X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=net%2Fipv6%2Ficmp.c;h=4ae661bc3677b4efc9a5a205086f4922d8c6806e;hb=7eca61eb6a57a3aae10e77d9306cda0b8c23cdab;hp=8e0f569b883e419acd311e79670d9ea916f0a034;hpb=0d3d077cd4f1154e63a9858e47fe3fb1ad0c03e5;p=safe%2Fjmp%2Flinux-2.6 diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 8e0f569..4ae661b 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -5,8 +5,6 @@ * Authors: * Pedro Roque * - * $Id: icmp.c,v 1.38 2002/02/08 03:57:19 davem Exp $ - * * Based on net/ipv4/icmp.c * * RFC 1885 @@ -37,11 +35,11 @@ #include #include #include -#include #include #include #include #include +#include #ifdef CONFIG_SYSCTL #include @@ -63,11 +61,16 @@ #include #include #include +#include +#include #include #include -DEFINE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics); +DEFINE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics) __read_mostly; +EXPORT_SYMBOL(icmpv6_statistics); +DEFINE_SNMP_STAT(struct icmpv6msg_mib, icmpv6msg_statistics) __read_mostly; +EXPORT_SYMBOL(icmpv6msg_statistics); /* * The ICMP socket(s). This is the most convenient way to flow control @@ -76,40 +79,45 @@ DEFINE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics); * * On SMP we have one ICMP socket per-cpu. */ -static DEFINE_PER_CPU(struct socket *, __icmpv6_socket) = NULL; -#define icmpv6_socket __get_cpu_var(__icmpv6_socket) +static inline struct sock *icmpv6_sk(struct net *net) +{ + return net->ipv6.icmp_sk[smp_processor_id()]; +} -static int icmpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp); +static int icmpv6_rcv(struct sk_buff *skb); -static struct inet6_protocol icmpv6_protocol = { +static const struct inet6_protocol icmpv6_protocol = { .handler = icmpv6_rcv, - .flags = INET6_PROTO_FINAL, + .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -static __inline__ int icmpv6_xmit_lock(void) +static __inline__ struct sock *icmpv6_xmit_lock(struct net *net) { + struct sock *sk; + local_bh_disable(); - if (unlikely(!spin_trylock(&icmpv6_socket->sk->sk_lock.slock))) { + sk = icmpv6_sk(net); + if (unlikely(!spin_trylock(&sk->sk_lock.slock))) { /* This can happen if the output path (f.e. SIT or * ip6ip6 tunnel) signals dst_link_failure() for an * outgoing ICMP6 packet. */ local_bh_enable(); - return 1; + return NULL; } - return 0; + return sk; } -static __inline__ void icmpv6_xmit_unlock(void) +static __inline__ void icmpv6_xmit_unlock(struct sock *sk) { - spin_unlock_bh(&icmpv6_socket->sk->sk_lock.slock); + spin_unlock_bh(&sk->sk_lock.slock); } -/* +/* * Slightly more convenient version of icmpv6_send. */ -void icmpv6_param_prob(struct sk_buff *skb, int code, int pos) +void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) { icmpv6_send(skb, ICMPV6_PARAMPROB, code, pos, skb->dev); kfree_skb(skb); @@ -128,9 +136,9 @@ void icmpv6_param_prob(struct sk_buff *skb, int code, int pos) static int is_ineligible(struct sk_buff *skb) { - int ptr = (u8*)(skb->nh.ipv6h+1) - skb->data; + int ptr = (u8 *)(ipv6_hdr(skb) + 1) - skb->data; int len = skb->len - ptr; - __u8 nexthdr = skb->nh.ipv6h->nexthdr; + __u8 nexthdr = ipv6_hdr(skb)->nexthdr; if (len < 0) return 1; @@ -150,15 +158,14 @@ static int is_ineligible(struct sk_buff *skb) return 0; } -static int sysctl_icmpv6_time = 1*HZ; - -/* - * Check the ICMP output rate limit +/* + * Check the ICMP output rate limit */ -static inline int icmpv6_xrlim_allow(struct sock *sk, int type, +static inline int icmpv6_xrlim_allow(struct sock *sk, u8 type, struct flowi *fl) { struct dst_entry *dst; + struct net *net = sock_net(sk); int res = 0; /* Informational messages are not limited. */ @@ -169,19 +176,20 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type, if (type == ICMPV6_PKT_TOOBIG) return 1; - /* + /* * Look up the output route. * XXX: perhaps the expire for routing entries cloned by * this lookup should be more aggressive (not longer than timeout). */ - dst = ip6_route_output(sk, fl); + dst = ip6_route_output(net, sk, fl); if (dst->error) { - IP6_INC_STATS(IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(net, ip6_dst_idev(dst), + IPSTATS_MIB_OUTNOROUTES); } else if (dst->dev && (dst->dev->flags&IFF_LOOPBACK)) { res = 1; } else { struct rt6_info *rt = (struct rt6_info *)dst; - int tmo = sysctl_icmpv6_time; + int tmo = net->ipv6.sysctl.icmpv6_time; /* Give more bandwidth to wider prefixes. */ if (rt->rt6i_dst.plen < 128) @@ -196,7 +204,7 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type, /* * an inline helper for the "simple" if statement below * checks if parameter problem report is caused by an - * unrecognized IPv6 option that has the Option Type + * unrecognized IPv6 option that has the Option Type * highest-order two bits set to 10 */ @@ -204,7 +212,7 @@ static __inline__ int opt_unrec(struct sk_buff *skb, __u32 offset) { u8 _optval, *op; - offset += skb->nh.raw - skb->data; + offset += skb_network_offset(skb); op = skb_header_pointer(skb, offset, sizeof(_optval), &_optval); if (op == NULL) return 1; @@ -220,33 +228,31 @@ static int icmpv6_push_pending_frames(struct sock *sk, struct flowi *fl, struct if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) goto out; - icmp6h = (struct icmp6hdr*) skb->h.raw; + icmp6h = icmp6_hdr(skb); memcpy(icmp6h, thdr, sizeof(struct icmp6hdr)); icmp6h->icmp6_cksum = 0; if (skb_queue_len(&sk->sk_write_queue) == 1) { - skb->csum = csum_partial((char *)icmp6h, + skb->csum = csum_partial(icmp6h, sizeof(struct icmp6hdr), skb->csum); icmp6h->icmp6_cksum = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, len, fl->proto, skb->csum); } else { - u32 tmp_csum = 0; + __wsum tmp_csum = 0; skb_queue_walk(&sk->sk_write_queue, skb) { tmp_csum = csum_add(tmp_csum, skb->csum); } - tmp_csum = csum_partial((char *)icmp6h, + tmp_csum = csum_partial(icmp6h, sizeof(struct icmp6hdr), tmp_csum); - tmp_csum = csum_ipv6_magic(&fl->fl6_src, - &fl->fl6_dst, - len, fl->proto, tmp_csum); - icmp6h->icmp6_cksum = tmp_csum; + icmp6h->icmp6_cksum = csum_ipv6_magic(&fl->fl6_src, + &fl->fl6_dst, + len, fl->proto, + tmp_csum); } - if (icmp6h->icmp6_cksum == 0) - icmp6h->icmp6_cksum = -1; ip6_push_pending_frames(sk); out: return err; @@ -255,34 +261,64 @@ out: struct icmpv6_msg { struct sk_buff *skb; int offset; + uint8_t type; }; static int icmpv6_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb) { struct icmpv6_msg *msg = (struct icmpv6_msg *) from; struct sk_buff *org_skb = msg->skb; - __u32 csum = 0; + __wsum csum = 0; csum = skb_copy_and_csum_bits(org_skb, msg->offset + offset, to, len, csum); skb->csum = csum_block_add(skb->csum, csum, odd); + if (!(msg->type & ICMPV6_INFOMSG_MASK)) + nf_ct_attach(skb, org_skb); return 0; } +#if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE) +static void mip6_addr_swap(struct sk_buff *skb) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + struct inet6_skb_parm *opt = IP6CB(skb); + struct ipv6_destopt_hao *hao; + struct in6_addr tmp; + int off; + + if (opt->dsthao) { + off = ipv6_find_tlv(skb, opt->dsthao, IPV6_TLV_HAO); + if (likely(off >= 0)) { + hao = (struct ipv6_destopt_hao *) + (skb_network_header(skb) + off); + ipv6_addr_copy(&tmp, &iph->saddr); + ipv6_addr_copy(&iph->saddr, &hao->addr); + ipv6_addr_copy(&hao->addr, &tmp); + } + } +} +#else +static inline void mip6_addr_swap(struct sk_buff *skb) {} +#endif + /* * Send an ICMP message in response to a packet in error */ -void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, +void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, struct net_device *dev) { + struct net *net = dev_net(skb->dev); struct inet6_dev *idev = NULL; - struct ipv6hdr *hdr = skb->nh.ipv6h; - struct sock *sk = icmpv6_socket->sk; - struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6hdr *hdr = ipv6_hdr(skb); + struct sock *sk; + struct ipv6_pinfo *np; struct in6_addr *saddr = NULL; struct dst_entry *dst; + struct dst_entry *dst2; struct icmp6hdr tmp_hdr; struct flowi fl; + struct flowi fl2; struct icmpv6_msg msg; int iif = 0; int addr_type = 0; @@ -290,18 +326,19 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, int hlimit; int err = 0; - if ((u8*)hdr < skb->head || (u8*)(hdr+1) > skb->tail) + if ((u8 *)hdr < skb->head || + (skb->network_header + sizeof(*hdr)) > skb->tail) return; /* - * Make sure we respect the rules + * Make sure we respect the rules * i.e. RFC 1885 2.4(e) * Rule (e.1) is enforced by not using icmpv6_send * in any code that processes icmp errors. */ addr_type = ipv6_addr_type(&hdr->daddr); - if (ipv6_chk_addr(&hdr->daddr, skb->dev, 0)) + if (ipv6_chk_addr(net, &hdr->daddr, skb->dev, 0)) saddr = &hdr->daddr; /* @@ -310,8 +347,8 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, if ((addr_type & IPV6_ADDR_MULTICAST || skb->pkt_type != PACKET_HOST)) { if (type != ICMPV6_PKT_TOOBIG && - !(type == ICMPV6_PARAMPROB && - code == ICMPV6_UNK_OPTION && + !(type == ICMPV6_PARAMPROB && + code == ICMPV6_UNK_OPTION && (opt_unrec(skb, info)))) return; @@ -328,24 +365,26 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, iif = skb->dev->ifindex; /* - * Must not send if we know that source is Anycast also. - * for now we don't know that. + * Must not send error if the source does not uniquely + * identify a single node (RFC2463 Section 2.4). + * We check unspecified / multicast addresses here, + * and anycast addresses will be checked later. */ if ((addr_type == IPV6_ADDR_ANY) || (addr_type & IPV6_ADDR_MULTICAST)) { - LIMIT_NETDEBUG( - printk(KERN_DEBUG "icmpv6_send: addr_any/mcast source\n")); + LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: addr_any/mcast source\n"); return; } - /* + /* * Never answer to a ICMP packet. */ if (is_ineligible(skb)) { - LIMIT_NETDEBUG( - printk(KERN_DEBUG "icmpv6_send: no reply to icmp error\n")); + LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: no reply to icmp error\n"); return; } + mip6_addr_swap(skb); + memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_ICMPV6; ipv6_addr_copy(&fl.fl6_dst, &hdr->saddr); @@ -354,9 +393,12 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, fl.oif = iif; fl.fl_icmp_type = type; fl.fl_icmp_code = code; + security_skb_classify_flow(skb, &fl); - if (icmpv6_xmit_lock()) + sk = icmpv6_xmit_lock(net); + if (sk == NULL) return; + np = inet6_sk(sk); if (!icmpv6_xrlim_allow(sk, type, &fl)) goto out; @@ -372,26 +414,69 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, err = ip6_dst_lookup(sk, &dst, &fl); if (err) goto out; - if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) + + /* + * We won't send icmp if the destination is known + * anycast. + */ + if (((struct rt6_info *)dst)->rt6i_flags & RTF_ANYCAST) { + LIMIT_NETDEBUG(KERN_DEBUG "icmpv6_send: acast source\n"); goto out_dst_release; + } + + /* No need to clone since we're just using its address. */ + dst2 = dst; + err = xfrm_lookup(net, &dst, &fl, sk, 0); + switch (err) { + case 0: + if (dst != dst2) + goto route_done; + break; + case -EPERM: + dst = NULL; + break; + default: + goto out; + } + + if (xfrm_decode_session_reverse(skb, &fl2, AF_INET6)) + goto relookup_failed; + + if (ip6_dst_lookup(sk, &dst2, &fl2)) + goto relookup_failed; + + err = xfrm_lookup(net, &dst2, &fl2, sk, XFRM_LOOKUP_ICMP); + switch (err) { + case 0: + dst_release(dst); + dst = dst2; + break; + case -EPERM: + goto out_dst_release; + default: +relookup_failed: + if (!dst) + goto out; + break; + } + +route_done: if (ipv6_addr_is_multicast(&fl.fl6_dst)) hlimit = np->mcast_hops; else hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); msg.skb = skb; - msg.offset = skb->nh.raw - skb->data; + msg.offset = skb_network_offset(skb); + msg.type = type; len = skb->len - msg.offset; len = min_t(unsigned int, len, IPV6_MIN_MTU - sizeof(struct ipv6hdr) -sizeof(struct icmp6hdr)); if (len < 0) { - LIMIT_NETDEBUG( - printk(KERN_DEBUG "icmp: len problem\n")); + LIMIT_NETDEBUG(KERN_DEBUG "icmp: len problem\n"); goto out_dst_release; } @@ -399,8 +484,8 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, err = ip6_append_data(sk, icmpv6_getfrag, &msg, len + sizeof(struct icmp6hdr), - sizeof(struct icmp6hdr), - hlimit, NULL, &fl, (struct rt6_info*)dst, + sizeof(struct icmp6hdr), hlimit, + np->tclass, NULL, &fl, (struct rt6_info*)dst, MSG_DONTWAIT); if (err) { ip6_flush_pending_frames(sk); @@ -408,26 +493,25 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, } err = icmpv6_push_pending_frames(sk, &fl, &tmp_hdr, len + sizeof(struct icmp6hdr)); - if (type >= ICMPV6_DEST_UNREACH && type <= ICMPV6_PARAMPROB) - ICMP6_INC_STATS_OFFSET_BH(idev, ICMP6_MIB_OUTDESTUNREACHS, type - ICMPV6_DEST_UNREACH); - ICMP6_INC_STATS_BH(idev, ICMP6_MIB_OUTMSGS); - out_put: if (likely(idev != NULL)) in6_dev_put(idev); out_dst_release: dst_release(dst); out: - icmpv6_xmit_unlock(); + icmpv6_xmit_unlock(sk); } +EXPORT_SYMBOL(icmpv6_send); + static void icmpv6_echo_reply(struct sk_buff *skb) { - struct sock *sk = icmpv6_socket->sk; + struct net *net = dev_net(skb->dev); + struct sock *sk; struct inet6_dev *idev; - struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6_pinfo *np; struct in6_addr *saddr = NULL; - struct icmp6hdr *icmph = (struct icmp6hdr *) skb->h.raw; + struct icmp6hdr *icmph = icmp6_hdr(skb); struct icmp6hdr tmp_hdr; struct flowi fl; struct icmpv6_msg msg; @@ -435,7 +519,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) int err = 0; int hlimit; - saddr = &skb->nh.ipv6h->daddr; + saddr = &ipv6_hdr(skb)->daddr; if (!ipv6_unicast_destination(skb)) saddr = NULL; @@ -445,14 +529,17 @@ static void icmpv6_echo_reply(struct sk_buff *skb) memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_ICMPV6; - ipv6_addr_copy(&fl.fl6_dst, &skb->nh.ipv6h->saddr); + ipv6_addr_copy(&fl.fl6_dst, &ipv6_hdr(skb)->saddr); if (saddr) ipv6_addr_copy(&fl.fl6_src, saddr); fl.oif = skb->dev->ifindex; fl.fl_icmp_type = ICMPV6_ECHO_REPLY; + security_skb_classify_flow(skb, &fl); - if (icmpv6_xmit_lock()) + sk = icmpv6_xmit_lock(net); + if (sk == NULL) return; + np = inet6_sk(sk); if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) fl.oif = np->mcast_oif; @@ -460,25 +547,24 @@ static void icmpv6_echo_reply(struct sk_buff *skb) err = ip6_dst_lookup(sk, &dst, &fl); if (err) goto out; - if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) - goto out_dst_release; + if ((err = xfrm_lookup(net, &dst, &fl, sk, 0)) < 0) + goto out; if (ipv6_addr_is_multicast(&fl.fl6_dst)) hlimit = np->mcast_hops; else hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); idev = in6_dev_get(skb->dev); msg.skb = skb; msg.offset = 0; + msg.type = ICMPV6_ECHO_REPLY; err = ip6_append_data(sk, icmpv6_getfrag, &msg, skb->len + sizeof(struct icmp6hdr), - sizeof(struct icmp6hdr), hlimit, NULL, &fl, + sizeof(struct icmp6hdr), hlimit, np->tclass, NULL, &fl, (struct rt6_info*)dst, MSG_DONTWAIT); if (err) { @@ -487,23 +573,17 @@ static void icmpv6_echo_reply(struct sk_buff *skb) } err = icmpv6_push_pending_frames(sk, &fl, &tmp_hdr, skb->len + sizeof(struct icmp6hdr)); - ICMP6_INC_STATS_BH(idev, ICMP6_MIB_OUTECHOREPLIES); - ICMP6_INC_STATS_BH(idev, ICMP6_MIB_OUTMSGS); - -out_put: +out_put: if (likely(idev != NULL)) in6_dev_put(idev); -out_dst_release: dst_release(dst); -out: - icmpv6_xmit_unlock(); +out: + icmpv6_xmit_unlock(sk); } -static void icmpv6_notify(struct sk_buff *skb, int type, int code, u32 info) +static void icmpv6_notify(struct sk_buff *skb, u8 type, u8 code, __be32 info) { - struct in6_addr *saddr, *daddr; - struct inet6_protocol *ipprot; - struct sock *sk; + const struct inet6_protocol *ipprot; int inner_offset; int hash; u8 nexthdr; @@ -525,9 +605,6 @@ static void icmpv6_notify(struct sk_buff *skb, int type, int code, u32 info) if (!pskb_may_pull(skb, inner_offset+8)) return; - saddr = &skb->nh.ipv6h->saddr; - daddr = &skb->nh.ipv6h->daddr; - /* BUGGG_FUTURE: we should try to parse exthdrs in this packet. Without this we will not able f.e. to make source routed pmtu discovery. @@ -543,66 +620,72 @@ static void icmpv6_notify(struct sk_buff *skb, int type, int code, u32 info) ipprot->err_handler(skb, NULL, type, code, inner_offset, info); rcu_read_unlock(); - read_lock(&raw_v6_lock); - if ((sk = sk_head(&raw_v6_htable[hash])) != NULL) { - while((sk = __raw_v6_lookup(sk, nexthdr, daddr, saddr))) { - rawv6_err(sk, skb, NULL, type, code, inner_offset, info); - sk = sk_next(sk); - } - } - read_unlock(&raw_v6_lock); + raw6_icmp_error(skb, nexthdr, type, code, inner_offset, info); } - + /* * Handle icmp messages */ -static int icmpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) +static int icmpv6_rcv(struct sk_buff *skb) { - struct sk_buff *skb = *pskb; struct net_device *dev = skb->dev; struct inet6_dev *idev = __in6_dev_get(dev); struct in6_addr *saddr, *daddr; struct ipv6hdr *orig_hdr; struct icmp6hdr *hdr; - int type; + u8 type; - ICMP6_INC_STATS_BH(idev, ICMP6_MIB_INMSGS); + if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { + struct sec_path *sp = skb_sec_path(skb); + int nh; - saddr = &skb->nh.ipv6h->saddr; - daddr = &skb->nh.ipv6h->daddr; + if (!(sp && sp->xvec[sp->len - 1]->props.flags & + XFRM_STATE_ICMP)) + goto drop_no_count; - /* Perform checksum. */ - if (skb->ip_summed == CHECKSUM_HW) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - if (csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, - skb->csum)) { - LIMIT_NETDEBUG( - printk(KERN_DEBUG "ICMPv6 hw checksum failed\n")); - skb->ip_summed = CHECKSUM_NONE; - } + if (!pskb_may_pull(skb, sizeof(*hdr) + sizeof(*orig_hdr))) + goto drop_no_count; + + nh = skb_network_offset(skb); + skb_set_network_header(skb, sizeof(*hdr)); + + if (!xfrm6_policy_check_reverse(NULL, XFRM_POLICY_IN, skb)) + goto drop_no_count; + + skb_set_network_header(skb, nh); } - if (skb->ip_summed == CHECKSUM_NONE) { - if (csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, - skb_checksum(skb, 0, skb->len, 0))) { - LIMIT_NETDEBUG( - printk(KERN_DEBUG "ICMPv6 checksum failed [%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x > %04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x]\n", - NIP6(*saddr), NIP6(*daddr))); + + ICMP6_INC_STATS_BH(dev_net(dev), idev, ICMP6_MIB_INMSGS); + + saddr = &ipv6_hdr(skb)->saddr; + daddr = &ipv6_hdr(skb)->daddr; + + /* Perform checksum. */ + switch (skb->ip_summed) { + case CHECKSUM_COMPLETE: + if (!csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, + skb->csum)) + break; + /* fall through */ + case CHECKSUM_NONE: + skb->csum = ~csum_unfold(csum_ipv6_magic(saddr, daddr, skb->len, + IPPROTO_ICMPV6, 0)); + if (__skb_checksum_complete(skb)) { + LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 checksum failed [%pI6 > %pI6]\n", + saddr, daddr); goto discard_it; } } - if (!pskb_pull(skb, sizeof(struct icmp6hdr))) + if (!pskb_pull(skb, sizeof(*hdr))) goto discard_it; - hdr = (struct icmp6hdr *) skb->h.raw; + hdr = icmp6_hdr(skb); type = hdr->icmp6_type; - if (type >= ICMPV6_DEST_UNREACH && type <= ICMPV6_PARAMPROB) - ICMP6_INC_STATS_OFFSET_BH(idev, ICMP6_MIB_INDESTUNREACHS, type - ICMPV6_DEST_UNREACH); - else if (type >= ICMPV6_ECHO_REQUEST && type <= NDISC_REDIRECT) - ICMP6_INC_STATS_OFFSET_BH(idev, ICMP6_MIB_INECHOS, type - ICMPV6_ECHO_REQUEST); + ICMP6MSGIN_INC_STATS_BH(dev_net(dev), idev, type); switch (type) { case ICMPV6_ECHO_REQUEST: @@ -621,7 +704,7 @@ static int icmpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) */ if (!pskb_may_pull(skb, sizeof(struct ipv6hdr))) goto discard_it; - hdr = (struct icmp6hdr *) skb->h.raw; + hdr = icmp6_hdr(skb); orig_hdr = (struct ipv6hdr *) (hdr + 1); rt6_pmtu_discovery(&orig_hdr->daddr, &orig_hdr->saddr, dev, ntohl(hdr->icmp6_mtu)); @@ -663,40 +746,64 @@ static int icmpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) break; default: - LIMIT_NETDEBUG( - printk(KERN_DEBUG "icmpv6: msg of unknown type\n")); + LIMIT_NETDEBUG(KERN_DEBUG "icmpv6: msg of unknown type\n"); /* informational */ if (type & ICMPV6_INFOMSG_MASK) break; - /* - * error of unknown type. - * must pass to upper level + /* + * error of unknown type. + * must pass to upper level */ icmpv6_notify(skb, type, hdr->icmp6_code, hdr->icmp6_mtu); - }; + } + kfree_skb(skb); return 0; discard_it: - ICMP6_INC_STATS_BH(idev, ICMP6_MIB_INERRORS); + ICMP6_INC_STATS_BH(dev_net(dev), idev, ICMP6_MIB_INERRORS); +drop_no_count: kfree_skb(skb); return 0; } -int __init icmpv6_init(struct net_proto_family *ops) +void icmpv6_flow_init(struct sock *sk, struct flowi *fl, + u8 type, + const struct in6_addr *saddr, + const struct in6_addr *daddr, + int oif) +{ + memset(fl, 0, sizeof(*fl)); + ipv6_addr_copy(&fl->fl6_src, saddr); + ipv6_addr_copy(&fl->fl6_dst, daddr); + fl->proto = IPPROTO_ICMPV6; + fl->fl_icmp_type = type; + fl->fl_icmp_code = 0; + fl->oif = oif; + security_sk_classify_flow(sk, fl); +} + +/* + * Special lock-class for __icmpv6_sk: + */ +static struct lock_class_key icmpv6_socket_sk_dst_lock_key; + +static int __net_init icmpv6_sk_init(struct net *net) { struct sock *sk; int err, i, j; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; + net->ipv6.icmp_sk = + kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL); + if (net->ipv6.icmp_sk == NULL) + return -ENOMEM; - err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, - &per_cpu(__icmpv6_socket, i)); + for_each_possible_cpu(i) { + err = inet_ctl_sock_create(&sk, PF_INET6, + SOCK_RAW, IPPROTO_ICMPV6, net); if (err < 0) { printk(KERN_ERR "Failed to initialize the ICMP6 control socket " @@ -705,50 +812,74 @@ int __init icmpv6_init(struct net_proto_family *ops) goto fail; } - sk = per_cpu(__icmpv6_socket, i)->sk; - sk->sk_allocation = GFP_ATOMIC; + net->ipv6.icmp_sk[i] = sk; + + /* + * Split off their lock-class, because sk->sk_dst_lock + * gets used from softirqs, which is safe for + * __icmpv6_sk (because those never get directly used + * via userspace syscalls), but unsafe for normal sockets. + */ + lockdep_set_class(&sk->sk_dst_lock, + &icmpv6_socket_sk_dst_lock_key); /* Enough space for 2 64K ICMP packets, including * sk_buff struct overhead. */ sk->sk_sndbuf = (2 * ((64 * 1024) + sizeof(struct sk_buff))); - - sk->sk_prot->unhash(sk); } + return 0; + fail: + for (j = 0; j < i; j++) + inet_ctl_sock_destroy(net->ipv6.icmp_sk[j]); + kfree(net->ipv6.icmp_sk); + return err; +} - if (inet6_add_protocol(&icmpv6_protocol, IPPROTO_ICMPV6) < 0) { - printk(KERN_ERR "Failed to register ICMP6 protocol\n"); - err = -EAGAIN; - goto fail; +static void __net_exit icmpv6_sk_exit(struct net *net) +{ + int i; + + for_each_possible_cpu(i) { + inet_ctl_sock_destroy(net->ipv6.icmp_sk[i]); } + kfree(net->ipv6.icmp_sk); +} - return 0; +static struct pernet_operations icmpv6_sk_ops = { + .init = icmpv6_sk_init, + .exit = icmpv6_sk_exit, +}; - fail: - for (j = 0; j < i; j++) { - if (!cpu_possible(j)) - continue; - sock_release(per_cpu(__icmpv6_socket, j)); - } +int __init icmpv6_init(void) +{ + int err; + + err = register_pernet_subsys(&icmpv6_sk_ops); + if (err < 0) + return err; + + err = -EAGAIN; + if (inet6_add_protocol(&icmpv6_protocol, IPPROTO_ICMPV6) < 0) + goto fail; + return 0; +fail: + printk(KERN_ERR "Failed to register ICMP6 protocol\n"); + unregister_pernet_subsys(&icmpv6_sk_ops); return err; } void icmpv6_cleanup(void) { - int i; - - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; - sock_release(per_cpu(__icmpv6_socket, i)); - } + unregister_pernet_subsys(&icmpv6_sk_ops); inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6); } -static struct icmp6_err { + +static const struct icmp6_err { int err; int fatal; } tab_unreach[] = { @@ -774,7 +905,7 @@ static struct icmp6_err { }, }; -int icmpv6_err_convert(int type, int code, int *err) +int icmpv6_err_convert(u8 type, u8 code, int *err) { int fatal = 0; @@ -792,7 +923,7 @@ int icmpv6_err_convert(int type, int code, int *err) case ICMPV6_PKT_TOOBIG: *err = EMSGSIZE; break; - + case ICMPV6_PARAMPROB: *err = EPROTO; fatal = 1; @@ -801,22 +932,37 @@ int icmpv6_err_convert(int type, int code, int *err) case ICMPV6_TIME_EXCEED: *err = EHOSTUNREACH; break; - }; + } return fatal; } +EXPORT_SYMBOL(icmpv6_err_convert); + #ifdef CONFIG_SYSCTL -ctl_table ipv6_icmp_table[] = { +ctl_table ipv6_icmp_table_template[] = { { - .ctl_name = NET_IPV6_ICMP_RATELIMIT, .procname = "ratelimit", - .data = &sysctl_icmpv6_time, + .data = &init_net.ipv6.sysctl.icmpv6_time, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = &proc_dointvec + .proc_handler = proc_dointvec_ms_jiffies, }, - { .ctl_name = 0 }, + { }, }; + +struct ctl_table *ipv6_icmp_sysctl_init(struct net *net) +{ + struct ctl_table *table; + + table = kmemdup(ipv6_icmp_table_template, + sizeof(ipv6_icmp_table_template), + GFP_KERNEL); + + if (table) + table[0].data = &net->ipv6.sysctl.icmpv6_time; + + return table; +} #endif