diff options
Diffstat (limited to 'net/ipv4')
-rw-r--r-- | net/ipv4/esp4.c | 3 | ||||
-rw-r--r-- | net/ipv4/fib_semantics.c | 18 | ||||
-rw-r--r-- | net/ipv4/fib_trie.c | 1 | ||||
-rw-r--r-- | net/ipv4/fou_core.c | 31 | ||||
-rw-r--r-- | net/ipv4/inet_timewait_sock.c | 16 | ||||
-rw-r--r-- | net/ipv4/metrics.c | 8 | ||||
-rw-r--r-- | net/ipv4/netfilter/iptable_nat.c | 18 | ||||
-rw-r--r-- | net/ipv4/nexthop.c | 7 | ||||
-rw-r--r-- | net/ipv4/route.c | 18 | ||||
-rw-r--r-- | net/ipv4/syncookies.c | 3 | ||||
-rw-r--r-- | net/ipv4/tcp.c | 10 | ||||
-rw-r--r-- | net/ipv4/tcp_bpf.c | 2 | ||||
-rw-r--r-- | net/ipv4/tcp_cong.c | 11 | ||||
-rw-r--r-- | net/ipv4/tcp_input.c | 71 | ||||
-rw-r--r-- | net/ipv4/tcp_ipv4.c | 27 | ||||
-rw-r--r-- | net/ipv4/tcp_metrics.c | 7 | ||||
-rw-r--r-- | net/ipv4/tcp_minisocks.c | 7 | ||||
-rw-r--r-- | net/ipv4/tcp_offload.c | 3 | ||||
-rw-r--r-- | net/ipv4/tcp_output.c | 18 | ||||
-rw-r--r-- | net/ipv4/tcp_timer.c | 37 | ||||
-rw-r--r-- | net/ipv4/udp.c | 4 | ||||
-rw-r--r-- | net/ipv4/udp_offload.c | 5 |
22 files changed, 212 insertions, 113 deletions
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index fe501d2186bc..eeace9b509ce 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -238,8 +238,7 @@ static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb) #else static int esp_output_tail_tcp(struct xfrm_state *x, struct sk_buff *skb) { - kfree_skb(skb); - + WARN_ON(1); return -EOPNOTSUPP; } #endif diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 5eb1b8d302bb..233d9d0437c2 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -1030,7 +1030,7 @@ bool fib_metrics_match(struct fib_config *cfg, struct fib_info *fi) bool ecn_ca = false; nla_strscpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(fi->fib_net, tmp, &ecn_ca); + val = tcp_ca_get_key_by_name(tmp, &ecn_ca); } else { if (nla_len(nla) != sizeof(u32)) return false; @@ -1459,8 +1459,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg, fi = kzalloc(struct_size(fi, fib_nh, nhs), GFP_KERNEL); if (!fi) goto failure; - fi->fib_metrics = ip_fib_metrics_init(fi->fib_net, cfg->fc_mx, - cfg->fc_mx_len, extack); + fi->fib_metrics = ip_fib_metrics_init(cfg->fc_mx, cfg->fc_mx_len, extack); if (IS_ERR(fi->fib_metrics)) { err = PTR_ERR(fi->fib_metrics); kfree(fi); @@ -2270,6 +2269,15 @@ void fib_select_path(struct net *net, struct fib_result *res, fib_select_default(fl4, res); check_saddr: - if (!fl4->saddr) - fl4->saddr = fib_result_prefsrc(net, res); + if (!fl4->saddr) { + struct net_device *l3mdev; + + l3mdev = dev_get_by_index_rcu(net, fl4->flowi4_l3mdev); + + if (!l3mdev || + l3mdev_master_dev_rcu(FIB_RES_DEV(*res)) == l3mdev) + fl4->saddr = fib_result_prefsrc(net, res); + else + fl4->saddr = inet_select_addr(l3mdev, 0, RT_SCOPE_LINK); + } } diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 9bdfdab906fe..77b97c48da5e 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -1628,6 +1628,7 @@ set_result: res->nhc = nhc; res->type = fa->fa_type; res->scope = fi->fib_scope; + res->dscp = fa->fa_dscp; res->fi = fi; res->table = tb; res->fa_head = &n->leaf; diff --git a/net/ipv4/fou_core.c b/net/ipv4/fou_core.c index 0c41076e31ed..4e0a7d038e21 100644 --- a/net/ipv4/fou_core.c +++ b/net/ipv4/fou_core.c @@ -50,7 +50,7 @@ struct fou_net { static inline struct fou *fou_from_sock(struct sock *sk) { - return sk->sk_user_data; + return rcu_dereference_sk_user_data(sk); } static int fou_recv_pull(struct sk_buff *skb, struct fou *fou, size_t len) @@ -233,9 +233,15 @@ static struct sk_buff *fou_gro_receive(struct sock *sk, struct sk_buff *skb) { const struct net_offload __rcu **offloads; - u8 proto = fou_from_sock(sk)->protocol; + struct fou *fou = fou_from_sock(sk); const struct net_offload *ops; struct sk_buff *pp = NULL; + u8 proto; + + if (!fou) + goto out; + + proto = fou->protocol; /* We can clear the encap_mark for FOU as we are essentially doing * one of two possible things. We are either adding an L4 tunnel @@ -263,14 +269,24 @@ static int fou_gro_complete(struct sock *sk, struct sk_buff *skb, int nhoff) { const struct net_offload __rcu **offloads; - u8 proto = fou_from_sock(sk)->protocol; + struct fou *fou = fou_from_sock(sk); const struct net_offload *ops; - int err = -ENOSYS; + u8 proto; + int err; + + if (!fou) { + err = -ENOENT; + goto out; + } + + proto = fou->protocol; offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]); - if (WARN_ON(!ops || !ops->callbacks.gro_complete)) + if (WARN_ON(!ops || !ops->callbacks.gro_complete)) { + err = -ENOSYS; goto out; + } err = ops->callbacks.gro_complete(skb, nhoff); @@ -322,6 +338,9 @@ static struct sk_buff *gue_gro_receive(struct sock *sk, skb_gro_remcsum_init(&grc); + if (!fou) + goto out; + off = skb_gro_offset(skb); len = off + sizeof(*guehdr); @@ -433,7 +452,7 @@ next_proto: offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads; ops = rcu_dereference(offloads[proto]); - if (WARN_ON_ONCE(!ops || !ops->callbacks.gro_receive)) + if (!ops || !ops->callbacks.gro_receive) goto out; pp = call_gro_receive(ops->callbacks.gro_receive, head, skb); diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index 757ae3a4e2f1..fff53144250c 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -279,14 +279,18 @@ void __inet_twsk_schedule(struct inet_timewait_sock *tw, int timeo, bool rearm) EXPORT_SYMBOL_GPL(__inet_twsk_schedule); /* Remove all non full sockets (TIME_WAIT and NEW_SYN_RECV) for dead netns */ -void inet_twsk_purge(struct inet_hashinfo *hashinfo, int family) +void inet_twsk_purge(struct inet_hashinfo *hashinfo) { + struct inet_ehash_bucket *head = &hashinfo->ehash[0]; + unsigned int ehash_mask = hashinfo->ehash_mask; struct hlist_nulls_node *node; unsigned int slot; struct sock *sk; - for (slot = 0; slot <= hashinfo->ehash_mask; slot++) { - struct inet_ehash_bucket *head = &hashinfo->ehash[slot]; + for (slot = 0; slot <= ehash_mask; slot++, head++) { + if (hlist_nulls_empty(&head->chain)) + continue; + restart_rcu: cond_resched(); rcu_read_lock(); @@ -298,15 +302,13 @@ restart: TCPF_NEW_SYN_RECV)) continue; - if (sk->sk_family != family || - refcount_read(&sock_net(sk)->ns.count)) + if (refcount_read(&sock_net(sk)->ns.count)) continue; if (unlikely(!refcount_inc_not_zero(&sk->sk_refcnt))) continue; - if (unlikely(sk->sk_family != family || - refcount_read(&sock_net(sk)->ns.count))) { + if (refcount_read(&sock_net(sk)->ns.count)) { sock_gen_put(sk); goto restart; } diff --git a/net/ipv4/metrics.c b/net/ipv4/metrics.c index 0e3ee1532848..8ddac1f595ed 100644 --- a/net/ipv4/metrics.c +++ b/net/ipv4/metrics.c @@ -7,7 +7,7 @@ #include <net/net_namespace.h> #include <net/tcp.h> -static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx, +static int ip_metrics_convert(struct nlattr *fc_mx, int fc_mx_len, u32 *metrics, struct netlink_ext_ack *extack) { @@ -31,7 +31,7 @@ static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx, char tmp[TCP_CA_NAME_MAX]; nla_strscpy(tmp, nla, sizeof(tmp)); - val = tcp_ca_get_key_by_name(net, tmp, &ecn_ca); + val = tcp_ca_get_key_by_name(tmp, &ecn_ca); if (val == TCP_CA_UNSPEC) { NL_SET_ERR_MSG(extack, "Unknown tcp congestion algorithm"); return -EINVAL; @@ -63,7 +63,7 @@ static int ip_metrics_convert(struct net *net, struct nlattr *fc_mx, return 0; } -struct dst_metrics *ip_fib_metrics_init(struct net *net, struct nlattr *fc_mx, +struct dst_metrics *ip_fib_metrics_init(struct nlattr *fc_mx, int fc_mx_len, struct netlink_ext_ack *extack) { @@ -77,7 +77,7 @@ struct dst_metrics *ip_fib_metrics_init(struct net *net, struct nlattr *fc_mx, if (unlikely(!fib_metrics)) return ERR_PTR(-ENOMEM); - err = ip_metrics_convert(net, fc_mx, fc_mx_len, fib_metrics->metrics, + err = ip_metrics_convert(fc_mx, fc_mx_len, fib_metrics->metrics, extack); if (!err) { refcount_set(&fib_metrics->refcnt, 1); diff --git a/net/ipv4/netfilter/iptable_nat.c b/net/ipv4/netfilter/iptable_nat.c index 56f6ecc43451..12ca666d6e2c 100644 --- a/net/ipv4/netfilter/iptable_nat.c +++ b/net/ipv4/netfilter/iptable_nat.c @@ -145,25 +145,27 @@ static struct pernet_operations iptable_nat_net_ops = { static int __init iptable_nat_init(void) { - int ret = xt_register_template(&nf_nat_ipv4_table, - iptable_nat_table_init); + int ret; + /* net->gen->ptr[iptable_nat_net_id] must be allocated + * before calling iptable_nat_table_init(). + */ + ret = register_pernet_subsys(&iptable_nat_net_ops); if (ret < 0) return ret; - ret = register_pernet_subsys(&iptable_nat_net_ops); - if (ret < 0) { - xt_unregister_template(&nf_nat_ipv4_table); - return ret; - } + ret = xt_register_template(&nf_nat_ipv4_table, + iptable_nat_table_init); + if (ret < 0) + unregister_pernet_subsys(&iptable_nat_net_ops); return ret; } static void __exit iptable_nat_exit(void) { - unregister_pernet_subsys(&iptable_nat_net_ops); xt_unregister_template(&nf_nat_ipv4_table); + unregister_pernet_subsys(&iptable_nat_net_ops); } module_init(iptable_nat_init); diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c index bbff68b5b5d4..8d41b0394219 100644 --- a/net/ipv4/nexthop.c +++ b/net/ipv4/nexthop.c @@ -676,9 +676,10 @@ static int nla_put_nh_group(struct sk_buff *skb, struct nh_group *nhg) p = nla_data(nla); for (i = 0; i < nhg->num_nh; ++i) { - p->id = nhg->nh_entries[i].nh->id; - p->weight = nhg->nh_entries[i].weight - 1; - p += 1; + *p++ = (struct nexthop_grp) { + .id = nhg->nh_entries[i].nh->id, + .weight = nhg->nh_entries[i].weight - 1, + }; } if (nhg->resilient && nla_put_nh_group_res(skb, nhg)) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 40b9c579c917..285482060082 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1275,7 +1275,7 @@ void ip_rt_get_source(u8 *addr, struct sk_buff *skb, struct rtable *rt) struct flowi4 fl4 = { .daddr = iph->daddr, .saddr = iph->saddr, - .flowi4_tos = RT_TOS(iph->tos), + .flowi4_tos = iph->tos & IPTOS_RT_MASK, .flowi4_oif = rt->dst.dev->ifindex, .flowi4_iif = skb->dev->ifindex, .flowi4_mark = skb->mark, @@ -2930,9 +2930,9 @@ EXPORT_SYMBOL_GPL(ip_route_output_tunnel); /* called with rcu_read_lock held */ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, - struct rtable *rt, u32 table_id, struct flowi4 *fl4, - struct sk_buff *skb, u32 portid, u32 seq, - unsigned int flags) + struct rtable *rt, u32 table_id, dscp_t dscp, + struct flowi4 *fl4, struct sk_buff *skb, u32 portid, + u32 seq, unsigned int flags) { struct rtmsg *r; struct nlmsghdr *nlh; @@ -2948,7 +2948,7 @@ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, r->rtm_family = AF_INET; r->rtm_dst_len = 32; r->rtm_src_len = 0; - r->rtm_tos = fl4 ? fl4->flowi4_tos : 0; + r->rtm_tos = inet_dscp_to_dsfield(dscp); r->rtm_table = table_id < 256 ? table_id : RT_TABLE_COMPAT; if (nla_put_u32(skb, RTA_TABLE, table_id)) goto nla_put_failure; @@ -3098,7 +3098,7 @@ static int fnhe_dump_bucket(struct net *net, struct sk_buff *skb, goto next; err = rt_fill_info(net, fnhe->fnhe_daddr, 0, rt, - table_id, NULL, skb, + table_id, 0, NULL, skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq, flags); if (err) @@ -3394,7 +3394,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, fri.tb_id = table_id; fri.dst = res.prefix; fri.dst_len = res.prefixlen; - fri.dscp = inet_dsfield_to_dscp(fl4.flowi4_tos); + fri.dscp = res.dscp; fri.type = rt->rt_type; fri.offload = 0; fri.trap = 0; @@ -3421,8 +3421,8 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, err = fib_dump_info(skb, NETLINK_CB(in_skb).portid, nlh->nlmsg_seq, RTM_NEWROUTE, &fri, 0); } else { - err = rt_fill_info(net, dst, src, rt, table_id, &fl4, skb, - NETLINK_CB(in_skb).portid, + err = rt_fill_info(net, dst, src, rt, table_id, res.dscp, &fl4, + skb, NETLINK_CB(in_skb).portid, nlh->nlmsg_seq, 0); } if (err < 0) diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 3b4dafefb4b0..e14356207795 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -424,7 +424,8 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) } /* Try to redo what tcp_v4_send_synack did. */ - req->rsk_window_clamp = tp->window_clamp ? :dst_metric(&rt->dst, RTAX_WINDOW); + req->rsk_window_clamp = READ_ONCE(tp->window_clamp) ? : + dst_metric(&rt->dst, RTAX_WINDOW); /* limit the window selection if the user enforce a smaller rx buffer */ full_space = tcp_full_space(sk); if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 2df05ea2e00f..1e3202f2b7a8 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -591,7 +591,7 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait) */ mask |= EPOLLOUT | EPOLLWRNORM; } - /* This barrier is coupled with smp_wmb() in tcp_reset() */ + /* This barrier is coupled with smp_wmb() in tcp_done_with_error() */ smp_rmb(); if (READ_ONCE(sk->sk_err) || !skb_queue_empty_lockless(&sk->sk_error_queue)) @@ -1723,7 +1723,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val) space = tcp_space_from_win(sk, val); if (space > sk->sk_rcvbuf) { WRITE_ONCE(sk->sk_rcvbuf, space); - tcp_sk(sk)->window_clamp = val; + WRITE_ONCE(tcp_sk(sk)->window_clamp, val); } return 0; } @@ -3386,7 +3386,7 @@ int tcp_set_window_clamp(struct sock *sk, int val) if (!val) { if (sk->sk_state != TCP_CLOSE) return -EINVAL; - tp->window_clamp = 0; + WRITE_ONCE(tp->window_clamp, 0); } else { u32 new_rcv_ssthresh, old_window_clamp = tp->window_clamp; u32 new_window_clamp = val < SOCK_MIN_RCVBUF / 2 ? @@ -3395,7 +3395,7 @@ int tcp_set_window_clamp(struct sock *sk, int val) if (new_window_clamp == old_window_clamp) return 0; - tp->window_clamp = new_window_clamp; + WRITE_ONCE(tp->window_clamp, new_window_clamp); if (new_window_clamp < old_window_clamp) { /* need to apply the reserved mem provisioning only * when shrinking the window clamp @@ -4020,7 +4020,7 @@ int do_tcp_getsockopt(struct sock *sk, int level, TCP_RTO_MAX / HZ); break; case TCP_WINDOW_CLAMP: - val = tp->window_clamp; + val = READ_ONCE(tp->window_clamp); break; case TCP_INFO: { struct tcp_info info; diff --git a/net/ipv4/tcp_bpf.c b/net/ipv4/tcp_bpf.c index 53b0d62fd2c2..fe6178715ba0 100644 --- a/net/ipv4/tcp_bpf.c +++ b/net/ipv4/tcp_bpf.c @@ -577,7 +577,7 @@ out_err: err = sk_stream_error(sk, msg->msg_flags, err); release_sock(sk); sk_psock_put(sk, psock); - return copied ? copied : err; + return copied > 0 ? copied : err; } enum { diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 1b34050a7538..95dbb2799be4 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -46,8 +46,7 @@ void tcp_set_ca_state(struct sock *sk, const u8 ca_state) } /* Must be called with rcu lock held */ -static struct tcp_congestion_ops *tcp_ca_find_autoload(struct net *net, - const char *name) +static struct tcp_congestion_ops *tcp_ca_find_autoload(const char *name) { struct tcp_congestion_ops *ca = tcp_ca_find(name); @@ -182,7 +181,7 @@ int tcp_update_congestion_control(struct tcp_congestion_ops *ca, struct tcp_cong return ret; } -u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca) +u32 tcp_ca_get_key_by_name(const char *name, bool *ecn_ca) { const struct tcp_congestion_ops *ca; u32 key = TCP_CA_UNSPEC; @@ -190,7 +189,7 @@ u32 tcp_ca_get_key_by_name(struct net *net, const char *name, bool *ecn_ca) might_sleep(); rcu_read_lock(); - ca = tcp_ca_find_autoload(net, name); + ca = tcp_ca_find_autoload(name); if (ca) { key = ca->key; *ecn_ca = ca->flags & TCP_CONG_NEEDS_ECN; @@ -287,7 +286,7 @@ int tcp_set_default_congestion_control(struct net *net, const char *name) int ret; rcu_read_lock(); - ca = tcp_ca_find_autoload(net, name); + ca = tcp_ca_find_autoload(name); if (!ca) { ret = -ENOENT; } else if (!bpf_try_module_get(ca, ca->owner)) { @@ -425,7 +424,7 @@ int tcp_set_congestion_control(struct sock *sk, const char *name, bool load, if (!load) ca = tcp_ca_find(name); else - ca = tcp_ca_find_autoload(sock_net(sk), name); + ca = tcp_ca_find_autoload(name); /* No change asking for existing value */ if (ca == icsk->icsk_ca_ops) { diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 2621c262f182..b565c8a8e7ba 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -243,9 +243,14 @@ static void tcp_measure_rcv_mss(struct sock *sk, const struct sk_buff *skb) */ if (unlikely(len != icsk->icsk_ack.rcv_mss)) { u64 val = (u64)skb->len << TCP_RMEM_TO_WIN_SCALE; + u8 old_ratio = tcp_sk(sk)->scaling_ratio; do_div(val, skb->truesize); tcp_sk(sk)->scaling_ratio = val ? val : 1; + + if (old_ratio != tcp_sk(sk)->scaling_ratio) + WRITE_ONCE(tcp_sk(sk)->window_clamp, + tcp_win_from_space(sk, sk->sk_rcvbuf)); } icsk->icsk_ack.rcv_mss = min_t(unsigned int, len, tcp_sk(sk)->advmss); @@ -570,19 +575,20 @@ static void tcp_init_buffer_space(struct sock *sk) maxwin = tcp_full_space(sk); if (tp->window_clamp >= maxwin) { - tp->window_clamp = maxwin; + WRITE_ONCE(tp->window_clamp, maxwin); if (tcp_app_win && maxwin > 4 * tp->advmss) - tp->window_clamp = max(maxwin - - (maxwin >> tcp_app_win), - 4 * tp->advmss); + WRITE_ONCE(tp->window_clamp, + max(maxwin - (maxwin >> tcp_app_win), + 4 * tp->advmss)); } /* Force reservation of one segment. */ if (tcp_app_win && tp->window_clamp > 2 * tp->advmss && tp->window_clamp + tp->advmss > maxwin) - tp->window_clamp = max(2 * tp->advmss, maxwin - tp->advmss); + WRITE_ONCE(tp->window_clamp, + max(2 * tp->advmss, maxwin - tp->advmss)); tp->rcv_ssthresh = min(tp->rcv_ssthresh, tp->window_clamp); tp->snd_cwnd_stamp = tcp_jiffies32; @@ -768,7 +774,8 @@ void tcp_rcv_space_adjust(struct sock *sk) WRITE_ONCE(sk->sk_rcvbuf, rcvbuf); /* Make the window clamp follow along. */ - tp->window_clamp = tcp_win_from_space(sk, rcvbuf); + WRITE_ONCE(tp->window_clamp, + tcp_win_from_space(sk, rcvbuf)); } } tp->rcvq_space.space = copied; @@ -2106,8 +2113,16 @@ void tcp_clear_retrans(struct tcp_sock *tp) static inline void tcp_init_undo(struct tcp_sock *tp) { tp->undo_marker = tp->snd_una; + /* Retransmission still in flight may cause DSACKs later. */ - tp->undo_retrans = tp->retrans_out ? : -1; + /* First, account for regular retransmits in flight: */ + tp->undo_retrans = tp->retrans_out; + /* Next, account for TLP retransmits in flight: */ + if (tp->tlp_high_seq && tp->tlp_retrans) + tp->undo_retrans++; + /* Finally, avoid 0, because undo_retrans==0 means "can undo now": */ + if (!tp->undo_retrans) + tp->undo_retrans = -1; } static bool tcp_is_rack(const struct sock *sk) @@ -2186,6 +2201,7 @@ void tcp_enter_loss(struct sock *sk) tcp_set_ca_state(sk, TCP_CA_Loss); tp->high_seq = tp->snd_nxt; + tp->tlp_high_seq = 0; tcp_ecn_queue_cwr(tp); /* F-RTO RFC5682 sec 3.1 step 1: retransmit SND.UNA if no previous @@ -4358,9 +4374,26 @@ static enum skb_drop_reason tcp_sequence(const struct tcp_sock *tp, return SKB_NOT_DROPPED_YET; } + +void tcp_done_with_error(struct sock *sk, int err) +{ + /* This barrier is coupled with smp_rmb() in tcp_poll() */ + WRITE_ONCE(sk->sk_err, err); + smp_wmb(); + + tcp_write_queue_purge(sk); + tcp_done(sk); + + if (!sock_flag(sk, SOCK_DEAD)) + sk_error_report(sk); +} +EXPORT_SYMBOL(tcp_done_with_error); + /* When we get a reset we do this. */ void tcp_reset(struct sock *sk, struct sk_buff *skb) { + int err; + trace_tcp_receive_reset(sk); /* mptcp can't tell us to ignore reset pkts, @@ -4372,24 +4405,17 @@ void tcp_reset(struct sock *sk, struct sk_buff *skb) /* We want the right error as BSD sees it (and indeed as we do). */ switch (sk->sk_state) { case TCP_SYN_SENT: - WRITE_ONCE(sk->sk_err, ECONNREFUSED); + err = ECONNREFUSED; break; case TCP_CLOSE_WAIT: - WRITE_ONCE(sk->sk_err, EPIPE); + err = EPIPE; break; case TCP_CLOSE: return; default: - WRITE_ONCE(sk->sk_err, ECONNRESET); + err = ECONNRESET; } - /* This barrier is coupled with smp_rmb() in tcp_poll() */ - smp_wmb(); - - tcp_write_queue_purge(sk); - tcp_done(sk); - - if (!sock_flag(sk, SOCK_DEAD)) - sk_error_report(sk); + tcp_done_with_error(sk, err); } /* @@ -5854,6 +5880,11 @@ static bool tcp_validate_incoming(struct sock *sk, struct sk_buff *skb, * RFC 5961 4.2 : Send a challenge ack */ if (th->syn) { + if (sk->sk_state == TCP_SYN_RECV && sk->sk_socket && th->ack && + TCP_SKB_CB(skb)->seq + 1 == TCP_SKB_CB(skb)->end_seq && + TCP_SKB_CB(skb)->seq + 1 == tp->rcv_nxt && + TCP_SKB_CB(skb)->ack_seq == tp->snd_nxt) + goto pass; syn_challenge: if (syn_inerr) TCP_INC_STATS(sock_net(sk), TCP_MIB_INERRS); @@ -5863,6 +5894,7 @@ syn_challenge: goto discard; } +pass: bpf_skops_parse_hdr(sk, skb); return true; @@ -6328,7 +6360,8 @@ consume: if (!tp->rx_opt.wscale_ok) { tp->rx_opt.snd_wscale = tp->rx_opt.rcv_wscale = 0; - tp->window_clamp = min(tp->window_clamp, 65535U); + WRITE_ONCE(tp->window_clamp, + min(tp->window_clamp, 65535U)); } if (tp->rx_opt.saw_tstamp) { diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 7c2ca4df0daa..96d235bcf5cb 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -94,6 +94,8 @@ EXPORT_SYMBOL(tcp_hashinfo); static DEFINE_PER_CPU(struct sock *, ipv4_tcp_sk); +static DEFINE_MUTEX(tcp_exit_batch_mutex); + static u32 tcp_v4_init_seq(const struct sk_buff *skb) { return secure_tcp_seq(ip_hdr(skb)->daddr, @@ -602,15 +604,10 @@ int tcp_v4_err(struct sk_buff *skb, u32 info) ip_icmp_error(sk, skb, err, th->dest, info, (u8 *)th); - if (!sock_owned_by_user(sk)) { - WRITE_ONCE(sk->sk_err, err); - - sk_error_report(sk); - - tcp_done(sk); - } else { + if (!sock_owned_by_user(sk)) + tcp_done_with_error(sk, err); + else WRITE_ONCE(sk->sk_err_soft, err); - } goto out; } @@ -3309,13 +3306,25 @@ static void __net_exit tcp_sk_exit_batch(struct list_head *net_exit_list) { struct net *net; - tcp_twsk_purge(net_exit_list, AF_INET); + /* make sure concurrent calls to tcp_sk_exit_batch from net_cleanup_work + * and failed setup_net error unwinding path are serialized. + * + * tcp_twsk_purge() handles twsk in any dead netns, not just those in + * net_exit_list, the thread that dismantles a particular twsk must + * do so without other thread progressing to refcount_dec_and_test() of + * tcp_death_row.tw_refcount. + */ + mutex_lock(&tcp_exit_batch_mutex); + + tcp_twsk_purge(net_exit_list); list_for_each_entry(net, net_exit_list, exit_list) { inet_pernet_hashinfo_free(net->ipv4.tcp_death_row.hashinfo); WARN_ON_ONCE(!refcount_dec_and_test(&net->ipv4.tcp_death_row.tw_refcount)); tcp_fastopen_ctx_destroy(net); } + + mutex_unlock(&tcp_exit_batch_mutex); } static struct pernet_operations __net_initdata tcp_sk_ops = { diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c index b71f94a5932a..e0883ba709b0 100644 --- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -899,11 +899,13 @@ static void tcp_metrics_flush_all(struct net *net) unsigned int row; for (row = 0; row < max_rows; row++, hb++) { - struct tcp_metrics_block __rcu **pp; + struct tcp_metrics_block __rcu **pp = &hb->chain; bool match; + if (!rcu_access_pointer(*pp)) + continue; + spin_lock_bh(&tcp_metrics_lock); - pp = &hb->chain; for (tm = deref_locked(*pp); tm; tm = deref_locked(*pp)) { match = net ? net_eq(tm_net(tm), net) : !refcount_read(&tm_net(tm)->ns.count); @@ -915,6 +917,7 @@ static void tcp_metrics_flush_all(struct net *net) } } spin_unlock_bh(&tcp_metrics_lock); + cond_resched(); } } diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 260bfb9ada38..0c04c460001f 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -363,7 +363,7 @@ void tcp_twsk_destructor(struct sock *sk) } EXPORT_SYMBOL_GPL(tcp_twsk_destructor); -void tcp_twsk_purge(struct list_head *net_exit_list, int family) +void tcp_twsk_purge(struct list_head *net_exit_list) { bool purged_once = false; struct net *net; @@ -371,14 +371,13 @@ void tcp_twsk_purge(struct list_head *net_exit_list, int family) list_for_each_entry(net, net_exit_list, exit_list) { if (net->ipv4.tcp_death_row.hashinfo->pernet) { /* Even if tw_refcount == 1, we must clean up kernel reqsk */ - inet_twsk_purge(net->ipv4.tcp_death_row.hashinfo, family); + inet_twsk_purge(net->ipv4.tcp_death_row.hashinfo); } else if (!purged_once) { - inet_twsk_purge(&tcp_hashinfo, family); + inet_twsk_purge(&tcp_hashinfo); purged_once = true; } } } -EXPORT_SYMBOL_GPL(tcp_twsk_purge); /* Warning : This function is called without sk_listener being locked. * Be sure to read socket fields once, as their value could change under us. diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index 8311c38267b5..69e6012ae82f 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -73,6 +73,9 @@ struct sk_buff *tcp_gso_segment(struct sk_buff *skb, if (thlen < sizeof(*th)) goto out; + if (unlikely(skb_checksum_start(skb) != skb_transport_header(skb))) + goto out; + if (!pskb_may_pull(skb, thlen)) goto out; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 5631041ae12c..15c49d559db5 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -203,16 +203,17 @@ static inline void tcp_event_ack_sent(struct sock *sk, u32 rcv_nxt) * This MUST be enforced by all callers. */ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, - __u32 *rcv_wnd, __u32 *window_clamp, + __u32 *rcv_wnd, __u32 *__window_clamp, int wscale_ok, __u8 *rcv_wscale, __u32 init_rcv_wnd) { unsigned int space = (__space < 0 ? 0 : __space); + u32 window_clamp = READ_ONCE(*__window_clamp); /* If no clamp set the clamp to the max possible scaled window */ - if (*window_clamp == 0) - (*window_clamp) = (U16_MAX << TCP_MAX_WSCALE); - space = min(*window_clamp, space); + if (window_clamp == 0) + window_clamp = (U16_MAX << TCP_MAX_WSCALE); + space = min(window_clamp, space); /* Quantize space offering to a multiple of mss if possible. */ if (space > mss) @@ -239,12 +240,13 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, /* Set window scaling on max possible window */ space = max_t(u32, space, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])); space = max_t(u32, space, READ_ONCE(sysctl_rmem_max)); - space = min_t(u32, space, *window_clamp); + space = min_t(u32, space, window_clamp); *rcv_wscale = clamp_t(int, ilog2(space) - 15, 0, TCP_MAX_WSCALE); } /* Set the clamp no higher than max representable value */ - (*window_clamp) = min_t(__u32, U16_MAX << (*rcv_wscale), *window_clamp); + WRITE_ONCE(*__window_clamp, + min_t(__u32, U16_MAX << (*rcv_wscale), window_clamp)); } EXPORT_SYMBOL(tcp_select_initial_window); @@ -3787,7 +3789,7 @@ static void tcp_connect_init(struct sock *sk) tcp_ca_dst_init(sk, dst); if (!tp->window_clamp) - tp->window_clamp = dst_metric(dst, RTAX_WINDOW); + WRITE_ONCE(tp->window_clamp, dst_metric(dst, RTAX_WINDOW)); tp->advmss = tcp_mss_clamp(tp, dst_metric_advmss(dst)); tcp_initialize_rcv_mss(sk); @@ -3795,7 +3797,7 @@ static void tcp_connect_init(struct sock *sk) /* limit the window selection if the user enforce a smaller rx buffer */ if (sk->sk_userlocks & SOCK_RCVBUF_LOCK && (tp->window_clamp > tcp_full_space(sk) || tp->window_clamp == 0)) - tp->window_clamp = tcp_full_space(sk); + WRITE_ONCE(tp->window_clamp, tcp_full_space(sk)); rcv_wnd = tcp_rwnd_init_bpf(sk); if (rcv_wnd == 0) diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 984ab4a0421e..64bcf384e9dd 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -69,11 +69,7 @@ u32 tcp_clamp_probe0_to_user_timeout(const struct sock *sk, u32 when) static void tcp_write_err(struct sock *sk) { - WRITE_ONCE(sk->sk_err, READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT); - sk_error_report(sk); - - tcp_write_queue_purge(sk); - tcp_done(sk); + tcp_done_with_error(sk, READ_ONCE(sk->sk_err_soft) ? : ETIMEDOUT); __NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPABORTONTIMEOUT); } @@ -457,17 +453,34 @@ static void tcp_fastopen_synack_timer(struct sock *sk, struct request_sock *req) static bool tcp_rtx_probe0_timed_out(const struct sock *sk, const struct sk_buff *skb) { + const struct inet_connection_sock *icsk = inet_csk(sk); + u32 user_timeout = READ_ONCE(icsk->icsk_user_timeout); const struct tcp_sock *tp = tcp_sk(sk); - const int timeout = TCP_RTO_MAX * 2; - u32 rcv_delta, rtx_delta; - - rcv_delta = inet_csk(sk)->icsk_timeout - tp->rcv_tstamp; - if (rcv_delta <= timeout) - return false; + int timeout = TCP_RTO_MAX * 2; + u32 rtx_delta; + s32 rcv_delta; rtx_delta = (u32)msecs_to_jiffies(tcp_time_stamp(tp) - (tp->retrans_stamp ?: tcp_skb_timestamp(skb))); + if (user_timeout) { + /* If user application specified a TCP_USER_TIMEOUT, + * it does not want win 0 packets to 'reset the timer' + * while retransmits are not making progress. + */ + if (rtx_delta > user_timeout) + return true; + timeout = min_t(u32, timeout, msecs_to_jiffies(user_timeout)); + } + + /* Note: timer interrupt might have been delayed by at least one jiffy, + * and tp->rcv_tstamp might very well have been written recently. + * rcv_delta can thus be negative. + */ + rcv_delta = icsk->icsk_timeout - tp->rcv_tstamp; + if (rcv_delta <= timeout) + return false; + return rtx_delta > timeout; } @@ -509,8 +522,6 @@ void tcp_retransmit_timer(struct sock *sk) if (WARN_ON_ONCE(!skb)) return; - tp->tlp_high_seq = 0; - if (!tp->snd_wnd && !sock_flag(sk, SOCK_DEAD) && !((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))) { /* Receiver dastardly shrinks window. Our retransmits diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 16ca211c8619..73fb814460b6 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -326,6 +326,8 @@ found: goto fail_unlock; } + sock_set_flag(sk, SOCK_RCU_FREE); + sk_add_node_rcu(sk, &hslot->head); hslot->count++; sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); @@ -342,7 +344,7 @@ found: hslot2->count++; spin_unlock(&hslot2->lock); } - sock_set_flag(sk, SOCK_RCU_FREE); + error = 0; fail_unlock: spin_unlock_bh(&hslot->lock); diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c index e5971890d637..f544016e6eb3 100644 --- a/net/ipv4/udp_offload.c +++ b/net/ipv4/udp_offload.c @@ -278,6 +278,11 @@ struct sk_buff *__udp_gso_segment(struct sk_buff *gso_skb, if (gso_skb->len <= sizeof(*uh) + mss) return ERR_PTR(-EINVAL); + if (unlikely(skb_checksum_start(gso_skb) != + skb_transport_header(gso_skb) && + !(skb_shinfo(gso_skb)->gso_type & SKB_GSO_FRAGLIST))) + return ERR_PTR(-EINVAL); + if (skb_gso_ok(gso_skb, features | NETIF_F_GSO_ROBUST)) { /* Packet is from an untrusted source, reset gso_segs. */ skb_shinfo(gso_skb)->gso_segs = DIV_ROUND_UP(gso_skb->len - sizeof(*uh), |