diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index eace3a9d87ad0..3b940b0e74159 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -664,7 +664,7 @@ static int vrf_finish_output6(struct net *net, struct sock *sk, skb->protocol = htons(ETH_P_IPV6); skb->dev = dev; - rcu_read_lock_bh(); + rcu_read_lock(); nexthop = rt6_nexthop((struct rt6_info *)dst, &ipv6_hdr(skb)->daddr); neigh = __ipv6_neigh_lookup_noref(dst->dev, nexthop); if (unlikely(!neigh)) @@ -672,10 +672,10 @@ static int vrf_finish_output6(struct net *net, struct sock *sk, if (!IS_ERR(neigh)) { sock_confirm_neigh(skb, neigh); ret = neigh_output(neigh, skb, false); - rcu_read_unlock_bh(); + rcu_read_unlock(); return ret; } - rcu_read_unlock_bh(); + rcu_read_unlock(); IP6_INC_STATS(dev_net(dst->dev), ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); @@ -889,7 +889,7 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s } } - rcu_read_lock_bh(); + rcu_read_lock(); neigh = ip_neigh_for_gw(rt, skb, &is_v6gw); if (!IS_ERR(neigh)) { @@ -898,11 +898,11 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s sock_confirm_neigh(skb, neigh); /* if crossing protocols, can not use the cached header */ ret = neigh_output(neigh, skb, is_v6gw); - rcu_read_unlock_bh(); + rcu_read_unlock(); return ret; } - rcu_read_unlock_bh(); + rcu_read_unlock(); vrf_tx_error(skb->dev, skb); return -EINVAL; } diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index ae2c6a3cec5db..1d7a0a6123c86 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -60,16 +60,6 @@ static inline bool lockdep_rtnl_is_held(void) #define rcu_dereference_rtnl(p) \ rcu_dereference_check(p, lockdep_rtnl_is_held()) -/** - * rcu_dereference_bh_rtnl - rcu_dereference_bh with debug checking - * @p: The pointer to read, prior to dereference - * - * Do an rcu_dereference_bh(p), but check caller either holds rcu_read_lock_bh() - * or RTNL. Note : Please prefer rtnl_dereference() or rcu_dereference_bh() - */ -#define rcu_dereference_bh_rtnl(p) \ - rcu_dereference_bh_check(p, lockdep_rtnl_is_held()) - /** * rtnl_dereference - fetch RCU pointer when updates are prevented by RTNL * @p: The pointer to read, prior to dereferencing diff --git a/include/net/arp.h b/include/net/arp.h index 4950191f6b2bf..46144bba3dade 100644 --- a/include/net/arp.h +++ b/include/net/arp.h @@ -38,11 +38,11 @@ static inline struct neighbour *__ipv4_neigh_lookup(struct net_device *dev, u32 { struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv4_neigh_lookup_noref(dev, key); if (n && !refcount_inc_not_zero(&n->refcnt)) n = NULL; - rcu_read_unlock_bh(); + rcu_read_unlock(); return n; } @@ -51,16 +51,10 @@ static inline void __ipv4_confirm_neigh(struct net_device *dev, u32 key) { struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv4_neigh_lookup_noref(dev, key); - if (n) { - unsigned long now = jiffies; - - /* avoid dirtying neighbour */ - if (READ_ONCE(n->confirmed) != now) - WRITE_ONCE(n->confirmed, now); - } - rcu_read_unlock_bh(); + neigh_confirm(n); + rcu_read_unlock(); } void arp_init(void); diff --git a/include/net/dst.h b/include/net/dst.h index 6aa252c3fc55c..86c18222ba1ce 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -23,7 +23,12 @@ struct sk_buff; struct dst_entry { - struct net_device *dev; + RH_KABI_REPLACE( + struct net_device *dev, + union { + struct net_device *dev; + struct net_device __rcu *dev_rcu; + }) struct dst_ops *ops; unsigned long _metrics; unsigned long expires; @@ -553,6 +558,41 @@ static inline void skb_dst_update_pmtu_no_confirm(struct sk_buff *skb, u32 mtu) dst->ops->update_pmtu(dst, NULL, skb, mtu, false); } +static inline struct net_device *dst_dev(const struct dst_entry *dst) +{ + return READ_ONCE(dst->dev); +} + +static inline struct net_device *dst_dev_rcu(const struct dst_entry *dst) +{ + return rcu_dereference(dst->dev_rcu); +} + +static inline struct net *dst_dev_net_rcu(const struct dst_entry *dst) +{ + return dev_net_rcu(dst_dev_rcu(dst)); +} + +static inline struct net_device *skb_dst_dev(const struct sk_buff *skb) +{ + return dst_dev(skb_dst(skb)); +} + +static inline struct net_device *skb_dst_dev_rcu(const struct sk_buff *skb) +{ + return dst_dev_rcu(skb_dst(skb)); +} + +static inline struct net *skb_dst_dev_net(const struct sk_buff *skb) +{ + return dev_net(skb_dst_dev(skb)); +} + +static inline struct net *skb_dst_dev_net_rcu(const struct sk_buff *skb) +{ + return dev_net_rcu(skb_dst_dev_rcu(skb)); +} + struct dst_entry *dst_blackhole_check(struct dst_entry *dst, u32 cookie); void dst_blackhole_update_pmtu(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb, u32 mtu, bool confirm_neigh); diff --git a/include/net/ip.h b/include/net/ip.h index faf5baa174ff0..86de7ad5e85cb 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -440,20 +440,43 @@ static inline bool ip_sk_ignore_df(const struct sock *sk) static inline unsigned int ip_dst_mtu_maybe_forward(const struct dst_entry *dst, bool forwarding) { - struct net *net = dev_net(dst->dev); - unsigned int mtu; + const struct rtable *rt = container_of(dst, struct rtable, dst); + const struct net_device *dev; + unsigned int mtu, res; + struct net *net; + + rcu_read_lock(); + dev = dst_dev_rcu(dst); + net = dev_net_rcu(dev); if (READ_ONCE(net->ipv4.sysctl_ip_fwd_use_pmtu) || ip_mtu_locked(dst) || - !forwarding) - return dst_mtu(dst); + !forwarding) { + mtu = rt->rt_pmtu; + if (mtu && time_before(jiffies, rt->dst.expires)) + goto out; + } /* 'forwarding = true' case should always honour route mtu */ mtu = dst_metric_raw(dst, RTAX_MTU); - if (!mtu) - mtu = min(READ_ONCE(dst->dev->mtu), IP_MAX_MTU); + if (mtu) + goto out; + + mtu = READ_ONCE(dev->mtu); + + if (unlikely(ip_mtu_locked(dst))) { + if (rt->rt_uses_gateway && mtu > 576) + mtu = 576; + } + +out: + mtu = min_t(unsigned int, mtu, IP_MAX_MTU); + + res = mtu - lwtunnel_headroom(dst->lwtstate, mtu); + + rcu_read_unlock(); - return mtu - lwtunnel_headroom(dst->lwtstate, mtu); + return res; } static inline unsigned int ip_skb_dst_mtu(struct sock *sk, diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 0bf09a9bca4e0..e466c0da4e642 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -316,12 +316,13 @@ static inline bool rt6_duplicate_nexthop(struct fib6_info *a, struct fib6_info * !lwtunnel_cmp_encap(nha->fib_nh_lws, nhb->fib_nh_lws); } -static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) +static inline unsigned int ip6_dst_mtu_maybe_forward(const struct dst_entry *dst, + bool forwarding) { struct inet6_dev *idev; unsigned int mtu; - if (dst_metric_locked(dst, RTAX_MTU)) { + if (!forwarding || dst_metric_locked(dst, RTAX_MTU)) { mtu = dst_metric_raw(dst, RTAX_MTU); if (mtu) goto out; @@ -329,7 +330,7 @@ static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst) mtu = IPV6_MIN_MTU; rcu_read_lock(); - idev = __in6_dev_get(dst->dev); + idev = __in6_dev_get(dst_dev_rcu(dst)); if (idev) mtu = idev->cnf.mtu6; rcu_read_unlock(); diff --git a/include/net/ndisc.h b/include/net/ndisc.h index fc5c161e93653..59e9ed7a23f22 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -395,11 +395,11 @@ static inline struct neighbour *__ipv6_neigh_lookup(struct net_device *dev, cons { struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv6_neigh_lookup_noref(dev, pkey); if (n && !refcount_inc_not_zero(&n->refcnt)) n = NULL; - rcu_read_unlock_bh(); + rcu_read_unlock(); return n; } @@ -409,16 +409,10 @@ static inline void __ipv6_confirm_neigh(struct net_device *dev, { struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv6_neigh_lookup_noref(dev, pkey); - if (n) { - unsigned long now = jiffies; - - /* avoid dirtying neighbour */ - if (READ_ONCE(n->confirmed) != now) - WRITE_ONCE(n->confirmed, now); - } - rcu_read_unlock_bh(); + neigh_confirm(n); + rcu_read_unlock(); } static inline void __ipv6_confirm_neigh_stub(struct net_device *dev, @@ -426,16 +420,10 @@ static inline void __ipv6_confirm_neigh_stub(struct net_device *dev, { struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv6_neigh_lookup_noref_stub(dev, pkey); - if (n) { - unsigned long now = jiffies; - - /* avoid dirtying neighbour */ - if (READ_ONCE(n->confirmed) != now) - WRITE_ONCE(n->confirmed, now); - } - rcu_read_unlock_bh(); + neigh_confirm(n); + rcu_read_unlock(); } /* uses ipv6_stub and is meant for use outside of IPv6 core */ diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 2a1905cd5207c..61c7e80489d61 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -290,14 +290,14 @@ static inline struct neighbour *___neigh_lookup_noref( const void *pkey, struct net_device *dev) { - struct neigh_hash_table *nht = rcu_dereference_bh(tbl->nht); + struct neigh_hash_table *nht = rcu_dereference(tbl->nht); struct neighbour *n; u32 hash_val; hash_val = hash(pkey, dev, nht->hash_rnd) >> (32 - nht->hash_shift); - for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]); + for (n = rcu_dereference(nht->hash_buckets[hash_val]); n != NULL; - n = rcu_dereference_bh(n->next)) { + n = rcu_dereference(n->next)) { if (n->dev == dev && key_eq(n, pkey)) return n; } @@ -312,6 +312,17 @@ static inline struct neighbour *__neigh_lookup_noref(struct neigh_table *tbl, return ___neigh_lookup_noref(tbl, tbl->key_eq, tbl->hash, pkey, dev); } +static inline void neigh_confirm(struct neighbour *n) +{ + if (n) { + unsigned long now = jiffies; + + /* avoid dirtying neighbour */ + if (READ_ONCE(n->confirmed) != now) + WRITE_ONCE(n->confirmed, now); + } +} + void neigh_table_init(int index, struct neigh_table *tbl); int neigh_table_clear(int index, struct neigh_table *tbl); struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, diff --git a/include/net/nexthop.h b/include/net/nexthop.h index 28085b995ddcf..2b12725de9c09 100644 --- a/include/net/nexthop.h +++ b/include/net/nexthop.h @@ -497,29 +497,6 @@ static inline struct fib6_nh *nexthop_fib6_nh(struct nexthop *nh) return NULL; } -/* Variant of nexthop_fib6_nh(). - * Caller should either hold rcu_read_lock_bh(), or RTNL. - */ -static inline struct fib6_nh *nexthop_fib6_nh_bh(struct nexthop *nh) -{ - struct nh_info *nhi; - - if (nh->is_group) { - struct nh_group *nh_grp; - - nh_grp = rcu_dereference_bh_rtnl(nh->nh_grp); - nh = nexthop_mpath_select(nh_grp, 0); - if (!nh) - return NULL; - } - - nhi = rcu_dereference_bh_rtnl(nh->nh_info); - if (nhi->family == AF_INET6) - return &nhi->fib6_nh; - - return NULL; -} - static inline struct net_device *fib6_info_nh_dev(struct fib6_info *f6i) { struct fib6_nh *fib6_nh; diff --git a/include/net/route.h b/include/net/route.h index 08df794364853..fcd1b29f58f6a 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -358,10 +358,15 @@ static inline int inet_iif(const struct sk_buff *skb) static inline int ip4_dst_hoplimit(const struct dst_entry *dst) { int hoplimit = dst_metric_raw(dst, RTAX_HOPLIMIT); - struct net *net = dev_net(dst->dev); - if (hoplimit == 0) + if (hoplimit == 0) { + const struct net *net; + + rcu_read_lock(); + net = dst_dev_net_rcu(dst); hoplimit = READ_ONCE(net->ipv4.sysctl_ip_default_ttl); + rcu_read_unlock(); + } return hoplimit; } diff --git a/include/net/sock.h b/include/net/sock.h index a9d88975c5aec..412f809d7d0a0 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2192,13 +2192,10 @@ static inline void sock_confirm_neigh(struct sk_buff *skb, struct neighbour *n) { if (skb_get_dst_pending_confirm(skb)) { struct sock *sk = skb->sk; - unsigned long now = jiffies; - /* avoid dirtying neighbour */ - if (READ_ONCE(n->confirmed) != now) - WRITE_ONCE(n->confirmed, now); if (sk && READ_ONCE(sk->sk_dst_pending_confirm)) WRITE_ONCE(sk->sk_dst_pending_confirm, 0); + neigh_confirm(n); } } diff --git a/net/core/dst.c b/net/core/dst.c index 81e7c996448aa..c37a917004b8d 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -154,7 +154,7 @@ void dst_dev_put(struct dst_entry *dst) dst->ops->ifdown(dst, dev, true); dst->input = dst_discard; dst->output = dst_discard_out; - dst->dev = blackhole_netdev; + rcu_assign_pointer(dst->dev_rcu, blackhole_netdev); dev_replace_track(dev, blackhole_netdev, &dst->dev_tracker, GFP_ATOMIC); } @@ -265,7 +265,7 @@ unsigned int dst_blackhole_mtu(const struct dst_entry *dst) { unsigned int mtu = dst_metric_raw(dst, RTAX_MTU); - return mtu ? : dst->dev->mtu; + return mtu ? : dst_dev(dst)->mtu; } EXPORT_SYMBOL_GPL(dst_blackhole_mtu); diff --git a/net/core/filter.c b/net/core/filter.c index 77d1877b39301..b07b292390ee4 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2191,7 +2191,7 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb, return -ENOMEM; } - rcu_read_lock_bh(); + rcu_read_lock(); if (!nh) { dst = skb_dst(skb); nexthop = rt6_nexthop(container_of(dst, struct rt6_info, dst), @@ -2204,13 +2204,15 @@ static int bpf_out_neigh_v6(struct net *net, struct sk_buff *skb, int ret; sock_confirm_neigh(skb, neigh); + local_bh_disable(); dev_xmit_recursion_inc(); ret = neigh_output(neigh, skb, false); dev_xmit_recursion_dec(); - rcu_read_unlock_bh(); + local_bh_enable(); + rcu_read_unlock(); return ret; } - rcu_read_unlock_bh(); + rcu_read_unlock(); if (dst) IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); out_drop: @@ -2289,7 +2291,7 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb, return -ENOMEM; } - rcu_read_lock_bh(); + rcu_read_lock(); if (!nh) { struct dst_entry *dst = skb_dst(skb); struct rtable *rt = container_of(dst, struct rtable, dst); @@ -2301,7 +2303,7 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb, } else if (nh->nh_family == AF_INET) { neigh = ip_neigh_gw4(dev, nh->ipv4_nh); } else { - rcu_read_unlock_bh(); + rcu_read_unlock(); goto out_drop; } @@ -2309,13 +2311,15 @@ static int bpf_out_neigh_v4(struct net *net, struct sk_buff *skb, int ret; sock_confirm_neigh(skb, neigh); + local_bh_disable(); dev_xmit_recursion_inc(); ret = neigh_output(neigh, skb, is_v6gw); dev_xmit_recursion_dec(); - rcu_read_unlock_bh(); + local_bh_enable(); + rcu_read_unlock(); return ret; } - rcu_read_unlock_bh(); + rcu_read_unlock(); out_drop: kfree_skb(skb); return -ENETDOWN; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 0a7cf42f04df2..e65bdd8fb7d37 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -532,7 +532,7 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, NEIGH_CACHE_STAT_INC(tbl, lookups); - rcu_read_lock_bh(); + rcu_read_lock(); n = __neigh_lookup_noref(tbl, pkey, dev); if (n) { if (!refcount_inc_not_zero(&n->refcnt)) @@ -540,7 +540,7 @@ struct neighbour *neigh_lookup(struct neigh_table *tbl, const void *pkey, NEIGH_CACHE_STAT_INC(tbl, hits); } - rcu_read_unlock_bh(); + rcu_read_unlock(); return n; } EXPORT_SYMBOL(neigh_lookup); @@ -555,13 +555,13 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, NEIGH_CACHE_STAT_INC(tbl, lookups); - rcu_read_lock_bh(); - nht = rcu_dereference_bh(tbl->nht); + rcu_read_lock(); + nht = rcu_dereference(tbl->nht); hash_val = tbl->hash(pkey, NULL, nht->hash_rnd) >> (32 - nht->hash_shift); - for (n = rcu_dereference_bh(nht->hash_buckets[hash_val]); + for (n = rcu_dereference(nht->hash_buckets[hash_val]); n != NULL; - n = rcu_dereference_bh(n->next)) { + n = rcu_dereference(n->next)) { if (!memcmp(n->primary_key, pkey, key_len) && net_eq(dev_net(n->dev), net)) { if (!refcount_inc_not_zero(&n->refcnt)) @@ -571,7 +571,7 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, } } - rcu_read_unlock_bh(); + rcu_read_unlock(); return n; } EXPORT_SYMBOL(neigh_lookup_nodev); @@ -2067,11 +2067,11 @@ static int neightbl_fill_info(struct sk_buff *skb, struct neigh_table *tbl, .ndtc_proxy_qlen = tbl->proxy_queue.qlen, }; - rcu_read_lock_bh(); - nht = rcu_dereference_bh(tbl->nht); + rcu_read_lock(); + nht = rcu_dereference(tbl->nht); ndc.ndtc_hash_rnd = nht->hash_rnd[0]; ndc.ndtc_hash_mask = ((1 << nht->hash_shift) - 1); - rcu_read_unlock_bh(); + rcu_read_unlock(); if (nla_put(skb, NDTA_CONFIG, sizeof(ndc), &ndc)) goto nla_put_failure; @@ -2562,15 +2562,15 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, if (filter->dev_idx || filter->master_idx) flags |= NLM_F_DUMP_FILTERED; - rcu_read_lock_bh(); - nht = rcu_dereference_bh(tbl->nht); + rcu_read_lock(); + nht = rcu_dereference(tbl->nht); for (h = s_h; h < (1 << nht->hash_shift); h++) { if (h > s_h) s_idx = 0; - for (n = rcu_dereference_bh(nht->hash_buckets[h]), idx = 0; + for (n = rcu_dereference(nht->hash_buckets[h]), idx = 0; n != NULL; - n = rcu_dereference_bh(n->next)) { + n = rcu_dereference(n->next)) { if (idx < s_idx || !net_eq(dev_net(n->dev), net)) goto next; if (neigh_ifindex_filtered(n->dev, filter->dev_idx) || @@ -2589,7 +2589,7 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, } rc = skb->len; out: - rcu_read_unlock_bh(); + rcu_read_unlock(); cb->args[1] = h; cb->args[2] = idx; return rc; @@ -2932,20 +2932,20 @@ void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void int chain; struct neigh_hash_table *nht; - rcu_read_lock_bh(); - nht = rcu_dereference_bh(tbl->nht); + rcu_read_lock(); + nht = rcu_dereference(tbl->nht); - read_lock(&tbl->lock); /* avoid resizes */ + read_lock_bh(&tbl->lock); /* avoid resizes */ for (chain = 0; chain < (1 << nht->hash_shift); chain++) { struct neighbour *n; - for (n = rcu_dereference_bh(nht->hash_buckets[chain]); + for (n = rcu_dereference(nht->hash_buckets[chain]); n != NULL; - n = rcu_dereference_bh(n->next)) + n = rcu_dereference(n->next)) cb(n, cookie); } - read_unlock(&tbl->lock); - rcu_read_unlock_bh(); + read_unlock_bh(&tbl->lock); + rcu_read_unlock(); } EXPORT_SYMBOL(neigh_for_each); @@ -2995,7 +2995,7 @@ int neigh_xmit(int index, struct net_device *dev, tbl = neigh_tables[index]; if (!tbl) goto out; - rcu_read_lock_bh(); + rcu_read_lock(); if (index == NEIGH_ARP_TABLE) { u32 key = *((u32 *)addr); @@ -3007,11 +3007,11 @@ int neigh_xmit(int index, struct net_device *dev, neigh = __neigh_create(tbl, addr, dev, false); err = PTR_ERR(neigh); if (IS_ERR(neigh)) { - rcu_read_unlock_bh(); + rcu_read_unlock(); goto out_kfree_skb; } err = neigh->output(neigh, skb); - rcu_read_unlock_bh(); + rcu_read_unlock(); } else if (index == NEIGH_LINK_TABLE) { err = dev_hard_header(skb, dev, ntohs(skb->protocol), @@ -3040,7 +3040,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) state->flags &= ~NEIGH_SEQ_IS_PNEIGH; for (bucket = 0; bucket < (1 << nht->hash_shift); bucket++) { - n = rcu_dereference_bh(nht->hash_buckets[bucket]); + n = rcu_dereference(nht->hash_buckets[bucket]); while (n) { if (!net_eq(dev_net(n->dev), net)) @@ -3058,7 +3058,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) if (n->nud_state & ~NUD_NOARP) break; next: - n = rcu_dereference_bh(n->next); + n = rcu_dereference(n->next); } if (n) @@ -3082,7 +3082,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (v) return n; } - n = rcu_dereference_bh(n->next); + n = rcu_dereference(n->next); while (1) { while (n) { @@ -3100,7 +3100,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (n->nud_state & ~NUD_NOARP) break; next: - n = rcu_dereference_bh(n->next); + n = rcu_dereference(n->next); } if (n) @@ -3109,7 +3109,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, if (++state->bucket >= (1 << nht->hash_shift)) break; - n = rcu_dereference_bh(nht->hash_buckets[state->bucket]); + n = rcu_dereference(nht->hash_buckets[state->bucket]); } if (n && pos) @@ -3211,7 +3211,7 @@ static void *neigh_get_idx_any(struct seq_file *seq, loff_t *pos) void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl, unsigned int neigh_seq_flags) __acquires(tbl->lock) - __acquires(rcu_bh) + __acquires(rcu) { struct neigh_seq_state *state = seq->private; @@ -3219,9 +3219,9 @@ void *neigh_seq_start(struct seq_file *seq, loff_t *pos, struct neigh_table *tbl state->bucket = 0; state->flags = (neigh_seq_flags & ~NEIGH_SEQ_IS_PNEIGH); - rcu_read_lock_bh(); - state->nht = rcu_dereference_bh(tbl->nht); - read_lock(&tbl->lock); + rcu_read_lock(); + state->nht = rcu_dereference(tbl->nht); + read_lock_bh(&tbl->lock); return *pos ? neigh_get_idx_any(seq, pos) : SEQ_START_TOKEN; } @@ -3256,13 +3256,13 @@ EXPORT_SYMBOL(neigh_seq_next); void neigh_seq_stop(struct seq_file *seq, void *v) __releases(tbl->lock) - __releases(rcu_bh) + __releases(rcu) { struct neigh_seq_state *state = seq->private; struct neigh_table *tbl = state->tbl; - read_unlock(&tbl->lock); - rcu_read_unlock_bh(); + read_unlock_bh(&tbl->lock); + rcu_read_unlock(); } EXPORT_SYMBOL(neigh_seq_stop); diff --git a/net/core/sock.c b/net/core/sock.c index 39f201d7377bf..c1f5786561d2c 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2283,10 +2283,12 @@ static void sk_trim_gso_size(struct sock *sk) void sk_setup_caps(struct sock *sk, struct dst_entry *dst) { + const struct net_device *dev; u32 max_segs = 1; - sk_dst_set(sk, dst); - sk->sk_route_caps = dst->dev->features; + rcu_read_lock(); + dev = dst_dev_rcu(dst); + sk->sk_route_caps = dev->features; if (sk_is_tcp(sk)) sk->sk_route_caps |= NETIF_F_GSO; if (sk->sk_route_caps & NETIF_F_GSO) @@ -2299,14 +2301,16 @@ void sk_setup_caps(struct sock *sk, struct dst_entry *dst) } else { sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM; /* pairs with the WRITE_ONCE() in netif_set_gso_max_size() */ - sk->sk_gso_max_size = READ_ONCE(dst->dev->gso_max_size); + sk->sk_gso_max_size = READ_ONCE(dev->gso_max_size); sk_trim_gso_size(sk); sk->sk_gso_max_size -= (MAX_TCP_HEADER + 1); /* pairs with the WRITE_ONCE() in netif_set_gso_max_segs() */ - max_segs = max_t(u32, READ_ONCE(dst->dev->gso_max_segs), 1); + max_segs = max_t(u32, READ_ONCE(dev->gso_max_segs), 1); } } sk->sk_gso_max_segs = max_segs; + sk_dst_set(sk, dst); + rcu_read_unlock(); } EXPORT_SYMBOL_GPL(sk_setup_caps); diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 019a13cad52d2..ecec47c6b215e 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -2211,7 +2211,7 @@ static bool fib_good_nh(const struct fib_nh *nh) if (nh->fib_nh_scope == RT_SCOPE_LINK) { struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); if (likely(nh->fib_nh_gw_family == AF_INET)) n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev, @@ -2224,7 +2224,7 @@ static bool fib_good_nh(const struct fib_nh *nh) if (n) state = n->nud_state; - rcu_read_unlock_bh(); + rcu_read_unlock(); } return !!(state & NUD_VALID); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 8820bddf285ee..4d0068e73d9bd 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -218,7 +218,7 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s return res; } - rcu_read_lock_bh(); + rcu_read_lock(); neigh = ip_neigh_for_gw(rt, skb, &is_v6gw); if (!IS_ERR(neigh)) { int res; @@ -226,10 +226,10 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s sock_confirm_neigh(skb, neigh); /* if crossing protocols, can not use the cached header */ res = neigh_output(neigh, skb, is_v6gw); - rcu_read_unlock_bh(); + rcu_read_unlock(); return res; } - rcu_read_unlock_bh(); + rcu_read_unlock(); net_dbg_ratelimited("%s: No header cache and no neighbour!\n", __func__); @@ -420,17 +420,22 @@ int ip_mc_output(struct net *net, struct sock *sk, struct sk_buff *skb) int ip_output(struct net *net, struct sock *sk, struct sk_buff *skb) { - struct net_device *dev = skb_dst(skb)->dev, *indev = skb->dev; + struct net_device *dev, *indev = skb->dev; + int ret_val; IP_UPD_PO_STATS(net, IPSTATS_MIB_OUT, skb->len); + rcu_read_lock(); + dev = skb_dst_dev_rcu(skb); skb->dev = dev; skb->protocol = htons(ETH_P_IP); - return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, - net, sk, skb, indev, dev, - ip_finish_output, - !(IPCB(skb)->flags & IPSKB_REROUTED)); + ret_val = NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, + net, sk, skb, indev, dev, + ip_finish_output, + !(IPCB(skb)->flags & IPSKB_REROUTED)); + rcu_read_unlock(); + return ret_val; } EXPORT_SYMBOL(ip_output); diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c index 686c2680bcc9a..2dd14d1e22df1 100644 --- a/net/ipv4/nexthop.c +++ b/net/ipv4/nexthop.c @@ -1124,13 +1124,13 @@ static bool ipv6_good_nh(const struct fib6_nh *nh) int state = NUD_REACHABLE; struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv6_neigh_lookup_noref_stub(nh->fib_nh_dev, &nh->fib_nh_gw6); if (n) state = n->nud_state; - rcu_read_unlock_bh(); + rcu_read_unlock(); return !!(state & NUD_VALID); } @@ -1140,14 +1140,14 @@ static bool ipv4_good_nh(const struct fib_nh *nh) int state = NUD_REACHABLE; struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev, (__force u32)nh->fib_nh_gw4); if (n) state = n->nud_state; - rcu_read_unlock_bh(); + rcu_read_unlock(); return !!(state & NUD_VALID); } diff --git a/net/ipv4/route.c b/net/ipv4/route.c index ba24f371287bc..1ae721b7fcfd4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -416,7 +416,7 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst, struct net_device *dev = dst->dev; struct neighbour *n; - rcu_read_lock_bh(); + rcu_read_lock(); if (likely(rt->rt_gw_family == AF_INET)) { n = ip_neigh_gw4(dev, rt->rt_gw4); @@ -432,7 +432,7 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst, if (!IS_ERR(n) && !refcount_inc_not_zero(&n->refcnt)) n = NULL; - rcu_read_unlock_bh(); + rcu_read_unlock(); return n; } @@ -1022,9 +1022,9 @@ out: kfree_skb_reason(skb, reason); static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) { struct dst_entry *dst = &rt->dst; - struct net *net = dev_net(dst->dev); struct fib_result res; bool lock = false; + struct net *net; u32 old_mtu; if (ip_mtu_locked(dst)) @@ -1034,6 +1034,8 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) if (old_mtu < mtu) return; + rcu_read_lock(); + net = dst_dev_net_rcu(dst); if (mtu < ip_rt_min_pmtu) { lock = true; mtu = min(old_mtu, ip_rt_min_pmtu); @@ -1041,9 +1043,8 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) if (rt->rt_pmtu == mtu && !lock && time_before(jiffies, dst->expires - ip_rt_mtu_expires / 2)) - return; + goto out; - rcu_read_lock(); if (fib_lookup(net, fl4, &res, 0) == 0) { struct fib_nh_common *nhc; @@ -1052,6 +1053,7 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) update_or_create_fnhe(nhc, fl4->daddr, 0, mtu, lock, jiffies + ip_rt_mtu_expires); } +out: rcu_read_unlock(); } @@ -1325,26 +1327,7 @@ static unsigned int ipv4_default_advmss(const struct dst_entry *dst) INDIRECT_CALLABLE_SCOPE unsigned int ipv4_mtu(const struct dst_entry *dst) { - const struct rtable *rt = (const struct rtable *)dst; - unsigned int mtu = rt->rt_pmtu; - - if (!mtu || time_after_eq(jiffies, rt->dst.expires)) - mtu = dst_metric_raw(dst, RTAX_MTU); - - if (mtu) - goto out; - - mtu = READ_ONCE(dst->dev->mtu); - - if (unlikely(ip_mtu_locked(dst))) { - if (rt->rt_uses_gateway && mtu > 576) - mtu = 576; - } - -out: - mtu = min_t(unsigned int, mtu, IP_MAX_MTU); - - return mtu - lwtunnel_headroom(dst->lwtstate, mtu); + return ip_dst_mtu_maybe_forward(dst, false); } EXPORT_INDIRECT_CALLABLE(ipv4_mtu); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 1cbb81183f2bc..329b55ab8c218 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1017,7 +1017,7 @@ static int ipv6_add_addr_hash(struct net_device *dev, struct inet6_ifaddr *ifa) unsigned int hash = inet6_addr_hash(dev_net(dev), &ifa->addr); int err = 0; - spin_lock(&addrconf_hash_lock); + spin_lock_bh(&addrconf_hash_lock); /* Ignore adding duplicate addresses on an interface */ if (ipv6_chk_same_addr(dev_net(dev), &ifa->addr, dev, hash)) { @@ -1027,7 +1027,7 @@ static int ipv6_add_addr_hash(struct net_device *dev, struct inet6_ifaddr *ifa) hlist_add_head_rcu(&ifa->addr_lst, &inet6_addr_lst[hash]); } - spin_unlock(&addrconf_hash_lock); + spin_unlock_bh(&addrconf_hash_lock); return err; } @@ -1121,15 +1121,15 @@ ipv6_add_addr(struct inet6_dev *idev, struct ifa6_config *cfg, /* For caller */ refcount_set(&ifa->refcnt, 1); - rcu_read_lock_bh(); + rcu_read_lock(); err = ipv6_add_addr_hash(idev->dev, ifa); if (err < 0) { - rcu_read_unlock_bh(); + rcu_read_unlock(); goto out; } - write_lock(&idev->lock); + write_lock_bh(&idev->lock); /* Add to inet6_dev unicast addr list. */ ipv6_link_dev_addr(idev, ifa); @@ -1140,9 +1140,9 @@ ipv6_add_addr(struct inet6_dev *idev, struct ifa6_config *cfg, } in6_ifa_hold(ifa); - write_unlock(&idev->lock); + write_unlock_bh(&idev->lock); - rcu_read_unlock_bh(); + rcu_read_unlock(); inet6addr_notifier_call_chain(NETDEV_UP, ifa); out: diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 3b148663645a1..d027d6d844c0c 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -2493,7 +2493,7 @@ static int ipv6_route_native_seq_show(struct seq_file *seq, void *v) const struct net_device *dev; if (rt->nh) - fib6_nh = nexthop_fib6_nh_bh(rt->nh); + fib6_nh = nexthop_fib6_nh(rt->nh); seq_printf(seq, "%pi6 %02x ", &rt->fib6_dst.addr, rt->fib6_dst.plen); @@ -2558,14 +2558,14 @@ static struct fib6_table *ipv6_route_seq_next_table(struct fib6_table *tbl, if (tbl) { h = (tbl->tb6_id & (FIB6_TABLE_HASHSZ - 1)) + 1; - node = rcu_dereference_bh(hlist_next_rcu(&tbl->tb6_hlist)); + node = rcu_dereference(hlist_next_rcu(&tbl->tb6_hlist)); } else { h = 0; node = NULL; } while (!node && h < FIB6_TABLE_HASHSZ) { - node = rcu_dereference_bh( + node = rcu_dereference( hlist_first_rcu(&net->ipv6.fib_table_hash[h++])); } return hlist_entry_safe(node, struct fib6_table, tb6_hlist); @@ -2595,7 +2595,7 @@ static void *ipv6_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) if (!v) goto iter_table; - n = rcu_dereference_bh(((struct fib6_info *)v)->fib6_next); + n = rcu_dereference(((struct fib6_info *)v)->fib6_next); if (n) return n; @@ -2621,12 +2621,12 @@ static void *ipv6_route_seq_next(struct seq_file *seq, void *v, loff_t *pos) } static void *ipv6_route_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(RCU_BH) + __acquires(RCU) { struct net *net = seq_file_net(seq); struct ipv6_route_iter *iter = seq->private; - rcu_read_lock_bh(); + rcu_read_lock(); iter->tbl = ipv6_route_seq_next_table(NULL, net); iter->skip = *pos; @@ -2647,7 +2647,7 @@ static bool ipv6_route_iter_active(struct ipv6_route_iter *iter) } static void ipv6_route_native_seq_stop(struct seq_file *seq, void *v) - __releases(RCU_BH) + __releases(RCU) { struct net *net = seq_file_net(seq); struct ipv6_route_iter *iter = seq->private; @@ -2655,7 +2655,7 @@ static void ipv6_route_native_seq_stop(struct seq_file *seq, void *v) if (ipv6_route_iter_active(iter)) fib6_walker_unlink(net, &iter->w); - rcu_read_unlock_bh(); + rcu_read_unlock(); } #if IS_BUILTIN(CONFIG_IPV6) && defined(CONFIG_BPF_SYSCALL) diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index ceb85c67ce395..bf6731d135582 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -58,18 +58,18 @@ DEFINE_STATIC_KEY_DEFERRED_FALSE(ipv6_flowlabel_exclusive, HZ); EXPORT_SYMBOL(ipv6_flowlabel_exclusive); #define for_each_fl_rcu(hash, fl) \ - for (fl = rcu_dereference_bh(fl_ht[(hash)]); \ + for (fl = rcu_dereference(fl_ht[(hash)]); \ fl != NULL; \ - fl = rcu_dereference_bh(fl->next)) + fl = rcu_dereference(fl->next)) #define for_each_fl_continue_rcu(fl) \ - for (fl = rcu_dereference_bh(fl->next); \ + for (fl = rcu_dereference(fl->next); \ fl != NULL; \ - fl = rcu_dereference_bh(fl->next)) + fl = rcu_dereference(fl->next)) #define for_each_sk_fl_rcu(np, sfl) \ - for (sfl = rcu_dereference_bh(np->ipv6_fl_list); \ + for (sfl = rcu_dereference(np->ipv6_fl_list); \ sfl != NULL; \ - sfl = rcu_dereference_bh(sfl->next)) + sfl = rcu_dereference(sfl->next)) static inline struct ip6_flowlabel *__fl_lookup(struct net *net, __be32 label) { @@ -86,11 +86,11 @@ static struct ip6_flowlabel *fl_lookup(struct net *net, __be32 label) { struct ip6_flowlabel *fl; - rcu_read_lock_bh(); + rcu_read_lock(); fl = __fl_lookup(net, label); if (fl && !atomic_inc_not_zero(&fl->users)) fl = NULL; - rcu_read_unlock_bh(); + rcu_read_unlock(); return fl; } @@ -217,6 +217,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net, fl->label = label & IPV6_FLOWLABEL_MASK; + rcu_read_lock(); spin_lock_bh(&ip6_fl_lock); if (label == 0) { for (;;) { @@ -240,6 +241,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net, if (lfl) { atomic_inc(&lfl->users); spin_unlock_bh(&ip6_fl_lock); + rcu_read_unlock(); return lfl; } } @@ -249,6 +251,7 @@ static struct ip6_flowlabel *fl_intern(struct net *net, rcu_assign_pointer(fl_ht[FL_HASH(fl->label)], fl); atomic_inc(&fl_size); spin_unlock_bh(&ip6_fl_lock); + rcu_read_unlock(); return NULL; } @@ -263,17 +266,17 @@ struct ip6_flowlabel *__fl6_sock_lookup(struct sock *sk, __be32 label) label &= IPV6_FLOWLABEL_MASK; - rcu_read_lock_bh(); + rcu_read_lock(); for_each_sk_fl_rcu(np, sfl) { struct ip6_flowlabel *fl = sfl->fl; if (fl->label == label && atomic_inc_not_zero(&fl->users)) { fl->lastuse = jiffies; - rcu_read_unlock_bh(); + rcu_read_unlock(); return fl; } } - rcu_read_unlock_bh(); + rcu_read_unlock(); return NULL; } EXPORT_SYMBOL_GPL(__fl6_sock_lookup); @@ -475,10 +478,10 @@ static int mem_check(struct sock *sk) if (room > FL_MAX_SIZE - FL_MAX_PER_SOCK) return 0; - rcu_read_lock_bh(); + rcu_read_lock(); for_each_sk_fl_rcu(np, sfl) count++; - rcu_read_unlock_bh(); + rcu_read_unlock(); if (room <= 0 || ((count >= FL_MAX_PER_SOCK || @@ -515,7 +518,7 @@ int ipv6_flowlabel_opt_get(struct sock *sk, struct in6_flowlabel_req *freq, return 0; } - rcu_read_lock_bh(); + rcu_read_lock(); for_each_sk_fl_rcu(np, sfl) { if (sfl->fl->label == (np->flow_label & IPV6_FLOWLABEL_MASK)) { @@ -527,11 +530,11 @@ int ipv6_flowlabel_opt_get(struct sock *sk, struct in6_flowlabel_req *freq, freq->flr_linger = sfl->fl->linger / HZ; spin_unlock_bh(&ip6_fl_lock); - rcu_read_unlock_bh(); + rcu_read_unlock(); return 0; } } - rcu_read_unlock_bh(); + rcu_read_unlock(); return -ENOENT; } @@ -581,16 +584,16 @@ static int ipv6_flowlabel_renew(struct sock *sk, struct in6_flowlabel_req *freq) struct ipv6_fl_socklist *sfl; int err; - rcu_read_lock_bh(); + rcu_read_lock(); for_each_sk_fl_rcu(np, sfl) { if (sfl->fl->label == freq->flr_label) { err = fl6_renew(sfl->fl, freq->flr_linger, freq->flr_expires); - rcu_read_unlock_bh(); + rcu_read_unlock(); return err; } } - rcu_read_unlock_bh(); + rcu_read_unlock(); if (freq->flr_share == IPV6_FL_S_NONE && ns_capable(net->user_ns, CAP_NET_ADMIN)) { @@ -641,11 +644,11 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq, if (freq->flr_label) { err = -EEXIST; - rcu_read_lock_bh(); + rcu_read_lock(); for_each_sk_fl_rcu(np, sfl) { if (sfl->fl->label == freq->flr_label) { if (freq->flr_flags & IPV6_FL_F_EXCL) { - rcu_read_unlock_bh(); + rcu_read_unlock(); goto done; } fl1 = sfl->fl; @@ -654,7 +657,7 @@ static int ipv6_flowlabel_get(struct sock *sk, struct in6_flowlabel_req *freq, break; } } - rcu_read_unlock_bh(); + rcu_read_unlock(); if (!fl1) fl1 = fl_lookup(net, freq->flr_label); @@ -809,7 +812,7 @@ static void *ip6fl_seq_start(struct seq_file *seq, loff_t *pos) state->pid_ns = proc_pid_ns(file_inode(seq->file)->i_sb); - rcu_read_lock_bh(); + rcu_read_lock(); return *pos ? ip6fl_get_idx(seq, *pos - 1) : SEQ_START_TOKEN; } @@ -828,7 +831,7 @@ static void *ip6fl_seq_next(struct seq_file *seq, void *v, loff_t *pos) static void ip6fl_seq_stop(struct seq_file *seq, void *v) __releases(RCU) { - rcu_read_unlock_bh(); + rcu_read_unlock(); } static int ip6fl_seq_show(struct seq_file *seq, void *v) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 40430d1e9d3f9..437dc38b85e4f 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -59,7 +59,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff *skb) { struct dst_entry *dst = skb_dst(skb); - struct net_device *dev = dst->dev; + struct net_device *dev = dst_dev_rcu(dst); struct inet6_dev *idev = ip6_dst_idev(dst); unsigned int hh_len = LL_RESERVED_SPACE(dev); const struct in6_addr *daddr, *nexthop; @@ -69,6 +69,7 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * /* Be paranoid, rather than too clever. */ if (unlikely(hh_len > skb_headroom(skb)) && dev->header_ops) { + /* idev stays alive because we hold rcu_read_lock(). */ skb = skb_expand_head(skb, hh_len); if (!skb) { IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); @@ -116,22 +117,21 @@ static int ip6_finish_output2(struct net *net, struct sock *sk, struct sk_buff * return res; } - rcu_read_lock_bh(); nexthop = rt6_nexthop((struct rt6_info *)dst, daddr); neigh = __ipv6_neigh_lookup_noref(dev, nexthop); - if (unlikely(!neigh)) - neigh = __neigh_create(&nd_tbl, nexthop, dev, false); - if (!IS_ERR(neigh)) { - sock_confirm_neigh(skb, neigh); - ret = neigh_output(neigh, skb, false); - rcu_read_unlock_bh(); - return ret; - } - rcu_read_unlock_bh(); - IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTNOROUTES); - kfree_skb_reason(skb, SKB_DROP_REASON_NEIGH_CREATEFAIL); - return -EINVAL; + if (unlikely(IS_ERR_OR_NULL(neigh))) { + if (unlikely(!neigh)) + neigh = __neigh_create(&nd_tbl, nexthop, dev, false); + if (IS_ERR(neigh)) { + IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTNOROUTES); + kfree_skb_reason(skb, SKB_DROP_REASON_NEIGH_CREATEFAIL); + return -EINVAL; + } + } + sock_confirm_neigh(skb, neigh); + ret = neigh_output(neigh, skb, false); + return ret; } static int @@ -211,22 +211,30 @@ static int ip6_finish_output(struct net *net, struct sock *sk, struct sk_buff *s int ip6_output(struct net *net, struct sock *sk, struct sk_buff *skb) { - struct net_device *dev = skb_dst(skb)->dev, *indev = skb->dev; - struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb)); + struct dst_entry *dst = skb_dst(skb); + struct net_device *dev, *indev = skb->dev; + struct inet6_dev *idev; + int ret; skb->protocol = htons(ETH_P_IPV6); + rcu_read_lock(); + dev = dst_dev_rcu(dst); + idev = ip6_dst_idev(dst); skb->dev = dev; if (unlikely(idev->cnf.disable_ipv6)) { IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); + rcu_read_unlock(); kfree_skb_reason(skb, SKB_DROP_REASON_IPV6DISABLED); return 0; } - return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, - net, sk, skb, indev, dev, - ip6_finish_output, - !(IP6CB(skb)->flags & IP6SKB_REROUTED)); + ret = NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, + net, sk, skb, indev, dev, + ip6_finish_output, + !(IP6CB(skb)->flags & IP6SKB_REROUTED)); + rcu_read_unlock(); + return ret; } EXPORT_SYMBOL(ip6_output); @@ -247,30 +255,35 @@ bool ip6_autoflowlabel(struct net *net, const struct ipv6_pinfo *np) int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, __u32 mark, struct ipv6_txoptions *opt, int tclass, u32 priority) { - struct net *net = sock_net(sk); const struct ipv6_pinfo *np = inet6_sk(sk); struct in6_addr *first_hop = &fl6->daddr; struct dst_entry *dst = skb_dst(skb); - struct net_device *dev = dst->dev; struct inet6_dev *idev = ip6_dst_idev(dst); struct hop_jumbo_hdr *hop_jumbo; int hoplen = sizeof(*hop_jumbo); + struct net *net = sock_net(sk); unsigned int head_room; + struct net_device *dev; struct ipv6hdr *hdr; u8 proto = fl6->flowi6_proto; int seg_len = skb->len; - int hlimit = -1; + int ret, hlimit = -1; u32 mtu; + rcu_read_lock(); + + dev = dst_dev_rcu(dst); head_room = sizeof(struct ipv6hdr) + hoplen + LL_RESERVED_SPACE(dev); if (opt) head_room += opt->opt_nflen + opt->opt_flen; if (unlikely(head_room > skb_headroom(skb))) { + /* idev stays alive while we hold rcu_read_lock(). */ skb = skb_expand_head(skb, head_room); if (!skb) { IP6_INC_STATS(net, idev, IPSTATS_MIB_OUTDISCARDS); - return -ENOBUFS; + ret = -ENOBUFS; + goto unlock; } } @@ -333,17 +346,21 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, * skb to its handler for processing */ skb = l3mdev_ip6_out((struct sock *)sk, skb); - if (unlikely(!skb)) - return 0; + if (unlikely(!skb)) { + ret = 0; + goto unlock; + } /* hooks should never assume socket lock is held. * we promote our socket to non const */ - return NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, - net, (struct sock *)sk, skb, NULL, dev, - dst_output); + ret = NF_HOOK(NFPROTO_IPV6, NF_INET_LOCAL_OUT, + net, (struct sock *)sk, skb, NULL, dev, + dst_output); + goto unlock; } + ret = -EMSGSIZE; skb->dev = dev; /* ipv6_local_error() does not require socket lock, * we promote our socket to non const @@ -352,7 +369,9 @@ int ip6_xmit(const struct sock *sk, struct sk_buff *skb, struct flowi6 *fl6, IP6_INC_STATS(net, idev, IPSTATS_MIB_FRAGFAILS); kfree_skb(skb); - return -EMSGSIZE; +unlock: + rcu_read_unlock(); + return ret; } EXPORT_SYMBOL(ip6_xmit); @@ -618,7 +637,7 @@ int ip6_forward(struct sk_buff *skb) } } - mtu = ip6_dst_mtu_forward(dst); + mtu = ip6_dst_mtu_maybe_forward(dst, true); if (mtu < IPV6_MIN_MTU) mtu = IPV6_MIN_MTU; @@ -1149,11 +1168,11 @@ static int ip6_dst_lookup_tail(struct net *net, const struct sock *sk, * dst entry of the nexthop router */ rt = (struct rt6_info *) *dst; - rcu_read_lock_bh(); + rcu_read_lock(); n = __ipv6_neigh_lookup_noref(rt->dst.dev, rt6_nexthop(rt, &fl6->daddr)); err = n && !(n->nud_state & NUD_VALID) ? -EINVAL : 0; - rcu_read_unlock_bh(); + rcu_read_unlock(); if (err) { struct inet6_ifaddr *ifp; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a5ad8fea1078c..257fd3daccc37 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -631,7 +631,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh) nh_gw = &fib6_nh->fib_nh_gw6; dev = fib6_nh->fib_nh_dev; - rcu_read_lock_bh(); + rcu_read_lock(); last_probe = READ_ONCE(fib6_nh->last_probe); idev = __in6_dev_get(dev); if (!idev) @@ -641,7 +641,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh) if (neigh->nud_state & NUD_VALID) goto out; - write_lock(&neigh->lock); + write_lock_bh(&neigh->lock); if (!(neigh->nud_state & NUD_VALID) && time_after(jiffies, neigh->updated + idev->cnf.rtr_probe_interval)) { @@ -649,7 +649,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh) if (work) __neigh_set_probe_once(neigh); } - write_unlock(&neigh->lock); + write_unlock_bh(&neigh->lock); } else if (time_after(jiffies, last_probe + idev->cnf.rtr_probe_interval)) { work = kmalloc(sizeof(*work), GFP_ATOMIC); @@ -667,7 +667,7 @@ static void rt6_probe(struct fib6_nh *fib6_nh) } out: - rcu_read_unlock_bh(); + rcu_read_unlock(); } #else static inline void rt6_probe(struct fib6_nh *fib6_nh) @@ -683,25 +683,25 @@ static enum rt6_nud_state rt6_check_neigh(const struct fib6_nh *fib6_nh) enum rt6_nud_state ret = RT6_NUD_FAIL_HARD; struct neighbour *neigh; - rcu_read_lock_bh(); + rcu_read_lock(); neigh = __ipv6_neigh_lookup_noref(fib6_nh->fib_nh_dev, &fib6_nh->fib_nh_gw6); if (neigh) { - read_lock(&neigh->lock); - if (neigh->nud_state & NUD_VALID) + u8 nud_state = READ_ONCE(neigh->nud_state); + + if (nud_state & NUD_VALID) ret = RT6_NUD_SUCCEED; #ifdef CONFIG_IPV6_ROUTER_PREF - else if (!(neigh->nud_state & NUD_FAILED)) + else if (!(nud_state & NUD_FAILED)) ret = RT6_NUD_SUCCEED; else ret = RT6_NUD_FAIL_PROBE; #endif - read_unlock(&neigh->lock); } else { ret = IS_ENABLED(CONFIG_IPV6_ROUTER_PREF) ? RT6_NUD_SUCCEED : RT6_NUD_FAIL_DO_RR; } - rcu_read_unlock_bh(); + rcu_read_unlock(); return ret; } @@ -3183,13 +3183,18 @@ static unsigned int ip6_default_advmss(const struct dst_entry *dst) { struct net_device *dev = dst->dev; unsigned int mtu = dst_mtu(dst); - struct net *net = dev_net(dev); + struct net *net; mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr); + rcu_read_lock(); + + net = dev_net_rcu(dev); if (mtu < net->ipv6.sysctl.ip6_rt_min_advmss) mtu = net->ipv6.sysctl.ip6_rt_min_advmss; + rcu_read_unlock(); + /* * Maximal non-jumbo IPv6 payload is IPV6_MAXPLEN and * corresponding MSS is IPV6_MAXPLEN - tcp_header_size. @@ -3203,25 +3208,7 @@ static unsigned int ip6_default_advmss(const struct dst_entry *dst) INDIRECT_CALLABLE_SCOPE unsigned int ip6_mtu(const struct dst_entry *dst) { - struct inet6_dev *idev; - unsigned int mtu; - - mtu = dst_metric_raw(dst, RTAX_MTU); - if (mtu) - goto out; - - mtu = IPV6_MIN_MTU; - - rcu_read_lock(); - idev = __in6_dev_get(dst->dev); - if (idev) - mtu = idev->cnf.mtu6; - rcu_read_unlock(); - -out: - mtu = min_t(unsigned int, mtu, IP6_MAX_MTU); - - return mtu - lwtunnel_headroom(dst->lwtstate, mtu); + return ip6_dst_mtu_maybe_forward(dst, false); } EXPORT_INDIRECT_CALLABLE(ip6_mtu); diff --git a/net/netfilter/nf_flow_table_core.c b/net/netfilter/nf_flow_table_core.c index e8d5d30fd00a7..fee0ccd7fee36 100644 --- a/net/netfilter/nf_flow_table_core.c +++ b/net/netfilter/nf_flow_table_core.c @@ -99,7 +99,7 @@ static int flow_offload_fill_route(struct flow_offload *flow, flow_tuple->mtu = ip_dst_mtu_maybe_forward(dst, true); break; case NFPROTO_IPV6: - flow_tuple->mtu = ip6_dst_mtu_forward(dst); + flow_tuple->mtu = ip6_dst_mtu_maybe_forward(dst, true); break; }