xref: /openbmc/linux/net/rxrpc/local_object.c (revision 37744fee)
1 // SPDX-License-Identifier: GPL-2.0-or-later
2 /* Local endpoint object management
3  *
4  * Copyright (C) 2016 Red Hat, Inc. All Rights Reserved.
5  * Written by David Howells (dhowells@redhat.com)
6  */
7 
8 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
9 
10 #include <linux/module.h>
11 #include <linux/net.h>
12 #include <linux/skbuff.h>
13 #include <linux/slab.h>
14 #include <linux/udp.h>
15 #include <linux/ip.h>
16 #include <linux/hashtable.h>
17 #include <net/sock.h>
18 #include <net/udp.h>
19 #include <net/af_rxrpc.h>
20 #include "ar-internal.h"
21 
22 static void rxrpc_local_processor(struct work_struct *);
23 static void rxrpc_local_rcu(struct rcu_head *);
24 
25 /*
26  * Compare a local to an address.  Return -ve, 0 or +ve to indicate less than,
27  * same or greater than.
28  *
29  * We explicitly don't compare the RxRPC service ID as we want to reject
30  * conflicting uses by differing services.  Further, we don't want to share
31  * addresses with different options (IPv6), so we don't compare those bits
32  * either.
33  */
34 static long rxrpc_local_cmp_key(const struct rxrpc_local *local,
35 				const struct sockaddr_rxrpc *srx)
36 {
37 	long diff;
38 
39 	diff = ((local->srx.transport_type - srx->transport_type) ?:
40 		(local->srx.transport_len - srx->transport_len) ?:
41 		(local->srx.transport.family - srx->transport.family));
42 	if (diff != 0)
43 		return diff;
44 
45 	switch (srx->transport.family) {
46 	case AF_INET:
47 		/* If the choice of UDP port is left up to the transport, then
48 		 * the endpoint record doesn't match.
49 		 */
50 		return ((u16 __force)local->srx.transport.sin.sin_port -
51 			(u16 __force)srx->transport.sin.sin_port) ?:
52 			memcmp(&local->srx.transport.sin.sin_addr,
53 			       &srx->transport.sin.sin_addr,
54 			       sizeof(struct in_addr));
55 #ifdef CONFIG_AF_RXRPC_IPV6
56 	case AF_INET6:
57 		/* If the choice of UDP6 port is left up to the transport, then
58 		 * the endpoint record doesn't match.
59 		 */
60 		return ((u16 __force)local->srx.transport.sin6.sin6_port -
61 			(u16 __force)srx->transport.sin6.sin6_port) ?:
62 			memcmp(&local->srx.transport.sin6.sin6_addr,
63 			       &srx->transport.sin6.sin6_addr,
64 			       sizeof(struct in6_addr));
65 #endif
66 	default:
67 		BUG();
68 	}
69 }
70 
71 /*
72  * Allocate a new local endpoint.
73  */
74 static struct rxrpc_local *rxrpc_alloc_local(struct rxrpc_net *rxnet,
75 					     const struct sockaddr_rxrpc *srx)
76 {
77 	struct rxrpc_local *local;
78 
79 	local = kzalloc(sizeof(struct rxrpc_local), GFP_KERNEL);
80 	if (local) {
81 		atomic_set(&local->usage, 1);
82 		atomic_set(&local->active_users, 1);
83 		local->rxnet = rxnet;
84 		INIT_LIST_HEAD(&local->link);
85 		INIT_WORK(&local->processor, rxrpc_local_processor);
86 		init_rwsem(&local->defrag_sem);
87 		skb_queue_head_init(&local->reject_queue);
88 		skb_queue_head_init(&local->event_queue);
89 		local->client_conns = RB_ROOT;
90 		spin_lock_init(&local->client_conns_lock);
91 		spin_lock_init(&local->lock);
92 		rwlock_init(&local->services_lock);
93 		local->debug_id = atomic_inc_return(&rxrpc_debug_id);
94 		memcpy(&local->srx, srx, sizeof(*srx));
95 		local->srx.srx_service = 0;
96 		trace_rxrpc_local(local->debug_id, rxrpc_local_new, 1, NULL);
97 	}
98 
99 	_leave(" = %p", local);
100 	return local;
101 }
102 
103 /*
104  * create the local socket
105  * - must be called with rxrpc_local_mutex locked
106  */
107 static int rxrpc_open_socket(struct rxrpc_local *local, struct net *net)
108 {
109 	struct sock *usk;
110 	int ret, opt;
111 
112 	_enter("%p{%d,%d}",
113 	       local, local->srx.transport_type, local->srx.transport.family);
114 
115 	/* create a socket to represent the local endpoint */
116 	ret = sock_create_kern(net, local->srx.transport.family,
117 			       local->srx.transport_type, 0, &local->socket);
118 	if (ret < 0) {
119 		_leave(" = %d [socket]", ret);
120 		return ret;
121 	}
122 
123 	/* set the socket up */
124 	usk = local->socket->sk;
125 	inet_sk(usk)->mc_loop = 0;
126 
127 	/* Enable CHECKSUM_UNNECESSARY to CHECKSUM_COMPLETE conversion */
128 	inet_inc_convert_csum(usk);
129 
130 	rcu_assign_sk_user_data(usk, local);
131 
132 	udp_sk(usk)->encap_type = UDP_ENCAP_RXRPC;
133 	udp_sk(usk)->encap_rcv = rxrpc_input_packet;
134 	udp_sk(usk)->encap_destroy = NULL;
135 	udp_sk(usk)->gro_receive = NULL;
136 	udp_sk(usk)->gro_complete = NULL;
137 
138 	udp_encap_enable();
139 #if IS_ENABLED(CONFIG_AF_RXRPC_IPV6)
140 	if (local->srx.transport.family == AF_INET6)
141 		udpv6_encap_enable();
142 #endif
143 	usk->sk_error_report = rxrpc_error_report;
144 
145 	/* if a local address was supplied then bind it */
146 	if (local->srx.transport_len > sizeof(sa_family_t)) {
147 		_debug("bind");
148 		ret = kernel_bind(local->socket,
149 				  (struct sockaddr *)&local->srx.transport,
150 				  local->srx.transport_len);
151 		if (ret < 0) {
152 			_debug("bind failed %d", ret);
153 			goto error;
154 		}
155 	}
156 
157 	switch (local->srx.transport.family) {
158 	case AF_INET6:
159 		/* we want to receive ICMPv6 errors */
160 		opt = 1;
161 		ret = kernel_setsockopt(local->socket, SOL_IPV6, IPV6_RECVERR,
162 					(char *) &opt, sizeof(opt));
163 		if (ret < 0) {
164 			_debug("setsockopt failed");
165 			goto error;
166 		}
167 
168 		/* Fall through and set IPv4 options too otherwise we don't get
169 		 * errors from IPv4 packets sent through the IPv6 socket.
170 		 */
171 		/* Fall through */
172 	case AF_INET:
173 		/* we want to receive ICMP errors */
174 		opt = 1;
175 		ret = kernel_setsockopt(local->socket, SOL_IP, IP_RECVERR,
176 					(char *) &opt, sizeof(opt));
177 		if (ret < 0) {
178 			_debug("setsockopt failed");
179 			goto error;
180 		}
181 
182 		/* we want to set the don't fragment bit */
183 		opt = IP_PMTUDISC_DO;
184 		ret = kernel_setsockopt(local->socket, SOL_IP, IP_MTU_DISCOVER,
185 					(char *) &opt, sizeof(opt));
186 		if (ret < 0) {
187 			_debug("setsockopt failed");
188 			goto error;
189 		}
190 
191 		/* We want receive timestamps. */
192 		opt = 1;
193 		ret = kernel_setsockopt(local->socket, SOL_SOCKET, SO_TIMESTAMPNS_OLD,
194 					(char *)&opt, sizeof(opt));
195 		if (ret < 0) {
196 			_debug("setsockopt failed");
197 			goto error;
198 		}
199 		break;
200 
201 	default:
202 		BUG();
203 	}
204 
205 	_leave(" = 0");
206 	return 0;
207 
208 error:
209 	kernel_sock_shutdown(local->socket, SHUT_RDWR);
210 	local->socket->sk->sk_user_data = NULL;
211 	sock_release(local->socket);
212 	local->socket = NULL;
213 
214 	_leave(" = %d", ret);
215 	return ret;
216 }
217 
218 /*
219  * Look up or create a new local endpoint using the specified local address.
220  */
221 struct rxrpc_local *rxrpc_lookup_local(struct net *net,
222 				       const struct sockaddr_rxrpc *srx)
223 {
224 	struct rxrpc_local *local;
225 	struct rxrpc_net *rxnet = rxrpc_net(net);
226 	struct list_head *cursor;
227 	const char *age;
228 	long diff;
229 	int ret;
230 
231 	_enter("{%d,%d,%pISp}",
232 	       srx->transport_type, srx->transport.family, &srx->transport);
233 
234 	mutex_lock(&rxnet->local_mutex);
235 
236 	for (cursor = rxnet->local_endpoints.next;
237 	     cursor != &rxnet->local_endpoints;
238 	     cursor = cursor->next) {
239 		local = list_entry(cursor, struct rxrpc_local, link);
240 
241 		diff = rxrpc_local_cmp_key(local, srx);
242 		if (diff < 0)
243 			continue;
244 		if (diff > 0)
245 			break;
246 
247 		/* Services aren't allowed to share transport sockets, so
248 		 * reject that here.  It is possible that the object is dying -
249 		 * but it may also still have the local transport address that
250 		 * we want bound.
251 		 */
252 		if (srx->srx_service) {
253 			local = NULL;
254 			goto addr_in_use;
255 		}
256 
257 		/* Found a match.  We replace a dying object.  Attempting to
258 		 * bind the transport socket may still fail if we're attempting
259 		 * to use a local address that the dying object is still using.
260 		 */
261 		if (!rxrpc_use_local(local))
262 			break;
263 
264 		age = "old";
265 		goto found;
266 	}
267 
268 	local = rxrpc_alloc_local(rxnet, srx);
269 	if (!local)
270 		goto nomem;
271 
272 	ret = rxrpc_open_socket(local, net);
273 	if (ret < 0)
274 		goto sock_error;
275 
276 	if (cursor != &rxnet->local_endpoints)
277 		list_replace_init(cursor, &local->link);
278 	else
279 		list_add_tail(&local->link, cursor);
280 	age = "new";
281 
282 found:
283 	mutex_unlock(&rxnet->local_mutex);
284 
285 	_net("LOCAL %s %d {%pISp}",
286 	     age, local->debug_id, &local->srx.transport);
287 
288 	_leave(" = %p", local);
289 	return local;
290 
291 nomem:
292 	ret = -ENOMEM;
293 sock_error:
294 	mutex_unlock(&rxnet->local_mutex);
295 	if (local)
296 		call_rcu(&local->rcu, rxrpc_local_rcu);
297 	_leave(" = %d", ret);
298 	return ERR_PTR(ret);
299 
300 addr_in_use:
301 	mutex_unlock(&rxnet->local_mutex);
302 	_leave(" = -EADDRINUSE");
303 	return ERR_PTR(-EADDRINUSE);
304 }
305 
306 /*
307  * Get a ref on a local endpoint.
308  */
309 struct rxrpc_local *rxrpc_get_local(struct rxrpc_local *local)
310 {
311 	const void *here = __builtin_return_address(0);
312 	int n;
313 
314 	n = atomic_inc_return(&local->usage);
315 	trace_rxrpc_local(local->debug_id, rxrpc_local_got, n, here);
316 	return local;
317 }
318 
319 /*
320  * Get a ref on a local endpoint unless its usage has already reached 0.
321  */
322 struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *local)
323 {
324 	const void *here = __builtin_return_address(0);
325 
326 	if (local) {
327 		int n = atomic_fetch_add_unless(&local->usage, 1, 0);
328 		if (n > 0)
329 			trace_rxrpc_local(local->debug_id, rxrpc_local_got,
330 					  n + 1, here);
331 		else
332 			local = NULL;
333 	}
334 	return local;
335 }
336 
337 /*
338  * Queue a local endpoint and pass the caller's reference to the work item.
339  */
340 void rxrpc_queue_local(struct rxrpc_local *local)
341 {
342 	const void *here = __builtin_return_address(0);
343 	unsigned int debug_id = local->debug_id;
344 	int n = atomic_read(&local->usage);
345 
346 	if (rxrpc_queue_work(&local->processor))
347 		trace_rxrpc_local(debug_id, rxrpc_local_queued, n, here);
348 	else
349 		rxrpc_put_local(local);
350 }
351 
352 /*
353  * Drop a ref on a local endpoint.
354  */
355 void rxrpc_put_local(struct rxrpc_local *local)
356 {
357 	const void *here = __builtin_return_address(0);
358 	unsigned int debug_id;
359 	int n;
360 
361 	if (local) {
362 		debug_id = local->debug_id;
363 
364 		n = atomic_dec_return(&local->usage);
365 		trace_rxrpc_local(debug_id, rxrpc_local_put, n, here);
366 
367 		if (n == 0)
368 			call_rcu(&local->rcu, rxrpc_local_rcu);
369 	}
370 }
371 
372 /*
373  * Start using a local endpoint.
374  */
375 struct rxrpc_local *rxrpc_use_local(struct rxrpc_local *local)
376 {
377 	local = rxrpc_get_local_maybe(local);
378 	if (!local)
379 		return NULL;
380 
381 	if (!__rxrpc_use_local(local)) {
382 		rxrpc_put_local(local);
383 		return NULL;
384 	}
385 
386 	return local;
387 }
388 
389 /*
390  * Cease using a local endpoint.  Once the number of active users reaches 0, we
391  * start the closure of the transport in the work processor.
392  */
393 void rxrpc_unuse_local(struct rxrpc_local *local)
394 {
395 	if (local) {
396 		if (__rxrpc_unuse_local(local)) {
397 			rxrpc_get_local(local);
398 			rxrpc_queue_local(local);
399 		}
400 	}
401 }
402 
403 /*
404  * Destroy a local endpoint's socket and then hand the record to RCU to dispose
405  * of.
406  *
407  * Closing the socket cannot be done from bottom half context or RCU callback
408  * context because it might sleep.
409  */
410 static void rxrpc_local_destroyer(struct rxrpc_local *local)
411 {
412 	struct socket *socket = local->socket;
413 	struct rxrpc_net *rxnet = local->rxnet;
414 
415 	_enter("%d", local->debug_id);
416 
417 	local->dead = true;
418 
419 	mutex_lock(&rxnet->local_mutex);
420 	list_del_init(&local->link);
421 	mutex_unlock(&rxnet->local_mutex);
422 
423 	rxrpc_clean_up_local_conns(local);
424 	rxrpc_service_connection_reaper(&rxnet->service_conn_reaper);
425 	ASSERT(!local->service);
426 
427 	if (socket) {
428 		local->socket = NULL;
429 		kernel_sock_shutdown(socket, SHUT_RDWR);
430 		socket->sk->sk_user_data = NULL;
431 		sock_release(socket);
432 	}
433 
434 	/* At this point, there should be no more packets coming in to the
435 	 * local endpoint.
436 	 */
437 	rxrpc_purge_queue(&local->reject_queue);
438 	rxrpc_purge_queue(&local->event_queue);
439 }
440 
441 /*
442  * Process events on an endpoint.  The work item carries a ref which
443  * we must release.
444  */
445 static void rxrpc_local_processor(struct work_struct *work)
446 {
447 	struct rxrpc_local *local =
448 		container_of(work, struct rxrpc_local, processor);
449 	bool again;
450 
451 	trace_rxrpc_local(local->debug_id, rxrpc_local_processing,
452 			  atomic_read(&local->usage), NULL);
453 
454 	do {
455 		again = false;
456 		if (!__rxrpc_use_local(local)) {
457 			rxrpc_local_destroyer(local);
458 			break;
459 		}
460 
461 		if (!skb_queue_empty(&local->reject_queue)) {
462 			rxrpc_reject_packets(local);
463 			again = true;
464 		}
465 
466 		if (!skb_queue_empty(&local->event_queue)) {
467 			rxrpc_process_local_events(local);
468 			again = true;
469 		}
470 
471 		__rxrpc_unuse_local(local);
472 	} while (again);
473 
474 	rxrpc_put_local(local);
475 }
476 
477 /*
478  * Destroy a local endpoint after the RCU grace period expires.
479  */
480 static void rxrpc_local_rcu(struct rcu_head *rcu)
481 {
482 	struct rxrpc_local *local = container_of(rcu, struct rxrpc_local, rcu);
483 
484 	_enter("%d", local->debug_id);
485 
486 	ASSERT(!work_pending(&local->processor));
487 
488 	_net("DESTROY LOCAL %d", local->debug_id);
489 	kfree(local);
490 	_leave("");
491 }
492 
493 /*
494  * Verify the local endpoint list is empty by this point.
495  */
496 void rxrpc_destroy_all_locals(struct rxrpc_net *rxnet)
497 {
498 	struct rxrpc_local *local;
499 
500 	_enter("");
501 
502 	flush_workqueue(rxrpc_workqueue);
503 
504 	if (!list_empty(&rxnet->local_endpoints)) {
505 		mutex_lock(&rxnet->local_mutex);
506 		list_for_each_entry(local, &rxnet->local_endpoints, link) {
507 			pr_err("AF_RXRPC: Leaked local %p {%d}\n",
508 			       local, atomic_read(&local->usage));
509 		}
510 		mutex_unlock(&rxnet->local_mutex);
511 		BUG();
512 	}
513 }
514