options.c 34.3 KB
Newer Older
P
Peter Krystad 已提交
1 2 3 4 5 6
// SPDX-License-Identifier: GPL-2.0
/* Multipath TCP
 *
 * Copyright (c) 2017 - 2019, Intel Corporation.
 */

7 8
#define pr_fmt(fmt) "MPTCP: " fmt

P
Peter Krystad 已提交
9
#include <linux/kernel.h>
10
#include <crypto/sha2.h>
P
Peter Krystad 已提交
11 12 13
#include <net/tcp.h>
#include <net/mptcp.h>
#include "protocol.h"
G
Geliang Tang 已提交
14
#include "mib.h"
P
Peter Krystad 已提交
15

16 17 18 19 20
static bool mptcp_cap_flag_sha256(u8 flags)
{
	return (flags & MPTCP_CAP_FLAG_MASK) == MPTCP_CAP_HMAC_SHA256;
}

21 22 23
static void mptcp_parse_option(const struct sk_buff *skb,
			       const unsigned char *ptr, int opsize,
			       struct mptcp_options_received *mp_opt)
P
Peter Krystad 已提交
24 25
{
	u8 subtype = *ptr >> 4;
26
	int expected_opsize;
P
Peter Krystad 已提交
27 28 29 30 31
	u8 version;
	u8 flags;

	switch (subtype) {
	case MPTCPOPT_MP_CAPABLE:
32 33 34 35 36 37 38 39 40 41 42 43 44
		/* strict size checking */
		if (!(TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN)) {
			if (skb->len > tcp_hdr(skb)->doff << 2)
				expected_opsize = TCPOLEN_MPTCP_MPC_ACK_DATA;
			else
				expected_opsize = TCPOLEN_MPTCP_MPC_ACK;
		} else {
			if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_ACK)
				expected_opsize = TCPOLEN_MPTCP_MPC_SYNACK;
			else
				expected_opsize = TCPOLEN_MPTCP_MPC_SYN;
		}
		if (opsize != expected_opsize)
P
Peter Krystad 已提交
45 46
			break;

47
		/* try to be gentle vs future versions on the initial syn */
P
Peter Krystad 已提交
48
		version = *ptr++ & MPTCP_VERSION_MASK;
49 50 51 52
		if (opsize != TCPOLEN_MPTCP_MPC_SYN) {
			if (version != MPTCP_SUPPORTED_VERSION)
				break;
		} else if (version < MPTCP_SUPPORTED_VERSION) {
P
Peter Krystad 已提交
53
			break;
54
		}
P
Peter Krystad 已提交
55 56

		flags = *ptr++;
57
		if (!mptcp_cap_flag_sha256(flags) ||
P
Peter Krystad 已提交
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77
		    (flags & MPTCP_CAP_EXTENSIBILITY))
			break;

		/* RFC 6824, Section 3.1:
		 * "For the Checksum Required bit (labeled "A"), if either
		 * host requires the use of checksums, checksums MUST be used.
		 * In other words, the only way for checksums not to be used
		 * is if both hosts in their SYNs set A=0."
		 *
		 * Section 3.3.0:
		 * "If a checksum is not present when its use has been
		 * negotiated, the receiver MUST close the subflow with a RST as
		 * it is considered broken."
		 *
		 * We don't implement DSS checksum - fall back to TCP.
		 */
		if (flags & MPTCP_CAP_CHECKSUM_REQD)
			break;

		mp_opt->mp_capable = 1;
78 79 80 81 82
		if (opsize >= TCPOLEN_MPTCP_MPC_SYNACK) {
			mp_opt->sndr_key = get_unaligned_be64(ptr);
			ptr += 8;
		}
		if (opsize >= TCPOLEN_MPTCP_MPC_ACK) {
P
Peter Krystad 已提交
83 84 85
			mp_opt->rcvr_key = get_unaligned_be64(ptr);
			ptr += 8;
		}
86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
		if (opsize == TCPOLEN_MPTCP_MPC_ACK_DATA) {
			/* Section 3.1.:
			 * "the data parameters in a MP_CAPABLE are semantically
			 * equivalent to those in a DSS option and can be used
			 * interchangeably."
			 */
			mp_opt->dss = 1;
			mp_opt->use_map = 1;
			mp_opt->mpc_map = 1;
			mp_opt->data_len = get_unaligned_be16(ptr);
			ptr += 2;
		}
		pr_debug("MP_CAPABLE version=%x, flags=%x, optlen=%d sndr=%llu, rcvr=%llu len=%d",
			 version, flags, opsize, mp_opt->sndr_key,
			 mp_opt->rcvr_key, mp_opt->data_len);
P
Peter Krystad 已提交
101 102
		break;

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
	case MPTCPOPT_MP_JOIN:
		mp_opt->mp_join = 1;
		if (opsize == TCPOLEN_MPTCP_MPJ_SYN) {
			mp_opt->backup = *ptr++ & MPTCPOPT_BACKUP;
			mp_opt->join_id = *ptr++;
			mp_opt->token = get_unaligned_be32(ptr);
			ptr += 4;
			mp_opt->nonce = get_unaligned_be32(ptr);
			ptr += 4;
			pr_debug("MP_JOIN bkup=%u, id=%u, token=%u, nonce=%u",
				 mp_opt->backup, mp_opt->join_id,
				 mp_opt->token, mp_opt->nonce);
		} else if (opsize == TCPOLEN_MPTCP_MPJ_SYNACK) {
			mp_opt->backup = *ptr++ & MPTCPOPT_BACKUP;
			mp_opt->join_id = *ptr++;
			mp_opt->thmac = get_unaligned_be64(ptr);
			ptr += 8;
			mp_opt->nonce = get_unaligned_be32(ptr);
			ptr += 4;
			pr_debug("MP_JOIN bkup=%u, id=%u, thmac=%llu, nonce=%u",
				 mp_opt->backup, mp_opt->join_id,
				 mp_opt->thmac, mp_opt->nonce);
		} else if (opsize == TCPOLEN_MPTCP_MPJ_ACK) {
			ptr += 2;
			memcpy(mp_opt->hmac, ptr, MPTCPOPT_HMAC_LEN);
			pr_debug("MP_JOIN hmac");
		} else {
			pr_warn("MP_JOIN bad option size");
			mp_opt->mp_join = 0;
		}
		break;

P
Peter Krystad 已提交
135 136
	case MPTCPOPT_DSS:
		pr_debug("DSS");
137 138
		ptr++;

139 140 141 142 143
		/* we must clear 'mpc_map' be able to detect MP_CAPABLE
		 * map vs DSS map in mptcp_incoming_options(), and reconstruct
		 * map info accordingly
		 */
		mp_opt->mpc_map = 0;
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
		flags = (*ptr++) & MPTCP_DSS_FLAG_MASK;
		mp_opt->data_fin = (flags & MPTCP_DSS_DATA_FIN) != 0;
		mp_opt->dsn64 = (flags & MPTCP_DSS_DSN64) != 0;
		mp_opt->use_map = (flags & MPTCP_DSS_HAS_MAP) != 0;
		mp_opt->ack64 = (flags & MPTCP_DSS_ACK64) != 0;
		mp_opt->use_ack = (flags & MPTCP_DSS_HAS_ACK);

		pr_debug("data_fin=%d dsn64=%d use_map=%d ack64=%d use_ack=%d",
			 mp_opt->data_fin, mp_opt->dsn64,
			 mp_opt->use_map, mp_opt->ack64,
			 mp_opt->use_ack);

		expected_opsize = TCPOLEN_MPTCP_DSS_BASE;

		if (mp_opt->use_ack) {
			if (mp_opt->ack64)
				expected_opsize += TCPOLEN_MPTCP_DSS_ACK64;
			else
				expected_opsize += TCPOLEN_MPTCP_DSS_ACK32;
		}

		if (mp_opt->use_map) {
			if (mp_opt->dsn64)
				expected_opsize += TCPOLEN_MPTCP_DSS_MAP64;
			else
				expected_opsize += TCPOLEN_MPTCP_DSS_MAP32;
		}

		/* RFC 6824, Section 3.3:
		 * If a checksum is present, but its use had
		 * not been negotiated in the MP_CAPABLE handshake,
		 * the checksum field MUST be ignored.
		 */
		if (opsize != expected_opsize &&
		    opsize != expected_opsize + TCPOLEN_MPTCP_DSS_CHECKSUM)
			break;

P
Peter Krystad 已提交
181
		mp_opt->dss = 1;
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

		if (mp_opt->use_ack) {
			if (mp_opt->ack64) {
				mp_opt->data_ack = get_unaligned_be64(ptr);
				ptr += 8;
			} else {
				mp_opt->data_ack = get_unaligned_be32(ptr);
				ptr += 4;
			}

			pr_debug("data_ack=%llu", mp_opt->data_ack);
		}

		if (mp_opt->use_map) {
			if (mp_opt->dsn64) {
				mp_opt->data_seq = get_unaligned_be64(ptr);
				ptr += 8;
			} else {
				mp_opt->data_seq = get_unaligned_be32(ptr);
				ptr += 4;
			}

			mp_opt->subflow_seq = get_unaligned_be32(ptr);
			ptr += 4;

			mp_opt->data_len = get_unaligned_be16(ptr);
			ptr += 2;

			pr_debug("data_seq=%llu subflow_seq=%u data_len=%u",
				 mp_opt->data_seq, mp_opt->subflow_seq,
				 mp_opt->data_len);
		}

P
Peter Krystad 已提交
215 216
		break;

P
Peter Krystad 已提交
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
	case MPTCPOPT_ADD_ADDR:
		mp_opt->echo = (*ptr++) & MPTCP_ADDR_ECHO;
		if (!mp_opt->echo) {
			if (opsize == TCPOLEN_MPTCP_ADD_ADDR ||
			    opsize == TCPOLEN_MPTCP_ADD_ADDR_PORT)
				mp_opt->family = MPTCP_ADDR_IPVERSION_4;
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
			else if (opsize == TCPOLEN_MPTCP_ADD_ADDR6 ||
				 opsize == TCPOLEN_MPTCP_ADD_ADDR6_PORT)
				mp_opt->family = MPTCP_ADDR_IPVERSION_6;
#endif
			else
				break;
		} else {
			if (opsize == TCPOLEN_MPTCP_ADD_ADDR_BASE ||
			    opsize == TCPOLEN_MPTCP_ADD_ADDR_BASE_PORT)
				mp_opt->family = MPTCP_ADDR_IPVERSION_4;
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
			else if (opsize == TCPOLEN_MPTCP_ADD_ADDR6_BASE ||
				 opsize == TCPOLEN_MPTCP_ADD_ADDR6_BASE_PORT)
				mp_opt->family = MPTCP_ADDR_IPVERSION_6;
#endif
			else
				break;
		}

		mp_opt->add_addr = 1;
		mp_opt->addr_id = *ptr++;
		if (mp_opt->family == MPTCP_ADDR_IPVERSION_4) {
			memcpy((u8 *)&mp_opt->addr.s_addr, (u8 *)ptr, 4);
			ptr += 4;
			if (opsize == TCPOLEN_MPTCP_ADD_ADDR_PORT ||
			    opsize == TCPOLEN_MPTCP_ADD_ADDR_BASE_PORT) {
				mp_opt->port = get_unaligned_be16(ptr);
				ptr += 2;
			}
		}
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
		else {
			memcpy(mp_opt->addr6.s6_addr, (u8 *)ptr, 16);
			ptr += 16;
			if (opsize == TCPOLEN_MPTCP_ADD_ADDR6_PORT ||
			    opsize == TCPOLEN_MPTCP_ADD_ADDR6_BASE_PORT) {
				mp_opt->port = get_unaligned_be16(ptr);
				ptr += 2;
			}
		}
#endif
		if (!mp_opt->echo) {
			mp_opt->ahmac = get_unaligned_be64(ptr);
			ptr += 8;
		}
269 270 271
		pr_debug("ADD_ADDR%s: id=%d, ahmac=%llu, echo=%d, port=%d",
			 (mp_opt->family == MPTCP_ADDR_IPVERSION_6) ? "6" : "",
			 mp_opt->addr_id, mp_opt->ahmac, mp_opt->echo, mp_opt->port);
P
Peter Krystad 已提交
272 273 274 275 276 277
		break;

	case MPTCPOPT_RM_ADDR:
		if (opsize != TCPOLEN_MPTCP_RM_ADDR_BASE)
			break;

278 279
		ptr++;

P
Peter Krystad 已提交
280 281 282 283 284
		mp_opt->rm_addr = 1;
		mp_opt->rm_id = *ptr++;
		pr_debug("RM_ADDR: id=%d", mp_opt->rm_id);
		break;

285 286 287 288 289 290 291 292 293
	case MPTCPOPT_MP_PRIO:
		if (opsize != TCPOLEN_MPTCP_PRIO)
			break;

		mp_opt->mp_prio = 1;
		mp_opt->backup = *ptr++ & MPTCP_PRIO_BKUP;
		pr_debug("MP_PRIO: prio=%d", mp_opt->backup);
		break;

294 295 296 297 298 299 300 301 302 303
	case MPTCPOPT_MP_FASTCLOSE:
		if (opsize != TCPOLEN_MPTCP_FASTCLOSE)
			break;

		ptr += 2;
		mp_opt->rcvr_key = get_unaligned_be64(ptr);
		ptr += 8;
		mp_opt->fastclose = 1;
		break;

P
Peter Krystad 已提交
304 305 306 307 308
	default:
		break;
	}
}

309
void mptcp_get_options(const struct sk_buff *skb,
310
		       struct mptcp_options_received *mp_opt)
311 312
{
	const struct tcphdr *th = tcp_hdr(skb);
313 314
	const unsigned char *ptr;
	int length;
315

316 317 318 319
	/* initialize option status */
	mp_opt->mp_capable = 0;
	mp_opt->mp_join = 0;
	mp_opt->add_addr = 0;
320
	mp_opt->ahmac = 0;
321
	mp_opt->fastclose = 0;
322
	mp_opt->port = 0;
323 324
	mp_opt->rm_addr = 0;
	mp_opt->dss = 0;
325
	mp_opt->mp_prio = 0;
326 327

	length = (th->doff * 4) - sizeof(struct tcphdr);
328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346
	ptr = (const unsigned char *)(th + 1);

	while (length > 0) {
		int opcode = *ptr++;
		int opsize;

		switch (opcode) {
		case TCPOPT_EOL:
			return;
		case TCPOPT_NOP:	/* Ref: RFC 793 section 3.1 */
			length--;
			continue;
		default:
			opsize = *ptr++;
			if (opsize < 2) /* "silly options" */
				return;
			if (opsize > length)
				return;	/* don't parse partial options */
			if (opcode == TCPOPT_MPTCP)
347
				mptcp_parse_option(skb, ptr, opsize, mp_opt);
348 349 350 351 352 353
			ptr += opsize - 2;
			length -= opsize;
		}
	}
}

354 355
bool mptcp_syn_options(struct sock *sk, const struct sk_buff *skb,
		       unsigned int *size, struct mptcp_out_options *opts)
356 357 358
{
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);

359 360 361 362
	/* we will use snd_isn to detect first pkt [re]transmission
	 * in mptcp_established_options_mp()
	 */
	subflow->snd_isn = TCP_SKB_CB(skb)->end_seq;
363 364 365 366
	if (subflow->request_mptcp) {
		opts->suboptions = OPTION_MPTCP_MPC_SYN;
		*size = TCPOLEN_MPTCP_MPC_SYN;
		return true;
367 368 369 370 371 372 373 374 375 376
	} else if (subflow->request_join) {
		pr_debug("remote_token=%u, nonce=%u", subflow->remote_token,
			 subflow->local_nonce);
		opts->suboptions = OPTION_MPTCP_MPJ_SYN;
		opts->join_id = subflow->local_id;
		opts->token = subflow->remote_token;
		opts->nonce = subflow->local_nonce;
		opts->backup = subflow->request_bkup;
		*size = TCPOLEN_MPTCP_MPJ_SYN;
		return true;
377 378 379 380
	}
	return false;
}

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
/* MP_JOIN client subflow must wait for 4th ack before sending any data:
 * TCP can't schedule delack timer before the subflow is fully established.
 * MPTCP uses the delack timer to do 3rd ack retransmissions
 */
static void schedule_3rdack_retransmission(struct sock *sk)
{
	struct inet_connection_sock *icsk = inet_csk(sk);
	struct tcp_sock *tp = tcp_sk(sk);
	unsigned long timeout;

	/* reschedule with a timeout above RTT, as we must look only for drop */
	if (tp->srtt_us)
		timeout = tp->srtt_us << 1;
	else
		timeout = TCP_TIMEOUT_INIT;

	WARN_ON_ONCE(icsk->icsk_ack.pending & ICSK_ACK_TIMER);
	icsk->icsk_ack.pending |= ICSK_ACK_SCHED | ICSK_ACK_TIMER;
	icsk->icsk_ack.timeout = timeout;
	sk_reset_timer(sk, &icsk->icsk_delack_timer, timeout);
}

static void clear_3rdack_retransmission(struct sock *sk)
{
	struct inet_connection_sock *icsk = inet_csk(sk);

	sk_stop_timer(sk, &icsk->icsk_delack_timer);
	icsk->icsk_ack.timeout = 0;
	icsk->icsk_ack.ato = 0;
	icsk->icsk_ack.pending &= ~(ICSK_ACK_SCHED | ICSK_ACK_TIMER);
}

413 414
static bool mptcp_established_options_mp(struct sock *sk, struct sk_buff *skb,
					 unsigned int *size,
415 416
					 unsigned int remaining,
					 struct mptcp_out_options *opts)
417 418
{
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
419 420 421
	struct mptcp_ext *mpext;
	unsigned int data_len;

422 423 424 425 426 427 428 429
	/* When skb is not available, we better over-estimate the emitted
	 * options len. A full DSS option (28 bytes) is longer than
	 * TCPOLEN_MPTCP_MPC_ACK_DATA(22) or TCPOLEN_MPTCP_MPJ_ACK(24), so
	 * tell the caller to defer the estimate to
	 * mptcp_established_options_dss(), which will reserve enough space.
	 */
	if (!skb)
		return false;
430

431 432 433 434 435 436
	/* MPC/MPJ needed only on 3rd ack packet */
	if (subflow->fully_established ||
	    subflow->snd_isn != TCP_SKB_CB(skb)->seq)
		return false;

	if (subflow->mp_capable) {
437 438
		mpext = mptcp_get_ext(skb);
		data_len = mpext ? mpext->data_len : 0;
439

440 441 442 443 444
		/* we will check ext_copy.data_len in mptcp_write_options() to
		 * discriminate between TCPOLEN_MPTCP_MPC_ACK_DATA and
		 * TCPOLEN_MPTCP_MPC_ACK
		 */
		opts->ext_copy.data_len = data_len;
445 446 447
		opts->suboptions = OPTION_MPTCP_MPC_ACK;
		opts->sndr_key = subflow->local_key;
		opts->rcvr_key = subflow->remote_key;
448 449 450 451 452 453 454 455 456 457 458 459 460 461 462

		/* Section 3.1.
		 * The MP_CAPABLE option is carried on the SYN, SYN/ACK, and ACK
		 * packets that start the first subflow of an MPTCP connection,
		 * as well as the first packet that carries data
		 */
		if (data_len > 0)
			*size = ALIGN(TCPOLEN_MPTCP_MPC_ACK_DATA, 4);
		else
			*size = TCPOLEN_MPTCP_MPC_ACK;

		pr_debug("subflow=%p, local_key=%llu, remote_key=%llu map_len=%d",
			 subflow, subflow->local_key, subflow->remote_key,
			 data_len);

463
		return true;
464 465 466 467 468 469 470 471
	} else if (subflow->mp_join) {
		opts->suboptions = OPTION_MPTCP_MPJ_ACK;
		memcpy(opts->hmac, subflow->hmac, MPTCPOPT_HMAC_LEN);
		*size = TCPOLEN_MPTCP_MPJ_ACK;
		pr_debug("subflow=%p", subflow);

		schedule_3rdack_retransmission(sk);
		return true;
472 473 474 475
	}
	return false;
}

476
static void mptcp_write_data_fin(struct mptcp_subflow_context *subflow,
477
				 struct sk_buff *skb, struct mptcp_ext *ext)
478
{
P
Paolo Abeni 已提交
479 480 481 482
	/* The write_seq value has already been incremented, so the actual
	 * sequence number for the DATA_FIN is one less.
	 */
	u64 data_fin_tx_seq = READ_ONCE(mptcp_sk(subflow->conn)->write_seq) - 1;
483

484
	if (!ext->use_map || !skb->len) {
485 486 487
		/* RFC6824 requires a DSS mapping with specific values
		 * if DATA_FIN is set but no data payload is mapped
		 */
488
		ext->data_fin = 1;
489 490
		ext->use_map = 1;
		ext->dsn64 = 1;
P
Paolo Abeni 已提交
491
		ext->data_seq = data_fin_tx_seq;
492 493
		ext->subflow_seq = 0;
		ext->data_len = 1;
494
	} else if (ext->data_seq + ext->data_len == data_fin_tx_seq) {
495 496 497
		/* If there's an existing DSS mapping and it is the
		 * final mapping, DATA_FIN consumes 1 additional byte of
		 * mapping space.
498
		 */
499
		ext->data_fin = 1;
500 501 502 503 504 505 506 507 508 509
		ext->data_len++;
	}
}

static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb,
					  unsigned int *size,
					  unsigned int remaining,
					  struct mptcp_out_options *opts)
{
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
510
	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
511
	unsigned int dss_size = 0;
512
	u64 snd_data_fin_enable;
513 514
	struct mptcp_ext *mpext;
	unsigned int ack_size;
515
	bool ret = false;
516

517
	mpext = skb ? mptcp_get_ext(skb) : NULL;
P
Paolo Abeni 已提交
518
	snd_data_fin_enable = mptcp_data_fin_enabled(msk);
519

520
	if (!skb || (mpext && mpext->use_map) || snd_data_fin_enable) {
521 522 523 524 525 526 527 528 529
		unsigned int map_size;

		map_size = TCPOLEN_MPTCP_DSS_BASE + TCPOLEN_MPTCP_DSS_MAP64;

		remaining -= map_size;
		dss_size = map_size;
		if (mpext)
			opts->ext_copy = *mpext;

530
		if (skb && snd_data_fin_enable)
531
			mptcp_write_data_fin(subflow, skb, &opts->ext_copy);
532 533 534
		ret = true;
	}

535 536 537
	/* passive sockets msk will set the 'can_ack' after accept(), even
	 * if the first subflow may have the already the remote key handy
	 */
538
	opts->ext_copy.use_ack = 0;
P
Paolo Abeni 已提交
539
	if (!READ_ONCE(msk->can_ack)) {
540 541
		*size = ALIGN(dss_size, 4);
		return ret;
542 543
	}

544
	if (READ_ONCE(msk->use_64bit_ack)) {
545
		ack_size = TCPOLEN_MPTCP_DSS_ACK64;
546
		opts->ext_copy.data_ack = READ_ONCE(msk->ack_seq);
547 548 549
		opts->ext_copy.ack64 = 1;
	} else {
		ack_size = TCPOLEN_MPTCP_DSS_ACK32;
550
		opts->ext_copy.data_ack32 = (uint32_t)READ_ONCE(msk->ack_seq);
551 552 553
		opts->ext_copy.ack64 = 0;
	}
	opts->ext_copy.use_ack = 1;
554
	WRITE_ONCE(msk->old_wspace, __mptcp_space((struct sock *)msk));
555 556 557 558 559 560 561 562 563 564 565

	/* Add kind/length/subtype/flag overhead if mapping is not populated */
	if (dss_size == 0)
		ack_size += TCPOLEN_MPTCP_DSS_BASE;

	dss_size += ack_size;

	*size = ALIGN(dss_size, 4);
	return true;
}

P
Peter Krystad 已提交
566 567 568
static u64 add_addr_generate_hmac(u64 key1, u64 key2, u8 addr_id,
				  struct in_addr *addr)
{
569
	u8 hmac[SHA256_DIGEST_SIZE];
P
Peter Krystad 已提交
570 571 572 573 574 575 576 577 578
	u8 msg[7];

	msg[0] = addr_id;
	memcpy(&msg[1], &addr->s_addr, 4);
	msg[5] = 0;
	msg[6] = 0;

	mptcp_crypto_hmac_sha(key1, key2, msg, 7, hmac);

579
	return get_unaligned_be64(&hmac[SHA256_DIGEST_SIZE - sizeof(u64)]);
P
Peter Krystad 已提交
580 581 582 583 584 585
}

#if IS_ENABLED(CONFIG_MPTCP_IPV6)
static u64 add_addr6_generate_hmac(u64 key1, u64 key2, u8 addr_id,
				   struct in6_addr *addr)
{
586
	u8 hmac[SHA256_DIGEST_SIZE];
P
Peter Krystad 已提交
587 588 589 590 591 592 593 594 595
	u8 msg[19];

	msg[0] = addr_id;
	memcpy(&msg[1], &addr->s6_addr, 16);
	msg[17] = 0;
	msg[18] = 0;

	mptcp_crypto_hmac_sha(key1, key2, msg, 19, hmac);

596
	return get_unaligned_be64(&hmac[SHA256_DIGEST_SIZE - sizeof(u64)]);
P
Peter Krystad 已提交
597 598 599
}
#endif

600
static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff *skb,
601 602 603
					       unsigned int *size,
					       unsigned int remaining,
					       struct mptcp_out_options *opts)
P
Peter Krystad 已提交
604 605 606
{
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
607 608
	bool drop_other_suboptions = false;
	unsigned int opt_size = *size;
609
	struct mptcp_addr_info saddr;
610
	bool echo;
611
	bool port;
612
	int len;
P
Peter Krystad 已提交
613

614 615
	if ((mptcp_pm_should_add_signal_ipv6(msk) ||
	     mptcp_pm_should_add_signal_port(msk)) &&
616 617 618
	    skb && skb_is_tcp_pure_ack(skb)) {
		pr_debug("drop other suboptions");
		opts->suboptions = 0;
619 620
		opts->ext_copy.use_ack = 0;
		opts->ext_copy.use_map = 0;
621 622 623 624
		remaining += opt_size;
		drop_other_suboptions = true;
	}

625
	if (!mptcp_pm_should_add_signal(msk) ||
626
	    !(mptcp_pm_add_addr_signal(msk, remaining, &saddr, &echo, &port)))
627 628
		return false;

629
	len = mptcp_add_addr_len(saddr.family, echo, port);
630 631
	if (remaining < len)
		return false;
P
Peter Krystad 已提交
632

633
	*size = len;
634 635
	if (drop_other_suboptions)
		*size -= opt_size;
636
	opts->addr_id = saddr.id;
637 638
	if (port)
		opts->port = ntohs(saddr.port);
639
	if (saddr.family == AF_INET) {
P
Peter Krystad 已提交
640
		opts->suboptions |= OPTION_MPTCP_ADD_ADDR;
641
		opts->addr = saddr.addr;
642 643 644 645 646 647
		if (!echo) {
			opts->ahmac = add_addr_generate_hmac(msk->local_key,
							     msk->remote_key,
							     opts->addr_id,
							     &opts->addr);
		}
P
Peter Krystad 已提交
648 649
	}
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
650
	else if (saddr.family == AF_INET6) {
P
Peter Krystad 已提交
651
		opts->suboptions |= OPTION_MPTCP_ADD_ADDR6;
652
		opts->addr6 = saddr.addr6;
653 654 655 656 657 658
		if (!echo) {
			opts->ahmac = add_addr6_generate_hmac(msk->local_key,
							      msk->remote_key,
							      opts->addr_id,
							      &opts->addr6);
		}
P
Peter Krystad 已提交
659 660
	}
#endif
661 662
	pr_debug("addr_id=%d, ahmac=%llu, echo=%d, port=%d",
		 opts->addr_id, opts->ahmac, echo, opts->port);
P
Peter Krystad 已提交
663 664 665 666

	return true;
}

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
static bool mptcp_established_options_rm_addr(struct sock *sk,
					      unsigned int *size,
					      unsigned int remaining,
					      struct mptcp_out_options *opts)
{
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
	u8 rm_id;

	if (!mptcp_pm_should_rm_signal(msk) ||
	    !(mptcp_pm_rm_addr_signal(msk, remaining, &rm_id)))
		return false;

	if (remaining < TCPOLEN_MPTCP_RM_ADDR_BASE)
		return false;

	*size = TCPOLEN_MPTCP_RM_ADDR_BASE;
	opts->suboptions |= OPTION_MPTCP_RM_ADDR;
	opts->rm_id = rm_id;

	pr_debug("rm_id=%d", opts->rm_id);

	return true;
}

692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713
static bool mptcp_established_options_mp_prio(struct sock *sk,
					      unsigned int *size,
					      unsigned int remaining,
					      struct mptcp_out_options *opts)
{
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);

	if (!subflow->send_mp_prio)
		return false;

	if (remaining < TCPOLEN_MPTCP_PRIO)
		return false;

	*size = TCPOLEN_MPTCP_PRIO;
	opts->suboptions |= OPTION_MPTCP_PRIO;
	opts->backup = subflow->request_bkup;

	pr_debug("prio=%d", opts->backup);

	return true;
}

714 715 716 717 718 719 720
bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
			       unsigned int *size, unsigned int remaining,
			       struct mptcp_out_options *opts)
{
	unsigned int opt_size = 0;
	bool ret = false;

P
Peter Krystad 已提交
721 722
	opts->suboptions = 0;

723 724 725
	if (unlikely(mptcp_check_fallback(sk)))
		return false;

726 727 728 729 730 731
	/* prevent adding of any MPTCP related options on reset packet
	 * until we support MP_TCPRST/MP_FASTCLOSE
	 */
	if (unlikely(skb && TCP_SKB_CB(skb)->tcp_flags & TCPHDR_RST))
		return false;

732
	if (mptcp_established_options_mp(sk, skb, &opt_size, remaining, opts))
733 734 735 736 737 738 739 740 741 742 743 744 745
		ret = true;
	else if (mptcp_established_options_dss(sk, skb, &opt_size, remaining,
					       opts))
		ret = true;

	/* we reserved enough space for the above options, and exceeding the
	 * TCP option space would be fatal
	 */
	if (WARN_ON_ONCE(opt_size > remaining))
		return false;

	*size += opt_size;
	remaining -= opt_size;
746
	if (mptcp_established_options_add_addr(sk, skb, &opt_size, remaining, opts)) {
P
Peter Krystad 已提交
747 748 749
		*size += opt_size;
		remaining -= opt_size;
		ret = true;
750 751 752 753
	} else if (mptcp_established_options_rm_addr(sk, &opt_size, remaining, opts)) {
		*size += opt_size;
		remaining -= opt_size;
		ret = true;
P
Peter Krystad 已提交
754
	}
755

756 757 758 759 760 761
	if (mptcp_established_options_mp_prio(sk, &opt_size, remaining, opts)) {
		*size += opt_size;
		remaining -= opt_size;
		ret = true;
	}

762 763 764
	return ret;
}

765 766 767 768 769 770 771 772 773 774 775 776
bool mptcp_synack_options(const struct request_sock *req, unsigned int *size,
			  struct mptcp_out_options *opts)
{
	struct mptcp_subflow_request_sock *subflow_req = mptcp_subflow_rsk(req);

	if (subflow_req->mp_capable) {
		opts->suboptions = OPTION_MPTCP_MPC_SYNACK;
		opts->sndr_key = subflow_req->local_key;
		*size = TCPOLEN_MPTCP_MPC_SYNACK;
		pr_debug("subflow_req=%p, local_key=%llu",
			 subflow_req, subflow_req->local_key);
		return true;
777 778 779 780 781 782 783 784 785 786 787
	} else if (subflow_req->mp_join) {
		opts->suboptions = OPTION_MPTCP_MPJ_SYNACK;
		opts->backup = subflow_req->backup;
		opts->join_id = subflow_req->local_id;
		opts->thmac = subflow_req->thmac;
		opts->nonce = subflow_req->local_nonce;
		pr_debug("req=%p, bkup=%u, id=%u, thmac=%llu, nonce=%u",
			 subflow_req, opts->backup, opts->join_id,
			 opts->thmac, opts->nonce);
		*size = TCPOLEN_MPTCP_MPJ_SYNACK;
		return true;
788 789 790 791
	}
	return false;
}

792
static bool check_fully_established(struct mptcp_sock *msk, struct sock *ssk,
793
				    struct mptcp_subflow_context *subflow,
P
Paolo Abeni 已提交
794 795
				    struct sk_buff *skb,
				    struct mptcp_options_received *mp_opt)
796 797
{
	/* here we can process OoO, in-window pkts, only in-sequence 4th ack
798
	 * will make the subflow fully established
799
	 */
800 801 802 803 804 805 806 807 808
	if (likely(subflow->fully_established)) {
		/* on passive sockets, check for 3rd ack retransmission
		 * note that msk is always set by subflow_syn_recv_sock()
		 * for mp_join subflows
		 */
		if (TCP_SKB_CB(skb)->seq == subflow->ssn_offset + 1 &&
		    TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq &&
		    subflow->mp_join && mp_opt->mp_join &&
		    READ_ONCE(msk->pm.server_side))
809
			tcp_send_ack(ssk);
810 811 812
		goto fully_established;
	}

813 814 815 816
	/* we must process OoO packets before the first subflow is fully
	 * established. OoO packets are instead a protocol violation
	 * for MP_JOIN subflows as the peer must not send any data
	 * before receiving the forth ack - cfr. RFC 8684 section 3.2.
817
	 */
818 819 820
	if (TCP_SKB_CB(skb)->seq != subflow->ssn_offset + 1) {
		if (subflow->mp_join)
			goto reset;
821
		return subflow->mp_capable;
822
	}
823

824
	if (mp_opt->dss && mp_opt->use_ack) {
825 826 827
		/* subflows are fully established as soon as we get any
		 * additional ack.
		 */
P
Paolo Abeni 已提交
828
		subflow->fully_established = 1;
829
		WRITE_ONCE(msk->fully_established, true);
830 831
		goto fully_established;
	}
832

833 834 835 836 837
	if (mp_opt->add_addr) {
		WRITE_ONCE(msk->fully_established, true);
		return true;
	}

838
	/* If the first established packet does not contain MP_CAPABLE + data
839 840
	 * then fallback to TCP. Fallback scenarios requires a reset for
	 * MP_JOIN subflows.
841 842
	 */
	if (!mp_opt->mp_capable) {
843 844
		if (subflow->mp_join)
			goto reset;
845
		subflow->mp_capable = 0;
846 847
		pr_fallback(msk);
		__mptcp_do_fallback(msk);
848 849
		return false;
	}
850

P
Paolo Abeni 已提交
851 852
	if (unlikely(!READ_ONCE(msk->pm.server_side)))
		pr_warn_once("bogus mpc option on established client sk");
853
	mptcp_subflow_fully_established(subflow, mp_opt);
854 855

fully_established:
856 857 858 859 860 861
	/* if the subflow is not already linked into the conn_list, we can't
	 * notify the PM: this subflow is still on the listener queue
	 * and the PM possibly acquiring the subflow lock could race with
	 * the listener close
	 */
	if (likely(subflow->pm_notified) || list_empty(&subflow->node))
862 863 864
		return true;

	subflow->pm_notified = 1;
865
	if (subflow->mp_join) {
866
		clear_3rdack_retransmission(ssk);
867
		mptcp_pm_subflow_established(msk, subflow);
868
	} else {
869
		mptcp_pm_fully_established(msk);
870
	}
871
	return true;
872 873 874 875

reset:
	mptcp_subflow_reset(ssk);
	return false;
876 877
}

878 879 880 881 882 883 884 885 886 887 888 889 890 891 892
static u64 expand_ack(u64 old_ack, u64 cur_ack, bool use_64bit)
{
	u32 old_ack32, cur_ack32;

	if (use_64bit)
		return cur_ack;

	old_ack32 = (u32)old_ack;
	cur_ack32 = (u32)cur_ack;
	cur_ack = (old_ack & GENMASK_ULL(63, 32)) + cur_ack32;
	if (unlikely(before(cur_ack32, old_ack32)))
		return cur_ack + (1LL << 32);
	return cur_ack;
}

893
static void ack_update_msk(struct mptcp_sock *msk,
894
			   struct sock *ssk,
895
			   struct mptcp_options_received *mp_opt)
896
{
897
	u64 new_wnd_end, new_snd_una, snd_nxt = READ_ONCE(msk->snd_nxt);
898
	struct sock *sk = (struct sock *)msk;
899 900 901
	u64 old_snd_una;

	mptcp_data_lock(sk);
902 903 904 905 906

	/* avoid ack expansion on update conflict, to reduce the risk of
	 * wrongly expanding to a future ack sequence number, which is way
	 * more dangerous than missing an ack
	 */
907
	old_snd_una = msk->snd_una;
908 909 910
	new_snd_una = expand_ack(old_snd_una, mp_opt->data_ack, mp_opt->ack64);

	/* ACK for data not even sent yet? Ignore. */
P
Paolo Abeni 已提交
911
	if (after64(new_snd_una, snd_nxt))
912 913
		new_snd_una = old_snd_una;

914 915
	new_wnd_end = new_snd_una + tcp_sk(ssk)->snd_wnd;

916
	if (after64(new_wnd_end, msk->wnd_end))
917
		msk->wnd_end = new_wnd_end;
918 919 920 921 922

	/* this assumes mptcp_incoming_options() is invoked after tcp_ack() */
	if (after64(msk->wnd_end, READ_ONCE(msk->snd_nxt)) &&
	    sk_stream_memory_free(ssk))
		__mptcp_check_push(sk, ssk);
923

924 925 926
	if (after64(new_snd_una, old_snd_una)) {
		msk->snd_una = new_snd_una;
		__mptcp_data_acked(sk);
927
	}
928
	mptcp_data_unlock(sk);
929 930
}

931
bool mptcp_update_rcv_data_fin(struct mptcp_sock *msk, u64 data_fin_seq, bool use_64bit)
932 933 934 935 936 937 938 939 940
{
	/* Skip if DATA_FIN was already received.
	 * If updating simultaneously with the recvmsg loop, values
	 * should match. If they mismatch, the peer is misbehaving and
	 * we will prefer the most recent information.
	 */
	if (READ_ONCE(msk->rcv_data_fin) || !READ_ONCE(msk->first))
		return false;

941 942
	WRITE_ONCE(msk->rcv_data_fin_seq,
		   expand_ack(READ_ONCE(msk->ack_seq), data_fin_seq, use_64bit));
943 944 945 946 947
	WRITE_ONCE(msk->rcv_data_fin, 1);

	return true;
}

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
static bool add_addr_hmac_valid(struct mptcp_sock *msk,
				struct mptcp_options_received *mp_opt)
{
	u64 hmac = 0;

	if (mp_opt->echo)
		return true;

	if (mp_opt->family == MPTCP_ADDR_IPVERSION_4)
		hmac = add_addr_generate_hmac(msk->remote_key,
					      msk->local_key,
					      mp_opt->addr_id, &mp_opt->addr);
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
	else
		hmac = add_addr6_generate_hmac(msk->remote_key,
					       msk->local_key,
					       mp_opt->addr_id, &mp_opt->addr6);
#endif

	pr_debug("msk=%p, ahmac=%llu, mp_opt->ahmac=%llu\n",
		 msk, (unsigned long long)hmac,
		 (unsigned long long)mp_opt->ahmac);

	return hmac == mp_opt->ahmac;
}

974
void mptcp_incoming_options(struct sock *sk, struct sk_buff *skb)
975
{
976
	struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk);
977
	struct mptcp_sock *msk = mptcp_sk(subflow->conn);
978
	struct mptcp_options_received mp_opt;
979 980
	struct mptcp_ext *mpext;

981 982 983 984 985 986 987
	if (__mptcp_check_fallback(msk)) {
		/* Keep it simple and unconditionally trigger send data cleanup and
		 * pending queue spooling. We will need to acquire the data lock
		 * for more accurate checks, and once the lock is acquired, such
		 * helpers are cheap.
		 */
		mptcp_data_lock(subflow->conn);
988 989
		if (sk_stream_memory_free(sk))
			__mptcp_check_push(subflow->conn, sk);
990 991
		__mptcp_data_acked(subflow->conn);
		mptcp_data_unlock(subflow->conn);
992
		return;
993
	}
994

995 996
	mptcp_get_options(skb, &mp_opt);
	if (!check_fully_established(msk, sk, subflow, skb, &mp_opt))
997
		return;
998

999 1000 1001 1002 1003 1004
	if (mp_opt.fastclose &&
	    msk->local_key == mp_opt.rcvr_key) {
		WRITE_ONCE(msk->rcv_fastclose, true);
		mptcp_schedule_work((struct sock *)msk);
	}

1005
	if (mp_opt.add_addr && add_addr_hmac_valid(msk, &mp_opt)) {
1006 1007
		struct mptcp_addr_info addr;

1008 1009 1010
		addr.port = htons(mp_opt.port);
		addr.id = mp_opt.addr_id;
		if (mp_opt.family == MPTCP_ADDR_IPVERSION_4) {
1011
			addr.family = AF_INET;
1012
			addr.addr = mp_opt.addr;
1013 1014
		}
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
1015
		else if (mp_opt.family == MPTCP_ADDR_IPVERSION_6) {
1016
			addr.family = AF_INET6;
1017
			addr.addr6 = mp_opt.addr6;
1018 1019
		}
#endif
G
Geliang Tang 已提交
1020
		if (!mp_opt.echo) {
1021
			mptcp_pm_add_addr_received(msk, &addr);
G
Geliang Tang 已提交
1022 1023
			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ADDADDR);
		} else {
1024
			mptcp_pm_del_add_timer(msk, &addr);
G
Geliang Tang 已提交
1025 1026
			MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ECHOADD);
		}
1027
		mp_opt.add_addr = 0;
1028 1029
	}

1030 1031 1032 1033 1034
	if (mp_opt.rm_addr) {
		mptcp_pm_rm_addr_received(msk, mp_opt.rm_id);
		mp_opt.rm_addr = 0;
	}

1035 1036
	if (mp_opt.mp_prio) {
		mptcp_pm_mp_prio_received(sk, mp_opt.backup);
G
Geliang Tang 已提交
1037
		MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_MPPRIORX);
1038 1039 1040
		mp_opt.mp_prio = 0;
	}

1041
	if (!mp_opt.dss)
1042 1043
		return;

1044 1045 1046
	/* we can't wait for recvmsg() to update the ack_seq, otherwise
	 * monodirectional flows will stuck
	 */
1047
	if (mp_opt.use_ack)
1048
		ack_update_msk(msk, sk, &mp_opt);
1049

1050 1051 1052 1053
	/* Zero-data-length packets are dropped by the caller and not
	 * propagated to the MPTCP layer, so the skb extension does not
	 * need to be allocated or populated. DATA_FIN information, if
	 * present, needs to be updated here before the skb is freed.
1054 1055 1056
	 */
	if (TCP_SKB_CB(skb)->seq == TCP_SKB_CB(skb)->end_seq) {
		if (mp_opt.data_fin && mp_opt.data_len == 1 &&
1057
		    mptcp_update_rcv_data_fin(msk, mp_opt.data_seq, mp_opt.dsn64) &&
1058 1059
		    schedule_work(&msk->work))
			sock_hold(subflow->conn);
1060 1061

		return;
1062 1063
	}

1064 1065 1066 1067 1068 1069
	mpext = skb_ext_add(skb, SKB_EXT_MPTCP);
	if (!mpext)
		return;

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

1070 1071
	if (mp_opt.use_map) {
		if (mp_opt.mpc_map) {
1072 1073 1074 1075 1076 1077 1078 1079 1080
			/* this is an MP_CAPABLE carrying MPTCP data
			 * we know this map the first chunk of data
			 */
			mptcp_crypto_key_sha(subflow->remote_key, NULL,
					     &mpext->data_seq);
			mpext->data_seq++;
			mpext->subflow_seq = 1;
			mpext->dsn64 = 1;
			mpext->mpc_map = 1;
1081
			mpext->data_fin = 0;
1082
		} else {
1083 1084 1085 1086
			mpext->data_seq = mp_opt.data_seq;
			mpext->subflow_seq = mp_opt.subflow_seq;
			mpext->dsn64 = mp_opt.dsn64;
			mpext->data_fin = mp_opt.data_fin;
1087
		}
1088
		mpext->data_len = mp_opt.data_len;
1089 1090 1091 1092
		mpext->use_map = 1;
	}
}

1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110
static void mptcp_set_rwin(const struct tcp_sock *tp)
{
	const struct sock *ssk = (const struct sock *)tp;
	const struct mptcp_subflow_context *subflow;
	struct mptcp_sock *msk;
	u64 ack_seq;

	subflow = mptcp_subflow_ctx(ssk);
	msk = mptcp_sk(subflow->conn);

	ack_seq = READ_ONCE(msk->ack_seq) + tp->rcv_wnd;

	if (after64(ack_seq, READ_ONCE(msk->rcv_wnd_sent)))
		WRITE_ONCE(msk->rcv_wnd_sent, ack_seq);
}

void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
			 struct mptcp_out_options *opts)
P
Peter Krystad 已提交
1111
{
1112
	if ((OPTION_MPTCP_MPC_SYN | OPTION_MPTCP_MPC_SYNACK |
P
Peter Krystad 已提交
1113 1114 1115 1116 1117
	     OPTION_MPTCP_MPC_ACK) & opts->suboptions) {
		u8 len;

		if (OPTION_MPTCP_MPC_SYN & opts->suboptions)
			len = TCPOLEN_MPTCP_MPC_SYN;
1118 1119
		else if (OPTION_MPTCP_MPC_SYNACK & opts->suboptions)
			len = TCPOLEN_MPTCP_MPC_SYNACK;
1120 1121
		else if (opts->ext_copy.data_len)
			len = TCPOLEN_MPTCP_MPC_ACK_DATA;
P
Peter Krystad 已提交
1122 1123 1124
		else
			len = TCPOLEN_MPTCP_MPC_ACK;

P
Peter Krystad 已提交
1125 1126 1127
		*ptr++ = mptcp_option(MPTCPOPT_MP_CAPABLE, len,
				      MPTCP_SUPPORTED_VERSION,
				      MPTCP_CAP_HMAC_SHA256);
1128 1129 1130 1131 1132

		if (!((OPTION_MPTCP_MPC_SYNACK | OPTION_MPTCP_MPC_ACK) &
		    opts->suboptions))
			goto mp_capable_done;

P
Peter Krystad 已提交
1133 1134
		put_unaligned_be64(opts->sndr_key, ptr);
		ptr += 2;
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145
		if (!((OPTION_MPTCP_MPC_ACK) & opts->suboptions))
			goto mp_capable_done;

		put_unaligned_be64(opts->rcvr_key, ptr);
		ptr += 2;
		if (!opts->ext_copy.data_len)
			goto mp_capable_done;

		put_unaligned_be32(opts->ext_copy.data_len << 16 |
				   TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
		ptr += 1;
P
Peter Krystad 已提交
1146
	}
1147

1148
mp_capable_done:
1149 1150 1151 1152 1153
	if ((OPTION_MPTCP_ADD_ADDR
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
	     | OPTION_MPTCP_ADD_ADDR6
#endif
	    ) & opts->suboptions) {
1154 1155 1156
		u8 len = TCPOLEN_MPTCP_ADD_ADDR_BASE;
		u8 echo = MPTCP_ADDR_ECHO;

1157 1158 1159 1160 1161
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
		if (OPTION_MPTCP_ADD_ADDR6 & opts->suboptions)
			len = TCPOLEN_MPTCP_ADD_ADDR6_BASE;
#endif

1162 1163 1164
		if (opts->port)
			len += TCPOLEN_MPTCP_PORT_LEN;

P
Peter Krystad 已提交
1165
		if (opts->ahmac) {
1166 1167
			len += sizeof(opts->ahmac);
			echo = 0;
P
Peter Krystad 已提交
1168 1169
		}

1170 1171
		*ptr++ = mptcp_option(MPTCPOPT_ADD_ADDR,
				      len, echo, opts->addr_id);
1172 1173 1174
		if (OPTION_MPTCP_ADD_ADDR & opts->suboptions) {
			memcpy((u8 *)ptr, (u8 *)&opts->addr.s_addr, 4);
			ptr += 1;
P
Peter Krystad 已提交
1175 1176
		}
#if IS_ENABLED(CONFIG_MPTCP_IPV6)
1177 1178 1179
		else if (OPTION_MPTCP_ADD_ADDR6 & opts->suboptions) {
			memcpy((u8 *)ptr, opts->addr6.s6_addr, 16);
			ptr += 4;
P
Peter Krystad 已提交
1180 1181 1182
		}
#endif

1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205
		if (!opts->port) {
			if (opts->ahmac) {
				put_unaligned_be64(opts->ahmac, ptr);
				ptr += 2;
			}
		} else {
			if (opts->ahmac) {
				u8 *bptr = (u8 *)ptr;

				put_unaligned_be16(opts->port, bptr);
				bptr += 2;
				put_unaligned_be64(opts->ahmac, bptr);
				bptr += 8;
				put_unaligned_be16(TCPOPT_NOP << 8 |
						   TCPOPT_NOP, bptr);

				ptr += 3;
			} else {
				put_unaligned_be32(opts->port << 16 |
						   TCPOPT_NOP << 8 |
						   TCPOPT_NOP, ptr);
				ptr += 1;
			}
P
Peter Krystad 已提交
1206 1207 1208 1209 1210 1211 1212 1213 1214
		}
	}

	if (OPTION_MPTCP_RM_ADDR & opts->suboptions) {
		*ptr++ = mptcp_option(MPTCPOPT_RM_ADDR,
				      TCPOLEN_MPTCP_RM_ADDR_BASE,
				      0, opts->rm_id);
	}

1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
	if (OPTION_MPTCP_PRIO & opts->suboptions) {
		const struct sock *ssk = (const struct sock *)tp;
		struct mptcp_subflow_context *subflow;

		subflow = mptcp_subflow_ctx(ssk);
		subflow->send_mp_prio = 0;

		*ptr++ = mptcp_option(MPTCPOPT_MP_PRIO,
				      TCPOLEN_MPTCP_PRIO,
				      opts->backup, TCPOPT_NOP);
	}

1227 1228 1229 1230 1231 1232 1233 1234 1235 1236
	if (OPTION_MPTCP_MPJ_SYN & opts->suboptions) {
		*ptr++ = mptcp_option(MPTCPOPT_MP_JOIN,
				      TCPOLEN_MPTCP_MPJ_SYN,
				      opts->backup, opts->join_id);
		put_unaligned_be32(opts->token, ptr);
		ptr += 1;
		put_unaligned_be32(opts->nonce, ptr);
		ptr += 1;
	}

1237 1238 1239 1240 1241 1242 1243 1244 1245 1246
	if (OPTION_MPTCP_MPJ_SYNACK & opts->suboptions) {
		*ptr++ = mptcp_option(MPTCPOPT_MP_JOIN,
				      TCPOLEN_MPTCP_MPJ_SYNACK,
				      opts->backup, opts->join_id);
		put_unaligned_be64(opts->thmac, ptr);
		ptr += 2;
		put_unaligned_be32(opts->nonce, ptr);
		ptr += 1;
	}

1247 1248 1249 1250 1251 1252 1253
	if (OPTION_MPTCP_MPJ_ACK & opts->suboptions) {
		*ptr++ = mptcp_option(MPTCPOPT_MP_JOIN,
				      TCPOLEN_MPTCP_MPJ_ACK, 0, 0);
		memcpy(ptr, opts->hmac, MPTCPOPT_HMAC_LEN);
		ptr += 5;
	}

1254 1255 1256 1257 1258 1259
	if (opts->ext_copy.use_ack || opts->ext_copy.use_map) {
		struct mptcp_ext *mpext = &opts->ext_copy;
		u8 len = TCPOLEN_MPTCP_DSS_BASE;
		u8 flags = 0;

		if (mpext->use_ack) {
1260 1261 1262 1263 1264 1265 1266
			flags = MPTCP_DSS_HAS_ACK;
			if (mpext->ack64) {
				len += TCPOLEN_MPTCP_DSS_ACK64;
				flags |= MPTCP_DSS_ACK64;
			} else {
				len += TCPOLEN_MPTCP_DSS_ACK32;
			}
1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279
		}

		if (mpext->use_map) {
			len += TCPOLEN_MPTCP_DSS_MAP64;

			/* Use only 64-bit mapping flags for now, add
			 * support for optional 32-bit mappings later.
			 */
			flags |= MPTCP_DSS_HAS_MAP | MPTCP_DSS_DSN64;
			if (mpext->data_fin)
				flags |= MPTCP_DSS_DATA_FIN;
		}

P
Peter Krystad 已提交
1280
		*ptr++ = mptcp_option(MPTCPOPT_DSS, len, 0, flags);
1281 1282

		if (mpext->use_ack) {
1283 1284 1285 1286 1287 1288 1289
			if (mpext->ack64) {
				put_unaligned_be64(mpext->data_ack, ptr);
				ptr += 2;
			} else {
				put_unaligned_be32(mpext->data_ack32, ptr);
				ptr += 1;
			}
1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300
		}

		if (mpext->use_map) {
			put_unaligned_be64(mpext->data_seq, ptr);
			ptr += 2;
			put_unaligned_be32(mpext->subflow_seq, ptr);
			ptr += 1;
			put_unaligned_be32(mpext->data_len << 16 |
					   TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
		}
	}
1301 1302 1303

	if (tp)
		mptcp_set_rwin(tp);
P
Peter Krystad 已提交
1304
}