diff options
Diffstat (limited to 'net/ipv6')
-rw-r--r-- | net/ipv6/addrconf.c | 457 | ||||
-rw-r--r-- | net/ipv6/af_inet6.c | 8 | ||||
-rw-r--r-- | net/ipv6/icmp.c | 30 | ||||
-rw-r--r-- | net/ipv6/ip6_fib.c | 54 | ||||
-rw-r--r-- | net/ipv6/ip6_input.c | 5 | ||||
-rw-r--r-- | net/ipv6/ip6_output.c | 92 | ||||
-rw-r--r-- | net/ipv6/ip6_tunnel.c | 7 | ||||
-rw-r--r-- | net/ipv6/ipcomp6.c | 3 | ||||
-rw-r--r-- | net/ipv6/ipv6_sockglue.c | 5 | ||||
-rw-r--r-- | net/ipv6/ipv6_syms.c | 2 | ||||
-rw-r--r-- | net/ipv6/mcast.c | 25 | ||||
-rw-r--r-- | net/ipv6/netfilter/Kconfig | 36 | ||||
-rw-r--r-- | net/ipv6/netfilter/Makefile | 6 | ||||
-rw-r--r-- | net/ipv6/netfilter/ip6_tables.c | 298 | ||||
-rw-r--r-- | net/ipv6/netfilter/ip6t_MARK.c | 8 | ||||
-rw-r--r-- | net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 556 | ||||
-rw-r--r-- | net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 272 | ||||
-rw-r--r-- | net/ipv6/netfilter/nf_conntrack_reasm.c | 897 | ||||
-rw-r--r-- | net/ipv6/proc.c | 4 | ||||
-rw-r--r-- | net/ipv6/raw.c | 46 | ||||
-rw-r--r-- | net/ipv6/reassembly.c | 41 | ||||
-rw-r--r-- | net/ipv6/route.c | 10 | ||||
-rw-r--r-- | net/ipv6/tcp_ipv6.c | 35 | ||||
-rw-r--r-- | net/ipv6/udp.c | 25 |
24 files changed, 2430 insertions, 492 deletions
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index a970b47..a16064b 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -35,6 +35,9 @@ * YOSHIFUJI Hideaki @USAGI : ARCnet support * YOSHIFUJI Hideaki @USAGI : convert /proc/net/if_inet6 to * seq_file. + * YOSHIFUJI Hideaki @USAGI : improved source address + * selection; consider scope, + * status etc. */ #include <linux/config.h> @@ -75,7 +78,7 @@ #ifdef CONFIG_IPV6_PRIVACY #include <linux/random.h> #include <linux/crypto.h> -#include <asm/scatterlist.h> +#include <linux/scatterlist.h> #endif #include <asm/uaccess.h> @@ -193,46 +196,51 @@ const struct in6_addr in6addr_any = IN6ADDR_ANY_INIT; #endif const struct in6_addr in6addr_loopback = IN6ADDR_LOOPBACK_INIT; -int ipv6_addr_type(const struct in6_addr *addr) +#define IPV6_ADDR_SCOPE_TYPE(scope) ((scope) << 16) + +static inline unsigned ipv6_addr_scope2type(unsigned scope) +{ + switch(scope) { + case IPV6_ADDR_SCOPE_NODELOCAL: + return (IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_NODELOCAL) | + IPV6_ADDR_LOOPBACK); + case IPV6_ADDR_SCOPE_LINKLOCAL: + return (IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_LINKLOCAL) | + IPV6_ADDR_LINKLOCAL); + case IPV6_ADDR_SCOPE_SITELOCAL: + return (IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_SITELOCAL) | + IPV6_ADDR_SITELOCAL); + } + return IPV6_ADDR_SCOPE_TYPE(scope); +} + +int __ipv6_addr_type(const struct in6_addr *addr) { - int type; u32 st; st = addr->s6_addr32[0]; - if ((st & htonl(0xFF000000)) == htonl(0xFF000000)) { - type = IPV6_ADDR_MULTICAST; - - switch((st & htonl(0x00FF0000))) { - case __constant_htonl(0x00010000): - type |= IPV6_ADDR_LOOPBACK; - break; - - case __constant_htonl(0x00020000): - type |= IPV6_ADDR_LINKLOCAL; - break; - - case __constant_htonl(0x00050000): - type |= IPV6_ADDR_SITELOCAL; - break; - }; - return type; - } - - type = IPV6_ADDR_UNICAST; - /* Consider all addresses with the first three bits different of - 000 and 111 as finished. + 000 and 111 as unicasts. */ if ((st & htonl(0xE0000000)) != htonl(0x00000000) && (st & htonl(0xE0000000)) != htonl(0xE0000000)) - return type; - - if ((st & htonl(0xFFC00000)) == htonl(0xFE800000)) - return (IPV6_ADDR_LINKLOCAL | type); + return (IPV6_ADDR_UNICAST | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_GLOBAL)); + if ((st & htonl(0xFF000000)) == htonl(0xFF000000)) { + /* multicast */ + /* addr-select 3.1 */ + return (IPV6_ADDR_MULTICAST | + ipv6_addr_scope2type(IPV6_ADDR_MC_SCOPE(addr))); + } + + if ((st & htonl(0xFFC00000)) == htonl(0xFE800000)) + return (IPV6_ADDR_LINKLOCAL | IPV6_ADDR_UNICAST | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_LINKLOCAL)); /* addr-select 3.1 */ if ((st & htonl(0xFFC00000)) == htonl(0xFEC00000)) - return (IPV6_ADDR_SITELOCAL | type); + return (IPV6_ADDR_SITELOCAL | IPV6_ADDR_UNICAST | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_SITELOCAL)); /* addr-select 3.1 */ if ((addr->s6_addr32[0] | addr->s6_addr32[1]) == 0) { if (addr->s6_addr32[2] == 0) { @@ -240,24 +248,20 @@ int ipv6_addr_type(const struct in6_addr *addr) return IPV6_ADDR_ANY; if (addr->s6_addr32[3] == htonl(0x00000001)) - return (IPV6_ADDR_LOOPBACK | type); + return (IPV6_ADDR_LOOPBACK | IPV6_ADDR_UNICAST | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_LINKLOCAL)); /* addr-select 3.4 */ - return (IPV6_ADDR_COMPATv4 | type); + return (IPV6_ADDR_COMPATv4 | IPV6_ADDR_UNICAST | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_GLOBAL)); /* addr-select 3.3 */ } if (addr->s6_addr32[2] == htonl(0x0000ffff)) - return IPV6_ADDR_MAPPED; + return (IPV6_ADDR_MAPPED | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_GLOBAL)); /* addr-select 3.3 */ } - st &= htonl(0xFF000000); - if (st == 0) - return IPV6_ADDR_RESERVED; - st &= htonl(0xFE000000); - if (st == htonl(0x02000000)) - return IPV6_ADDR_RESERVED; /* for NSAP */ - if (st == htonl(0x04000000)) - return IPV6_ADDR_RESERVED; /* for IPX */ - return type; + return (IPV6_ADDR_RESERVED | + IPV6_ADDR_SCOPE_TYPE(IPV6_ADDR_SCOPE_GLOBAL)); /* addr-select 3.4 */ } static void addrconf_del_timer(struct inet6_ifaddr *ifp) @@ -805,138 +809,276 @@ out: #endif /* - * Choose an appropriate source address - * should do: - * i) get an address with an appropriate scope - * ii) see if there is a specific route for the destination and use - * an address of the attached interface - * iii) don't use deprecated addresses + * Choose an appropriate source address (RFC3484) */ -static int inline ipv6_saddr_pref(const struct inet6_ifaddr *ifp, u8 invpref) +struct ipv6_saddr_score { + int addr_type; + unsigned int attrs; + int matchlen; + unsigned int scope; + unsigned int rule; +}; + +#define IPV6_SADDR_SCORE_LOCAL 0x0001 +#define IPV6_SADDR_SCORE_PREFERRED 0x0004 +#define IPV6_SADDR_SCORE_HOA 0x0008 +#define IPV6_SADDR_SCORE_OIF 0x0010 +#define IPV6_SADDR_SCORE_LABEL 0x0020 +#define IPV6_SADDR_SCORE_PRIVACY 0x0040 + +static int inline ipv6_saddr_preferred(int type) { - int pref; - pref = ifp->flags&IFA_F_DEPRECATED ? 0 : 2; -#ifdef CONFIG_IPV6_PRIVACY - pref |= (ifp->flags^invpref)&IFA_F_TEMPORARY ? 0 : 1; -#endif - return pref; + if (type & (IPV6_ADDR_MAPPED|IPV6_ADDR_COMPATv4| + IPV6_ADDR_LOOPBACK|IPV6_ADDR_RESERVED)) + return 1; + return 0; } -#ifdef CONFIG_IPV6_PRIVACY -#define IPV6_GET_SADDR_MAXSCORE(score) ((score) == 3) -#else -#define IPV6_GET_SADDR_MAXSCORE(score) (score) -#endif +/* static matching label */ +static int inline ipv6_saddr_label(const struct in6_addr *addr, int type) +{ + /* + * prefix (longest match) label + * ----------------------------- + * ::1/128 0 + * ::/0 1 + * 2002::/16 2 + * ::/96 3 + * ::ffff:0:0/96 4 + */ + if (type & IPV6_ADDR_LOOPBACK) + return 0; + else if (type & IPV6_ADDR_COMPATv4) + return 3; + else if (type & IPV6_ADDR_MAPPED) + return 4; + else if (addr->s6_addr16[0] == htons(0x2002)) + return 2; + return 1; +} -int ipv6_dev_get_saddr(struct net_device *dev, +int ipv6_dev_get_saddr(struct net_device *daddr_dev, struct in6_addr *daddr, struct in6_addr *saddr) { - struct inet6_ifaddr *ifp = NULL; - struct inet6_ifaddr *match = NULL; - struct inet6_dev *idev; - int scope; - int err; - int hiscore = -1, score; + struct ipv6_saddr_score hiscore; + struct inet6_ifaddr *ifa_result = NULL; + int daddr_type = __ipv6_addr_type(daddr); + int daddr_scope = __ipv6_addr_src_scope(daddr_type); + u32 daddr_label = ipv6_saddr_label(daddr, daddr_type); + struct net_device *dev; - scope = ipv6_addr_scope(daddr); + memset(&hiscore, 0, sizeof(hiscore)); - /* - * known dev - * search dev and walk through dev addresses - */ + read_lock(&dev_base_lock); + read_lock(&addrconf_lock); - if (dev) { - if (dev->flags & IFF_LOOPBACK) - scope = IFA_HOST; + for (dev = dev_base; dev; dev=dev->next) { + struct inet6_dev *idev; + struct inet6_ifaddr *ifa; + + /* Rule 0: Candidate Source Address (section 4) + * - multicast and link-local destination address, + * the set of candidate source address MUST only + * include addresses assigned to interfaces + * belonging to the same link as the outgoing + * interface. + * (- For site-local destination addresses, the + * set of candidate source addresses MUST only + * include addresses assigned to interfaces + * belonging to the same site as the outgoing + * interface.) + */ + if ((daddr_type & IPV6_ADDR_MULTICAST || + daddr_scope <= IPV6_ADDR_SCOPE_LINKLOCAL) && + daddr_dev && dev != daddr_dev) + continue; - read_lock(&addrconf_lock); idev = __in6_dev_get(dev); - if (idev) { - read_lock_bh(&idev->lock); - for (ifp=idev->addr_list; ifp; ifp=ifp->if_next) { - if (ifp->scope == scope) { - if (ifp->flags&IFA_F_TENTATIVE) - continue; -#ifdef CONFIG_IPV6_PRIVACY - score = ipv6_saddr_pref(ifp, idev->cnf.use_tempaddr > 1 ? IFA_F_TEMPORARY : 0); -#else - score = ipv6_saddr_pref(ifp, 0); -#endif - if (score <= hiscore) - continue; + if (!idev) + continue; - if (match) - in6_ifa_put(match); - match = ifp; - hiscore = score; - in6_ifa_hold(ifp); + read_lock_bh(&idev->lock); + for (ifa = idev->addr_list; ifa; ifa = ifa->if_next) { + struct ipv6_saddr_score score; - if (IPV6_GET_SADDR_MAXSCORE(score)) { - read_unlock_bh(&idev->lock); - read_unlock(&addrconf_lock); - goto out; - } + score.addr_type = __ipv6_addr_type(&ifa->addr); + + /* Rule 0: Candidate Source Address (section 4) + * - In any case, anycast addresses, multicast + * addresses, and the unspecified address MUST + * NOT be included in a candidate set. + */ + if (unlikely(score.addr_type == IPV6_ADDR_ANY || + score.addr_type & IPV6_ADDR_MULTICAST)) { + LIMIT_NETDEBUG(KERN_DEBUG + "ADDRCONF: unspecified / multicast address" + "assigned as unicast address on %s", + dev->name); + continue; + } + + score.attrs = 0; + score.matchlen = 0; + score.scope = 0; + score.rule = 0; + + if (ifa_result == NULL) { + /* record it if the first available entry */ + goto record_it; + } + + /* Rule 1: Prefer same address */ + if (hiscore.rule < 1) { + if (ipv6_addr_equal(&ifa_result->addr, daddr)) + hiscore.attrs |= IPV6_SADDR_SCORE_LOCAL; + hiscore.rule++; + } + if (ipv6_addr_equal(&ifa->addr, daddr)) { + score.attrs |= IPV6_SADDR_SCORE_LOCAL; + if (!(hiscore.attrs & IPV6_SADDR_SCORE_LOCAL)) { + score.rule = 1; + goto record_it; } + } else { + if (hiscore.attrs & IPV6_SADDR_SCORE_LOCAL) + continue; } - read_unlock_bh(&idev->lock); - } - read_unlock(&addrconf_lock); - } - if (scope == IFA_LINK) - goto out; + /* Rule 2: Prefer appropriate scope */ + if (hiscore.rule < 2) { + hiscore.scope = __ipv6_addr_src_scope(hiscore.addr_type); + hiscore.rule++; + } + score.scope = __ipv6_addr_src_scope(score.addr_type); + if (hiscore.scope < score.scope) { + if (hiscore.scope < daddr_scope) { + score.rule = 2; + goto record_it; + } else + continue; + } else if (score.scope < hiscore.scope) { + if (score.scope < daddr_scope) + continue; + else { + score.rule = 2; + goto record_it; + } + } - /* - * dev == NULL or search failed for specified dev - */ + /* Rule 3: Avoid deprecated address */ + if (hiscore.rule < 3) { + if (ipv6_saddr_preferred(hiscore.addr_type) || + !(ifa_result->flags & IFA_F_DEPRECATED)) + hiscore.attrs |= IPV6_SADDR_SCORE_PREFERRED; + hiscore.rule++; + } + if (ipv6_saddr_preferred(score.addr_type) || + !(ifa->flags & IFA_F_DEPRECATED)) { + score.attrs |= IPV6_SADDR_SCORE_PREFERRED; + if (!(hiscore.attrs & IPV6_SADDR_SCORE_PREFERRED)) { + score.rule = 3; + goto record_it; + } + } else { + if (hiscore.attrs & IPV6_SADDR_SCORE_PREFERRED) + continue; + } - read_lock(&dev_base_lock); - read_lock(&addrconf_lock); - for (dev = dev_base; dev; dev=dev->next) { - idev = __in6_dev_get(dev); - if (idev) { - read_lock_bh(&idev->lock); - for (ifp=idev->addr_list; ifp; ifp=ifp->if_next) { - if (ifp->scope == scope) { - if (ifp->flags&IFA_F_TENTATIVE) - continue; -#ifdef CONFIG_IPV6_PRIVACY - score = ipv6_saddr_pref(ifp, idev->cnf.use_tempaddr > 1 ? IFA_F_TEMPORARY : 0); -#else - score = ipv6_saddr_pref(ifp, 0); -#endif - if (score <= hiscore) - continue; + /* Rule 4: Prefer home address -- not implemented yet */ - if (match) - in6_ifa_put(match); - match = ifp; - hiscore = score; - in6_ifa_hold(ifp); + /* Rule 5: Prefer outgoing interface */ + if (hiscore.rule < 5) { + if (daddr_dev == NULL || + daddr_dev == ifa_result->idev->dev) + hiscore.attrs |= IPV6_SADDR_SCORE_OIF; + hiscore.rule++; + } + if (daddr_dev == NULL || + daddr_dev == ifa->idev->dev) { + score.attrs |= IPV6_SADDR_SCORE_OIF; + if (!(hiscore.attrs & IPV6_SADDR_SCORE_OIF)) { + score.rule = 5; + goto record_it; + } + } else { + if (hiscore.attrs & IPV6_SADDR_SCORE_OIF) + continue; + } - if (IPV6_GET_SADDR_MAXSCORE(score)) { - read_unlock_bh(&idev->lock); - goto out_unlock_base; - } + /* Rule 6: Prefer matching label */ + if (hiscore.rule < 6) { + if (ipv6_saddr_label(&ifa_result->addr, hiscore.addr_type) == daddr_label) + hiscore.attrs |= IPV6_SADDR_SCORE_LABEL; + hiscore.rule++; + } + if (ipv6_saddr_label(&ifa->addr, score.addr_type) == daddr_label) { + score.attrs |= IPV6_SADDR_SCORE_LABEL; + if (!(hiscore.attrs & IPV6_SADDR_SCORE_LABEL)) { + score.rule = 6; + goto record_it; } + } else { + if (hiscore.attrs & IPV6_SADDR_SCORE_LABEL) + continue; } - read_unlock_bh(&idev->lock); + +#ifdef CONFIG_IPV6_PRIVACY + /* Rule 7: Prefer public address + * Note: prefer temprary address if use_tempaddr >= 2 + */ + if (hiscore.rule < 7) { + if ((!(ifa_result->flags & IFA_F_TEMPORARY)) ^ + (ifa_result->idev->cnf.use_tempaddr >= 2)) + hiscore.attrs |= IPV6_SADDR_SCORE_PRIVACY; + hiscore.rule++; + } + if ((!(ifa->flags & IFA_F_TEMPORARY)) ^ + (ifa->idev->cnf.use_tempaddr >= 2)) { + score.attrs |= IPV6_SADDR_SCORE_PRIVACY; + if (!(hiscore.attrs & IPV6_SADDR_SCORE_PRIVACY)) { + score.rule = 7; + goto record_it; + } + } else { + if (hiscore.attrs & IPV6_SADDR_SCORE_PRIVACY) + continue; + } +#endif + /* Rule 8: Use longest matching prefix */ + if (hiscore.rule < 8) { + hiscore.matchlen = ipv6_addr_diff(&ifa_result->addr, daddr); + hiscore.rule++; + } + score.matchlen = ipv6_addr_diff(&ifa->addr, daddr); + if (score.matchlen > hiscore.matchlen) { + score.rule = 8; + goto record_it; + } +#if 0 + else if (score.matchlen < hiscore.matchlen) + continue; +#endif + + /* Final Rule: choose first available one */ + continue; +record_it: + if (ifa_result) + in6_ifa_put(ifa_result); + in6_ifa_hold(ifa); + ifa_result = ifa; + hiscore = score; } + read_unlock_bh(&idev->lock); } - -out_unlock_base: read_unlock(&addrconf_lock); read_unlock(&dev_base_lock); -out: - err = -EADDRNOTAVAIL; - if (match) { - ipv6_addr_copy(saddr, &match->addr); - err = 0; - in6_ifa_put(match); - } - - return err; + if (!ifa_result) + return -EADDRNOTAVAIL; + + ipv6_addr_copy(saddr, &ifa_result->addr); + in6_ifa_put(ifa_result); + return 0; } @@ -1217,12 +1359,8 @@ static int __ipv6_regen_rndid(struct inet6_dev *idev) struct net_device *dev; struct scatterlist sg[2]; - sg[0].page = virt_to_page(idev->entropy); - sg[0].offset = offset_in_page(idev->entropy); - sg[0].length = 8; - sg[1].page = virt_to_page(idev->work_eui64); - sg[1].offset = offset_in_page(idev->work_eui64); - sg[1].length = 8; + sg_set_buf(&sg[0], idev->entropy, 8); + sg_set_buf(&sg[1], idev->work_eui64, 8); dev = idev->dev; @@ -2167,7 +2305,7 @@ static int addrconf_ifdown(struct net_device *dev, int how) /* Step 5: netlink notification of this interface */ idev->tstamp = jiffies; - inet6_ifinfo_notify(RTM_NEWLINK, idev); + inet6_ifinfo_notify(RTM_DELLINK, idev); /* Shot the device (if unregistered) */ @@ -2489,7 +2627,7 @@ static void addrconf_verify(unsigned long foo) for (i=0; i < IN6_ADDR_HSIZE; i++) { restart: - write_lock(&addrconf_hash_lock); + read_lock(&addrconf_hash_lock); for (ifp=inet6_addr_lst[i]; ifp; ifp=ifp->lst_next) { unsigned long age; #ifdef CONFIG_IPV6_PRIVACY @@ -2511,7 +2649,7 @@ restart: if (age >= ifp->valid_lft) { spin_unlock(&ifp->lock); in6_ifa_hold(ifp); - write_unlock(&addrconf_hash_lock); + read_unlock(&addrconf_hash_lock); ipv6_del_addr(ifp); goto restart; } else if (age >= ifp->prefered_lft) { @@ -2530,7 +2668,7 @@ restart: if (deprecate) { in6_ifa_hold(ifp); - write_unlock(&addrconf_hash_lock); + read_unlock(&addrconf_hash_lock); ipv6_ifa_notify(0, ifp); in6_ifa_put(ifp); @@ -2548,7 +2686,7 @@ restart: in6_ifa_hold(ifp); in6_ifa_hold(ifpub); spin_unlock(&ifp->lock); - write_unlock(&addrconf_hash_lock); + read_unlock(&addrconf_hash_lock); ipv6_create_tempaddr(ifpub, ifp); in6_ifa_put(ifpub); in6_ifa_put(ifp); @@ -2565,7 +2703,7 @@ restart: spin_unlock(&ifp->lock); } } - write_unlock(&addrconf_hash_lock); + read_unlock(&addrconf_hash_lock); } addr_chk_timer.expires = time_before(next, jiffies + HZ) ? jiffies + HZ : next; @@ -2954,8 +3092,7 @@ static int inet6_fill_ifinfo(struct sk_buff *skb, struct inet6_dev *idev, nlmsg_failure: rtattr_failure: - if (array) - kfree(array); + kfree(array); skb_trim(skb, b - skb->data); return -1; } diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 4f8795a..c63b8ce 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -699,12 +699,14 @@ static int __init inet6_init(void) /* Register the family here so that the init calls below will * be able to create sockets. (?? is this dangerous ??) */ - (void) sock_register(&inet6_family_ops); + err = sock_register(&inet6_family_ops); + if (err) + goto out_unregister_raw_proto; /* Initialise ipv6 mibs */ err = init_ipv6_mibs(); if (err) - goto out_unregister_raw_proto; + goto out_unregister_sock; /* * ipngwg API draft makes clear that the correct semantics @@ -796,6 +798,8 @@ icmp_fail: ipv6_sysctl_unregister(); #endif cleanup_ipv6_mibs(); +out_unregister_sock: + sock_unregister(PF_INET6); out_unregister_raw_proto: proto_unregister(&rawv6_prot); out_unregister_udp_proto: diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index b7185fb..1bdf0fb 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -585,17 +585,16 @@ static int icmpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) daddr = &skb->nh.ipv6h->daddr; /* 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(KERN_DEBUG "ICMPv6 hw checksum failed\n"); - skb->ip_summed = CHECKSUM_NONE; - } - } - if (skb->ip_summed == CHECKSUM_NONE) { - if (csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, - skb_checksum(skb, 0, skb->len, 0))) { + switch (skb->ip_summed) { + case CHECKSUM_HW: + if (!csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, + skb->csum)) + break; + /* fall through */ + case CHECKSUM_NONE: + skb->csum = ~csum_ipv6_magic(saddr, daddr, skb->len, + IPPROTO_ICMPV6, 0); + if (__skb_checksum_complete(skb)) { LIMIT_NETDEBUG(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)); goto discard_it; @@ -700,10 +699,7 @@ int __init icmpv6_init(struct net_proto_family *ops) struct sock *sk; int err, i, j; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; - + for_each_cpu(i) { err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &per_cpu(__icmpv6_socket, i)); if (err < 0) { @@ -749,9 +745,7 @@ void icmpv6_cleanup(void) { int i; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; + for_each_cpu(i) { sock_release(per_cpu(__icmpv6_socket, i)); } inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 4fcc5a7..1bf6d9a 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -127,56 +127,6 @@ static __inline__ int addr_bit_set(void *token, int fn_bit) return htonl(1 << ((~fn_bit)&0x1F)) & addr[fn_bit>>5]; } -/* - * find the first different bit between two addresses - * length of address must be a multiple of 32bits - */ - -static __inline__ int addr_diff(void *token1, void *token2, int addrlen) -{ - __u32 *a1 = token1; - __u32 *a2 = token2; - int i; - - addrlen >>= 2; - - for (i = 0; i < addrlen; i++) { - __u32 xb; - - xb = a1[i] ^ a2[i]; - - if (xb) { - int j = 31; - - xb = ntohl(xb); - - while ((xb & (1 << j)) == 0) - j--; - - return (i * 32 + 31 - j); - } - } - - /* - * we should *never* get to this point since that - * would mean the addrs are equal - * - * However, we do get to it 8) And exacly, when - * addresses are equal 8) - * - * ip route add 1111::/128 via ... - * ip route add 1111::/64 via ... - * and we are here. - * - * Ideally, this function should stop comparison - * at prefix length. It does not, but it is still OK, - * if returned value is greater than prefix length. - * --ANK (980803) - */ - - return addrlen<<5; -} - static __inline__ struct fib6_node * node_alloc(void) { struct fib6_node *fn; @@ -296,11 +246,11 @@ insert_above: /* find 1st bit in difference between the 2 addrs. - See comment in addr_diff: bit may be an invalid value, + See comment in __ipv6_addr_diff: bit may be an invalid value, but if it is >= plen, the value is ignored in any case. */ - bit = addr_diff(addr, &key->addr, addrlen); + bit = __ipv6_addr_diff(addr, &key->addr, addrlen); /* * (intermediate)[in] diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 6e34804..a6026d2 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -176,6 +176,11 @@ resubmit: if (ipprot->flags & INET6_PROTO_FINAL) { struct ipv6hdr *hdr; + /* Free reference early: we don't need it any more, + and it may hold ip_conntrack module loaded + indefinitely. */ + nf_reset(skb); + skb_postpull_rcsum(skb, skb->nh.raw, skb->h.raw - skb->nh.raw); hdr = skb->nh.ipv6h; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 563b442..c1fa693 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -147,7 +147,8 @@ static int ip6_output2(struct sk_buff *skb) int ip6_output(struct sk_buff *skb) { - if (skb->len > dst_mtu(skb->dst) || dst_allfrag(skb->dst)) + if ((skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->ufo_size) || + dst_allfrag(skb->dst)) return ip6_fragment(skb, ip6_output2); else return ip6_output2(skb); @@ -440,9 +441,15 @@ static void ip6_copy_metadata(struct sk_buff *to, struct sk_buff *from) #ifdef CONFIG_NETFILTER to->nfmark = from->nfmark; /* Connection association is same as pre-frag packet */ + nf_conntrack_put(to->nfct); to->nfct = from->nfct; nf_conntrack_get(to->nfct); to->nfctinfo = from->nfctinfo; +#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) + nf_conntrack_put_reasm(to->nfct_reasm); + to->nfct_reasm = from->nfct_reasm; + nf_conntrack_get_reasm(to->nfct_reasm); +#endif #ifdef CONFIG_BRIDGE_NETFILTER nf_bridge_put(to->nf_bridge); to->nf_bridge = from->nf_bridge; @@ -586,8 +593,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)) skb->next = NULL; } - if (tmp_hdr) - kfree(tmp_hdr); + kfree(tmp_hdr); if (err == 0) { IP6_INC_STATS(IPSTATS_MIB_FRAGOKS); @@ -768,6 +774,65 @@ out_err_release: *dst = NULL; return err; } +inline int ip6_ufo_append_data(struct sock *sk, + int getfrag(void *from, char *to, int offset, int len, + int odd, struct sk_buff *skb), + void *from, int length, int hh_len, int fragheaderlen, + int transhdrlen, int mtu,unsigned int flags) + +{ + struct sk_buff *skb; + int err; + + /* There is support for UDP large send offload by network + * device, so create one single skb packet containing complete + * udp datagram + */ + if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL) { + skb = sock_alloc_send_skb(sk, + hh_len + fragheaderlen + transhdrlen + 20, + (flags & MSG_DONTWAIT), &err); + if (skb == NULL) + return -ENOMEM; + + /* reserve space for Hardware header */ + skb_reserve(skb, hh_len); + + /* create space for UDP/IP header */ + skb_put(skb,fragheaderlen + transhdrlen); + + /* initialize network header pointer */ + skb->nh.raw = skb->data; + + /* initialize protocol header pointer */ + skb->h.raw = skb->data + fragheaderlen; + + skb->ip_summed = CHECKSUM_HW; + skb->csum = 0; + sk->sk_sndmsg_off = 0; + } + + err = skb_append_datato_frags(sk,skb, getfrag, from, + (length - transhdrlen)); + if (!err) { + struct frag_hdr fhdr; + + /* specify the length of each IP datagram fragment*/ + skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen) - + sizeof(struct frag_hdr); + ipv6_select_ident(skb, &fhdr); + skb_shinfo(skb)->ip6_frag_id = fhdr.identification; + __skb_queue_tail(&sk->sk_write_queue, skb); + + return 0; + } + /* There is not enough support do UPD LSO, + * so follow normal path + */ + kfree_skb(skb); + + return err; +} int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb), @@ -860,6 +925,15 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, */ inet->cork.length += length; + if (((length > mtu) && (sk->sk_protocol == IPPROTO_UDP)) && + (rt->u.dst.dev->features & NETIF_F_UFO)) { + + if(ip6_ufo_append_data(sk, getfrag, from, length, hh_len, + fragheaderlen, transhdrlen, mtu, flags)) + goto error; + + return 0; + } if ((skb = skb_peek_tail(&sk->sk_write_queue)) == NULL) goto alloc_new_skb; @@ -1117,10 +1191,8 @@ int ip6_push_pending_frames(struct sock *sk) out: inet->cork.flags &= ~IPCORK_OPT; - if (np->cork.opt) { - kfree(np->cork.opt); - np->cork.opt = NULL; - } + kfree(np->cork.opt); + np->cork.opt = NULL; if (np->cork.rt) { dst_release(&np->cork.rt->u.dst); np->cork.rt = NULL; @@ -1145,10 +1217,8 @@ void ip6_flush_pending_frames(struct sock *sk) inet->cork.flags &= ~IPCORK_OPT; - if (np->cork.opt) { - kfree(np->cork.opt); - np->cork.opt = NULL; - } + kfree(np->cork.opt); + np->cork.opt = NULL; if (np->cork.rt) { dst_release(&np->cork.rt->u.dst); np->cork.rt = NULL; diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index cf94372..e315d0f 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -525,6 +525,7 @@ ip6ip6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) if ((t = ip6ip6_tnl_lookup(&ipv6h->saddr, &ipv6h->daddr)) != NULL) { if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) { + read_unlock(&ip6ip6_lock); kfree_skb(skb); return 0; } @@ -756,8 +757,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev) } ip6_tnl_dst_store(t, dst); - if (opt) - kfree(opt); + kfree(opt); t->recursion--; return 0; @@ -766,8 +766,7 @@ tx_err_link_failure: dst_link_failure(skb); tx_err_dst_release: dst_release(dst); - if (opt) - kfree(opt); + kfree(opt); tx_err: stats->tx_errors++; stats->tx_dropped++; diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 85bfbc6..55917fb 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -130,8 +130,7 @@ static int ipcomp6_input(struct xfrm_state *x, struct xfrm_decap_state *decap, s out_put_cpu: put_cpu(); out: - if (tmp_hdr) - kfree(tmp_hdr); + kfree(tmp_hdr); if (err) goto error_out; return nexthdr; diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 8567873..25757ad 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -80,8 +80,7 @@ int ip6_ra_control(struct sock *sk, int sel, void (*destructor)(struct sock *)) if (ra->sk == sk) { if (sel>=0) { write_unlock_bh(&ip6_ra_lock); - if (new_ra) - kfree(new_ra); + kfree(new_ra); return -EADDRINUSE; } @@ -288,7 +287,7 @@ int ipv6_setsockopt(struct sock *sk, int level, int optname, { struct ipv6_txoptions *opt; if (optlen == 0) - optval = 0; + optval = NULL; /* hop-by-hop / destination options are privileged option */ retv = -EPERM; diff --git a/net/ipv6/ipv6_syms.c b/net/ipv6/ipv6_syms.c index 37a4a99..1648278 100644 --- a/net/ipv6/ipv6_syms.c +++ b/net/ipv6/ipv6_syms.c @@ -7,7 +7,7 @@ #include <net/ip6_route.h> #include <net/xfrm.h> -EXPORT_SYMBOL(ipv6_addr_type); +EXPORT_SYMBOL(__ipv6_addr_type); EXPORT_SYMBOL(icmpv6_send); EXPORT_SYMBOL(icmpv6_statistics); EXPORT_SYMBOL(icmpv6_err_convert); diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 39a96c7..f15e04a 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -164,7 +164,7 @@ static int ip6_mc_leave_src(struct sock *sk, struct ipv6_mc_socklist *iml, #define MLDV2_MASK(value, nb) ((nb)>=32 ? (value) : ((1<<(nb))-1) & (value)) #define MLDV2_EXP(thresh, nbmant, nbexp, value) \ ((value) < (thresh) ? (value) : \ - ((MLDV2_MASK(value, nbmant) | (1<<(nbmant+nbexp))) << \ + ((MLDV2_MASK(value, nbmant) | (1<<(nbmant))) << \ (MLDV2_MASK((value) >> (nbmant), nbexp) + (nbexp)))) #define MLDV2_QQIC(value) MLDV2_EXP(0x80, 4, 3, value) @@ -545,8 +545,10 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf) sock_kfree_s(sk, newpsl, IP6_SFLSIZE(newpsl->sl_max)); goto done; } - } else + } else { newpsl = NULL; + (void) ip6_mc_add_src(idev, group, gsf->gf_fmode, 0, NULL, 0); + } psl = pmc->sflist; if (psl) { (void) ip6_mc_del_src(idev, group, pmc->sfmode, @@ -1087,7 +1089,7 @@ static void mld_marksources(struct ifmcaddr6 *pmc, int nsrcs, int igmp6_event_query(struct sk_buff *skb) { - struct mld2_query *mlh2 = (struct mld2_query *) skb->h.raw; + struct mld2_query *mlh2 = NULL; struct ifmcaddr6 *ma; struct in6_addr *group; unsigned long max_delay; @@ -1140,6 +1142,13 @@ int igmp6_event_query(struct sk_buff *skb) /* clear deleted report items */ mld_clear_delrec(idev); } else if (len >= 28) { + int srcs_offset = sizeof(struct mld2_query) - + sizeof(struct icmp6hdr); + if (!pskb_may_pull(skb, srcs_offset)) { + in6_dev_put(idev); + return -EINVAL; + } + mlh2 = (struct mld2_query *) skb->h.raw; max_delay = (MLDV2_MRC(ntohs(mlh2->mrc))*HZ)/1000; if (!max_delay) max_delay = 1; @@ -1156,7 +1165,15 @@ int igmp6_event_query(struct sk_buff *skb) return 0; } /* mark sources to include, if group & source-specific */ - mark = mlh2->nsrcs != 0; + if (mlh2->nsrcs != 0) { + if (!pskb_may_pull(skb, srcs_offset + + mlh2->nsrcs * sizeof(struct in6_addr))) { + in6_dev_put(idev); + return -EINVAL; + } + mlh2 = (struct mld2_query *) skb->h.raw; + mark = 1; + } } else { in6_dev_put(idev); return -EINVAL; diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index bb7ccfe..060d612 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -5,10 +5,20 @@ menu "IPv6: Netfilter Configuration (EXPERIMENTAL)" depends on INET && IPV6 && NETFILTER && EXPERIMENTAL -#tristate 'Connection tracking (required for masq/NAT)' CONFIG_IP6_NF_CONNTRACK -#if [ "$CONFIG_IP6_NF_CONNTRACK" != "n" ]; then -# dep_tristate ' FTP protocol support' CONFIG_IP6_NF_FTP $CONFIG_IP6_NF_CONNTRACK -#fi +config NF_CONNTRACK_IPV6 + tristate "IPv6 support for new connection tracking (EXPERIMENTAL)" + depends on EXPERIMENTAL && NF_CONNTRACK + ---help--- + Connection tracking keeps a record of what packets have passed + through your machine, in order to figure out how they are related + into connections. + + This is IPv6 support on Layer 3 independent connection tracking. + Layer 3 independent connection tracking is experimental scheme + which generalize ip_conntrack to support other layer 3 protocols. + + To compile it as a module, choose M here. If unsure, say N. + config IP6_NF_QUEUE tristate "IP6 Userspace queueing via NETLINK (OBSOLETE)" ---help--- @@ -114,7 +124,6 @@ config IP6_NF_MATCH_OWNER To compile it as a module, choose M here. If unsure, say N. -# dep_tristate ' MAC address match support' CONFIG_IP6_NF_MATCH_MAC $CONFIG_IP6_NF_IPTABLES config IP6_NF_MATCH_MARK tristate "netfilter MARK match support" depends on IP6_NF_IPTABLES @@ -170,15 +179,6 @@ config IP6_NF_MATCH_PHYSDEV To compile it as a module, choose M here. If unsure, say N. -# dep_tristate ' Multiple port match support' CONFIG_IP6_NF_MATCH_MULTIPORT $CONFIG_IP6_NF_IPTABLES -# dep_tristate ' TOS match support' CONFIG_IP6_NF_MATCH_TOS $CONFIG_IP6_NF_IPTABLES -# if [ "$CONFIG_IP6_NF_CONNTRACK" != "n" ]; then -# dep_tristate ' Connection state match support' CONFIG_IP6_NF_MATCH_STATE $CONFIG_IP6_NF_CONNTRACK $CONFIG_IP6_NF_IPTABLES -# fi -# if [ "$CONFIG_EXPERIMENTAL" = "y" ]; then -# dep_tristate ' Unclean match support (EXPERIMENTAL)' CONFIG_IP6_NF_MATCH_UNCLEAN $CONFIG_IP6_NF_IPTABLES -# dep_tristate ' Owner match support (EXPERIMENTAL)' CONFIG_IP6_NF_MATCH_OWNER $CONFIG_IP6_NF_IPTABLES -# fi # The targets config IP6_NF_FILTER tristate "Packet filtering" @@ -220,12 +220,6 @@ config IP6_NF_TARGET_NFQUEUE To compile it as a module, choose M here. If unsure, say N. -# if [ "$CONFIG_IP6_NF_FILTER" != "n" ]; then -# dep_tristate ' REJECT target support' CONFIG_IP6_NF_TARGET_REJECT $CONFIG_IP6_NF_FILTER -# if [ "$CONFIG_EXPERIMENTAL" = "y" ]; then -# dep_tristate ' MIRROR target support (EXPERIMENTAL)' CONFIG_IP6_NF_TARGET_MIRROR $CONFIG_IP6_NF_FILTER -# fi -# fi config IP6_NF_MANGLE tristate "Packet mangling" depends on IP6_NF_IPTABLES @@ -236,7 +230,6 @@ config IP6_NF_MANGLE To compile it as a module, choose M here. If unsure, say N. -# dep_tristate ' TOS target support' CONFIG_IP6_NF_TARGET_TOS $CONFIG_IP_NF_MANGLE config IP6_NF_TARGET_MARK tristate "MARK target support" depends on IP6_NF_MANGLE @@ -266,7 +259,6 @@ config IP6_NF_TARGET_HL To compile it as a module, choose M here. If unsure, say N. -#dep_tristate ' LOG target support' CONFIG_IP6_NF_TARGET_LOG $CONFIG_IP6_NF_IPTABLES config IP6_NF_RAW tristate 'raw table support (required for TRACE)' depends on IP6_NF_IPTABLES diff --git a/net/ipv6/netfilter/Makefile b/net/ipv6/netfilter/Makefile index 2b2c370..9ab5b2c 100644 --- a/net/ipv6/netfilter/Makefile +++ b/net/ipv6/netfilter/Makefile @@ -27,3 +27,9 @@ obj-$(CONFIG_IP6_NF_TARGET_LOG) += ip6t_LOG.o obj-$(CONFIG_IP6_NF_RAW) += ip6table_raw.o obj-$(CONFIG_IP6_NF_MATCH_HL) += ip6t_hl.o obj-$(CONFIG_IP6_NF_TARGET_REJECT) += ip6t_REJECT.o + +# objects for l3 independent conntrack +nf_conntrack_ipv6-objs := nf_conntrack_l3proto_ipv6.o nf_conntrack_proto_icmpv6.o nf_conntrack_reasm.o + +# l3 independent conntrack +obj-$(CONFIG_NF_CONNTRACK_IPV6) += nf_conntrack_ipv6.o diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 21deec2..7d49222 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -2,7 +2,7 @@ * Packet matching code. * * Copyright (C) 1999 Paul `Rusty' Russell & Michael J. Neuling - * Copyright (C) 2000-2002 Netfilter core team <coreteam@netfilter.org> + * Copyright (C) 2000-2005 Netfilter Core Team <coreteam@netfilter.org> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -23,7 +23,6 @@ #include <linux/tcp.h> #include <linux/udp.h> #include <linux/icmpv6.h> -#include <net/ip.h> #include <net/ipv6.h> #include <asm/uaccess.h> #include <asm/semaphore.h> @@ -80,13 +79,12 @@ static DECLARE_MUTEX(ip6t_mutex); #define inline #endif -/* Locking is simple: we assume at worst case there will be one packet - in user context and one from bottom halves (or soft irq if Alexey's - softnet patch was applied). - +/* We keep a set of rules for each CPU, so we can avoid write-locking - them; doing a readlock_bh() stops packets coming through if we're - in user context. + them in the softirq when updating the counters and therefore + only need to read-lock in the softirq; doing a write_lock_bh() in user + context stops packets coming through and allows user context to read + the counters or update the rules. To be cache friendly on SMP, we arrange them like so: [ n-entries ] @@ -356,7 +354,7 @@ ip6t_do_table(struct sk_buff **pskb, struct ip6t_table *table, void *userdata) { - static const char nulldevname[IFNAMSIZ]; + static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); int offset = 0; unsigned int protoff = 0; int hotdrop = 0; @@ -369,7 +367,6 @@ ip6t_do_table(struct sk_buff **pskb, /* Initialization */ indev = in ? in->name : nulldevname; outdev = out ? out->name : nulldevname; - /* We handle fragments by dealing with the first fragment as * if it was a normal packet. All other fragments are treated * normally, except that they will NEVER match rules that ask @@ -497,75 +494,145 @@ ip6t_do_table(struct sk_buff **pskb, #endif } -/* If it succeeds, returns element and locks mutex */ -static inline void * -find_inlist_lock_noload(struct list_head *head, - const char *name, - int *error, - struct semaphore *mutex) +/* + * These are weird, but module loading must not be done with mutex + * held (since they will register), and we have to have a single + * function to use try_then_request_module(). + */ + +/* Find table by name, grabs mutex & ref. Returns ERR_PTR() on error. */ +static inline struct ip6t_table *find_table_lock(const char *name) { - void *ret; + struct ip6t_table *t; -#if 1 - duprintf("find_inlist: searching for `%s' in %s.\n", - name, head == &ip6t_target ? "ip6t_target" - : head == &ip6t_match ? "ip6t_match" - : head == &ip6t_tables ? "ip6t_tables" : "UNKNOWN"); -#endif + if (down_interruptible(&ip6t_mutex) != 0) + return ERR_PTR(-EINTR); - *error = down_interruptible(mutex); - if (*error != 0) - return NULL; + list_for_each_entry(t, &ip6t_tables, list) + if (strcmp(t->name, name) == 0 && try_module_get(t->me)) + return t; + up(&ip6t_mutex); + return NULL; +} + +/* Find match, grabs ref. Returns ERR_PTR() on error. */ +static inline struct ip6t_match *find_match(const char *name, u8 revision) +{ + struct ip6t_match *m; + int err = 0; - ret = list_named_find(head, name); - if (!ret) { - *error = -ENOENT; - up(mutex); + if (down_interruptible(&ip6t_mutex) != 0) + return ERR_PTR(-EINTR); + + list_for_each_entry(m, &ip6t_match, list) { + if (strcmp(m->name, name) == 0) { + if (m->revision == revision) { + if (try_module_get(m->me)) { + up(&ip6t_mutex); + return m; + } + } else + err = -EPROTOTYPE; /* Found something. */ + } } - return ret; + up(&ip6t_mutex); + return ERR_PTR(err); } -#ifndef CONFIG_KMOD -#define find_inlist_lock(h,n,p,e,m) find_inlist_lock_noload((h),(n),(e),(m)) -#else -static void * -find_inlist_lock(struct list_head *head, - const char *name, - const char *prefix, - int *error, - struct semaphore *mutex) +/* Find target, grabs ref. Returns ERR_PTR() on error. */ +static inline struct ip6t_target *find_target(const char *name, u8 revision) { - void *ret; + struct ip6t_target *t; + int err = 0; - ret = find_inlist_lock_noload(head, name, error, mutex); - if (!ret) { - duprintf("find_inlist: loading `%s%s'.\n", prefix, name); - request_module("%s%s", prefix, name); - ret = find_inlist_lock_noload(head, name, error, mutex); + if (down_interruptible(&ip6t_mutex) != 0) + return ERR_PTR(-EINTR); + + list_for_each_entry(t, &ip6t_target, list) { + if (strcmp(t->name, name) == 0) { + if (t->revision == revision) { + if (try_module_get(t->me)) { + up(&ip6t_mutex); + return t; + } + } else + err = -EPROTOTYPE; /* Found something. */ + } } + up(&ip6t_mutex); + return ERR_PTR(err); +} - return ret; +struct ip6t_target *ip6t_find_target(const char *name, u8 revision) +{ + struct ip6t_target *target; + + target = try_then_request_module(find_target(name, revision), + "ip6t_%s", name); + if (IS_ERR(target) || !target) + return NULL; + return target; } -#endif -static inline struct ip6t_table * -ip6t_find_table_lock(const char *name, int *error, struct semaphore *mutex) +static int match_revfn(const char *name, u8 revision, int *bestp) { - return find_inlist_lock(&ip6t_tables, name, "ip6table_", error, mutex); + struct ip6t_match *m; + int have_rev = 0; + + list_for_each_entry(m, &ip6t_match, list) { + if (strcmp(m->name, name) == 0) { + if (m->revision > *bestp) + *bestp = m->revision; + if (m->revision == revision) + have_rev = 1; + } + } + return have_rev; } -static inline struct ip6t_match * -find_match_lock(const char *name, int *error, struct semaphore *mutex) +static int target_revfn(const char *name, u8 revision, int *bestp) { - return find_inlist_lock(&ip6t_match, name, "ip6t_", error, mutex); + struct ip6t_target *t; + int have_rev = 0; + + list_for_each_entry(t, &ip6t_target, list) { + if (strcmp(t->name, name) == 0) { + if (t->revision > *bestp) + *bestp = t->revision; + if (t->revision == revision) + have_rev = 1; + } + } + return have_rev; } -static struct ip6t_target * -ip6t_find_target_lock(const char *name, int *error, struct semaphore *mutex) +/* Returns true or fals (if no such extension at all) */ +static inline int find_revision(const char *name, u8 revision, + int (*revfn)(const char *, u8, int *), + int *err) { - return find_inlist_lock(&ip6t_target, name, "ip6t_", error, mutex); + int have_rev, best = -1; + + if (down_interruptible(&ip6t_mutex) != 0) { + *err = -EINTR; + return 1; + } + have_rev = revfn(name, revision, &best); + up(&ip6t_mutex); + + /* Nothing at all? Return 0 to try loading module. */ + if (best == -1) { + *err = -ENOENT; + return 0; + } + + *err = best; + if (!have_rev) + *err = -EPROTONOSUPPORT; + return 1; } + /* All zeroes == unconditional rule. */ static inline int unconditional(const struct ip6t_ip6 *ipv6) @@ -725,20 +792,16 @@ check_match(struct ip6t_entry_match *m, unsigned int hookmask, unsigned int *i) { - int ret; struct ip6t_match *match; - match = find_match_lock(m->u.user.name, &ret, &ip6t_mutex); - if (!match) { - // duprintf("check_match: `%s' not found\n", m->u.name); - return ret; - } - if (!try_module_get(match->me)) { - up(&ip6t_mutex); - return -ENOENT; + match = try_then_request_module(find_match(m->u.user.name, + m->u.user.revision), + "ip6t_%s", m->u.user.name); + if (IS_ERR(match) || !match) { + duprintf("check_match: `%s' not found\n", m->u.user.name); + return match ? PTR_ERR(match) : -ENOENT; } m->u.kernel.match = match; - up(&ip6t_mutex); if (m->u.kernel.match->checkentry && !m->u.kernel.match->checkentry(name, ipv6, m->data, @@ -776,22 +839,16 @@ check_entry(struct ip6t_entry *e, const char *name, unsigned int size, goto cleanup_matches; t = ip6t_get_target(e); - target = ip6t_find_target_lock(t->u.user.name, &ret, &ip6t_mutex); - if (!target) { + target = try_then_request_module(find_target(t->u.user.name, + t->u.user.revision), + "ip6t_%s", t->u.user.name); + if (IS_ERR(target) || !target) { duprintf("check_entry: `%s' not found\n", t->u.user.name); - goto cleanup_matches; - } - if (!try_module_get(target->me)) { - up(&ip6t_mutex); - ret = -ENOENT; + ret = target ? PTR_ERR(target) : -ENOENT; goto cleanup_matches; } t->u.kernel.target = target; - up(&ip6t_mutex); - if (!t->u.kernel.target) { - ret = -EBUSY; - goto cleanup_matches; - } + if (t->u.kernel.target == &ip6t_standard_target) { if (!standard_check(t, size)) { ret = -EINVAL; @@ -1118,8 +1175,8 @@ get_entries(const struct ip6t_get_entries *entries, int ret; struct ip6t_table *t; - t = ip6t_find_table_lock(entries->name, &ret, &ip6t_mutex); - if (t) { + t = find_table_lock(entries->name); + if (t && !IS_ERR(t)) { duprintf("t->private->number = %u\n", t->private->number); if (entries->size == t->private->size) @@ -1131,10 +1188,10 @@ get_entries(const struct ip6t_get_entries *entries, entries->size); ret = -EINVAL; } + module_put(t->me); up(&ip6t_mutex); } else - duprintf("get_entries: Can't find %s!\n", - entries->name); + ret = t ? PTR_ERR(t) : -ENOENT; return ret; } @@ -1182,22 +1239,19 @@ do_replace(void __user *user, unsigned int len) duprintf("ip_tables: Translated table\n"); - t = ip6t_find_table_lock(tmp.name, &ret, &ip6t_mutex); - if (!t) + t = try_then_request_module(find_table_lock(tmp.name), + "ip6table_%s", tmp.name); + if (!t || IS_ERR(t)) { + ret = t ? PTR_ERR(t) : -ENOENT; goto free_newinfo_counters_untrans; + } /* You lied! */ if (tmp.valid_hooks != t->valid_hooks) { duprintf("Valid hook crap: %08X vs %08X\n", tmp.valid_hooks, t->valid_hooks); ret = -EINVAL; - goto free_newinfo_counters_untrans_unlock; - } - - /* Get a reference in advance, we're not allowed fail later */ - if (!try_module_get(t->me)) { - ret = -EBUSY; - goto free_newinfo_counters_untrans_unlock; + goto put_module; } oldinfo = replace_table(t, tmp.num_counters, newinfo, &ret); @@ -1219,7 +1273,6 @@ do_replace(void __user *user, unsigned int len) /* Decrease module usage counts and free resource */ IP6T_ENTRY_ITERATE(oldinfo->entries, oldinfo->size, cleanup_entry,NULL); vfree(oldinfo); - /* Silent error: too late now. */ if (copy_to_user(tmp.counters, counters, sizeof(struct ip6t_counters) * tmp.num_counters) != 0) ret = -EFAULT; @@ -1229,7 +1282,6 @@ do_replace(void __user *user, unsigned int len) put_module: module_put(t->me); - free_newinfo_counters_untrans_unlock: up(&ip6t_mutex); free_newinfo_counters_untrans: IP6T_ENTRY_ITERATE(newinfo->entries, newinfo->size, cleanup_entry,NULL); @@ -1268,7 +1320,7 @@ do_add_counters(void __user *user, unsigned int len) unsigned int i; struct ip6t_counters_info tmp, *paddc; struct ip6t_table *t; - int ret; + int ret = 0; if (copy_from_user(&tmp, user, sizeof(tmp)) != 0) return -EFAULT; @@ -1285,9 +1337,11 @@ do_add_counters(void __user *user, unsigned int len) goto free; } - t = ip6t_find_table_lock(tmp.name, &ret, &ip6t_mutex); - if (!t) + t = find_table_lock(tmp.name); + if (!t || IS_ERR(t)) { + ret = t ? PTR_ERR(t) : -ENOENT; goto free; + } write_lock_bh(&t->lock); if (t->private->number != paddc->num_counters) { @@ -1304,6 +1358,7 @@ do_add_counters(void __user *user, unsigned int len) unlock_up_free: write_unlock_bh(&t->lock); up(&ip6t_mutex); + module_put(t->me); free: vfree(paddc); @@ -1360,8 +1415,10 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) break; } name[IP6T_TABLE_MAXNAMELEN-1] = '\0'; - t = ip6t_find_table_lock(name, &ret, &ip6t_mutex); - if (t) { + + t = try_then_request_module(find_table_lock(name), + "ip6table_%s", name); + if (t && !IS_ERR(t)) { struct ip6t_getinfo info; info.valid_hooks = t->valid_hooks; @@ -1377,9 +1434,10 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) ret = -EFAULT; else ret = 0; - up(&ip6t_mutex); - } + module_put(t->me); + } else + ret = t ? PTR_ERR(t) : -ENOENT; } break; @@ -1400,6 +1458,31 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) break; } + case IP6T_SO_GET_REVISION_MATCH: + case IP6T_SO_GET_REVISION_TARGET: { + struct ip6t_get_revision rev; + int (*revfn)(const char *, u8, int *); + + if (*len != sizeof(rev)) { + ret = -EINVAL; + break; + } + if (copy_from_user(&rev, user, sizeof(rev)) != 0) { + ret = -EFAULT; + break; + } + + if (cmd == IP6T_SO_GET_REVISION_TARGET) + revfn = target_revfn; + else + revfn = match_revfn; + + try_then_request_module(find_revision(rev.name, rev.revision, + revfn, &ret), + "ip6t_%s", rev.name); + break; + } + default: duprintf("do_ip6t_get_ctl: unknown request %i\n", cmd); ret = -EINVAL; @@ -1417,12 +1500,7 @@ ip6t_register_target(struct ip6t_target *target) ret = down_interruptible(&ip6t_mutex); if (ret != 0) return ret; - - if (!list_named_insert(&ip6t_target, target)) { - duprintf("ip6t_register_target: `%s' already in list!\n", - target->name); - ret = -EINVAL; - } + list_add(&target->list, &ip6t_target); up(&ip6t_mutex); return ret; } @@ -1444,11 +1522,7 @@ ip6t_register_match(struct ip6t_match *match) if (ret != 0) return ret; - if (!list_named_insert(&ip6t_match, match)) { - duprintf("ip6t_register_match: `%s' already in list!\n", - match->name); - ret = -EINVAL; - } + list_add(&match->list, &ip6t_match); up(&ip6t_mutex); return ret; diff --git a/net/ipv6/netfilter/ip6t_MARK.c b/net/ipv6/netfilter/ip6t_MARK.c index 81924fc..eab8fb8 100644 --- a/net/ipv6/netfilter/ip6t_MARK.c +++ b/net/ipv6/netfilter/ip6t_MARK.c @@ -56,8 +56,12 @@ checkentry(const char *tablename, return 1; } -static struct ip6t_target ip6t_mark_reg -= { { NULL, NULL }, "MARK", target, checkentry, NULL, THIS_MODULE }; +static struct ip6t_target ip6t_mark_reg = { + .name = "MARK", + .target = target, + .checkentry = checkentry, + .me = THIS_MODULE +}; static int __init init(void) { diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c new file mode 100644 index 0000000..753a3ae --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -0,0 +1,556 @@ +/* + * Copyright (C)2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - support Layer 3 protocol independent connection tracking. + * Based on the original ip_conntrack code which had the following + * copyright information: + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + * + * 23 Mar 2004: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - add get_features() to support various size of conntrack + * structures. + */ + +#include <linux/config.h> +#include <linux/types.h> +#include <linux/ipv6.h> +#include <linux/in6.h> +#include <linux/netfilter.h> +#include <linux/module.h> +#include <linux/skbuff.h> +#include <linux/icmp.h> +#include <linux/sysctl.h> +#include <net/ipv6.h> + +#include <linux/netfilter_ipv6.h> +#include <net/netfilter/nf_conntrack.h> +#include <net/netfilter/nf_conntrack_helper.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_l3proto.h> +#include <net/netfilter/nf_conntrack_core.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +DECLARE_PER_CPU(struct ip_conntrack_stat, nf_conntrack_stat); + +static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) +{ + u_int32_t _addrs[8], *ap; + + ap = skb_header_pointer(skb, nhoff + offsetof(struct ipv6hdr, saddr), + sizeof(_addrs), _addrs); + if (ap == NULL) + return 0; + + memcpy(tuple->src.u3.ip6, ap, sizeof(tuple->src.u3.ip6)); + memcpy(tuple->dst.u3.ip6, ap + 4, sizeof(tuple->dst.u3.ip6)); + + return 1; +} + +static int ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + memcpy(tuple->src.u3.ip6, orig->dst.u3.ip6, sizeof(tuple->src.u3.ip6)); + memcpy(tuple->dst.u3.ip6, orig->src.u3.ip6, sizeof(tuple->dst.u3.ip6)); + + return 1; +} + +static int ipv6_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "src=%x:%x:%x:%x:%x:%x:%x:%x dst=%x:%x:%x:%x:%x:%x:%x:%x ", + NIP6(*((struct in6_addr *)tuple->src.u3.ip6)), + NIP6(*((struct in6_addr *)tuple->dst.u3.ip6))); +} + +static int ipv6_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* + * Based on ipv6_skip_exthdr() in net/ipv6/exthdr.c + * + * This function parses (probably truncated) exthdr set "hdr" + * of length "len". "nexthdrp" initially points to some place, + * where type of the first header can be found. + * + * It skips all well-known exthdrs, and returns pointer to the start + * of unparsable area i.e. the first header with unknown type. + * if success, *nexthdr is updated by type/protocol of this header. + * + * NOTES: - it may return pointer pointing beyond end of packet, + * if the last recognized header is truncated in the middle. + * - if packet is truncated, so that all parsed headers are skipped, + * it returns -1. + * - if packet is fragmented, return pointer of the fragment header. + * - ESP is unparsable for now and considered like + * normal payload protocol. + * - Note also special handling of AUTH header. Thanks to IPsec wizards. + */ + +int nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, + int len) +{ + u8 nexthdr = *nexthdrp; + + while (ipv6_ext_hdr(nexthdr)) { + struct ipv6_opt_hdr hdr; + int hdrlen; + + if (len < (int)sizeof(struct ipv6_opt_hdr)) + return -1; + if (nexthdr == NEXTHDR_NONE) + break; + if (nexthdr == NEXTHDR_FRAGMENT) + break; + if (skb_copy_bits(skb, start, &hdr, sizeof(hdr))) + BUG(); + if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hdr.hdrlen+2)<<2; + else + hdrlen = ipv6_optlen(&hdr); + + nexthdr = hdr.nexthdr; + len -= hdrlen; + start += hdrlen; + } + + *nexthdrp = nexthdr; + return start; +} + +static int +ipv6_prepare(struct sk_buff **pskb, unsigned int hooknum, unsigned int *dataoff, + u_int8_t *protonum) +{ + unsigned int extoff; + unsigned char pnum; + int protoff; + + extoff = (u8*)((*pskb)->nh.ipv6h + 1) - (*pskb)->data; + pnum = (*pskb)->nh.ipv6h->nexthdr; + + protoff = nf_ct_ipv6_skip_exthdr(*pskb, extoff, &pnum, + (*pskb)->len - extoff); + + /* + * (protoff == (*pskb)->len) mean that the packet doesn't have no data + * except of IPv6 & ext headers. but it's tracked anyway. - YK + */ + if ((protoff < 0) || (protoff > (*pskb)->len)) { + DEBUGP("ip6_conntrack_core: can't find proto in pkt\n"); + NF_CT_STAT_INC(error); + NF_CT_STAT_INC(invalid); + return -NF_ACCEPT; + } + + *dataoff = protoff; + *protonum = pnum; + return NF_ACCEPT; +} + +static u_int32_t ipv6_get_features(const struct nf_conntrack_tuple *tuple) +{ + return NF_CT_F_BASIC; +} + +static unsigned int ipv6_confirm(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + + /* This is where we call the helper: as the packet goes out. */ + ct = nf_ct_get(*pskb, &ctinfo); + if (ct && ct->helper) { + unsigned int ret, protoff; + unsigned int extoff = (u8*)((*pskb)->nh.ipv6h + 1) + - (*pskb)->data; + unsigned char pnum = (*pskb)->nh.ipv6h->nexthdr; + + protoff = nf_ct_ipv6_skip_exthdr(*pskb, extoff, &pnum, + (*pskb)->len - extoff); + if (protoff < 0 || protoff > (*pskb)->len || + pnum == NEXTHDR_FRAGMENT) { + DEBUGP("proto header not found\n"); + return NF_ACCEPT; + } + + ret = ct->helper->help(pskb, protoff, ct, ctinfo); + if (ret != NF_ACCEPT) + return ret; + } + + /* We've seen it coming out the other side: confirm it */ + + return nf_conntrack_confirm(pskb); +} + +extern struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb); +extern void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, + struct net_device *out, + int (*okfn)(struct sk_buff *)); +static unsigned int ipv6_defrag(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm; + + /* Previously seen (loopback)? */ + if ((*pskb)->nfct) + return NF_ACCEPT; + + reasm = nf_ct_frag6_gather(*pskb); + + /* queued */ + if (reasm == NULL) + return NF_STOLEN; + + /* error occured or not fragmented */ + if (reasm == *pskb) + return NF_ACCEPT; + + nf_ct_frag6_output(hooknum, reasm, (struct net_device *)in, + (struct net_device *)out, okfn); + + return NF_STOLEN; +} + +static unsigned int ipv6_conntrack_in(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *reasm = (*pskb)->nfct_reasm; + + /* This packet is fragmented and has reassembled packet. */ + if (reasm) { + /* Reassembled packet isn't parsed yet ? */ + if (!reasm->nfct) { + unsigned int ret; + + ret = nf_conntrack_in(PF_INET6, hooknum, &reasm); + if (ret != NF_ACCEPT) + return ret; + } + nf_conntrack_get(reasm->nfct); + (*pskb)->nfct = reasm->nfct; + return NF_ACCEPT; + } + + return nf_conntrack_in(PF_INET6, hooknum, pskb); +} + +static unsigned int ipv6_conntrack_local(unsigned int hooknum, + struct sk_buff **pskb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + /* root is playing with raw sockets. */ + if ((*pskb)->len < sizeof(struct ipv6hdr)) { + if (net_ratelimit()) + printk("ipv6_conntrack_local: packet too short\n"); + return NF_ACCEPT; + } + return ipv6_conntrack_in(hooknum, pskb, in, out, okfn); +} + +/* Connection tracking may drop packets, but never alters them, so + make it the first hook. */ +static struct nf_hook_ops ipv6_conntrack_defrag_ops = { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, +}; + +static struct nf_hook_ops ipv6_conntrack_in_ops = { + .hook = ipv6_conntrack_in, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_PRE_ROUTING, + .priority = NF_IP6_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv6_conntrack_local_out_ops = { + .hook = ipv6_conntrack_local, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK, +}; + +static struct nf_hook_ops ipv6_conntrack_defrag_local_out_ops = { + .hook = ipv6_defrag, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_OUT, + .priority = NF_IP6_PRI_CONNTRACK_DEFRAG, +}; + +/* Refragmenter; last chance. */ +static struct nf_hook_ops ipv6_conntrack_out_ops = { + .hook = ipv6_confirm, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_POST_ROUTING, + .priority = NF_IP6_PRI_LAST, +}; + +static struct nf_hook_ops ipv6_conntrack_local_in_ops = { + .hook = ipv6_confirm, + .owner = THIS_MODULE, + .pf = PF_INET6, + .hooknum = NF_IP6_LOCAL_IN, + .priority = NF_IP6_PRI_LAST-1, +}; + +#ifdef CONFIG_SYSCTL + +/* From nf_conntrack_proto_icmpv6.c */ +extern unsigned long nf_ct_icmpv6_timeout; + +/* From nf_conntrack_frag6.c */ +extern unsigned long nf_ct_frag6_timeout; +extern unsigned int nf_ct_frag6_low_thresh; +extern unsigned int nf_ct_frag6_high_thresh; + +static struct ctl_table_header *nf_ct_ipv6_sysctl_header; + +static ctl_table nf_ct_sysctl_table[] = { + { + .ctl_name = NET_NF_CONNTRACK_ICMPV6_TIMEOUT, + .procname = "nf_conntrack_icmpv6_timeout", + .data = &nf_ct_icmpv6_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_TIMEOUT, + .procname = "nf_conntrack_frag6_timeout", + .data = &nf_ct_frag6_timeout, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec_jiffies, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_LOW_THRESH, + .procname = "nf_conntrack_frag6_low_thresh", + .data = &nf_ct_frag6_low_thresh, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = NET_NF_CONNTRACK_FRAG6_HIGH_THRESH, + .procname = "nf_conntrack_frag6_high_thresh", + .data = &nf_ct_frag6_high_thresh, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_netfilter_table[] = { + { + .ctl_name = NET_NETFILTER, + .procname = "netfilter", + .mode = 0555, + .child = nf_ct_sysctl_table, + }, + { .ctl_name = 0 } +}; + +static ctl_table nf_ct_net_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = nf_ct_netfilter_table, + }, + { .ctl_name = 0 } +}; +#endif + +struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6 = { + .l3proto = PF_INET6, + .name = "ipv6", + .pkt_to_tuple = ipv6_pkt_to_tuple, + .invert_tuple = ipv6_invert_tuple, + .print_tuple = ipv6_print_tuple, + .print_conntrack = ipv6_print_conntrack, + .prepare = ipv6_prepare, + .get_features = ipv6_get_features, + .me = THIS_MODULE, +}; + +extern struct nf_conntrack_protocol nf_conntrack_protocol_tcp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_udp6; +extern struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6; +extern int nf_ct_frag6_init(void); +extern void nf_ct_frag6_cleanup(void); +static int init_or_cleanup(int init) +{ + int ret = 0; + + if (!init) goto cleanup; + + ret = nf_ct_frag6_init(); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't initialize frag6.\n"); + goto cleanup_nothing; + } + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_tcp6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register tcp.\n"); + goto cleanup_frag6; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_udp6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register udp.\n"); + goto cleanup_tcp; + } + + ret = nf_conntrack_protocol_register(&nf_conntrack_protocol_icmpv6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register icmpv6.\n"); + goto cleanup_udp; + } + + ret = nf_conntrack_l3proto_register(&nf_conntrack_l3proto_ipv6); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register ipv6\n"); + goto cleanup_icmpv6; + } + + ret = nf_register_hook(&ipv6_conntrack_defrag_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register pre-routing defrag " + "hook.\n"); + goto cleanup_ipv6; + } + + ret = nf_register_hook(&ipv6_conntrack_defrag_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local_out defrag " + "hook.\n"); + goto cleanup_defragops; + } + + ret = nf_register_hook(&ipv6_conntrack_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register pre-routing hook.\n"); + goto cleanup_defraglocalops; + } + + ret = nf_register_hook(&ipv6_conntrack_local_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local out hook.\n"); + goto cleanup_inops; + } + + ret = nf_register_hook(&ipv6_conntrack_out_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register post-routing hook.\n"); + goto cleanup_inandlocalops; + } + + ret = nf_register_hook(&ipv6_conntrack_local_in_ops); + if (ret < 0) { + printk("nf_conntrack_ipv6: can't register local in hook.\n"); + goto cleanup_inoutandlocalops; + } + +#ifdef CONFIG_SYSCTL + nf_ct_ipv6_sysctl_header = register_sysctl_table(nf_ct_net_table, 0); + if (nf_ct_ipv6_sysctl_header == NULL) { + printk("nf_conntrack: can't register to sysctl.\n"); + ret = -ENOMEM; + goto cleanup_localinops; + } +#endif + return ret; + + cleanup: + synchronize_net(); +#ifdef CONFIG_SYSCTL + unregister_sysctl_table(nf_ct_ipv6_sysctl_header); + cleanup_localinops: +#endif + nf_unregister_hook(&ipv6_conntrack_local_in_ops); + cleanup_inoutandlocalops: + nf_unregister_hook(&ipv6_conntrack_out_ops); + cleanup_inandlocalops: + nf_unregister_hook(&ipv6_conntrack_local_out_ops); + cleanup_inops: + nf_unregister_hook(&ipv6_conntrack_in_ops); + cleanup_defraglocalops: + nf_unregister_hook(&ipv6_conntrack_defrag_local_out_ops); + cleanup_defragops: + nf_unregister_hook(&ipv6_conntrack_defrag_ops); + cleanup_ipv6: + nf_conntrack_l3proto_unregister(&nf_conntrack_l3proto_ipv6); + cleanup_icmpv6: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_icmpv6); + cleanup_udp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_udp6); + cleanup_tcp: + nf_conntrack_protocol_unregister(&nf_conntrack_protocol_tcp6); + cleanup_frag6: + nf_ct_frag6_cleanup(); + cleanup_nothing: + return ret; +} + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Yasuyuki KOZAKAI @USAGI <yasuyuki.kozakai@toshiba.co.jp>"); + +static int __init init(void) +{ + need_nf_conntrack(); + return init_or_cleanup(1); +} + +static void __exit fini(void) +{ + init_or_cleanup(0); +} + +module_init(init); +module_exit(fini); + +void need_ip6_conntrack(void) +{ +} + +EXPORT_SYMBOL(need_ip6_conntrack); diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c new file mode 100644 index 0000000..c0f1da5 --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -0,0 +1,272 @@ +/* + * Copyright (C)2003,2004 USAGI/WIDE Project + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * 16 Dec 2003: Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * - ICMPv6 tracking support. Derived from the original ip_conntrack code + * net/ipv4/netfilter/ip_conntrack_proto_icmp.c which had the following + * copyright information: + * (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2004 Netfilter Core Team <coreteam@netfilter.org> + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/timer.h> +#include <linux/module.h> +#include <linux/netfilter.h> +#include <linux/in6.h> +#include <linux/icmpv6.h> +#include <linux/ipv6.h> +#include <net/ipv6.h> +#include <net/ip6_checksum.h> +#include <linux/seq_file.h> +#include <linux/netfilter_ipv6.h> +#include <net/netfilter/nf_conntrack_tuple.h> +#include <net/netfilter/nf_conntrack_protocol.h> +#include <net/netfilter/nf_conntrack_core.h> +#include <net/netfilter/ipv6/nf_conntrack_icmpv6.h> + +unsigned long nf_ct_icmpv6_timeout = 30*HZ; + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +static int icmpv6_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct icmp6hdr _hdr, *hp; + + hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (hp == NULL) + return 0; + tuple->dst.u.icmp.type = hp->icmp6_type; + tuple->src.u.icmp.id = hp->icmp6_identifier; + tuple->dst.u.icmp.code = hp->icmp6_code; + + return 1; +} + +static int icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) +{ + /* Add 1; spaces filled with 0. */ + static u_int8_t invmap[] = { + [ICMPV6_ECHO_REQUEST - 128] = ICMPV6_ECHO_REPLY + 1, + [ICMPV6_ECHO_REPLY - 128] = ICMPV6_ECHO_REQUEST + 1, + [ICMPV6_NI_QUERY - 128] = ICMPV6_NI_QUERY + 1, + [ICMPV6_NI_REPLY - 128] = ICMPV6_NI_REPLY +1 + }; + + __u8 type = orig->dst.u.icmp.type - 128; + if (type >= sizeof(invmap) || !invmap[type]) + return 0; + + tuple->src.u.icmp.id = orig->src.u.icmp.id; + tuple->dst.u.icmp.type = invmap[type] - 1; + tuple->dst.u.icmp.code = orig->dst.u.icmp.code; + return 1; +} + +/* Print out the per-protocol part of the tuple. */ +static int icmpv6_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "type=%u code=%u id=%u ", + tuple->dst.u.icmp.type, + tuple->dst.u.icmp.code, + ntohs(tuple->src.u.icmp.id)); +} + +/* Print out the private part of the conntrack. */ +static int icmpv6_print_conntrack(struct seq_file *s, + const struct nf_conn *conntrack) +{ + return 0; +} + +/* Returns verdict for packet, or -1 for invalid. */ +static int icmpv6_packet(struct nf_conn *ct, + const struct sk_buff *skb, + unsigned int dataoff, + enum ip_conntrack_info ctinfo, + int pf, + unsigned int hooknum) +{ + /* Try to delete connection immediately after all replies: + won't actually vanish as we still have skb, and del_timer + means this will only run once even if count hits zero twice + (theoretically possible with SMP) */ + if (CTINFO2DIR(ctinfo) == IP_CT_DIR_REPLY) { + if (atomic_dec_and_test(&ct->proto.icmp.count) + && del_timer(&ct->timeout)) + ct->timeout.function((unsigned long)ct); + } else { + atomic_inc(&ct->proto.icmp.count); + nf_conntrack_event_cache(IPCT_PROTOINFO_VOLATILE, skb); + nf_ct_refresh_acct(ct, ctinfo, skb, nf_ct_icmpv6_timeout); + } + + return NF_ACCEPT; +} + +/* Called when a new connection for this protocol found. */ +static int icmpv6_new(struct nf_conn *conntrack, + const struct sk_buff *skb, + unsigned int dataoff) +{ + static u_int8_t valid_new[] = { + [ICMPV6_ECHO_REQUEST - 128] = 1, + [ICMPV6_NI_QUERY - 128] = 1 + }; + + if (conntrack->tuplehash[0].tuple.dst.u.icmp.type - 128 >= sizeof(valid_new) + || !valid_new[conntrack->tuplehash[0].tuple.dst.u.icmp.type - 128]) { + /* Can't create a new ICMPv6 `conn' with this. */ + DEBUGP("icmp: can't create new conn with type %u\n", + conntrack->tuplehash[0].tuple.dst.u.icmp.type); + NF_CT_DUMP_TUPLE(&conntrack->tuplehash[0].tuple); + return 0; + } + atomic_set(&conntrack->proto.icmp.count, 0); + return 1; +} + +extern int +nf_ct_ipv6_skip_exthdr(struct sk_buff *skb, int start, u8 *nexthdrp, int len); +extern struct nf_conntrack_l3proto nf_conntrack_l3proto_ipv6; +static int +icmpv6_error_message(struct sk_buff *skb, + unsigned int icmp6off, + enum ip_conntrack_info *ctinfo, + unsigned int hooknum) +{ + struct nf_conntrack_tuple intuple, origtuple; + struct nf_conntrack_tuple_hash *h; + struct icmp6hdr _hdr, *hp; + unsigned int inip6off; + struct nf_conntrack_protocol *inproto; + u_int8_t inprotonum; + unsigned int inprotoff; + + NF_CT_ASSERT(skb->nfct == NULL); + + hp = skb_header_pointer(skb, icmp6off, sizeof(_hdr), &_hdr); + if (hp == NULL) { + DEBUGP("icmpv6_error: Can't get ICMPv6 hdr.\n"); + return -NF_ACCEPT; + } + + inip6off = icmp6off + sizeof(_hdr); + if (skb_copy_bits(skb, inip6off+offsetof(struct ipv6hdr, nexthdr), + &inprotonum, sizeof(inprotonum)) != 0) { + DEBUGP("icmpv6_error: Can't get nexthdr in inner IPv6 header.\n"); + return -NF_ACCEPT; + } + inprotoff = nf_ct_ipv6_skip_exthdr(skb, + inip6off + sizeof(struct ipv6hdr), + &inprotonum, + skb->len - inip6off + - sizeof(struct ipv6hdr)); + + if ((inprotoff < 0) || (inprotoff > skb->len) || + (inprotonum == NEXTHDR_FRAGMENT)) { + DEBUGP("icmpv6_error: Can't get protocol header in ICMPv6 payload.\n"); + return -NF_ACCEPT; + } + + inproto = nf_ct_find_proto(PF_INET6, inprotonum); + + /* Are they talking about one of our connections? */ + if (!nf_ct_get_tuple(skb, inip6off, inprotoff, PF_INET6, inprotonum, + &origtuple, &nf_conntrack_l3proto_ipv6, inproto)) { + DEBUGP("icmpv6_error: Can't get tuple\n"); + return -NF_ACCEPT; + } + + /* Ordinarily, we'd expect the inverted tupleproto, but it's + been preserved inside the ICMP. */ + if (!nf_ct_invert_tuple(&intuple, &origtuple, + &nf_conntrack_l3proto_ipv6, inproto)) { + DEBUGP("icmpv6_error: Can't invert tuple\n"); + return -NF_ACCEPT; + } + + *ctinfo = IP_CT_RELATED; + + h = nf_conntrack_find_get(&intuple, NULL); + if (!h) { + DEBUGP("icmpv6_error: no match\n"); + return -NF_ACCEPT; + } else { + if (NF_CT_DIRECTION(h) == IP_CT_DIR_REPLY) + *ctinfo += IP_CT_IS_REPLY; + } + + /* Update skb to refer to this connection */ + skb->nfct = &nf_ct_tuplehash_to_ctrack(h)->ct_general; + skb->nfctinfo = *ctinfo; + return -NF_ACCEPT; +} + +static int +icmpv6_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, int pf, unsigned int hooknum) +{ + struct icmp6hdr _ih, *icmp6h; + + icmp6h = skb_header_pointer(skb, dataoff, sizeof(_ih), &_ih); + if (icmp6h == NULL) { + if (LOG_INVALID(IPPROTO_ICMPV6)) + nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL, + "nf_ct_icmpv6: short packet "); + return -NF_ACCEPT; + } + + if (hooknum != NF_IP6_PRE_ROUTING) + goto skipped; + + /* Ignore it if the checksum's bogus. */ + if (csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr, + skb->len - dataoff, IPPROTO_ICMPV6, + skb_checksum(skb, dataoff, + skb->len - dataoff, 0))) { + nf_log_packet(PF_INET6, 0, skb, NULL, NULL, NULL, + "nf_ct_icmpv6: ICMPv6 checksum failed\n"); + return -NF_ACCEPT; + } + +skipped: + + /* is not error message ? */ + if (icmp6h->icmp6_type >= 128) + return NF_ACCEPT; + + return icmpv6_error_message(skb, dataoff, ctinfo, hooknum); +} + +struct nf_conntrack_protocol nf_conntrack_protocol_icmpv6 = +{ + .l3proto = PF_INET6, + .proto = IPPROTO_ICMPV6, + .name = "icmpv6", + .pkt_to_tuple = icmpv6_pkt_to_tuple, + .invert_tuple = icmpv6_invert_tuple, + .print_tuple = icmpv6_print_tuple, + .print_conntrack = icmpv6_print_conntrack, + .packet = icmpv6_packet, + .new = icmpv6_new, + .error = icmpv6_error, +}; + +EXPORT_SYMBOL(nf_conntrack_protocol_icmpv6); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c new file mode 100644 index 0000000..c2c52af --- /dev/null +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -0,0 +1,897 @@ +/* + * IPv6 fragment reassembly for connection tracking + * + * Copyright (C)2004 USAGI/WIDE Project + * + * Author: + * Yasuyuki Kozakai @USAGI <yasuyuki.kozakai@toshiba.co.jp> + * + * Based on: net/ipv6/reassembly.c + * + * 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/config.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/socket.h> +#include <linux/sockios.h> +#include <linux/jiffies.h> +#include <linux/net.h> +#include <linux/list.h> +#include <linux/netdevice.h> +#include <linux/in6.h> +#include <linux/ipv6.h> +#include <linux/icmpv6.h> +#include <linux/random.h> +#include <linux/jhash.h> + +#include <net/sock.h> +#include <net/snmp.h> + +#include <net/ipv6.h> +#include <net/protocol.h> +#include <net/transp_v6.h> +#include <net/rawv6.h> +#include <net/ndisc.h> +#include <net/addrconf.h> +#include <linux/sysctl.h> +#include <linux/netfilter.h> +#include <linux/netfilter_ipv6.h> +#include <linux/kernel.h> +#include <linux/module.h> + +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(format, args...) +#endif + +#define NF_CT_FRAG6_HIGH_THRESH 262144 /* == 256*1024 */ +#define NF_CT_FRAG6_LOW_THRESH 196608 /* == 192*1024 */ +#define NF_CT_FRAG6_TIMEOUT IPV6_FRAG_TIMEOUT + +unsigned int nf_ct_frag6_high_thresh = 256*1024; +unsigned int nf_ct_frag6_low_thresh = 192*1024; +unsigned long nf_ct_frag6_timeout = IPV6_FRAG_TIMEOUT; + +struct nf_ct_frag6_skb_cb +{ + struct inet6_skb_parm h; + int offset; + struct sk_buff *orig; +}; + +#define NFCT_FRAG6_CB(skb) ((struct nf_ct_frag6_skb_cb*)((skb)->cb)) + +struct nf_ct_frag6_queue +{ + struct nf_ct_frag6_queue *next; + struct list_head lru_list; /* lru list member */ + + __u32 id; /* fragment id */ + struct in6_addr saddr; + struct in6_addr daddr; + + spinlock_t lock; + atomic_t refcnt; + struct timer_list timer; /* expire timer */ + struct sk_buff *fragments; + int len; + int meat; + struct timeval stamp; + unsigned int csum; + __u8 last_in; /* has first/last segment arrived? */ +#define COMPLETE 4 +#define FIRST_IN 2 +#define LAST_IN 1 + __u16 nhoffset; + struct nf_ct_frag6_queue **pprev; +}; + +/* Hash table. */ + +#define FRAG6Q_HASHSZ 64 + +static struct nf_ct_frag6_queue *nf_ct_frag6_hash[FRAG6Q_HASHSZ]; +static rwlock_t nf_ct_frag6_lock = RW_LOCK_UNLOCKED; +static u32 nf_ct_frag6_hash_rnd; +static LIST_HEAD(nf_ct_frag6_lru_list); +int nf_ct_frag6_nqueues = 0; + +static __inline__ void __fq_unlink(struct nf_ct_frag6_queue *fq) +{ + if (fq->next) + fq->next->pprev = fq->pprev; + *fq->pprev = fq->next; + list_del(&fq->lru_list); + nf_ct_frag6_nqueues--; +} + +static __inline__ void fq_unlink(struct nf_ct_frag6_queue *fq) +{ + write_lock(&nf_ct_frag6_lock); + __fq_unlink(fq); + write_unlock(&nf_ct_frag6_lock); +} + +static unsigned int ip6qhashfn(u32 id, struct in6_addr *saddr, + struct in6_addr *daddr) +{ + u32 a, b, c; + + a = saddr->s6_addr32[0]; + b = saddr->s6_addr32[1]; + c = saddr->s6_addr32[2]; + + a += JHASH_GOLDEN_RATIO; + b += JHASH_GOLDEN_RATIO; + c += nf_ct_frag6_hash_rnd; + __jhash_mix(a, b, c); + + a += saddr->s6_addr32[3]; + b += daddr->s6_addr32[0]; + c += daddr->s6_addr32[1]; + __jhash_mix(a, b, c); + + a += daddr->s6_addr32[2]; + b += daddr->s6_addr32[3]; + c += id; + __jhash_mix(a, b, c); + + return c & (FRAG6Q_HASHSZ - 1); +} + +static struct timer_list nf_ct_frag6_secret_timer; +int nf_ct_frag6_secret_interval = 10 * 60 * HZ; + +static void nf_ct_frag6_secret_rebuild(unsigned long dummy) +{ + unsigned long now = jiffies; + int i; + + write_lock(&nf_ct_frag6_lock); + get_random_bytes(&nf_ct_frag6_hash_rnd, sizeof(u32)); + for (i = 0; i < FRAG6Q_HASHSZ; i++) { + struct nf_ct_frag6_queue *q; + + q = nf_ct_frag6_hash[i]; + while (q) { + struct nf_ct_frag6_queue *next = q->next; + unsigned int hval = ip6qhashfn(q->id, + &q->saddr, + &q->daddr); + + if (hval != i) { + /* Unlink. */ + if (q->next) + q->next->pprev = q->pprev; + *q->pprev = q->next; + + /* Relink to new hash chain. */ + if ((q->next = nf_ct_frag6_hash[hval]) != NULL) + q->next->pprev = &q->next; + nf_ct_frag6_hash[hval] = q; + q->pprev = &nf_ct_frag6_hash[hval]; + } + + q = next; + } + } + write_unlock(&nf_ct_frag6_lock); + + mod_timer(&nf_ct_frag6_secret_timer, now + nf_ct_frag6_secret_interval); +} + +atomic_t nf_ct_frag6_mem = ATOMIC_INIT(0); + +/* Memory Tracking Functions. */ +static inline void frag_kfree_skb(struct sk_buff *skb, unsigned int *work) +{ + if (work) + *work -= skb->truesize; + atomic_sub(skb->truesize, &nf_ct_frag6_mem); + if (NFCT_FRAG6_CB(skb)->orig) + kfree_skb(NFCT_FRAG6_CB(skb)->orig); + + kfree_skb(skb); +} + +static inline void frag_free_queue(struct nf_ct_frag6_queue *fq, + unsigned int *work) +{ + if (work) + *work -= sizeof(struct nf_ct_frag6_queue); + atomic_sub(sizeof(struct nf_ct_frag6_queue), &nf_ct_frag6_mem); + kfree(fq); +} + +static inline struct nf_ct_frag6_queue *frag_alloc_queue(void) +{ + struct nf_ct_frag6_queue *fq = kmalloc(sizeof(struct nf_ct_frag6_queue), GFP_ATOMIC); + + if (!fq) + return NULL; + atomic_add(sizeof(struct nf_ct_frag6_queue), &nf_ct_frag6_mem); + return fq; +} + +/* Destruction primitives. */ + +/* Complete destruction of fq. */ +static void nf_ct_frag6_destroy(struct nf_ct_frag6_queue *fq, + unsigned int *work) +{ + struct sk_buff *fp; + + BUG_TRAP(fq->last_in&COMPLETE); + BUG_TRAP(del_timer(&fq->timer) == 0); + + /* Release all fragment data. */ + fp = fq->fragments; + while (fp) { + struct sk_buff *xp = fp->next; + + frag_kfree_skb(fp, work); + fp = xp; + } + + frag_free_queue(fq, work); +} + +static __inline__ void fq_put(struct nf_ct_frag6_queue *fq, unsigned int *work) +{ + if (atomic_dec_and_test(&fq->refcnt)) + nf_ct_frag6_destroy(fq, work); +} + +/* Kill fq entry. It is not destroyed immediately, + * because caller (and someone more) holds reference count. + */ +static __inline__ void fq_kill(struct nf_ct_frag6_queue *fq) +{ + if (del_timer(&fq->timer)) + atomic_dec(&fq->refcnt); + + if (!(fq->last_in & COMPLETE)) { + fq_unlink(fq); + atomic_dec(&fq->refcnt); + fq->last_in |= COMPLETE; + } +} + +static void nf_ct_frag6_evictor(void) +{ + struct nf_ct_frag6_queue *fq; + struct list_head *tmp; + unsigned int work; + + work = atomic_read(&nf_ct_frag6_mem); + if (work <= nf_ct_frag6_low_thresh) + return; + + work -= nf_ct_frag6_low_thresh; + while (work > 0) { + read_lock(&nf_ct_frag6_lock); + if (list_empty(&nf_ct_frag6_lru_list)) { + read_unlock(&nf_ct_frag6_lock); + return; + } + tmp = nf_ct_frag6_lru_list.next; + BUG_ON(tmp == NULL); + fq = list_entry(tmp, struct nf_ct_frag6_queue, lru_list); + atomic_inc(&fq->refcnt); + read_unlock(&nf_ct_frag6_lock); + + spin_lock(&fq->lock); + if (!(fq->last_in&COMPLETE)) + fq_kill(fq); + spin_unlock(&fq->lock); + + fq_put(fq, &work); + } +} + +static void nf_ct_frag6_expire(unsigned long data) +{ + struct nf_ct_frag6_queue *fq = (struct nf_ct_frag6_queue *) data; + + spin_lock(&fq->lock); + + if (fq->last_in & COMPLETE) + goto out; + + fq_kill(fq); + +out: + spin_unlock(&fq->lock); + fq_put(fq, NULL); +} + +/* Creation primitives. */ + + +static struct nf_ct_frag6_queue *nf_ct_frag6_intern(unsigned int hash, + struct nf_ct_frag6_queue *fq_in) +{ + struct nf_ct_frag6_queue *fq; + + write_lock(&nf_ct_frag6_lock); +#ifdef CONFIG_SMP + for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + if (fq->id == fq_in->id && + !ipv6_addr_cmp(&fq_in->saddr, &fq->saddr) && + !ipv6_addr_cmp(&fq_in->daddr, &fq->daddr)) { + atomic_inc(&fq->refcnt); + write_unlock(&nf_ct_frag6_lock); + fq_in->last_in |= COMPLETE; + fq_put(fq_in, NULL); + return fq; + } + } +#endif + fq = fq_in; + + if (!mod_timer(&fq->timer, jiffies + nf_ct_frag6_timeout)) + atomic_inc(&fq->refcnt); + + atomic_inc(&fq->refcnt); + if ((fq->next = nf_ct_frag6_hash[hash]) != NULL) + fq->next->pprev = &fq->next; + nf_ct_frag6_hash[hash] = fq; + fq->pprev = &nf_ct_frag6_hash[hash]; + INIT_LIST_HEAD(&fq->lru_list); + list_add_tail(&fq->lru_list, &nf_ct_frag6_lru_list); + nf_ct_frag6_nqueues++; + write_unlock(&nf_ct_frag6_lock); + return fq; +} + + +static struct nf_ct_frag6_queue * +nf_ct_frag6_create(unsigned int hash, u32 id, struct in6_addr *src, struct in6_addr *dst) +{ + struct nf_ct_frag6_queue *fq; + + if ((fq = frag_alloc_queue()) == NULL) { + DEBUGP("Can't alloc new queue\n"); + goto oom; + } + + memset(fq, 0, sizeof(struct nf_ct_frag6_queue)); + + fq->id = id; + ipv6_addr_copy(&fq->saddr, src); + ipv6_addr_copy(&fq->daddr, dst); + + init_timer(&fq->timer); + fq->timer.function = nf_ct_frag6_expire; + fq->timer.data = (long) fq; + fq->lock = SPIN_LOCK_UNLOCKED; + atomic_set(&fq->refcnt, 1); + + return nf_ct_frag6_intern(hash, fq); + +oom: + return NULL; +} + +static __inline__ struct nf_ct_frag6_queue * +fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) +{ + struct nf_ct_frag6_queue *fq; + unsigned int hash = ip6qhashfn(id, src, dst); + + read_lock(&nf_ct_frag6_lock); + for (fq = nf_ct_frag6_hash[hash]; fq; fq = fq->next) { + if (fq->id == id && + !ipv6_addr_cmp(src, &fq->saddr) && + !ipv6_addr_cmp(dst, &fq->daddr)) { + atomic_inc(&fq->refcnt); + read_unlock(&nf_ct_frag6_lock); + return fq; + } + } + read_unlock(&nf_ct_frag6_lock); + + return nf_ct_frag6_create(hash, id, src, dst); +} + + +static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, + struct frag_hdr *fhdr, int nhoff) +{ + struct sk_buff *prev, *next; + int offset, end; + + if (fq->last_in & COMPLETE) { + DEBUGP("Allready completed\n"); + goto err; + } + + offset = ntohs(fhdr->frag_off) & ~0x7; + end = offset + (ntohs(skb->nh.ipv6h->payload_len) - + ((u8 *) (fhdr + 1) - (u8 *) (skb->nh.ipv6h + 1))); + + if ((unsigned int)end > IPV6_MAXPLEN) { + DEBUGP("offset is too large.\n"); + return -1; + } + + if (skb->ip_summed == CHECKSUM_HW) + skb->csum = csum_sub(skb->csum, + csum_partial(skb->nh.raw, + (u8*)(fhdr + 1) - skb->nh.raw, + 0)); + + /* Is this the final fragment? */ + if (!(fhdr->frag_off & htons(IP6_MF))) { + /* If we already have some bits beyond end + * or have different end, the segment is corrupted. + */ + if (end < fq->len || + ((fq->last_in & LAST_IN) && end != fq->len)) { + DEBUGP("already received last fragment\n"); + goto err; + } + fq->last_in |= LAST_IN; + fq->len = end; + } else { + /* Check if the fragment is rounded to 8 bytes. + * Required by the RFC. + */ + if (end & 0x7) { + /* RFC2460 says always send parameter problem in + * this case. -DaveM + */ + DEBUGP("the end of this fragment is not rounded to 8 bytes.\n"); + return -1; + } + if (end > fq->len) { + /* Some bits beyond end -> corruption. */ + if (fq->last_in & LAST_IN) { + DEBUGP("last packet already reached.\n"); + goto err; + } + fq->len = end; + } + } + + if (end == offset) + goto err; + + /* Point into the IP datagram 'data' part. */ + if (!pskb_pull(skb, (u8 *) (fhdr + 1) - skb->data)) { + DEBUGP("queue: message is too short.\n"); + goto err; + } + if (end-offset < skb->len) { + if (pskb_trim(skb, end - offset)) { + DEBUGP("Can't trim\n"); + goto err; + } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_NONE; + } + + /* Find out which fragments are in front and at the back of us + * in the chain of fragments so far. We must know where to put + * this fragment, right? + */ + prev = NULL; + for (next = fq->fragments; next != NULL; next = next->next) { + if (NFCT_FRAG6_CB(next)->offset >= offset) + break; /* bingo! */ + prev = next; + } + + /* We found where to put this one. Check for overlap with + * preceding fragment, and, if needed, align things so that + * any overlaps are eliminated. + */ + if (prev) { + int i = (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset; + + if (i > 0) { + offset += i; + if (end <= offset) { + DEBUGP("overlap\n"); + goto err; + } + if (!pskb_pull(skb, i)) { + DEBUGP("Can't pull\n"); + goto err; + } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->ip_summed = CHECKSUM_NONE; + } + } + + /* Look for overlap with succeeding segments. + * If we can merge fragments, do it. + */ + while (next && NFCT_FRAG6_CB(next)->offset < end) { + /* overlap is 'i' bytes */ + int i = end - NFCT_FRAG6_CB(next)->offset; + + if (i < next->len) { + /* Eat head of the next overlapped fragment + * and leave the loop. The next ones cannot overlap. + */ + DEBUGP("Eat head of the overlapped parts.: %d", i); + if (!pskb_pull(next, i)) + goto err; + + /* next fragment */ + NFCT_FRAG6_CB(next)->offset += i; + fq->meat -= i; + if (next->ip_summed != CHECKSUM_UNNECESSARY) + next->ip_summed = CHECKSUM_NONE; + break; + } else { + struct sk_buff *free_it = next; + + /* Old fragmnet is completely overridden with + * new one drop it. + */ + next = next->next; + + if (prev) + prev->next = next; + else + fq->fragments = next; + + fq->meat -= free_it->len; + frag_kfree_skb(free_it, NULL); + } + } + + NFCT_FRAG6_CB(skb)->offset = offset; + + /* Insert this fragment in the chain of fragments. */ + skb->next = next; + if (prev) + prev->next = skb; + else + fq->fragments = skb; + + skb->dev = NULL; + skb_get_timestamp(skb, &fq->stamp); + fq->meat += skb->len; + atomic_add(skb->truesize, &nf_ct_frag6_mem); + + /* The first fragment. + * nhoffset is obtained from the first fragment, of course. + */ + if (offset == 0) { + fq->nhoffset = nhoff; + fq->last_in |= FIRST_IN; + } + write_lock(&nf_ct_frag6_lock); + list_move_tail(&fq->lru_list, &nf_ct_frag6_lru_list); + write_unlock(&nf_ct_frag6_lock); + return 0; + +err: + return -1; +} + +/* + * Check if this packet is complete. + * Returns NULL on failure by any reason, and pointer + * to current nexthdr field in reassembled frame. + * + * It is called with locked fq, and caller must check that + * queue is eligible for reassembly i.e. it is not COMPLETE, + * the last and the first frames arrived and all the bits are here. + */ +static struct sk_buff * +nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev) +{ + struct sk_buff *fp, *op, *head = fq->fragments; + int payload_len; + + fq_kill(fq); + + BUG_TRAP(head != NULL); + BUG_TRAP(NFCT_FRAG6_CB(head)->offset == 0); + + /* Unfragmented part is taken from the first segment. */ + payload_len = (head->data - head->nh.raw) - sizeof(struct ipv6hdr) + fq->len - sizeof(struct frag_hdr); + if (payload_len > IPV6_MAXPLEN) { + DEBUGP("payload len is too large.\n"); + goto out_oversize; + } + + /* Head of list must not be cloned. */ + if (skb_cloned(head) && pskb_expand_head(head, 0, 0, GFP_ATOMIC)) { + DEBUGP("skb is cloned but can't expand head"); + goto out_oom; + } + + /* If the first fragment is fragmented itself, we split + * it to two chunks: the first with data and paged part + * and the second, holding only fragments. */ + if (skb_shinfo(head)->frag_list) { + struct sk_buff *clone; + int i, plen = 0; + + if ((clone = alloc_skb(0, GFP_ATOMIC)) == NULL) { + DEBUGP("Can't alloc skb\n"); + goto out_oom; + } + clone->next = head->next; + head->next = clone; + skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list; + skb_shinfo(head)->frag_list = NULL; + for (i=0; i<skb_shinfo(head)->nr_frags; i++) + plen += skb_shinfo(head)->frags[i].size; + clone->len = clone->data_len = head->data_len - plen; + head->data_len -= clone->len; + head->len -= clone->len; + clone->csum = 0; + clone->ip_summed = head->ip_summed; + + NFCT_FRAG6_CB(clone)->orig = NULL; + atomic_add(clone->truesize, &nf_ct_frag6_mem); + } + + /* We have to remove fragment header from datagram and to relocate + * header in order to calculate ICV correctly. */ + head->nh.raw[fq->nhoffset] = head->h.raw[0]; + memmove(head->head + sizeof(struct frag_hdr), head->head, + (head->data - head->head) - sizeof(struct frag_hdr)); + head->mac.raw += sizeof(struct frag_hdr); + head->nh.raw += sizeof(struct frag_hdr); + + skb_shinfo(head)->frag_list = head->next; + head->h.raw = head->data; + skb_push(head, head->data - head->nh.raw); + atomic_sub(head->truesize, &nf_ct_frag6_mem); + + for (fp=head->next; fp; fp = fp->next) { + head->data_len += fp->len; + head->len += fp->len; + if (head->ip_summed != fp->ip_summed) + head->ip_summed = CHECKSUM_NONE; + else if (head->ip_summed == CHECKSUM_HW) + head->csum = csum_add(head->csum, fp->csum); + head->truesize += fp->truesize; + atomic_sub(fp->truesize, &nf_ct_frag6_mem); + } + + head->next = NULL; + head->dev = dev; + skb_set_timestamp(head, &fq->stamp); + head->nh.ipv6h->payload_len = htons(payload_len); + + /* Yes, and fold redundant checksum back. 8) */ + if (head->ip_summed == CHECKSUM_HW) + head->csum = csum_partial(head->nh.raw, head->h.raw-head->nh.raw, head->csum); + + fq->fragments = NULL; + + /* all original skbs are linked into the NFCT_FRAG6_CB(head).orig */ + fp = skb_shinfo(head)->frag_list; + if (NFCT_FRAG6_CB(fp)->orig == NULL) + /* at above code, head skb is divided into two skbs. */ + fp = fp->next; + + op = NFCT_FRAG6_CB(head)->orig; + for (; fp; fp = fp->next) { + struct sk_buff *orig = NFCT_FRAG6_CB(fp)->orig; + + op->next = orig; + op = orig; + NFCT_FRAG6_CB(fp)->orig = NULL; + } + + return head; + +out_oversize: + if (net_ratelimit()) + printk(KERN_DEBUG "nf_ct_frag6_reasm: payload len = %d\n", payload_len); + goto out_fail; +out_oom: + if (net_ratelimit()) + printk(KERN_DEBUG "nf_ct_frag6_reasm: no memory for reassembly\n"); +out_fail: + return NULL; +} + +/* + * find the header just before Fragment Header. + * + * if success return 0 and set ... + * (*prevhdrp): the value of "Next Header Field" in the header + * just before Fragment Header. + * (*prevhoff): the offset of "Next Header Field" in the header + * just before Fragment Header. + * (*fhoff) : the offset of Fragment Header. + * + * Based on ipv6_skip_hdr() in net/ipv6/exthdr.c + * + */ +static int +find_prev_fhdr(struct sk_buff *skb, u8 *prevhdrp, int *prevhoff, int *fhoff) +{ + u8 nexthdr = skb->nh.ipv6h->nexthdr; + u8 prev_nhoff = (u8 *)&skb->nh.ipv6h->nexthdr - skb->data; + int start = (u8 *)(skb->nh.ipv6h+1) - skb->data; + int len = skb->len - start; + u8 prevhdr = NEXTHDR_IPV6; + + while (nexthdr != NEXTHDR_FRAGMENT) { + struct ipv6_opt_hdr hdr; + int hdrlen; + + if (!ipv6_ext_hdr(nexthdr)) { + return -1; + } + if (len < (int)sizeof(struct ipv6_opt_hdr)) { + DEBUGP("too short\n"); + return -1; + } + if (nexthdr == NEXTHDR_NONE) { + DEBUGP("next header is none\n"); + return -1; + } + if (skb_copy_bits(skb, start, &hdr, sizeof(hdr))) + BUG(); + if (nexthdr == NEXTHDR_AUTH) + hdrlen = (hdr.hdrlen+2)<<2; + else + hdrlen = ipv6_optlen(&hdr); + + prevhdr = nexthdr; + prev_nhoff = start; + + nexthdr = hdr.nexthdr; + len -= hdrlen; + start += hdrlen; + } + + if (len < 0) + return -1; + + *prevhdrp = prevhdr; + *prevhoff = prev_nhoff; + *fhoff = start; + + return 0; +} + +struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb) +{ + struct sk_buff *clone; + struct net_device *dev = skb->dev; + struct frag_hdr *fhdr; + struct nf_ct_frag6_queue *fq; + struct ipv6hdr *hdr; + int fhoff, nhoff; + u8 prevhdr; + struct sk_buff *ret_skb = NULL; + + /* Jumbo payload inhibits frag. header */ + if (skb->nh.ipv6h->payload_len == 0) { + DEBUGP("payload len = 0\n"); + return skb; + } + + if (find_prev_fhdr(skb, &prevhdr, &nhoff, &fhoff) < 0) + return skb; + + clone = skb_clone(skb, GFP_ATOMIC); + if (clone == NULL) { + DEBUGP("Can't clone skb\n"); + return skb; + } + + NFCT_FRAG6_CB(clone)->orig = skb; + + if (!pskb_may_pull(clone, fhoff + sizeof(*fhdr))) { + DEBUGP("message is too short.\n"); + goto ret_orig; + } + + clone->h.raw = clone->data + fhoff; + hdr = clone->nh.ipv6h; + fhdr = (struct frag_hdr *)clone->h.raw; + + if (!(fhdr->frag_off & htons(0xFFF9))) { + DEBUGP("Invalid fragment offset\n"); + /* It is not a fragmented frame */ + goto ret_orig; + } + + if (atomic_read(&nf_ct_frag6_mem) > nf_ct_frag6_high_thresh) + nf_ct_frag6_evictor(); + + fq = fq_find(fhdr->identification, &hdr->saddr, &hdr->daddr); + if (fq == NULL) { + DEBUGP("Can't find and can't create new queue\n"); + goto ret_orig; + } + + spin_lock(&fq->lock); + + if (nf_ct_frag6_queue(fq, clone, fhdr, nhoff) < 0) { + spin_unlock(&fq->lock); + DEBUGP("Can't insert skb to queue\n"); + fq_put(fq, NULL); + goto ret_orig; + } + + if (fq->last_in == (FIRST_IN|LAST_IN) && fq->meat == fq->len) { + ret_skb = nf_ct_frag6_reasm(fq, dev); + if (ret_skb == NULL) + DEBUGP("Can't reassemble fragmented packets\n"); + } + spin_unlock(&fq->lock); + + fq_put(fq, NULL); + return ret_skb; + +ret_orig: + kfree_skb(clone); + return skb; +} + +void nf_ct_frag6_output(unsigned int hooknum, struct sk_buff *skb, + struct net_device *in, struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + struct sk_buff *s, *s2; + + for (s = NFCT_FRAG6_CB(skb)->orig; s;) { + nf_conntrack_put_reasm(s->nfct_reasm); + nf_conntrack_get_reasm(skb); + s->nfct_reasm = skb; + + s2 = s->next; + NF_HOOK_THRESH(PF_INET6, hooknum, s, in, out, okfn, + NF_IP6_PRI_CONNTRACK_DEFRAG + 1); + s = s2; + } + nf_conntrack_put_reasm(skb); +} + +int nf_ct_frag6_kfree_frags(struct sk_buff *skb) +{ + struct sk_buff *s, *s2; + + for (s = NFCT_FRAG6_CB(skb)->orig; s; s = s2) { + + s2 = s->next; + kfree_skb(s); + } + + kfree_skb(skb); + + return 0; +} + +int nf_ct_frag6_init(void) +{ + nf_ct_frag6_hash_rnd = (u32) ((num_physpages ^ (num_physpages>>7)) ^ + (jiffies ^ (jiffies >> 6))); + + init_timer(&nf_ct_frag6_secret_timer); + nf_ct_frag6_secret_timer.function = nf_ct_frag6_secret_rebuild; + nf_ct_frag6_secret_timer.expires = jiffies + + nf_ct_frag6_secret_interval; + add_timer(&nf_ct_frag6_secret_timer); + + return 0; +} + +void nf_ct_frag6_cleanup(void) +{ + del_timer(&nf_ct_frag6_secret_timer); + nf_ct_frag6_low_thresh = 0; + nf_ct_frag6_evictor(); +} diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 334a596..50a13e7 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -140,9 +140,7 @@ fold_field(void *mib[], int offt) unsigned long res = 0; int i; - for (i = 0; i < NR_CPUS; i++) { - if (!cpu_possible(i)) - continue; + for_each_cpu(i) { res += *(((unsigned long *)per_cpu_ptr(mib[0], i)) + offt); res += *(((unsigned long *)per_cpu_ptr(mib[1], i)) + offt); } diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index d77d335..a66900c 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -174,8 +174,10 @@ int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) struct sk_buff *clone = skb_clone(skb, GFP_ATOMIC); /* Not releasing hash table! */ - if (clone) + if (clone) { + nf_reset(clone); rawv6_rcv(sk, clone); + } } sk = __raw_v6_lookup(sk_next(sk), nexthdr, daddr, saddr, IP6CB(skb)->iif); @@ -296,13 +298,10 @@ void rawv6_err(struct sock *sk, struct sk_buff *skb, static inline int rawv6_rcv_skb(struct sock * sk, struct sk_buff * skb) { if ((raw6_sk(sk)->checksum || sk->sk_filter) && - skb->ip_summed != CHECKSUM_UNNECESSARY) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { - /* FIXME: increment a raw6 drops counter here */ - kfree_skb(skb); - return 0; - } - skb->ip_summed = CHECKSUM_UNNECESSARY; + skb_checksum_complete(skb)) { + /* FIXME: increment a raw6 drops counter here */ + kfree_skb(skb); + return 0; } /* Charge it to the socket. */ @@ -335,32 +334,25 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb) if (!rp->checksum) skb->ip_summed = CHECKSUM_UNNECESSARY; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) { - if (skb->ip_summed == CHECKSUM_HW) { - skb_postpull_rcsum(skb, skb->nh.raw, - skb->h.raw - skb->nh.raw); + if (skb->ip_summed == CHECKSUM_HW) { + skb_postpull_rcsum(skb, skb->nh.raw, + skb->h.raw - skb->nh.raw); + if (!csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, inet->num, skb->csum)) skb->ip_summed = CHECKSUM_UNNECESSARY; - if (csum_ipv6_magic(&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr, - skb->len, inet->num, skb->csum)) { - LIMIT_NETDEBUG(KERN_DEBUG "raw v6 hw csum failure.\n"); - skb->ip_summed = CHECKSUM_NONE; - } - } - if (skb->ip_summed == CHECKSUM_NONE) - skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr, - skb->len, inet->num, 0); } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) + skb->csum = ~csum_ipv6_magic(&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, + skb->len, inet->num, 0); if (inet->hdrincl) { - if (skb->ip_summed != CHECKSUM_UNNECESSARY && - (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { + if (skb_checksum_complete(skb)) { /* FIXME: increment a raw6 drops counter here */ kfree_skb(skb); return 0; } - skb->ip_summed = CHECKSUM_UNNECESSARY; } rawv6_rcv_skb(sk, skb); @@ -405,7 +397,7 @@ static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk, if (skb->ip_summed==CHECKSUM_UNNECESSARY) { err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied); } else if (msg->msg_flags&MSG_TRUNC) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) + if (__skb_checksum_complete(skb)) goto csum_copy_err; err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied); } else { diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index e4fe9ee..5d316cb 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -74,7 +74,7 @@ struct ip6frag_skb_cb struct frag_queue { - struct frag_queue *next; + struct hlist_node list; struct list_head lru_list; /* lru list member */ __u32 id; /* fragment id */ @@ -95,14 +95,13 @@ struct frag_queue #define FIRST_IN 2 #define LAST_IN 1 __u16 nhoffset; - struct frag_queue **pprev; }; /* Hash table. */ #define IP6Q_HASHSZ 64 -static struct frag_queue *ip6_frag_hash[IP6Q_HASHSZ]; +static struct hlist_head ip6_frag_hash[IP6Q_HASHSZ]; static DEFINE_RWLOCK(ip6_frag_lock); static u32 ip6_frag_hash_rnd; static LIST_HEAD(ip6_frag_lru_list); @@ -110,9 +109,7 @@ int ip6_frag_nqueues = 0; static __inline__ void __fq_unlink(struct frag_queue *fq) { - if(fq->next) - fq->next->pprev = fq->pprev; - *fq->pprev = fq->next; + hlist_del(&fq->list); list_del(&fq->lru_list); ip6_frag_nqueues--; } @@ -163,28 +160,21 @@ static void ip6_frag_secret_rebuild(unsigned long dummy) get_random_bytes(&ip6_frag_hash_rnd, sizeof(u32)); for (i = 0; i < IP6Q_HASHSZ; i++) { struct frag_queue *q; + struct hlist_node *p, *n; - q = ip6_frag_hash[i]; - while (q) { - struct frag_queue *next = q->next; + hlist_for_each_entry_safe(q, p, n, &ip6_frag_hash[i], list) { unsigned int hval = ip6qhashfn(q->id, &q->saddr, &q->daddr); if (hval != i) { - /* Unlink. */ - if (q->next) - q->next->pprev = q->pprev; - *q->pprev = q->next; + hlist_del(&q->list); /* Relink to new hash chain. */ - if ((q->next = ip6_frag_hash[hval]) != NULL) - q->next->pprev = &q->next; - ip6_frag_hash[hval] = q; - q->pprev = &ip6_frag_hash[hval]; - } + hlist_add_head(&q->list, + &ip6_frag_hash[hval]); - q = next; + } } } write_unlock(&ip6_frag_lock); @@ -337,10 +327,13 @@ static struct frag_queue *ip6_frag_intern(unsigned int hash, struct frag_queue *fq_in) { struct frag_queue *fq; +#ifdef CONFIG_SMP + struct hlist_node *n; +#endif write_lock(&ip6_frag_lock); #ifdef CONFIG_SMP - for (fq = ip6_frag_hash[hash]; fq; fq = fq->next) { + hlist_for_each_entry(fq, n, &ip6_frag_hash[hash], list) { if (fq->id == fq_in->id && ipv6_addr_equal(&fq_in->saddr, &fq->saddr) && ipv6_addr_equal(&fq_in->daddr, &fq->daddr)) { @@ -358,10 +351,7 @@ static struct frag_queue *ip6_frag_intern(unsigned int hash, atomic_inc(&fq->refcnt); atomic_inc(&fq->refcnt); - if((fq->next = ip6_frag_hash[hash]) != NULL) - fq->next->pprev = &fq->next; - ip6_frag_hash[hash] = fq; - fq->pprev = &ip6_frag_hash[hash]; + hlist_add_head(&fq->list, &ip6_frag_hash[hash]); INIT_LIST_HEAD(&fq->lru_list); list_add_tail(&fq->lru_list, &ip6_frag_lru_list); ip6_frag_nqueues++; @@ -401,10 +391,11 @@ static __inline__ struct frag_queue * fq_find(u32 id, struct in6_addr *src, struct in6_addr *dst) { struct frag_queue *fq; + struct hlist_node *n; unsigned int hash = ip6qhashfn(id, src, dst); read_lock(&ip6_frag_lock); - for(fq = ip6_frag_hash[hash]; fq; fq = fq->next) { + hlist_for_each_entry(fq, n, &ip6_frag_hash[hash], list) { if (fq->id == id && ipv6_addr_equal(src, &fq->saddr) && ipv6_addr_equal(dst, &fq->daddr)) { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 5d5bbb4..a7a537b 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -483,7 +483,7 @@ restart: goto out; } - rt = rt6_device_match(rt, skb->dev->ifindex, 0); + rt = rt6_device_match(rt, skb->dev->ifindex, strict); BACKTRACK(); if (!rt->rt6i_nexthop && !(rt->rt6i_flags & RTF_NONEXTHOP)) { @@ -1701,16 +1701,14 @@ static void fib6_dump_end(struct netlink_callback *cb) fib6_walker_unlink(w); kfree(w); } - if (cb->args[1]) { - cb->done = (void*)cb->args[1]; - cb->args[1] = 0; - } + cb->done = (void*)cb->args[1]; + cb->args[1] = 0; } static int fib6_dump_done(struct netlink_callback *cb) { fib6_dump_end(cb); - return cb->done(cb); + return cb->done ? cb->done(cb) : 0; } int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index d693cb9..62c0e5b 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -114,16 +114,9 @@ static int tcp_v6_get_port(struct sock *sk, unsigned short snum) int low = sysctl_local_port_range[0]; int high = sysctl_local_port_range[1]; int remaining = (high - low) + 1; - int rover; + int rover = net_random() % (high - low) + low; - spin_lock(&tcp_hashinfo.portalloc_lock); - if (tcp_hashinfo.port_rover < low) - rover = low; - else - rover = tcp_hashinfo.port_rover; - do { rover++; - if (rover > high) - rover = low; + do { head = &tcp_hashinfo.bhash[inet_bhashfn(rover, tcp_hashinfo.bhash_size)]; spin_lock(&head->lock); inet_bind_bucket_for_each(tb, node, &head->chain) @@ -132,9 +125,9 @@ static int tcp_v6_get_port(struct sock *sk, unsigned short snum) break; next: spin_unlock(&head->lock); + if (++rover > high) + rover = low; } while (--remaining > 0); - tcp_hashinfo.port_rover = rover; - spin_unlock(&tcp_hashinfo.portalloc_lock); /* Exhausted local port range during search? It is not * possible for us to be holding one of the bind hash @@ -1408,20 +1401,18 @@ out: static int tcp_v6_checksum_init(struct sk_buff *skb) { if (skb->ip_summed == CHECKSUM_HW) { - skb->ip_summed = CHECKSUM_UNNECESSARY; if (!tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr,skb->csum)) + &skb->nh.ipv6h->daddr,skb->csum)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; return 0; - LIMIT_NETDEBUG(KERN_DEBUG "hw tcp v6 csum failed\n"); + } } + + skb->csum = ~tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, + &skb->nh.ipv6h->daddr, 0); + if (skb->len <= 76) { - if (tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr,skb_checksum(skb, 0, skb->len, 0))) - return -1; - skb->ip_summed = CHECKSUM_UNNECESSARY; - } else { - skb->csum = ~tcp_v6_check(skb->h.th,skb->len,&skb->nh.ipv6h->saddr, - &skb->nh.ipv6h->daddr,0); + return __skb_checksum_complete(skb); } return 0; } @@ -1582,7 +1573,7 @@ static int tcp_v6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) goto discard_it; if ((skb->ip_summed != CHECKSUM_UNNECESSARY && - tcp_v6_checksum_init(skb) < 0)) + tcp_v6_checksum_init(skb))) goto bad_packet; th = skb->h.th; diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index e2b87cc..5cc8731 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -248,7 +248,7 @@ try_again: err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, copied); } else if (msg->msg_flags&MSG_TRUNC) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) + if (__skb_checksum_complete(skb)) goto csum_copy_err; err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov, copied); @@ -363,13 +363,10 @@ static inline int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) return -1; } - if (skb->ip_summed != CHECKSUM_UNNECESSARY) { - if ((unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) { - UDP6_INC_STATS_BH(UDP_MIB_INERRORS); - kfree_skb(skb); - return 0; - } - skb->ip_summed = CHECKSUM_UNNECESSARY; + if (skb_checksum_complete(skb)) { + UDP6_INC_STATS_BH(UDP_MIB_INERRORS); + kfree_skb(skb); + return 0; } if (sock_queue_rcv_skb(sk,skb)<0) { @@ -491,13 +488,10 @@ static int udpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) uh = skb->h.uh; } - if (skb->ip_summed==CHECKSUM_HW) { + if (skb->ip_summed == CHECKSUM_HW && + !csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum)) skb->ip_summed = CHECKSUM_UNNECESSARY; - if (csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, skb->csum)) { - LIMIT_NETDEBUG(KERN_DEBUG "udp v6 hw csum failure.\n"); - skb->ip_summed = CHECKSUM_NONE; - } - } + if (skb->ip_summed != CHECKSUM_UNNECESSARY) skb->csum = ~csum_ipv6_magic(saddr, daddr, ulen, IPPROTO_UDP, 0); @@ -521,8 +515,7 @@ static int udpv6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard; - if (skb->ip_summed != CHECKSUM_UNNECESSARY && - (unsigned short)csum_fold(skb_checksum(skb, 0, skb->len, skb->csum))) + if (skb_checksum_complete(skb)) goto discard; UDP6_INC_STATS_BH(UDP_MIB_NOPORTS); |