1 // SPDX-License-Identifier: GPL-2.0-or-later 2 /* 3 * NET3: Garbage Collector For AF_UNIX sockets 4 * 5 * Garbage Collector: 6 * Copyright (C) Barak A. Pearlmutter. 7 * 8 * Chopped about by Alan Cox 22/3/96 to make it fit the AF_UNIX socket problem. 9 * If it doesn't work blame me, it worked when Barak sent it. 10 * 11 * Assumptions: 12 * 13 * - object w/ a bit 14 * - free list 15 * 16 * Current optimizations: 17 * 18 * - explicit stack instead of recursion 19 * - tail recurse on first born instead of immediate push/pop 20 * - we gather the stuff that should not be killed into tree 21 * and stack is just a path from root to the current pointer. 22 * 23 * Future optimizations: 24 * 25 * - don't just push entire root set; process in place 26 * 27 * Fixes: 28 * Alan Cox 07 Sept 1997 Vmalloc internal stack as needed. 29 * Cope with changing max_files. 30 * Al Viro 11 Oct 1998 31 * Graph may have cycles. That is, we can send the descriptor 32 * of foo to bar and vice versa. Current code chokes on that. 33 * Fix: move SCM_RIGHTS ones into the separate list and then 34 * skb_free() them all instead of doing explicit fput's. 35 * Another problem: since fput() may block somebody may 36 * create a new unix_socket when we are in the middle of sweep 37 * phase. Fix: revert the logic wrt MARKED. Mark everything 38 * upon the beginning and unmark non-junk ones. 39 * 40 * [12 Oct 1998] AAARGH! New code purges all SCM_RIGHTS 41 * sent to connect()'ed but still not accept()'ed sockets. 42 * Fixed. Old code had slightly different problem here: 43 * extra fput() in situation when we passed the descriptor via 44 * such socket and closed it (descriptor). That would happen on 45 * each unix_gc() until the accept(). Since the struct file in 46 * question would go to the free list and might be reused... 47 * That might be the reason of random oopses on filp_close() 48 * in unrelated processes. 49 * 50 * AV 28 Feb 1999 51 * Kill the explicit allocation of stack. Now we keep the tree 52 * with root in dummy + pointer (gc_current) to one of the nodes. 53 * Stack is represented as path from gc_current to dummy. Unmark 54 * now means "add to tree". Push == "make it a son of gc_current". 55 * Pop == "move gc_current to parent". We keep only pointers to 56 * parents (->gc_tree). 57 * AV 1 Mar 1999 58 * Damn. Added missing check for ->dead in listen queues scanning. 59 * 60 * Miklos Szeredi 25 Jun 2007 61 * Reimplement with a cycle collecting algorithm. This should 62 * solve several problems with the previous code, like being racy 63 * wrt receive and holding up unrelated socket operations. 64 */ 65 66 #include <linux/kernel.h> 67 #include <linux/string.h> 68 #include <linux/socket.h> 69 #include <linux/un.h> 70 #include <linux/net.h> 71 #include <linux/fs.h> 72 #include <linux/skbuff.h> 73 #include <linux/netdevice.h> 74 #include <linux/file.h> 75 #include <linux/proc_fs.h> 76 #include <linux/mutex.h> 77 #include <linux/wait.h> 78 79 #include <net/sock.h> 80 #include <net/af_unix.h> 81 #include <net/scm.h> 82 #include <net/tcp_states.h> 83 84 struct unix_sock *unix_get_socket(struct file *filp) 85 { 86 struct inode *inode = file_inode(filp); 87 88 /* Socket ? */ 89 if (S_ISSOCK(inode->i_mode) && !(filp->f_mode & FMODE_PATH)) { 90 struct socket *sock = SOCKET_I(inode); 91 const struct proto_ops *ops; 92 struct sock *sk = sock->sk; 93 94 ops = READ_ONCE(sock->ops); 95 96 /* PF_UNIX ? */ 97 if (sk && ops && ops->family == PF_UNIX) 98 return unix_sk(sk); 99 } 100 101 return NULL; 102 } 103 104 DEFINE_SPINLOCK(unix_gc_lock); 105 unsigned int unix_tot_inflight; 106 static LIST_HEAD(gc_candidates); 107 static LIST_HEAD(gc_inflight_list); 108 109 /* Keep the number of times in flight count for the file 110 * descriptor if it is for an AF_UNIX socket. 111 */ 112 void unix_inflight(struct user_struct *user, struct file *filp) 113 { 114 struct unix_sock *u = unix_get_socket(filp); 115 116 spin_lock(&unix_gc_lock); 117 118 if (u) { 119 if (!u->inflight) { 120 WARN_ON_ONCE(!list_empty(&u->link)); 121 list_add_tail(&u->link, &gc_inflight_list); 122 } else { 123 WARN_ON_ONCE(list_empty(&u->link)); 124 } 125 u->inflight++; 126 127 /* Paired with READ_ONCE() in wait_for_unix_gc() */ 128 WRITE_ONCE(unix_tot_inflight, unix_tot_inflight + 1); 129 } 130 131 WRITE_ONCE(user->unix_inflight, user->unix_inflight + 1); 132 133 spin_unlock(&unix_gc_lock); 134 } 135 136 void unix_notinflight(struct user_struct *user, struct file *filp) 137 { 138 struct unix_sock *u = unix_get_socket(filp); 139 140 spin_lock(&unix_gc_lock); 141 142 if (u) { 143 WARN_ON_ONCE(!u->inflight); 144 WARN_ON_ONCE(list_empty(&u->link)); 145 146 u->inflight--; 147 if (!u->inflight) 148 list_del_init(&u->link); 149 150 /* Paired with READ_ONCE() in wait_for_unix_gc() */ 151 WRITE_ONCE(unix_tot_inflight, unix_tot_inflight - 1); 152 } 153 154 WRITE_ONCE(user->unix_inflight, user->unix_inflight - 1); 155 156 spin_unlock(&unix_gc_lock); 157 } 158 159 static void scan_inflight(struct sock *x, void (*func)(struct unix_sock *), 160 struct sk_buff_head *hitlist) 161 { 162 struct sk_buff *skb; 163 struct sk_buff *next; 164 165 spin_lock(&x->sk_receive_queue.lock); 166 skb_queue_walk_safe(&x->sk_receive_queue, skb, next) { 167 /* Do we have file descriptors ? */ 168 if (UNIXCB(skb).fp) { 169 bool hit = false; 170 /* Process the descriptors of this socket */ 171 int nfd = UNIXCB(skb).fp->count; 172 struct file **fp = UNIXCB(skb).fp->fp; 173 174 while (nfd--) { 175 /* Get the socket the fd matches if it indeed does so */ 176 struct unix_sock *u = unix_get_socket(*fp++); 177 178 /* Ignore non-candidates, they could have been added 179 * to the queues after starting the garbage collection 180 */ 181 if (u && test_bit(UNIX_GC_CANDIDATE, &u->gc_flags)) { 182 hit = true; 183 184 func(u); 185 } 186 } 187 if (hit && hitlist != NULL) { 188 __skb_unlink(skb, &x->sk_receive_queue); 189 __skb_queue_tail(hitlist, skb); 190 } 191 } 192 } 193 spin_unlock(&x->sk_receive_queue.lock); 194 } 195 196 static void scan_children(struct sock *x, void (*func)(struct unix_sock *), 197 struct sk_buff_head *hitlist) 198 { 199 if (x->sk_state != TCP_LISTEN) { 200 scan_inflight(x, func, hitlist); 201 } else { 202 struct sk_buff *skb; 203 struct sk_buff *next; 204 struct unix_sock *u; 205 LIST_HEAD(embryos); 206 207 /* For a listening socket collect the queued embryos 208 * and perform a scan on them as well. 209 */ 210 spin_lock(&x->sk_receive_queue.lock); 211 skb_queue_walk_safe(&x->sk_receive_queue, skb, next) { 212 u = unix_sk(skb->sk); 213 214 /* An embryo cannot be in-flight, so it's safe 215 * to use the list link. 216 */ 217 WARN_ON_ONCE(!list_empty(&u->link)); 218 list_add_tail(&u->link, &embryos); 219 } 220 spin_unlock(&x->sk_receive_queue.lock); 221 222 while (!list_empty(&embryos)) { 223 u = list_entry(embryos.next, struct unix_sock, link); 224 scan_inflight(&u->sk, func, hitlist); 225 list_del_init(&u->link); 226 } 227 } 228 } 229 230 static void dec_inflight(struct unix_sock *usk) 231 { 232 usk->inflight--; 233 } 234 235 static void inc_inflight(struct unix_sock *usk) 236 { 237 usk->inflight++; 238 } 239 240 static void inc_inflight_move_tail(struct unix_sock *u) 241 { 242 u->inflight++; 243 244 /* If this still might be part of a cycle, move it to the end 245 * of the list, so that it's checked even if it was already 246 * passed over 247 */ 248 if (test_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags)) 249 list_move_tail(&u->link, &gc_candidates); 250 } 251 252 static bool gc_in_progress; 253 254 static void __unix_gc(struct work_struct *work) 255 { 256 struct sk_buff_head hitlist; 257 struct unix_sock *u, *next; 258 LIST_HEAD(not_cycle_list); 259 struct list_head cursor; 260 261 spin_lock(&unix_gc_lock); 262 263 /* First, select candidates for garbage collection. Only 264 * in-flight sockets are considered, and from those only ones 265 * which don't have any external reference. 266 * 267 * Holding unix_gc_lock will protect these candidates from 268 * being detached, and hence from gaining an external 269 * reference. Since there are no possible receivers, all 270 * buffers currently on the candidates' queues stay there 271 * during the garbage collection. 272 * 273 * We also know that no new candidate can be added onto the 274 * receive queues. Other, non candidate sockets _can_ be 275 * added to queue, so we must make sure only to touch 276 * candidates. 277 * 278 * Embryos, though never candidates themselves, affect which 279 * candidates are reachable by the garbage collector. Before 280 * being added to a listener's queue, an embryo may already 281 * receive data carrying SCM_RIGHTS, potentially making the 282 * passed socket a candidate that is not yet reachable by the 283 * collector. It becomes reachable once the embryo is 284 * enqueued. Therefore, we must ensure that no SCM-laden 285 * embryo appears in a (candidate) listener's queue between 286 * consecutive scan_children() calls. 287 */ 288 list_for_each_entry_safe(u, next, &gc_inflight_list, link) { 289 struct sock *sk = &u->sk; 290 long total_refs; 291 292 total_refs = file_count(sk->sk_socket->file); 293 294 WARN_ON_ONCE(!u->inflight); 295 WARN_ON_ONCE(total_refs < u->inflight); 296 if (total_refs == u->inflight) { 297 list_move_tail(&u->link, &gc_candidates); 298 __set_bit(UNIX_GC_CANDIDATE, &u->gc_flags); 299 __set_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags); 300 301 if (sk->sk_state == TCP_LISTEN) { 302 unix_state_lock(sk); 303 unix_state_unlock(sk); 304 } 305 } 306 } 307 308 /* Now remove all internal in-flight reference to children of 309 * the candidates. 310 */ 311 list_for_each_entry(u, &gc_candidates, link) 312 scan_children(&u->sk, dec_inflight, NULL); 313 314 /* Restore the references for children of all candidates, 315 * which have remaining references. Do this recursively, so 316 * only those remain, which form cyclic references. 317 * 318 * Use a "cursor" link, to make the list traversal safe, even 319 * though elements might be moved about. 320 */ 321 list_add(&cursor, &gc_candidates); 322 while (cursor.next != &gc_candidates) { 323 u = list_entry(cursor.next, struct unix_sock, link); 324 325 /* Move cursor to after the current position. */ 326 list_move(&cursor, &u->link); 327 328 if (u->inflight) { 329 list_move_tail(&u->link, ¬_cycle_list); 330 __clear_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags); 331 scan_children(&u->sk, inc_inflight_move_tail, NULL); 332 } 333 } 334 list_del(&cursor); 335 336 /* Now gc_candidates contains only garbage. Restore original 337 * inflight counters for these as well, and remove the skbuffs 338 * which are creating the cycle(s). 339 */ 340 skb_queue_head_init(&hitlist); 341 list_for_each_entry(u, &gc_candidates, link) { 342 scan_children(&u->sk, inc_inflight, &hitlist); 343 344 #if IS_ENABLED(CONFIG_AF_UNIX_OOB) 345 if (u->oob_skb) { 346 kfree_skb(u->oob_skb); 347 u->oob_skb = NULL; 348 } 349 #endif 350 } 351 352 /* not_cycle_list contains those sockets which do not make up a 353 * cycle. Restore these to the inflight list. 354 */ 355 while (!list_empty(¬_cycle_list)) { 356 u = list_entry(not_cycle_list.next, struct unix_sock, link); 357 __clear_bit(UNIX_GC_CANDIDATE, &u->gc_flags); 358 list_move_tail(&u->link, &gc_inflight_list); 359 } 360 361 spin_unlock(&unix_gc_lock); 362 363 /* Here we are. Hitlist is filled. Die. */ 364 __skb_queue_purge(&hitlist); 365 366 spin_lock(&unix_gc_lock); 367 368 /* All candidates should have been detached by now. */ 369 WARN_ON_ONCE(!list_empty(&gc_candidates)); 370 371 /* Paired with READ_ONCE() in wait_for_unix_gc(). */ 372 WRITE_ONCE(gc_in_progress, false); 373 374 spin_unlock(&unix_gc_lock); 375 } 376 377 static DECLARE_WORK(unix_gc_work, __unix_gc); 378 379 void unix_gc(void) 380 { 381 WRITE_ONCE(gc_in_progress, true); 382 queue_work(system_unbound_wq, &unix_gc_work); 383 } 384 385 #define UNIX_INFLIGHT_TRIGGER_GC 16000 386 #define UNIX_INFLIGHT_SANE_USER (SCM_MAX_FD * 8) 387 388 void wait_for_unix_gc(struct scm_fp_list *fpl) 389 { 390 /* If number of inflight sockets is insane, 391 * force a garbage collect right now. 392 * 393 * Paired with the WRITE_ONCE() in unix_inflight(), 394 * unix_notinflight(), and __unix_gc(). 395 */ 396 if (READ_ONCE(unix_tot_inflight) > UNIX_INFLIGHT_TRIGGER_GC && 397 !READ_ONCE(gc_in_progress)) 398 unix_gc(); 399 400 /* Penalise users who want to send AF_UNIX sockets 401 * but whose sockets have not been received yet. 402 */ 403 if (!fpl || !fpl->count_unix || 404 READ_ONCE(fpl->user->unix_inflight) < UNIX_INFLIGHT_SANE_USER) 405 return; 406 407 if (READ_ONCE(gc_in_progress)) 408 flush_work(&unix_gc_work); 409 } 410