cregit-Linux how code gets into the kernel

Release 4.14 net/ipv6/raw.c

Directory: net/ipv6
 *      RAW sockets for IPv6
 *      Linux INET6 implementation
 *      Authors:
 *      Pedro Roque             <>
 *      Adapted from linux/net/ipv4/raw.c
 *      Fixes:
 *      Hideaki YOSHIFUJI       :       sin6_scope_id support
 *      YOSHIFUJI,H.@USAGI      :       raw checksum (RFC2292(bis) compliance)
 *      Kazunori MIYAZAWA @USAGI:       change process style to use ip6_append_data
 *      This program is free software; you can redistribute it and/or
 *      modify it under the terms of the GNU General Public License
 *      as published by the Free Software Foundation; either version
 *      2 of the License, or (at your option) any later version.

#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/slab.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/in6.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/icmpv6.h>
#include <linux/netfilter.h>
#include <linux/netfilter_ipv6.h>
#include <linux/skbuff.h>
#include <linux/compat.h>
#include <linux/uaccess.h>
#include <asm/ioctls.h>

#include <net/net_namespace.h>
#include <net/ip.h>
#include <net/sock.h>
#include <net/snmp.h>

#include <net/ipv6.h>
#include <net/ndisc.h>
#include <net/protocol.h>
#include <net/ip6_route.h>
#include <net/ip6_checksum.h>
#include <net/addrconf.h>
#include <net/transp_v6.h>
#include <net/udp.h>
#include <net/inet_common.h>
#include <net/tcp_states.h>
#include <net/mip6.h>
#include <linux/mroute6.h>

#include <net/raw.h>
#include <net/rawv6.h>
#include <net/xfrm.h>

#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/export.h>

#define	ICMPV6_HDRLEN	4	
/* ICMPv6 header, RFC 4443 Section 2.1 */

struct raw_hashinfo raw_v6_hashinfo = {
	.lock = __RW_LOCK_UNLOCKED(raw_v6_hashinfo.lock),


struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, unsigned short num, const struct in6_addr *loc_addr, const struct in6_addr *rmt_addr, int dif, int sdif) { bool is_multicast = ipv6_addr_is_multicast(loc_addr); sk_for_each_from(sk) if (inet_sk(sk)->inet_num == num) { if (!net_eq(sock_net(sk), net)) continue; if (!ipv6_addr_any(&sk->sk_v6_daddr) && !ipv6_addr_equal(&sk->sk_v6_daddr, rmt_addr)) continue; if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif && sk->sk_bound_dev_if != sdif) continue; if (!ipv6_addr_any(&sk->sk_v6_rcv_saddr)) { if (ipv6_addr_equal(&sk->sk_v6_rcv_saddr, loc_addr)) goto found; if (is_multicast && inet6_mc_check(sk, loc_addr, rmt_addr)) goto found; continue; } goto found; } sk = NULL; found: return sk; }


Linus Torvalds (pre-git)8849.72%212.50%
Arnaldo Carvalho de Melo2212.43%16.25%
Andrew McDonald169.04%16.25%
Hideaki Yoshifuji / 吉藤英明147.91%425.00%
Eric Dumazet126.78%425.00%
Pavel Emelyanov116.21%16.25%
David Ahern95.08%16.25%
David S. Miller31.69%16.25%
David L Stevens21.13%16.25%

EXPORT_SYMBOL_GPL(__raw_v6_lookup); /* * 0 - deliver * 1 - block */
static int icmpv6_filter(const struct sock *sk, const struct sk_buff *skb) { struct icmp6hdr _hdr; const struct icmp6hdr *hdr; /* We require only the four bytes of the ICMPv6 header, not any * additional bytes of message body in "struct icmp6hdr". */ hdr = skb_header_pointer(skb, skb_transport_offset(skb), ICMPV6_HDRLEN, &_hdr); if (hdr) { const __u32 *data = &raw6_sk(sk)->[0]; unsigned int type = hdr->icmp6_type; return (data[type >> 5] & (1U << (type & 31))) != 0; } return 1; }


Linus Torvalds4038.83%228.57%
Eric Dumazet3533.98%114.29%
Linus Torvalds (pre-git)2524.27%228.57%
Werner Almesberger21.94%114.29%
David S. Miller10.97%114.29%

#if IS_ENABLED(CONFIG_IPV6_MIP6) typedef int mh_filter_t(struct sock *sock, struct sk_buff *skb); static mh_filter_t __rcu *mh_filter __read_mostly;
int rawv6_mh_filter_register(mh_filter_t filter) { rcu_assign_pointer(mh_filter, filter); return 0; }


Masahide Nakamura1688.89%133.33%
Eric Dumazet211.11%266.67%

int rawv6_mh_filter_unregister(mh_filter_t filter) { RCU_INIT_POINTER(mh_filter, NULL); synchronize_rcu(); return 0; }


Masahide Nakamura1990.48%133.33%
Eric Dumazet14.76%133.33%
Stephen Hemminger14.76%133.33%

EXPORT_SYMBOL(rawv6_mh_filter_unregister); #endif /* * demultiplex raw sockets. * (should consider queueing the skb in the sock receive_queue * without calling rawv6.c) * * Caller owns SKB so we must make clones. */
static bool ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) { const struct in6_addr *saddr; const struct in6_addr *daddr; struct sock *sk; bool delivered = false; __u8 hash; struct net *net; saddr = &ipv6_hdr(skb)->saddr; daddr = saddr + 1; hash = nexthdr & (RAW_HTABLE_SIZE - 1); read_lock(&raw_v6_hashinfo.lock); sk = sk_head(&[hash]); if (!sk) goto out; net = dev_net(skb->dev); sk = __raw_v6_lookup(net, sk, nexthdr, daddr, saddr, inet6_iif(skb), inet6_sdif(skb)); while (sk) { int filtered; delivered = true; switch (nexthdr) { case IPPROTO_ICMPV6: filtered = icmpv6_filter(sk, skb); break; #if IS_ENABLED(CONFIG_IPV6_MIP6) case IPPROTO_MH: { /* XXX: To validate MH only once for each packet, * this is placed here. It should be after checking * xfrm policy, however it doesn't. The checking xfrm * policy is placed in rawv6_rcv() because it is * required for each socket. */ mh_filter_t *filter; filter = rcu_dereference(mh_filter); filtered = filter ? (*filter)(sk, skb) : 0; break; } #endif default: filtered = 0; break; } if (filtered < 0) break; if (filtered == 0) { struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); /* Not releasing hash table! */ if (clone) { nf_reset(clone); rawv6_rcv(sk, clone); } } sk = __raw_v6_lookup(net, sk_next(sk), nexthdr, daddr, saddr, inet6_iif(skb), inet6_sdif(skb)); } out: read_unlock(&raw_v6_hashinfo.lock); return delivered; }


Linus Torvalds (pre-git)12341.00%28.33%
Masahide Nakamura6722.33%28.33%
Pavel Emelyanov258.33%312.50%
David S. Miller227.33%28.33%
Eric Dumazet113.67%312.50%
Arnaldo Carvalho de Melo113.67%312.50%
David Ahern103.33%14.17%
Patrick McHardy93.00%14.17%
Hideaki Yoshifuji / 吉藤英明72.33%28.33%
Yasuyuki Kozakai72.33%14.17%
Andrew McDonald41.33%14.17%
Duan Jiong20.67%14.17%
Ian Morris10.33%14.17%
Américo Wang10.33%14.17%

bool raw6_local_deliver(struct sk_buff *skb, int nexthdr) { struct sock *raw_sk; raw_sk = sk_head(&[nexthdr & (RAW_HTABLE_SIZE - 1)]); if (raw_sk && !ipv6_raw_deliver(skb, nexthdr)) raw_sk = NULL; return raw_sk != NULL; }


Pavel Emelyanov5696.55%250.00%
David S. Miller11.72%125.00%
Eric Dumazet11.72%125.00%

/* This cleans up af_inet6 a bit. -DaveM */
static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct sockaddr_in6 *addr = (struct sockaddr_in6 *) uaddr; __be32 v4addr = 0; int addr_type; int err; if (addr_len < SIN6_LEN_RFC2133) return -EINVAL; if (addr->sin6_family != AF_INET6) return -EINVAL; addr_type = ipv6_addr_type(&addr->sin6_addr); /* Raw sockets are IPv6 only */ if (addr_type == IPV6_ADDR_MAPPED) return -EADDRNOTAVAIL; lock_sock(sk); err = -EINVAL; if (sk->sk_state != TCP_CLOSE) goto out; rcu_read_lock(); /* Check if the address belongs to the host. */ if (addr_type != IPV6_ADDR_ANY) { struct net_device *dev = NULL; if (__ipv6_addr_needs_scope_id(addr_type)) { if (addr_len >= sizeof(struct sockaddr_in6) && addr->sin6_scope_id) { /* Override any existing binding, if another * one is supplied by user. */ sk->sk_bound_dev_if = addr->sin6_scope_id; } /* Binding to link-local address requires an interface */ if (!sk->sk_bound_dev_if) goto out_unlock; err = -ENODEV; dev = dev_get_by_index_rcu(sock_net(sk), sk->sk_bound_dev_if); if (!dev) goto out_unlock; } /* ipv4 addr of the socket is invalid. Only the * unspecified and mapped address have a v4 equivalent. */ v4addr = LOOPBACK4_IPV6; if (!(addr_type & IPV6_ADDR_MULTICAST) && !sock_net(sk)->ipv6.sysctl.ip_nonlocal_bind) { err = -EADDRNOTAVAIL; if (!ipv6_chk_addr(sock_net(sk), &addr->sin6_addr, dev, 0)) { goto out_unlock; } } } inet->inet_rcv_saddr = inet->inet_saddr = v4addr; sk->sk_v6_rcv_saddr = addr->sin6_addr; if (!(addr_type & IPV6_ADDR_MULTICAST)) np->saddr = addr->sin6_addr; err = 0; out_unlock: rcu_read_unlock(); out: release_sock(sk); return err; }


Linus Torvalds (pre-git)20562.31%417.39%
Ville Nuorvala3410.33%28.70%
David S. Miller226.69%28.70%
Eric Dumazet216.38%313.04%
Hannes Frederic Sowa154.56%28.70%
Tom Herbert123.65%14.35%
Hideaki Yoshifuji / 吉藤英明72.13%28.70%
Arnaldo Carvalho de Melo51.52%28.70%
Daniel Lezcano20.61%14.35%
Alexey Dobriyan20.61%14.35%
Pavel Emelyanov20.61%14.35%
Al Viro10.30%14.35%
Eric W. Biedermann10.30%14.35%

static void rawv6_err(struct sock *sk, struct sk_buff *skb, struct inet6_skb_parm *opt, u8 type, u8 code, int offset, __be32 info) { struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); int err; int harderr; /* Report error on raw socket, if: 1. User requested recverr. 2. Socket is connected (otherwise the error indication is useless without recverr and error is hard. */ if (!np->recverr && sk->sk_state != TCP_ESTABLISHED) return; harderr = icmpv6_err_convert(type, code, &err); if (type == ICMPV6_PKT_TOOBIG) { ip6_sk_update_pmtu(skb, sk, info); harderr = (np->pmtudisc == IPV6_PMTUDISC_DO); } if (type == NDISC_REDIRECT) { ip6_sk_redirect(skb, sk); return; } if (np->recverr) { u8 *payload = skb->data; if (!inet->hdrincl) payload += offset; ipv6_icmp_error(sk, skb, err, 0, ntohl(info), payload); } if (np->recverr || harderr) { sk->sk_err = err; sk->sk_error_report(sk); } }


Linus Torvalds (pre-git)11558.38%323.08%
David S. Miller4824.37%323.08%
Linus Torvalds2311.68%17.69%
Arnaldo Carvalho de Melo42.03%215.38%
Duan Jiong31.52%17.69%
Brian Haley21.02%17.69%
Pavel Emelyanov10.51%17.69%
Al Viro10.51%17.69%

void raw6_icmp_error(struct sk_buff *skb, int nexthdr, u8 type, u8 code, int inner_offset, __be32 info) { struct sock *sk; int hash; const struct in6_addr *saddr, *daddr; struct net *net; hash = nexthdr & (RAW_HTABLE_SIZE - 1); read_lock(&raw_v6_hashinfo.lock); sk = sk_head(&[hash]); if (sk) { /* Note: ipv6_hdr(skb) != skb->data */ const struct ipv6hdr *ip6h = (const struct ipv6hdr *)skb->data; saddr = &ip6h->saddr; daddr = &ip6h->daddr; net = dev_net(skb->dev); while ((sk = __raw_v6_lookup(net, sk, nexthdr, saddr, daddr, inet6_iif(skb), inet6_iif(skb)))) { rawv6_err(sk, skb, NULL, type, code, inner_offset, info); sk = sk_next(sk); } } read_unlock(&raw_v6_hashinfo.lock); }


Pavel Emelyanov15683.42%333.33%
Hideaki Yoshifuji / 吉藤英明2010.70%222.22%
David Ahern52.67%111.11%
Eric Dumazet31.60%111.11%
Brian Haley21.07%111.11%
Duan Jiong10.53%111.11%

static inline int rawv6_rcv_skb(struct sock *sk, struct sk_buff *skb) { if ((raw6_sk(sk)->checksum || rcu_access_pointer(sk->sk_filter)) && skb_checksum_complete(skb)) { atomic_inc(&sk->sk_drops); kfree_skb(skb); return NET_RX_DROP; } /* Charge it to the socket. */ skb_dst_drop(skb); if (sock_queue_rcv_skb(sk, skb) < 0) { kfree_skb(skb); return NET_RX_DROP; } return 0; }


Linus Torvalds (pre-git)4045.45%110.00%
Alexey Kuznetsov1820.45%110.00%
Eric Dumazet910.23%330.00%
Hideaki Yoshifuji / 吉藤英明910.23%110.00%
Wang Chen89.09%110.00%
Yang Hongyang22.27%110.00%
Arnaldo Carvalho de Melo11.14%110.00%
Herbert Xu11.14%110.00%

/* * This is next to useless... * if we demultiplex in network layer we don't need the extra call * just to queue the skb... * maybe we could have the network decide upon a hint if it * should call raw_rcv for demultiplexing */
int rawv6_rcv(struct sock *sk, struct sk_buff *skb) { struct inet_sock *inet = inet_sk(sk); struct raw6_sock *rp = raw6_sk(sk); if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) { atomic_inc(&sk->sk_drops); kfree_skb(skb); return NET_RX_DROP; } if (!rp->checksum) skb->ip_summed = CHECKSUM_UNNECESSARY; if (skb->ip_summed == CHECKSUM_COMPLETE) { skb_postpull_rcsum(skb, skb_network_header(skb), skb_network_header_len(skb)); if (!csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->len, inet->inet_num, skb->csum)) skb->ip_summed = CHECKSUM_UNNECESSARY; } if (!skb_csum_unnecessary(skb)) skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->len, inet->inet_num, 0)); if (inet->hdrincl) { if (skb_checksum_complete(skb)) { atomic_inc(&sk->sk_drops); kfree_skb(skb); return NET_RX_DROP; } } rawv6_rcv_skb(sk, skb); return 0; }


Alexey Kuznetsov9741.99%15.00%
Linus Torvalds (pre-git)3213.85%315.00%
Kazunori Miyazawa229.52%15.00%
Arnaldo Carvalho de Melo229.52%525.00%
Wang Chen166.93%15.00%
Linus Torvalds104.33%15.00%
Patrick McHardy104.33%210.00%
David S. Miller93.90%15.00%
Herbert Xu73.03%210.00%
Al Viro31.30%15.00%
Eric Dumazet20.87%15.00%
Yang Hongyang10.43%15.00%

/* * This should be easy, if there is something there * we return it, otherwise we block. */
static int rawv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int noblock, int flags, int *addr_len) { struct ipv6_pinfo *np = inet6_sk(sk); DECLARE_SOCKADDR(struct sockaddr_in6 *, sin6, msg->msg_name); struct sk_buff *skb; size_t copied; int err; if (flags & MSG_OOB) return -EOPNOTSUPP; if (flags & MSG_ERRQUEUE) return ipv6_recv_error(sk, msg, len, addr_len); if (np->rxpmtu && np->rxopt.bits.rxpmtu) return ipv6_recv_rxpmtu(sk, msg, len, addr_len); skb = skb_recv_datagram(sk, flags, noblock, &err); if (!skb) goto out; copied = skb->len; if (copied > len) { copied = len; msg->msg_flags |= MSG_TRUNC; } if (skb_csum_unnecessary(skb)) { err = skb_copy_datagram_msg(skb, 0, msg, copied); } else if (msg->msg_flags&MSG_TRUNC) { if (__skb_checksum_complete(skb)) goto csum_copy_err; err = skb_copy_datagram_msg(skb, 0, msg, copied); } else { err = skb_copy_and_csum_datagram_msg(skb, 0, msg); if (err == -EINVAL) goto csum_copy_err; } if (err) goto out_free; /* Copy the address. */ if (sin6) { sin6->sin6_family = AF_INET6; sin6->sin6_port = 0; sin6->sin6_addr = ipv6_hdr(skb)->saddr; sin6->sin6_flowinfo = 0; sin6->sin6_scope_id = ipv6_iface_scope_id(&sin6->sin6_addr, inet6_iif(skb)); *addr_len = sizeof(*sin6); } sock_recv_ts_and_drops(msg, sk, skb); if (np->rxopt.all) ip6_datagram_recv_ctl(sk, msg, skb); err = copied; if (flags & MSG_TRUNC) err = skb->len; out_free: skb_free_datagram(sk, skb); out: return err; csum_copy_err: skb_kill_datagram(sk, skb, flags); /* Error for blocking case is chosen to masquerade as some normal condition. */ err = (flags&MSG_DONTWAIT) ? -EAGAIN : -EHOSTUNREACH; goto out; }


Linus Torvalds (pre-git)20952.64%825.81%
Alexey Kuznetsov8621.66%13.23%
Brian Haley246.05%13.23%
David S. Miller246.05%39.68%
Hannes Frederic Sowa164.03%39.68%
Herbert Xu82.02%39.68%
Steffen Hurrle71.76%13.23%
Tetsuo Handa61.51%13.23%
Stephen Hemminger41.01%13.23%
Arnaldo Carvalho de Melo30.76%13.23%
Hideaki Yoshifuji / 吉藤英明30.76%13.23%
Duan Jiong10.25%13.23%
Tom Parkin10.25%13.23%
Al Viro10.25%13.23%
Alexey Dobriyan10.25%13.23%
Stephen Rothwell10.25%13.23%
Neil Horman10.25%13.23%
Linus Torvalds10.25%13.23%

static int rawv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, struct raw6_sock *rp) { struct sk_buff *skb; int err = 0; int offset; int len; int total_len; __wsum tmp_csum; __sum16 csum; if (!rp->checksum) goto send; skb = skb_peek(&sk->sk_write_queue); if (!skb) goto out; offset = rp->offset; total_len = inet_sk(sk)->cork.base.length; if (offset >= total_len - 1) { err = -EINVAL; ip6_flush_pending_frames(sk); goto out; } /* should be check HW csum miyazawa */ if (skb_queue_len(&sk->sk_write_queue) == 1) { /* * Only one fragment on the socket. */ tmp_csum = skb->csum; } else { struct sk_buff *csum_skb = NULL; tmp_csum = 0; skb_queue_walk(&sk->sk_write_queue, skb) { tmp_csum = csum_add(tmp_csum, skb->csum); if (csum_skb) continue; len = skb->len - skb_transport_offset(skb); if (offset >= len) { offset -= len; continue; } csum_skb = skb; } skb = csum_skb; } offset += skb_transport_offset(skb); err = skb_copy_bits(skb, offset, &csum, 2); if (err < 0) { ip6_flush_pending_frames(sk); goto out; } /* in case cksum was not initialized */ if (unlikely(csum)) tmp_csum = csum_sub(tmp_csum, csum_unfold(csum)); csum = csum_ipv6_magic(&fl6->saddr, &fl6->daddr, total_len, fl6->flowi6_proto, tmp_csum); if (csum == 0 && fl6->flowi6_proto == IPPROTO_UDP) csum = CSUM_MANGLED_0; BUG_ON(skb_store_bits(skb, offset, &csum, 2)); send: err = ip6_push_pending_frames(sk); out: return err; }


Herbert Xu10431.23%28.33%
Kazunori Miyazawa8926.73%14.17%
Linus Torvalds (pre-git)3610.81%312.50%
Hideaki Yoshifuji / 吉藤英明3510.51%416.67%
Dave Jones175.11%14.17%
David S. Miller144.20%312.50%
Arnaldo Carvalho de Melo133.90%312.50%
Al Viro92.70%312.50%
Fabian Frederick51.50%14.17%
Brian Haley41.20%14.17%
Himangi Saraogi41.20%14.17%
Alexey Kuznetsov30.90%14.17%

static int rawv6_send_hdrinc(struct sock *sk, struct msghdr *msg, int length, struct flowi6 *fl6, struct dst_entry **dstp, unsigned int flags) { struct ipv6_pinfo *np = inet6_sk(sk); struct net *net = sock_net(sk); struct ipv6hdr *iph; struct sk_buff *skb; int err; struct rt6_info *rt = (struct rt6_info *)*dstp; int hlen = LL_RESERVED_SPACE(rt->; int tlen = rt->>needed_tailroom; if (length > rt->>mtu) { ipv6_local_error(sk, EMSGSIZE, fl6, rt