1// SPDX-License-Identifier: GPL-2.0-or-later 2/* 3 * NETLINK Kernel-user communication protocol. 4 * 5 * Authors: Alan Cox <alan@lxorguk.ukuu.org.uk> 6 * Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> 7 * Patrick McHardy <kaber@trash.net> 8 * 9 * Tue Jun 26 14:36:48 MEST 2001 Herbert "herp" Rosmanith 10 * added netlink_proto_exit 11 * Tue Jan 22 18:32:44 BRST 2002 Arnaldo C. de Melo <acme@conectiva.com.br> 12 * use nlk_sk, as sk->protinfo is on a diet 8) 13 * Fri Jul 22 19:51:12 MEST 2005 Harald Welte <laforge@gnumonks.org> 14 * - inc module use count of module that owns 15 * the kernel socket in case userspace opens 16 * socket of same protocol 17 * - remove all module support, since netlink is 18 * mandatory if CONFIG_NET=y these days 19 */ 20 21#include <linux/module.h> 22 23#include <linux/capability.h> 24#include <linux/kernel.h> 25#include <linux/init.h> 26#include <linux/signal.h> 27#include <linux/sched.h> 28#include <linux/errno.h> 29#include <linux/string.h> 30#include <linux/stat.h> 31#include <linux/socket.h> 32#include <linux/un.h> 33#include <linux/fcntl.h> 34#include <linux/termios.h> 35#include <linux/sockios.h> 36#include <linux/net.h> 37#include <linux/fs.h> 38#include <linux/slab.h> 39#include <linux/uaccess.h> 40#include <linux/skbuff.h> 41#include <linux/netdevice.h> 42#include <linux/rtnetlink.h> 43#include <linux/proc_fs.h> 44#include <linux/seq_file.h> 45#include <linux/notifier.h> 46#include <linux/security.h> 47#include <linux/jhash.h> 48#include <linux/jiffies.h> 49#include <linux/random.h> 50#include <linux/bitops.h> 51#include <linux/mm.h> 52#include <linux/types.h> 53#include <linux/audit.h> 54#include <linux/mutex.h> 55#include <linux/vmalloc.h> 56#include <linux/if_arp.h> 57#include <linux/rhashtable.h> 58#include <asm/cacheflush.h> 59#include <linux/hash.h> 60#include <linux/genetlink.h> 61#include <linux/net_namespace.h> 62#include <linux/nospec.h> 63#include <linux/btf_ids.h> 64 65#include <net/net_namespace.h> 66#include <net/netns/generic.h> 67#include <net/sock.h> 68#include <net/scm.h> 69#include <net/netlink.h> 70 71#include "af_netlink.h" 72 73struct listeners { 74 struct rcu_head rcu; 75 unsigned long masks[]; 76}; 77 78/* state bits */ 79#define NETLINK_S_CONGESTED 0x0 80 81static inline int netlink_is_kernel(struct sock *sk) 82{ 83 return nlk_sk(sk)->flags & NETLINK_F_KERNEL_SOCKET; 84} 85 86struct netlink_table *nl_table __read_mostly; 87EXPORT_SYMBOL_GPL(nl_table); 88 89static DECLARE_WAIT_QUEUE_HEAD(nl_table_wait); 90 91static struct lock_class_key nlk_cb_mutex_keys[MAX_LINKS]; 92 93static const char *const nlk_cb_mutex_key_strings[MAX_LINKS + 1] = { 94 "nlk_cb_mutex-ROUTE", 95 "nlk_cb_mutex-1", 96 "nlk_cb_mutex-USERSOCK", 97 "nlk_cb_mutex-FIREWALL", 98 "nlk_cb_mutex-SOCK_DIAG", 99 "nlk_cb_mutex-NFLOG", 100 "nlk_cb_mutex-XFRM", 101 "nlk_cb_mutex-SELINUX", 102 "nlk_cb_mutex-ISCSI", 103 "nlk_cb_mutex-AUDIT", 104 "nlk_cb_mutex-FIB_LOOKUP", 105 "nlk_cb_mutex-CONNECTOR", 106 "nlk_cb_mutex-NETFILTER", 107 "nlk_cb_mutex-IP6_FW", 108 "nlk_cb_mutex-DNRTMSG", 109 "nlk_cb_mutex-KOBJECT_UEVENT", 110 "nlk_cb_mutex-GENERIC", 111 "nlk_cb_mutex-17", 112 "nlk_cb_mutex-SCSITRANSPORT", 113 "nlk_cb_mutex-ECRYPTFS", 114 "nlk_cb_mutex-RDMA", 115 "nlk_cb_mutex-CRYPTO", 116 "nlk_cb_mutex-SMC", 117 "nlk_cb_mutex-23", 118 "nlk_cb_mutex-24", 119 "nlk_cb_mutex-25", 120 "nlk_cb_mutex-26", 121 "nlk_cb_mutex-27", 122 "nlk_cb_mutex-28", 123 "nlk_cb_mutex-29", 124 "nlk_cb_mutex-30", 125 "nlk_cb_mutex-31", 126 "nlk_cb_mutex-MAX_LINKS" 127}; 128 129static int netlink_dump(struct sock *sk); 130 131/* nl_table locking explained: 132 * Lookup and traversal are protected with an RCU read-side lock. Insertion 133 * and removal are protected with per bucket lock while using RCU list 134 * modification primitives and may run in parallel to RCU protected lookups. 135 * Destruction of the Netlink socket may only occur *after* nl_table_lock has 136 * been acquired * either during or after the socket has been removed from 137 * the list and after an RCU grace period. 138 */ 139DEFINE_RWLOCK(nl_table_lock); 140EXPORT_SYMBOL_GPL(nl_table_lock); 141static atomic_t nl_table_users = ATOMIC_INIT(0); 142 143#define nl_deref_protected(X) rcu_dereference_protected(X, lockdep_is_held(&nl_table_lock)); 144 145static BLOCKING_NOTIFIER_HEAD(netlink_chain); 146 147 148static const struct rhashtable_params netlink_rhashtable_params; 149 150static inline u32 netlink_group_mask(u32 group) 151{ 152 if (group > 32) 153 return 0; 154 return group ? 1 << (group - 1) : 0; 155} 156 157static struct sk_buff *netlink_to_full_skb(const struct sk_buff *skb, 158 gfp_t gfp_mask) 159{ 160 unsigned int len = skb->len; 161 struct sk_buff *new; 162 163 new = alloc_skb(len, gfp_mask); 164 if (new == NULL) 165 return NULL; 166 167 NETLINK_CB(new).portid = NETLINK_CB(skb).portid; 168 NETLINK_CB(new).dst_group = NETLINK_CB(skb).dst_group; 169 NETLINK_CB(new).creds = NETLINK_CB(skb).creds; 170 171 skb_put_data(new, skb->data, len); 172 return new; 173} 174 175static unsigned int netlink_tap_net_id; 176 177struct netlink_tap_net { 178 struct list_head netlink_tap_all; 179 struct mutex netlink_tap_lock; 180}; 181 182int netlink_add_tap(struct netlink_tap *nt) 183{ 184 struct net *net = dev_net(nt->dev); 185 struct netlink_tap_net *nn = net_generic(net, netlink_tap_net_id); 186 187 if (unlikely(nt->dev->type != ARPHRD_NETLINK)) 188 return -EINVAL; 189 190 mutex_lock(&nn->netlink_tap_lock); 191 list_add_rcu(&nt->list, &nn->netlink_tap_all); 192 mutex_unlock(&nn->netlink_tap_lock); 193 194 __module_get(nt->module); 195 196 return 0; 197} 198EXPORT_SYMBOL_GPL(netlink_add_tap); 199 200static int __netlink_remove_tap(struct netlink_tap *nt) 201{ 202 struct net *net = dev_net(nt->dev); 203 struct netlink_tap_net *nn = net_generic(net, netlink_tap_net_id); 204 bool found = false; 205 struct netlink_tap *tmp; 206 207 mutex_lock(&nn->netlink_tap_lock); 208 209 list_for_each_entry(tmp, &nn->netlink_tap_all, list) { 210 if (nt == tmp) { 211 list_del_rcu(&nt->list); 212 found = true; 213 goto out; 214 } 215 } 216 217 pr_warn("__netlink_remove_tap: %p not found\n", nt); 218out: 219 mutex_unlock(&nn->netlink_tap_lock); 220 221 if (found) 222 module_put(nt->module); 223 224 return found ? 0 : -ENODEV; 225} 226 227int netlink_remove_tap(struct netlink_tap *nt) 228{ 229 int ret; 230 231 ret = __netlink_remove_tap(nt); 232 synchronize_net(); 233 234 return ret; 235} 236EXPORT_SYMBOL_GPL(netlink_remove_tap); 237 238static __net_init int netlink_tap_init_net(struct net *net) 239{ 240 struct netlink_tap_net *nn = net_generic(net, netlink_tap_net_id); 241 242 INIT_LIST_HEAD(&nn->netlink_tap_all); 243 mutex_init(&nn->netlink_tap_lock); 244 return 0; 245} 246 247static struct pernet_operations netlink_tap_net_ops = { 248 .init = netlink_tap_init_net, 249 .id = &netlink_tap_net_id, 250 .size = sizeof(struct netlink_tap_net), 251}; 252 253static bool netlink_filter_tap(const struct sk_buff *skb) 254{ 255 struct sock *sk = skb->sk; 256 257 /* We take the more conservative approach and 258 * whitelist socket protocols that may pass. 259 */ 260 switch (sk->sk_protocol) { 261 case NETLINK_ROUTE: 262 case NETLINK_USERSOCK: 263 case NETLINK_SOCK_DIAG: 264 case NETLINK_NFLOG: 265 case NETLINK_XFRM: 266 case NETLINK_FIB_LOOKUP: 267 case NETLINK_NETFILTER: 268 case NETLINK_GENERIC: 269 return true; 270 } 271 272 return false; 273} 274 275static int __netlink_deliver_tap_skb(struct sk_buff *skb, 276 struct net_device *dev) 277{ 278 struct sk_buff *nskb; 279 struct sock *sk = skb->sk; 280 int ret = -ENOMEM; 281 282 if (!net_eq(dev_net(dev), sock_net(sk))) 283 return 0; 284 285 dev_hold(dev); 286 287 if (is_vmalloc_addr(skb->head)) 288 nskb = netlink_to_full_skb(skb, GFP_ATOMIC); 289 else 290 nskb = skb_clone(skb, GFP_ATOMIC); 291 if (nskb) { 292 nskb->dev = dev; 293 nskb->protocol = htons((u16) sk->sk_protocol); 294 nskb->pkt_type = netlink_is_kernel(sk) ? 295 PACKET_KERNEL : PACKET_USER; 296 skb_reset_network_header(nskb); 297 ret = dev_queue_xmit(nskb); 298 if (unlikely(ret > 0)) 299 ret = net_xmit_errno(ret); 300 } 301 302 dev_put(dev); 303 return ret; 304} 305 306static void __netlink_deliver_tap(struct sk_buff *skb, struct netlink_tap_net *nn) 307{ 308 int ret; 309 struct netlink_tap *tmp; 310 311 if (!netlink_filter_tap(skb)) 312 return; 313 314 list_for_each_entry_rcu(tmp, &nn->netlink_tap_all, list) { 315 ret = __netlink_deliver_tap_skb(skb, tmp->dev); 316 if (unlikely(ret)) 317 break; 318 } 319} 320 321static void netlink_deliver_tap(struct net *net, struct sk_buff *skb) 322{ 323 struct netlink_tap_net *nn = net_generic(net, netlink_tap_net_id); 324 325 rcu_read_lock(); 326 327 if (unlikely(!list_empty(&nn->netlink_tap_all))) 328 __netlink_deliver_tap(skb, nn); 329 330 rcu_read_unlock(); 331} 332 333static void netlink_deliver_tap_kernel(struct sock *dst, struct sock *src, 334 struct sk_buff *skb) 335{ 336 if (!(netlink_is_kernel(dst) && netlink_is_kernel(src))) 337 netlink_deliver_tap(sock_net(dst), skb); 338} 339 340static void netlink_overrun(struct sock *sk) 341{ 342 struct netlink_sock *nlk = nlk_sk(sk); 343 344 if (!(nlk->flags & NETLINK_F_RECV_NO_ENOBUFS)) { 345 if (!test_and_set_bit(NETLINK_S_CONGESTED, 346 &nlk_sk(sk)->state)) { 347 sk->sk_err = ENOBUFS; 348 sk->sk_error_report(sk); 349 } 350 } 351 atomic_inc(&sk->sk_drops); 352} 353 354static void netlink_rcv_wake(struct sock *sk) 355{ 356 struct netlink_sock *nlk = nlk_sk(sk); 357 358 if (skb_queue_empty_lockless(&sk->sk_receive_queue)) 359 clear_bit(NETLINK_S_CONGESTED, &nlk->state); 360 if (!test_bit(NETLINK_S_CONGESTED, &nlk->state)) 361 wake_up_interruptible(&nlk->wait); 362} 363 364static void netlink_skb_destructor(struct sk_buff *skb) 365{ 366 if (is_vmalloc_addr(skb->head)) { 367 if (!skb->cloned || 368 !atomic_dec_return(&(skb_shinfo(skb)->dataref))) 369 vfree_atomic(skb->head); 370 371 skb->head = NULL; 372 } 373 if (skb->sk != NULL) 374 sock_rfree(skb); 375} 376 377static void netlink_skb_set_owner_r(struct sk_buff *skb, struct sock *sk) 378{ 379 WARN_ON(skb->sk != NULL); 380 skb->sk = sk; 381 skb->destructor = netlink_skb_destructor; 382 atomic_add(skb->truesize, &sk->sk_rmem_alloc); 383 sk_mem_charge(sk, skb->truesize); 384} 385 386static void netlink_sock_destruct(struct sock *sk) 387{ 388 struct netlink_sock *nlk = nlk_sk(sk); 389 390 if (nlk->cb_running) { 391 if (nlk->cb.done) 392 nlk->cb.done(&nlk->cb); 393 module_put(nlk->cb.module); 394 kfree_skb(nlk->cb.skb); 395 } 396 397 skb_queue_purge(&sk->sk_receive_queue); 398 399 if (!sock_flag(sk, SOCK_DEAD)) { 400 printk(KERN_ERR "Freeing alive netlink socket %p\n", sk); 401 return; 402 } 403 404 WARN_ON(atomic_read(&sk->sk_rmem_alloc)); 405 WARN_ON(refcount_read(&sk->sk_wmem_alloc)); 406 WARN_ON(nlk_sk(sk)->groups); 407} 408 409static void netlink_sock_destruct_work(struct work_struct *work) 410{ 411 struct netlink_sock *nlk = container_of(work, struct netlink_sock, 412 work); 413 414 sk_free(&nlk->sk); 415} 416 417/* This lock without WQ_FLAG_EXCLUSIVE is good on UP and it is _very_ bad on 418 * SMP. Look, when several writers sleep and reader wakes them up, all but one 419 * immediately hit write lock and grab all the cpus. Exclusive sleep solves 420 * this, _but_ remember, it adds useless work on UP machines. 421 */ 422 423void netlink_table_grab(void) 424 __acquires(nl_table_lock) 425{ 426 might_sleep(); 427 428 write_lock_irq(&nl_table_lock); 429 430 if (atomic_read(&nl_table_users)) { 431 DECLARE_WAITQUEUE(wait, current); 432 433 add_wait_queue_exclusive(&nl_table_wait, &wait); 434 for (;;) { 435 set_current_state(TASK_UNINTERRUPTIBLE); 436 if (atomic_read(&nl_table_users) == 0) 437 break; 438 write_unlock_irq(&nl_table_lock); 439 schedule(); 440 write_lock_irq(&nl_table_lock); 441 } 442 443 __set_current_state(TASK_RUNNING); 444 remove_wait_queue(&nl_table_wait, &wait); 445 } 446} 447 448void netlink_table_ungrab(void) 449 __releases(nl_table_lock) 450{ 451 write_unlock_irq(&nl_table_lock); 452 wake_up(&nl_table_wait); 453} 454 455static inline void 456netlink_lock_table(void) 457{ 458 unsigned long flags; 459 460 /* read_lock() synchronizes us to netlink_table_grab */ 461 462 read_lock_irqsave(&nl_table_lock, flags); 463 atomic_inc(&nl_table_users); 464 read_unlock_irqrestore(&nl_table_lock, flags); 465} 466 467static inline void 468netlink_unlock_table(void) 469{ 470 if (atomic_dec_and_test(&nl_table_users)) 471 wake_up(&nl_table_wait); 472} 473 474struct netlink_compare_arg 475{ 476 possible_net_t pnet; 477 u32 portid; 478}; 479 480/* Doing sizeof directly may yield 4 extra bytes on 64-bit. */ 481#define netlink_compare_arg_len \ 482 (offsetof(struct netlink_compare_arg, portid) + sizeof(u32)) 483 484static inline int netlink_compare(struct rhashtable_compare_arg *arg, 485 const void *ptr) 486{ 487 const struct netlink_compare_arg *x = arg->key; 488 const struct netlink_sock *nlk = ptr; 489 490 return nlk->portid != x->portid || 491 !net_eq(sock_net(&nlk->sk), read_pnet(&x->pnet)); 492} 493 494static void netlink_compare_arg_init(struct netlink_compare_arg *arg, 495 struct net *net, u32 portid) 496{ 497 memset(arg, 0, sizeof(*arg)); 498 write_pnet(&arg->pnet, net); 499 arg->portid = portid; 500} 501 502static struct sock *__netlink_lookup(struct netlink_table *table, u32 portid, 503 struct net *net) 504{ 505 struct netlink_compare_arg arg; 506 507 netlink_compare_arg_init(&arg, net, portid); 508 return rhashtable_lookup_fast(&table->hash, &arg, 509 netlink_rhashtable_params); 510} 511 512static int __netlink_insert(struct netlink_table *table, struct sock *sk) 513{ 514 struct netlink_compare_arg arg; 515 516 netlink_compare_arg_init(&arg, sock_net(sk), nlk_sk(sk)->portid); 517 return rhashtable_lookup_insert_key(&table->hash, &arg, 518 &nlk_sk(sk)->node, 519 netlink_rhashtable_params); 520} 521 522static struct sock *netlink_lookup(struct net *net, int protocol, u32 portid) 523{ 524 struct netlink_table *table = &nl_table[protocol]; 525 struct sock *sk; 526 527 rcu_read_lock(); 528 sk = __netlink_lookup(table, portid, net); 529 if (sk) 530 sock_hold(sk); 531 rcu_read_unlock(); 532 533 return sk; 534} 535 536static const struct proto_ops netlink_ops; 537 538static void 539netlink_update_listeners(struct sock *sk) 540{ 541 struct netlink_table *tbl = &nl_table[sk->sk_protocol]; 542 unsigned long mask; 543 unsigned int i; 544 struct listeners *listeners; 545 546 listeners = nl_deref_protected(tbl->listeners); 547 if (!listeners) 548 return; 549 550 for (i = 0; i < NLGRPLONGS(tbl->groups); i++) { 551 mask = 0; 552 sk_for_each_bound(sk, &tbl->mc_list) { 553 if (i < NLGRPLONGS(nlk_sk(sk)->ngroups)) 554 mask |= nlk_sk(sk)->groups[i]; 555 } 556 listeners->masks[i] = mask; 557 } 558 /* this function is only called with the netlink table "grabbed", which 559 * makes sure updates are visible before bind or setsockopt return. */ 560} 561 562static int netlink_insert(struct sock *sk, u32 portid) 563{ 564 struct netlink_table *table = &nl_table[sk->sk_protocol]; 565 int err; 566 567 lock_sock(sk); 568 569 err = nlk_sk(sk)->portid == portid ? 0 : -EBUSY; 570 if (nlk_sk(sk)->bound) 571 goto err; 572 573 /* portid can be read locklessly from netlink_getname(). */ 574 WRITE_ONCE(nlk_sk(sk)->portid, portid); 575 576 sock_hold(sk); 577 578 err = __netlink_insert(table, sk); 579 if (err) { 580 /* In case the hashtable backend returns with -EBUSY 581 * from here, it must not escape to the caller. 582 */ 583 if (unlikely(err == -EBUSY)) 584 err = -EOVERFLOW; 585 if (err == -EEXIST) 586 err = -EADDRINUSE; 587 sock_put(sk); 588 goto err; 589 } 590 591 /* We need to ensure that the socket is hashed and visible. */ 592 smp_wmb(); 593 /* Paired with lockless reads from netlink_bind(), 594 * netlink_connect() and netlink_sendmsg(). 595 */ 596 WRITE_ONCE(nlk_sk(sk)->bound, portid); 597 598err: 599 release_sock(sk); 600 return err; 601} 602 603static void netlink_remove(struct sock *sk) 604{ 605 struct netlink_table *table; 606 607 table = &nl_table[sk->sk_protocol]; 608 if (!rhashtable_remove_fast(&table->hash, &nlk_sk(sk)->node, 609 netlink_rhashtable_params)) { 610 WARN_ON(refcount_read(&sk->sk_refcnt) == 1); 611 __sock_put(sk); 612 } 613 614 netlink_table_grab(); 615 if (nlk_sk(sk)->subscriptions) { 616 __sk_del_bind_node(sk); 617 netlink_update_listeners(sk); 618 } 619 if (sk->sk_protocol == NETLINK_GENERIC) 620 atomic_inc(&genl_sk_destructing_cnt); 621 netlink_table_ungrab(); 622} 623 624static struct proto netlink_proto = { 625 .name = "NETLINK", 626 .owner = THIS_MODULE, 627 .obj_size = sizeof(struct netlink_sock), 628}; 629 630static int __netlink_create(struct net *net, struct socket *sock, 631 struct mutex *cb_mutex, int protocol, 632 int kern) 633{ 634 struct sock *sk; 635 struct netlink_sock *nlk; 636 637 sock->ops = &netlink_ops; 638 639 sk = sk_alloc(net, PF_NETLINK, GFP_KERNEL, &netlink_proto, kern); 640 if (!sk) 641 return -ENOMEM; 642 643 sock_init_data(sock, sk); 644 645 nlk = nlk_sk(sk); 646 if (cb_mutex) { 647 nlk->cb_mutex = cb_mutex; 648 } else { 649 nlk->cb_mutex = &nlk->cb_def_mutex; 650 mutex_init(nlk->cb_mutex); 651 lockdep_set_class_and_name(nlk->cb_mutex, 652 nlk_cb_mutex_keys + protocol, 653 nlk_cb_mutex_key_strings[protocol]); 654 } 655 init_waitqueue_head(&nlk->wait); 656 657 sk->sk_destruct = netlink_sock_destruct; 658 sk->sk_protocol = protocol; 659 return 0; 660} 661 662static int netlink_create(struct net *net, struct socket *sock, int protocol, 663 int kern) 664{ 665 struct module *module = NULL; 666 struct mutex *cb_mutex; 667 struct netlink_sock *nlk; 668 int (*bind)(struct net *net, int group); 669 void (*unbind)(struct net *net, int group); 670 int err = 0; 671 672 sock->state = SS_UNCONNECTED; 673 674 if (sock->type != SOCK_RAW && sock->type != SOCK_DGRAM) 675 return -ESOCKTNOSUPPORT; 676 677 if (protocol < 0 || protocol >= MAX_LINKS) 678 return -EPROTONOSUPPORT; 679 protocol = array_index_nospec(protocol, MAX_LINKS); 680 681 netlink_lock_table(); 682#ifdef CONFIG_MODULES 683 if (!nl_table[protocol].registered) { 684 netlink_unlock_table(); 685 request_module("net-pf-%d-proto-%d", PF_NETLINK, protocol); 686 netlink_lock_table(); 687 } 688#endif 689 if (nl_table[protocol].registered && 690 try_module_get(nl_table[protocol].module)) 691 module = nl_table[protocol].module; 692 else 693 err = -EPROTONOSUPPORT; 694 cb_mutex = nl_table[protocol].cb_mutex; 695 bind = nl_table[protocol].bind; 696 unbind = nl_table[protocol].unbind; 697 netlink_unlock_table(); 698 699 if (err < 0) 700 goto out; 701 702 err = __netlink_create(net, sock, cb_mutex, protocol, kern); 703 if (err < 0) 704 goto out_module; 705 706 local_bh_disable(); 707 sock_prot_inuse_add(net, &netlink_proto, 1); 708 local_bh_enable(); 709 710 nlk = nlk_sk(sock->sk); 711 nlk->module = module; 712 nlk->netlink_bind = bind; 713 nlk->netlink_unbind = unbind; 714out: 715 return err; 716 717out_module: 718 module_put(module); 719 goto out; 720} 721 722static void deferred_put_nlk_sk(struct rcu_head *head) 723{ 724 struct netlink_sock *nlk = container_of(head, struct netlink_sock, rcu); 725 struct sock *sk = &nlk->sk; 726 727 kfree(nlk->groups); 728 nlk->groups = NULL; 729 730 if (!refcount_dec_and_test(&sk->sk_refcnt)) 731 return; 732 733 if (nlk->cb_running && nlk->cb.done) { 734 INIT_WORK(&nlk->work, netlink_sock_destruct_work); 735 schedule_work(&nlk->work); 736 return; 737 } 738 739 sk_free(sk); 740} 741 742static int netlink_release(struct socket *sock) 743{ 744 struct sock *sk = sock->sk; 745 struct netlink_sock *nlk; 746 747 if (!sk) 748 return 0; 749 750 netlink_remove(sk); 751 sock_orphan(sk); 752 nlk = nlk_sk(sk); 753 754 /* 755 * OK. Socket is unlinked, any packets that arrive now 756 * will be purged. 757 */ 758 759 /* must not acquire netlink_table_lock in any way again before unbind 760 * and notifying genetlink is done as otherwise it might deadlock 761 */ 762 if (nlk->netlink_unbind) { 763 int i; 764 765 for (i = 0; i < nlk->ngroups; i++) 766 if (test_bit(i, nlk->groups)) 767 nlk->netlink_unbind(sock_net(sk), i + 1); 768 } 769 if (sk->sk_protocol == NETLINK_GENERIC && 770 atomic_dec_return(&genl_sk_destructing_cnt) == 0) 771 wake_up(&genl_sk_destructing_waitq); 772 773 sock->sk = NULL; 774 wake_up_interruptible_all(&nlk->wait); 775 776 skb_queue_purge(&sk->sk_write_queue); 777 778 if (nlk->portid && nlk->bound) { 779 struct netlink_notify n = { 780 .net = sock_net(sk), 781 .protocol = sk->sk_protocol, 782 .portid = nlk->portid, 783 }; 784 blocking_notifier_call_chain(&netlink_chain, 785 NETLINK_URELEASE, &n); 786 } 787 788 module_put(nlk->module); 789 790 if (netlink_is_kernel(sk)) { 791 netlink_table_grab(); 792 BUG_ON(nl_table[sk->sk_protocol].registered == 0); 793 if (--nl_table[sk->sk_protocol].registered == 0) { 794 struct listeners *old; 795 796 old = nl_deref_protected(nl_table[sk->sk_protocol].listeners); 797 RCU_INIT_POINTER(nl_table[sk->sk_protocol].listeners, NULL); 798 kfree_rcu(old, rcu); 799 nl_table[sk->sk_protocol].module = NULL; 800 nl_table[sk->sk_protocol].bind = NULL; 801 nl_table[sk->sk_protocol].unbind = NULL; 802 nl_table[sk->sk_protocol].flags = 0; 803 nl_table[sk->sk_protocol].registered = 0; 804 } 805 netlink_table_ungrab(); 806 } 807 808 local_bh_disable(); 809 sock_prot_inuse_add(sock_net(sk), &netlink_proto, -1); 810 local_bh_enable(); 811 call_rcu(&nlk->rcu, deferred_put_nlk_sk); 812 return 0; 813} 814 815static int netlink_autobind(struct socket *sock) 816{ 817 struct sock *sk = sock->sk; 818 struct net *net = sock_net(sk); 819 struct netlink_table *table = &nl_table[sk->sk_protocol]; 820 s32 portid = task_tgid_vnr(current); 821 int err; 822 s32 rover = -4096; 823 bool ok; 824 825retry: 826 cond_resched(); 827 rcu_read_lock(); 828 ok = !__netlink_lookup(table, portid, net); 829 rcu_read_unlock(); 830 if (!ok) { 831 /* Bind collision, search negative portid values. */ 832 if (rover == -4096) 833 /* rover will be in range [S32_MIN, -4097] */ 834 rover = S32_MIN + prandom_u32_max(-4096 - S32_MIN); 835 else if (rover >= -4096) 836 rover = -4097; 837 portid = rover--; 838 goto retry; 839 } 840 841 err = netlink_insert(sk, portid); 842 if (err == -EADDRINUSE) 843 goto retry; 844 845 /* If 2 threads race to autobind, that is fine. */ 846 if (err == -EBUSY) 847 err = 0; 848 849 return err; 850} 851 852/** 853 * __netlink_ns_capable - General netlink message capability test 854 * @nsp: NETLINK_CB of the socket buffer holding a netlink command from userspace. 855 * @user_ns: The user namespace of the capability to use 856 * @cap: The capability to use 857 * 858 * Test to see if the opener of the socket we received the message 859 * from had when the netlink socket was created and the sender of the 860 * message has the capability @cap in the user namespace @user_ns. 861 */ 862bool __netlink_ns_capable(const struct netlink_skb_parms *nsp, 863 struct user_namespace *user_ns, int cap) 864{ 865 return ((nsp->flags & NETLINK_SKB_DST) || 866 file_ns_capable(nsp->sk->sk_socket->file, user_ns, cap)) && 867 ns_capable(user_ns, cap); 868} 869EXPORT_SYMBOL(__netlink_ns_capable); 870 871/** 872 * netlink_ns_capable - General netlink message capability test 873 * @skb: socket buffer holding a netlink command from userspace 874 * @user_ns: The user namespace of the capability to use 875 * @cap: The capability to use 876 * 877 * Test to see if the opener of the socket we received the message 878 * from had when the netlink socket was created and the sender of the 879 * message has the capability @cap in the user namespace @user_ns. 880 */ 881bool netlink_ns_capable(const struct sk_buff *skb, 882 struct user_namespace *user_ns, int cap) 883{ 884 return __netlink_ns_capable(&NETLINK_CB(skb), user_ns, cap); 885} 886EXPORT_SYMBOL(netlink_ns_capable); 887 888/** 889 * netlink_capable - Netlink global message capability test 890 * @skb: socket buffer holding a netlink command from userspace 891 * @cap: The capability to use 892 * 893 * Test to see if the opener of the socket we received the message 894 * from had when the netlink socket was created and the sender of the 895 * message has the capability @cap in all user namespaces. 896 */ 897bool netlink_capable(const struct sk_buff *skb, int cap) 898{ 899 return netlink_ns_capable(skb, &init_user_ns, cap); 900} 901EXPORT_SYMBOL(netlink_capable); 902 903/** 904 * netlink_net_capable - Netlink network namespace message capability test 905 * @skb: socket buffer holding a netlink command from userspace 906 * @cap: The capability to use 907 * 908 * Test to see if the opener of the socket we received the message 909 * from had when the netlink socket was created and the sender of the 910 * message has the capability @cap over the network namespace of 911 * the socket we received the message from. 912 */ 913bool netlink_net_capable(const struct sk_buff *skb, int cap) 914{ 915 return netlink_ns_capable(skb, sock_net(skb->sk)->user_ns, cap); 916} 917EXPORT_SYMBOL(netlink_net_capable); 918 919static inline int netlink_allowed(const struct socket *sock, unsigned int flag) 920{ 921 return (nl_table[sock->sk->sk_protocol].flags & flag) || 922 ns_capable(sock_net(sock->sk)->user_ns, CAP_NET_ADMIN); 923} 924 925static void 926netlink_update_subscriptions(struct sock *sk, unsigned int subscriptions) 927{ 928 struct netlink_sock *nlk = nlk_sk(sk); 929 930 if (nlk->subscriptions && !subscriptions) 931 __sk_del_bind_node(sk); 932 else if (!nlk->subscriptions && subscriptions) 933 sk_add_bind_node(sk, &nl_table[sk->sk_protocol].mc_list); 934 nlk->subscriptions = subscriptions; 935} 936 937static int netlink_realloc_groups(struct sock *sk) 938{ 939 struct netlink_sock *nlk = nlk_sk(sk); 940 unsigned int groups; 941 unsigned long *new_groups; 942 int err = 0; 943 944 netlink_table_grab(); 945 946 groups = nl_table[sk->sk_protocol].groups; 947 if (!nl_table[sk->sk_protocol].registered) { 948 err = -ENOENT; 949 goto out_unlock; 950 } 951 952 if (nlk->ngroups >= groups) 953 goto out_unlock; 954 955 new_groups = krealloc(nlk->groups, NLGRPSZ(groups), GFP_ATOMIC); 956 if (new_groups == NULL) { 957 err = -ENOMEM; 958 goto out_unlock; 959 } 960 memset((char *)new_groups + NLGRPSZ(nlk->ngroups), 0, 961 NLGRPSZ(groups) - NLGRPSZ(nlk->ngroups)); 962 963 nlk->groups = new_groups; 964 nlk->ngroups = groups; 965 out_unlock: 966 netlink_table_ungrab(); 967 return err; 968} 969 970static void netlink_undo_bind(int group, long unsigned int groups, 971 struct sock *sk) 972{ 973 struct netlink_sock *nlk = nlk_sk(sk); 974 int undo; 975 976 if (!nlk->netlink_unbind) 977 return; 978 979 for (undo = 0; undo < group; undo++) 980 if (test_bit(undo, &groups)) 981 nlk->netlink_unbind(sock_net(sk), undo + 1); 982} 983 984static int netlink_bind(struct socket *sock, struct sockaddr *addr, 985 int addr_len) 986{ 987 struct sock *sk = sock->sk; 988 struct net *net = sock_net(sk); 989 struct netlink_sock *nlk = nlk_sk(sk); 990 struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr; 991 int err = 0; 992 unsigned long groups; 993 bool bound; 994 995 if (addr_len < sizeof(struct sockaddr_nl)) 996 return -EINVAL; 997 998 if (nladdr->nl_family != AF_NETLINK) 999 return -EINVAL; 1000 groups = nladdr->nl_groups; 1001 1002 /* Only superuser is allowed to listen multicasts */ 1003 if (groups) { 1004 if (!netlink_allowed(sock, NL_CFG_F_NONROOT_RECV)) 1005 return -EPERM; 1006 err = netlink_realloc_groups(sk); 1007 if (err) 1008 return err; 1009 } 1010 1011 if (nlk->ngroups < BITS_PER_LONG) 1012 groups &= (1UL << nlk->ngroups) - 1; 1013 1014 /* Paired with WRITE_ONCE() in netlink_insert() */ 1015 bound = READ_ONCE(nlk->bound); 1016 if (bound) { 1017 /* Ensure nlk->portid is up-to-date. */ 1018 smp_rmb(); 1019 1020 if (nladdr->nl_pid != nlk->portid) 1021 return -EINVAL; 1022 } 1023 1024 if (nlk->netlink_bind && groups) { 1025 int group; 1026 1027 /* nl_groups is a u32, so cap the maximum groups we can bind */ 1028 for (group = 0; group < BITS_PER_TYPE(u32); group++) { 1029 if (!test_bit(group, &groups)) 1030 continue; 1031 err = nlk->netlink_bind(net, group + 1); 1032 if (!err) 1033 continue; 1034 netlink_undo_bind(group, groups, sk); 1035 return err; 1036 } 1037 } 1038 1039 /* No need for barriers here as we return to user-space without 1040 * using any of the bound attributes. 1041 */ 1042 netlink_lock_table(); 1043 if (!bound) { 1044 err = nladdr->nl_pid ? 1045 netlink_insert(sk, nladdr->nl_pid) : 1046 netlink_autobind(sock); 1047 if (err) { 1048 netlink_undo_bind(BITS_PER_TYPE(u32), groups, sk); 1049 goto unlock; 1050 } 1051 } 1052 1053 if (!groups && (nlk->groups == NULL || !(u32)nlk->groups[0])) 1054 goto unlock; 1055 netlink_unlock_table(); 1056 1057 netlink_table_grab(); 1058 netlink_update_subscriptions(sk, nlk->subscriptions + 1059 hweight32(groups) - 1060 hweight32(nlk->groups[0])); 1061 nlk->groups[0] = (nlk->groups[0] & ~0xffffffffUL) | groups; 1062 netlink_update_listeners(sk); 1063 netlink_table_ungrab(); 1064 1065 return 0; 1066 1067unlock: 1068 netlink_unlock_table(); 1069 return err; 1070} 1071 1072static int netlink_connect(struct socket *sock, struct sockaddr *addr, 1073 int alen, int flags) 1074{ 1075 int err = 0; 1076 struct sock *sk = sock->sk; 1077 struct netlink_sock *nlk = nlk_sk(sk); 1078 struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr; 1079 1080 if (alen < sizeof(addr->sa_family)) 1081 return -EINVAL; 1082 1083 if (addr->sa_family == AF_UNSPEC) { 1084 /* paired with READ_ONCE() in netlink_getsockbyportid() */ 1085 WRITE_ONCE(sk->sk_state, NETLINK_UNCONNECTED); 1086 /* dst_portid and dst_group can be read locklessly */ 1087 WRITE_ONCE(nlk->dst_portid, 0); 1088 WRITE_ONCE(nlk->dst_group, 0); 1089 return 0; 1090 } 1091 if (addr->sa_family != AF_NETLINK) 1092 return -EINVAL; 1093 1094 if (alen < sizeof(struct sockaddr_nl)) 1095 return -EINVAL; 1096 1097 if ((nladdr->nl_groups || nladdr->nl_pid) && 1098 !netlink_allowed(sock, NL_CFG_F_NONROOT_SEND)) 1099 return -EPERM; 1100 1101 /* No need for barriers here as we return to user-space without 1102 * using any of the bound attributes. 1103 * Paired with WRITE_ONCE() in netlink_insert(). 1104 */ 1105 if (!READ_ONCE(nlk->bound)) 1106 err = netlink_autobind(sock); 1107 1108 if (err == 0) { 1109 /* paired with READ_ONCE() in netlink_getsockbyportid() */ 1110 WRITE_ONCE(sk->sk_state, NETLINK_CONNECTED); 1111 /* dst_portid and dst_group can be read locklessly */ 1112 WRITE_ONCE(nlk->dst_portid, nladdr->nl_pid); 1113 WRITE_ONCE(nlk->dst_group, ffs(nladdr->nl_groups)); 1114 } 1115 1116 return err; 1117} 1118 1119static int netlink_getname(struct socket *sock, struct sockaddr *addr, 1120 int peer) 1121{ 1122 struct sock *sk = sock->sk; 1123 struct netlink_sock *nlk = nlk_sk(sk); 1124 DECLARE_SOCKADDR(struct sockaddr_nl *, nladdr, addr); 1125 1126 nladdr->nl_family = AF_NETLINK; 1127 nladdr->nl_pad = 0; 1128 1129 if (peer) { 1130 /* Paired with WRITE_ONCE() in netlink_connect() */ 1131 nladdr->nl_pid = READ_ONCE(nlk->dst_portid); 1132 nladdr->nl_groups = netlink_group_mask(READ_ONCE(nlk->dst_group)); 1133 } else { 1134 /* Paired with WRITE_ONCE() in netlink_insert() */ 1135 nladdr->nl_pid = READ_ONCE(nlk->portid); 1136 netlink_lock_table(); 1137 nladdr->nl_groups = nlk->groups ? nlk->groups[0] : 0; 1138 netlink_unlock_table(); 1139 } 1140 return sizeof(*nladdr); 1141} 1142 1143static int netlink_ioctl(struct socket *sock, unsigned int cmd, 1144 unsigned long arg) 1145{ 1146 /* try to hand this ioctl down to the NIC drivers. 1147 */ 1148 return -ENOIOCTLCMD; 1149} 1150 1151static struct sock *netlink_getsockbyportid(struct sock *ssk, u32 portid) 1152{ 1153 struct sock *sock; 1154 struct netlink_sock *nlk; 1155 1156 sock = netlink_lookup(sock_net(ssk), ssk->sk_protocol, portid); 1157 if (!sock) 1158 return ERR_PTR(-ECONNREFUSED); 1159 1160 /* Don't bother queuing skb if kernel socket has no input function */ 1161 nlk = nlk_sk(sock); 1162 /* dst_portid and sk_state can be changed in netlink_connect() */ 1163 if (READ_ONCE(sock->sk_state) == NETLINK_CONNECTED && 1164 READ_ONCE(nlk->dst_portid) != nlk_sk(ssk)->portid) { 1165 sock_put(sock); 1166 return ERR_PTR(-ECONNREFUSED); 1167 } 1168 return sock; 1169} 1170 1171struct sock *netlink_getsockbyfilp(struct file *filp) 1172{ 1173 struct inode *inode = file_inode(filp); 1174 struct sock *sock; 1175 1176 if (!S_ISSOCK(inode->i_mode)) 1177 return ERR_PTR(-ENOTSOCK); 1178 1179 sock = SOCKET_I(inode)->sk; 1180 if (sock->sk_family != AF_NETLINK) 1181 return ERR_PTR(-EINVAL); 1182 1183 sock_hold(sock); 1184 return sock; 1185} 1186 1187static struct sk_buff *netlink_alloc_large_skb(unsigned int size, 1188 int broadcast) 1189{ 1190 struct sk_buff *skb; 1191 void *data; 1192 1193 if (size <= NLMSG_GOODSIZE || broadcast) 1194 return alloc_skb(size, GFP_KERNEL); 1195 1196 size = SKB_DATA_ALIGN(size) + 1197 SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); 1198 1199 data = vmalloc(size); 1200 if (data == NULL) 1201 return NULL; 1202 1203 skb = __build_skb(data, size); 1204 if (skb == NULL) 1205 vfree(data); 1206 else 1207 skb->destructor = netlink_skb_destructor; 1208 1209 return skb; 1210} 1211 1212/* 1213 * Attach a skb to a netlink socket. 1214 * The caller must hold a reference to the destination socket. On error, the 1215 * reference is dropped. The skb is not send to the destination, just all 1216 * all error checks are performed and memory in the queue is reserved. 1217 * Return values: 1218 * < 0: error. skb freed, reference to sock dropped. 1219 * 0: continue 1220 * 1: repeat lookup - reference dropped while waiting for socket memory. 1221 */ 1222int netlink_attachskb(struct sock *sk, struct sk_buff *skb, 1223 long *timeo, struct sock *ssk) 1224{ 1225 struct netlink_sock *nlk; 1226 1227 nlk = nlk_sk(sk); 1228 1229 if ((atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf || 1230 test_bit(NETLINK_S_CONGESTED, &nlk->state))) { 1231 DECLARE_WAITQUEUE(wait, current); 1232 if (!*timeo) { 1233 if (!ssk || netlink_is_kernel(ssk)) 1234 netlink_overrun(sk); 1235 sock_put(sk); 1236 kfree_skb(skb); 1237 return -EAGAIN; 1238 } 1239 1240 __set_current_state(TASK_INTERRUPTIBLE); 1241 add_wait_queue(&nlk->wait, &wait); 1242 1243 if ((atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf || 1244 test_bit(NETLINK_S_CONGESTED, &nlk->state)) && 1245 !sock_flag(sk, SOCK_DEAD)) 1246 *timeo = schedule_timeout(*timeo); 1247 1248 __set_current_state(TASK_RUNNING); 1249 remove_wait_queue(&nlk->wait, &wait); 1250 sock_put(sk); 1251 1252 if (signal_pending(current)) { 1253 kfree_skb(skb); 1254 return sock_intr_errno(*timeo); 1255 } 1256 return 1; 1257 } 1258 netlink_skb_set_owner_r(skb, sk); 1259 return 0; 1260} 1261 1262static int __netlink_sendskb(struct sock *sk, struct sk_buff *skb) 1263{ 1264 int len = skb->len; 1265 1266 netlink_deliver_tap(sock_net(sk), skb); 1267 1268 skb_queue_tail(&sk->sk_receive_queue, skb); 1269 sk->sk_data_ready(sk); 1270 return len; 1271} 1272 1273int netlink_sendskb(struct sock *sk, struct sk_buff *skb) 1274{ 1275 int len = __netlink_sendskb(sk, skb); 1276 1277 sock_put(sk); 1278 return len; 1279} 1280 1281void netlink_detachskb(struct sock *sk, struct sk_buff *skb) 1282{ 1283 kfree_skb(skb); 1284 sock_put(sk); 1285} 1286 1287static struct sk_buff *netlink_trim(struct sk_buff *skb, gfp_t allocation) 1288{ 1289 int delta; 1290 1291 WARN_ON(skb->sk != NULL); 1292 delta = skb->end - skb->tail; 1293 if (is_vmalloc_addr(skb->head) || delta * 2 < skb->truesize) 1294 return skb; 1295 1296 if (skb_shared(skb)) { 1297 struct sk_buff *nskb = skb_clone(skb, allocation); 1298 if (!nskb) 1299 return skb; 1300 consume_skb(skb); 1301 skb = nskb; 1302 } 1303 1304 pskb_expand_head(skb, 0, -delta, 1305 (allocation & ~__GFP_DIRECT_RECLAIM) | 1306 __GFP_NOWARN | __GFP_NORETRY); 1307 return skb; 1308} 1309 1310static int netlink_unicast_kernel(struct sock *sk, struct sk_buff *skb, 1311 struct sock *ssk) 1312{ 1313 int ret; 1314 struct netlink_sock *nlk = nlk_sk(sk); 1315 1316 ret = -ECONNREFUSED; 1317 if (nlk->netlink_rcv != NULL) { 1318 ret = skb->len; 1319 netlink_skb_set_owner_r(skb, sk); 1320 NETLINK_CB(skb).sk = ssk; 1321 netlink_deliver_tap_kernel(sk, ssk, skb); 1322 nlk->netlink_rcv(skb); 1323 consume_skb(skb); 1324 } else { 1325 kfree_skb(skb); 1326 } 1327 sock_put(sk); 1328 return ret; 1329} 1330 1331int netlink_unicast(struct sock *ssk, struct sk_buff *skb, 1332 u32 portid, int nonblock) 1333{ 1334 struct sock *sk; 1335 int err; 1336 long timeo; 1337 1338 skb = netlink_trim(skb, gfp_any()); 1339 1340 timeo = sock_sndtimeo(ssk, nonblock); 1341retry: 1342 sk = netlink_getsockbyportid(ssk, portid); 1343 if (IS_ERR(sk)) { 1344 kfree_skb(skb); 1345 return PTR_ERR(sk); 1346 } 1347 if (netlink_is_kernel(sk)) 1348 return netlink_unicast_kernel(sk, skb, ssk); 1349 1350 if (sk_filter(sk, skb)) { 1351 err = skb->len; 1352 kfree_skb(skb); 1353 sock_put(sk); 1354 return err; 1355 } 1356 1357 err = netlink_attachskb(sk, skb, &timeo, ssk); 1358 if (err == 1) 1359 goto retry; 1360 if (err) 1361 return err; 1362 1363 return netlink_sendskb(sk, skb); 1364} 1365EXPORT_SYMBOL(netlink_unicast); 1366 1367int netlink_has_listeners(struct sock *sk, unsigned int group) 1368{ 1369 int res = 0; 1370 struct listeners *listeners; 1371 1372 BUG_ON(!netlink_is_kernel(sk)); 1373 1374 rcu_read_lock(); 1375 listeners = rcu_dereference(nl_table[sk->sk_protocol].listeners); 1376 1377 if (listeners && group - 1 < nl_table[sk->sk_protocol].groups) 1378 res = test_bit(group - 1, listeners->masks); 1379 1380 rcu_read_unlock(); 1381 1382 return res; 1383} 1384EXPORT_SYMBOL_GPL(netlink_has_listeners); 1385 1386bool netlink_strict_get_check(struct sk_buff *skb) 1387{ 1388 const struct netlink_sock *nlk = nlk_sk(NETLINK_CB(skb).sk); 1389 1390 return nlk->flags & NETLINK_F_STRICT_CHK; 1391} 1392EXPORT_SYMBOL_GPL(netlink_strict_get_check); 1393 1394static int netlink_broadcast_deliver(struct sock *sk, struct sk_buff *skb) 1395{ 1396 struct netlink_sock *nlk = nlk_sk(sk); 1397 1398 if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf && 1399 !test_bit(NETLINK_S_CONGESTED, &nlk->state)) { 1400 netlink_skb_set_owner_r(skb, sk); 1401 __netlink_sendskb(sk, skb); 1402 return atomic_read(&sk->sk_rmem_alloc) > (sk->sk_rcvbuf >> 1); 1403 } 1404 return -1; 1405} 1406 1407struct netlink_broadcast_data { 1408 struct sock *exclude_sk; 1409 struct net *net; 1410 u32 portid; 1411 u32 group; 1412 int failure; 1413 int delivery_failure; 1414 int congested; 1415 int delivered; 1416 gfp_t allocation; 1417 struct sk_buff *skb, *skb2; 1418 int (*tx_filter)(struct sock *dsk, struct sk_buff *skb, void *data); 1419 void *tx_data; 1420}; 1421 1422static void do_one_broadcast(struct sock *sk, 1423 struct netlink_broadcast_data *p) 1424{ 1425 struct netlink_sock *nlk = nlk_sk(sk); 1426 int val; 1427 1428 if (p->exclude_sk == sk) 1429 return; 1430 1431 if (nlk->portid == p->portid || p->group - 1 >= nlk->ngroups || 1432 !test_bit(p->group - 1, nlk->groups)) 1433 return; 1434 1435 if (!net_eq(sock_net(sk), p->net)) { 1436 if (!(nlk->flags & NETLINK_F_LISTEN_ALL_NSID)) 1437 return; 1438 1439 if (!peernet_has_id(sock_net(sk), p->net)) 1440 return; 1441 1442 if (!file_ns_capable(sk->sk_socket->file, p->net->user_ns, 1443 CAP_NET_BROADCAST)) 1444 return; 1445 } 1446 1447 if (p->failure) { 1448 netlink_overrun(sk); 1449 return; 1450 } 1451 1452 sock_hold(sk); 1453 if (p->skb2 == NULL) { 1454 if (skb_shared(p->skb)) { 1455 p->skb2 = skb_clone(p->skb, p->allocation); 1456 } else { 1457 p->skb2 = skb_get(p->skb); 1458 /* 1459 * skb ownership may have been set when 1460 * delivered to a previous socket. 1461 */ 1462 skb_orphan(p->skb2); 1463 } 1464 } 1465 if (p->skb2 == NULL) { 1466 netlink_overrun(sk); 1467 /* Clone failed. Notify ALL listeners. */ 1468 p->failure = 1; 1469 if (nlk->flags & NETLINK_F_BROADCAST_SEND_ERROR) 1470 p->delivery_failure = 1; 1471 goto out; 1472 } 1473 if (p->tx_filter && p->tx_filter(sk, p->skb2, p->tx_data)) { 1474 kfree_skb(p->skb2); 1475 p->skb2 = NULL; 1476 goto out; 1477 } 1478 if (sk_filter(sk, p->skb2)) { 1479 kfree_skb(p->skb2); 1480 p->skb2 = NULL; 1481 goto out; 1482 } 1483 NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); 1484 if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) 1485 NETLINK_CB(p->skb2).nsid_is_set = true; 1486 val = netlink_broadcast_deliver(sk, p->skb2); 1487 if (val < 0) { 1488 netlink_overrun(sk); 1489 if (nlk->flags & NETLINK_F_BROADCAST_SEND_ERROR) 1490 p->delivery_failure = 1; 1491 } else { 1492 p->congested |= val; 1493 p->delivered = 1; 1494 p->skb2 = NULL; 1495 } 1496out: 1497 sock_put(sk); 1498} 1499 1500int netlink_broadcast_filtered(struct sock *ssk, struct sk_buff *skb, u32 portid, 1501 u32 group, gfp_t allocation, 1502 int (*filter)(struct sock *dsk, struct sk_buff *skb, void *data), 1503 void *filter_data) 1504{ 1505 struct net *net = sock_net(ssk); 1506 struct netlink_broadcast_data info; 1507 struct sock *sk; 1508 1509 skb = netlink_trim(skb, allocation); 1510 1511 info.exclude_sk = ssk; 1512 info.net = net; 1513 info.portid = portid; 1514 info.group = group; 1515 info.failure = 0; 1516 info.delivery_failure = 0; 1517 info.congested = 0; 1518 info.delivered = 0; 1519 info.allocation = allocation; 1520 info.skb = skb; 1521 info.skb2 = NULL; 1522 info.tx_filter = filter; 1523 info.tx_data = filter_data; 1524 1525 /* While we sleep in clone, do not allow to change socket list */ 1526 1527 netlink_lock_table(); 1528 1529 sk_for_each_bound(sk, &nl_table[ssk->sk_protocol].mc_list) 1530 do_one_broadcast(sk, &info); 1531 1532 consume_skb(skb); 1533 1534 netlink_unlock_table(); 1535 1536 if (info.delivery_failure) { 1537 kfree_skb(info.skb2); 1538 return -ENOBUFS; 1539 } 1540 consume_skb(info.skb2); 1541 1542 if (info.delivered) { 1543 if (info.congested && gfpflags_allow_blocking(allocation)) 1544 yield(); 1545 return 0; 1546 } 1547 return -ESRCH; 1548} 1549EXPORT_SYMBOL(netlink_broadcast_filtered); 1550 1551int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 portid, 1552 u32 group, gfp_t allocation) 1553{ 1554 return netlink_broadcast_filtered(ssk, skb, portid, group, allocation, 1555 NULL, NULL); 1556} 1557EXPORT_SYMBOL(netlink_broadcast); 1558 1559struct netlink_set_err_data { 1560 struct sock *exclude_sk; 1561 u32 portid; 1562 u32 group; 1563 int code; 1564}; 1565 1566static int do_one_set_err(struct sock *sk, struct netlink_set_err_data *p) 1567{ 1568 struct netlink_sock *nlk = nlk_sk(sk); 1569 int ret = 0; 1570 1571 if (sk == p->exclude_sk) 1572 goto out; 1573 1574 if (!net_eq(sock_net(sk), sock_net(p->exclude_sk))) 1575 goto out; 1576 1577 if (nlk->portid == p->portid || p->group - 1 >= nlk->ngroups || 1578 !test_bit(p->group - 1, nlk->groups)) 1579 goto out; 1580 1581 if (p->code == ENOBUFS && nlk->flags & NETLINK_F_RECV_NO_ENOBUFS) { 1582 ret = 1; 1583 goto out; 1584 } 1585 1586 sk->sk_err = p->code; 1587 sk->sk_error_report(sk); 1588out: 1589 return ret; 1590} 1591 1592/** 1593 * netlink_set_err - report error to broadcast listeners 1594 * @ssk: the kernel netlink socket, as returned by netlink_kernel_create() 1595 * @portid: the PORTID of a process that we want to skip (if any) 1596 * @group: the broadcast group that will notice the error 1597 * @code: error code, must be negative (as usual in kernelspace) 1598 * 1599 * This function returns the number of broadcast listeners that have set the 1600 * NETLINK_NO_ENOBUFS socket option. 1601 */ 1602int netlink_set_err(struct sock *ssk, u32 portid, u32 group, int code) 1603{ 1604 struct netlink_set_err_data info; 1605 unsigned long flags; 1606 struct sock *sk; 1607 int ret = 0; 1608 1609 info.exclude_sk = ssk; 1610 info.portid = portid; 1611 info.group = group; 1612 /* sk->sk_err wants a positive error value */ 1613 info.code = -code; 1614 1615 read_lock_irqsave(&nl_table_lock, flags); 1616 1617 sk_for_each_bound(sk, &nl_table[ssk->sk_protocol].mc_list) 1618 ret += do_one_set_err(sk, &info); 1619 1620 read_unlock_irqrestore(&nl_table_lock, flags); 1621 return ret; 1622} 1623EXPORT_SYMBOL(netlink_set_err); 1624 1625/* must be called with netlink table grabbed */ 1626static void netlink_update_socket_mc(struct netlink_sock *nlk, 1627 unsigned int group, 1628 int is_new) 1629{ 1630 int old, new = !!is_new, subscriptions; 1631 1632 old = test_bit(group - 1, nlk->groups); 1633 subscriptions = nlk->subscriptions - old + new; 1634 if (new) 1635 __set_bit(group - 1, nlk->groups); 1636 else 1637 __clear_bit(group - 1, nlk->groups); 1638 netlink_update_subscriptions(&nlk->sk, subscriptions); 1639 netlink_update_listeners(&nlk->sk); 1640} 1641 1642static int netlink_setsockopt(struct socket *sock, int level, int optname, 1643 sockptr_t optval, unsigned int optlen) 1644{ 1645 struct sock *sk = sock->sk; 1646 struct netlink_sock *nlk = nlk_sk(sk); 1647 unsigned int val = 0; 1648 int err; 1649 1650 if (level != SOL_NETLINK) 1651 return -ENOPROTOOPT; 1652 1653 if (optlen >= sizeof(int) && 1654 copy_from_sockptr(&val, optval, sizeof(val))) 1655 return -EFAULT; 1656 1657 switch (optname) { 1658 case NETLINK_PKTINFO: 1659 if (val) 1660 nlk->flags |= NETLINK_F_RECV_PKTINFO; 1661 else 1662 nlk->flags &= ~NETLINK_F_RECV_PKTINFO; 1663 err = 0; 1664 break; 1665 case NETLINK_ADD_MEMBERSHIP: 1666 case NETLINK_DROP_MEMBERSHIP: { 1667 if (!netlink_allowed(sock, NL_CFG_F_NONROOT_RECV)) 1668 return -EPERM; 1669 err = netlink_realloc_groups(sk); 1670 if (err) 1671 return err; 1672 if (!val || val - 1 >= nlk->ngroups) 1673 return -EINVAL; 1674 if (optname == NETLINK_ADD_MEMBERSHIP && nlk->netlink_bind) { 1675 err = nlk->netlink_bind(sock_net(sk), val); 1676 if (err) 1677 return err; 1678 } 1679 netlink_table_grab(); 1680 netlink_update_socket_mc(nlk, val, 1681 optname == NETLINK_ADD_MEMBERSHIP); 1682 netlink_table_ungrab(); 1683 if (optname == NETLINK_DROP_MEMBERSHIP && nlk->netlink_unbind) 1684 nlk->netlink_unbind(sock_net(sk), val); 1685 1686 err = 0; 1687 break; 1688 } 1689 case NETLINK_BROADCAST_ERROR: 1690 if (val) 1691 nlk->flags |= NETLINK_F_BROADCAST_SEND_ERROR; 1692 else 1693 nlk->flags &= ~NETLINK_F_BROADCAST_SEND_ERROR; 1694 err = 0; 1695 break; 1696 case NETLINK_NO_ENOBUFS: 1697 if (val) { 1698 nlk->flags |= NETLINK_F_RECV_NO_ENOBUFS; 1699 clear_bit(NETLINK_S_CONGESTED, &nlk->state); 1700 wake_up_interruptible(&nlk->wait); 1701 } else { 1702 nlk->flags &= ~NETLINK_F_RECV_NO_ENOBUFS; 1703 } 1704 err = 0; 1705 break; 1706 case NETLINK_LISTEN_ALL_NSID: 1707 if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_BROADCAST)) 1708 return -EPERM; 1709 1710 if (val) 1711 nlk->flags |= NETLINK_F_LISTEN_ALL_NSID; 1712 else 1713 nlk->flags &= ~NETLINK_F_LISTEN_ALL_NSID; 1714 err = 0; 1715 break; 1716 case NETLINK_CAP_ACK: 1717 if (val) 1718 nlk->flags |= NETLINK_F_CAP_ACK; 1719 else 1720 nlk->flags &= ~NETLINK_F_CAP_ACK; 1721 err = 0; 1722 break; 1723 case NETLINK_EXT_ACK: 1724 if (val) 1725 nlk->flags |= NETLINK_F_EXT_ACK; 1726 else 1727 nlk->flags &= ~NETLINK_F_EXT_ACK; 1728 err = 0; 1729 break; 1730 case NETLINK_GET_STRICT_CHK: 1731 if (val) 1732 nlk->flags |= NETLINK_F_STRICT_CHK; 1733 else 1734 nlk->flags &= ~NETLINK_F_STRICT_CHK; 1735 err = 0; 1736 break; 1737 default: 1738 err = -ENOPROTOOPT; 1739 } 1740 return err; 1741} 1742 1743static int netlink_getsockopt(struct socket *sock, int level, int optname, 1744 char __user *optval, int __user *optlen) 1745{ 1746 struct sock *sk = sock->sk; 1747 struct netlink_sock *nlk = nlk_sk(sk); 1748 unsigned int flag; 1749 int len, val; 1750 1751 if (level != SOL_NETLINK) 1752 return -ENOPROTOOPT; 1753 1754 if (get_user(len, optlen)) 1755 return -EFAULT; 1756 if (len < 0) 1757 return -EINVAL; 1758 1759 switch (optname) { 1760 case NETLINK_PKTINFO: 1761 flag = NETLINK_F_RECV_PKTINFO; 1762 break; 1763 case NETLINK_BROADCAST_ERROR: 1764 flag = NETLINK_F_BROADCAST_SEND_ERROR; 1765 break; 1766 case NETLINK_NO_ENOBUFS: 1767 flag = NETLINK_F_RECV_NO_ENOBUFS; 1768 break; 1769 case NETLINK_LIST_MEMBERSHIPS: { 1770 int pos, idx, shift, err = 0; 1771 1772 netlink_lock_table(); 1773 for (pos = 0; pos * 8 < nlk->ngroups; pos += sizeof(u32)) { 1774 if (len - pos < sizeof(u32)) 1775 break; 1776 1777 idx = pos / sizeof(unsigned long); 1778 shift = (pos % sizeof(unsigned long)) * 8; 1779 if (put_user((u32)(nlk->groups[idx] >> shift), 1780 (u32 __user *)(optval + pos))) { 1781 err = -EFAULT; 1782 break; 1783 } 1784 } 1785 if (put_user(ALIGN(BITS_TO_BYTES(nlk->ngroups), sizeof(u32)), optlen)) 1786 err = -EFAULT; 1787 netlink_unlock_table(); 1788 return err; 1789 } 1790 case NETLINK_CAP_ACK: 1791 flag = NETLINK_F_CAP_ACK; 1792 break; 1793 case NETLINK_EXT_ACK: 1794 flag = NETLINK_F_EXT_ACK; 1795 break; 1796 case NETLINK_GET_STRICT_CHK: 1797 flag = NETLINK_F_STRICT_CHK; 1798 break; 1799 default: 1800 return -ENOPROTOOPT; 1801 } 1802 1803 if (len < sizeof(int)) 1804 return -EINVAL; 1805 1806 len = sizeof(int); 1807 val = nlk->flags & flag ? 1 : 0; 1808 1809 if (put_user(len, optlen) || 1810 copy_to_user(optval, &val, len)) 1811 return -EFAULT; 1812 1813 return 0; 1814} 1815 1816static void netlink_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb) 1817{ 1818 struct nl_pktinfo info; 1819 1820 info.group = NETLINK_CB(skb).dst_group; 1821 put_cmsg(msg, SOL_NETLINK, NETLINK_PKTINFO, sizeof(info), &info); 1822} 1823 1824static void netlink_cmsg_listen_all_nsid(struct sock *sk, struct msghdr *msg, 1825 struct sk_buff *skb) 1826{ 1827 if (!NETLINK_CB(skb).nsid_is_set) 1828 return; 1829 1830 put_cmsg(msg, SOL_NETLINK, NETLINK_LISTEN_ALL_NSID, sizeof(int), 1831 &NETLINK_CB(skb).nsid); 1832} 1833 1834static int netlink_sendmsg(struct socket *sock, struct msghdr *msg, size_t len) 1835{ 1836 struct sock *sk = sock->sk; 1837 struct netlink_sock *nlk = nlk_sk(sk); 1838 DECLARE_SOCKADDR(struct sockaddr_nl *, addr, msg->msg_name); 1839 u32 dst_portid; 1840 u32 dst_group; 1841 struct sk_buff *skb; 1842 int err; 1843 struct scm_cookie scm; 1844 u32 netlink_skb_flags = 0; 1845 1846 if (msg->msg_flags & MSG_OOB) 1847 return -EOPNOTSUPP; 1848 1849 if (len == 0) { 1850 pr_warn_once("Zero length message leads to an empty skb\n"); 1851 return -ENODATA; 1852 } 1853 1854 err = scm_send(sock, msg, &scm, true); 1855 if (err < 0) 1856 return err; 1857 1858 if (msg->msg_namelen) { 1859 err = -EINVAL; 1860 if (msg->msg_namelen < sizeof(struct sockaddr_nl)) 1861 goto out; 1862 if (addr->nl_family != AF_NETLINK) 1863 goto out; 1864 dst_portid = addr->nl_pid; 1865 dst_group = ffs(addr->nl_groups); 1866 err = -EPERM; 1867 if ((dst_group || dst_portid) && 1868 !netlink_allowed(sock, NL_CFG_F_NONROOT_SEND)) 1869 goto out; 1870 netlink_skb_flags |= NETLINK_SKB_DST; 1871 } else { 1872 /* Paired with WRITE_ONCE() in netlink_connect() */ 1873 dst_portid = READ_ONCE(nlk->dst_portid); 1874 dst_group = READ_ONCE(nlk->dst_group); 1875 } 1876 1877 /* Paired with WRITE_ONCE() in netlink_insert() */ 1878 if (!READ_ONCE(nlk->bound)) { 1879 err = netlink_autobind(sock); 1880 if (err) 1881 goto out; 1882 } else { 1883 /* Ensure nlk is hashed and visible. */ 1884 smp_rmb(); 1885 } 1886 1887 err = -EMSGSIZE; 1888 if (len > sk->sk_sndbuf - 32) 1889 goto out; 1890 err = -ENOBUFS; 1891 skb = netlink_alloc_large_skb(len, dst_group); 1892 if (skb == NULL) 1893 goto out; 1894 1895 NETLINK_CB(skb).portid = nlk->portid; 1896 NETLINK_CB(skb).dst_group = dst_group; 1897 NETLINK_CB(skb).creds = scm.creds; 1898 NETLINK_CB(skb).flags = netlink_skb_flags; 1899 1900 err = -EFAULT; 1901 if (memcpy_from_msg(skb_put(skb, len), msg, len)) { 1902 kfree_skb(skb); 1903 goto out; 1904 } 1905 1906 err = security_netlink_send(sk, skb); 1907 if (err) { 1908 kfree_skb(skb); 1909 goto out; 1910 } 1911 1912 if (dst_group) { 1913 refcount_inc(&skb->users); 1914 netlink_broadcast(sk, skb, dst_portid, dst_group, GFP_KERNEL); 1915 } 1916 err = netlink_unicast(sk, skb, dst_portid, msg->msg_flags & MSG_DONTWAIT); 1917 1918out: 1919 scm_destroy(&scm); 1920 return err; 1921} 1922 1923static int netlink_recvmsg(struct socket *sock, struct msghdr *msg, size_t len, 1924 int flags) 1925{ 1926 struct scm_cookie scm; 1927 struct sock *sk = sock->sk; 1928 struct netlink_sock *nlk = nlk_sk(sk); 1929 int noblock = flags & MSG_DONTWAIT; 1930 size_t copied; 1931 struct sk_buff *skb, *data_skb; 1932 int err, ret; 1933 1934 if (flags & MSG_OOB) 1935 return -EOPNOTSUPP; 1936 1937 copied = 0; 1938 1939 skb = skb_recv_datagram(sk, flags, noblock, &err); 1940 if (skb == NULL) 1941 goto out; 1942 1943 data_skb = skb; 1944 1945#ifdef CONFIG_COMPAT_NETLINK_MESSAGES 1946 if (unlikely(skb_shinfo(skb)->frag_list)) { 1947 /* 1948 * If this skb has a frag_list, then here that means that we 1949 * will have to use the frag_list skb's data for compat tasks 1950 * and the regular skb's data for normal (non-compat) tasks. 1951 * 1952 * If we need to send the compat skb, assign it to the 1953 * 'data_skb' variable so that it will be used below for data 1954 * copying. We keep 'skb' for everything else, including 1955 * freeing both later. 1956 */ 1957 if (flags & MSG_CMSG_COMPAT) 1958 data_skb = skb_shinfo(skb)->frag_list; 1959 } 1960#endif 1961 1962 /* Record the max length of recvmsg() calls for future allocations */ 1963 nlk->max_recvmsg_len = max(nlk->max_recvmsg_len, len); 1964 nlk->max_recvmsg_len = min_t(size_t, nlk->max_recvmsg_len, 1965 SKB_WITH_OVERHEAD(32768)); 1966 1967 copied = data_skb->len; 1968 if (len < copied) { 1969 msg->msg_flags |= MSG_TRUNC; 1970 copied = len; 1971 } 1972 1973 err = skb_copy_datagram_msg(data_skb, 0, msg, copied); 1974 1975 if (msg->msg_name) { 1976 DECLARE_SOCKADDR(struct sockaddr_nl *, addr, msg->msg_name); 1977 addr->nl_family = AF_NETLINK; 1978 addr->nl_pad = 0; 1979 addr->nl_pid = NETLINK_CB(skb).portid; 1980 addr->nl_groups = netlink_group_mask(NETLINK_CB(skb).dst_group); 1981 msg->msg_namelen = sizeof(*addr); 1982 } 1983 1984 if (nlk->flags & NETLINK_F_RECV_PKTINFO) 1985 netlink_cmsg_recv_pktinfo(msg, skb); 1986 if (nlk->flags & NETLINK_F_LISTEN_ALL_NSID) 1987 netlink_cmsg_listen_all_nsid(sk, msg, skb); 1988 1989 memset(&scm, 0, sizeof(scm)); 1990 scm.creds = *NETLINK_CREDS(skb); 1991 if (flags & MSG_TRUNC) 1992 copied = data_skb->len; 1993 1994 skb_free_datagram(sk, skb); 1995 1996 if (READ_ONCE(nlk->cb_running) && 1997 atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf / 2) { 1998 ret = netlink_dump(sk); 1999 if (ret) { 2000 sk->sk_err = -ret; 2001 sk->sk_error_report(sk); 2002 } 2003 } 2004 2005 scm_recv(sock, msg, &scm, flags); 2006out: 2007 netlink_rcv_wake(sk); 2008 return err ? : copied; 2009} 2010 2011static void netlink_data_ready(struct sock *sk) 2012{ 2013 BUG(); 2014} 2015 2016/* 2017 * We export these functions to other modules. They provide a 2018 * complete set of kernel non-blocking support for message 2019 * queueing. 2020 */ 2021 2022struct sock * 2023__netlink_kernel_create(struct net *net, int unit, struct module *module, 2024 struct netlink_kernel_cfg *cfg) 2025{ 2026 struct socket *sock; 2027 struct sock *sk; 2028 struct netlink_sock *nlk; 2029 struct listeners *listeners = NULL; 2030 struct mutex *cb_mutex = cfg ? cfg->cb_mutex : NULL; 2031 unsigned int groups; 2032 2033 BUG_ON(!nl_table); 2034 2035 if (unit < 0 || unit >= MAX_LINKS) 2036 return NULL; 2037 2038 if (sock_create_lite(PF_NETLINK, SOCK_DGRAM, unit, &sock)) 2039 return NULL; 2040 2041 if (__netlink_create(net, sock, cb_mutex, unit, 1) < 0) 2042 goto out_sock_release_nosk; 2043 2044 sk = sock->sk; 2045 2046 if (!cfg || cfg->groups < 32) 2047 groups = 32; 2048 else 2049 groups = cfg->groups; 2050 2051 listeners = kzalloc(sizeof(*listeners) + NLGRPSZ(groups), GFP_KERNEL); 2052 if (!listeners) 2053 goto out_sock_release; 2054 2055 sk->sk_data_ready = netlink_data_ready; 2056 if (cfg && cfg->input) 2057 nlk_sk(sk)->netlink_rcv = cfg->input; 2058 2059 if (netlink_insert(sk, 0)) 2060 goto out_sock_release; 2061 2062 nlk = nlk_sk(sk); 2063 nlk->flags |= NETLINK_F_KERNEL_SOCKET; 2064 2065 netlink_table_grab(); 2066 if (!nl_table[unit].registered) { 2067 nl_table[unit].groups = groups; 2068 rcu_assign_pointer(nl_table[unit].listeners, listeners); 2069 nl_table[unit].cb_mutex = cb_mutex; 2070 nl_table[unit].module = module; 2071 if (cfg) { 2072 nl_table[unit].bind = cfg->bind; 2073 nl_table[unit].unbind = cfg->unbind; 2074 nl_table[unit].flags = cfg->flags; 2075 if (cfg->compare) 2076 nl_table[unit].compare = cfg->compare; 2077 } 2078 nl_table[unit].registered = 1; 2079 } else { 2080 kfree(listeners); 2081 nl_table[unit].registered++; 2082 } 2083 netlink_table_ungrab(); 2084 return sk; 2085 2086out_sock_release: 2087 kfree(listeners); 2088 netlink_kernel_release(sk); 2089 return NULL; 2090 2091out_sock_release_nosk: 2092 sock_release(sock); 2093 return NULL; 2094} 2095EXPORT_SYMBOL(__netlink_kernel_create); 2096 2097void 2098netlink_kernel_release(struct sock *sk) 2099{ 2100 if (sk == NULL || sk->sk_socket == NULL) 2101 return; 2102 2103 sock_release(sk->sk_socket); 2104} 2105EXPORT_SYMBOL(netlink_kernel_release); 2106 2107int __netlink_change_ngroups(struct sock *sk, unsigned int groups) 2108{ 2109 struct listeners *new, *old; 2110 struct netlink_table *tbl = &nl_table[sk->sk_protocol]; 2111 2112 if (groups < 32) 2113 groups = 32; 2114 2115 if (NLGRPSZ(tbl->groups) < NLGRPSZ(groups)) { 2116 new = kzalloc(sizeof(*new) + NLGRPSZ(groups), GFP_ATOMIC); 2117 if (!new) 2118 return -ENOMEM; 2119 old = nl_deref_protected(tbl->listeners); 2120 memcpy(new->masks, old->masks, NLGRPSZ(tbl->groups)); 2121 rcu_assign_pointer(tbl->listeners, new); 2122 2123 kfree_rcu(old, rcu); 2124 } 2125 tbl->groups = groups; 2126 2127 return 0; 2128} 2129 2130/** 2131 * netlink_change_ngroups - change number of multicast groups 2132 * 2133 * This changes the number of multicast groups that are available 2134 * on a certain netlink family. Note that it is not possible to 2135 * change the number of groups to below 32. Also note that it does 2136 * not implicitly call netlink_clear_multicast_users() when the 2137 * number of groups is reduced. 2138 * 2139 * @sk: The kernel netlink socket, as returned by netlink_kernel_create(). 2140 * @groups: The new number of groups. 2141 */ 2142int netlink_change_ngroups(struct sock *sk, unsigned int groups) 2143{ 2144 int err; 2145 2146 netlink_table_grab(); 2147 err = __netlink_change_ngroups(sk, groups); 2148 netlink_table_ungrab(); 2149 2150 return err; 2151} 2152 2153void __netlink_clear_multicast_users(struct sock *ksk, unsigned int group) 2154{ 2155 struct sock *sk; 2156 struct netlink_table *tbl = &nl_table[ksk->sk_protocol]; 2157 2158 sk_for_each_bound(sk, &tbl->mc_list) 2159 netlink_update_socket_mc(nlk_sk(sk), group, 0); 2160} 2161 2162struct nlmsghdr * 2163__nlmsg_put(struct sk_buff *skb, u32 portid, u32 seq, int type, int len, int flags) 2164{ 2165 struct nlmsghdr *nlh; 2166 int size = nlmsg_msg_size(len); 2167 2168 nlh = skb_put(skb, NLMSG_ALIGN(size)); 2169 nlh->nlmsg_type = type; 2170 nlh->nlmsg_len = size; 2171 nlh->nlmsg_flags = flags; 2172 nlh->nlmsg_pid = portid; 2173 nlh->nlmsg_seq = seq; 2174 if (!__builtin_constant_p(size) || NLMSG_ALIGN(size) - size != 0) 2175 memset(nlmsg_data(nlh) + len, 0, NLMSG_ALIGN(size) - size); 2176 return nlh; 2177} 2178EXPORT_SYMBOL(__nlmsg_put); 2179 2180/* 2181 * It looks a bit ugly. 2182 * It would be better to create kernel thread. 2183 */ 2184 2185static int netlink_dump_done(struct netlink_sock *nlk, struct sk_buff *skb, 2186 struct netlink_callback *cb, 2187 struct netlink_ext_ack *extack) 2188{ 2189 struct nlmsghdr *nlh; 2190 2191 nlh = nlmsg_put_answer(skb, cb, NLMSG_DONE, sizeof(nlk->dump_done_errno), 2192 NLM_F_MULTI | cb->answer_flags); 2193 if (WARN_ON(!nlh)) 2194 return -ENOBUFS; 2195 2196 nl_dump_check_consistent(cb, nlh); 2197 memcpy(nlmsg_data(nlh), &nlk->dump_done_errno, sizeof(nlk->dump_done_errno)); 2198 2199 if (extack->_msg && nlk->flags & NETLINK_F_EXT_ACK) { 2200 nlh->nlmsg_flags |= NLM_F_ACK_TLVS; 2201 if (!nla_put_string(skb, NLMSGERR_ATTR_MSG, extack->_msg)) 2202 nlmsg_end(skb, nlh); 2203 } 2204 2205 return 0; 2206} 2207 2208static int netlink_dump(struct sock *sk) 2209{ 2210 struct netlink_sock *nlk = nlk_sk(sk); 2211 struct netlink_ext_ack extack = {}; 2212 struct netlink_callback *cb; 2213 struct sk_buff *skb = NULL; 2214 struct module *module; 2215 int err = -ENOBUFS; 2216 int alloc_min_size; 2217 int alloc_size; 2218 2219 mutex_lock(nlk->cb_mutex); 2220 if (!nlk->cb_running) { 2221 err = -EINVAL; 2222 goto errout_skb; 2223 } 2224 2225 if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf) 2226 goto errout_skb; 2227 2228 /* NLMSG_GOODSIZE is small to avoid high order allocations being 2229 * required, but it makes sense to _attempt_ a 16K bytes allocation 2230 * to reduce number of system calls on dump operations, if user 2231 * ever provided a big enough buffer. 2232 */ 2233 cb = &nlk->cb; 2234 alloc_min_size = max_t(int, cb->min_dump_alloc, NLMSG_GOODSIZE); 2235 2236 if (alloc_min_size < nlk->max_recvmsg_len) { 2237 alloc_size = nlk->max_recvmsg_len; 2238 skb = alloc_skb(alloc_size, 2239 (GFP_KERNEL & ~__GFP_DIRECT_RECLAIM) | 2240 __GFP_NOWARN | __GFP_NORETRY); 2241 } 2242 if (!skb) { 2243 alloc_size = alloc_min_size; 2244 skb = alloc_skb(alloc_size, GFP_KERNEL); 2245 } 2246 if (!skb) 2247 goto errout_skb; 2248 2249 /* Trim skb to allocated size. User is expected to provide buffer as 2250 * large as max(min_dump_alloc, 16KiB (mac_recvmsg_len capped at 2251 * netlink_recvmsg())). dump will pack as many smaller messages as 2252 * could fit within the allocated skb. skb is typically allocated 2253 * with larger space than required (could be as much as near 2x the 2254 * requested size with align to next power of 2 approach). Allowing 2255 * dump to use the excess space makes it difficult for a user to have a 2256 * reasonable static buffer based on the expected largest dump of a 2257 * single netdev. The outcome is MSG_TRUNC error. 2258 */ 2259 skb_reserve(skb, skb_tailroom(skb) - alloc_size); 2260 2261 /* Make sure malicious BPF programs can not read unitialized memory 2262 * from skb->head -> skb->data 2263 */ 2264 skb_reset_network_header(skb); 2265 skb_reset_mac_header(skb); 2266 2267 netlink_skb_set_owner_r(skb, sk); 2268 2269 if (nlk->dump_done_errno > 0) { 2270 cb->extack = &extack; 2271 nlk->dump_done_errno = cb->dump(skb, cb); 2272 cb->extack = NULL; 2273 } 2274 2275 if (nlk->dump_done_errno > 0 || 2276 skb_tailroom(skb) < nlmsg_total_size(sizeof(nlk->dump_done_errno))) { 2277 mutex_unlock(nlk->cb_mutex); 2278 2279 if (sk_filter(sk, skb)) 2280 kfree_skb(skb); 2281 else 2282 __netlink_sendskb(sk, skb); 2283 return 0; 2284 } 2285 2286 if (netlink_dump_done(nlk, skb, cb, &extack)) 2287 goto errout_skb; 2288 2289#ifdef CONFIG_COMPAT_NETLINK_MESSAGES 2290 /* frag_list skb's data is used for compat tasks 2291 * and the regular skb's data for normal (non-compat) tasks. 2292 * See netlink_recvmsg(). 2293 */ 2294 if (unlikely(skb_shinfo(skb)->frag_list)) { 2295 if (netlink_dump_done(nlk, skb_shinfo(skb)->frag_list, cb, &extack)) 2296 goto errout_skb; 2297 } 2298#endif 2299 2300 if (sk_filter(sk, skb)) 2301 kfree_skb(skb); 2302 else 2303 __netlink_sendskb(sk, skb); 2304 2305 if (cb->done) 2306 cb->done(cb); 2307 2308 WRITE_ONCE(nlk->cb_running, false); 2309 module = cb->module; 2310 skb = cb->skb; 2311 mutex_unlock(nlk->cb_mutex); 2312 module_put(module); 2313 consume_skb(skb); 2314 return 0; 2315 2316errout_skb: 2317 mutex_unlock(nlk->cb_mutex); 2318 kfree_skb(skb); 2319 return err; 2320} 2321 2322int __netlink_dump_start(struct sock *ssk, struct sk_buff *skb, 2323 const struct nlmsghdr *nlh, 2324 struct netlink_dump_control *control) 2325{ 2326 struct netlink_sock *nlk, *nlk2; 2327 struct netlink_callback *cb; 2328 struct sock *sk; 2329 int ret; 2330 2331 refcount_inc(&skb->users); 2332 2333 sk = netlink_lookup(sock_net(ssk), ssk->sk_protocol, NETLINK_CB(skb).portid); 2334 if (sk == NULL) { 2335 ret = -ECONNREFUSED; 2336 goto error_free; 2337 } 2338 2339 nlk = nlk_sk(sk); 2340 mutex_lock(nlk->cb_mutex); 2341 /* A dump is in progress... */ 2342 if (nlk->cb_running) { 2343 ret = -EBUSY; 2344 goto error_unlock; 2345 } 2346 /* add reference of module which cb->dump belongs to */ 2347 if (!try_module_get(control->module)) { 2348 ret = -EPROTONOSUPPORT; 2349 goto error_unlock; 2350 } 2351 2352 cb = &nlk->cb; 2353 memset(cb, 0, sizeof(*cb)); 2354 cb->dump = control->dump; 2355 cb->done = control->done; 2356 cb->nlh = nlh; 2357 cb->data = control->data; 2358 cb->module = control->module; 2359 cb->min_dump_alloc = control->min_dump_alloc; 2360 cb->skb = skb; 2361 2362 nlk2 = nlk_sk(NETLINK_CB(skb).sk); 2363 cb->strict_check = !!(nlk2->flags & NETLINK_F_STRICT_CHK); 2364 2365 if (control->start) { 2366 ret = control->start(cb); 2367 if (ret) 2368 goto error_put; 2369 } 2370 2371 WRITE_ONCE(nlk->cb_running, true); 2372 nlk->dump_done_errno = INT_MAX; 2373 2374 mutex_unlock(nlk->cb_mutex); 2375 2376 ret = netlink_dump(sk); 2377 2378 sock_put(sk); 2379 2380 if (ret) 2381 return ret; 2382 2383 /* We successfully started a dump, by returning -EINTR we 2384 * signal not to send ACK even if it was requested. 2385 */ 2386 return -EINTR; 2387 2388error_put: 2389 module_put(control->module); 2390error_unlock: 2391 sock_put(sk); 2392 mutex_unlock(nlk->cb_mutex); 2393error_free: 2394 kfree_skb(skb); 2395 return ret; 2396} 2397EXPORT_SYMBOL(__netlink_dump_start); 2398 2399void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err, 2400 const struct netlink_ext_ack *extack) 2401{ 2402 struct sk_buff *skb; 2403 struct nlmsghdr *rep; 2404 struct nlmsgerr *errmsg; 2405 size_t payload = sizeof(*errmsg); 2406 size_t tlvlen = 0; 2407 struct netlink_sock *nlk = nlk_sk(NETLINK_CB(in_skb).sk); 2408 unsigned int flags = 0; 2409 bool nlk_has_extack = nlk->flags & NETLINK_F_EXT_ACK; 2410 2411 /* Error messages get the original request appened, unless the user 2412 * requests to cap the error message, and get extra error data if 2413 * requested. 2414 */ 2415 if (nlk_has_extack && extack && extack->_msg) 2416 tlvlen += nla_total_size(strlen(extack->_msg) + 1); 2417 2418 if (err && !(nlk->flags & NETLINK_F_CAP_ACK)) 2419 payload += nlmsg_len(nlh); 2420 else 2421 flags |= NLM_F_CAPPED; 2422 if (err && nlk_has_extack && extack && extack->bad_attr) 2423 tlvlen += nla_total_size(sizeof(u32)); 2424 if (nlk_has_extack && extack && extack->cookie_len) 2425 tlvlen += nla_total_size(extack->cookie_len); 2426 if (err && nlk_has_extack && extack && extack->policy) 2427 tlvlen += netlink_policy_dump_attr_size_estimate(extack->policy); 2428 2429 if (tlvlen) 2430 flags |= NLM_F_ACK_TLVS; 2431 2432 skb = nlmsg_new(payload + tlvlen, GFP_KERNEL); 2433 if (!skb) { 2434 NETLINK_CB(in_skb).sk->sk_err = ENOBUFS; 2435 NETLINK_CB(in_skb).sk->sk_error_report(NETLINK_CB(in_skb).sk); 2436 return; 2437 } 2438 2439 rep = __nlmsg_put(skb, NETLINK_CB(in_skb).portid, nlh->nlmsg_seq, 2440 NLMSG_ERROR, payload, flags); 2441 errmsg = nlmsg_data(rep); 2442 errmsg->error = err; 2443 memcpy(&errmsg->msg, nlh, payload > sizeof(*errmsg) ? nlh->nlmsg_len : sizeof(*nlh)); 2444 2445 if (nlk_has_extack && extack) { 2446 if (extack->_msg) { 2447 WARN_ON(nla_put_string(skb, NLMSGERR_ATTR_MSG, 2448 extack->_msg)); 2449 } 2450 if (err && extack->bad_attr && 2451 !WARN_ON((u8 *)extack->bad_attr < in_skb->data || 2452 (u8 *)extack->bad_attr >= in_skb->data + 2453 in_skb->len)) 2454 WARN_ON(nla_put_u32(skb, NLMSGERR_ATTR_OFFS, 2455 (u8 *)extack->bad_attr - 2456 (u8 *)nlh)); 2457 if (extack->cookie_len) 2458 WARN_ON(nla_put(skb, NLMSGERR_ATTR_COOKIE, 2459 extack->cookie_len, extack->cookie)); 2460 if (extack->policy) 2461 netlink_policy_dump_write_attr(skb, extack->policy, 2462 NLMSGERR_ATTR_POLICY); 2463 } 2464 2465 nlmsg_end(skb, rep); 2466 2467 netlink_unicast(in_skb->sk, skb, NETLINK_CB(in_skb).portid, MSG_DONTWAIT); 2468} 2469EXPORT_SYMBOL(netlink_ack); 2470 2471int netlink_rcv_skb(struct sk_buff *skb, int (*cb)(struct sk_buff *, 2472 struct nlmsghdr *, 2473 struct netlink_ext_ack *)) 2474{ 2475 struct netlink_ext_ack extack; 2476 struct nlmsghdr *nlh; 2477 int err; 2478 2479 while (skb->len >= nlmsg_total_size(0)) { 2480 int msglen; 2481 2482 memset(&extack, 0, sizeof(extack)); 2483 nlh = nlmsg_hdr(skb); 2484 err = 0; 2485 2486 if (nlh->nlmsg_len < NLMSG_HDRLEN || skb->len < nlh->nlmsg_len) 2487 return 0; 2488 2489 /* Only requests are handled by the kernel */ 2490 if (!(nlh->nlmsg_flags & NLM_F_REQUEST)) 2491 goto ack; 2492 2493 /* Skip control messages */ 2494 if (nlh->nlmsg_type < NLMSG_MIN_TYPE) 2495 goto ack; 2496 2497 err = cb(skb, nlh, &extack); 2498 if (err == -EINTR) 2499 goto skip; 2500 2501ack: 2502 if (nlh->nlmsg_flags & NLM_F_ACK || err) 2503 netlink_ack(skb, nlh, err, &extack); 2504 2505skip: 2506 msglen = NLMSG_ALIGN(nlh->nlmsg_len); 2507 if (msglen > skb->len) 2508 msglen = skb->len; 2509 skb_pull(skb, msglen); 2510 } 2511 2512 return 0; 2513} 2514EXPORT_SYMBOL(netlink_rcv_skb); 2515 2516/** 2517 * nlmsg_notify - send a notification netlink message 2518 * @sk: netlink socket to use 2519 * @skb: notification message 2520 * @portid: destination netlink portid for reports or 0 2521 * @group: destination multicast group or 0 2522 * @report: 1 to report back, 0 to disable 2523 * @flags: allocation flags 2524 */ 2525int nlmsg_notify(struct sock *sk, struct sk_buff *skb, u32 portid, 2526 unsigned int group, int report, gfp_t flags) 2527{ 2528 int err = 0; 2529 2530 if (group) { 2531 int exclude_portid = 0; 2532 2533 if (report) { 2534 refcount_inc(&skb->users); 2535 exclude_portid = portid; 2536 } 2537 2538 /* errors reported via destination sk->sk_err, but propagate 2539 * delivery errors if NETLINK_BROADCAST_ERROR flag is set */ 2540 err = nlmsg_multicast(sk, skb, exclude_portid, group, flags); 2541 if (err == -ESRCH) 2542 err = 0; 2543 } 2544 2545 if (report) { 2546 int err2; 2547 2548 err2 = nlmsg_unicast(sk, skb, portid); 2549 if (!err) 2550 err = err2; 2551 } 2552 2553 return err; 2554} 2555EXPORT_SYMBOL(nlmsg_notify); 2556 2557#ifdef CONFIG_PROC_FS 2558struct nl_seq_iter { 2559 struct seq_net_private p; 2560 struct rhashtable_iter hti; 2561 int link; 2562}; 2563 2564static void netlink_walk_start(struct nl_seq_iter *iter) 2565{ 2566 rhashtable_walk_enter(&nl_table[iter->link].hash, &iter->hti); 2567 rhashtable_walk_start(&iter->hti); 2568} 2569 2570static void netlink_walk_stop(struct nl_seq_iter *iter) 2571{ 2572 rhashtable_walk_stop(&iter->hti); 2573 rhashtable_walk_exit(&iter->hti); 2574} 2575 2576static void *__netlink_seq_next(struct seq_file *seq) 2577{ 2578 struct nl_seq_iter *iter = seq->private; 2579 struct netlink_sock *nlk; 2580 2581 do { 2582 for (;;) { 2583 nlk = rhashtable_walk_next(&iter->hti); 2584 2585 if (IS_ERR(nlk)) { 2586 if (PTR_ERR(nlk) == -EAGAIN) 2587 continue; 2588 2589 return nlk; 2590 } 2591 2592 if (nlk) 2593 break; 2594 2595 netlink_walk_stop(iter); 2596 if (++iter->link >= MAX_LINKS) 2597 return NULL; 2598 2599 netlink_walk_start(iter); 2600 } 2601 } while (sock_net(&nlk->sk) != seq_file_net(seq)); 2602 2603 return nlk; 2604} 2605 2606static void *netlink_seq_start(struct seq_file *seq, loff_t *posp) 2607 __acquires(RCU) 2608{ 2609 struct nl_seq_iter *iter = seq->private; 2610 void *obj = SEQ_START_TOKEN; 2611 loff_t pos; 2612 2613 iter->link = 0; 2614 2615 netlink_walk_start(iter); 2616 2617 for (pos = *posp; pos && obj && !IS_ERR(obj); pos--) 2618 obj = __netlink_seq_next(seq); 2619 2620 return obj; 2621} 2622 2623static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos) 2624{ 2625 ++*pos; 2626 return __netlink_seq_next(seq); 2627} 2628 2629static void netlink_native_seq_stop(struct seq_file *seq, void *v) 2630{ 2631 struct nl_seq_iter *iter = seq->private; 2632 2633 if (iter->link >= MAX_LINKS) 2634 return; 2635 2636 netlink_walk_stop(iter); 2637} 2638 2639 2640static int netlink_native_seq_show(struct seq_file *seq, void *v) 2641{ 2642 if (v == SEQ_START_TOKEN) { 2643 seq_puts(seq, 2644 "sk Eth Pid Groups " 2645 "Rmem Wmem Dump Locks Drops Inode\n"); 2646 } else { 2647 struct sock *s = v; 2648 struct netlink_sock *nlk = nlk_sk(s); 2649 2650 seq_printf(seq, "%pK %-3d %-10u %08x %-8d %-8d %-5d %-8d %-8u %-8lu\n", 2651 s, 2652 s->sk_protocol, 2653 nlk->portid, 2654 nlk->groups ? (u32)nlk->groups[0] : 0, 2655 sk_rmem_alloc_get(s), 2656 sk_wmem_alloc_get(s), 2657 READ_ONCE(nlk->cb_running), 2658 refcount_read(&s->sk_refcnt), 2659 atomic_read(&s->sk_drops), 2660 sock_i_ino(s) 2661 ); 2662 2663 } 2664 return 0; 2665} 2666 2667#ifdef CONFIG_BPF_SYSCALL 2668struct bpf_iter__netlink { 2669 __bpf_md_ptr(struct bpf_iter_meta *, meta); 2670 __bpf_md_ptr(struct netlink_sock *, sk); 2671}; 2672 2673DEFINE_BPF_ITER_FUNC(netlink, struct bpf_iter_meta *meta, struct netlink_sock *sk) 2674 2675static int netlink_prog_seq_show(struct bpf_prog *prog, 2676 struct bpf_iter_meta *meta, 2677 void *v) 2678{ 2679 struct bpf_iter__netlink ctx; 2680 2681 meta->seq_num--; /* skip SEQ_START_TOKEN */ 2682 ctx.meta = meta; 2683 ctx.sk = nlk_sk((struct sock *)v); 2684 return bpf_iter_run_prog(prog, &ctx); 2685} 2686 2687static int netlink_seq_show(struct seq_file *seq, void *v) 2688{ 2689 struct bpf_iter_meta meta; 2690 struct bpf_prog *prog; 2691 2692 meta.seq = seq; 2693 prog = bpf_iter_get_info(&meta, false); 2694 if (!prog) 2695 return netlink_native_seq_show(seq, v); 2696 2697 if (v != SEQ_START_TOKEN) 2698 return netlink_prog_seq_show(prog, &meta, v); 2699 2700 return 0; 2701} 2702 2703static void netlink_seq_stop(struct seq_file *seq, void *v) 2704{ 2705 struct bpf_iter_meta meta; 2706 struct bpf_prog *prog; 2707 2708 if (!v) { 2709 meta.seq = seq; 2710 prog = bpf_iter_get_info(&meta, true); 2711 if (prog) 2712 (void)netlink_prog_seq_show(prog, &meta, v); 2713 } 2714 2715 netlink_native_seq_stop(seq, v); 2716} 2717#else 2718static int netlink_seq_show(struct seq_file *seq, void *v) 2719{ 2720 return netlink_native_seq_show(seq, v); 2721} 2722 2723static void netlink_seq_stop(struct seq_file *seq, void *v) 2724{ 2725 netlink_native_seq_stop(seq, v); 2726} 2727#endif 2728 2729static const struct seq_operations netlink_seq_ops = { 2730 .start = netlink_seq_start, 2731 .next = netlink_seq_next, 2732 .stop = netlink_seq_stop, 2733 .show = netlink_seq_show, 2734}; 2735#endif 2736 2737int netlink_register_notifier(struct notifier_block *nb) 2738{ 2739 return blocking_notifier_chain_register(&netlink_chain, nb); 2740} 2741EXPORT_SYMBOL(netlink_register_notifier); 2742 2743int netlink_unregister_notifier(struct notifier_block *nb) 2744{ 2745 return blocking_notifier_chain_unregister(&netlink_chain, nb); 2746} 2747EXPORT_SYMBOL(netlink_unregister_notifier); 2748 2749static const struct proto_ops netlink_ops = { 2750 .family = PF_NETLINK, 2751 .owner = THIS_MODULE, 2752 .release = netlink_release, 2753 .bind = netlink_bind, 2754 .connect = netlink_connect, 2755 .socketpair = sock_no_socketpair, 2756 .accept = sock_no_accept, 2757 .getname = netlink_getname, 2758 .poll = datagram_poll, 2759 .ioctl = netlink_ioctl, 2760 .listen = sock_no_listen, 2761 .shutdown = sock_no_shutdown, 2762 .setsockopt = netlink_setsockopt, 2763 .getsockopt = netlink_getsockopt, 2764 .sendmsg = netlink_sendmsg, 2765 .recvmsg = netlink_recvmsg, 2766 .mmap = sock_no_mmap, 2767 .sendpage = sock_no_sendpage, 2768}; 2769 2770static const struct net_proto_family netlink_family_ops = { 2771 .family = PF_NETLINK, 2772 .create = netlink_create, 2773 .owner = THIS_MODULE, /* for consistency 8) */ 2774}; 2775 2776static int __net_init netlink_net_init(struct net *net) 2777{ 2778#ifdef CONFIG_PROC_FS 2779 if (!proc_create_net("netlink", 0, net->proc_net, &netlink_seq_ops, 2780 sizeof(struct nl_seq_iter))) 2781 return -ENOMEM; 2782#endif 2783 return 0; 2784} 2785 2786static void __net_exit netlink_net_exit(struct net *net) 2787{ 2788#ifdef CONFIG_PROC_FS 2789 remove_proc_entry("netlink", net->proc_net); 2790#endif 2791} 2792 2793static void __init netlink_add_usersock_entry(void) 2794{ 2795 struct listeners *listeners; 2796 int groups = 32; 2797 2798 listeners = kzalloc(sizeof(*listeners) + NLGRPSZ(groups), GFP_KERNEL); 2799 if (!listeners) 2800 panic("netlink_add_usersock_entry: Cannot allocate listeners\n"); 2801 2802 netlink_table_grab(); 2803 2804 nl_table[NETLINK_USERSOCK].groups = groups; 2805 rcu_assign_pointer(nl_table[NETLINK_USERSOCK].listeners, listeners); 2806 nl_table[NETLINK_USERSOCK].module = THIS_MODULE; 2807 nl_table[NETLINK_USERSOCK].registered = 1; 2808 nl_table[NETLINK_USERSOCK].flags = NL_CFG_F_NONROOT_SEND; 2809 2810 netlink_table_ungrab(); 2811} 2812 2813static struct pernet_operations __net_initdata netlink_net_ops = { 2814 .init = netlink_net_init, 2815 .exit = netlink_net_exit, 2816}; 2817 2818static inline u32 netlink_hash(const void *data, u32 len, u32 seed) 2819{ 2820 const struct netlink_sock *nlk = data; 2821 struct netlink_compare_arg arg; 2822 2823 netlink_compare_arg_init(&arg, sock_net(&nlk->sk), nlk->portid); 2824 return jhash2((u32 *)&arg, netlink_compare_arg_len / sizeof(u32), seed); 2825} 2826 2827static const struct rhashtable_params netlink_rhashtable_params = { 2828 .head_offset = offsetof(struct netlink_sock, node), 2829 .key_len = netlink_compare_arg_len, 2830 .obj_hashfn = netlink_hash, 2831 .obj_cmpfn = netlink_compare, 2832 .automatic_shrinking = true, 2833}; 2834 2835#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) 2836BTF_ID_LIST(btf_netlink_sock_id) 2837BTF_ID(struct, netlink_sock) 2838 2839static const struct bpf_iter_seq_info netlink_seq_info = { 2840 .seq_ops = &netlink_seq_ops, 2841 .init_seq_private = bpf_iter_init_seq_net, 2842 .fini_seq_private = bpf_iter_fini_seq_net, 2843 .seq_priv_size = sizeof(struct nl_seq_iter), 2844}; 2845 2846static struct bpf_iter_reg netlink_reg_info = { 2847 .target = "netlink", 2848 .ctx_arg_info_size = 1, 2849 .ctx_arg_info = { 2850 { offsetof(struct bpf_iter__netlink, sk), 2851 PTR_TO_BTF_ID_OR_NULL }, 2852 }, 2853 .seq_info = &netlink_seq_info, 2854}; 2855 2856static int __init bpf_iter_register(void) 2857{ 2858 netlink_reg_info.ctx_arg_info[0].btf_id = *btf_netlink_sock_id; 2859 return bpf_iter_reg_target(&netlink_reg_info); 2860} 2861#endif 2862 2863static int __init netlink_proto_init(void) 2864{ 2865 int i; 2866 int err = proto_register(&netlink_proto, 0); 2867 2868 if (err != 0) 2869 goto out; 2870 2871#if defined(CONFIG_BPF_SYSCALL) && defined(CONFIG_PROC_FS) 2872 err = bpf_iter_register(); 2873 if (err) 2874 goto out; 2875#endif 2876 2877 BUILD_BUG_ON(sizeof(struct netlink_skb_parms) > sizeof_field(struct sk_buff, cb)); 2878 2879 nl_table = kcalloc(MAX_LINKS, sizeof(*nl_table), GFP_KERNEL); 2880 if (!nl_table) 2881 goto panic; 2882 2883 for (i = 0; i < MAX_LINKS; i++) { 2884 if (rhashtable_init(&nl_table[i].hash, 2885 &netlink_rhashtable_params) < 0) { 2886 while (--i > 0) 2887 rhashtable_destroy(&nl_table[i].hash); 2888 kfree(nl_table); 2889 goto panic; 2890 } 2891 } 2892 2893 netlink_add_usersock_entry(); 2894 2895 sock_register(&netlink_family_ops); 2896 register_pernet_subsys(&netlink_net_ops); 2897 register_pernet_subsys(&netlink_tap_net_ops); 2898 /* The netlink device handler may be needed early. */ 2899 rtnetlink_init(); 2900out: 2901 return err; 2902panic: 2903 panic("netlink_init: Cannot allocate nl_table\n"); 2904} 2905 2906core_initcall(netlink_proto_init); 2907