input.c 25.8 KB
Newer Older
1 2
/* RxRPC packet reception
 *
3
 * Copyright (C) 2007, 2016 Red Hat, Inc. All Rights Reserved.
4 5 6 7 8 9 10 11
 * Written by David Howells (dhowells@redhat.com)
 *
 * 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.
 */

12 13
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

14 15 16 17 18 19 20 21
#include <linux/module.h>
#include <linux/net.h>
#include <linux/skbuff.h>
#include <linux/errqueue.h>
#include <linux/udp.h>
#include <linux/in.h>
#include <linux/in6.h>
#include <linux/icmp.h>
22
#include <linux/gfp.h>
23 24 25
#include <net/sock.h>
#include <net/af_rxrpc.h>
#include <net/ip.h>
26
#include <net/udp.h>
27
#include <net/net_namespace.h>
28 29
#include "ar-internal.h"

30 31 32 33 34 35 36 37 38
static void rxrpc_proto_abort(const char *why,
			      struct rxrpc_call *call, rxrpc_seq_t seq)
{
	if (rxrpc_abort_call(why, call, seq, RX_PROTOCOL_ERROR, EBADMSG)) {
		set_bit(RXRPC_CALL_EV_ABORT, &call->events);
		rxrpc_queue_call(call);
	}
}

39 40 41 42 43 44 45 46
/*
 * Ping the other end to fill our RTT cache and to retrieve the rwind
 * and MTU parameters.
 */
static void rxrpc_send_ping(struct rxrpc_call *call, struct sk_buff *skb,
			    int skew)
{
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
47
	ktime_t now = skb->tstamp;
48

49 50 51
	if (call->peer->rtt_usage < 3 ||
	    ktime_before(ktime_add_ms(call->peer->rtt_last_req, 1000), now))
		rxrpc_propose_ACK(call, RXRPC_ACK_PING, skew, sp->hdr.serial,
52 53
				  true, true,
				  rxrpc_propose_ack_ping_for_params);
54 55
}

56
/*
57
 * Apply a hard ACK by advancing the Tx window.
58
 */
59
static void rxrpc_rotate_tx_window(struct rxrpc_call *call, rxrpc_seq_t to)
60
{
61 62
	struct sk_buff *skb, *list = NULL;
	int ix;
63
	u8 annotation;
64

65
	spin_lock(&call->lock);
66

67 68 69 70
	while (before(call->tx_hard_ack, to)) {
		call->tx_hard_ack++;
		ix = call->tx_hard_ack & RXRPC_RXTX_BUFF_MASK;
		skb = call->rxtx_buffer[ix];
71
		annotation = call->rxtx_annotations[ix];
D
David Howells 已提交
72
		rxrpc_see_skb(skb, rxrpc_skb_tx_rotated);
73 74 75 76
		call->rxtx_buffer[ix] = NULL;
		call->rxtx_annotations[ix] = 0;
		skb->next = list;
		list = skb;
77 78 79

		if (annotation & RXRPC_TX_ANNO_LAST)
			set_bit(RXRPC_CALL_TX_LAST, &call->flags);
80
	}
81

82
	spin_unlock(&call->lock);
83

84 85 86
	trace_rxrpc_transmit(call, (test_bit(RXRPC_CALL_TX_LAST, &call->flags) ?
				    rxrpc_transmit_rotate_last :
				    rxrpc_transmit_rotate));
87 88
	wake_up(&call->waitq);

89 90 91 92
	while (list) {
		skb = list;
		list = skb->next;
		skb->next = NULL;
D
David Howells 已提交
93
		rxrpc_free_skb(skb, rxrpc_skb_tx_freed);
94
	}
95
}
96

97 98 99 100 101 102
/*
 * End the transmission phase of a call.
 *
 * This occurs when we get an ACKALL packet, the first DATA packet of a reply,
 * or a final ACK packet.
 */
103 104
static bool rxrpc_end_tx_phase(struct rxrpc_call *call, bool reply_begun,
			       const char *abort_why)
105
{
106

107
	ASSERT(test_bit(RXRPC_CALL_TX_LAST, &call->flags));
108

109
	write_lock(&call->state_lock);
110

111
	switch (call->state) {
112
	case RXRPC_CALL_CLIENT_SEND_REQUEST:
113
	case RXRPC_CALL_CLIENT_AWAIT_REPLY:
114 115 116 117
		if (reply_begun)
			call->state = RXRPC_CALL_CLIENT_RECV_REPLY;
		else
			call->state = RXRPC_CALL_CLIENT_AWAIT_REPLY;
118
		break;
119

120 121 122 123
	case RXRPC_CALL_SERVER_AWAIT_ACK:
		__rxrpc_call_completed(call);
		rxrpc_notify_socket(call);
		break;
124 125 126

	default:
		goto bad_state;
127 128
	}

129
	write_unlock(&call->state_lock);
130 131 132 133 134
	if (call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY) {
		trace_rxrpc_transmit(call, rxrpc_transmit_await_reply);
	} else {
		trace_rxrpc_transmit(call, rxrpc_transmit_end);
	}
135 136
	_leave(" = ok");
	return true;
137 138 139 140 141 142 143 144 145 146 147 148 149 150 151

bad_state:
	write_unlock(&call->state_lock);
	kdebug("end_tx %s", rxrpc_call_states[call->state]);
	rxrpc_proto_abort(abort_why, call, call->tx_top);
	return false;
}

/*
 * Begin the reply reception phase of a call.
 */
static bool rxrpc_receiving_reply(struct rxrpc_call *call)
{
	rxrpc_seq_t top = READ_ONCE(call->tx_top);

152 153 154 155 156 157 158 159 160
	if (call->ackr_reason) {
		spin_lock_bh(&call->lock);
		call->ackr_reason = 0;
		call->resend_at = call->expire_at;
		call->ack_at = call->expire_at;
		spin_unlock_bh(&call->lock);
		rxrpc_set_timer(call, rxrpc_timer_init_for_reply);
	}

161 162 163 164 165 166 167 168 169 170
	if (!test_bit(RXRPC_CALL_TX_LAST, &call->flags))
		rxrpc_rotate_tx_window(call, top);
	if (!test_bit(RXRPC_CALL_TX_LAST, &call->flags)) {
		rxrpc_proto_abort("TXL", call, top);
		return false;
	}
	if (!rxrpc_end_tx_phase(call, true, "ETD"))
		return false;
	call->tx_phase = false;
	return true;
171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
}

/*
 * Scan a jumbo packet to validate its structure and to work out how many
 * subpackets it contains.
 *
 * A jumbo packet is a collection of consecutive packets glued together with
 * little headers between that indicate how to change the initial header for
 * each subpacket.
 *
 * RXRPC_JUMBO_PACKET must be set on all but the last subpacket - and all but
 * the last are RXRPC_JUMBO_DATALEN in size.  The last subpacket may be of any
 * size.
 */
static bool rxrpc_validate_jumbo(struct sk_buff *skb)
{
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
	unsigned int offset = sp->offset;
189
	unsigned int len = skb->len;
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206
	int nr_jumbo = 1;
	u8 flags = sp->hdr.flags;

	do {
		nr_jumbo++;
		if (len - offset < RXRPC_JUMBO_SUBPKTLEN)
			goto protocol_error;
		if (flags & RXRPC_LAST_PACKET)
			goto protocol_error;
		offset += RXRPC_JUMBO_DATALEN;
		if (skb_copy_bits(skb, offset, &flags, 1) < 0)
			goto protocol_error;
		offset += sizeof(struct rxrpc_jumbo_header);
	} while (flags & RXRPC_JUMBO_PACKET);

	sp->nr_jumbo = nr_jumbo;
	return true;
207

208 209
protocol_error:
	return false;
210 211 212
}

/*
213 214 215 216 217 218 219 220 221 222 223
 * Handle reception of a duplicate packet.
 *
 * We have to take care to avoid an attack here whereby we're given a series of
 * jumbograms, each with a sequence number one before the preceding one and
 * filled up to maximum UDP size.  If they never send us the first packet in
 * the sequence, they can cause us to have to hold on to around 2MiB of kernel
 * space until the call times out.
 *
 * We limit the space usage by only accepting three duplicate jumbo packets per
 * call.  After that, we tell the other side we're no longer accepting jumbos
 * (that information is encoded in the ACK packet).
224
 */
225
static void rxrpc_input_dup_data(struct rxrpc_call *call, rxrpc_seq_t seq,
226
				 u8 annotation, bool *_jumbo_bad)
227
{
228 229 230
	/* Discard normal packets that are duplicates. */
	if (annotation == 0)
		return;
231

232 233 234 235
	/* Skip jumbo subpackets that are duplicates.  When we've had three or
	 * more partially duplicate jumbo packets, we refuse to take any more
	 * jumbos for this call.
	 */
236 237 238
	if (!*_jumbo_bad) {
		call->nr_jumbo_bad++;
		*_jumbo_bad = true;
239 240
	}
}
241

242 243 244 245 246 247 248 249 250 251 252
/*
 * Process a DATA packet, adding the packet to the Rx ring.
 */
static void rxrpc_input_data(struct rxrpc_call *call, struct sk_buff *skb,
			     u16 skew)
{
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
	unsigned int offset = sp->offset;
	unsigned int ix;
	rxrpc_serial_t serial = sp->hdr.serial, ack_serial = 0;
	rxrpc_seq_t seq = sp->hdr.seq, hard_ack;
253
	bool immediate_ack = false, jumbo_bad = false, queued;
254 255
	u16 len;
	u8 ack = 0, flags, annotation = 0;
256

257
	_enter("{%u,%u},{%u,%u}",
258
	       call->rx_hard_ack, call->rx_top, skb->len, seq);
259

260 261
	_proto("Rx DATA %%%u { #%u f=%02x }",
	       sp->hdr.serial, seq, sp->hdr.flags);
262

263 264
	if (call->state >= RXRPC_CALL_COMPLETE)
		return;
265

266 267 268
	/* Received data implicitly ACKs all of the request packets we sent
	 * when we're acting as a client.
	 */
269 270 271
	if ((call->state == RXRPC_CALL_CLIENT_SEND_REQUEST ||
	     call->state == RXRPC_CALL_CLIENT_AWAIT_REPLY) &&
	    !rxrpc_receiving_reply(call))
272
		return;
273

274
	call->ackr_prev_seq = seq;
275

276 277
	hard_ack = READ_ONCE(call->rx_hard_ack);
	if (after(seq, hard_ack + call->rx_winsize)) {
278
		ack = RXRPC_ACK_EXCEEDS_WINDOW;
279 280
		ack_serial = serial;
		goto ack;
281 282
	}

283 284
	flags = sp->hdr.flags;
	if (flags & RXRPC_JUMBO_PACKET) {
285
		if (call->nr_jumbo_bad > 3) {
286 287 288
			ack = RXRPC_ACK_NOSPACE;
			ack_serial = serial;
			goto ack;
289
		}
290
		annotation = 1;
291 292
	}

293 294 295
next_subpacket:
	queued = false;
	ix = seq & RXRPC_RXTX_BUFF_MASK;
296
	len = skb->len;
297 298 299 300
	if (flags & RXRPC_JUMBO_PACKET)
		len = RXRPC_JUMBO_DATALEN;

	if (flags & RXRPC_LAST_PACKET) {
301
		if (test_bit(RXRPC_CALL_RX_LAST, &call->flags) &&
302 303 304 305 306 307
		    seq != call->rx_top)
			return rxrpc_proto_abort("LSN", call, seq);
	} else {
		if (test_bit(RXRPC_CALL_RX_LAST, &call->flags) &&
		    after_eq(seq, call->rx_top))
			return rxrpc_proto_abort("LSA", call, seq);
308 309
	}

310 311 312 313 314 315 316 317 318 319 320 321
	if (before_eq(seq, hard_ack)) {
		ack = RXRPC_ACK_DUPLICATE;
		ack_serial = serial;
		goto skip;
	}

	if (flags & RXRPC_REQUEST_ACK && !ack) {
		ack = RXRPC_ACK_REQUESTED;
		ack_serial = serial;
	}

	if (call->rxtx_buffer[ix]) {
322
		rxrpc_input_dup_data(call, seq, annotation, &jumbo_bad);
323 324 325
		if (ack != RXRPC_ACK_DUPLICATE) {
			ack = RXRPC_ACK_DUPLICATE;
			ack_serial = serial;
326
		}
327 328
		immediate_ack = true;
		goto skip;
329 330
	}

331 332 333 334 335 336 337 338
	/* Queue the packet.  We use a couple of memory barriers here as need
	 * to make sure that rx_top is perceived to be set after the buffer
	 * pointer and that the buffer pointer is set after the annotation and
	 * the skb data.
	 *
	 * Barriers against rxrpc_recvmsg_data() and rxrpc_rotate_rx_window()
	 * and also rxrpc_fill_out_ack().
	 */
D
David Howells 已提交
339
	rxrpc_get_skb(skb, rxrpc_skb_rx_got);
340 341 342
	call->rxtx_annotations[ix] = annotation;
	smp_wmb();
	call->rxtx_buffer[ix] = skb;
343
	if (after(seq, call->rx_top)) {
344
		smp_store_release(&call->rx_top, seq);
345 346 347 348 349 350 351 352
	} else if (before(seq, call->rx_top)) {
		/* Send an immediate ACK if we fill in a hole */
		if (!ack) {
			ack = RXRPC_ACK_DELAY;
			ack_serial = serial;
		}
		immediate_ack = true;
	}
353
	if (flags & RXRPC_LAST_PACKET) {
354
		set_bit(RXRPC_CALL_RX_LAST, &call->flags);
355 356 357 358
		trace_rxrpc_receive(call, rxrpc_receive_queue_last, serial, seq);
	} else {
		trace_rxrpc_receive(call, rxrpc_receive_queue, serial, seq);
	}
359 360 361 362 363 364 365 366 367
	queued = true;

	if (after_eq(seq, call->rx_expect_next)) {
		if (after(seq, call->rx_expect_next)) {
			_net("OOS %u > %u", seq, call->rx_expect_next);
			ack = RXRPC_ACK_OUT_OF_SEQUENCE;
			ack_serial = serial;
		}
		call->rx_expect_next = seq + 1;
368 369
	}

370 371 372 373 374 375 376 377 378 379 380
skip:
	offset += len;
	if (flags & RXRPC_JUMBO_PACKET) {
		if (skb_copy_bits(skb, offset, &flags, 1) < 0)
			return rxrpc_proto_abort("XJF", call, seq);
		offset += sizeof(struct rxrpc_jumbo_header);
		seq++;
		serial++;
		annotation++;
		if (flags & RXRPC_JUMBO_PACKET)
			annotation |= RXRPC_RX_ANNO_JLAST;
381 382 383 384 385 386 387 388 389
		if (after(seq, hard_ack + call->rx_winsize)) {
			ack = RXRPC_ACK_EXCEEDS_WINDOW;
			ack_serial = serial;
			if (!jumbo_bad) {
				call->nr_jumbo_bad++;
				jumbo_bad = true;
			}
			goto ack;
		}
390 391 392 393

		_proto("Rx DATA Jumbo %%%u", serial);
		goto next_subpacket;
	}
394

395 396 397 398
	if (queued && flags & RXRPC_LAST_PACKET && !ack) {
		ack = RXRPC_ACK_DELAY;
		ack_serial = serial;
	}
399

400 401 402
ack:
	if (ack)
		rxrpc_propose_ACK(call, ack, skew, ack_serial,
403 404
				  immediate_ack, true,
				  rxrpc_propose_ack_input_data);
405

406 407 408
	if (sp->hdr.seq == READ_ONCE(call->rx_hard_ack) + 1)
		rxrpc_notify_socket(call);
	_leave(" [queued]");
409 410
}

411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
/*
 * Process a requested ACK.
 */
static void rxrpc_input_requested_ack(struct rxrpc_call *call,
				      ktime_t resp_time,
				      rxrpc_serial_t orig_serial,
				      rxrpc_serial_t ack_serial)
{
	struct rxrpc_skb_priv *sp;
	struct sk_buff *skb;
	ktime_t sent_at;
	int ix;

	for (ix = 0; ix < RXRPC_RXTX_BUFF_SIZE; ix++) {
		skb = call->rxtx_buffer[ix];
		if (!skb)
			continue;

		sp = rxrpc_skb(skb);
		if (sp->hdr.serial != orig_serial)
			continue;
		smp_rmb();
		sent_at = skb->tstamp;
		goto found;
	}
	return;

found:
	rxrpc_peer_add_rtt(call, rxrpc_rtt_rx_requested_ack,
			   orig_serial, ack_serial, sent_at, resp_time);
}

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
/*
 * Process a ping response.
 */
static void rxrpc_input_ping_response(struct rxrpc_call *call,
				      ktime_t resp_time,
				      rxrpc_serial_t orig_serial,
				      rxrpc_serial_t ack_serial)
{
	rxrpc_serial_t ping_serial;
	ktime_t ping_time;

	ping_time = call->ackr_ping_time;
	smp_rmb();
	ping_serial = call->ackr_ping;

	if (!test_bit(RXRPC_CALL_PINGING, &call->flags) ||
	    before(orig_serial, ping_serial))
		return;
	clear_bit(RXRPC_CALL_PINGING, &call->flags);
	if (after(orig_serial, ping_serial))
		return;

	rxrpc_peer_add_rtt(call, rxrpc_rtt_rx_ping_response,
			   orig_serial, ack_serial, ping_time, resp_time);
}

469
/*
470
 * Process the extra information that may be appended to an ACK packet
471
 */
472 473
static void rxrpc_input_ackinfo(struct rxrpc_call *call, struct sk_buff *skb,
				struct rxrpc_ackinfo *ackinfo)
474
{
475 476 477
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
	struct rxrpc_peer *peer;
	unsigned int mtu;
478
	u32 rwind = ntohl(ackinfo->rwind);
479 480 481 482

	_proto("Rx ACK %%%u Info { rx=%u max=%u rwin=%u jm=%u }",
	       sp->hdr.serial,
	       ntohl(ackinfo->rxMTU), ntohl(ackinfo->maxMTU),
483
	       rwind, ntohl(ackinfo->jumbo_max));
484

485 486 487
	if (rwind > RXRPC_RXTX_BUFF_SIZE - 1)
		rwind = RXRPC_RXTX_BUFF_SIZE - 1;
	call->tx_winsize = rwind;
488 489 490 491 492 493 494 495 496 497 498 499

	mtu = min(ntohl(ackinfo->rxMTU), ntohl(ackinfo->maxMTU));

	peer = call->peer;
	if (mtu < peer->maxdata) {
		spin_lock_bh(&peer->lock);
		peer->maxdata = mtu;
		peer->mtu = mtu + peer->hdrsize;
		spin_unlock_bh(&peer->lock);
		_net("Net MTU %u (maxdata %u)", peer->mtu, peer->maxdata);
	}
}
500

501 502 503 504 505 506 507 508 509 510 511 512 513 514
/*
 * Process individual soft ACKs.
 *
 * Each ACK in the array corresponds to one packet and can be either an ACK or
 * a NAK.  If we get find an explicitly NAK'd packet we resend immediately;
 * packets that lie beyond the end of the ACK list are scheduled for resend by
 * the timer on the basis that the peer might just not have processed them at
 * the time the ACK was sent.
 */
static void rxrpc_input_soft_acks(struct rxrpc_call *call, u8 *acks,
				  rxrpc_seq_t seq, int nr_acks)
{
	bool resend = false;
	int ix;
D
David Howells 已提交
515
	u8 annotation, anno_type;
516 517 518

	for (; nr_acks > 0; nr_acks--, seq++) {
		ix = seq & RXRPC_RXTX_BUFF_MASK;
D
David Howells 已提交
519 520 521
		annotation = call->rxtx_annotations[ix];
		anno_type = annotation & RXRPC_TX_ANNO_MASK;
		annotation &= ~RXRPC_TX_ANNO_MASK;
522
		switch (*acks++) {
523
		case RXRPC_ACK_TYPE_ACK:
D
David Howells 已提交
524 525 526 527
			if (anno_type == RXRPC_TX_ANNO_ACK)
				continue;
			call->rxtx_annotations[ix] =
				RXRPC_TX_ANNO_ACK | annotation;
528 529
			break;
		case RXRPC_ACK_TYPE_NACK:
D
David Howells 已提交
530
			if (anno_type == RXRPC_TX_ANNO_NAK)
531
				continue;
532 533
			if (anno_type == RXRPC_TX_ANNO_RETRANS)
				continue;
D
David Howells 已提交
534 535
			call->rxtx_annotations[ix] =
				RXRPC_TX_ANNO_NAK | annotation;
536 537 538 539
			resend = true;
			break;
		default:
			return rxrpc_proto_abort("SFT", call, 0);
540 541
		}
	}
542 543 544 545

	if (resend &&
	    !test_and_set_bit(RXRPC_CALL_EV_RESEND, &call->events))
		rxrpc_queue_call(call);
546 547 548
}

/*
549 550 551 552 553 554 555 556
 * Process an ACK packet.
 *
 * ack.firstPacket is the sequence number of the first soft-ACK'd/NAK'd packet
 * in the ACK array.  Anything before that is hard-ACK'd and may be discarded.
 *
 * A hard-ACK means that a packet has been processed and may be discarded; a
 * soft-ACK means that the packet may be discarded and retransmission
 * requested.  A phase is complete when all packets are hard-ACK'd.
557
 */
558 559
static void rxrpc_input_ack(struct rxrpc_call *call, struct sk_buff *skb,
			    u16 skew)
560
{
561
	u8 ack_reason;
562
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
563 564 565 566 567
	union {
		struct rxrpc_ackpacket ack;
		struct rxrpc_ackinfo info;
		u8 acks[RXRPC_MAXACKS];
	} buf;
568
	rxrpc_serial_t acked_serial;
569 570 571 572 573 574 575 576
	rxrpc_seq_t first_soft_ack, hard_ack;
	int nr_acks, offset;

	_enter("");

	if (skb_copy_bits(skb, sp->offset, &buf.ack, sizeof(buf.ack)) < 0) {
		_debug("extraction failure");
		return rxrpc_proto_abort("XAK", call, 0);
577
	}
578 579
	sp->offset += sizeof(buf.ack);

580
	acked_serial = ntohl(buf.ack.serial);
581 582 583
	first_soft_ack = ntohl(buf.ack.firstPacket);
	hard_ack = first_soft_ack - 1;
	nr_acks = buf.ack.nAcks;
584 585
	ack_reason = (buf.ack.reason < RXRPC_ACK__INVALID ?
		      buf.ack.reason : RXRPC_ACK__INVALID);
586

587
	trace_rxrpc_rx_ack(call, first_soft_ack, ack_reason, nr_acks);
588

589 590 591 592 593
	_proto("Rx ACK %%%u { m=%hu f=#%u p=#%u s=%%%u r=%s n=%u }",
	       sp->hdr.serial,
	       ntohs(buf.ack.maxSkew),
	       first_soft_ack,
	       ntohl(buf.ack.previousPacket),
594
	       acked_serial,
595
	       rxrpc_ack_names[ack_reason],
596 597
	       buf.ack.nAcks);

598 599 600
	if (buf.ack.reason == RXRPC_ACK_PING_RESPONSE)
		rxrpc_input_ping_response(call, skb->tstamp, acked_serial,
					  sp->hdr.serial);
601 602 603
	if (buf.ack.reason == RXRPC_ACK_REQUESTED)
		rxrpc_input_requested_ack(call, skb->tstamp, acked_serial,
					  sp->hdr.serial);
604

605 606 607
	if (buf.ack.reason == RXRPC_ACK_PING) {
		_proto("Rx ACK %%%u PING Request", sp->hdr.serial);
		rxrpc_propose_ACK(call, RXRPC_ACK_PING_RESPONSE,
608 609
				  skew, sp->hdr.serial, true, true,
				  rxrpc_propose_ack_respond_to_ping);
610
	} else if (sp->hdr.flags & RXRPC_REQUEST_ACK) {
611
		rxrpc_propose_ACK(call, RXRPC_ACK_REQUESTED,
612 613
				  skew, sp->hdr.serial, true, true,
				  rxrpc_propose_ack_respond_to_ack);
614 615
	}

616
	offset = sp->offset + nr_acks + 3;
617
	if (skb->len >= offset + sizeof(buf.info)) {
618 619 620 621
		if (skb_copy_bits(skb, offset, &buf.info, sizeof(buf.info)) < 0)
			return rxrpc_proto_abort("XAI", call, 0);
		rxrpc_input_ackinfo(call, skb, &buf.info);
	}
622

623 624
	if (first_soft_ack == 0)
		return rxrpc_proto_abort("AK0", call, 0);
625

626 627 628 629 630 631 632
	/* Ignore ACKs unless we are or have just been transmitting. */
	switch (call->state) {
	case RXRPC_CALL_CLIENT_SEND_REQUEST:
	case RXRPC_CALL_CLIENT_AWAIT_REPLY:
	case RXRPC_CALL_SERVER_SEND_REPLY:
	case RXRPC_CALL_SERVER_AWAIT_ACK:
		break;
633
	default:
634 635
		return;
	}
636

637
	/* Discard any out-of-order or duplicate ACKs. */
638
	if (before_eq(sp->hdr.serial, call->acks_latest)) {
639 640 641 642 643
		_debug("discard ACK %d <= %d",
		       sp->hdr.serial, call->acks_latest);
		return;
	}
	call->acks_latest = sp->hdr.serial;
644

645 646 647
	if (before(hard_ack, call->tx_hard_ack) ||
	    after(hard_ack, call->tx_top))
		return rxrpc_proto_abort("AKW", call, 0);
648 649
	if (nr_acks > call->tx_top - hard_ack)
		return rxrpc_proto_abort("AKN", call, 0);
650

651 652
	if (after(hard_ack, call->tx_hard_ack))
		rxrpc_rotate_tx_window(call, hard_ack);
653

654 655 656 657 658 659 660 661
	if (nr_acks > 0) {
		if (skb_copy_bits(skb, sp->offset, buf.acks, nr_acks) < 0)
			return rxrpc_proto_abort("XSA", call, 0);
		rxrpc_input_soft_acks(call, buf.acks, first_soft_ack, nr_acks);
	}

	if (test_bit(RXRPC_CALL_TX_LAST, &call->flags)) {
		rxrpc_end_tx_phase(call, false, "ETA");
662
		return;
663
	}
664 665 666 667

}

/*
668
 * Process an ACKALL packet.
669
 */
670
static void rxrpc_input_ackall(struct rxrpc_call *call, struct sk_buff *skb)
671
{
672
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
673

674
	_proto("Rx ACKALL %%%u", sp->hdr.serial);
675

676 677 678
	rxrpc_rotate_tx_window(call, call->tx_top);
	if (test_bit(RXRPC_CALL_TX_LAST, &call->flags))
		rxrpc_end_tx_phase(call, false, "ETL");
679
}
680

681 682 683 684 685 686 687 688
/*
 * Process an ABORT packet.
 */
static void rxrpc_input_abort(struct rxrpc_call *call, struct sk_buff *skb)
{
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
	__be32 wtmp;
	u32 abort_code = RX_CALL_DEAD;
689

690
	_enter("");
691

692 693 694
	if (skb->len >= 4 &&
	    skb_copy_bits(skb, sp->offset, &wtmp, sizeof(wtmp)) >= 0)
		abort_code = ntohl(wtmp);
695

696
	_proto("Rx ABORT %%%u { %x }", sp->hdr.serial, abort_code);
697

698 699 700
	if (rxrpc_set_call_completion(call, RXRPC_CALL_REMOTELY_ABORTED,
				      abort_code, ECONNABORTED))
		rxrpc_notify_socket(call);
701 702 703
}

/*
704
 * Process an incoming call packet.
705
 */
706 707
static void rxrpc_input_call_packet(struct rxrpc_call *call,
				    struct sk_buff *skb, u16 skew)
708
{
709
	struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
710

711
	_enter("%p,%p", call, skb);
712

713 714 715 716
	switch (sp->hdr.type) {
	case RXRPC_PACKET_TYPE_DATA:
		rxrpc_input_data(call, skb, skew);
		break;
717

718 719
	case RXRPC_PACKET_TYPE_ACK:
		rxrpc_input_ack(call, skb, skew);
720 721
		break;

722 723
	case RXRPC_PACKET_TYPE_BUSY:
		_proto("Rx BUSY %%%u", sp->hdr.serial);
724

725 726 727 728 729
		/* Just ignore BUSY packets from the server; the retry and
		 * lifespan timers will take care of business.  BUSY packets
		 * from the client don't make sense.
		 */
		break;
730

731 732 733
	case RXRPC_PACKET_TYPE_ABORT:
		rxrpc_input_abort(call, skb);
		break;
734

735 736 737
	case RXRPC_PACKET_TYPE_ACKALL:
		rxrpc_input_ackall(call, skb);
		break;
738

739 740 741
	default:
		_proto("Rx %s %%%u", rxrpc_pkts[sp->hdr.type], sp->hdr.serial);
		break;
742
	}
743

744 745 746 747 748
	_leave("");
}

/*
 * post connection-level events to the connection
749 750
 * - this includes challenges, responses, some aborts and call terminal packet
 *   retransmission.
751
 */
752
static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn,
753 754 755 756 757
				      struct sk_buff *skb)
{
	_enter("%p,%p", conn, skb);

	skb_queue_tail(&conn->rx_queue, skb);
758
	rxrpc_queue_conn(conn);
759 760
}

761 762 763 764 765 766 767 768 769 770
/*
 * post endpoint-level events to the local endpoint
 * - this includes debug and version messages
 */
static void rxrpc_post_packet_to_local(struct rxrpc_local *local,
				       struct sk_buff *skb)
{
	_enter("%p,%p", local, skb);

	skb_queue_tail(&local->event_queue, skb);
771
	rxrpc_queue_local(local);
772 773
}

774 775 776 777 778 779 780 781 782 783 784
/*
 * put a packet up for transport-level abort
 */
static void rxrpc_reject_packet(struct rxrpc_local *local, struct sk_buff *skb)
{
	CHECK_SLAB_OKAY(&local->usage);

	skb_queue_tail(&local->reject_queue, skb);
	rxrpc_queue_local(local);
}

785 786 787 788 789 790 791 792 793
/*
 * Extract the wire header from a packet and translate the byte order.
 */
static noinline
int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb)
{
	struct rxrpc_wire_header whdr;

	/* dig out the RxRPC connection details */
794
	if (skb_copy_bits(skb, 0, &whdr, sizeof(whdr)) < 0)
795 796 797 798 799 800 801 802 803 804 805 806 807 808
		return -EBADMSG;

	memset(sp, 0, sizeof(*sp));
	sp->hdr.epoch		= ntohl(whdr.epoch);
	sp->hdr.cid		= ntohl(whdr.cid);
	sp->hdr.callNumber	= ntohl(whdr.callNumber);
	sp->hdr.seq		= ntohl(whdr.seq);
	sp->hdr.serial		= ntohl(whdr.serial);
	sp->hdr.flags		= whdr.flags;
	sp->hdr.type		= whdr.type;
	sp->hdr.userStatus	= whdr.userStatus;
	sp->hdr.securityIndex	= whdr.securityIndex;
	sp->hdr._rsvd		= ntohs(whdr._rsvd);
	sp->hdr.serviceId	= ntohs(whdr.serviceId);
809
	sp->offset = sizeof(whdr);
810 811 812
	return 0;
}

813 814 815
/*
 * handle data received on the local endpoint
 * - may be called in interrupt context
816 817 818 819
 *
 * The socket is locked by the caller and this prevents the socket from being
 * shut down and the local endpoint from going away, thus sk_user_data will not
 * be cleared until this function returns.
820
 */
821
void rxrpc_data_ready(struct sock *udp_sk)
822
{
823
	struct rxrpc_connection *conn;
824 825
	struct rxrpc_channel *chan;
	struct rxrpc_call *call;
826
	struct rxrpc_skb_priv *sp;
827
	struct rxrpc_local *local = udp_sk->sk_user_data;
828
	struct sk_buff *skb;
829
	unsigned int channel;
830
	int ret, skew;
831

832
	_enter("%p", udp_sk);
833 834 835

	ASSERT(!irqs_disabled());

836
	skb = skb_recv_datagram(udp_sk, 0, 1, &ret);
837 838 839 840 841 842 843
	if (!skb) {
		if (ret == -EAGAIN)
			return;
		_debug("UDP socket error %d", ret);
		return;
	}

D
David Howells 已提交
844
	rxrpc_new_skb(skb, rxrpc_skb_rx_received);
845 846 847 848 849

	_net("recv skb %p", skb);

	/* we'll probably need to checksum it (didn't call sock_recvmsg) */
	if (skb_checksum_complete(skb)) {
D
David Howells 已提交
850
		rxrpc_free_skb(skb, rxrpc_skb_rx_freed);
851
		__UDP_INC_STATS(&init_net, UDP_MIB_INERRORS, 0);
852 853 854 855
		_leave(" [CSUM failed]");
		return;
	}

856
	__UDP_INC_STATS(&init_net, UDP_MIB_INDATAGRAMS, 0);
857

858 859 860
	/* The socket buffer we have is owned by UDP, with UDP's data all over
	 * it, but we really want our own data there.
	 */
861 862 863
	skb_orphan(skb);
	sp = rxrpc_skb(skb);

864 865 866 867
	/* dig out the RxRPC connection details */
	if (rxrpc_extract_header(sp, skb) < 0)
		goto bad_message;

868 869 870
	if (IS_ENABLED(CONFIG_AF_RXRPC_INJECT_LOSS)) {
		static int lose;
		if ((lose++ & 7) == 7) {
871
			trace_rxrpc_rx_lose(sp);
872 873 874 875 876
			rxrpc_lose_skb(skb, rxrpc_skb_rx_lost);
			return;
		}
	}

877
	trace_rxrpc_rx_packet(sp);
878 879 880

	_net("Rx RxRPC %s ep=%x call=%x:%x",
	     sp->hdr.flags & RXRPC_CLIENT_INITIATED ? "ToServer" : "ToClient",
881
	     sp->hdr.epoch, sp->hdr.cid, sp->hdr.callNumber);
882

883 884
	if (sp->hdr.type >= RXRPC_N_PACKET_TYPES ||
	    !((RXRPC_SUPPORTED_PACKET_TYPES >> sp->hdr.type) & 1)) {
885 886 887 888
		_proto("Rx Bad Packet Type %u", sp->hdr.type);
		goto bad_message;
	}

889 890
	switch (sp->hdr.type) {
	case RXRPC_PACKET_TYPE_VERSION:
891 892
		rxrpc_post_packet_to_local(local, skb);
		goto out;
893

894 895 896 897 898 899 900 901 902 903 904 905
	case RXRPC_PACKET_TYPE_BUSY:
		if (sp->hdr.flags & RXRPC_CLIENT_INITIATED)
			goto discard;

	case RXRPC_PACKET_TYPE_DATA:
		if (sp->hdr.callNumber == 0)
			goto bad_message;
		if (sp->hdr.flags & RXRPC_JUMBO_PACKET &&
		    !rxrpc_validate_jumbo(skb))
			goto bad_message;
		break;
	}
906

907 908 909
	rcu_read_lock();

	conn = rxrpc_find_connection_rcu(local, skb);
910 911 912
	if (conn) {
		if (sp->hdr.securityIndex != conn->security_ix)
			goto wrong_security;
913

914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929
		if (sp->hdr.callNumber == 0) {
			/* Connection-level packet */
			_debug("CONN %p {%d}", conn, conn->debug_id);
			rxrpc_post_packet_to_conn(conn, skb);
			goto out_unlock;
		}

		/* Note the serial number skew here */
		skew = (int)sp->hdr.serial - (int)conn->hi_serial;
		if (skew >= 0) {
			if (skew > 0)
				conn->hi_serial = sp->hdr.serial;
		} else {
			skew = -skew;
			skew = min(skew, 65535);
		}
930

931
		/* Call-bound packets are routed by connection channel. */
932 933
		channel = sp->hdr.cid & RXRPC_CHANNELMASK;
		chan = &conn->channels[channel];
934 935 936 937 938 939

		/* Ignore really old calls */
		if (sp->hdr.callNumber < chan->last_call)
			goto discard_unlock;

		if (sp->hdr.callNumber == chan->last_call) {
940 941
			/* For the previous service call, if completed successfully, we
			 * discard all further packets.
942
			 */
D
David Howells 已提交
943
			if (rxrpc_conn_is_service(conn) &&
944 945 946 947
			    (chan->last_type == RXRPC_PACKET_TYPE_ACK ||
			     sp->hdr.type == RXRPC_PACKET_TYPE_ABORT))
				goto discard_unlock;

948 949
			/* But otherwise we need to retransmit the final packet from
			 * data cached in the connection record.
950 951 952 953
			 */
			rxrpc_post_packet_to_conn(conn, skb);
			goto out_unlock;
		}
954

955
		call = rcu_dereference(chan->call);
956 957 958 959
	} else {
		skew = 0;
		call = NULL;
	}
960

961 962 963 964 965 966 967 968 969 970 971 972
	if (!call || atomic_read(&call->usage) == 0) {
		if (!(sp->hdr.type & RXRPC_CLIENT_INITIATED) ||
		    sp->hdr.callNumber == 0 ||
		    sp->hdr.type != RXRPC_PACKET_TYPE_DATA)
			goto bad_message_unlock;
		if (sp->hdr.seq != 1)
			goto discard_unlock;
		call = rxrpc_new_incoming_call(local, conn, skb);
		if (!call) {
			rcu_read_unlock();
			goto reject_packet;
		}
973
		rxrpc_send_ping(call, skb, skew);
974
	}
975

976 977 978
	rxrpc_input_call_packet(call, skb, skew);
	goto discard_unlock;

979
discard_unlock:
980
	rcu_read_unlock();
981
discard:
D
David Howells 已提交
982
	rxrpc_free_skb(skb, rxrpc_skb_rx_freed);
983
out:
984
	trace_rxrpc_rx_done(0, 0);
985 986
	return;

987
out_unlock:
988
	rcu_read_unlock();
989
	goto out;
990

991 992 993 994 995 996
wrong_security:
	rcu_read_unlock();
	trace_rxrpc_abort("SEC", sp->hdr.cid, sp->hdr.callNumber, sp->hdr.seq,
			  RXKADINCONSISTENCY, EBADMSG);
	skb->priority = RXKADINCONSISTENCY;
	goto post_abort;
997

998 999
bad_message_unlock:
	rcu_read_unlock();
1000
bad_message:
1001 1002
	trace_rxrpc_abort("BAD", sp->hdr.cid, sp->hdr.callNumber, sp->hdr.seq,
			  RX_PROTOCOL_ERROR, EBADMSG);
1003
	skb->priority = RX_PROTOCOL_ERROR;
1004 1005
post_abort:
	skb->mark = RXRPC_SKB_MARK_LOCAL_ABORT;
1006 1007
reject_packet:
	trace_rxrpc_rx_done(skb->mark, skb->priority);
1008 1009 1010
	rxrpc_reject_packet(local, skb);
	_leave(" [badmsg]");
}