summaryrefslogtreecommitdiff
path: root/drivers/net/wireguard/send.c
blob: 6687db69980351c6ba68bb5757626a7dbdbd0e56 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
 */

#include "queueing.h"
#include "timers.h"
#include "device.h"
#include "peer.h"
#include "socket.h"
#include "messages.h"
#include "cookie.h"

#include <linux/uio.h>
#include <linux/inetdevice.h>
#include <linux/socket.h>
#include <net/ip_tunnels.h>
#include <net/udp.h>
#include <net/sock.h>

static void wg_packet_send_handshake_initiation(struct wg_peer *peer)
{
	struct message_handshake_initiation packet;

	if (!wg_birthdate_has_expired(atomic64_read(&peer->last_sent_handshake),
				      REKEY_TIMEOUT))
		return; /* This function is rate limited. */

	atomic64_set(&peer->last_sent_handshake, ktime_get_coarse_boottime_ns());
	net_dbg_ratelimited("%s: Sending handshake initiation to peer %llu (%pISpfsc)\n",
			    peer->device->dev->name, peer->internal_id,
			    &peer->endpoint.addr);

	if (wg_noise_handshake_create_initiation(&packet, &peer->handshake)) {
		wg_cookie_add_mac_to_packet(&packet, sizeof(packet), peer);
		wg_timers_any_authenticated_packet_traversal(peer);
		wg_timers_any_authenticated_packet_sent(peer);
		atomic64_set(&peer->last_sent_handshake,
			     ktime_get_coarse_boottime_ns());
		wg_socket_send_buffer_to_peer(peer, &packet, sizeof(packet),
					      HANDSHAKE_DSCP);
		wg_timers_handshake_initiated(peer);
	}
}

void wg_packet_handshake_send_worker(struct work_struct *work)
{
	struct wg_peer *peer = container_of(work, struct wg_peer,
					    transmit_handshake_work);

	wg_packet_send_handshake_initiation(peer);
	wg_peer_put(peer);
}

void wg_packet_send_queued_handshake_initiation(struct wg_peer *peer,
						bool is_retry)
{
	if (!is_retry)
		peer->timer_handshake_attempts = 0;

	rcu_read_lock_bh();
	/* We check last_sent_handshake here in addition to the actual function
	 * we're queueing up, so that we don't queue things if not strictly
	 * necessary:
	 */
	if (!wg_birthdate_has_expired(atomic64_read(&peer->last_sent_handshake),
				      REKEY_TIMEOUT) ||
			unlikely(READ_ONCE(peer->is_dead)))
		goto out;

	wg_peer_get(peer);
	/* Queues up calling packet_send_queued_handshakes(peer), where we do a
	 * peer_put(peer) after:
	 */
	if (!queue_work(peer->device->handshake_send_wq,
			&peer->transmit_handshake_work))
		/* If the work was already queued, we want to drop the
		 * extra reference:
		 */
		wg_peer_put(peer);
out:
	rcu_read_unlock_bh();
}

void wg_packet_send_handshake_response(struct wg_peer *peer)
{
	struct message_handshake_response packet;

	atomic64_set(&peer->last_sent_handshake, ktime_get_coarse_boottime_ns());
	net_dbg_ratelimited("%s: Sending handshake response to peer %llu (%pISpfsc)\n",
			    peer->device->dev->name, peer->internal_id,
			    &peer->endpoint.addr);

	if (wg_noise_handshake_create_response(&packet, &peer->handshake)) {
		wg_cookie_add_mac_to_packet(&packet, sizeof(packet), peer);
		if (wg_noise_handshake_begin_session(&peer->handshake,
						     &peer->keypairs)) {
			wg_timers_session_derived(peer);
			wg_timers_any_authenticated_packet_traversal(peer);
			wg_timers_any_authenticated_packet_sent(peer);
			atomic64_set(&peer->last_sent_handshake,
				     ktime_get_coarse_boottime_ns());
			wg_socket_send_buffer_to_peer(peer, &packet,
						      sizeof(packet),
						      HANDSHAKE_DSCP);
		}
	}
}

void wg_packet_send_handshake_cookie(struct wg_device *wg,
				     struct sk_buff *initiating_skb,
				     __le32 sender_index)
{
	struct message_handshake_cookie packet;

	net_dbg_skb_ratelimited("%s: Sending cookie response for denied handshake message for %pISpfsc\n",
				wg->dev->name, initiating_skb);
	wg_cookie_message_create(&packet, initiating_skb, sender_index,
				 &wg->cookie_checker);
	wg_socket_send_buffer_as_reply_to_skb(wg, initiating_skb, &packet,
					      sizeof(packet));
}

static void keep_key_fresh(struct wg_peer *peer)
{
	struct noise_keypair *keypair;
	bool send;

	rcu_read_lock_bh();
	keypair = rcu_dereference_bh(peer->keypairs.current_keypair);
	send = keypair && READ_ONCE(keypair->sending.is_valid) &&
	       (atomic64_read(&keypair->sending.counter.counter) > REKEY_AFTER_MESSAGES ||
		(keypair->i_am_the_initiator &&
		 wg_birthdate_has_expired(keypair->sending.birthdate, REKEY_AFTER_TIME)));
	rcu_read_unlock_bh();

	if (unlikely(send))
		wg_packet_send_queued_handshake_initiation(peer, false);
}

static unsigned int calculate_skb_padding(struct sk_buff *skb)
{
	unsigned int padded_size, last_unit = skb->len;

	if (unlikely(!PACKET_CB(skb)->mtu))
		return ALIGN(last_unit, MESSAGE_PADDING_MULTIPLE) - last_unit;

	/* We do this modulo business with the MTU, just in case the networking
	 * layer gives us a packet that's bigger than the MTU. In that case, we
	 * wouldn't want the final subtraction to overflow in the case of the
	 * padded_size being clamped. Fortunately, that's very rarely the case,
	 * so we optimize for that not happening.
	 */
	if (unlikely(last_unit > PACKET_CB(skb)->mtu))
		last_unit %= PACKET_CB(skb)->mtu;

	padded_size = min(PACKET_CB(skb)->mtu,
			  ALIGN(last_unit, MESSAGE_PADDING_MULTIPLE));
	return padded_size - last_unit;
}

static bool encrypt_packet(struct sk_buff *skb, struct noise_keypair *keypair)
{
	unsigned int padding_len, plaintext_len, trailer_len;
	struct scatterlist sg[MAX_SKB_FRAGS + 8];
	struct message_data *header;
	struct sk_buff *trailer;
	int num_frags;

	/* Calculate lengths. */
	padding_len = calculate_skb_padding(skb);
	trailer_len = padding_len + noise_encrypted_len(0);
	plaintext_len = skb->len + padding_len;

	/* Expand data section to have room for padding and auth tag. */
	num_frags = skb_cow_data(skb, trailer_len, &trailer);
	if (unlikely(num_frags < 0 || num_frags > ARRAY_SIZE(sg)))
		return false;

	/* Set the padding to zeros, and make sure it and the auth tag are part
	 * of the skb.
	 */
	memset(skb_tail_pointer(trailer), 0, padding_len);

	/* Expand head section to have room for our header and the network
	 * stack's headers.
	 */
	if (unlikely(skb_cow_head(skb, DATA_PACKET_HEAD_ROOM) < 0))
		return false;

	/* Finalize checksum calculation for the inner packet, if required. */
	if (unlikely(skb->ip_summed == CHECKSUM_PARTIAL &&
		     skb_checksum_help(skb)))
		return false;

	/* Only after checksumming can we safely add on the padding at the end
	 * and the header.
	 */
	skb_set_inner_network_header(skb, 0);
	header = (struct message_data *)skb_push(skb, sizeof(*header));
	header->header.type = cpu_to_le32(MESSAGE_DATA);
	header->key_idx = keypair->remote_index;
	header->counter = cpu_to_le64(PACKET_CB(skb)->nonce);
	pskb_put(skb, trailer, trailer_len);

	/* Now we can encrypt the scattergather segments */
	sg_init_table(sg, num_frags);
	if (skb_to_sgvec(skb, sg, sizeof(struct message_data),
			 noise_encrypted_len(plaintext_len)) <= 0)
		return false;
	return chacha20poly1305_encrypt_sg_inplace(sg, plaintext_len, NULL, 0,
						   PACKET_CB(skb)->nonce,
						   keypair->sending.key);
}

void wg_packet_send_keepalive(struct wg_peer *peer)
{
	struct sk_buff *skb;

	if (skb_queue_empty(&peer->staged_packet_queue)) {
		skb = alloc_skb(DATA_PACKET_HEAD_ROOM + MESSAGE_MINIMUM_LENGTH,
				GFP_ATOMIC);
		if (unlikely(!skb))
			return;
		skb_reserve(skb, DATA_PACKET_HEAD_ROOM);
		skb->dev = peer->device->dev;
		PACKET_CB(skb)->mtu = skb->dev->mtu;
		skb_queue_tail(&peer->staged_packet_queue, skb);
		net_dbg_ratelimited("%s: Sending keepalive packet to peer %llu (%pISpfsc)\n",
				    peer->device->dev->name, peer->internal_id,
				    &peer->endpoint.addr);
	}

	wg_packet_send_staged_packets(peer);
}

static void wg_packet_create_data_done(struct sk_buff *first,
				       struct wg_peer *peer)
{
	struct sk_buff *skb, *next;
	bool is_keepalive, data_sent = false;

	wg_timers_any_authenticated_packet_traversal(peer);
	wg_timers_any_authenticated_packet_sent(peer);
	skb_list_walk_safe(first, skb, next) {
		is_keepalive = skb->len == message_data_len(0);
		if (likely(!wg_socket_send_skb_to_peer(peer, skb,
				PACKET_CB(skb)->ds) && !is_keepalive))
			data_sent = true;
	}

	if (likely(data_sent))
		wg_timers_data_sent(peer);

	keep_key_fresh(peer);
}

void wg_packet_tx_worker(struct work_struct *work)
{
	struct crypt_queue *queue = container_of(work, struct crypt_queue,
						 work);
	struct noise_keypair *keypair;
	enum packet_state state;
	struct sk_buff *first;
	struct wg_peer *peer;

	while ((first = __ptr_ring_peek(&queue->ring)) != NULL &&
	       (state = atomic_read_acquire(&PACKET_CB(first)->state)) !=
		       PACKET_STATE_UNCRYPTED) {
		__ptr_ring_discard_one(&queue->ring);
		peer = PACKET_PEER(first);
		keypair = PACKET_CB(first)->keypair;

		if (likely(state == PACKET_STATE_CRYPTED))
			wg_packet_create_data_done(first, peer);
		else
			kfree_skb_list(first);

		wg_noise_keypair_put(keypair, false);
		wg_peer_put(peer);
		if (need_resched())
			cond_resched();
	}
}

void wg_packet_encrypt_worker(struct work_struct *work)
{
	struct crypt_queue *queue = container_of(work, struct multicore_worker,
						 work)->ptr;
	struct sk_buff *first, *skb, *next;

	while ((first = ptr_ring_consume_bh(&queue->ring)) != NULL) {
		enum packet_state state = PACKET_STATE_CRYPTED;

		skb_list_walk_safe(first, skb, next) {
			if (likely(encrypt_packet(skb,
					PACKET_CB(first)->keypair))) {
				wg_reset_packet(skb);
			} else {
				state = PACKET_STATE_DEAD;
				break;
			}
		}
		wg_queue_enqueue_per_peer(&PACKET_PEER(first)->tx_queue, first,
					  state);
		if (need_resched())
			cond_resched();
	}
}

static void wg_packet_create_data(struct sk_buff *first)
{
	struct wg_peer *peer = PACKET_PEER(first);
	struct wg_device *wg = peer->device;
	int ret = -EINVAL;

	rcu_read_lock_bh();
	if (unlikely(READ_ONCE(peer->is_dead)))
		goto err;

	ret = wg_queue_enqueue_per_device_and_peer(&wg->encrypt_queue,
						   &peer->tx_queue, first,
						   wg->packet_crypt_wq,
						   &wg->encrypt_queue.last_cpu);
	if (unlikely(ret == -EPIPE))
		wg_queue_enqueue_per_peer(&peer->tx_queue, first,
					  PACKET_STATE_DEAD);
err:
	rcu_read_unlock_bh();
	if (likely(!ret || ret == -EPIPE))
		return;
	wg_noise_keypair_put(PACKET_CB(first)->keypair, false);
	wg_peer_put(peer);
	kfree_skb_list(first);
}

void wg_packet_purge_staged_packets(struct wg_peer *peer)
{
	spin_lock_bh(&peer->staged_packet_queue.lock);
	peer->device->dev->stats.tx_dropped += peer->staged_packet_queue.qlen;
	__skb_queue_purge(&peer->staged_packet_queue);
	spin_unlock_bh(&peer->staged_packet_queue.lock);
}

void wg_packet_send_staged_packets(struct wg_peer *peer)
{
	struct noise_symmetric_key *key;
	struct noise_keypair *keypair;
	struct sk_buff_head packets;
	struct sk_buff *skb;

	/* Steal the current queue into our local one. */
	__skb_queue_head_init(&packets);
	spin_lock_bh(&peer->staged_packet_queue.lock);
	skb_queue_splice_init(&peer->staged_packet_queue, &packets);
	spin_unlock_bh(&peer->staged_packet_queue.lock);
	if (unlikely(skb_queue_empty(&packets)))
		return;

	/* First we make sure we have a valid reference to a valid key. */
	rcu_read_lock_bh();
	keypair = wg_noise_keypair_get(
		rcu_dereference_bh(peer->keypairs.current_keypair));
	rcu_read_unlock_bh();
	if (unlikely(!keypair))
		goto out_nokey;
	key = &keypair->sending;
	if (unlikely(!READ_ONCE(key->is_valid)))
		goto out_nokey;
	if (unlikely(wg_birthdate_has_expired(key->birthdate,
					      REJECT_AFTER_TIME)))
		goto out_invalid;

	/* After we know we have a somewhat valid key, we now try to assign
	 * nonces to all of the packets in the queue. If we can't assign nonces
	 * for all of them, we just consider it a failure and wait for the next
	 * handshake.
	 */
	skb_queue_walk(&packets, skb) {
		/* 0 for no outer TOS: no leak. TODO: at some later point, we
		 * might consider using flowi->tos as outer instead.
		 */
		PACKET_CB(skb)->ds = ip_tunnel_ecn_encap(0, ip_hdr(skb), skb);
		PACKET_CB(skb)->nonce =
				atomic64_inc_return(&key->counter.counter) - 1;
		if (unlikely(PACKET_CB(skb)->nonce >= REJECT_AFTER_MESSAGES))
			goto out_invalid;
	}

	packets.prev->next = NULL;
	wg_peer_get(keypair->entry.peer);
	PACKET_CB(packets.next)->keypair = keypair;
	wg_packet_create_data(packets.next);
	return;

out_invalid:
	WRITE_ONCE(key->is_valid, false);
out_nokey:
	wg_noise_keypair_put(keypair, false);

	/* We orphan the packets if we're waiting on a handshake, so that they
	 * don't block a socket's pool.
	 */
	skb_queue_walk(&packets, skb)
		skb_orphan(skb);
	/* Then we put them back on the top of the queue. We're not too
	 * concerned about accidentally getting things a little out of order if
	 * packets are being added really fast, because this queue is for before
	 * packets can even be sent and it's small anyway.
	 */
	spin_lock_bh(&peer->staged_packet_queue.lock);
	skb_queue_splice(&packets, &peer->staged_packet_queue);
	spin_unlock_bh(&peer->staged_packet_queue.lock);

	/* If we're exiting because there's something wrong with the key, it
	 * means we should initiate a new handshake.
	 */
	wg_packet_send_queued_handshake_initiation(peer, false);
}