      target_container);
716

717 718
			memcpy(&key_addrs->v6addrs, &iph->saddr,
			       sizeof(key_addrs->v6addrs));
719
			key_control->addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
720
		}
721

722 723 724 725 726 727
		if ((dissector_uses_key(flow_dissector,
					FLOW_DISSECTOR_KEY_FLOW_LABEL) ||
		     (flags & FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL)) &&
		    ip6_flowlabel(iph)) {
			__be32 flow_label = ip6_flowlabel(iph);

728 729
			if (dissector_uses_key(flow_dissector,
					       FLOW_DISSECTOR_KEY_FLOW_LABEL)) {
730 731 732 733
				key_tags = skb_flow_dissector_target(flow_dissector,
								     FLOW_DISSECTOR_KEY_FLOW_LABEL,
								     target_container);
				key_tags->flow_label = ntohl(flow_label);
734
			}
735 736 737 738
			if (flags & FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL) {
				fdret = FLOW_DISSECT_RET_OUT_GOOD;
				break;
			}
739 740
		}

741 742 743
		__skb_flow_dissect_ipv6(skb, flow_dissector,
					target_container, data, iph);

744
		if (flags & FLOW_DISSECTOR_F_STOP_AT_L3)
745
			fdret = FLOW_DISSECT_RET_OUT_GOOD;
746

E
Eric Dumazet 已提交
747 748
		break;
	}
749 750
	case htons(ETH_P_8021AD):
	case htons(ETH_P_8021Q): {
E
Eric Dumazet 已提交
751
		const struct vlan_hdr *vlan;
752 753
		struct vlan_hdr _vlan;
		bool vlan_tag_present = skb && skb_vlan_tag_present(skb);
E
Eric Dumazet 已提交
754

755
		if (vlan_tag_present)
756 757
			proto = skb->protocol;

758
		if (!vlan_tag_present || eth_type_vlan(skb->protocol)) {
759 760
			vlan = __skb_header_pointer(skb, nhoff, sizeof(_vlan),
						    data, hlen, &_vlan);
761 762 763 764 765
			if (!vlan) {
				fdret = FLOW_DISSECT_RET_OUT_BAD;
				break;
			}

766 767
			proto = vlan->h_vlan_encapsulated_proto;
			nhoff += sizeof(*vlan);
768 769 770 771
			if (skip_vlan) {
				fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
				break;
			}
772
		}
E
Eric Dumazet 已提交
773

774
		skip_vlan = true;
775
		if (dissector_uses_key(flow_dissector,
776 777 778
				       FLOW_DISSECTOR_KEY_VLAN)) {
			key_vlan = skb_flow_dissector_target(flow_dissector,
							     FLOW_DISSECTOR_KEY_VLAN,
T
Tom Herbert 已提交
779 780
							     target_container);

781
			if (vlan_tag_present) {
782 783 784 785 786
				key_vlan->vlan_id = skb_vlan_tag_get_id(skb);
				key_vlan->vlan_priority =
					(skb_vlan_tag_get_prio(skb) >> VLAN_PRIO_SHIFT);
			} else {
				key_vlan->vlan_id = ntohs(vlan->h_vlan_TCI) &
787
					VLAN_VID_MASK;
788 789 790 791
				key_vlan->vlan_priority =
					(ntohs(vlan->h_vlan_TCI) &
					 VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT;
			}
T
Tom Herbert 已提交
792 793
		}

794 795
		fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
		break;
E
Eric Dumazet 已提交
796
	}
797
	case htons(ETH_P_PPP_SES): {
E
Eric Dumazet 已提交
798 799 800 801
		struct {
			struct pppoe_hdr hdr;
			__be16 proto;
		} *hdr, _hdr;
802
		hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr), data, hlen, &_hdr);
803 804 805 806 807
		if (!hdr) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}

E
Eric Dumazet 已提交
808 809 810
		proto = hdr->proto;
		nhoff += PPPOE_SES_HLEN;
		switch (proto) {
811
		case htons(PPP_IP):
812 813 814
			proto = htons(ETH_P_IP);
			fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
			break;
815
		case htons(PPP_IPV6):
816 817 818
			proto = htons(ETH_P_IPV6);
			fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
			break;
E
Eric Dumazet 已提交
819
		default:
820 821
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
E
Eric Dumazet 已提交
822
		}
823
		break;
E
Eric Dumazet 已提交
824
	}
E
Erik Hugne 已提交
825
	case htons(ETH_P_TIPC): {
826 827 828 829
		struct tipc_basic_hdr *hdr, _hdr;

		hdr = __skb_header_pointer(skb, nhoff, sizeof(_hdr),
					   data, hlen, &_hdr);
830 831 832 833
		if (!hdr) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}
834

835
		if (dissector_uses_key(flow_dissector,
836
				       FLOW_DISSECTOR_KEY_TIPC)) {
837
			key_addrs = skb_flow_dissector_target(flow_dissector,
838
							      FLOW_DISSECTOR_KEY_TIPC,
839
							      target_container);
840 841
			key_addrs->tipckey.key = tipc_hdr_rps_key(hdr);
			key_control->addr_type = FLOW_DISSECTOR_KEY_TIPC;
842
		}
843 844
		fdret = FLOW_DISSECT_RET_OUT_GOOD;
		break;
E
Erik Hugne 已提交
845
	}
846 847

	case htons(ETH_P_MPLS_UC):
848
	case htons(ETH_P_MPLS_MC):
849
		fdret = __skb_flow_dissect_mpls(skb, flow_dissector,
850
						target_container, data,
851 852
						nhoff, hlen);
		break;
853
	case htons(ETH_P_FCOE):
854 855 856 857
		if ((hlen - nhoff) < FCOE_HEADER_LEN) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}
858 859

		nhoff += FCOE_HEADER_LEN;
860 861
		fdret = FLOW_DISSECT_RET_OUT_GOOD;
		break;
S
Simon Horman 已提交
862 863

	case htons(ETH_P_ARP):
864
	case htons(ETH_P_RARP):
865
		fdret = __skb_flow_dissect_arp(skb, flow_dissector,
866
					       target_container, data,
867 868 869
					       nhoff, hlen);
		break;

870 871 872 873 874
	case htons(ETH_P_BATMAN):
		fdret = __skb_flow_dissect_batadv(skb, key_control, data,
						  &proto, &nhoff, hlen, flags);
		break;

875 876 877 878 879 880 881 882 883 884
	default:
		fdret = FLOW_DISSECT_RET_OUT_BAD;
		break;
	}

	/* Process result of proto processing */
	switch (fdret) {
	case FLOW_DISSECT_RET_OUT_GOOD:
		goto out_good;
	case FLOW_DISSECT_RET_PROTO_AGAIN:
885 886 887
		if (skb_flow_dissect_allowed(&num_hdrs))
			goto proto_again;
		goto out_good;
888 889 890 891
	case FLOW_DISSECT_RET_CONTINUE:
	case FLOW_DISSECT_RET_IPPROTO_AGAIN:
		break;
	case FLOW_DISSECT_RET_OUT_BAD:
E
Eric Dumazet 已提交
892
	default:
893
		goto out_bad;
E
Eric Dumazet 已提交
894 895
	}

896
ip_proto_again:
897 898
	fdret = FLOW_DISSECT_RET_CONTINUE;

E
Eric Dumazet 已提交
899
	switch (ip_proto) {
900
	case IPPROTO_GRE:
901
		fdret = __skb_flow_dissect_gre(skb, key_control, flow_dissector,
902
					       target_container, data,
903 904 905
					       &proto, &nhoff, &hlen, flags);
		break;

906 907 908 909 910 911 912 913 914 915
	case NEXTHDR_HOP:
	case NEXTHDR_ROUTING:
	case NEXTHDR_DEST: {
		u8 _opthdr[2], *opthdr;

		if (proto != htons(ETH_P_IPV6))
			break;

		opthdr = __skb_header_pointer(skb, nhoff, sizeof(_opthdr),
					      data, hlen, &_opthdr);
916 917 918 919
		if (!opthdr) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}
920

921 922
		ip_proto = opthdr[0];
		nhoff += (opthdr[1] + 1) << 3;
923

924 925
		fdret = FLOW_DISSECT_RET_IPPROTO_AGAIN;
		break;
926
	}
927 928 929 930 931 932 933 934 935
	case NEXTHDR_FRAGMENT: {
		struct frag_hdr _fh, *fh;

		if (proto != htons(ETH_P_IPV6))
			break;

		fh = __skb_header_pointer(skb, nhoff, sizeof(_fh),
					  data, hlen, &_fh);

936 937 938 939
		if (!fh) {
			fdret = FLOW_DISSECT_RET_OUT_BAD;
			break;
		}
940

941
		key_control->flags |= FLOW_DIS_IS_FRAGMENT;
942 943

		nhoff += sizeof(_fh);
944
		ip_proto = fh->nexthdr;
945 946

		if (!(fh->frag_off & htons(IP6_OFFSET))) {
947
			key_control->flags |= FLOW_DIS_FIRST_FRAG;
948 949 950 951
			if (flags & FLOW_DISSECTOR_F_PARSE_1ST_FRAG) {
				fdret = FLOW_DISSECT_RET_IPPROTO_AGAIN;
				break;
			}
952
		}
953 954 955

		fdret = FLOW_DISSECT_RET_OUT_GOOD;
		break;
956
	}
E
Eric Dumazet 已提交
957
	case IPPROTO_IPIP:
T
Tom Herbert 已提交
958
		proto = htons(ETH_P_IP);
959

960
		key_control->flags |= FLOW_DIS_ENCAPSULATION;
961 962 963 964 965 966 967
		if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) {
			fdret = FLOW_DISSECT_RET_OUT_GOOD;
			break;
		}

		fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
		break;
968

969 970
	case IPPROTO_IPV6:
		proto = htons(ETH_P_IPV6);
971

972
		key_control->flags |= FLOW_DIS_ENCAPSULATION;
973 974 975 976 977 978 979 980
		if (flags & FLOW_DISSECTOR_F_STOP_AT_ENCAP) {
			fdret = FLOW_DISSECT_RET_OUT_GOOD;
			break;
		}

		fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
		break;

981

982 983
	case IPPROTO_MPLS:
		proto = htons(ETH_P_MPLS_UC);
984 985 986
		fdret = FLOW_DISSECT_RET_PROTO_AGAIN;
		break;

987 988 989 990
	case IPPROTO_TCP:
		__skb_flow_dissect_tcp(skb, flow_dissector, target_container,
				       data, nhoff, hlen);
		break;
991

E
Eric Dumazet 已提交
992 993 994 995
	default:
		break;
	}

996 997
	if (dissector_uses_key(flow_dissector,
			       FLOW_DISSECTOR_KEY_PORTS)) {
998 999 1000 1001 1002 1003
		key_ports = skb_flow_dissector_target(flow_dissector,
						      FLOW_DISSECTOR_KEY_PORTS,
						      target_container);
		key_ports->ports = __skb_flow_get_ports(skb, nhoff, ip_proto,
							data, hlen);
	}
1004

S
Simon Horman 已提交
1005 1006 1007 1008 1009 1010 1011 1012
	if (dissector_uses_key(flow_dissector,
			       FLOW_DISSECTOR_KEY_ICMP)) {
		key_icmp = skb_flow_dissector_target(flow_dissector,
						     FLOW_DISSECTOR_KEY_ICMP,
						     target_container);
		key_icmp->icmp = skb_flow_get_be16(skb, nhoff, data, hlen);
	}

1013 1014 1015
	/* Process result of IP proto processing */
	switch (fdret) {
	case FLOW_DISSECT_RET_PROTO_AGAIN:
1016 1017 1018
		if (skb_flow_dissect_allowed(&num_hdrs))
			goto proto_again;
		break;
1019
	case FLOW_DISSECT_RET_IPPROTO_AGAIN:
1020 1021 1022
		if (skb_flow_dissect_allowed(&num_hdrs))
			goto ip_proto_again;
		break;
1023 1024 1025 1026 1027 1028 1029 1030
	case FLOW_DISSECT_RET_OUT_GOOD:
	case FLOW_DISSECT_RET_CONTINUE:
		break;
	case FLOW_DISSECT_RET_OUT_BAD:
	default:
		goto out_bad;
	}

1031 1032 1033
out_good:
	ret = true;

1034 1035
	key_control->thoff = (u16)nhoff;
out:
1036 1037 1038 1039
	key_basic->n_proto = proto;
	key_basic->ip_proto = ip_proto;

	return ret;
1040 1041 1042 1043 1044

out_bad:
	ret = false;
	key_control->thoff = min_t(u16, nhoff, skb ? skb->len : hlen);
	goto out;
E
Eric Dumazet 已提交
1045
}
1046
EXPORT_SYMBOL(__skb_flow_dissect);
1047 1048

static u32 hashrnd __read_mostly;
1049 1050 1051 1052 1053
static __always_inline void __flow_hash_secret_init(void)
{
	net_get_random_once(&hashrnd, sizeof(hashrnd));
}

1054 1055
static __always_inline u32 __flow_hash_words(const u32 *words, u32 length,
					     u32 keyval)
1056 1057 1058 1059
{
	return jhash2(words, length, keyval);
}

1060
static inline const u32 *flow_keys_hash_start(const struct flow_keys *flow)
1061
{
1062 1063
	const void *p = flow;

1064
	BUILD_BUG_ON(FLOW_KEYS_HASH_OFFSET % sizeof(u32));
1065
	return (const u32 *)(p + FLOW_KEYS_HASH_OFFSET);
1066 1067
}

1068
static inline size_t flow_keys_hash_length(const struct flow_keys *flow)
1069
{
1070
	size_t diff = FLOW_KEYS_HASH_OFFSET + sizeof(flow->addrs);
1071
	BUILD_BUG_ON((sizeof(*flow) - FLOW_KEYS_HASH_OFFSET) % sizeof(u32));
1072 1073 1074 1075 1076 1077 1078 1079 1080 1081
	BUILD_BUG_ON(offsetof(typeof(*flow), addrs) !=
		     sizeof(*flow) - sizeof(flow->addrs));

	switch (flow->control.addr_type) {
	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
		diff -= sizeof(flow->addrs.v4addrs);
		break;
	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
		diff -= sizeof(flow->addrs.v6addrs);
		break;
1082 1083
	case FLOW_DISSECTOR_KEY_TIPC:
		diff -= sizeof(flow->addrs.tipckey);
T
Tom Herbert 已提交
1084
		break;
1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096
	}
	return (sizeof(*flow) - diff) / sizeof(u32);
}

__be32 flow_get_u32_src(const struct flow_keys *flow)
{
	switch (flow->control.addr_type) {
	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
		return flow->addrs.v4addrs.src;
	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
		return (__force __be32)ipv6_addr_hash(
			&flow->addrs.v6addrs.src);
1097 1098
	case FLOW_DISSECTOR_KEY_TIPC:
		return flow->addrs.tipckey.key;
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
	default:
		return 0;
	}
}
EXPORT_SYMBOL(flow_get_u32_src);

__be32 flow_get_u32_dst(const struct flow_keys *flow)
{
	switch (flow->control.addr_type) {
	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
		return flow->addrs.v4addrs.dst;
	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
		return (__force __be32)ipv6_addr_hash(
			&flow->addrs.v6addrs.dst);
	default:
		return 0;
	}
}
EXPORT_SYMBOL(flow_get_u32_dst);

static inline void __flow_hash_consistentify(struct flow_keys *keys)
{
	int addr_diff, i;

	switch (keys->control.addr_type) {
	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
		addr_diff = (__force u32)keys->addrs.v4addrs.dst -
			    (__force u32)keys->addrs.v4addrs.src;
		if ((addr_diff < 0) ||
		    (addr_diff == 0 &&
		     ((__force u16)keys->ports.dst <
		      (__force u16)keys->ports.src))) {
			swap(keys->addrs.v4addrs.src, keys->addrs.v4addrs.dst);
			swap(keys->ports.src, keys->ports.dst);
		}
		break;
	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
		addr_diff = memcmp(&keys->addrs.v6addrs.dst,
				   &keys->addrs.v6addrs.src,
				   sizeof(keys->addrs.v6addrs.dst));
		if ((addr_diff < 0) ||
		    (addr_diff == 0 &&
		     ((__force u16)keys->ports.dst <
		      (__force u16)keys->ports.src))) {
			for (i = 0; i < 4; i++)
				swap(keys->addrs.v6addrs.src.s6_addr32[i],
				     keys->addrs.v6addrs.dst.s6_addr32[i]);
			swap(keys->ports.src, keys->ports.dst);
		}
		break;
	}
1150 1151
}

T
Tom Herbert 已提交
1152
static inline u32 __flow_hash_from_keys(struct flow_keys *keys, u32 keyval)
1153 1154 1155
{
	u32 hash;

1156
	__flow_hash_consistentify(keys);
1157

1158
	hash = __flow_hash_words(flow_keys_hash_start(keys),
1159
				 flow_keys_hash_length(keys), keyval);
1160 1161 1162 1163 1164 1165 1166 1167
	if (!hash)
		hash = 1;

	return hash;
}

u32 flow_hash_from_keys(struct flow_keys *keys)
{
T
Tom Herbert 已提交
1168 1169
	__flow_hash_secret_init();
	return __flow_hash_from_keys(keys, hashrnd);
1170 1171 1172
}
EXPORT_SYMBOL(flow_hash_from_keys);

T
Tom Herbert 已提交
1173 1174 1175
static inline u32 ___skb_get_hash(const struct sk_buff *skb,
				  struct flow_keys *keys, u32 keyval)
{
1176 1177
	skb_flow_dissect_flow_keys(skb, keys,
				   FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL);
T
Tom Herbert 已提交
1178 1179 1180 1181

	return __flow_hash_from_keys(keys, keyval);
}

T
Tom Herbert 已提交
1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200
struct _flow_keys_digest_data {
	__be16	n_proto;
	u8	ip_proto;
	u8	padding;
	__be32	ports;
	__be32	src;
	__be32	dst;
};

void make_flow_keys_digest(struct flow_keys_digest *digest,
			   const struct flow_keys *flow)
{
	struct _flow_keys_digest_data *data =
	    (struct _flow_keys_digest_data *)digest;

	BUILD_BUG_ON(sizeof(*data) > sizeof(*digest));

	memset(digest, 0, sizeof(*digest));

1201 1202 1203
	data->n_proto = flow->basic.n_proto;
	data->ip_proto = flow->basic.ip_proto;
	data->ports = flow->ports.ports;
1204 1205
	data->src = flow->addrs.v4addrs.src;
	data->dst = flow->addrs.v4addrs.dst;
T
Tom Herbert 已提交
1206 1207 1208
}
EXPORT_SYMBOL(make_flow_keys_digest);

1209 1210
static struct flow_dissector flow_keys_dissector_symmetric __read_mostly;

1211
u32 __skb_get_hash_symmetric(const struct sk_buff *skb)
1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225
{
	struct flow_keys keys;

	__flow_hash_secret_init();

	memset(&keys, 0, sizeof(keys));
	__skb_flow_dissect(skb, &flow_keys_dissector_symmetric, &keys,
			   NULL, 0, 0, 0,
			   FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL);

	return __flow_hash_from_keys(&keys, hashrnd);
}
EXPORT_SYMBOL_GPL(__skb_get_hash_symmetric);

1226 1227 1228 1229 1230
/**
 * __skb_get_hash: calculate a flow hash
 * @skb: sk_buff to calculate flow hash from
 *
 * This function calculates a flow hash based on src/dst addresses
1231 1232
 * and src/dst port numbers.  Sets hash in skb to non-zero hash value
 * on success, zero indicates no valid hash.  Also, sets l4_hash in skb
1233 1234
 * if hash is a canonical 4-tuple hash over transport ports.
 */
1235
void __skb_get_hash(struct sk_buff *skb)
1236 1237
{
	struct flow_keys keys;
1238
	u32 hash;
1239

T
Tom Herbert 已提交
1240 1241
	__flow_hash_secret_init();

1242 1243 1244
	hash = ___skb_get_hash(skb, &keys, hashrnd);

	__skb_set_sw_hash(skb, hash, flow_keys_have_l4(&keys));
1245
}
1246
EXPORT_SYMBOL(__skb_get_hash);
1247

T
Tom Herbert 已提交
1248 1249 1250 1251 1252 1253 1254 1255
__u32 skb_get_hash_perturb(const struct sk_buff *skb, u32 perturb)
{
	struct flow_keys keys;

	return ___skb_get_hash(skb, &keys, perturb);
}
EXPORT_SYMBOL(skb_get_hash_perturb);

1256 1257
u32 __skb_get_poff(const struct sk_buff *skb, void *data,
		   const struct flow_keys *keys, int hlen)
1258
{
1259
	u32 poff = keys->control.thoff;
1260

1261 1262 1263 1264 1265
	/* skip L4 headers for fragments after the first */
	if ((keys->control.flags & FLOW_DIS_IS_FRAGMENT) &&
	    !(keys->control.flags & FLOW_DIS_FIRST_FRAG))
		return poff;

1266
	switch (keys->basic.ip_proto) {
1267
	case IPPROTO_TCP: {
1268 1269 1270
		/* access doff as u8 to avoid unaligned access */
		const u8 *doff;
		u8 _doff;
1271

1272 1273 1274
		doff = __skb_header_pointer(skb, poff + 12, sizeof(_doff),
					    data, hlen, &_doff);
		if (!doff)
1275 1276
			return poff;

1277
		poff += max_t(u32, sizeof(struct tcphdr), (*doff & 0xF0) >> 2);
1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
		break;
	}
	case IPPROTO_UDP:
	case IPPROTO_UDPLITE:
		poff += sizeof(struct udphdr);
		break;
	/* For the rest, we do not really care about header
	 * extensions at this point for now.
	 */
	case IPPROTO_ICMP:
		poff += sizeof(struct icmphdr);
		break;
	case IPPROTO_ICMPV6:
		poff += sizeof(struct icmp6hdr);
		break;
	case IPPROTO_IGMP:
		poff += sizeof(struct igmphdr);
		break;
	case IPPROTO_DCCP:
		poff += sizeof(struct dccp_hdr);
		break;
	case IPPROTO_SCTP:
		poff += sizeof(struct sctphdr);
		break;
	}

	return poff;
}

1307 1308 1309 1310 1311 1312
/**
 * skb_get_poff - get the offset to the payload
 * @skb: sk_buff to get the payload offset from
 *
 * The function will get the offset to the payload as far as it could
 * be dissected.  The main user is currently BPF, so that we can dynamically
1313 1314 1315 1316 1317 1318 1319
 * truncate packets without needing to push actual payload to the user
 * space and can analyze headers only, instead.
 */
u32 skb_get_poff(const struct sk_buff *skb)
{
	struct flow_keys keys;

1320
	if (!skb_flow_dissect_flow_keys(skb, &keys, 0))
1321 1322 1323 1324
		return 0;

	return __skb_get_poff(skb, skb->data, &keys, skb_headlen(skb));
}
1325

1326
__u32 __get_hash_from_flowi6(const struct flowi6 *fl6, struct flow_keys *keys)
1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344
{
	memset(keys, 0, sizeof(*keys));

	memcpy(&keys->addrs.v6addrs.src, &fl6->saddr,
	    sizeof(keys->addrs.v6addrs.src));
	memcpy(&keys->addrs.v6addrs.dst, &fl6->daddr,
	    sizeof(keys->addrs.v6addrs.dst));
	keys->control.addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
	keys->ports.src = fl6->fl6_sport;
	keys->ports.dst = fl6->fl6_dport;
	keys->keyid.keyid = fl6->fl6_gre_key;
	keys->tags.flow_label = (__force u32)fl6->flowlabel;
	keys->basic.ip_proto = fl6->flowi6_proto;

	return flow_hash_from_keys(keys);
}
EXPORT_SYMBOL(__get_hash_from_flowi6);

1345
__u32 __get_hash_from_flowi4(const struct flowi4 *fl4, struct flow_keys *keys)
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360
{
	memset(keys, 0, sizeof(*keys));

	keys->addrs.v4addrs.src = fl4->saddr;
	keys->addrs.v4addrs.dst = fl4->daddr;
	keys->control.addr_type = FLOW_DISSECTOR_KEY_IPV4_ADDRS;
	keys->ports.src = fl4->fl4_sport;
	keys->ports.dst = fl4->fl4_dport;
	keys->keyid.keyid = fl4->fl4_gre_key;
	keys->basic.ip_proto = fl4->flowi4_proto;

	return flow_hash_from_keys(keys);
}
EXPORT_SYMBOL(__get_hash_from_flowi4);

1361
static const struct flow_dissector_key flow_keys_dissector_keys[] = {
1362 1363 1364 1365
	{
		.key_id = FLOW_DISSECTOR_KEY_CONTROL,
		.offset = offsetof(struct flow_keys, control),
	},
1366 1367 1368 1369 1370 1371
	{
		.key_id = FLOW_DISSECTOR_KEY_BASIC,
		.offset = offsetof(struct flow_keys, basic),
	},
	{
		.key_id = FLOW_DISSECTOR_KEY_IPV4_ADDRS,
1372 1373 1374 1375 1376
		.offset = offsetof(struct flow_keys, addrs.v4addrs),
	},
	{
		.key_id = FLOW_DISSECTOR_KEY_IPV6_ADDRS,
		.offset = offsetof(struct flow_keys, addrs.v6addrs),
1377
	},
T
Tom Herbert 已提交
1378
	{
1379 1380
		.key_id = FLOW_DISSECTOR_KEY_TIPC,
		.offset = offsetof(struct flow_keys, addrs.tipckey),
T
Tom Herbert 已提交
1381
	},
1382 1383 1384 1385
	{
		.key_id = FLOW_DISSECTOR_KEY_PORTS,
		.offset = offsetof(struct flow_keys, ports),
	},
T
Tom Herbert 已提交
1386
	{
1387 1388
		.key_id = FLOW_DISSECTOR_KEY_VLAN,
		.offset = offsetof(struct flow_keys, vlan),
T
Tom Herbert 已提交
1389
	},
1390 1391 1392 1393
	{
		.key_id = FLOW_DISSECTOR_KEY_FLOW_LABEL,
		.offset = offsetof(struct flow_keys, tags),
	},
T
Tom Herbert 已提交
1394 1395 1396 1397
	{
		.key_id = FLOW_DISSECTOR_KEY_GRE_KEYID,
		.offset = offsetof(struct flow_keys, keyid),
	},
1398 1399
};

1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422
static const struct flow_dissector_key flow_keys_dissector_symmetric_keys[] = {
	{
		.key_id = FLOW_DISSECTOR_KEY_CONTROL,
		.offset = offsetof(struct flow_keys, control),
	},
	{
		.key_id = FLOW_DISSECTOR_KEY_BASIC,
		.offset = offsetof(struct flow_keys, basic),
	},
	{
		.key_id = FLOW_DISSECTOR_KEY_IPV4_ADDRS,
		.offset = offsetof(struct flow_keys, addrs.v4addrs),
	},
	{
		.key_id = FLOW_DISSECTOR_KEY_IPV6_ADDRS,
		.offset = offsetof(struct flow_keys, addrs.v6addrs),
	},
	{
		.key_id = FLOW_DISSECTOR_KEY_PORTS,
		.offset = offsetof(struct flow_keys, ports),
	},
};

1423
static const struct flow_dissector_key flow_keys_buf_dissector_keys[] = {
1424 1425 1426 1427
	{
		.key_id = FLOW_DISSECTOR_KEY_CONTROL,
		.offset = offsetof(struct flow_keys, control),
	},
1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443
	{
		.key_id = FLOW_DISSECTOR_KEY_BASIC,
		.offset = offsetof(struct flow_keys, basic),
	},
};

struct flow_dissector flow_keys_dissector __read_mostly;
EXPORT_SYMBOL(flow_keys_dissector);

struct flow_dissector flow_keys_buf_dissector __read_mostly;

static int __init init_default_flow_dissectors(void)
{
	skb_flow_dissector_init(&flow_keys_dissector,
				flow_keys_dissector_keys,
				ARRAY_SIZE(flow_keys_dissector_keys));
1444 1445 1446
	skb_flow_dissector_init(&flow_keys_dissector_symmetric,
				flow_keys_dissector_symmetric_keys,
				ARRAY_SIZE(flow_keys_dissector_symmetric_keys));
1447 1448 1449 1450 1451 1452
	skb_flow_dissector_init(&flow_keys_buf_dissector,
				flow_keys_buf_dissector_keys,
				ARRAY_SIZE(flow_keys_buf_dissector_keys));
	return 0;
}

1453
core_initcall(init_default_flow_dissectors);