diff options
Diffstat (limited to 'net/ipv6')
-rw-r--r-- | net/ipv6/addrconf.c | 27 | ||||
-rw-r--r-- | net/ipv6/addrconf_core.c | 2 | ||||
-rw-r--r-- | net/ipv6/af_inet6.c | 3 | ||||
-rw-r--r-- | net/ipv6/icmp.c | 28 | ||||
-rw-r--r-- | net/ipv6/ila/ila_main.c | 5 | ||||
-rw-r--r-- | net/ipv6/ip6_fib.c | 18 | ||||
-rw-r--r-- | net/ipv6/ndisc.c | 8 | ||||
-rw-r--r-- | net/ipv6/output_core.c | 30 | ||||
-rw-r--r-- | net/ipv6/route.c | 554 | ||||
-rw-r--r-- | net/ipv6/seg6.c | 5 | ||||
-rw-r--r-- | net/ipv6/tcp_ipv6.c | 57 |
11 files changed, 366 insertions, 371 deletions
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4ae17a966ae3..2e8d1d2d8d3d 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -173,7 +173,8 @@ static int addrconf_ifdown(struct net_device *dev, int how); static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, int plen, const struct net_device *dev, - u32 flags, u32 noflags); + u32 flags, u32 noflags, + bool no_gw); static void addrconf_dad_start(struct inet6_ifaddr *ifp); static void addrconf_dad_work(struct work_struct *w); @@ -1230,10 +1231,8 @@ cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, bool del_r { struct fib6_info *f6i; - f6i = addrconf_get_prefix_route(&ifp->addr, - ifp->prefix_len, - ifp->idev->dev, - 0, RTF_GATEWAY | RTF_DEFAULT); + f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len, + ifp->idev->dev, 0, RTF_DEFAULT, true); if (f6i) { if (del_rt) ip6_del_rt(dev_net(ifp->idev->dev), f6i); @@ -2402,7 +2401,8 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, u32 metric, static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, int plen, const struct net_device *dev, - u32 flags, u32 noflags) + u32 flags, u32 noflags, + bool no_gw) { struct fib6_node *fn; struct fib6_info *rt = NULL; @@ -2419,7 +2419,9 @@ static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, goto out; for_each_fib6_node_rt_rcu(fn) { - if (rt->fib6_nh.nh_dev->ifindex != dev->ifindex) + if (rt->fib6_nh.fib_nh_dev->ifindex != dev->ifindex) + continue; + if (no_gw && rt->fib6_nh.fib_nh_has_gw) continue; if ((rt->fib6_flags & flags) != flags) continue; @@ -2717,7 +2719,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) pinfo->prefix_len, dev, RTF_ADDRCONF | RTF_PREFIX_RT, - RTF_GATEWAY | RTF_DEFAULT); + RTF_DEFAULT, true); if (rt) { /* Autoconf prefix route */ @@ -4588,10 +4590,8 @@ static int modify_prefix_route(struct inet6_ifaddr *ifp, struct fib6_info *f6i; u32 prio; - f6i = addrconf_get_prefix_route(&ifp->addr, - ifp->prefix_len, - ifp->idev->dev, - 0, RTF_GATEWAY | RTF_DEFAULT); + f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len, + ifp->idev->dev, 0, RTF_DEFAULT, true); if (!f6i) return -ENOENT; @@ -5972,7 +5972,8 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp) struct fib6_info *rt; rt = addrconf_get_prefix_route(&ifp->peer_addr, 128, - ifp->idev->dev, 0, 0); + ifp->idev->dev, 0, 0, + false); if (rt) ip6_del_rt(net, rt); } diff --git a/net/ipv6/addrconf_core.c b/net/ipv6/addrconf_core.c index 6c79af056d9b..945b66e3008f 100644 --- a/net/ipv6/addrconf_core.c +++ b/net/ipv6/addrconf_core.c @@ -5,7 +5,7 @@ #include <linux/export.h> #include <net/ipv6.h> -#include <net/addrconf.h> +#include <net/ipv6_stubs.h> #include <net/ip.h> /* if ipv6 module registers this function is used by xfrm to force all diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 2f45d2a3e3a3..1789bf99c419 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -56,6 +56,7 @@ #include <net/transp_v6.h> #include <net/ip6_route.h> #include <net/addrconf.h> +#include <net/ipv6_stubs.h> #include <net/ndisc.h> #ifdef CONFIG_IPV6_TUNNEL #include <net/ip6_tunnel.h> @@ -847,6 +848,8 @@ static int __net_init inet6_net_init(struct net *net) net->ipv6.sysctl.bindv6only = 0; net->ipv6.sysctl.icmpv6_time = 1*HZ; net->ipv6.sysctl.icmpv6_echo_ignore_all = 0; + net->ipv6.sysctl.icmpv6_echo_ignore_multicast = 0; + net->ipv6.sysctl.icmpv6_echo_ignore_anycast = 0; net->ipv6.sysctl.flowlabel_consistency = 1; net->ipv6.sysctl.auto_flowlabels = IP6_DEFAULT_AUTO_FLOW_LABELS; net->ipv6.sysctl.idgen_retries = 3; diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 802faa2fcc0e..cc14b9998941 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -683,12 +683,20 @@ static void icmpv6_echo_reply(struct sk_buff *skb) struct dst_entry *dst; struct ipcm6_cookie ipc6; u32 mark = IP6_REPLY_MARK(net, skb->mark); + bool acast; + + if (ipv6_addr_is_multicast(&ipv6_hdr(skb)->daddr) && + net->ipv6.sysctl.icmpv6_echo_ignore_multicast) + return; saddr = &ipv6_hdr(skb)->daddr; + acast = ipv6_anycast_destination(skb_dst(skb), saddr); + if (acast && net->ipv6.sysctl.icmpv6_echo_ignore_anycast) + return; + if (!ipv6_unicast_destination(skb) && - !(net->ipv6.sysctl.anycast_src_echo_reply && - ipv6_anycast_destination(skb_dst(skb), saddr))) + !(net->ipv6.sysctl.anycast_src_echo_reply && acast)) saddr = NULL; memcpy(&tmp_hdr, icmph, sizeof(tmp_hdr)); @@ -1115,6 +1123,20 @@ static struct ctl_table ipv6_icmp_table_template[] = { .mode = 0644, .proc_handler = proc_dointvec, }, + { + .procname = "echo_ignore_multicast", + .data = &init_net.ipv6.sysctl.icmpv6_echo_ignore_multicast, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "echo_ignore_anycast", + .data = &init_net.ipv6.sysctl.icmpv6_echo_ignore_anycast, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { }, }; @@ -1129,6 +1151,8 @@ struct ctl_table * __net_init ipv6_icmp_sysctl_init(struct net *net) if (table) { table[0].data = &net->ipv6.sysctl.icmpv6_time; table[1].data = &net->ipv6.sysctl.icmpv6_echo_ignore_all; + table[2].data = &net->ipv6.sysctl.icmpv6_echo_ignore_multicast; + table[3].data = &net->ipv6.sysctl.icmpv6_echo_ignore_anycast; } return table; } diff --git a/net/ipv6/ila/ila_main.c b/net/ipv6/ila/ila_main.c index 18fac76b9520..8d31a5066d0c 100644 --- a/net/ipv6/ila/ila_main.c +++ b/net/ipv6/ila/ila_main.c @@ -17,19 +17,16 @@ static const struct genl_ops ila_nl_ops[] = { { .cmd = ILA_CMD_ADD, .doit = ila_xlat_nl_cmd_add_mapping, - .policy = ila_nl_policy, .flags = GENL_ADMIN_PERM, }, { .cmd = ILA_CMD_DEL, .doit = ila_xlat_nl_cmd_del_mapping, - .policy = ila_nl_policy, .flags = GENL_ADMIN_PERM, }, { .cmd = ILA_CMD_FLUSH, .doit = ila_xlat_nl_cmd_flush, - .policy = ila_nl_policy, .flags = GENL_ADMIN_PERM, }, { @@ -38,7 +35,6 @@ static const struct genl_ops ila_nl_ops[] = { .start = ila_xlat_nl_dump_start, .dumpit = ila_xlat_nl_dump, .done = ila_xlat_nl_dump_done, - .policy = ila_nl_policy, }, }; @@ -49,6 +45,7 @@ struct genl_family ila_nl_family __ro_after_init = { .name = ILA_GENL_NAME, .version = ILA_GENL_VERSION, .maxattr = ILA_ATTR_MAX, + .policy = ila_nl_policy, .netnsok = true, .parallel_ops = true, .module = THIS_MODULE, diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 6613d8dbb0e5..8c00609a1513 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -199,10 +199,7 @@ void fib6_info_destroy_rcu(struct rcu_head *head) free_percpu(f6i->rt6i_pcpu); } - lwtstate_put(f6i->fib6_nh.nh_lwtstate); - - if (f6i->fib6_nh.nh_dev) - dev_put(f6i->fib6_nh.nh_dev); + fib6_nh_release(&f6i->fib6_nh); ip_fib_metrics_put(f6i->fib6_metrics); @@ -2297,6 +2294,7 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v) { struct fib6_info *rt = v; struct ipv6_route_iter *iter = seq->private; + unsigned int flags = rt->fib6_flags; const struct net_device *dev; seq_printf(seq, "%pi6 %02x ", &rt->fib6_dst.addr, rt->fib6_dst.plen); @@ -2306,15 +2304,17 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v) #else seq_puts(seq, "00000000000000000000000000000000 00 "); #endif - if (rt->fib6_flags & RTF_GATEWAY) - seq_printf(seq, "%pi6", &rt->fib6_nh.nh_gw); - else + if (rt->fib6_nh.fib_nh_has_gw) { + flags |= RTF_GATEWAY; + seq_printf(seq, "%pi6", &rt->fib6_nh.fib_nh_gw6); + } else { seq_puts(seq, "00000000000000000000000000000000"); + } - dev = rt->fib6_nh.nh_dev; + dev = rt->fib6_nh.fib_nh_dev; seq_printf(seq, " %08x %08x %08x %08x %8s\n", rt->fib6_metric, atomic_read(&rt->fib6_ref), 0, - rt->fib6_flags, dev ? dev->name : ""); + flags, dev ? dev->name : ""); iter->w.leaf = NULL; return 0; } diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 659ecf4e4b3c..66c8b294e02b 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1276,8 +1276,8 @@ static void ndisc_router_discovery(struct sk_buff *skb) rt = rt6_get_dflt_router(net, &ipv6_hdr(skb)->saddr, skb->dev); if (rt) { - neigh = ip6_neigh_lookup(&rt->fib6_nh.nh_gw, - rt->fib6_nh.nh_dev, NULL, + neigh = ip6_neigh_lookup(&rt->fib6_nh.fib_nh_gw6, + rt->fib6_nh.fib_nh_dev, NULL, &ipv6_hdr(skb)->saddr); if (!neigh) { ND_PRINTK(0, err, @@ -1306,8 +1306,8 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } - neigh = ip6_neigh_lookup(&rt->fib6_nh.nh_gw, - rt->fib6_nh.nh_dev, NULL, + neigh = ip6_neigh_lookup(&rt->fib6_nh.fib_nh_gw6, + rt->fib6_nh.fib_nh_dev, NULL, &ipv6_hdr(skb)->saddr); if (!neigh) { ND_PRINTK(0, err, diff --git a/net/ipv6/output_core.c b/net/ipv6/output_core.c index 4fe7c90962dd..868ae23dbae1 100644 --- a/net/ipv6/output_core.c +++ b/net/ipv6/output_core.c @@ -10,15 +10,25 @@ #include <net/secure_seq.h> #include <linux/netfilter.h> -static u32 __ipv6_select_ident(struct net *net, u32 hashrnd, +static u32 __ipv6_select_ident(struct net *net, const struct in6_addr *dst, const struct in6_addr *src) { + const struct { + struct in6_addr dst; + struct in6_addr src; + } __aligned(SIPHASH_ALIGNMENT) combined = { + .dst = *dst, + .src = *src, + }; u32 hash, id; - hash = __ipv6_addr_jhash(dst, hashrnd); - hash = __ipv6_addr_jhash(src, hash); - hash ^= net_hash_mix(net); + /* Note the following code is not safe, but this is okay. */ + if (unlikely(siphash_key_is_zero(&net->ipv4.ip_id_key))) + get_random_bytes(&net->ipv4.ip_id_key, + sizeof(net->ipv4.ip_id_key)); + + hash = siphash(&combined, sizeof(combined), &net->ipv4.ip_id_key); /* Treat id of 0 as unset and if we get 0 back from ip_idents_reserve, * set the hight order instead thus minimizing possible future @@ -41,7 +51,6 @@ static u32 __ipv6_select_ident(struct net *net, u32 hashrnd, */ __be32 ipv6_proxy_select_ident(struct net *net, struct sk_buff *skb) { - static u32 ip6_proxy_idents_hashrnd __read_mostly; struct in6_addr buf[2]; struct in6_addr *addrs; u32 id; @@ -53,11 +62,7 @@ __be32 ipv6_proxy_select_ident(struct net *net, struct sk_buff *skb) if (!addrs) return 0; - net_get_random_once(&ip6_proxy_idents_hashrnd, - sizeof(ip6_proxy_idents_hashrnd)); - - id = __ipv6_select_ident(net, ip6_proxy_idents_hashrnd, - &addrs[1], &addrs[0]); + id = __ipv6_select_ident(net, &addrs[1], &addrs[0]); return htonl(id); } EXPORT_SYMBOL_GPL(ipv6_proxy_select_ident); @@ -66,12 +71,9 @@ __be32 ipv6_select_ident(struct net *net, const struct in6_addr *daddr, const struct in6_addr *saddr) { - static u32 ip6_idents_hashrnd __read_mostly; u32 id; - net_get_random_once(&ip6_idents_hashrnd, sizeof(ip6_idents_hashrnd)); - - id = __ipv6_select_ident(net, ip6_idents_hashrnd, daddr, saddr); + id = __ipv6_select_ident(net, daddr, saddr); return htonl(id); } EXPORT_SYMBOL(ipv6_select_ident); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 0302e0eb07af..6e89151693d0 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -441,14 +441,14 @@ struct fib6_info *fib6_multipath_select(const struct net *net, if (!fl6->mp_hash) fl6->mp_hash = rt6_multipath_hash(net, fl6, skb, NULL); - if (fl6->mp_hash <= atomic_read(&match->fib6_nh.nh_upper_bound)) + if (fl6->mp_hash <= atomic_read(&match->fib6_nh.fib_nh_upper_bound)) return match; list_for_each_entry_safe(sibling, next_sibling, &match->fib6_siblings, fib6_siblings) { int nh_upper_bound; - nh_upper_bound = atomic_read(&sibling->fib6_nh.nh_upper_bound); + nh_upper_bound = atomic_read(&sibling->fib6_nh.fib_nh_upper_bound); if (fl6->mp_hash > nh_upper_bound) continue; if (rt6_score_route(sibling, oif, strict) < 0) @@ -473,13 +473,13 @@ static inline struct fib6_info *rt6_device_match(struct net *net, struct fib6_info *sprt; if (!oif && ipv6_addr_any(saddr) && - !(rt->fib6_nh.nh_flags & RTNH_F_DEAD)) + !(rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD)) return rt; for (sprt = rt; sprt; sprt = rcu_dereference(sprt->fib6_next)) { - const struct net_device *dev = sprt->fib6_nh.nh_dev; + const struct net_device *dev = sprt->fib6_nh.fib_nh_dev; - if (sprt->fib6_nh.nh_flags & RTNH_F_DEAD) + if (sprt->fib6_nh.fib_nh_flags & RTNH_F_DEAD) continue; if (oif) { @@ -495,7 +495,7 @@ static inline struct fib6_info *rt6_device_match(struct net *net, if (oif && flags & RT6_LOOKUP_F_IFACE) return net->ipv6.fib6_null_entry; - return rt->fib6_nh.nh_flags & RTNH_F_DEAD ? net->ipv6.fib6_null_entry : rt; + return rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD ? net->ipv6.fib6_null_entry : rt; } #ifdef CONFIG_IPV6_ROUTER_PREF @@ -533,11 +533,11 @@ static void rt6_probe(struct fib6_info *rt) * Router Reachability Probe MUST be rate-limited * to no more than one per minute. */ - if (!rt || !(rt->fib6_flags & RTF_GATEWAY)) + if (!rt || !rt->fib6_nh.fib_nh_has_gw) return; - nh_gw = &rt->fib6_nh.nh_gw; - dev = rt->fib6_nh.nh_dev; + nh_gw = &rt->fib6_nh.fib_nh_gw6; + dev = rt->fib6_nh.fib_nh_dev; rcu_read_lock_bh(); idev = __in6_dev_get(dev); neigh = __ipv6_neigh_lookup_noref(dev, nh_gw); @@ -582,7 +582,7 @@ static inline void rt6_probe(struct fib6_info *rt) */ static inline int rt6_check_dev(struct fib6_info *rt, int oif) { - const struct net_device *dev = rt->fib6_nh.nh_dev; + const struct net_device *dev = rt->fib6_nh.fib_nh_dev; if (!oif || dev->ifindex == oif) return 2; @@ -595,12 +595,12 @@ static inline enum rt6_nud_state rt6_check_neigh(struct fib6_info *rt) struct neighbour *neigh; if (rt->fib6_flags & RTF_NONEXTHOP || - !(rt->fib6_flags & RTF_GATEWAY)) + !rt->fib6_nh.fib_nh_has_gw) return RT6_NUD_SUCCEED; rcu_read_lock_bh(); - neigh = __ipv6_neigh_lookup_noref(rt->fib6_nh.nh_dev, - &rt->fib6_nh.nh_gw); + neigh = __ipv6_neigh_lookup_noref(rt->fib6_nh.fib_nh_dev, + &rt->fib6_nh.fib_nh_gw6); if (neigh) { read_lock(&neigh->lock); if (neigh->nud_state & NUD_VALID) @@ -639,21 +639,6 @@ static int rt6_score_route(struct fib6_info *rt, int oif, int strict) return m; } -/* called with rc_read_lock held */ -static inline bool fib6_ignore_linkdown(const struct fib6_info *f6i) -{ - const struct net_device *dev = fib6_info_nh_dev(f6i); - bool rc = false; - - if (dev) { - const struct inet6_dev *idev = __in6_dev_get(dev); - - rc = !!idev->cnf.ignore_routes_with_linkdown; - } - - return rc; -} - static struct fib6_info *find_match(struct fib6_info *rt, int oif, int strict, int *mpri, struct fib6_info *match, bool *do_rr) @@ -661,11 +646,11 @@ static struct fib6_info *find_match(struct fib6_info *rt, int oif, int strict, int m; bool match_do_rr = false; - if (rt->fib6_nh.nh_flags & RTNH_F_DEAD) + if (rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD) goto out; - if (fib6_ignore_linkdown(rt) && - rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN && + if (ip6_ignore_linkdown(rt->fib6_nh.fib_nh_dev) && + rt->fib6_nh.fib_nh_flags & RTNH_F_LINKDOWN && !(strict & RT6_LOOKUP_F_IGNORE_LINKSTATE)) goto out; @@ -784,7 +769,7 @@ static struct fib6_info *rt6_select(struct net *net, struct fib6_node *fn, static bool rt6_is_gw_or_nonexthop(const struct fib6_info *rt) { - return (rt->fib6_flags & (RTF_NONEXTHOP | RTF_GATEWAY)); + return (rt->fib6_flags & RTF_NONEXTHOP) || rt->fib6_nh.fib_nh_has_gw; } #ifdef CONFIG_IPV6_ROUTE_INFO @@ -870,7 +855,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, /* called with rcu_lock held */ static struct net_device *ip6_rt_get_dev_rcu(struct fib6_info *rt) { - struct net_device *dev = rt->fib6_nh.nh_dev; + struct net_device *dev = rt->fib6_nh.fib_nh_dev; if (rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST)) { /* for copies of local routes, dst->dev needs to be the @@ -964,8 +949,8 @@ static void ip6_rt_init_dst(struct rt6_info *rt, struct fib6_info *ort) rt->dst.input = ip6_forward; } - if (ort->fib6_nh.nh_lwtstate) { - rt->dst.lwtstate = lwtstate_get(ort->fib6_nh.nh_lwtstate); + if (ort->fib6_nh.fib_nh_lws) { + rt->dst.lwtstate = lwtstate_get(ort->fib6_nh.fib_nh_lws); lwtunnel_set_redirect(&rt->dst); } @@ -989,8 +974,11 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct fib6_info *ort) rt->rt6i_dst = ort->fib6_dst; rt->rt6i_idev = dev ? in6_dev_get(dev) : NULL; - rt->rt6i_gateway = ort->fib6_nh.nh_gw; rt->rt6i_flags = ort->fib6_flags; + if (ort->fib6_nh.fib_nh_has_gw) { + rt->rt6i_gateway = ort->fib6_nh.fib_nh_gw6; + rt->rt6i_flags |= RTF_GATEWAY; + } rt6_set_from(rt, ort); #ifdef CONFIG_IPV6_SUBTREES rt->rt6i_src = ort->fib6_src; @@ -1015,14 +1003,13 @@ static struct fib6_node* fib6_backtrack(struct fib6_node *fn, } } -static bool ip6_hold_safe(struct net *net, struct rt6_info **prt, - bool null_fallback) +static bool ip6_hold_safe(struct net *net, struct rt6_info **prt) { struct rt6_info *rt = *prt; if (dst_hold_safe(&rt->dst)) return true; - if (null_fallback) { + if (net) { rt = net->ipv6.ip6_null_entry; dst_hold(&rt->dst); } else { @@ -1036,7 +1023,7 @@ static bool ip6_hold_safe(struct net *net, struct rt6_info **prt, static struct rt6_info *ip6_create_rt_rcu(struct fib6_info *rt) { unsigned short flags = fib6_info_dst_flags(rt); - struct net_device *dev = rt->fib6_nh.nh_dev; + struct net_device *dev = rt->fib6_nh.fib_nh_dev; struct rt6_info *nrt; if (!fib6_info_hold_safe(rt)) @@ -1095,7 +1082,7 @@ restart: /* Search through exception table */ rt = rt6_find_cached_rt(f6i, &fl6->daddr, &fl6->saddr); if (rt) { - if (ip6_hold_safe(net, &rt, true)) + if (ip6_hold_safe(net, &rt)) dst_use_noref(&rt->dst, jiffies); } else if (f6i == net->ipv6.fib6_null_entry) { rt = net->ipv6.ip6_null_entry; @@ -1242,7 +1229,7 @@ static struct rt6_info *rt6_get_pcpu_route(struct fib6_info *rt) pcpu_rt = *p; if (pcpu_rt) - ip6_hold_safe(NULL, &pcpu_rt, false); + ip6_hold_safe(NULL, &pcpu_rt); return pcpu_rt; } @@ -1420,7 +1407,7 @@ static unsigned int fib6_mtu(const struct fib6_info *rt) mtu = min_t(unsigned int, mtu, IP6_MAX_MTU); - return mtu - lwtunnel_headroom(rt->fib6_nh.nh_lwtstate, mtu); + return mtu - lwtunnel_headroom(rt->fib6_nh.fib_nh_lws, mtu); } static int rt6_insert_exception(struct rt6_info *nrt, @@ -1867,13 +1854,13 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, /*Search through exception table */ rt = rt6_find_cached_rt(f6i, &fl6->daddr, &fl6->saddr); if (rt) { - if (ip6_hold_safe(net, &rt, true)) + if (ip6_hold_safe(net, &rt)) dst_use_noref(&rt->dst, jiffies); rcu_read_unlock(); return rt; } else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) && - !(f6i->fib6_flags & RTF_GATEWAY))) { + !f6i->fib6_nh.fib_nh_has_gw)) { /* Create a RTF_CACHE clone which will not be * owned by the fib6 tree. It is for the special case where * the daddr in the skb during the neighbor look-up is different @@ -2437,22 +2424,22 @@ static struct rt6_info *__ip6_route_redirect(struct net *net, fn = fib6_node_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr); restart: for_each_fib6_node_rt_rcu(fn) { - if (rt->fib6_nh.nh_flags & RTNH_F_DEAD) + if (rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD) continue; if (fib6_check_expired(rt)) continue; if (rt->fib6_flags & RTF_REJECT) break; - if (!(rt->fib6_flags & RTF_GATEWAY)) + if (!rt->fib6_nh.fib_nh_has_gw) continue; - if (fl6->flowi6_oif != rt->fib6_nh.nh_dev->ifindex) + if (fl6->flowi6_oif != rt->fib6_nh.fib_nh_dev->ifindex) continue; /* rt_cache's gateway might be different from its 'parent' * in the case of an ip redirect. * So we keep searching in the exception table if the gateway * is different. */ - if (!ipv6_addr_equal(&rdfl->gateway, &rt->fib6_nh.nh_gw)) { + if (!ipv6_addr_equal(&rdfl->gateway, &rt->fib6_nh.fib_nh_gw6)) { rt_cache = rt6_find_cached_rt(rt, &fl6->daddr, &fl6->saddr); @@ -2482,7 +2469,7 @@ restart: out: if (ret) - ip6_hold_safe(net, &ret, true); + ip6_hold_safe(net, &ret); else ret = ip6_create_rt_rcu(rt); @@ -2899,17 +2886,143 @@ out: return err; } +static bool fib6_is_reject(u32 flags, struct net_device *dev, int addr_type) +{ + if ((flags & RTF_REJECT) || + (dev && (dev->flags & IFF_LOOPBACK) && + !(addr_type & IPV6_ADDR_LOOPBACK) && + !(flags & RTF_LOCAL))) + return true; + + return false; +} + +int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh, + struct fib6_config *cfg, gfp_t gfp_flags, + struct netlink_ext_ack *extack) +{ + struct net_device *dev = NULL; + struct inet6_dev *idev = NULL; + int addr_type; + int err; + + fib6_nh->fib_nh_family = AF_INET6; + + err = -ENODEV; + if (cfg->fc_ifindex) { + dev = dev_get_by_index(net, cfg->fc_ifindex); + if (!dev) + goto out; + idev = in6_dev_get(dev); + if (!idev) + goto out; + } + + if (cfg->fc_flags & RTNH_F_ONLINK) { + if (!dev) { + NL_SET_ERR_MSG(extack, + "Nexthop device required for onlink"); + goto out; + } + + if (!(dev->flags & IFF_UP)) { + NL_SET_ERR_MSG(extack, "Nexthop device is not up"); + err = -ENETDOWN; + goto out; + } + + fib6_nh->fib_nh_flags |= RTNH_F_ONLINK; + } + + fib6_nh->fib_nh_weight = 1; + + /* We cannot add true routes via loopback here, + * they would result in kernel looping; promote them to reject routes + */ + addr_type = ipv6_addr_type(&cfg->fc_dst); + if (fib6_is_reject(cfg->fc_flags, dev, addr_type)) { + /* hold loopback dev/idev if we haven't done so. */ + if (dev != net->loopback_dev) { + if (dev) { + dev_put(dev); + in6_dev_put(idev); + } + dev = net->loopback_dev; + dev_hold(dev); + idev = in6_dev_get(dev); + if (!idev) { + err = -ENODEV; + goto out; + } + } + goto set_dev; + } + + if (cfg->fc_flags & RTF_GATEWAY) { + err = ip6_validate_gw(net, cfg, &dev, &idev, extack); + if (err) + goto out; + + fib6_nh->fib_nh_gw6 = cfg->fc_gateway; + fib6_nh->fib_nh_has_gw = 1; + } + + err = -ENODEV; + if (!dev) + goto out; + + if (idev->cnf.disable_ipv6) { + NL_SET_ERR_MSG(extack, "IPv6 is disabled on nexthop device"); + err = -EACCES; + goto out; + } + + if (!(dev->flags & IFF_UP) && !cfg->fc_ignore_dev_down) { + NL_SET_ERR_MSG(extack, "Nexthop device is not up"); + err = -ENETDOWN; + goto out; + } + + if (!(cfg->fc_flags & (RTF_LOCAL | RTF_ANYCAST)) && + !netif_carrier_ok(dev)) + fib6_nh->fib_nh_flags |= RTNH_F_LINKDOWN; + + err = fib_nh_common_init(&fib6_nh->nh_common, cfg->fc_encap, + cfg->fc_encap_type, cfg, gfp_flags, extack); + if (err) + goto out; +set_dev: + fib6_nh->fib_nh_dev = dev; + fib6_nh->fib_nh_oif = dev->ifindex; + err = 0; +out: + if (idev) + in6_dev_put(idev); + + if (err) { + lwtstate_put(fib6_nh->fib_nh_lws); + fib6_nh->fib_nh_lws = NULL; + if (dev) + dev_put(dev); + } + + return err; +} + +void fib6_nh_release(struct fib6_nh *fib6_nh) +{ + fib_nh_common_release(&fib6_nh->nh_common); +} + static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, gfp_t gfp_flags, struct netlink_ext_ack *extack) { struct net *net = cfg->fc_nlinfo.nl_net; struct fib6_info *rt = NULL; - struct net_device *dev = NULL; - struct inet6_dev *idev = NULL; struct fib6_table *table; - int addr_type; int err = -EINVAL; + int addr_type; /* RTF_PCPU is an internal flag; can not be set by userspace */ if (cfg->fc_flags & RTF_PCPU) { @@ -2943,33 +3056,6 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, goto out; } #endif - if (cfg->fc_ifindex) { - err = -ENODEV; - dev = dev_get_by_index(net, cfg->fc_ifindex); - if (!dev) - goto out; - idev = in6_dev_get(dev); - if (!idev) - goto out; - } - - if (cfg->fc_metric == 0) - cfg->fc_metric = IP6_RT_PRIO_USER; - - if (cfg->fc_flags & RTNH_F_ONLINK) { - if (!dev) { - NL_SET_ERR_MSG(extack, - "Nexthop device required for onlink"); - err = -ENODEV; - goto out; - } - - if (!(dev->flags & IFF_UP)) { - NL_SET_ERR_MSG(extack, "Nexthop device is not up"); - err = -ENETDOWN; - goto out; - } - } err = -ENOBUFS; if (cfg->fc_nlinfo.nlh && @@ -3013,18 +3099,10 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, cfg->fc_protocol = RTPROT_BOOT; rt->fib6_protocol = cfg->fc_protocol; - addr_type = ipv6_addr_type(&cfg->fc_dst); - - if (cfg->fc_encap) { - struct lwtunnel_state *lwtstate; - - err = lwtunnel_build_state(cfg->fc_encap_type, - cfg->fc_encap, AF_INET6, cfg, - &lwtstate, extack); - if (err) - goto out; - rt->fib6_nh.nh_lwtstate = lwtstate_get(lwtstate); - } + rt->fib6_table = table; + rt->fib6_metric = cfg->fc_metric; + rt->fib6_type = cfg->fc_type; + rt->fib6_flags = cfg->fc_flags & ~RTF_GATEWAY; ipv6_addr_prefix(&rt->fib6_dst.addr, &cfg->fc_dst, cfg->fc_dst_len); rt->fib6_dst.plen = cfg->fc_dst_len; @@ -3035,62 +3113,20 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, ipv6_addr_prefix(&rt->fib6_src.addr, &cfg->fc_src, cfg->fc_src_len); rt->fib6_src.plen = cfg->fc_src_len; #endif - - rt->fib6_metric = cfg->fc_metric; - rt->fib6_nh.nh_weight = 1; - - rt->fib6_type = cfg->fc_type; + err = fib6_nh_init(net, &rt->fib6_nh, cfg, gfp_flags, extack); + if (err) + goto out; /* We cannot add true routes via loopback here, - they would result in kernel looping; promote them to reject routes + * they would result in kernel looping; promote them to reject routes */ - if ((cfg->fc_flags & RTF_REJECT) || - (dev && (dev->flags & IFF_LOOPBACK) && - !(addr_type & IPV6_ADDR_LOOPBACK) && - !(cfg->fc_flags & RTF_LOCAL))) { - /* hold loopback dev/idev if we haven't done so. */ - if (dev != net->loopback_dev) { - if (dev) { - dev_put(dev); - in6_dev_put(idev); - } - dev = net->loopback_dev; - dev_hold(dev); - idev = in6_dev_get(dev); - if (!idev) { - err = -ENODEV; - goto out; - } - } - rt->fib6_flags = RTF_REJECT|RTF_NONEXTHOP; - goto install_route; - } - - if (cfg->fc_flags & RTF_GATEWAY) { - err = ip6_validate_gw(net, cfg, &dev, &idev, extack); - if (err) - goto out; - - rt->fib6_nh.nh_gw = cfg->fc_gateway; - } - - err = -ENODEV; - if (!dev) - goto out; - - if (idev->cnf.disable_ipv6) { - NL_SET_ERR_MSG(extack, "IPv6 is disabled on nexthop device"); - err = -EACCES; - goto out; - } - - if (!(dev->flags & IFF_UP)) { - NL_SET_ERR_MSG(extack, "Nexthop device is not up"); - err = -ENETDOWN; - goto out; - } + addr_type = ipv6_addr_type(&cfg->fc_dst); + if (fib6_is_reject(cfg->fc_flags, rt->fib6_nh.fib_nh_dev, addr_type)) + rt->fib6_flags = RTF_REJECT | RTF_NONEXTHOP; if (!ipv6_addr_any(&cfg->fc_prefsrc)) { + struct net_device *dev = fib6_info_nh_dev(rt); + if (!ipv6_chk_addr(net, &cfg->fc_prefsrc, dev, 0)) { NL_SET_ERR_MSG(extack, "Invalid source address"); err = -EINVAL; @@ -3101,26 +3137,8 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, } else rt->fib6_prefsrc.plen = 0; - rt->fib6_flags = cfg->fc_flags; - -install_route: - if (!(rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST)) && - !netif_carrier_ok(dev)) - rt->fib6_nh.nh_flags |= RTNH_F_LINKDOWN; - rt->fib6_nh.nh_flags |= (cfg->fc_flags & RTNH_F_ONLINK); - rt->fib6_nh.nh_dev = dev; - rt->fib6_table = table; - - if (idev) - in6_dev_put(idev); - return rt; out: - if (dev) - dev_put(dev); - if (idev) - in6_dev_put(idev); - fib6_info_release(rt); return ERR_PTR(err); } @@ -3261,6 +3279,8 @@ static int ip6_route_del(struct fib6_config *cfg, if (fn) { for_each_fib6_node_rt_rcu(fn) { + struct fib6_nh *nh; + if (cfg->fc_flags & RTF_CACHE) { int rc; @@ -3275,12 +3295,14 @@ static int ip6_route_del(struct fib6_config *cfg, } continue; } + + nh = &rt->fib6_nh; if (cfg->fc_ifindex && - (!rt->fib6_nh.nh_dev || - rt->fib6_nh.nh_dev->ifindex != cfg->fc_ifindex)) + (!nh->fib_nh_dev || + nh->fib_nh_dev->ifindex != cfg->fc_ifindex)) continue; if (cfg->fc_flags & RTF_GATEWAY && - !ipv6_addr_equal(&cfg->fc_gateway, &rt->fib6_nh.nh_gw)) + !ipv6_addr_equal(&cfg->fc_gateway, &nh->fib_nh_gw6)) continue; if (cfg->fc_metric && cfg->fc_metric != rt->fib6_metric) continue; @@ -3451,11 +3473,12 @@ static struct fib6_info *rt6_get_route_info(struct net *net, goto out; for_each_fib6_node_rt_rcu(fn) { - if (rt->fib6_nh.nh_dev->ifindex != ifindex) + if (rt->fib6_nh.fib_nh_dev->ifindex != ifindex) continue; - if ((rt->fib6_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY)) + if (!(rt->fib6_flags & RTF_ROUTEINFO) || + !rt->fib6_nh.fib_nh_has_gw) continue; - if (!ipv6_addr_equal(&rt->fib6_nh.nh_gw, gwaddr)) + if (!ipv6_addr_equal(&rt->fib6_nh.fib_nh_gw6, gwaddr)) continue; if (!fib6_info_hold_safe(rt)) continue; @@ -3513,9 +3536,11 @@ struct fib6_info *rt6_get_dflt_router(struct net *net, rcu_read_lock(); for_each_fib6_node_rt_rcu(&table->tb6_root) { - if (dev == rt->fib6_nh.nh_dev && + struct fib6_nh *nh = &rt->fib6_nh; + + if (dev == nh->fib_nh_dev && ((rt->fib6_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) && - ipv6_addr_equal(&rt->fib6_nh.nh_gw, addr)) + ipv6_addr_equal(&nh->fib_nh_gw6, addr)) break; } if (rt && !fib6_info_hold_safe(rt)) @@ -3606,7 +3631,7 @@ static void rtmsg_to_fib6_config(struct net *net, .fc_table = l3mdev_fib_table_by_index(net, rtmsg->rtmsg_ifindex) ? : RT6_TABLE_MAIN, .fc_ifindex = rtmsg->rtmsg_ifindex, - .fc_metric = rtmsg->rtmsg_metric, + .fc_metric = rtmsg->rtmsg_metric ? : IP6_RT_PRIO_USER, .fc_expires = rtmsg->rtmsg_info, .fc_dst_len = rtmsg->rtmsg_dst_len, .fc_src_len = rtmsg->rtmsg_src_len, @@ -3717,36 +3742,26 @@ struct fib6_info *addrconf_f6i_alloc(struct net *net, const struct in6_addr *addr, bool anycast, gfp_t gfp_flags) { - u32 tb_id; - struct net_device *dev = idev->dev; - struct fib6_info *f6i; - - f6i = fib6_info_alloc(gfp_flags); - if (!f6i) - return ERR_PTR(-ENOMEM); + struct fib6_config cfg = { + .fc_table = l3mdev_fib_table(idev->dev) ? : RT6_TABLE_LOCAL, + .fc_ifindex = idev->dev->ifindex, + .fc_flags = RTF_UP | RTF_ADDRCONF | RTF_NONEXTHOP, + .fc_dst = *addr, + .fc_dst_len = 128, + .fc_protocol = RTPROT_KERNEL, + .fc_nlinfo.nl_net = net, + .fc_ignore_dev_down = true, + }; - f6i->fib6_metrics = ip_fib_metrics_init(net, NULL, 0, NULL); - f6i->dst_nocount = true; - f6i->dst_host = true; - f6i->fib6_protocol = RTPROT_KERNEL; - f6i->fib6_flags = RTF_UP | RTF_NONEXTHOP; if (anycast) { - f6i->fib6_type = RTN_ANYCAST; - f6i->fib6_flags |= RTF_ANYCAST; + cfg.fc_type = RTN_ANYCAST; + cfg.fc_flags |= RTF_ANYCAST; } else { - f6i->fib6_type = RTN_LOCAL; - f6i->fib6_flags |= RTF_LOCAL; + cfg.fc_type = RTN_LOCAL; + cfg.fc_flags |= RTF_LOCAL; } - f6i->fib6_nh.nh_gw = *addr; - dev_hold(dev); - f6i->fib6_nh.nh_dev = dev; - f6i->fib6_dst.addr = *addr; - f6i->fib6_dst.plen = 128; - tb_id = l3mdev_fib_table(idev->dev) ? : RT6_TABLE_LOCAL; - f6i->fib6_table = fib6_get_table(net, tb_id); - - return f6i; + return ip6_route_info_create(&cfg, gfp_flags, NULL); } /* remove deleted ip from prefsrc entries */ @@ -3762,7 +3777,7 @@ static int fib6_remove_prefsrc(struct fib6_info *rt, void *arg) struct net *net = ((struct arg_dev_net_ip *)arg)->net; struct in6_addr *addr = ((struct arg_dev_net_ip *)arg)->addr; - if (((void *)rt->fib6_nh.nh_dev == dev || !dev) && + if (((void *)rt->fib6_nh.fib_nh_dev == dev || !dev) && rt != net->ipv6.fib6_null_entry && ipv6_addr_equal(addr, &rt->fib6_prefsrc.addr)) { spin_lock_bh(&rt6_exception_lock); @@ -3784,7 +3799,7 @@ void rt6_remove_prefsrc(struct inet6_ifaddr *ifp) fib6_clean_all(net, fib6_remove_prefsrc, &adni); } -#define RTF_RA_ROUTER (RTF_ADDRCONF | RTF_DEFAULT | RTF_GATEWAY) +#define RTF_RA_ROUTER (RTF_ADDRCONF | RTF_DEFAULT) /* Remove routers and update dst entries when gateway turn into host. */ static int fib6_clean_tohost(struct fib6_info *rt, void *arg) @@ -3792,7 +3807,8 @@ static int fib6_clean_tohost(struct fib6_info *rt, void *arg) struct in6_addr *gateway = (struct in6_addr *)arg; if (((rt->fib6_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) && - ipv6_addr_equal(gateway, &rt->fib6_nh.nh_gw)) { + rt->fib6_nh.fib_nh_has_gw && + ipv6_addr_equal(gateway, &rt->fib6_nh.fib_nh_gw6)) { return -1; } @@ -3840,9 +3856,9 @@ static struct fib6_info *rt6_multipath_first_sibling(const struct fib6_info *rt) static bool rt6_is_dead(const struct fib6_info *rt) { - if (rt->fib6_nh.nh_flags & RTNH_F_DEAD || - (rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN && - fib6_ignore_linkdown(rt))) + if (rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD || + (rt->fib6_nh.fib_nh_flags & RTNH_F_LINKDOWN && + ip6_ignore_linkdown(rt->fib6_nh.fib_nh_dev))) return true; return false; @@ -3854,11 +3870,11 @@ static int rt6_multipath_total_weight(const struct fib6_info *rt) int total = 0; if (!rt6_is_dead(rt)) - total += rt->fib6_nh.nh_weight; + total += rt->fib6_nh.fib_nh_weight; list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) { if (!rt6_is_dead(iter)) - total += iter->fib6_nh.nh_weight; + total += iter->fib6_nh.fib_nh_weight; } return total; @@ -3869,11 +3885,11 @@ static void rt6_upper_bound_set(struct fib6_info *rt, int *weight, int total) int upper_bound = -1; if (!rt6_is_dead(rt)) { - *weight += rt->fib6_nh.nh_weight; + *weight += rt->fib6_nh.fib_nh_weight; upper_bound = DIV_ROUND_CLOSEST_ULL((u64) (*weight) << 31, total) - 1; } - atomic_set(&rt->fib6_nh.nh_upper_bound, upper_bound); + atomic_set(&rt->fib6_nh.fib_nh_upper_bound, upper_bound); } static void rt6_multipath_upper_bound_set(struct fib6_info *rt, int total) @@ -3916,8 +3932,9 @@ static int fib6_ifup(struct fib6_info *rt, void *p_arg) const struct arg_netdev_event *arg = p_arg; struct net *net = dev_net(arg->dev); - if (rt != net->ipv6.fib6_null_entry && rt->fib6_nh.nh_dev == arg->dev) { - rt->fib6_nh.nh_flags &= ~arg->nh_flags; + if (rt != net->ipv6.fib6_null_entry && + rt->fib6_nh.fib_nh_dev == arg->dev) { + rt->fib6_nh.fib_nh_flags &= ~arg->nh_flags; fib6_update_sernum_upto_root(net, rt); rt6_multipath_rebalance(rt); } @@ -3945,10 +3962,10 @@ static bool rt6_multipath_uses_dev(const struct fib6_info *rt, { struct fib6_info *iter; - if (rt->fib6_nh.nh_dev == dev) + if (rt->fib6_nh.fib_nh_dev == dev) return true; list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) - if (iter->fib6_nh.nh_dev == dev) + if (iter->fib6_nh.fib_nh_dev == dev) return true; return false; @@ -3969,12 +3986,12 @@ static unsigned int rt6_multipath_dead_count(const struct fib6_info *rt, struct fib6_info *iter; unsigned int dead = 0; - if (rt->fib6_nh.nh_dev == down_dev || - rt->fib6_nh.nh_flags & RTNH_F_DEAD) + if (rt->fib6_nh.fib_nh_dev == down_dev || + rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD) dead++; list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) - if (iter->fib6_nh.nh_dev == down_dev || - iter->fib6_nh.nh_flags & RTNH_F_DEAD) + if (iter->fib6_nh.fib_nh_dev == down_dev || + iter->fib6_nh.fib_nh_flags & RTNH_F_DEAD) dead++; return dead; @@ -3986,11 +4003,11 @@ static void rt6_multipath_nh_flags_set(struct fib6_info *rt, { struct fib6_info *iter; - if (rt->fib6_nh.nh_dev == dev) - rt->fib6_nh.nh_flags |= nh_flags; + if (rt->fib6_nh.fib_nh_dev == dev) + rt->fib6_nh.fib_nh_flags |= nh_flags; list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) - if (iter->fib6_nh.nh_dev == dev) - iter->fib6_nh.nh_flags |= nh_flags; + if (iter->fib6_nh.fib_nh_dev == dev) + iter->fib6_nh.fib_nh_flags |= nh_flags; } /* called with write lock held for table with rt */ @@ -4005,12 +4022,12 @@ static int fib6_ifdown(struct fib6_info *rt, void *p_arg) switch (arg->event) { case NETDEV_UNREGISTER: - return rt->fib6_nh.nh_dev == dev ? -1 : 0; + return rt->fib6_nh.fib_nh_dev == dev ? -1 : 0; case NETDEV_DOWN: if (rt->should_flush) return -1; if (!rt->fib6_nsiblings) - return rt->fib6_nh.nh_dev == dev ? -1 : 0; + return rt->fib6_nh.fib_nh_dev == dev ? -1 : 0; if (rt6_multipath_uses_dev(rt, dev)) { unsigned int count; @@ -4026,10 +4043,10 @@ static int fib6_ifdown(struct fib6_info *rt, void *p_arg) } return -2; case NETDEV_CHANGE: - if (rt->fib6_nh.nh_dev != dev || + if (rt->fib6_nh.fib_nh_dev != dev || rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST)) break; - rt->fib6_nh.nh_flags |= RTNH_F_LINKDOWN; + rt->fib6_nh.fib_nh_flags |= RTNH_F_LINKDOWN; rt6_multipath_rebalance(rt); break; } @@ -4085,7 +4102,7 @@ static int rt6_mtu_change_route(struct fib6_info *rt, void *p_arg) Since RFC 1981 doesn't include administrative MTU increase update PMTU increase is a MUST. (i.e. jumbo frame) */ - if (rt->fib6_nh.nh_dev == arg->dev && + if (rt->fib6_nh.fib_nh_dev == arg->dev && !fib6_metric_locked(rt, RTAX_MTU)) { u32 mtu = rt->fib6_pmtu; @@ -4376,7 +4393,7 @@ static int ip6_route_multipath_add(struct fib6_config *cfg, goto cleanup; } - rt->fib6_nh.nh_weight = rtnh->rtnh_hops + 1; + rt->fib6_nh.fib_nh_weight = rtnh->rtnh_hops + 1; err = ip6_route_info_append(info->nl_net, &rt6_nh_list, rt, &r_cfg); @@ -4526,6 +4543,9 @@ static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, if (err < 0) return err; + if (cfg.fc_metric == 0) + cfg.fc_metric = IP6_RT_PRIO_USER; + if (cfg.fc_mp) return ip6_route_multipath_add(&cfg, extack); else @@ -4540,7 +4560,7 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt) nexthop_len = nla_total_size(0) /* RTA_MULTIPATH */ + NLA_ALIGN(sizeof(struct rtnexthop)) + nla_total_size(16) /* RTA_GATEWAY */ - + lwtunnel_get_encap_size(rt->fib6_nh.nh_lwtstate); + + lwtunnel_get_encap_size(rt->fib6_nh.fib_nh_lws); nexthop_len *= rt->fib6_nsiblings; } @@ -4558,77 +4578,10 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt) + nla_total_size(sizeof(struct rta_cacheinfo)) + nla_total_size(TCP_CA_NAME_MAX) /* RTAX_CC_ALGO */ + nla_total_size(1) /* RTA_PREF */ - + lwtunnel_get_encap_size(rt->fib6_nh.nh_lwtstate) + + lwtunnel_get_encap_size(rt->fib6_nh.fib_nh_lws) + nexthop_len; } -static int rt6_nexthop_info(struct sk_buff *skb, struct fib6_info *rt, - unsigned int *flags, bool skip_oif) -{ - if (rt->fib6_nh.nh_flags & RTNH_F_DEAD) - *flags |= RTNH_F_DEAD; - - if (rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN) { - *flags |= RTNH_F_LINKDOWN; - - rcu_read_lock(); - if (fib6_ignore_linkdown(rt)) - *flags |= RTNH_F_DEAD; - rcu_read_unlock(); - } - - if (rt->fib6_flags & RTF_GATEWAY) { - if (nla_put_in6_addr(skb, RTA_GATEWAY, &rt->fib6_nh.nh_gw) < 0) - goto nla_put_failure; - } - - *flags |= (rt->fib6_nh.nh_flags & RTNH_F_ONLINK); - if (rt->fib6_nh.nh_flags & RTNH_F_OFFLOAD) - *flags |= RTNH_F_OFFLOAD; - - /* not needed for multipath encoding b/c it has a rtnexthop struct */ - if (!skip_oif && rt->fib6_nh.nh_dev && - nla_put_u32(skb, RTA_OIF, rt->fib6_nh.nh_dev->ifindex)) - goto nla_put_failure; - - if (rt->fib6_nh.nh_lwtstate && - lwtunnel_fill_encap(skb, rt->fib6_nh.nh_lwtstate) < 0) - goto nla_put_failure; - - return 0; - -nla_put_failure: - return -EMSGSIZE; -} - -/* add multipath next hop */ -static int rt6_add_nexthop(struct sk_buff *skb, struct fib6_info *rt) -{ - const struct net_device *dev = rt->fib6_nh.nh_dev; - struct rtnexthop *rtnh; - unsigned int flags = 0; - - rtnh = nla_reserve_nohdr(skb, sizeof(*rtnh)); - if (!rtnh) - goto nla_put_failure; - - rtnh->rtnh_hops = rt->fib6_nh.nh_weight - 1; - rtnh->rtnh_ifindex = dev ? dev->ifindex : 0; - - if (rt6_nexthop_info(skb, rt, &flags, true) < 0) - goto nla_put_failure; - - rtnh->rtnh_flags = flags; - - /* length of rtnetlink header + attributes */ - rtnh->rtnh_len = nlmsg_get_pos(skb) - (void *)rtnh; - - return 0; - -nla_put_failure: - return -EMSGSIZE; -} - static int rt6_fill_node(struct net *net, struct sk_buff *skb, struct fib6_info *rt, struct dst_entry *dst, struct in6_addr *dest, struct in6_addr *src, @@ -4745,18 +4698,21 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb, if (!mp) goto nla_put_failure; - if (rt6_add_nexthop(skb, rt) < 0) + if (fib_add_nexthop(skb, &rt->fib6_nh.nh_common, + rt->fib6_nh.fib_nh_weight) < 0) goto nla_put_failure; list_for_each_entry_safe(sibling, next_sibling, &rt->fib6_siblings, fib6_siblings) { - if (rt6_add_nexthop(skb, sibling) < 0) + if (fib_add_nexthop(skb, &sibling->fib6_nh.nh_common, + sibling->fib6_nh.fib_nh_weight) < 0) goto nla_put_failure; } nla_nest_end(skb, mp); } else { - if (rt6_nexthop_info(skb, rt, &rtm->rtm_flags, false) < 0) + if (fib_nexthop_info(skb, &rt->fib6_nh.nh_common, + &rtm->rtm_flags, false) < 0) goto nla_put_failure; } @@ -4783,7 +4739,7 @@ nla_put_failure: static bool fib6_info_uses_dev(const struct fib6_info *f6i, const struct net_device *dev) { - if (f6i->fib6_nh.nh_dev == dev) + if (f6i->fib6_nh.fib_nh_dev == dev) return true; if (f6i->fib6_nsiblings) { @@ -4791,7 +4747,7 @@ static bool fib6_info_uses_dev(const struct fib6_info *f6i, list_for_each_entry_safe(sibling, next_sibling, &f6i->fib6_siblings, fib6_siblings) { - if (sibling->fib6_nh.nh_dev == dev) + if (sibling->fib6_nh.fib_nh_dev == dev) return true; } } @@ -5076,7 +5032,7 @@ static int ip6_route_dev_notify(struct notifier_block *this, return NOTIFY_OK; if (event == NETDEV_REGISTER) { - net->ipv6.fib6_null_entry->fib6_nh.nh_dev = dev; + net->ipv6.fib6_null_entry->fib6_nh.fib_nh_dev = dev; net->ipv6.ip6_null_entry->dst.dev = dev; net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev); #ifdef CONFIG_IPV6_MULTIPLE_TABLES @@ -5411,7 +5367,7 @@ void __init ip6_route_init_special_entries(void) /* Registering of the loopback is done before this portion of code, * the loopback reference in rt6_info will not be taken, do it * manually for init_net */ - init_net.ipv6.fib6_null_entry->fib6_nh.nh_dev = init_net.loopback_dev; + init_net.ipv6.fib6_null_entry->fib6_nh.fib_nh_dev = init_net.loopback_dev; init_net.ipv6.ip6_null_entry->dst.dev = init_net.loopback_dev; init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); #ifdef CONFIG_IPV6_MULTIPLE_TABLES diff --git a/net/ipv6/seg6.c b/net/ipv6/seg6.c index 9b2f272ca164..ceff773471e7 100644 --- a/net/ipv6/seg6.c +++ b/net/ipv6/seg6.c @@ -399,7 +399,6 @@ static const struct genl_ops seg6_genl_ops[] = { { .cmd = SEG6_CMD_SETHMAC, .doit = seg6_genl_sethmac, - .policy = seg6_genl_policy, .flags = GENL_ADMIN_PERM, }, { @@ -407,19 +406,16 @@ static const struct genl_ops seg6_genl_ops[] = { .start = seg6_genl_dumphmac_start, .dumpit = seg6_genl_dumphmac, .done = seg6_genl_dumphmac_done, - .policy = seg6_genl_policy, .flags = GENL_ADMIN_PERM, }, { .cmd = SEG6_CMD_SET_TUNSRC, .doit = seg6_genl_set_tunsrc, - .policy = seg6_genl_policy, .flags = GENL_ADMIN_PERM, }, { .cmd = SEG6_CMD_GET_TUNSRC, .doit = seg6_genl_get_tunsrc, - .policy = seg6_genl_policy, .flags = GENL_ADMIN_PERM, }, }; @@ -429,6 +425,7 @@ static struct genl_family seg6_genl_family __ro_after_init = { .name = SEG6_GENL_NAME, .version = SEG6_GENL_VERSION, .maxattr = SEG6_ATTR_MAX, + .policy = seg6_genl_policy, .netnsok = true, .parallel_ops = true, .ops = seg6_genl_ops, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 44d431849d39..82018bdce863 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -90,6 +90,18 @@ static struct tcp_md5sig_key *tcp_v6_md5_do_lookup(const struct sock *sk, } #endif +/* Helper returning the inet6 address from a given tcp socket. + * It can be used in TCP stack instead of inet6_sk(sk). + * This avoids a dereference and allow compiler optimizations. + * It is a specialized version of inet6_sk_generic(). + */ +static struct ipv6_pinfo *tcp_inet6_sk(const struct sock *sk) +{ + unsigned int offset = sizeof(struct tcp6_sock) - sizeof(struct ipv6_pinfo); + + return (struct ipv6_pinfo *)(((u8 *)sk) + offset); +} + static void inet6_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); @@ -99,7 +111,7 @@ static void inet6_sk_rx_dst_set(struct sock *sk, const struct sk_buff *skb) sk->sk_rx_dst = dst; inet_sk(sk)->rx_dst_ifindex = skb->skb_iif; - inet6_sk(sk)->rx_dst_cookie = rt6_get_cookie(rt); + tcp_inet6_sk(sk)->rx_dst_cookie = rt6_get_cookie(rt); } } @@ -138,7 +150,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr, struct sockaddr_in6 *usin = (struct sockaddr_in6 *) uaddr; struct inet_sock *inet = inet_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); - struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6_pinfo *np = tcp_inet6_sk(sk); struct tcp_sock *tp = tcp_sk(sk); struct in6_addr *saddr = NULL, *final_p, final; struct ipv6_txoptions *opt; @@ -390,7 +402,7 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (sk->sk_state == TCP_CLOSE) goto out; - if (ipv6_hdr(skb)->hop_limit < inet6_sk(sk)->min_hopcount) { + if (ipv6_hdr(skb)->hop_limit < tcp_inet6_sk(sk)->min_hopcount) { __NET_INC_STATS(net, LINUX_MIB_TCPMINTTLDROP); goto out; } @@ -405,7 +417,7 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, goto out; } - np = inet6_sk(sk); + np = tcp_inet6_sk(sk); if (type == NDISC_REDIRECT) { if (!sock_owned_by_user(sk)) { @@ -478,7 +490,7 @@ static int tcp_v6_send_synack(const struct sock *sk, struct dst_entry *dst, enum tcp_synack_type synack_type) { struct inet_request_sock *ireq = inet_rsk(req); - struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6_pinfo *np = tcp_inet6_sk(sk); struct ipv6_txoptions *opt; struct flowi6 *fl6 = &fl->u.ip6; struct sk_buff *skb; @@ -737,7 +749,7 @@ static void tcp_v6_init_req(struct request_sock *req, { bool l3_slave = ipv6_l3mdev_skb(TCP_SKB_CB(skb)->header.h6.flags); struct inet_request_sock *ireq = inet_rsk(req); - const struct ipv6_pinfo *np = inet6_sk(sk_listener); + const struct ipv6_pinfo *np = tcp_inet6_sk(sk_listener); ireq->ir_v6_rmt_addr = ipv6_hdr(skb)->saddr; ireq->ir_v6_loc_addr = ipv6_hdr(skb)->daddr; @@ -1066,9 +1078,8 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * { struct inet_request_sock *ireq; struct ipv6_pinfo *newnp; - const struct ipv6_pinfo *np = inet6_sk(sk); + const struct ipv6_pinfo *np = tcp_inet6_sk(sk); struct ipv6_txoptions *opt; - struct tcp6_sock *newtcp6sk; struct inet_sock *newinet; struct tcp_sock *newtp; struct sock *newsk; @@ -1088,11 +1099,10 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * if (!newsk) return NULL; - newtcp6sk = (struct tcp6_sock *)newsk; - inet_sk(newsk)->pinet6 = &newtcp6sk->inet6; + inet_sk(newsk)->pinet6 = tcp_inet6_sk(newsk); newinet = inet_sk(newsk); - newnp = inet6_sk(newsk); + newnp = tcp_inet6_sk(newsk); newtp = tcp_sk(newsk); memcpy(newnp, np, sizeof(struct ipv6_pinfo)); @@ -1156,12 +1166,11 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * ip6_dst_store(newsk, dst, NULL, NULL); inet6_sk_rx_dst_set(newsk, skb); - newtcp6sk = (struct tcp6_sock *)newsk; - inet_sk(newsk)->pinet6 = &newtcp6sk->inet6; + inet_sk(newsk)->pinet6 = tcp_inet6_sk(newsk); newtp = tcp_sk(newsk); newinet = inet_sk(newsk); - newnp = inet6_sk(newsk); + newnp = tcp_inet6_sk(newsk); memcpy(newnp, np, sizeof(struct ipv6_pinfo)); @@ -1276,9 +1285,9 @@ out: */ static int tcp_v6_do_rcv(struct sock *sk, struct sk_buff *skb) { - struct ipv6_pinfo *np = inet6_sk(sk); - struct tcp_sock *tp; + struct ipv6_pinfo *np = tcp_inet6_sk(sk); struct sk_buff *opt_skb = NULL; + struct tcp_sock *tp; /* Imagine: socket is IPv6. IPv4 packet arrives, goes to IPv4 receive handler and backlogged. @@ -1428,6 +1437,7 @@ static void tcp_v6_fill_cb(struct sk_buff *skb, const struct ipv6hdr *hdr, static int tcp_v6_rcv(struct sk_buff *skb) { + struct sk_buff *skb_to_free; int sdif = inet6_sdif(skb); const struct tcphdr *th; const struct ipv6hdr *hdr; @@ -1524,7 +1534,7 @@ process: return 0; } } - if (hdr->hop_limit < inet6_sk(sk)->min_hopcount) { + if (hdr->hop_limit < tcp_inet6_sk(sk)->min_hopcount) { __NET_INC_STATS(net, LINUX_MIB_TCPMINTTLDROP); goto discard_and_relse; } @@ -1554,12 +1564,17 @@ process: tcp_segs_in(tcp_sk(sk), skb); ret = 0; if (!sock_owned_by_user(sk)) { + skb_to_free = sk->sk_rx_skb_cache; + sk->sk_rx_skb_cache = NULL; ret = tcp_v6_do_rcv(sk, skb); - } else if (tcp_add_backlog(sk, skb)) { - goto discard_and_relse; + } else { + if (tcp_add_backlog(sk, skb)) + goto discard_and_relse; + skb_to_free = NULL; } bh_unlock_sock(sk); - + if (skb_to_free) + __kfree_skb(skb_to_free); put_and_return: if (refcounted) sock_put(sk); @@ -1669,7 +1684,7 @@ static void tcp_v6_early_demux(struct sk_buff *skb) struct dst_entry *dst = READ_ONCE(sk->sk_rx_dst); if (dst) - dst = dst_check(dst, inet6_sk(sk)->rx_dst_cookie); + dst = dst_check(dst, tcp_inet6_sk(sk)->rx_dst_cookie); if (dst && inet_sk(sk)->rx_dst_ifindex == skb->skb_iif) skb_dst_set_noref(skb, dst); |