input.c 21.6 KB
Newer Older
1 2
/*
 *  net/dccp/input.c
3
 *
4 5 6 7 8 9 10 11 12 13 14 15 16 17
 *  An implementation of the DCCP protocol
 *  Arnaldo Carvalho de Melo <acme@conectiva.com.br>
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 */

#include <linux/dccp.h>
#include <linux/skbuff.h>

#include <net/sock.h>

18
#include "ackvec.h"
19 20 21
#include "ccid.h"
#include "dccp.h"

22 23 24
/* rate-limit for syncs in reply to sequence-invalid packets; RFC 4340, 7.5.4 */
int sysctl_dccp_sync_ratelimit	__read_mostly = HZ / 8;

25
static void dccp_enqueue_skb(struct sock *sk, struct sk_buff *skb)
26 27 28 29 30 31 32
{
	__skb_pull(skb, dccp_hdr(skb)->dccph_doff * 4);
	__skb_queue_tail(&sk->sk_receive_queue, skb);
	skb_set_owner_r(skb, sk);
	sk->sk_data_ready(sk, 0);
}

33 34 35 36 37 38 39 40 41 42 43 44 45
static void dccp_fin(struct sock *sk, struct sk_buff *skb)
{
	/*
	 * On receiving Close/CloseReq, both RD/WR shutdown are performed.
	 * RFC 4340, 8.3 says that we MAY send further Data/DataAcks after
	 * receiving the closing segment, but there is no guarantee that such
	 * data will be processed at all.
	 */
	sk->sk_shutdown = SHUTDOWN_MASK;
	sock_set_flag(sk, SOCK_DONE);
	dccp_enqueue_skb(sk, skb);
}

46
static int dccp_rcv_close(struct sock *sk, struct sk_buff *skb)
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
	int queued = 0;

	switch (sk->sk_state) {
	/*
	 * We ignore Close when received in one of the following states:
	 *  - CLOSED		(may be a late or duplicate packet)
	 *  - PASSIVE_CLOSEREQ	(the peer has sent a CloseReq earlier)
	 *  - RESPOND		(already handled by dccp_check_req)
	 */
	case DCCP_CLOSING:
		/*
		 * Simultaneous-close: receiving a Close after sending one. This
		 * can happen if both client and server perform active-close and
		 * will result in an endless ping-pong of crossing and retrans-
		 * mitted Close packets, which only terminates when one of the
		 * nodes times out (min. 64 seconds). Quicker convergence can be
		 * achieved when one of the nodes acts as tie-breaker.
		 * This is ok as both ends are done with data transfer and each
		 * end is just waiting for the other to acknowledge termination.
		 */
		if (dccp_sk(sk)->dccps_role != DCCP_ROLE_CLIENT)
			break;
		/* fall through */
	case DCCP_REQUESTING:
	case DCCP_ACTIVE_CLOSEREQ:
		dccp_send_reset(sk, DCCP_RESET_CODE_CLOSED);
		dccp_done(sk);
		break;
	case DCCP_OPEN:
	case DCCP_PARTOPEN:
		/* Give waiting application a chance to read pending data */
		queued = 1;
		dccp_fin(sk, skb);
		dccp_set_state(sk, DCCP_PASSIVE_CLOSE);
		/* fall through */
	case DCCP_PASSIVE_CLOSE:
		/*
		 * Retransmitted Close: we have already enqueued the first one.
		 */
		sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
	}
	return queued;
90 91
}

92
static int dccp_rcv_closereq(struct sock *sk, struct sk_buff *skb)
93
{
94 95
	int queued = 0;

96 97 98 99 100 101 102
	/*
	 *   Step 7: Check for unexpected packet types
	 *      If (S.is_server and P.type == CloseReq)
	 *	  Send Sync packet acknowledging P.seqno
	 *	  Drop packet and return
	 */
	if (dccp_sk(sk)->dccps_role != DCCP_ROLE_CLIENT) {
103
		dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq, DCCP_PKT_SYNC);
104
		return queued;
105 106
	}

107 108 109 110
	/* Step 13: process relevant Client states < CLOSEREQ */
	switch (sk->sk_state) {
	case DCCP_REQUESTING:
		dccp_send_close(sk, 0);
111
		dccp_set_state(sk, DCCP_CLOSING);
112 113 114 115 116 117 118 119 120 121 122 123
		break;
	case DCCP_OPEN:
	case DCCP_PARTOPEN:
		/* Give waiting application a chance to read pending data */
		queued = 1;
		dccp_fin(sk, skb);
		dccp_set_state(sk, DCCP_PASSIVE_CLOSEREQ);
		/* fall through */
	case DCCP_PASSIVE_CLOSEREQ:
		sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_HUP);
	}
	return queued;
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
static u8 dccp_reset_code_convert(const u8 code)
{
	const u8 error_code[] = {
	[DCCP_RESET_CODE_CLOSED]	     = 0,	/* normal termination */
	[DCCP_RESET_CODE_UNSPECIFIED]	     = 0,	/* nothing known */
	[DCCP_RESET_CODE_ABORTED]	     = ECONNRESET,

	[DCCP_RESET_CODE_NO_CONNECTION]	     = ECONNREFUSED,
	[DCCP_RESET_CODE_CONNECTION_REFUSED] = ECONNREFUSED,
	[DCCP_RESET_CODE_TOO_BUSY]	     = EUSERS,
	[DCCP_RESET_CODE_AGGRESSION_PENALTY] = EDQUOT,

	[DCCP_RESET_CODE_PACKET_ERROR]	     = ENOMSG,
	[DCCP_RESET_CODE_BAD_INIT_COOKIE]    = EBADR,
	[DCCP_RESET_CODE_BAD_SERVICE_CODE]   = EBADRQC,
	[DCCP_RESET_CODE_OPTION_ERROR]	     = EILSEQ,
	[DCCP_RESET_CODE_MANDATORY_ERROR]    = EOPNOTSUPP,
	};

	return code >= DCCP_MAX_RESET_CODES ? 0 : error_code[code];
}

static void dccp_rcv_reset(struct sock *sk, struct sk_buff *skb)
{
	u8 err = dccp_reset_code_convert(dccp_hdr_reset(skb)->dccph_reset_code);

	sk->sk_err = err;

	/* Queue the equivalent of TCP fin so that dccp_recvmsg exits the loop */
	dccp_fin(sk, skb);

	if (err && !sock_flag(sk, SOCK_DEAD))
158
		sk_wake_async(sk, SOCK_WAKE_IO, POLL_ERR);
159 160 161
	dccp_time_wait(sk, DCCP_TIME_WAIT, 0);
}

162
static void dccp_event_ack_recv(struct sock *sk, struct sk_buff *skb)
163 164 165
{
	struct dccp_sock *dp = dccp_sk(sk);

166
	if (dp->dccps_hc_rx_ackvec != NULL)
167 168
		dccp_ackvec_check_rcv_ackno(dp->dccps_hc_rx_ackvec, sk,
					    DCCP_SKB_CB(skb)->dccpd_ack_seq);
169 170
}

171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
static void dccp_deliver_input_to_ccids(struct sock *sk, struct sk_buff *skb)
{
	const struct dccp_sock *dp = dccp_sk(sk);

	/* Don't deliver to RX CCID when node has shut down read end. */
	if (!(sk->sk_shutdown & RCV_SHUTDOWN))
		ccid_hc_rx_packet_recv(dp->dccps_hc_rx_ccid, sk, skb);
	/*
	 * Until the TX queue has been drained, we can not honour SHUT_WR, since
	 * we need received feedback as input to adjust congestion control.
	 */
	if (sk->sk_write_queue.qlen > 0 || !(sk->sk_shutdown & SEND_SHUTDOWN))
		ccid_hc_tx_packet_recv(dp->dccps_hc_tx_ccid, sk, skb);
}

186 187 188 189
static int dccp_check_seqno(struct sock *sk, struct sk_buff *skb)
{
	const struct dccp_hdr *dh = dccp_hdr(skb);
	struct dccp_sock *dp = dccp_sk(sk);
190 191
	u64 lswl, lawl, seqno = DCCP_SKB_CB(skb)->dccpd_seq,
			ackno = DCCP_SKB_CB(skb)->dccpd_ack_seq;
192 193 194 195 196 197 198 199 200 201 202 203 204

	/*
	 *   Step 5: Prepare sequence numbers for Sync
	 *     If P.type == Sync or P.type == SyncAck,
	 *	  If S.AWL <= P.ackno <= S.AWH and P.seqno >= S.SWL,
	 *	     / * P is valid, so update sequence number variables
	 *		 accordingly.  After this update, P will pass the tests
	 *		 in Step 6.  A SyncAck is generated if necessary in
	 *		 Step 15 * /
	 *	     Update S.GSR, S.SWL, S.SWH
	 *	  Otherwise,
	 *	     Drop packet and return
	 */
205
	if (dh->dccph_type == DCCP_PKT_SYNC ||
206
	    dh->dccph_type == DCCP_PKT_SYNCACK) {
207 208 209
		if (between48(ackno, dp->dccps_awl, dp->dccps_awh) &&
		    dccp_delta_seqno(dp->dccps_swl, seqno) >= 0)
			dccp_update_gsr(sk, seqno);
210 211
		else
			return -1;
212
	}
213

214 215 216 217 218 219 220 221 222 223 224
	/*
	 *   Step 6: Check sequence numbers
	 *      Let LSWL = S.SWL and LAWL = S.AWL
	 *      If P.type == CloseReq or P.type == Close or P.type == Reset,
	 *	  LSWL := S.GSR + 1, LAWL := S.GAR
	 *      If LSWL <= P.seqno <= S.SWH
	 *	     and (P.ackno does not exist or LAWL <= P.ackno <= S.AWH),
	 *	  Update S.GSR, S.SWL, S.SWH
	 *	  If P.type != Sync,
	 *	     Update S.GAR
	 */
225 226 227 228
	lswl = dp->dccps_swl;
	lawl = dp->dccps_awl;

	if (dh->dccph_type == DCCP_PKT_CLOSEREQ ||
229 230
	    dh->dccph_type == DCCP_PKT_CLOSE ||
	    dh->dccph_type == DCCP_PKT_RESET) {
231
		lswl = ADD48(dp->dccps_gsr, 1);
232 233 234
		lawl = dp->dccps_gar;
	}

235 236 237 238
	if (between48(seqno, lswl, dp->dccps_swh) &&
	    (ackno == DCCP_PKT_WITHOUT_ACK_SEQ ||
	     between48(ackno, lawl, dp->dccps_awh))) {
		dccp_update_gsr(sk, seqno);
239 240

		if (dh->dccph_type != DCCP_PKT_SYNC &&
241 242
		    (ackno != DCCP_PKT_WITHOUT_ACK_SEQ))
			dp->dccps_gar = ackno;
243
	} else {
G
Gerrit Renker 已提交
244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260
		unsigned long now = jiffies;
		/*
		 *   Step 6: Check sequence numbers
		 *      Otherwise,
		 *         If P.type == Reset,
		 *            Send Sync packet acknowledging S.GSR
		 *         Otherwise,
		 *            Send Sync packet acknowledging P.seqno
		 *      Drop packet and return
		 *
		 *   These Syncs are rate-limited as per RFC 4340, 7.5.4:
		 *   at most 1 / (dccp_sync_rate_limit * HZ) Syncs per second.
		 */
		if (time_before(now, (dp->dccps_rate_last +
				      sysctl_dccp_sync_ratelimit)))
			return 0;

261 262 263 264
		DCCP_WARN("DCCP: Step 6 failed for %s packet, "
			  "(LSWL(%llu) <= P.seqno(%llu) <= S.SWH(%llu)) and "
			  "(P.ackno %s or LAWL(%llu) <= P.ackno(%llu) <= S.AWH(%llu), "
			  "sending SYNC...\n",  dccp_packet_name(dh->dccph_type),
265
			  (unsigned long long) lswl, (unsigned long long) seqno,
266
			  (unsigned long long) dp->dccps_swh,
267 268 269
			  (ackno == DCCP_PKT_WITHOUT_ACK_SEQ) ? "doesn't exist"
							      : "exists",
			  (unsigned long long) lawl, (unsigned long long) ackno,
270
			  (unsigned long long) dp->dccps_awh);
G
Gerrit Renker 已提交
271 272 273

		dp->dccps_rate_last = now;

G
Gerrit Renker 已提交
274 275
		if (dh->dccph_type == DCCP_PKT_RESET)
			seqno = dp->dccps_gsr;
276
		dccp_send_sync(sk, seqno, DCCP_PKT_SYNC);
277 278 279 280 281 282
		return -1;
	}

	return 0;
}

283 284
static int __dccp_rcv_established(struct sock *sk, struct sk_buff *skb,
				  const struct dccp_hdr *dh, const unsigned len)
285 286 287 288 289 290 291
{
	struct dccp_sock *dp = dccp_sk(sk);

	switch (dccp_hdr(skb)->dccph_type) {
	case DCCP_PKT_DATAACK:
	case DCCP_PKT_DATA:
		/*
292 293 294
		 * FIXME: schedule DATA_DROPPED (RFC 4340, 11.7.2) if and when
		 * - sk_shutdown == RCV_SHUTDOWN, use Code 1, "Not Listening"
		 * - sk_receive_queue is full, use Code 2, "Receive Buffer"
295
		 */
296
		dccp_enqueue_skb(sk, skb);
297 298 299 300 301 302 303 304 305 306 307
		return 0;
	case DCCP_PKT_ACK:
		goto discard;
	case DCCP_PKT_RESET:
		/*
		 *  Step 9: Process Reset
		 *	If P.type == Reset,
		 *		Tear down connection
		 *		S.state := TIMEWAIT
		 *		Set TIMEWAIT timer
		 *		Drop packet and return
308 309
		 */
		dccp_rcv_reset(sk, skb);
310 311
		return 0;
	case DCCP_PKT_CLOSEREQ:
312 313
		if (dccp_rcv_closereq(sk, skb))
			return 0;
314 315
		goto discard;
	case DCCP_PKT_CLOSE:
316 317 318
		if (dccp_rcv_close(sk, skb))
			return 0;
		goto discard;
319
	case DCCP_PKT_REQUEST:
320 321
		/* Step 7
		 *   or (S.is_server and P.type == Response)
322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337
		 *   or (S.is_client and P.type == Request)
		 *   or (S.state >= OPEN and P.type == Request
		 *	and P.seqno >= S.OSR)
		 *    or (S.state >= OPEN and P.type == Response
		 *	and P.seqno >= S.OSR)
		 *    or (S.state == RESPOND and P.type == Data),
		 *  Send Sync packet acknowledging P.seqno
		 *  Drop packet and return
		 */
		if (dp->dccps_role != DCCP_ROLE_LISTEN)
			goto send_sync;
		goto check_seq;
	case DCCP_PKT_RESPONSE:
		if (dp->dccps_role != DCCP_ROLE_CLIENT)
			goto send_sync;
check_seq:
338 339
		if (dccp_delta_seqno(dp->dccps_osr,
				     DCCP_SKB_CB(skb)->dccpd_seq) >= 0) {
340
send_sync:
341 342
			dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq,
				       DCCP_PKT_SYNC);
343 344
		}
		break;
345 346 347 348
	case DCCP_PKT_SYNC:
		dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq,
			       DCCP_PKT_SYNCACK);
		/*
349
		 * From RFC 4340, sec. 5.7
350 351 352
		 *
		 * As with DCCP-Ack packets, DCCP-Sync and DCCP-SyncAck packets
		 * MAY have non-zero-length application data areas, whose
353
		 * contents receivers MUST ignore.
354 355
		 */
		goto discard;
356 357 358 359 360 361 362 363
	}

	DCCP_INC_STATS_BH(DCCP_MIB_INERRS);
discard:
	__kfree_skb(skb);
	return 0;
}

364 365 366 367 368 369 370 371
int dccp_rcv_established(struct sock *sk, struct sk_buff *skb,
			 const struct dccp_hdr *dh, const unsigned len)
{
	struct dccp_sock *dp = dccp_sk(sk);

	if (dccp_check_seqno(sk, skb))
		goto discard;

372
	if (dccp_parse_options(sk, NULL, skb))
373
		return 1;
374 375 376 377

	if (DCCP_SKB_CB(skb)->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ)
		dccp_event_ack_recv(sk, skb);

378
	if (dp->dccps_hc_rx_ackvec != NULL &&
379 380 381 382
	    dccp_ackvec_add(dp->dccps_hc_rx_ackvec, sk,
			    DCCP_SKB_CB(skb)->dccpd_seq,
			    DCCP_ACKVEC_STATE_RECEIVED))
		goto discard;
383
	dccp_deliver_input_to_ccids(sk, skb);
384 385 386 387 388 389 390

	return __dccp_rcv_established(sk, skb, dh, len);
discard:
	__kfree_skb(skb);
	return 0;
}

391 392
EXPORT_SYMBOL_GPL(dccp_rcv_established);

393 394 395 396 397
static int dccp_rcv_request_sent_state_process(struct sock *sk,
					       struct sk_buff *skb,
					       const struct dccp_hdr *dh,
					       const unsigned len)
{
398
	/*
399 400 401 402 403 404 405 406 407 408 409 410 411
	 *  Step 4: Prepare sequence numbers in REQUEST
	 *     If S.state == REQUEST,
	 *	  If (P.type == Response or P.type == Reset)
	 *		and S.AWL <= P.ackno <= S.AWH,
	 *	     / * Set sequence number variables corresponding to the
	 *		other endpoint, so P will pass the tests in Step 6 * /
	 *	     Set S.GSR, S.ISR, S.SWL, S.SWH
	 *	     / * Response processing continues in Step 10; Reset
	 *		processing continues in Step 9 * /
	*/
	if (dh->dccph_type == DCCP_PKT_RESPONSE) {
		const struct inet_connection_sock *icsk = inet_csk(sk);
		struct dccp_sock *dp = dccp_sk(sk);
412
		long tstamp = dccp_timestamp();
413

414 415 416 417 418 419 420
		if (!between48(DCCP_SKB_CB(skb)->dccpd_ack_seq,
			       dp->dccps_awl, dp->dccps_awh)) {
			dccp_pr_debug("invalid ackno: S.AWL=%llu, "
				      "P.ackno=%llu, S.AWH=%llu \n",
				      (unsigned long long)dp->dccps_awl,
			   (unsigned long long)DCCP_SKB_CB(skb)->dccpd_ack_seq,
				      (unsigned long long)dp->dccps_awh);
421 422
			goto out_invalid_packet;
		}
423

424 425 426 427 428
		/*
		 * If option processing (Step 8) failed, return 1 here so that
		 * dccp_v4_do_rcv() sends a Reset. The Reset code depends on
		 * the option type and is set in dccp_parse_options().
		 */
429
		if (dccp_parse_options(sk, NULL, skb))
430
			return 1;
431

432 433 434 435
		/* Obtain usec RTT sample from SYN exchange (used by CCID 3) */
		if (likely(dp->dccps_options_received.dccpor_timestamp_echo))
			dp->dccps_syn_rtt = dccp_sample_rtt(sk, 10 * (tstamp -
			    dp->dccps_options_received.dccpor_timestamp_echo));
436

437 438 439 440 441 442
		/* Stop the REQUEST timer */
		inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS);
		WARN_ON(sk->sk_send_head == NULL);
		kfree_skb(sk->sk_send_head);
		sk->sk_send_head = NULL;

443
		/*
444 445 446 447 448
		 * Set ISR, GSR from packet. ISS was set in dccp_v{4,6}_connect
		 * and GSS in dccp_transmit_skb(). Setting AWL/AWH and SWL/SWH
		 * is done as part of activating the feature values below, since
		 * these settings depend on the local/remote Sequence Window
		 * features, which were undefined or not confirmed until now.
449
		 */
450
		dp->dccps_gsr = dp->dccps_isr = DCCP_SKB_CB(skb)->dccpd_seq;
451

452
		dccp_sync_mss(sk, icsk->icsk_pmtu_cookie);
453 454 455 456

		/*
		 *    Step 10: Process REQUEST state (second part)
		 *       If S.state == REQUEST,
457 458 459 460 461 462
		 *	  / * If we get here, P is a valid Response from the
		 *	      server (see Step 4), and we should move to
		 *	      PARTOPEN state. PARTOPEN means send an Ack,
		 *	      don't send Data packets, retransmit Acks
		 *	      periodically, and always include any Init Cookie
		 *	      from the Response * /
463 464
		 *	  S.state := PARTOPEN
		 *	  Set PARTOPEN timer
465
		 *	  Continue with S.state == PARTOPEN
466 467
		 *	  / * Step 12 will send the Ack completing the
		 *	      three-way handshake * /
468 469 470
		 */
		dccp_set_state(sk, DCCP_PARTOPEN);

471 472 473 474 475 476 477 478 479
		/*
		 * If feature negotiation was successful, activate features now;
		 * an activation failure means that this host could not activate
		 * one ore more features (e.g. insufficient memory), which would
		 * leave at least one feature in an undefined state.
		 */
		if (dccp_feat_activate_values(sk, &dp->dccps_featneg))
			goto unable_to_proceed;

480
		/* Make sure socket is routed, for correct metrics. */
481
		icsk->icsk_af_ops->rebuild_header(sk);
482 483 484

		if (!sock_flag(sk, SOCK_DEAD)) {
			sk->sk_state_change(sk);
485
			sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT);
486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
		}

		if (sk->sk_write_pending || icsk->icsk_ack.pingpong ||
		    icsk->icsk_accept_queue.rskq_defer_accept) {
			/* Save one ACK. Data will be ready after
			 * several ticks, if write_pending is set.
			 *
			 * It may be deleted, but with this feature tcpdumps
			 * look so _wonderfully_ clever, that I was not able
			 * to stand against the temptation 8)     --ANK
			 */
			/*
			 * OK, in DCCP we can as well do a similar trick, its
			 * even in the draft, but there is no need for us to
			 * schedule an ack here, as dccp_sendmsg does this for
			 * us, also stated in the draft. -acme
			 */
			__kfree_skb(skb);
			return 0;
505
		}
506 507 508 509 510
		dccp_send_ack(sk);
		return -1;
	}

out_invalid_packet:
511 512
	/* dccp_v4_do_rcv will send a reset */
	DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_PACKET_ERROR;
513
	return 1;
514 515 516 517 518 519 520 521 522 523

unable_to_proceed:
	DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_ABORTED;
	/*
	 * We mark this socket as no longer usable, so that the loop in
	 * dccp_sendmsg() terminates and the application gets notified.
	 */
	dccp_set_state(sk, DCCP_CLOSED);
	sk->sk_err = ECOMM;
	return 1;
524 525 526 527 528 529 530 531 532 533 534 535 536
}

static int dccp_rcv_respond_partopen_state_process(struct sock *sk,
						   struct sk_buff *skb,
						   const struct dccp_hdr *dh,
						   const unsigned len)
{
	int queued = 0;

	switch (dh->dccph_type) {
	case DCCP_PKT_RESET:
		inet_csk_clear_xmit_timer(sk, ICSK_TIME_DACK);
		break;
537 538 539
	case DCCP_PKT_DATA:
		if (sk->sk_state == DCCP_RESPOND)
			break;
540 541 542
	case DCCP_PKT_DATAACK:
	case DCCP_PKT_ACK:
		/*
543 544 545 546 547 548
		 * FIXME: we should be reseting the PARTOPEN (DELACK) timer
		 * here but only if we haven't used the DELACK timer for
		 * something else, like sending a delayed ack for a TIMESTAMP
		 * echo, etc, for now were not clearing it, sending an extra
		 * ACK when there is nothing else to do in DELACK is not a big
		 * deal after all.
549 550 551 552 553 554 555 556 557
		 */

		/* Stop the PARTOPEN timer */
		if (sk->sk_state == DCCP_PARTOPEN)
			inet_csk_clear_xmit_timer(sk, ICSK_TIME_DACK);

		dccp_sk(sk)->dccps_osr = DCCP_SKB_CB(skb)->dccpd_seq;
		dccp_set_state(sk, DCCP_OPEN);

558 559
		if (dh->dccph_type == DCCP_PKT_DATAACK ||
		    dh->dccph_type == DCCP_PKT_DATA) {
560
			__dccp_rcv_established(sk, skb, dh, len);
561
			queued = 1; /* packet was queued
562
				       (by __dccp_rcv_established) */
563 564 565 566 567 568 569 570 571 572 573
		}
		break;
	}

	return queued;
}

int dccp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
			   struct dccp_hdr *dh, unsigned len)
{
	struct dccp_sock *dp = dccp_sk(sk);
574
	struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
575 576 577
	const int old_state = sk->sk_state;
	int queued = 0;

578 579 580 581
	/*
	 *  Step 3: Process LISTEN state
	 *
	 *     If S.state == LISTEN,
582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598
	 *	 If P.type == Request or P contains a valid Init Cookie option,
	 *	      (* Must scan the packet's options to check for Init
	 *		 Cookies.  Only Init Cookies are processed here,
	 *		 however; other options are processed in Step 8.  This
	 *		 scan need only be performed if the endpoint uses Init
	 *		 Cookies *)
	 *	      (* Generate a new socket and switch to that socket *)
	 *	      Set S := new socket for this port pair
	 *	      S.state = RESPOND
	 *	      Choose S.ISS (initial seqno) or set from Init Cookies
	 *	      Initialize S.GAR := S.ISS
	 *	      Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init
	 *	      Cookies Continue with S.state == RESPOND
	 *	      (* A Response packet will be generated in Step 11 *)
	 *	 Otherwise,
	 *	      Generate Reset(No Connection) unless P.type == Reset
	 *	      Drop packet and return
599 600 601
	 */
	if (sk->sk_state == DCCP_LISTEN) {
		if (dh->dccph_type == DCCP_PKT_REQUEST) {
602 603
			if (inet_csk(sk)->icsk_af_ops->conn_request(sk,
								    skb) < 0)
604 605 606 607 608 609
				return 1;
			goto discard;
		}
		if (dh->dccph_type == DCCP_PKT_RESET)
			goto discard;

610 611
		/* Caller (dccp_v4_do_rcv) will send Reset */
		dcb->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION;
612 613 614
		return 1;
	}

615
	if (sk->sk_state != DCCP_REQUESTING && sk->sk_state != DCCP_RESPOND) {
616 617 618 619 620 621
		if (dccp_check_seqno(sk, skb))
			goto discard;

		/*
		 * Step 8: Process options and mark acknowledgeable
		 */
622
		if (dccp_parse_options(sk, NULL, skb))
623
			return 1;
624

625
		if (dcb->dccpd_ack_seq != DCCP_PKT_WITHOUT_ACK_SEQ)
626 627
			dccp_event_ack_recv(sk, skb);

628
		if (dp->dccps_hc_rx_ackvec != NULL &&
629
		    dccp_ackvec_add(dp->dccps_hc_rx_ackvec, sk,
630 631 632
				    DCCP_SKB_CB(skb)->dccpd_seq,
				    DCCP_ACKVEC_STATE_RECEIVED))
			goto discard;
633

634
		dccp_deliver_input_to_ccids(sk, skb);
635 636 637 638 639 640 641 642 643 644 645
	}

	/*
	 *  Step 9: Process Reset
	 *	If P.type == Reset,
	 *		Tear down connection
	 *		S.state := TIMEWAIT
	 *		Set TIMEWAIT timer
	 *		Drop packet and return
	*/
	if (dh->dccph_type == DCCP_PKT_RESET) {
646
		dccp_rcv_reset(sk, skb);
647 648 649
		return 0;
		/*
		 *   Step 7: Check for unexpected packet types
650
		 *      If (S.is_server and P.type == Response)
651 652 653 654 655 656
		 *	    or (S.is_client and P.type == Request)
		 *	    or (S.state == RESPOND and P.type == Data),
		 *	  Send Sync packet acknowledging P.seqno
		 *	  Drop packet and return
		 */
	} else if ((dp->dccps_role != DCCP_ROLE_CLIENT &&
657
		    dh->dccph_type == DCCP_PKT_RESPONSE) ||
658 659
		    (dp->dccps_role == DCCP_ROLE_CLIENT &&
		     dh->dccph_type == DCCP_PKT_REQUEST) ||
660 661
		    (sk->sk_state == DCCP_RESPOND &&
		     dh->dccph_type == DCCP_PKT_DATA)) {
662
		dccp_send_sync(sk, dcb->dccpd_seq, DCCP_PKT_SYNC);
663
		goto discard;
664
	} else if (dh->dccph_type == DCCP_PKT_CLOSEREQ) {
665 666
		if (dccp_rcv_closereq(sk, skb))
			return 0;
667 668
		goto discard;
	} else if (dh->dccph_type == DCCP_PKT_CLOSE) {
669 670 671
		if (dccp_rcv_close(sk, skb))
			return 0;
		goto discard;
672 673 674 675
	}

	switch (sk->sk_state) {
	case DCCP_CLOSED:
676
		dcb->dccpd_reset_code = DCCP_RESET_CODE_NO_CONNECTION;
677 678 679 680 681 682 683 684 685 686 687 688
		return 1;

	case DCCP_REQUESTING:
		queued = dccp_rcv_request_sent_state_process(sk, skb, dh, len);
		if (queued >= 0)
			return queued;

		__kfree_skb(skb);
		return 0;

	case DCCP_RESPOND:
	case DCCP_PARTOPEN:
689 690
		queued = dccp_rcv_respond_partopen_state_process(sk, skb,
								 dh, len);
691 692 693
		break;
	}

694 695
	if (dh->dccph_type == DCCP_PKT_ACK ||
	    dh->dccph_type == DCCP_PKT_DATAACK) {
696 697 698
		switch (old_state) {
		case DCCP_PARTOPEN:
			sk->sk_state_change(sk);
699
			sk_wake_async(sk, SOCK_WAKE_IO, POLL_OUT);
700 701
			break;
		}
702 703 704
	} else if (unlikely(dh->dccph_type == DCCP_PKT_SYNC)) {
		dccp_send_sync(sk, dcb->dccpd_seq, DCCP_PKT_SYNCACK);
		goto discard;
705 706
	}

707
	if (!queued) {
708 709 710 711 712
discard:
		__kfree_skb(skb);
	}
	return 0;
}
713 714

EXPORT_SYMBOL_GPL(dccp_rcv_state_process);
715 716

/**
717 718 719 720
 *  dccp_sample_rtt  -  Validate and finalise computation of RTT sample
 *  @delta:	number of microseconds between packet and acknowledgment
 *  The routine is kept generic to work in different contexts. It should be
 *  called immediately when the ACK used for the RTT sample arrives.
721
 */
722
u32 dccp_sample_rtt(struct sock *sk, long delta)
723
{
724 725
	/* dccpor_elapsed_time is either zeroed out or set and > 0 */
	delta -= dccp_sk(sk)->dccps_options_received.dccpor_elapsed_time * 10;
726 727

	if (unlikely(delta <= 0)) {
728
		DCCP_WARN("unusable RTT sample %ld, using min\n", delta);
729 730
		return DCCP_SANE_RTT_MIN;
	}
731 732
	if (unlikely(delta > DCCP_SANE_RTT_MAX)) {
		DCCP_WARN("RTT sample %ld too large, using max\n", delta);
733 734 735 736 737 738 739
		return DCCP_SANE_RTT_MAX;
	}

	return delta;
}

EXPORT_SYMBOL_GPL(dccp_sample_rtt);