1// SPDX-License-Identifier: GPL-2.0-only 2/* 3 * INET An implementation of the TCP/IP protocol suite for the LINUX 4 * operating system. INET is implemented using the BSD Socket 5 * interface as the means of communication with the user level. 6 * 7 * The Internet Protocol (IP) output module. 8 * 9 * Authors: Ross Biro 10 * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> 11 * Donald Becker, <becker@super.org> 12 * Alan Cox, <Alan.Cox@linux.org> 13 * Richard Underwood 14 * Stefan Becker, <stefanb@yello.ping.de> 15 * Jorge Cwik, <jorge@laser.satlink.net> 16 * Arnt Gulbrandsen, <agulbra@nvg.unit.no> 17 * Hirokazu Takahashi, <taka@valinux.co.jp> 18 * 19 * See ip_input.c for original log 20 * 21 * Fixes: 22 * Alan Cox : Missing nonblock feature in ip_build_xmit. 23 * Mike Kilburn : htons() missing in ip_build_xmit. 24 * Bradford Johnson: Fix faulty handling of some frames when 25 * no route is found. 26 * Alexander Demenshin: Missing sk/skb free in ip_queue_xmit 27 * (in case if packet not accepted by 28 * output firewall rules) 29 * Mike McLagan : Routing by source 30 * Alexey Kuznetsov: use new route cache 31 * Andi Kleen: Fix broken PMTU recovery and remove 32 * some redundant tests. 33 * Vitaly E. Lavrov : Transparent proxy revived after year coma. 34 * Andi Kleen : Replace ip_reply with ip_send_reply. 35 * Andi Kleen : Split fast and slow ip_build_xmit path 36 * for decreased register pressure on x86 37 * and more readability. 38 * Marc Boucher : When call_out_firewall returns FW_QUEUE, 39 * silently drop skb instead of failing with -EPERM. 40 * Detlev Wengorz : Copy protocol for fragments. 41 * Hirokazu Takahashi: HW checksumming for outgoing UDP 42 * datagrams. 43 * Hirokazu Takahashi: sendfile() on UDP works now. 44 */ 45 46#include <linux/uaccess.h> 47#include <linux/module.h> 48#include <linux/types.h> 49#include <linux/kernel.h> 50#include <linux/mm.h> 51#include <linux/string.h> 52#include <linux/errno.h> 53#include <linux/highmem.h> 54#include <linux/slab.h> 55 56#include <linux/socket.h> 57#include <linux/sockios.h> 58#include <linux/in.h> 59#include <linux/inet.h> 60#include <linux/netdevice.h> 61#include <linux/etherdevice.h> 62#include <linux/proc_fs.h> 63#include <linux/stat.h> 64#include <linux/init.h> 65 66#include <net/snmp.h> 67#include <net/ip.h> 68#include <net/protocol.h> 69#include <net/route.h> 70#include <net/xfrm.h> 71#include <linux/skbuff.h> 72#include <net/sock.h> 73#include <net/arp.h> 74#include <net/icmp.h> 75#include <net/checksum.h> 76#include <net/gso.h> 77#include <net/inetpeer.h> 78#include <net/inet_ecn.h> 79#include <net/lwtunnel.h> 80#include <linux/bpf-cgroup.h> 81#include <linux/igmp.h> 82#include <linux/netfilter_ipv4.h> 83#include <linux/netfilter_bridge.h> 84#include <linux/netlink.h> 85#include <linux/tcp.h> 86 87static int 88ip_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, 89 unsigned int mtu, 90 int (*output)(struct net *, struct sock *, struct sk_buff *)); 91 92/* Generate a checksum for an outgoing IP datagram. */ 93void ip_send_check(struct iphdr *iph) 94{ 95 iph->check = 0; 96 iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl); 97} 98EXPORT_SYMBOL(ip_send_check); 99 100int __ip_local_out(struct net *net, struct sock *sk, struct sk_buff *skb) 101{ 102 struct iphdr *iph = ip_hdr(skb); 103 104 iph_set_totlen(iph, skb->len); 105 ip_send_check(iph); 106 107 /* if egress device is enslaved to an L3 master device pass the 108 * skb to its handler for processing 109 */ 110 skb = l3mdev_ip_out(sk, skb); 111 if (unlikely(!skb)) 112 return 0; 113 114 skb->protocol = htons(ETH_P_IP); 115 116 return nf_hook(NFPROTO_IPV4, NF_INET_LOCAL_OUT, 117 net, sk, skb, NULL, skb_dst(skb)->dev, 118 dst_output); 119} 120 121int ip_local_out(struct net *net, struct sock *sk, struct sk_buff *skb) 122{ 123 int err; 124 125 err = __ip_local_out(net, sk, skb); 126 if (likely(err == 1)) 127 err = dst_output(net, sk, skb); 128 129 return err; 130} 131EXPORT_SYMBOL_GPL(ip_local_out); 132 133static inline int ip_select_ttl(const struct inet_sock *inet, 134 const struct dst_entry *dst) 135{ 136 int ttl = READ_ONCE(inet->uc_ttl); 137 138 if (ttl < 0) 139 ttl = ip4_dst_hoplimit(dst); 140 return ttl; 141} 142 143/* 144 * Add an ip header to a skbuff and send it out. 145 * 146 */ 147int ip_build_and_send_pkt(struct sk_buff *skb, const struct sock *sk, 148 __be32 saddr, __be32 daddr, struct ip_options_rcu *opt, 149 u8 tos) 150{ 151 const struct inet_sock *inet = inet_sk(sk); 152 struct rtable *rt = skb_rtable(skb); 153 struct net *net = sock_net(sk); 154 struct iphdr *iph; 155 156 /* Build the IP header. */ 157 skb_push(skb, sizeof(struct iphdr) + (opt ? opt->opt.optlen : 0)); 158 skb_reset_network_header(skb); 159 iph = ip_hdr(skb); 160 iph->version = 4; 161 iph->ihl = 5; 162 iph->tos = tos; 163 iph->ttl = ip_select_ttl(inet, &rt->dst); 164 iph->daddr = (opt && opt->opt.srr ? opt->opt.faddr : daddr); 165 iph->saddr = saddr; 166 iph->protocol = sk->sk_protocol; 167 /* Do not bother generating IPID for small packets (eg SYNACK) */ 168 if (skb->len <= IPV4_MIN_MTU || ip_dont_fragment(sk, &rt->dst)) { 169 iph->frag_off = htons(IP_DF); 170 iph->id = 0; 171 } else { 172 iph->frag_off = 0; 173 /* TCP packets here are SYNACK with fat IPv4/TCP options. 174 * Avoid using the hashed IP ident generator. 175 */ 176 if (sk->sk_protocol == IPPROTO_TCP) 177 iph->id = (__force __be16)get_random_u16(); 178 else 179 __ip_select_ident(net, iph, 1); 180 } 181 182 if (opt && opt->opt.optlen) { 183 iph->ihl += opt->opt.optlen>>2; 184 ip_options_build(skb, &opt->opt, daddr, rt); 185 } 186 187 skb->priority = READ_ONCE(sk->sk_priority); 188 if (!skb->mark) 189 skb->mark = READ_ONCE(sk->sk_mark); 190 191 /* Send it out. */ 192 return ip_local_out(net, skb->sk, skb); 193} 194EXPORT_SYMBOL_GPL(ip_build_and_send_pkt); 195 196static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *skb) 197{ 198 struct dst_entry *dst = skb_dst(skb); 199 struct rtable *rt = (struct rtable *)dst; 200 struct net_device *dev = dst->dev; 201 unsigned int hh_len = LL_RESERVED_SPACE(dev); 202 struct neighbour *neigh; 203 bool is_v6gw = false; 204 205 if (rt->rt_type == RTN_MULTICAST) { 206 IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTMCAST, skb->len); 207 } else if (rt->rt_type == RTN_BROADCAST) 208 IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTBCAST, skb->len); 209 210 /* OUTOCTETS should be counted after fragment */ 211 IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len); 212 213 if (unlikely(skb_headroom(skb) < hh_len && dev->header_ops)) { 214 skb = skb_expand_head(skb, hh_len); 215 if (!skb) 216 return -ENOMEM; 217 } 218 219 if (lwtunnel_xmit_redirect(dst->lwtstate)) { 220 int res = lwtunnel_xmit(skb); 221 222 if (res != LWTUNNEL_XMIT_CONTINUE) 223 return res; 224 } 225 226 rcu_read_lock(); 227 neigh = ip_neigh_for_gw(rt, skb, &is_v6gw); 228 if (!IS_ERR(neigh)) { 229 int res; 230 231 sock_confirm_neigh(skb, neigh); 232 /* if crossing protocols, can not use the cached header */ 233 res = neigh_output(neigh, skb, is_v6gw); 234 rcu_read_unlock(); 235 return res; 236 } 237 rcu_read_unlock(); 238 239 net_dbg_ratelimited("%s: No header cache and no neighbour!\n", 240 __func__); 241 kfree_skb_reason(skb, SKB_DROP_REASON_NEIGH_CREATEFAIL); 242 return PTR_ERR(neigh); 243} 244 245static int ip_finish_output_gso(struct net *net, struct sock *sk, 246 struct sk_buff *skb, unsigned int mtu) 247{ 248 struct sk_buff *segs, *nskb; 249 netdev_features_t features; 250 int ret = 0; 251 252 /* common case: seglen is <= mtu 253 */ 254 if (skb_gso_validate_network_len(skb, mtu)) 255 return ip_finish_output2(net, sk, skb); 256 257 /* Slowpath - GSO segment length exceeds the egress MTU. 258 * 259 * This can happen in several cases: 260 * - Forwarding of a TCP GRO skb, when DF flag is not set. 261 * - Forwarding of an skb that arrived on a virtualization interface 262 * (virtio-net/vhost/tap) with TSO/GSO size set by other network 263 * stack. 264 * - Local GSO skb transmitted on an NETIF_F_TSO tunnel stacked over an 265 * interface with a smaller MTU. 266 * - Arriving GRO skb (or GSO skb in a virtualized environment) that is 267 * bridged to a NETIF_F_TSO tunnel stacked over an interface with an 268 * insufficient MTU. 269 */ 270 features = netif_skb_features(skb); 271 BUILD_BUG_ON(sizeof(*IPCB(skb)) > SKB_GSO_CB_OFFSET); 272 segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK); 273 if (IS_ERR_OR_NULL(segs)) { 274 kfree_skb(skb); 275 return -ENOMEM; 276 } 277 278 consume_skb(skb); 279 280 skb_list_walk_safe(segs, segs, nskb) { 281 int err; 282 283 skb_mark_not_on_list(segs); 284 err = ip_fragment(net, sk, segs, mtu, ip_finish_output2); 285 286 if (err && ret == 0) 287 ret = err; 288 } 289 290 return ret; 291} 292 293static int __ip_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb) 294{ 295 unsigned int mtu; 296 297#if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM) 298 /* Policy lookup after SNAT yielded a new policy */ 299 if (skb_dst(skb)->xfrm) { 300 IPCB(skb)->flags |= IPSKB_REROUTED; 301 return dst_output(net, sk, skb); 302 } 303#endif 304 mtu = ip_skb_dst_mtu(sk, skb); 305 if (skb_is_gso(skb)) 306 return ip_finish_output_gso(net, sk, skb, mtu); 307 308 if (skb->len > mtu || IPCB(skb)->frag_max_size) 309 return ip_fragment(net, sk, skb, mtu, ip_finish_output2); 310 311 return ip_finish_output2(net, sk, skb); 312} 313 314static int ip_finish_output(struct net *net, struct sock *sk, struct sk_buff *skb) 315{ 316 int ret; 317 318 ret = BPF_CGROUP_RUN_PROG_INET_EGRESS(sk, skb); 319 switch (ret) { 320 case NET_XMIT_SUCCESS: 321 return __ip_finish_output(net, sk, skb); 322 case NET_XMIT_CN: 323 return __ip_finish_output(net, sk, skb) ? : ret; 324 default: 325 kfree_skb_reason(skb, SKB_DROP_REASON_BPF_CGROUP_EGRESS); 326 return ret; 327 } 328} 329 330static int ip_mc_finish_output(struct net *net, struct sock *sk, 331 struct sk_buff *skb) 332{ 333 struct rtable *new_rt; 334 bool do_cn = false; 335 int ret, err; 336 337 ret = BPF_CGROUP_RUN_PROG_INET_EGRESS(sk, skb); 338 switch (ret) { 339 case NET_XMIT_CN: 340 do_cn = true; 341 fallthrough; 342 case NET_XMIT_SUCCESS: 343 break; 344 default: 345 kfree_skb_reason(skb, SKB_DROP_REASON_BPF_CGROUP_EGRESS); 346 return ret; 347 } 348 349 /* Reset rt_iif so that inet_iif() will return skb->skb_iif. Setting 350 * this to non-zero causes ipi_ifindex in in_pktinfo to be overwritten, 351 * see ipv4_pktinfo_prepare(). 352 */ 353 new_rt = rt_dst_clone(net->loopback_dev, skb_rtable(skb)); 354 if (new_rt) { 355 new_rt->rt_iif = 0; 356 skb_dst_drop(skb); 357 skb_dst_set(skb, &new_rt->dst); 358 } 359 360 err = dev_loopback_xmit(net, sk, skb); 361 return (do_cn && err) ? ret : err; 362} 363 364int ip_mc_output(struct net *net, struct sock *sk, struct sk_buff *skb) 365{ 366 struct rtable *rt = skb_rtable(skb); 367 struct net_device *dev = rt->dst.dev; 368 369 /* 370 * If the indicated interface is up and running, send the packet. 371 */ 372 skb->dev = dev; 373 skb->protocol = htons(ETH_P_IP); 374 375 /* 376 * Multicasts are looped back for other local users 377 */ 378 379 if (rt->rt_flags&RTCF_MULTICAST) { 380 if (sk_mc_loop(sk) 381#ifdef CONFIG_IP_MROUTE 382 /* Small optimization: do not loopback not local frames, 383 which returned after forwarding; they will be dropped 384 by ip_mr_input in any case. 385 Note, that local frames are looped back to be delivered 386 to local recipients. 387 388 This check is duplicated in ip_mr_input at the moment. 389 */ 390 && 391 ((rt->rt_flags & RTCF_LOCAL) || 392 !(IPCB(skb)->flags & IPSKB_FORWARDED)) 393#endif 394 ) { 395 struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); 396 if (newskb) 397 NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, 398 net, sk, newskb, NULL, newskb->dev, 399 ip_mc_finish_output); 400 } 401 402 /* Multicasts with ttl 0 must not go beyond the host */ 403 404 if (ip_hdr(skb)->ttl == 0) { 405 kfree_skb(skb); 406 return 0; 407 } 408 } 409 410 if (rt->rt_flags&RTCF_BROADCAST) { 411 struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); 412 if (newskb) 413 NF_HOOK(NFPROTO_IPV4, NF_INET_POST_ROUTING, 414 net, sk, newskb, NULL, newskb->dev, 415 ip_mc_finish_output); 416 } 417 418 return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, 419 net, sk, skb, NULL, skb->dev, 420 ip_finish_output, 421 !(IPCB(skb)->flags & IPSKB_REROUTED)); 422} 423 424int ip_output(struct net *net, struct sock *sk, struct sk_buff *skb) 425{ 426 struct net_device *dev = skb_dst(skb)->dev, *indev = skb->dev; 427 428 skb->dev = dev; 429 skb->protocol = htons(ETH_P_IP); 430 431 return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, 432 net, sk, skb, indev, dev, 433 ip_finish_output, 434 !(IPCB(skb)->flags & IPSKB_REROUTED)); 435} 436EXPORT_SYMBOL(ip_output); 437 438/* 439 * copy saddr and daddr, possibly using 64bit load/stores 440 * Equivalent to : 441 * iph->saddr = fl4->saddr; 442 * iph->daddr = fl4->daddr; 443 */ 444static void ip_copy_addrs(struct iphdr *iph, const struct flowi4 *fl4) 445{ 446 BUILD_BUG_ON(offsetof(typeof(*fl4), daddr) != 447 offsetof(typeof(*fl4), saddr) + sizeof(fl4->saddr)); 448 449 iph->saddr = fl4->saddr; 450 iph->daddr = fl4->daddr; 451} 452 453/* Note: skb->sk can be different from sk, in case of tunnels */ 454int __ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, 455 __u8 tos) 456{ 457 struct inet_sock *inet = inet_sk(sk); 458 struct net *net = sock_net(sk); 459 struct ip_options_rcu *inet_opt; 460 struct flowi4 *fl4; 461 struct rtable *rt; 462 struct iphdr *iph; 463 int res; 464 465 /* Skip all of this if the packet is already routed, 466 * f.e. by something like SCTP. 467 */ 468 rcu_read_lock(); 469 inet_opt = rcu_dereference(inet->inet_opt); 470 fl4 = &fl->u.ip4; 471 rt = skb_rtable(skb); 472 if (rt) 473 goto packet_routed; 474 475 /* Make sure we can route this packet. */ 476 rt = (struct rtable *)__sk_dst_check(sk, 0); 477 if (!rt) { 478 __be32 daddr; 479 480 /* Use correct destination address if we have options. */ 481 daddr = inet->inet_daddr; 482 if (inet_opt && inet_opt->opt.srr) 483 daddr = inet_opt->opt.faddr; 484 485 /* If this fails, retransmit mechanism of transport layer will 486 * keep trying until route appears or the connection times 487 * itself out. 488 */ 489 rt = ip_route_output_ports(net, fl4, sk, 490 daddr, inet->inet_saddr, 491 inet->inet_dport, 492 inet->inet_sport, 493 sk->sk_protocol, 494 RT_CONN_FLAGS_TOS(sk, tos), 495 sk->sk_bound_dev_if); 496 if (IS_ERR(rt)) 497 goto no_route; 498 sk_setup_caps(sk, &rt->dst); 499 } 500 skb_dst_set_noref(skb, &rt->dst); 501 502packet_routed: 503 if (inet_opt && inet_opt->opt.is_strictroute && rt->rt_uses_gateway) 504 goto no_route; 505 506 /* OK, we know where to send it, allocate and build IP header. */ 507 skb_push(skb, sizeof(struct iphdr) + (inet_opt ? inet_opt->opt.optlen : 0)); 508 skb_reset_network_header(skb); 509 iph = ip_hdr(skb); 510 *((__be16 *)iph) = htons((4 << 12) | (5 << 8) | (tos & 0xff)); 511 if (ip_dont_fragment(sk, &rt->dst) && !skb->ignore_df) 512 iph->frag_off = htons(IP_DF); 513 else 514 iph->frag_off = 0; 515 iph->ttl = ip_select_ttl(inet, &rt->dst); 516 iph->protocol = sk->sk_protocol; 517 ip_copy_addrs(iph, fl4); 518 519 /* Transport layer set skb->h.foo itself. */ 520 521 if (inet_opt && inet_opt->opt.optlen) { 522 iph->ihl += inet_opt->opt.optlen >> 2; 523 ip_options_build(skb, &inet_opt->opt, inet->inet_daddr, rt); 524 } 525 526 ip_select_ident_segs(net, skb, sk, 527 skb_shinfo(skb)->gso_segs ?: 1); 528 529 /* TODO : should we use skb->sk here instead of sk ? */ 530 skb->priority = READ_ONCE(sk->sk_priority); 531 skb->mark = READ_ONCE(sk->sk_mark); 532 533 res = ip_local_out(net, sk, skb); 534 rcu_read_unlock(); 535 return res; 536 537no_route: 538 rcu_read_unlock(); 539 IP_INC_STATS(net, IPSTATS_MIB_OUTNOROUTES); 540 kfree_skb_reason(skb, SKB_DROP_REASON_IP_OUTNOROUTES); 541 return -EHOSTUNREACH; 542} 543EXPORT_SYMBOL(__ip_queue_xmit); 544 545int ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl) 546{ 547 return __ip_queue_xmit(sk, skb, fl, inet_sk(sk)->tos); 548} 549EXPORT_SYMBOL(ip_queue_xmit); 550 551static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) 552{ 553 to->pkt_type = from->pkt_type; 554 to->priority = from->priority; 555 to->protocol = from->protocol; 556 to->skb_iif = from->skb_iif; 557 skb_dst_drop(to); 558 skb_dst_copy(to, from); 559 to->dev = from->dev; 560 to->mark = from->mark; 561 562 skb_copy_hash(to, from); 563 564#ifdef CONFIG_NET_SCHED 565 to->tc_index = from->tc_index; 566#endif 567 nf_copy(to, from); 568 skb_ext_copy(to, from); 569#if IS_ENABLED(CONFIG_IP_VS) 570 to->ipvs_property = from->ipvs_property; 571#endif 572 skb_copy_secmark(to, from); 573} 574 575static int ip_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, 576 unsigned int mtu, 577 int (*output)(struct net *, struct sock *, struct sk_buff *)) 578{ 579 struct iphdr *iph = ip_hdr(skb); 580 581 if ((iph->frag_off & htons(IP_DF)) == 0) 582 return ip_do_fragment(net, sk, skb, output); 583 584 if (unlikely(!skb->ignore_df || 585 (IPCB(skb)->frag_max_size && 586 IPCB(skb)->frag_max_size > mtu))) { 587 IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); 588 icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, 589 htonl(mtu)); 590 kfree_skb(skb); 591 return -EMSGSIZE; 592 } 593 594 return ip_do_fragment(net, sk, skb, output); 595} 596 597void ip_fraglist_init(struct sk_buff *skb, struct iphdr *iph, 598 unsigned int hlen, struct ip_fraglist_iter *iter) 599{ 600 unsigned int first_len = skb_pagelen(skb); 601 602 iter->frag = skb_shinfo(skb)->frag_list; 603 skb_frag_list_init(skb); 604 605 iter->offset = 0; 606 iter->iph = iph; 607 iter->hlen = hlen; 608 609 skb->data_len = first_len - skb_headlen(skb); 610 skb->len = first_len; 611 iph->tot_len = htons(first_len); 612 iph->frag_off = htons(IP_MF); 613 ip_send_check(iph); 614} 615EXPORT_SYMBOL(ip_fraglist_init); 616 617void ip_fraglist_prepare(struct sk_buff *skb, struct ip_fraglist_iter *iter) 618{ 619 unsigned int hlen = iter->hlen; 620 struct iphdr *iph = iter->iph; 621 struct sk_buff *frag; 622 623 frag = iter->frag; 624 frag->ip_summed = CHECKSUM_NONE; 625 skb_reset_transport_header(frag); 626 __skb_push(frag, hlen); 627 skb_reset_network_header(frag); 628 memcpy(skb_network_header(frag), iph, hlen); 629 iter->iph = ip_hdr(frag); 630 iph = iter->iph; 631 iph->tot_len = htons(frag->len); 632 ip_copy_metadata(frag, skb); 633 iter->offset += skb->len - hlen; 634 iph->frag_off = htons(iter->offset >> 3); 635 if (frag->next) 636 iph->frag_off |= htons(IP_MF); 637 /* Ready, complete checksum */ 638 ip_send_check(iph); 639} 640EXPORT_SYMBOL(ip_fraglist_prepare); 641 642void ip_frag_init(struct sk_buff *skb, unsigned int hlen, 643 unsigned int ll_rs, unsigned int mtu, bool DF, 644 struct ip_frag_state *state) 645{ 646 struct iphdr *iph = ip_hdr(skb); 647 648 state->DF = DF; 649 state->hlen = hlen; 650 state->ll_rs = ll_rs; 651 state->mtu = mtu; 652 653 state->left = skb->len - hlen; /* Space per frame */ 654 state->ptr = hlen; /* Where to start from */ 655 656 state->offset = (ntohs(iph->frag_off) & IP_OFFSET) << 3; 657 state->not_last_frag = iph->frag_off & htons(IP_MF); 658} 659EXPORT_SYMBOL(ip_frag_init); 660 661static void ip_frag_ipcb(struct sk_buff *from, struct sk_buff *to, 662 bool first_frag) 663{ 664 /* Copy the flags to each fragment. */ 665 IPCB(to)->flags = IPCB(from)->flags; 666 667 /* ANK: dirty, but effective trick. Upgrade options only if 668 * the segment to be fragmented was THE FIRST (otherwise, 669 * options are already fixed) and make it ONCE 670 * on the initial skb, so that all the following fragments 671 * will inherit fixed options. 672 */ 673 if (first_frag) 674 ip_options_fragment(from); 675} 676 677struct sk_buff *ip_frag_next(struct sk_buff *skb, struct ip_frag_state *state) 678{ 679 unsigned int len = state->left; 680 struct sk_buff *skb2; 681 struct iphdr *iph; 682 683 /* IF: it doesn't fit, use 'mtu' - the data space left */ 684 if (len > state->mtu) 685 len = state->mtu; 686 /* IF: we are not sending up to and including the packet end 687 then align the next start on an eight byte boundary */ 688 if (len < state->left) { 689 len &= ~7; 690 } 691 692 /* Allocate buffer */ 693 skb2 = alloc_skb(len + state->hlen + state->ll_rs, GFP_ATOMIC); 694 if (!skb2) 695 return ERR_PTR(-ENOMEM); 696 697 /* 698 * Set up data on packet 699 */ 700 701 ip_copy_metadata(skb2, skb); 702 skb_reserve(skb2, state->ll_rs); 703 skb_put(skb2, len + state->hlen); 704 skb_reset_network_header(skb2); 705 skb2->transport_header = skb2->network_header + state->hlen; 706 707 /* 708 * Charge the memory for the fragment to any owner 709 * it might possess 710 */ 711 712 if (skb->sk) 713 skb_set_owner_w(skb2, skb->sk); 714 715 /* 716 * Copy the packet header into the new buffer. 717 */ 718 719 skb_copy_from_linear_data(skb, skb_network_header(skb2), state->hlen); 720 721 /* 722 * Copy a block of the IP datagram. 723 */ 724 if (skb_copy_bits(skb, state->ptr, skb_transport_header(skb2), len)) 725 BUG(); 726 state->left -= len; 727 728 /* 729 * Fill in the new header fields. 730 */ 731 iph = ip_hdr(skb2); 732 iph->frag_off = htons((state->offset >> 3)); 733 if (state->DF) 734 iph->frag_off |= htons(IP_DF); 735 736 /* 737 * Added AC : If we are fragmenting a fragment that's not the 738 * last fragment then keep MF on each bit 739 */ 740 if (state->left > 0 || state->not_last_frag) 741 iph->frag_off |= htons(IP_MF); 742 state->ptr += len; 743 state->offset += len; 744 745 iph->tot_len = htons(len + state->hlen); 746 747 ip_send_check(iph); 748 749 return skb2; 750} 751EXPORT_SYMBOL(ip_frag_next); 752 753/* 754 * This IP datagram is too large to be sent in one piece. Break it up into 755 * smaller pieces (each of size equal to IP header plus 756 * a block of the data of the original IP data part) that will yet fit in a 757 * single device frame, and queue such a frame for sending. 758 */ 759 760int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb, 761 int (*output)(struct net *, struct sock *, struct sk_buff *)) 762{ 763 struct iphdr *iph; 764 struct sk_buff *skb2; 765 bool mono_delivery_time = skb->mono_delivery_time; 766 struct rtable *rt = skb_rtable(skb); 767 unsigned int mtu, hlen, ll_rs; 768 struct ip_fraglist_iter iter; 769 ktime_t tstamp = skb->tstamp; 770 struct ip_frag_state state; 771 int err = 0; 772 773 /* for offloaded checksums cleanup checksum before fragmentation */ 774 if (skb->ip_summed == CHECKSUM_PARTIAL && 775 (err = skb_checksum_help(skb))) 776 goto fail; 777 778 /* 779 * Point into the IP datagram header. 780 */ 781 782 iph = ip_hdr(skb); 783 784 mtu = ip_skb_dst_mtu(sk, skb); 785 if (IPCB(skb)->frag_max_size && IPCB(skb)->frag_max_size < mtu) 786 mtu = IPCB(skb)->frag_max_size; 787 788 /* 789 * Setup starting values. 790 */ 791 792 hlen = iph->ihl * 4; 793 mtu = mtu - hlen; /* Size of data space */ 794 IPCB(skb)->flags |= IPSKB_FRAG_COMPLETE; 795 ll_rs = LL_RESERVED_SPACE(rt->dst.dev); 796 797 /* When frag_list is given, use it. First, check its validity: 798 * some transformers could create wrong frag_list or break existing 799 * one, it is not prohibited. In this case fall back to copying. 800 * 801 * LATER: this step can be merged to real generation of fragments, 802 * we can switch to copy when see the first bad fragment. 803 */ 804 if (skb_has_frag_list(skb)) { 805 struct sk_buff *frag, *frag2; 806 unsigned int first_len = skb_pagelen(skb); 807 808 if (first_len - hlen > mtu || 809 ((first_len - hlen) & 7) || 810 ip_is_fragment(iph) || 811 skb_cloned(skb) || 812 skb_headroom(skb) < ll_rs) 813 goto slow_path; 814 815 skb_walk_frags(skb, frag) { 816 /* Correct geometry. */ 817 if (frag->len > mtu || 818 ((frag->len & 7) && frag->next) || 819 skb_headroom(frag) < hlen + ll_rs) 820 goto slow_path_clean; 821 822 /* Partially cloned skb? */ 823 if (skb_shared(frag)) 824 goto slow_path_clean; 825 826 BUG_ON(frag->sk); 827 if (skb->sk) { 828 frag->sk = skb->sk; 829 frag->destructor = sock_wfree; 830 } 831 skb->truesize -= frag->truesize; 832 } 833 834 /* Everything is OK. Generate! */ 835 ip_fraglist_init(skb, iph, hlen, &iter); 836 837 for (;;) { 838 /* Prepare header of the next frame, 839 * before previous one went down. */ 840 if (iter.frag) { 841 bool first_frag = (iter.offset == 0); 842 843 IPCB(iter.frag)->flags = IPCB(skb)->flags; 844 ip_fraglist_prepare(skb, &iter); 845 if (first_frag && IPCB(skb)->opt.optlen) { 846 /* ipcb->opt is not populated for frags 847 * coming from __ip_make_skb(), 848 * ip_options_fragment() needs optlen 849 */ 850 IPCB(iter.frag)->opt.optlen = 851 IPCB(skb)->opt.optlen; 852 ip_options_fragment(iter.frag); 853 ip_send_check(iter.iph); 854 } 855 } 856 857 skb_set_delivery_time(skb, tstamp, mono_delivery_time); 858 err = output(net, sk, skb); 859 860 if (!err) 861 IP_INC_STATS(net, IPSTATS_MIB_FRAGCREATES); 862 if (err || !iter.frag) 863 break; 864 865 skb = ip_fraglist_next(&iter); 866 } 867 868 if (err == 0) { 869 IP_INC_STATS(net, IPSTATS_MIB_FRAGOKS); 870 return 0; 871 } 872 873 kfree_skb_list(iter.frag); 874 875 IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); 876 return err; 877 878slow_path_clean: 879 skb_walk_frags(skb, frag2) { 880 if (frag2 == frag) 881 break; 882 frag2->sk = NULL; 883 frag2->destructor = NULL; 884 skb->truesize += frag2->truesize; 885 } 886 } 887 888slow_path: 889 /* 890 * Fragment the datagram. 891 */ 892 893 ip_frag_init(skb, hlen, ll_rs, mtu, IPCB(skb)->flags & IPSKB_FRAG_PMTU, 894 &state); 895 896 /* 897 * Keep copying data until we run out. 898 */ 899 900 while (state.left > 0) { 901 bool first_frag = (state.offset == 0); 902 903 skb2 = ip_frag_next(skb, &state); 904 if (IS_ERR(skb2)) { 905 err = PTR_ERR(skb2); 906 goto fail; 907 } 908 ip_frag_ipcb(skb, skb2, first_frag); 909 910 /* 911 * Put this fragment into the sending queue. 912 */ 913 skb_set_delivery_time(skb2, tstamp, mono_delivery_time); 914 err = output(net, sk, skb2); 915 if (err) 916 goto fail; 917 918 IP_INC_STATS(net, IPSTATS_MIB_FRAGCREATES); 919 } 920 consume_skb(skb); 921 IP_INC_STATS(net, IPSTATS_MIB_FRAGOKS); 922 return err; 923 924fail: 925 kfree_skb(skb); 926 IP_INC_STATS(net, IPSTATS_MIB_FRAGFAILS); 927 return err; 928} 929EXPORT_SYMBOL(ip_do_fragment); 930 931int 932ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb) 933{ 934 struct msghdr *msg = from; 935 936 if (skb->ip_summed == CHECKSUM_PARTIAL) { 937 if (!copy_from_iter_full(to, len, &msg->msg_iter)) 938 return -EFAULT; 939 } else { 940 __wsum csum = 0; 941 if (!csum_and_copy_from_iter_full(to, len, &csum, &msg->msg_iter)) 942 return -EFAULT; 943 skb->csum = csum_block_add(skb->csum, csum, odd); 944 } 945 return 0; 946} 947EXPORT_SYMBOL(ip_generic_getfrag); 948 949static int __ip_append_data(struct sock *sk, 950 struct flowi4 *fl4, 951 struct sk_buff_head *queue, 952 struct inet_cork *cork, 953 struct page_frag *pfrag, 954 int getfrag(void *from, char *to, int offset, 955 int len, int odd, struct sk_buff *skb), 956 void *from, int length, int transhdrlen, 957 unsigned int flags) 958{ 959 struct inet_sock *inet = inet_sk(sk); 960 struct ubuf_info *uarg = NULL; 961 struct sk_buff *skb; 962 struct ip_options *opt = cork->opt; 963 int hh_len; 964 int exthdrlen; 965 int mtu; 966 int copy; 967 int err; 968 int offset = 0; 969 bool zc = false; 970 unsigned int maxfraglen, fragheaderlen, maxnonfragsize; 971 int csummode = CHECKSUM_NONE; 972 struct rtable *rt = (struct rtable *)cork->dst; 973 unsigned int wmem_alloc_delta = 0; 974 bool paged, extra_uref = false; 975 u32 tskey = 0; 976 977 skb = skb_peek_tail(queue); 978 979 exthdrlen = !skb ? rt->dst.header_len : 0; 980 mtu = cork->gso_size ? IP_MAX_MTU : cork->fragsize; 981 paged = !!cork->gso_size; 982 983 if (cork->tx_flags & SKBTX_ANY_TSTAMP && 984 READ_ONCE(sk->sk_tsflags) & SOF_TIMESTAMPING_OPT_ID) 985 tskey = atomic_inc_return(&sk->sk_tskey) - 1; 986 987 hh_len = LL_RESERVED_SPACE(rt->dst.dev); 988 989 fragheaderlen = sizeof(struct iphdr) + (opt ? opt->optlen : 0); 990 maxfraglen = ((mtu - fragheaderlen) & ~7) + fragheaderlen; 991 maxnonfragsize = ip_sk_ignore_df(sk) ? IP_MAX_MTU : mtu; 992 993 if (cork->length + length > maxnonfragsize - fragheaderlen) { 994 ip_local_error(sk, EMSGSIZE, fl4->daddr, inet->inet_dport, 995 mtu - (opt ? opt->optlen : 0)); 996 return -EMSGSIZE; 997 } 998 999 /* 1000 * transhdrlen > 0 means that this is the first fragment and we wish 1001 * it won't be fragmented in the future. 1002 */ 1003 if (transhdrlen && 1004 length + fragheaderlen <= mtu && 1005 rt->dst.dev->features & (NETIF_F_HW_CSUM | NETIF_F_IP_CSUM) && 1006 (!(flags & MSG_MORE) || cork->gso_size) && 1007 (!exthdrlen || (rt->dst.dev->features & NETIF_F_HW_ESP_TX_CSUM))) 1008 csummode = CHECKSUM_PARTIAL; 1009 1010 if ((flags & MSG_ZEROCOPY) && length) { 1011 struct msghdr *msg = from; 1012 1013 if (getfrag == ip_generic_getfrag && msg->msg_ubuf) { 1014 if (skb_zcopy(skb) && msg->msg_ubuf != skb_zcopy(skb)) 1015 return -EINVAL; 1016 1017 /* Leave uarg NULL if can't zerocopy, callers should 1018 * be able to handle it. 1019 */ 1020 if ((rt->dst.dev->features & NETIF_F_SG) && 1021 csummode == CHECKSUM_PARTIAL) { 1022 paged = true; 1023 zc = true; 1024 uarg = msg->msg_ubuf; 1025 } 1026 } else if (sock_flag(sk, SOCK_ZEROCOPY)) { 1027 uarg = msg_zerocopy_realloc(sk, length, skb_zcopy(skb)); 1028 if (!uarg) 1029 return -ENOBUFS; 1030 extra_uref = !skb_zcopy(skb); /* only ref on new uarg */ 1031 if (rt->dst.dev->features & NETIF_F_SG && 1032 csummode == CHECKSUM_PARTIAL) { 1033 paged = true; 1034 zc = true; 1035 } else { 1036 uarg_to_msgzc(uarg)->zerocopy = 0; 1037 skb_zcopy_set(skb, uarg, &extra_uref); 1038 } 1039 } 1040 } else if ((flags & MSG_SPLICE_PAGES) && length) { 1041 if (inet_test_bit(HDRINCL, sk)) 1042 return -EPERM; 1043 if (rt->dst.dev->features & NETIF_F_SG && 1044 getfrag == ip_generic_getfrag) 1045 /* We need an empty buffer to attach stuff to */ 1046 paged = true; 1047 else 1048 flags &= ~MSG_SPLICE_PAGES; 1049 } 1050 1051 cork->length += length; 1052 1053 /* So, what's going on in the loop below? 1054 * 1055 * We use calculated fragment length to generate chained skb, 1056 * each of segments is IP fragment ready for sending to network after 1057 * adding appropriate IP header. 1058 */ 1059 1060 if (!skb) 1061 goto alloc_new_skb; 1062 1063 while (length > 0) { 1064 /* Check if the remaining data fits into current packet. */ 1065 copy = mtu - skb->len; 1066 if (copy < length) 1067 copy = maxfraglen - skb->len; 1068 if (copy <= 0) { 1069 char *data; 1070 unsigned int datalen; 1071 unsigned int fraglen; 1072 unsigned int fraggap; 1073 unsigned int alloclen, alloc_extra; 1074 unsigned int pagedlen; 1075 struct sk_buff *skb_prev; 1076alloc_new_skb: 1077 skb_prev = skb; 1078 if (skb_prev) 1079 fraggap = skb_prev->len - maxfraglen; 1080 else 1081 fraggap = 0; 1082 1083 /* 1084 * If remaining data exceeds the mtu, 1085 * we know we need more fragment(s). 1086 */ 1087 datalen = length + fraggap; 1088 if (datalen > mtu - fragheaderlen) 1089 datalen = maxfraglen - fragheaderlen; 1090 fraglen = datalen + fragheaderlen; 1091 pagedlen = 0; 1092 1093 alloc_extra = hh_len + 15; 1094 alloc_extra += exthdrlen; 1095 1096 /* The last fragment gets additional space at tail. 1097 * Note, with MSG_MORE we overallocate on fragments, 1098 * because we have no idea what fragment will be 1099 * the last. 1100 */ 1101 if (datalen == length + fraggap) 1102 alloc_extra += rt->dst.trailer_len; 1103 1104 if ((flags & MSG_MORE) && 1105 !(rt->dst.dev->features&NETIF_F_SG)) 1106 alloclen = mtu; 1107 else if (!paged && 1108 (fraglen + alloc_extra < SKB_MAX_ALLOC || 1109 !(rt->dst.dev->features & NETIF_F_SG))) 1110 alloclen = fraglen; 1111 else { 1112 alloclen = fragheaderlen + transhdrlen; 1113 pagedlen = datalen - transhdrlen; 1114 } 1115 1116 alloclen += alloc_extra; 1117 1118 if (transhdrlen) { 1119 skb = sock_alloc_send_skb(sk, alloclen, 1120 (flags & MSG_DONTWAIT), &err); 1121 } else { 1122 skb = NULL; 1123 if (refcount_read(&sk->sk_wmem_alloc) + wmem_alloc_delta <= 1124 2 * sk->sk_sndbuf) 1125 skb = alloc_skb(alloclen, 1126 sk->sk_allocation); 1127 if (unlikely(!skb)) 1128 err = -ENOBUFS; 1129 } 1130 if (!skb) 1131 goto error; 1132 1133 /* 1134 * Fill in the control structures 1135 */ 1136 skb->ip_summed = csummode; 1137 skb->csum = 0; 1138 skb_reserve(skb, hh_len); 1139 1140 /* 1141 * Find where to start putting bytes. 1142 */ 1143 data = skb_put(skb, fraglen + exthdrlen - pagedlen); 1144 skb_set_network_header(skb, exthdrlen); 1145 skb->transport_header = (skb->network_header + 1146 fragheaderlen); 1147 data += fragheaderlen + exthdrlen; 1148 1149 if (fraggap) { 1150 skb->csum = skb_copy_and_csum_bits( 1151 skb_prev, maxfraglen, 1152 data + transhdrlen, fraggap); 1153 skb_prev->csum = csum_sub(skb_prev->csum, 1154 skb->csum); 1155 data += fraggap; 1156 pskb_trim_unique(skb_prev, maxfraglen); 1157 } 1158 1159 copy = datalen - transhdrlen - fraggap - pagedlen; 1160 /* [!] NOTE: copy will be negative if pagedlen>0 1161 * because then the equation reduces to -fraggap. 1162 */ 1163 if (copy > 0 && getfrag(from, data + transhdrlen, offset, copy, fraggap, skb) < 0) { 1164 err = -EFAULT; 1165 kfree_skb(skb); 1166 goto error; 1167 } else if (flags & MSG_SPLICE_PAGES) { 1168 copy = 0; 1169 } 1170 1171 offset += copy; 1172 length -= copy + transhdrlen; 1173 transhdrlen = 0; 1174 exthdrlen = 0; 1175 csummode = CHECKSUM_NONE; 1176 1177 /* only the initial fragment is time stamped */ 1178 skb_shinfo(skb)->tx_flags = cork->tx_flags; 1179 cork->tx_flags = 0; 1180 skb_shinfo(skb)->tskey = tskey; 1181 tskey = 0; 1182 skb_zcopy_set(skb, uarg, &extra_uref); 1183 1184 if ((flags & MSG_CONFIRM) && !skb_prev) 1185 skb_set_dst_pending_confirm(skb, 1); 1186 1187 /* 1188 * Put the packet on the pending queue. 1189 */ 1190 if (!skb->destructor) { 1191 skb->destructor = sock_wfree; 1192 skb->sk = sk; 1193 wmem_alloc_delta += skb->truesize; 1194 } 1195 __skb_queue_tail(queue, skb); 1196 continue; 1197 } 1198 1199 if (copy > length) 1200 copy = length; 1201 1202 if (!(rt->dst.dev->features&NETIF_F_SG) && 1203 skb_tailroom(skb) >= copy) { 1204 unsigned int off; 1205 1206 off = skb->len; 1207 if (getfrag(from, skb_put(skb, copy), 1208 offset, copy, off, skb) < 0) { 1209 __skb_trim(skb, off); 1210 err = -EFAULT; 1211 goto error; 1212 } 1213 } else if (flags & MSG_SPLICE_PAGES) { 1214 struct msghdr *msg = from; 1215 1216 err = -EIO; 1217 if (WARN_ON_ONCE(copy > msg->msg_iter.count)) 1218 goto error; 1219 1220 err = skb_splice_from_iter(skb, &msg->msg_iter, copy, 1221 sk->sk_allocation); 1222 if (err < 0) 1223 goto error; 1224 copy = err; 1225 wmem_alloc_delta += copy; 1226 } else if (!zc) { 1227 int i = skb_shinfo(skb)->nr_frags; 1228 1229 err = -ENOMEM; 1230 if (!sk_page_frag_refill(sk, pfrag)) 1231 goto error; 1232 1233 skb_zcopy_downgrade_managed(skb); 1234 if (!skb_can_coalesce(skb, i, pfrag->page, 1235 pfrag->offset)) { 1236 err = -EMSGSIZE; 1237 if (i == MAX_SKB_FRAGS) 1238 goto error; 1239 1240 __skb_fill_page_desc(skb, i, pfrag->page, 1241 pfrag->offset, 0); 1242 skb_shinfo(skb)->nr_frags = ++i; 1243 get_page(pfrag->page); 1244 } 1245 copy = min_t(int, copy, pfrag->size - pfrag->offset); 1246 if (getfrag(from, 1247 page_address(pfrag->page) + pfrag->offset, 1248 offset, copy, skb->len, skb) < 0) 1249 goto error_efault; 1250 1251 pfrag->offset += copy; 1252 skb_frag_size_add(&skb_shinfo(skb)->frags[i - 1], copy); 1253 skb_len_add(skb, copy); 1254 wmem_alloc_delta += copy; 1255 } else { 1256 err = skb_zerocopy_iter_dgram(skb, from, copy); 1257 if (err < 0) 1258 goto error; 1259 } 1260 offset += copy; 1261 length -= copy; 1262 } 1263 1264 if (wmem_alloc_delta) 1265 refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc); 1266 return 0; 1267 1268error_efault: 1269 err = -EFAULT; 1270error: 1271 net_zcopy_put_abort(uarg, extra_uref); 1272 cork->length -= length; 1273 IP_INC_STATS(sock_net(sk), IPSTATS_MIB_OUTDISCARDS); 1274 refcount_add(wmem_alloc_delta, &sk->sk_wmem_alloc); 1275 return err; 1276} 1277 1278static int ip_setup_cork(struct sock *sk, struct inet_cork *cork, 1279 struct ipcm_cookie *ipc, struct rtable **rtp) 1280{ 1281 struct ip_options_rcu *opt; 1282 struct rtable *rt; 1283 1284 rt = *rtp; 1285 if (unlikely(!rt)) 1286 return -EFAULT; 1287 1288 cork->fragsize = ip_sk_use_pmtu(sk) ? 1289 dst_mtu(&rt->dst) : READ_ONCE(rt->dst.dev->mtu); 1290 1291 if (!inetdev_valid_mtu(cork->fragsize)) 1292 return -ENETUNREACH; 1293 1294 /* 1295 * setup for corking. 1296 */ 1297 opt = ipc->opt; 1298 if (opt) { 1299 if (!cork->opt) { 1300 cork->opt = kmalloc(sizeof(struct ip_options) + 40, 1301 sk->sk_allocation); 1302 if (unlikely(!cork->opt)) 1303 return -ENOBUFS; 1304 } 1305 memcpy(cork->opt, &opt->opt, sizeof(struct ip_options) + opt->opt.optlen); 1306 cork->flags |= IPCORK_OPT; 1307 cork->addr = ipc->addr; 1308 } 1309 1310 cork->gso_size = ipc->gso_size; 1311 1312 cork->dst = &rt->dst; 1313 /* We stole this route, caller should not release it. */ 1314 *rtp = NULL; 1315 1316 cork->length = 0; 1317 cork->ttl = ipc->ttl; 1318 cork->tos = ipc->tos; 1319 cork->mark = ipc->sockc.mark; 1320 cork->priority = ipc->priority; 1321 cork->transmit_time = ipc->sockc.transmit_time; 1322 cork->tx_flags = 0; 1323 sock_tx_timestamp(sk, ipc->sockc.tsflags, &cork->tx_flags); 1324 1325 return 0; 1326} 1327 1328/* 1329 * ip_append_data() can make one large IP datagram from many pieces of 1330 * data. Each piece will be held on the socket until 1331 * ip_push_pending_frames() is called. Each piece can be a page or 1332 * non-page data. 1333 * 1334 * Not only UDP, other transport protocols - e.g. raw sockets - can use 1335 * this interface potentially. 1336 * 1337 * LATER: length must be adjusted by pad at tail, when it is required. 1338 */ 1339int ip_append_data(struct sock *sk, struct flowi4 *fl4, 1340 int getfrag(void *from, char *to, int offset, int len, 1341 int odd, struct sk_buff *skb), 1342 void *from, int length, int transhdrlen, 1343 struct ipcm_cookie *ipc, struct rtable **rtp, 1344 unsigned int flags) 1345{ 1346 struct inet_sock *inet = inet_sk(sk); 1347 int err; 1348 1349 if (flags&MSG_PROBE) 1350 return 0; 1351 1352 if (skb_queue_empty(&sk->sk_write_queue)) { 1353 err = ip_setup_cork(sk, &inet->cork.base, ipc, rtp); 1354 if (err) 1355 return err; 1356 } else { 1357 transhdrlen = 0; 1358 } 1359 1360 return __ip_append_data(sk, fl4, &sk->sk_write_queue, &inet->cork.base, 1361 sk_page_frag(sk), getfrag, 1362 from, length, transhdrlen, flags); 1363} 1364 1365static void ip_cork_release(struct inet_cork *cork) 1366{ 1367 cork->flags &= ~IPCORK_OPT; 1368 kfree(cork->opt); 1369 cork->opt = NULL; 1370 dst_release(cork->dst); 1371 cork->dst = NULL; 1372} 1373 1374/* 1375 * Combined all pending IP fragments on the socket as one IP datagram 1376 * and push them out. 1377 */ 1378struct sk_buff *__ip_make_skb(struct sock *sk, 1379 struct flowi4 *fl4, 1380 struct sk_buff_head *queue, 1381 struct inet_cork *cork) 1382{ 1383 struct sk_buff *skb, *tmp_skb; 1384 struct sk_buff **tail_skb; 1385 struct inet_sock *inet = inet_sk(sk); 1386 struct net *net = sock_net(sk); 1387 struct ip_options *opt = NULL; 1388 struct rtable *rt = (struct rtable *)cork->dst; 1389 struct iphdr *iph; 1390 __be16 df = 0; 1391 __u8 ttl; 1392 1393 skb = __skb_dequeue(queue); 1394 if (!skb) 1395 goto out; 1396 tail_skb = &(skb_shinfo(skb)->frag_list); 1397 1398 /* move skb->data to ip header from ext header */ 1399 if (skb->data < skb_network_header(skb)) 1400 __skb_pull(skb, skb_network_offset(skb)); 1401 while ((tmp_skb = __skb_dequeue(queue)) != NULL) { 1402 __skb_pull(tmp_skb, skb_network_header_len(skb)); 1403 *tail_skb = tmp_skb; 1404 tail_skb = &(tmp_skb->next); 1405 skb->len += tmp_skb->len; 1406 skb->data_len += tmp_skb->len; 1407 skb->truesize += tmp_skb->truesize; 1408 tmp_skb->destructor = NULL; 1409 tmp_skb->sk = NULL; 1410 } 1411 1412 /* Unless user demanded real pmtu discovery (IP_PMTUDISC_DO), we allow 1413 * to fragment the frame generated here. No matter, what transforms 1414 * how transforms change size of the packet, it will come out. 1415 */ 1416 skb->ignore_df = ip_sk_ignore_df(sk); 1417 1418 /* DF bit is set when we want to see DF on outgoing frames. 1419 * If ignore_df is set too, we still allow to fragment this frame 1420 * locally. */ 1421 if (inet->pmtudisc == IP_PMTUDISC_DO || 1422 inet->pmtudisc == IP_PMTUDISC_PROBE || 1423 (skb->len <= dst_mtu(&rt->dst) && 1424 ip_dont_fragment(sk, &rt->dst))) 1425 df = htons(IP_DF); 1426 1427 if (cork->flags & IPCORK_OPT) 1428 opt = cork->opt; 1429 1430 if (cork->ttl != 0) 1431 ttl = cork->ttl; 1432 else if (rt->rt_type == RTN_MULTICAST) 1433 ttl = inet->mc_ttl; 1434 else 1435 ttl = ip_select_ttl(inet, &rt->dst); 1436 1437 iph = ip_hdr(skb); 1438 iph->version = 4; 1439 iph->ihl = 5; 1440 iph->tos = (cork->tos != -1) ? cork->tos : inet->tos; 1441 iph->frag_off = df; 1442 iph->ttl = ttl; 1443 iph->protocol = sk->sk_protocol; 1444 ip_copy_addrs(iph, fl4); 1445 ip_select_ident(net, skb, sk); 1446 1447 if (opt) { 1448 iph->ihl += opt->optlen >> 2; 1449 ip_options_build(skb, opt, cork->addr, rt); 1450 } 1451 1452 skb->priority = (cork->tos != -1) ? cork->priority: sk->sk_priority; 1453 skb->mark = cork->mark; 1454 skb->tstamp = cork->transmit_time; 1455 /* 1456 * Steal rt from cork.dst to avoid a pair of atomic_inc/atomic_dec 1457 * on dst refcount 1458 */ 1459 cork->dst = NULL; 1460 skb_dst_set(skb, &rt->dst); 1461 1462 if (iph->protocol == IPPROTO_ICMP) { 1463 u8 icmp_type; 1464 1465 /* For such sockets, transhdrlen is zero when do ip_append_data(), 1466 * so icmphdr does not in skb linear region and can not get icmp_type 1467 * by icmp_hdr(skb)->type. 1468 */ 1469 if (sk->sk_type == SOCK_RAW && 1470 !inet_test_bit(HDRINCL, sk)) 1471 icmp_type = fl4->fl4_icmp_type; 1472 else 1473 icmp_type = icmp_hdr(skb)->type; 1474 icmp_out_count(net, icmp_type); 1475 } 1476 1477 ip_cork_release(cork); 1478out: 1479 return skb; 1480} 1481 1482int ip_send_skb(struct net *net, struct sk_buff *skb) 1483{ 1484 int err; 1485 1486 err = ip_local_out(net, skb->sk, skb); 1487 if (err) { 1488 if (err > 0) 1489 err = net_xmit_errno(err); 1490 if (err) 1491 IP_INC_STATS(net, IPSTATS_MIB_OUTDISCARDS); 1492 } 1493 1494 return err; 1495} 1496 1497int ip_push_pending_frames(struct sock *sk, struct flowi4 *fl4) 1498{ 1499 struct sk_buff *skb; 1500 1501 skb = ip_finish_skb(sk, fl4); 1502 if (!skb) 1503 return 0; 1504 1505 /* Netfilter gets whole the not fragmented skb. */ 1506 return ip_send_skb(sock_net(sk), skb); 1507} 1508 1509/* 1510 * Throw away all pending data on the socket. 1511 */ 1512static void __ip_flush_pending_frames(struct sock *sk, 1513 struct sk_buff_head *queue, 1514 struct inet_cork *cork) 1515{ 1516 struct sk_buff *skb; 1517 1518 while ((skb = __skb_dequeue_tail(queue)) != NULL) 1519 kfree_skb(skb); 1520 1521 ip_cork_release(cork); 1522} 1523 1524void ip_flush_pending_frames(struct sock *sk) 1525{ 1526 __ip_flush_pending_frames(sk, &sk->sk_write_queue, &inet_sk(sk)->cork.base); 1527} 1528 1529struct sk_buff *ip_make_skb(struct sock *sk, 1530 struct flowi4 *fl4, 1531 int getfrag(void *from, char *to, int offset, 1532 int len, int odd, struct sk_buff *skb), 1533 void *from, int length, int transhdrlen, 1534 struct ipcm_cookie *ipc, struct rtable **rtp, 1535 struct inet_cork *cork, unsigned int flags) 1536{ 1537 struct sk_buff_head queue; 1538 int err; 1539 1540 if (flags & MSG_PROBE) 1541 return NULL; 1542 1543 __skb_queue_head_init(&queue); 1544 1545 cork->flags = 0; 1546 cork->addr = 0; 1547 cork->opt = NULL; 1548 err = ip_setup_cork(sk, cork, ipc, rtp); 1549 if (err) 1550 return ERR_PTR(err); 1551 1552 err = __ip_append_data(sk, fl4, &queue, cork, 1553 ¤t->task_frag, getfrag, 1554 from, length, transhdrlen, flags); 1555 if (err) { 1556 __ip_flush_pending_frames(sk, &queue, cork); 1557 return ERR_PTR(err); 1558 } 1559 1560 return __ip_make_skb(sk, fl4, &queue, cork); 1561} 1562 1563/* 1564 * Fetch data from kernel space and fill in checksum if needed. 1565 */ 1566static int ip_reply_glue_bits(void *dptr, char *to, int offset, 1567 int len, int odd, struct sk_buff *skb) 1568{ 1569 __wsum csum; 1570 1571 csum = csum_partial_copy_nocheck(dptr+offset, to, len); 1572 skb->csum = csum_block_add(skb->csum, csum, odd); 1573 return 0; 1574} 1575 1576/* 1577 * Generic function to send a packet as reply to another packet. 1578 * Used to send some TCP resets/acks so far. 1579 */ 1580void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, 1581 const struct ip_options *sopt, 1582 __be32 daddr, __be32 saddr, 1583 const struct ip_reply_arg *arg, 1584 unsigned int len, u64 transmit_time, u32 txhash) 1585{ 1586 struct ip_options_data replyopts; 1587 struct ipcm_cookie ipc; 1588 struct flowi4 fl4; 1589 struct rtable *rt = skb_rtable(skb); 1590 struct net *net = sock_net(sk); 1591 struct sk_buff *nskb; 1592 int err; 1593 int oif; 1594 1595 if (__ip_options_echo(net, &replyopts.opt.opt, skb, sopt)) 1596 return; 1597 1598 ipcm_init(&ipc); 1599 ipc.addr = daddr; 1600 ipc.sockc.transmit_time = transmit_time; 1601 1602 if (replyopts.opt.opt.optlen) { 1603 ipc.opt = &replyopts.opt; 1604 1605 if (replyopts.opt.opt.srr) 1606 daddr = replyopts.opt.opt.faddr; 1607 } 1608 1609 oif = arg->bound_dev_if; 1610 if (!oif && netif_index_is_l3_master(net, skb->skb_iif)) 1611 oif = skb->skb_iif; 1612 1613 flowi4_init_output(&fl4, oif, 1614 IP4_REPLY_MARK(net, skb->mark) ?: sk->sk_mark, 1615 RT_TOS(arg->tos), 1616 RT_SCOPE_UNIVERSE, ip_hdr(skb)->protocol, 1617 ip_reply_arg_flowi_flags(arg), 1618 daddr, saddr, 1619 tcp_hdr(skb)->source, tcp_hdr(skb)->dest, 1620 arg->uid); 1621 security_skb_classify_flow(skb, flowi4_to_flowi_common(&fl4)); 1622 rt = ip_route_output_flow(net, &fl4, sk); 1623 if (IS_ERR(rt)) 1624 return; 1625 1626 inet_sk(sk)->tos = arg->tos & ~INET_ECN_MASK; 1627 1628 sk->sk_protocol = ip_hdr(skb)->protocol; 1629 sk->sk_bound_dev_if = arg->bound_dev_if; 1630 sk->sk_sndbuf = READ_ONCE(sysctl_wmem_default); 1631 ipc.sockc.mark = fl4.flowi4_mark; 1632 err = ip_append_data(sk, &fl4, ip_reply_glue_bits, arg->iov->iov_base, 1633 len, 0, &ipc, &rt, MSG_DONTWAIT); 1634 if (unlikely(err)) { 1635 ip_flush_pending_frames(sk); 1636 goto out; 1637 } 1638 1639 nskb = skb_peek(&sk->sk_write_queue); 1640 if (nskb) { 1641 if (arg->csumoffset >= 0) 1642 *((__sum16 *)skb_transport_header(nskb) + 1643 arg->csumoffset) = csum_fold(csum_add(nskb->csum, 1644 arg->csum)); 1645 nskb->ip_summed = CHECKSUM_NONE; 1646 nskb->mono_delivery_time = !!transmit_time; 1647 if (txhash) 1648 skb_set_hash(nskb, txhash, PKT_HASH_TYPE_L4); 1649 ip_push_pending_frames(sk, &fl4); 1650 } 1651out: 1652 ip_rt_put(rt); 1653} 1654 1655void __init ip_init(void) 1656{ 1657 ip_rt_init(); 1658 inet_initpeers(); 1659 1660#if defined(CONFIG_IP_MULTICAST) 1661 igmp_mc_init(); 1662#endif 1663} 1664