diff options
Diffstat (limited to 'net/ipv4')
43 files changed, 897 insertions, 352 deletions
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index bd2901604842..d83071dccd74 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -331,8 +331,8 @@ config NET_FOU_IP_TUNNELS When this option is enabled IP tunnels can be configured to use FOU or GUE encapsulation. -config GENEVE - tristate "Generic Network Virtualization Encapsulation (Geneve)" +config GENEVE_CORE + tristate "Generic Network Virtualization Encapsulation library" depends on INET select NET_UDP_TUNNEL ---help--- diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 518c04ed666e..b36236dd6014 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -56,7 +56,7 @@ obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o obj-$(CONFIG_MEMCG_KMEM) += tcp_memcontrol.o obj-$(CONFIG_NETLABEL) += cipso_ipv4.o -obj-$(CONFIG_GENEVE) += geneve.o +obj-$(CONFIG_GENEVE_CORE) += geneve_core.o obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \ xfrm4_output.o xfrm4_protocol.o diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 8b47a4d79d04..cc858ef44451 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -317,7 +317,7 @@ lookup_protocol: WARN_ON(!answer_prot->slab); err = -ENOBUFS; - sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot); + sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot, kern); if (!sk) goto out; @@ -488,7 +488,8 @@ int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) inet->inet_saddr = 0; /* Use device */ /* Make sure we are allowed to bind here. */ - if (sk->sk_prot->get_port(sk, snum)) { + if ((snum || !inet->bind_address_no_port) && + sk->sk_prot->get_port(sk, snum)) { inet->inet_saddr = inet->inet_rcv_saddr = 0; err = -EADDRINUSE; goto out_release_sock; @@ -1430,7 +1431,7 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, struct net *net) { struct socket *sock; - int rc = sock_create_kern(family, type, protocol, &sock); + int rc = sock_create_kern(net, family, type, protocol, &sock); if (rc == 0) { *sk = sock->sk; @@ -1440,8 +1441,6 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, * we do not wish this socket to see incoming packets. */ (*sk)->sk_prot->unhash(*sk); - - sk_change_net(*sk, net); } return rc; } @@ -1597,7 +1596,7 @@ static __net_init int inet_init_net(struct net *net) */ seqlock_init(&net->ipv4.ip_local_ports.lock); net->ipv4.ip_local_ports.range[0] = 32768; - net->ipv4.ip_local_ports.range[1] = 61000; + net->ipv4.ip_local_ports.range[1] = 60999; seqlock_init(&net->ipv4.ping_group_range.lock); /* diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 421a80b09b62..30b544f025ac 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -256,7 +256,8 @@ static int esp_output(struct xfrm_state *x, struct sk_buff *skb) aead_givcrypt_set_crypt(req, sg, sg, clen, iv); aead_givcrypt_set_assoc(req, asg, assoclen); aead_givcrypt_set_giv(req, esph->enc_data, - XFRM_SKB_CB(skb)->seq.output.low); + XFRM_SKB_CB(skb)->seq.output.low + + ((u64)XFRM_SKB_CB(skb)->seq.output.hi << 32)); ESP_SKB_CB(skb)->tmp = tmp; err = crypto_aead_givencrypt(req); diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 8d695b6659c7..28ec3c1823bf 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -713,8 +713,6 @@ static void fib_info_hash_move(struct hlist_head *new_info_hash, struct hlist_head *dest; unsigned int new_hash; - hlist_del(&fi->fib_hash); - new_hash = fib_info_hashfn(fi); dest = &new_info_hash[new_hash]; hlist_add_head(&fi->fib_hash, dest); @@ -731,8 +729,6 @@ static void fib_info_hash_move(struct hlist_head *new_info_hash, struct hlist_head *ldest; unsigned int new_hash; - hlist_del(&fi->fib_lhash); - new_hash = fib_laddr_hashfn(fi->fib_prefsrc); ldest = &new_laddrhash[new_hash]; hlist_add_head(&fi->fib_lhash, ldest); diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index e13fcc602da2..3c699c4e90a4 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -72,6 +72,7 @@ #include <linux/list.h> #include <linux/slab.h> #include <linux/export.h> +#include <linux/vmalloc.h> #include <net/net_namespace.h> #include <net/ip.h> #include <net/protocol.h> @@ -324,13 +325,15 @@ static inline void empty_child_dec(struct key_vector *n) static struct key_vector *leaf_new(t_key key, struct fib_alias *fa) { - struct tnode *kv = kmem_cache_alloc(trie_leaf_kmem, GFP_KERNEL); - struct key_vector *l = kv->kv; + struct key_vector *l; + struct tnode *kv; + kv = kmem_cache_alloc(trie_leaf_kmem, GFP_KERNEL); if (!kv) return NULL; /* initialize key vector */ + l = kv->kv; l->key = key; l->pos = 0; l->bits = 0; @@ -345,24 +348,26 @@ static struct key_vector *leaf_new(t_key key, struct fib_alias *fa) static struct key_vector *tnode_new(t_key key, int pos, int bits) { - struct tnode *tnode = tnode_alloc(bits); unsigned int shift = pos + bits; - struct key_vector *tn = tnode->kv; + struct key_vector *tn; + struct tnode *tnode; /* verify bits and pos their msb bits clear and values are valid */ BUG_ON(!bits || (shift > KEYLENGTH)); - pr_debug("AT %p s=%zu %zu\n", tnode, TNODE_SIZE(0), - sizeof(struct key_vector *) << bits); - + tnode = tnode_alloc(bits); if (!tnode) return NULL; + pr_debug("AT %p s=%zu %zu\n", tnode, TNODE_SIZE(0), + sizeof(struct key_vector *) << bits); + if (bits == KEYLENGTH) tnode->full_children = 1; else tnode->empty_children = 1ul << bits; + tn = tnode->kv; tn->key = (shift < KEYLENGTH) ? (key >> shift) << shift : 0; tn->pos = pos; tn->bits = bits; @@ -1164,14 +1169,15 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) state = fa->fa_state; new_fa->fa_state = state & ~FA_S_ACCESSED; new_fa->fa_slen = fa->fa_slen; + new_fa->tb_id = tb->tb_id; - err = netdev_switch_fib_ipv4_add(key, plen, fi, - new_fa->fa_tos, - cfg->fc_type, - cfg->fc_nlflags, - tb->tb_id); + err = switchdev_fib_ipv4_add(key, plen, fi, + new_fa->fa_tos, + cfg->fc_type, + cfg->fc_nlflags, + tb->tb_id); if (err) { - netdev_switch_fib_ipv4_abort(fi); + switchdev_fib_ipv4_abort(fi); kmem_cache_free(fn_alias_kmem, new_fa); goto out; } @@ -1215,12 +1221,10 @@ int fib_table_insert(struct fib_table *tb, struct fib_config *cfg) new_fa->tb_id = tb->tb_id; /* (Optionally) offload fib entry to switch hardware. */ - err = netdev_switch_fib_ipv4_add(key, plen, fi, tos, - cfg->fc_type, - cfg->fc_nlflags, - tb->tb_id); + err = switchdev_fib_ipv4_add(key, plen, fi, tos, cfg->fc_type, + cfg->fc_nlflags, tb->tb_id); if (err) { - netdev_switch_fib_ipv4_abort(fi); + switchdev_fib_ipv4_abort(fi); goto out_free_new_fa; } @@ -1239,7 +1243,7 @@ succeeded: return 0; out_sw_fib_del: - netdev_switch_fib_ipv4_del(key, plen, fi, tos, cfg->fc_type, tb->tb_id); + switchdev_fib_ipv4_del(key, plen, fi, tos, cfg->fc_type, tb->tb_id); out_free_new_fa: kmem_cache_free(fn_alias_kmem, new_fa); out: @@ -1517,8 +1521,8 @@ int fib_table_delete(struct fib_table *tb, struct fib_config *cfg) if (!fa_to_delete) return -ESRCH; - netdev_switch_fib_ipv4_del(key, plen, fa_to_delete->fa_info, tos, - cfg->fc_type, tb->tb_id); + switchdev_fib_ipv4_del(key, plen, fa_to_delete->fa_info, tos, + cfg->fc_type, tb->tb_id); rtmsg_fib(RTM_DELROUTE, htonl(key), fa_to_delete, plen, tb->tb_id, &cfg->fc_nlinfo, 0); @@ -1764,13 +1768,12 @@ void fib_table_flush_external(struct fib_table *tb) /* record local slen */ slen = fa->fa_slen; - if (!fi || !(fi->fib_flags & RTNH_F_EXTERNAL)) + if (!fi || !(fi->fib_flags & RTNH_F_OFFLOAD)) continue; - netdev_switch_fib_ipv4_del(n->key, - KEYLENGTH - fa->fa_slen, - fi, fa->fa_tos, - fa->fa_type, tb->tb_id); + switchdev_fib_ipv4_del(n->key, KEYLENGTH - fa->fa_slen, + fi, fa->fa_tos, fa->fa_type, + tb->tb_id); } /* update leaf slen */ @@ -1835,10 +1838,9 @@ int fib_table_flush(struct fib_table *tb) continue; } - netdev_switch_fib_ipv4_del(n->key, - KEYLENGTH - fa->fa_slen, - fi, fa->fa_tos, - fa->fa_type, tb->tb_id); + switchdev_fib_ipv4_del(n->key, KEYLENGTH - fa->fa_slen, + fi, fa->fa_tos, fa->fa_type, + tb->tb_id); hlist_del_rcu(&fa->fa_list); fib_release_info(fa->fa_info); alias_free_mem_rcu(fa); @@ -2056,11 +2058,12 @@ static struct key_vector *fib_trie_get_next(struct fib_trie_iter *iter) static struct key_vector *fib_trie_get_first(struct fib_trie_iter *iter, struct trie *t) { - struct key_vector *n, *pn = t->kv; + struct key_vector *n, *pn; if (!t) return NULL; + pn = t->kv; n = rcu_dereference(pn->tnode[0]); if (!n) return NULL; diff --git a/net/ipv4/geneve.c b/net/ipv4/geneve_core.c index 8986e63f3bda..311a4ba6950a 100644 --- a/net/ipv4/geneve.c +++ b/net/ipv4/geneve_core.c @@ -60,11 +60,6 @@ struct geneve_net { static int geneve_net_id; -static inline struct genevehdr *geneve_hdr(const struct sk_buff *skb) -{ - return (struct genevehdr *)(udp_hdr(skb) + 1); -} - static struct geneve_sock *geneve_find_sock(struct net *net, sa_family_t family, __be16 port) { @@ -435,7 +430,7 @@ static int __init geneve_init_module(void) if (rc) return rc; - pr_info("Geneve driver\n"); + pr_info("Geneve core logic\n"); return 0; } @@ -449,5 +444,4 @@ module_exit(geneve_cleanup_module); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Jesse Gross <jesse@nicira.com>"); -MODULE_DESCRIPTION("Driver for GENEVE encapsulated traffic"); -MODULE_ALIAS_RTNL_LINK("geneve"); +MODULE_DESCRIPTION("Driver library for GENEVE encapsulated traffic"); diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index a3a697f5ffba..651cdf648ec4 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1339,6 +1339,168 @@ out: } EXPORT_SYMBOL(ip_mc_inc_group); +static int ip_mc_check_iphdr(struct sk_buff *skb) +{ + const struct iphdr *iph; + unsigned int len; + unsigned int offset = skb_network_offset(skb) + sizeof(*iph); + + if (!pskb_may_pull(skb, offset)) + return -EINVAL; + + iph = ip_hdr(skb); + + if (iph->version != 4 || ip_hdrlen(skb) < sizeof(*iph)) + return -EINVAL; + + offset += ip_hdrlen(skb) - sizeof(*iph); + + if (!pskb_may_pull(skb, offset)) + return -EINVAL; + + iph = ip_hdr(skb); + + if (unlikely(ip_fast_csum((u8 *)iph, iph->ihl))) + return -EINVAL; + + len = skb_network_offset(skb) + ntohs(iph->tot_len); + if (skb->len < len || len < offset) + return -EINVAL; + + skb_set_transport_header(skb, offset); + + return 0; +} + +static int ip_mc_check_igmp_reportv3(struct sk_buff *skb) +{ + unsigned int len = skb_transport_offset(skb); + + len += sizeof(struct igmpv3_report); + + return pskb_may_pull(skb, len) ? 0 : -EINVAL; +} + +static int ip_mc_check_igmp_query(struct sk_buff *skb) +{ + unsigned int len = skb_transport_offset(skb); + + len += sizeof(struct igmphdr); + if (skb->len < len) + return -EINVAL; + + /* IGMPv{1,2}? */ + if (skb->len != len) { + /* or IGMPv3? */ + len += sizeof(struct igmpv3_query) - sizeof(struct igmphdr); + if (skb->len < len || !pskb_may_pull(skb, len)) + return -EINVAL; + } + + /* RFC2236+RFC3376 (IGMPv2+IGMPv3) require the multicast link layer + * all-systems destination addresses (224.0.0.1) for general queries + */ + if (!igmp_hdr(skb)->group && + ip_hdr(skb)->daddr != htonl(INADDR_ALLHOSTS_GROUP)) + return -EINVAL; + + return 0; +} + +static int ip_mc_check_igmp_msg(struct sk_buff *skb) +{ + switch (igmp_hdr(skb)->type) { + case IGMP_HOST_LEAVE_MESSAGE: + case IGMP_HOST_MEMBERSHIP_REPORT: + case IGMPV2_HOST_MEMBERSHIP_REPORT: + /* fall through */ + return 0; + case IGMPV3_HOST_MEMBERSHIP_REPORT: + return ip_mc_check_igmp_reportv3(skb); + case IGMP_HOST_MEMBERSHIP_QUERY: + return ip_mc_check_igmp_query(skb); + default: + return -ENOMSG; + } +} + +static inline __sum16 ip_mc_validate_checksum(struct sk_buff *skb) +{ + return skb_checksum_simple_validate(skb); +} + +static int __ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed) + +{ + struct sk_buff *skb_chk; + unsigned int transport_len; + unsigned int len = skb_transport_offset(skb) + sizeof(struct igmphdr); + int ret; + + transport_len = ntohs(ip_hdr(skb)->tot_len) - ip_hdrlen(skb); + + skb_get(skb); + skb_chk = skb_checksum_trimmed(skb, transport_len, + ip_mc_validate_checksum); + if (!skb_chk) + return -EINVAL; + + if (!pskb_may_pull(skb_chk, len)) { + kfree_skb(skb_chk); + return -EINVAL; + } + + ret = ip_mc_check_igmp_msg(skb_chk); + if (ret) { + kfree_skb(skb_chk); + return ret; + } + + if (skb_trimmed) + *skb_trimmed = skb_chk; + else + kfree_skb(skb_chk); + + return 0; +} + +/** + * ip_mc_check_igmp - checks whether this is a sane IGMP packet + * @skb: the skb to validate + * @skb_trimmed: to store an skb pointer trimmed to IPv4 packet tail (optional) + * + * Checks whether an IPv4 packet is a valid IGMP packet. If so sets + * skb network and transport headers accordingly and returns zero. + * + * -EINVAL: A broken packet was detected, i.e. it violates some internet + * standard + * -ENOMSG: IP header validation succeeded but it is not an IGMP packet. + * -ENOMEM: A memory allocation failure happened. + * + * Optionally, an skb pointer might be provided via skb_trimmed (or set it + * to NULL): After parsing an IGMP packet successfully it will point to + * an skb which has its tail aligned to the IP packet end. This might + * either be the originally provided skb or a trimmed, cloned version if + * the skb frame had data beyond the IP packet. A cloned skb allows us + * to leave the original skb and its full frame unchanged (which might be + * desirable for layer 2 frame jugglers). + * + * The caller needs to release a reference count from any returned skb_trimmed. + */ +int ip_mc_check_igmp(struct sk_buff *skb, struct sk_buff **skb_trimmed) +{ + int ret = ip_mc_check_iphdr(skb); + + if (ret < 0) + return ret; + + if (ip_hdr(skb)->protocol != IPPROTO_IGMP) + return -ENOMSG; + + return __ip_mc_check_igmp(skb, skb_trimmed); +} +EXPORT_SYMBOL(ip_mc_check_igmp); + /* * Resend IGMP JOIN report; used by netdev notifier. */ diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 5c3dd6267ed3..60021d0d9326 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -99,6 +99,7 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum) struct net *net = sock_net(sk); int smallest_size = -1, smallest_rover; kuid_t uid = sock_i_uid(sk); + int attempt_half = (sk->sk_reuse == SK_CAN_REUSE) ? 1 : 0; local_bh_disable(); if (!snum) { @@ -106,6 +107,14 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum) again: inet_get_local_port_range(net, &low, &high); + if (attempt_half) { + int half = low + ((high - low) >> 1); + + if (attempt_half == 1) + high = half; + else + low = half; + } remaining = (high - low) + 1; smallest_rover = rover = prandom_u32() % remaining + low; @@ -127,11 +136,6 @@ again: (tb->num_owners < smallest_size || smallest_size == -1)) { smallest_size = tb->num_owners; smallest_rover = rover; - if (atomic_read(&hashinfo->bsockets) > (high - low) + 1 && - !inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, false)) { - snum = smallest_rover; - goto tb_found; - } } if (!inet_csk(sk)->icsk_af_ops->bind_conflict(sk, tb, false)) { snum = rover; @@ -159,6 +163,11 @@ again: snum = smallest_rover; goto have_snum; } + if (attempt_half == 1) { + /* OK we now try the upper half of the range */ + attempt_half = 2; + goto again; + } goto fail; } /* OK, here is the one we will use. HEAD is @@ -564,6 +573,40 @@ int inet_rtx_syn_ack(struct sock *parent, struct request_sock *req) } EXPORT_SYMBOL(inet_rtx_syn_ack); +/* return true if req was found in the syn_table[] */ +static bool reqsk_queue_unlink(struct request_sock_queue *queue, + struct request_sock *req) +{ + struct listen_sock *lopt = queue->listen_opt; + struct request_sock **prev; + bool found = false; + + spin_lock(&queue->syn_wait_lock); + + for (prev = &lopt->syn_table[req->rsk_hash]; *prev != NULL; + prev = &(*prev)->dl_next) { + if (*prev == req) { + *prev = req->dl_next; + found = true; + break; + } + } + + spin_unlock(&queue->syn_wait_lock); + if (del_timer(&req->rsk_timer)) + reqsk_put(req); + return found; +} + +void inet_csk_reqsk_queue_drop(struct sock *sk, struct request_sock *req) +{ + if (reqsk_queue_unlink(&inet_csk(sk)->icsk_accept_queue, req)) { + reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req); + reqsk_put(req); + } +} +EXPORT_SYMBOL(inet_csk_reqsk_queue_drop); + static void reqsk_timer_handler(unsigned long data) { struct request_sock *req = (struct request_sock *)data; diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index bb77ebdae3b3..4d32262c7502 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -224,14 +224,16 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, handler->idiag_get_info(sk, r, info); if (sk->sk_state < TCP_TIME_WAIT) { - int err = 0; + union tcp_cc_info info; + size_t sz = 0; + int attr; rcu_read_lock(); ca_ops = READ_ONCE(icsk->icsk_ca_ops); if (ca_ops && ca_ops->get_info) - err = ca_ops->get_info(sk, ext, skb); + sz = ca_ops->get_info(sk, ext, &attr, &info); rcu_read_unlock(); - if (err < 0) + if (sz && nla_put(skb, attr, sz, &info) < 0) goto errout; } diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index c6fb80bd5826..5f9b063bbe8a 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -18,6 +18,7 @@ #include <linux/sched.h> #include <linux/slab.h> #include <linux/wait.h> +#include <linux/vmalloc.h> #include <net/inet_connection_sock.h> #include <net/inet_hashtables.h> @@ -90,10 +91,6 @@ void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb, const unsigned short snum) { - struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; - - atomic_inc(&hashinfo->bsockets); - inet_sk(sk)->inet_num = snum; sk_add_bind_node(sk, &tb->owners); tb->num_owners++; @@ -111,8 +108,6 @@ static void __inet_put_port(struct sock *sk) struct inet_bind_hashbucket *head = &hashinfo->bhash[bhash]; struct inet_bind_bucket *tb; - atomic_dec(&hashinfo->bsockets); - spin_lock(&head->lock); tb = inet_csk(sk)->icsk_bind_hash; __sk_del_bind_node(sk); @@ -399,9 +394,10 @@ not_unique: return -EADDRNOTAVAIL; } -static inline u32 inet_sk_port_offset(const struct sock *sk) +static u32 inet_sk_port_offset(const struct sock *sk) { const struct inet_sock *inet = inet_sk(sk); + return secure_ipv4_port_ephemeral(inet->inet_rcv_saddr, inet->inet_daddr, inet->inet_dport); @@ -507,8 +503,14 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, inet_get_local_port_range(net, &low, &high); remaining = (high - low) + 1; + /* By starting with offset being an even number, + * we tend to leave about 50% of ports for other uses, + * like bind(0). + */ + offset &= ~1; + local_bh_disable(); - for (i = 1; i <= remaining; i++) { + for (i = 0; i < remaining; i++) { port = low + (i + offset) % remaining; if (inet_is_local_reserved_port(net, port)) continue; @@ -552,7 +554,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, return -EADDRNOTAVAIL; ok: - hint += i; + hint += (i + 2) & ~1; /* Head lock still held and bh's disabled */ inet_bind_hash(sk, tb, port); @@ -599,7 +601,11 @@ out: int inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk) { - return __inet_hash_connect(death_row, sk, inet_sk_port_offset(sk), + u32 port_offset = 0; + + if (!inet_sk(sk)->inet_num) + port_offset = inet_sk_port_offset(sk); + return __inet_hash_connect(death_row, sk, port_offset, __inet_check_established); } EXPORT_SYMBOL_GPL(inet_hash_connect); @@ -608,7 +614,6 @@ void inet_hashinfo_init(struct inet_hashinfo *h) { int i; - atomic_set(&h->bsockets, 0); for (i = 0; i < INET_LHTABLE_SIZE; i++) { spin_lock_init(&h->listening_hash[i].lock); INIT_HLIST_NULLS_HEAD(&h->listening_hash[i].head, @@ -616,3 +621,33 @@ void inet_hashinfo_init(struct inet_hashinfo *h) } } EXPORT_SYMBOL_GPL(inet_hashinfo_init); + +int inet_ehash_locks_alloc(struct inet_hashinfo *hashinfo) +{ + unsigned int i, nblocks = 1; + + if (sizeof(spinlock_t) != 0) { + /* allocate 2 cache lines or at least one spinlock per cpu */ + nblocks = max_t(unsigned int, + 2 * L1_CACHE_BYTES / sizeof(spinlock_t), + 1); + nblocks = roundup_pow_of_two(nblocks * num_possible_cpus()); + + /* no more locks than number of hash buckets */ + nblocks = min(nblocks, hashinfo->ehash_mask + 1); + + hashinfo->ehash_locks = kmalloc_array(nblocks, sizeof(spinlock_t), + GFP_KERNEL | __GFP_NOWARN); + if (!hashinfo->ehash_locks) + hashinfo->ehash_locks = vmalloc(nblocks * sizeof(spinlock_t)); + + if (!hashinfo->ehash_locks) + return -ENOMEM; + + for (i = 0; i < nblocks; i++) + spin_lock_init(&hashinfo->ehash_locks[i]); + } + hashinfo->ehash_locks_mask = nblocks - 1; + return 0; +} +EXPORT_SYMBOL_GPL(inet_ehash_locks_alloc); diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index 00ec8d5d7e7e..2ffbd16b79e0 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -170,7 +170,7 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, } EXPORT_SYMBOL_GPL(__inet_twsk_hashdance); -void tw_timer_handler(unsigned long data) +static void tw_timer_handler(unsigned long data) { struct inet_timewait_sock *tw = (struct inet_timewait_sock *)data; diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 939992c456f3..2d3aa408fbdc 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -39,17 +39,21 @@ #include <net/route.h> #include <net/xfrm.h> -static bool ip_may_fragment(const struct sk_buff *skb) -{ - return unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0) || - skb->ignore_df; -} - static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) { if (skb->len <= mtu) return false; + if (unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0)) + return false; + + /* original fragment exceeds mtu and DF is set */ + if (unlikely(IPCB(skb)->frag_max_size > mtu)) + return true; + + if (skb->ignore_df) + return false; + if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) return false; @@ -82,6 +86,9 @@ int ip_forward(struct sk_buff *skb) if (skb->pkt_type != PACKET_HOST) goto drop; + if (unlikely(skb->sk)) + goto drop; + if (skb_warn_if_lro(skb)) goto drop; @@ -111,7 +118,7 @@ int ip_forward(struct sk_buff *skb) IPCB(skb)->flags |= IPSKB_FORWARDED; mtu = ip_dst_mtu_maybe_forward(&rt->dst, true); - if (!ip_may_fragment(skb) && ip_exceeds_mtu(skb, mtu)) { + if (ip_exceeds_mtu(skb, mtu)) { IP_INC_STATS(dev_net(rt->dst.dev), IPSTATS_MIB_FRAGFAILS); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu)); diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index cc1da6d9cb35..a50dc6d408d1 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -75,6 +75,7 @@ struct ipq { __be16 id; u8 protocol; u8 ecn; /* RFC3168 support */ + u16 max_df_size; /* largest frag with DF set seen */ int iif; unsigned int rid; struct inet_peer *peer; @@ -173,6 +174,15 @@ static void ipq_kill(struct ipq *ipq) inet_frag_kill(&ipq->q, &ip4_frags); } +static bool frag_expire_skip_icmp(u32 user) +{ + return user == IP_DEFRAG_AF_PACKET || + ip_defrag_user_in_between(user, IP_DEFRAG_CONNTRACK_IN, + __IP_DEFRAG_CONNTRACK_IN_END) || + ip_defrag_user_in_between(user, IP_DEFRAG_CONNTRACK_BRIDGE_IN, + __IP_DEFRAG_CONNTRACK_BRIDGE_IN); +} + /* * Oops, a fragment queue timed out. Kill it and send an ICMP reply. */ @@ -217,10 +227,8 @@ static void ip_expire(unsigned long arg) /* Only an end host needs to send an ICMP * "Fragment Reassembly Timeout" message, per RFC792. */ - if (qp->user == IP_DEFRAG_AF_PACKET || - ((qp->user >= IP_DEFRAG_CONNTRACK_IN) && - (qp->user <= __IP_DEFRAG_CONNTRACK_IN_END) && - (skb_rtable(head)->rt_type != RTN_LOCAL))) + if (frag_expire_skip_icmp(qp->user) && + (skb_rtable(head)->rt_type != RTN_LOCAL)) goto out_rcu_unlock; /* Send an ICMP "Fragment Reassembly Timeout" message. */ @@ -319,6 +327,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) { struct sk_buff *prev, *next; struct net_device *dev; + unsigned int fragsize; int flags, offset; int ihl, end; int err = -ENOENT; @@ -474,9 +483,14 @@ found: if (offset == 0) qp->q.flags |= INET_FRAG_FIRST_IN; + fragsize = skb->len + ihl; + + if (fragsize > qp->q.max_size) + qp->q.max_size = fragsize; + if (ip_hdr(skb)->frag_off & htons(IP_DF) && - skb->len + ihl > qp->q.max_size) - qp->q.max_size = skb->len + ihl; + fragsize > qp->max_df_size) + qp->max_df_size = fragsize; if (qp->q.flags == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && qp->q.meat == qp->q.len) { @@ -606,13 +620,27 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev, head->next = NULL; head->dev = dev; head->tstamp = qp->q.stamp; - IPCB(head)->frag_max_size = qp->q.max_size; + IPCB(head)->frag_max_size = max(qp->max_df_size, qp->q.max_size); iph = ip_hdr(head); - /* max_size != 0 implies at least one fragment had IP_DF set */ - iph->frag_off = qp->q.max_size ? htons(IP_DF) : 0; iph->tot_len = htons(len); iph->tos |= ecn; + + /* When we set IP_DF on a refragmented skb we must also force a + * call to ip_fragment to avoid forwarding a DF-skb of size s while + * original sender only sent fragments of size f (where f < s). + * + * We only set DF/IPSKB_FRAG_PMTU if such DF fragment was the largest + * frag seen to avoid sending tiny DF-fragments in case skb was built + * from one very small df-fragment and one large non-df frag. + */ + if (qp->max_df_size == qp->q.max_size) { + IPCB(head)->flags |= IPSKB_FRAG_PMTU; + iph->frag_off = htons(IP_DF); + } else { + iph->frag_off = 0; + } + IP_INC_STATS_BH(net, IPSTATS_MIB_REASMOKS); qp->q.fragments = NULL; qp->q.fragments_tail = NULL; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index c65b93a7b711..f5f5ef1cebd5 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -83,6 +83,10 @@ int sysctl_ip_default_ttl __read_mostly = IPDEFTTL; EXPORT_SYMBOL(sysctl_ip_default_ttl); +static int ip_fragment(struct sock *sk, struct sk_buff *skb, + unsigned int mtu, + int (*output)(struct sock *, struct sk_buff *)); + /* Generate a checksum for an outgoing IP datagram. */ void ip_send_check(struct iphdr *iph) { @@ -91,7 +95,7 @@ void ip_send_check(struct iphdr *iph) } EXPORT_SYMBOL(ip_send_check); -int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb) +static int __ip_local_out_sk(struct sock *sk, struct sk_buff *skb) { struct iphdr *iph = ip_hdr(skb); @@ -216,7 +220,8 @@ static inline int ip_finish_output2(struct sock *sk, struct sk_buff *skb) return -EINVAL; } -static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb) +static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb, + unsigned int mtu) { netdev_features_t features; struct sk_buff *segs; @@ -224,7 +229,7 @@ static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb) /* common case: locally created skb or seglen is <= mtu */ if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) || - skb_gso_network_seglen(skb) <= ip_skb_dst_mtu(skb)) + skb_gso_network_seglen(skb) <= mtu) return ip_finish_output2(sk, skb); /* Slowpath - GSO segment length is exceeding the dst MTU. @@ -248,7 +253,7 @@ static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb) int err; segs->next = NULL; - err = ip_fragment(sk, segs, ip_finish_output2); + err = ip_fragment(sk, segs, mtu, ip_finish_output2); if (err && ret == 0) ret = err; @@ -260,6 +265,8 @@ static int ip_finish_output_gso(struct sock *sk, struct sk_buff *skb) static int ip_finish_output(struct sock *sk, struct sk_buff *skb) { + unsigned int mtu; + #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM) /* Policy lookup after SNAT yielded a new policy */ if (skb_dst(skb)->xfrm) { @@ -267,11 +274,12 @@ static int ip_finish_output(struct sock *sk, struct sk_buff *skb) return dst_output_sk(sk, skb); } #endif + mtu = ip_skb_dst_mtu(skb); if (skb_is_gso(skb)) - return ip_finish_output_gso(sk, skb); + return ip_finish_output_gso(sk, skb, mtu); - if (skb->len > ip_skb_dst_mtu(skb)) - return ip_fragment(sk, skb, ip_finish_output2); + if (skb->len > mtu || (IPCB(skb)->flags & IPSKB_FRAG_PMTU)) + return ip_fragment(sk, skb, mtu, ip_finish_output2); return ip_finish_output2(sk, skb); } @@ -478,6 +486,31 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) skb_copy_secmark(to, from); } +static int ip_fragment(struct sock *sk, struct sk_buff *skb, + unsigned int mtu, + int (*output)(struct sock *, struct sk_buff *)) +{ + struct iphdr *iph = ip_hdr(skb); + + if ((iph->frag_off & htons(IP_DF)) == 0) + return ip_do_fragment(sk, skb, output); + + if (unlikely(!skb->ignore_df || + (IPCB(skb)->frag_max_size && + IPCB(skb)->frag_max_size > mtu))) { + struct rtable *rt = skb_rtable(skb); + struct net_device *dev = rt->dst.dev; + + IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS); + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, + htonl(mtu)); + kfree_skb(skb); + return -EMSGSIZE; + } + + return ip_do_fragment(sk, skb, output); +} + /* * This IP datagram is too large to be sent in one piece. Break it up into * smaller pieces (each of size equal to IP header plus @@ -485,8 +518,8 @@ static void ip_copy_metadata(struct sk_buff *to, struct sk_buff *from) * single device frame, and queue such a frame for sending. */ -int ip_fragment(struct sock *sk, struct sk_buff *skb, - int (*output)(struct sock *, struct sk_buff *)) +int ip_do_fragment(struct sock *sk, struct sk_buff *skb, + int (*output)(struct sock *, struct sk_buff *)) { struct iphdr *iph; int ptr; @@ -507,15 +540,8 @@ int ip_fragment(struct sock *sk, struct sk_buff *skb, iph = ip_hdr(skb); mtu = ip_skb_dst_mtu(skb); - if (unlikely(((iph->frag_off & htons(IP_DF)) && !skb->ignore_df) || - (IPCB(skb)->frag_max_size && - IPCB(skb)->frag_max_size > mtu))) { - IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS); - icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, - htonl(mtu)); - kfree_skb(skb); - return -EMSGSIZE; - } + if (IPCB(skb)->frag_max_size && IPCB(skb)->frag_max_size < mtu) + mtu = IPCB(skb)->frag_max_size; /* * Setup starting values. @@ -711,6 +737,9 @@ slow_path: iph = ip_hdr(skb2); iph->frag_off = htons((offset >> 3)); + if (IPCB(skb)->flags & IPSKB_FRAG_PMTU) + iph->frag_off |= htons(IP_DF); + /* ANK: dirty, but effective trick. Upgrade options only if * the segment to be fragmented was THE FIRST (otherwise, * options are already fixed) and make it ONCE @@ -751,7 +780,7 @@ fail: IP_INC_STATS(dev_net(dev), IPSTATS_MIB_FRAGFAILS); return err; } -EXPORT_SYMBOL(ip_fragment); +EXPORT_SYMBOL(ip_do_fragment); int ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb) @@ -1217,11 +1246,9 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, } while (size > 0) { - int i; - - if (skb_is_gso(skb)) + if (skb_is_gso(skb)) { len = size; - else { + } else { /* Check if the remaining data fits into current packet. */ len = mtu - skb->len; @@ -1273,15 +1300,10 @@ ssize_t ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page, continue; } - i = skb_shinfo(skb)->nr_frags; if (len > size) len = size; - if (skb_can_coalesce(skb, i, page, offset)) { - skb_frag_size_add(&skb_shinfo(skb)->frags[i-1], len); - } else if (i < MAX_SKB_FRAGS) { - get_page(page); - skb_fill_page_desc(skb, i, page, offset, len); - } else { + + if (skb_append_pagefrags(skb, page, offset, len)) { err = -EMSGSIZE; goto error; } diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 7cfb0893f263..04ae2992a5cd 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -582,6 +582,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, case IP_TRANSPARENT: case IP_MINTTL: case IP_NODEFRAG: + case IP_BIND_ADDRESS_NO_PORT: case IP_UNICAST_IF: case IP_MULTICAST_TTL: case IP_MULTICAST_ALL: @@ -732,6 +733,9 @@ static int do_ip_setsockopt(struct sock *sk, int level, } inet->nodefrag = val ? 1 : 0; break; + case IP_BIND_ADDRESS_NO_PORT: + inet->bind_address_no_port = val ? 1 : 0; + break; case IP_MTU_DISCOVER: if (val < IP_PMTUDISC_DONT || val > IP_PMTUDISC_OMIT) goto e_inval; @@ -1324,6 +1328,9 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, case IP_NODEFRAG: val = inet->nodefrag; break; + case IP_BIND_ADDRESS_NO_PORT: + val = inet->bind_address_no_port; + break; case IP_MTU_DISCOVER: val = inet->pmtudisc; break; diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index ce63ab21b6cd..6a51a71a6c67 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -98,7 +98,7 @@ int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto) return -ENOMEM; eh = (struct ethhdr *)skb->data; - if (likely(ntohs(eh->h_proto) >= ETH_P_802_3_MIN)) + if (likely(eth_proto_is_802_3(eh->h_proto))) skb->protocol = eh->h_proto; else skb->protocol = htons(ETH_P_802_2); @@ -165,6 +165,8 @@ struct rtnl_link_stats64 *ip_tunnel_get_stats64(struct net_device *dev, { int i; + netdev_stats_to_stats64(tot, &dev->stats); + for_each_possible_cpu(i) { const struct pcpu_sw_netstats *tstats = per_cpu_ptr(dev->tstats, i); @@ -185,22 +187,6 @@ struct rtnl_link_stats64 *ip_tunnel_get_stats64(struct net_device *dev, tot->tx_bytes += tx_bytes; } - tot->multicast = dev->stats.multicast; - - tot->rx_crc_errors = dev->stats.rx_crc_errors; - tot->rx_fifo_errors = dev->stats.rx_fifo_errors; - tot->rx_length_errors = dev->stats.rx_length_errors; - tot->rx_frame_errors = dev->stats.rx_frame_errors; - tot->rx_errors = dev->stats.rx_errors; - - tot->tx_fifo_errors = dev->stats.tx_fifo_errors; - tot->tx_carrier_errors = dev->stats.tx_carrier_errors; - tot->tx_dropped = dev->stats.tx_dropped; - tot->tx_aborted_errors = dev->stats.tx_aborted_errors; - tot->tx_errors = dev->stats.tx_errors; - - tot->collisions = dev->stats.collisions; - return tot; } EXPORT_SYMBOL_GPL(ip_tunnel_get_stats64); diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c index 9f7269f3c54a..0c152087ca15 100644 --- a/net/ipv4/ip_vti.c +++ b/net/ipv4/ip_vti.c @@ -65,7 +65,6 @@ static int vti_input(struct sk_buff *skb, int nexthdr, __be32 spi, goto drop; XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = tunnel; - skb->mark = be32_to_cpu(tunnel->parms.i_key); return xfrm_input(skb, nexthdr, spi, encap_type); } @@ -91,6 +90,8 @@ static int vti_rcv_cb(struct sk_buff *skb, int err) struct pcpu_sw_netstats *tstats; struct xfrm_state *x; struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4; + u32 orig_mark = skb->mark; + int ret; if (!tunnel) return 1; @@ -107,7 +108,11 @@ static int vti_rcv_cb(struct sk_buff *skb, int err) x = xfrm_input_state(skb); family = x->inner_mode->afinfo->family; - if (!xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family)) + skb->mark = be32_to_cpu(tunnel->parms.i_key); + ret = xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family); + skb->mark = orig_mark; + + if (!ret) return -EPERM; skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev))); @@ -216,8 +221,6 @@ static netdev_tx_t vti_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) memset(&fl, 0, sizeof(fl)); - skb->mark = be32_to_cpu(tunnel->parms.o_key); - switch (skb->protocol) { case htons(ETH_P_IP): xfrm_decode_session(skb, &fl, AF_INET); @@ -233,6 +236,9 @@ static netdev_tx_t vti_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) return NETDEV_TX_OK; } + /* override mark with tunnel output key */ + fl.flowi_mark = be32_to_cpu(tunnel->parms.o_key); + return vti_xmit(skb, dev, &fl); } diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index ff96396ebec5..254238daf58b 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -251,7 +251,8 @@ ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return -EINVAL; } - p.i_key = p.o_key = p.i_flags = p.o_flags = 0; + p.i_key = p.o_key = 0; + p.i_flags = p.o_flags = 0; if (p.iph.ttl) p.iph.frag_off |= htons(IP_DF); diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 13bfe84bf3ca..a61200754f4b 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1075,6 +1075,9 @@ static int do_replace(struct net *net, const void __user *user, /* overflow check */ if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters)) return -ENOMEM; + if (tmp.num_counters == 0) + return -EINVAL; + tmp.name[sizeof(tmp.name)-1] = 0; newinfo = xt_alloc_table_info(tmp.size); @@ -1499,6 +1502,9 @@ static int compat_do_replace(struct net *net, void __user *user, return -ENOMEM; if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters)) return -ENOMEM; + if (tmp.num_counters == 0) + return -EINVAL; + tmp.name[sizeof(tmp.name)-1] = 0; newinfo = xt_alloc_table_info(tmp.size); diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index c69db7fa25ee..e7abf5145edc 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1262,6 +1262,9 @@ do_replace(struct net *net, const void __user *user, unsigned int len) /* overflow check */ if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters)) return -ENOMEM; + if (tmp.num_counters == 0) + return -EINVAL; + tmp.name[sizeof(tmp.name)-1] = 0; newinfo = xt_alloc_table_info(tmp.size); @@ -1441,7 +1444,6 @@ static int compat_find_calc_match(struct xt_entry_match *m, const char *name, const struct ipt_ip *ip, - unsigned int hookmask, int *size) { struct xt_match *match; @@ -1510,8 +1512,7 @@ check_compat_entry_size_and_hooks(struct compat_ipt_entry *e, entry_offset = (void *)e - (void *)base; j = 0; xt_ematch_foreach(ematch, e) { - ret = compat_find_calc_match(ematch, name, - &e->ip, e->comefrom, &off); + ret = compat_find_calc_match(ematch, name, &e->ip, &off); if (ret != 0) goto release_matches; ++j; @@ -1809,6 +1810,9 @@ compat_do_replace(struct net *net, void __user *user, unsigned int len) return -ENOMEM; if (tmp.num_counters >= INT_MAX / sizeof(struct xt_counters)) return -ENOMEM; + if (tmp.num_counters == 0) + return -EINVAL; + tmp.name[sizeof(tmp.name)-1] = 0; newinfo = xt_alloc_table_info(tmp.size); diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 771ab3d01ad3..45cb16a6a4a3 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -367,6 +367,11 @@ static int clusterip_tg_check(const struct xt_tgchk_param *par) struct clusterip_config *config; int ret; + if (par->nft_compat) { + pr_err("cannot use CLUSTERIP target from nftables compat\n"); + return -EOPNOTSUPP; + } + if (cipinfo->hash_mode != CLUSTERIP_HASHMODE_SIP && cipinfo->hash_mode != CLUSTERIP_HASHMODE_SIP_SPT && cipinfo->hash_mode != CLUSTERIP_HASHMODE_SIP_SPT_DPT) { diff --git a/net/ipv4/netfilter/ipt_SYNPROXY.c b/net/ipv4/netfilter/ipt_SYNPROXY.c index e9e67793055f..fe8cc183411e 100644 --- a/net/ipv4/netfilter/ipt_SYNPROXY.c +++ b/net/ipv4/netfilter/ipt_SYNPROXY.c @@ -18,7 +18,7 @@ #include <net/netfilter/nf_conntrack_synproxy.h> static struct iphdr * -synproxy_build_ip(struct sk_buff *skb, u32 saddr, u32 daddr) +synproxy_build_ip(struct sk_buff *skb, __be32 saddr, __be32 daddr) { struct iphdr *iph; @@ -220,7 +220,7 @@ synproxy_send_client_ack(const struct synproxy_net *snet, nth->ack_seq = th->ack_seq; tcp_flag_word(nth) = TCP_FLAG_ACK; nth->doff = tcp_hdr_size / 4; - nth->window = ntohs(htons(th->window) >> opts->wscale); + nth->window = htons(ntohs(th->window) >> opts->wscale); nth->check = 0; nth->urg_ptr = 0; diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index a93f260cf24c..05ff44b758df 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -158,6 +158,7 @@ void ping_unhash(struct sock *sk) if (sk_hashed(sk)) { write_lock_bh(&ping_table.lock); hlist_nulls_del(&sk->sk_nulls_node); + sk_nulls_node_init(&sk->sk_nulls_node); sock_put(sk); isk->inet_num = 0; isk->inet_sport = 0; diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index e1f3b911dd1e..da5d483e236a 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -298,6 +298,8 @@ static const struct snmp_mib snmp4_net_list[] = { SNMP_MIB_ITEM("TCPACKSkippedFinWait2", LINUX_MIB_TCPACKSKIPPEDFINWAIT2), SNMP_MIB_ITEM("TCPACKSkippedTimeWait", LINUX_MIB_TCPACKSKIPPEDTIMEWAIT), SNMP_MIB_ITEM("TCPACKSkippedChallenge", LINUX_MIB_TCPACKSKIPPEDCHALLENGE), + SNMP_MIB_ITEM("TCPWinProbe", LINUX_MIB_TCPWINPROBE), + SNMP_MIB_ITEM("TCPKeepAlive", LINUX_MIB_TCPKEEPALIVE), SNMP_MIB_SENTINEL }; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index a78540f28276..f6055984c307 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -457,12 +457,9 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst, } #define IP_IDENTS_SZ 2048u -struct ip_ident_bucket { - atomic_t id; - u32 stamp32; -}; -static struct ip_ident_bucket *ip_idents __read_mostly; +static atomic_t *ip_idents __read_mostly; +static u32 *ip_tstamps __read_mostly; /* In order to protect privacy, we add a perturbation to identifiers * if one generator is seldom used. This makes hard for an attacker @@ -470,15 +467,16 @@ static struct ip_ident_bucket *ip_idents __read_mostly; */ u32 ip_idents_reserve(u32 hash, int segs) { - struct ip_ident_bucket *bucket = ip_idents + hash % IP_IDENTS_SZ; - u32 old = ACCESS_ONCE(bucket->stamp32); + u32 *p_tstamp = ip_tstamps + hash % IP_IDENTS_SZ; + atomic_t *p_id = ip_idents + hash % IP_IDENTS_SZ; + u32 old = ACCESS_ONCE(*p_tstamp); u32 now = (u32)jiffies; u32 delta = 0; - if (old != now && cmpxchg(&bucket->stamp32, old, now) == old) + if (old != now && cmpxchg(p_tstamp, old, now) == old) delta = prandom_u32_max(now - old); - return atomic_add_return(segs + delta, &bucket->id) - segs; + return atomic_add_return(segs + delta, p_id) - segs; } EXPORT_SYMBOL(ip_idents_reserve); @@ -902,6 +900,10 @@ static int ip_error(struct sk_buff *skb) bool send; int code; + /* IP on this device is disabled. */ + if (!in_dev) + goto out; + net = dev_net(rt->dst.dev); if (!IN_DEV_FORWARD(in_dev)) { switch (rt->dst.error) { @@ -962,10 +964,7 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu) if (dst_metric_locked(dst, RTAX_MTU)) return; - if (dst->dev->mtu < mtu) - return; - - if (rt->rt_pmtu && rt->rt_pmtu < mtu) + if (ipv4_mtu(dst) < mtu) return; if (mtu < ip_rt_min_pmtu) @@ -2096,7 +2095,8 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4) goto out; } if (ipv4_is_local_multicast(fl4->daddr) || - ipv4_is_lbcast(fl4->daddr)) { + ipv4_is_lbcast(fl4->daddr) || + fl4->flowi4_proto == IPPROTO_IGMP) { if (!fl4->saddr) fl4->saddr = inet_select_addr(dev_out, 0, RT_SCOPE_LINK); @@ -2741,6 +2741,10 @@ int __init ip_rt_init(void) prandom_bytes(ip_idents, IP_IDENTS_SZ * sizeof(*ip_idents)); + ip_tstamps = kcalloc(IP_IDENTS_SZ, sizeof(*ip_tstamps), GFP_KERNEL); + if (!ip_tstamps) + panic("IP: failed to allocate ip_tstamps\n"); + for_each_possible_cpu(cpu) { struct uncached_list *ul = &per_cpu(rt_uncached_list, cpu); diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index df849e5a10f1..d70b1f603692 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -219,9 +219,9 @@ int __cookie_v4_check(const struct iphdr *iph, const struct tcphdr *th, } EXPORT_SYMBOL_GPL(__cookie_v4_check); -static struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, - struct request_sock *req, - struct dst_entry *dst) +struct sock *tcp_get_cookie_sock(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct dst_entry *dst) { struct inet_connection_sock *icsk = inet_csk(sk); struct sock *child; @@ -235,7 +235,7 @@ static struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, } return child; } - +EXPORT_SYMBOL(tcp_get_cookie_sock); /* * when syncookies are in effect and tcp timestamps are enabled we stored @@ -391,7 +391,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb) ireq->rcv_wscale = rcv_wscale; ireq->ecn_ok = cookie_ecn_ok(&tcp_opt, sock_net(sk), &rt->dst); - ret = get_cookie_sock(sk, skb, req, &rt->dst); + ret = tcp_get_cookie_sock(sk, skb, req, &rt->dst); /* ip_queue_xmit() depends on our flow being setup * Normal sockets get it right from inet_csk_route_child_sock() */ diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index c3852a7ff3c7..433231ccfb17 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -41,11 +41,19 @@ static int tcp_syn_retries_min = 1; static int tcp_syn_retries_max = MAX_TCP_SYNCNT; static int ip_ping_group_range_min[] = { 0, 0 }; static int ip_ping_group_range_max[] = { GID_T_MAX, GID_T_MAX }; +static int min_sndbuf = SOCK_MIN_SNDBUF; +static int min_rcvbuf = SOCK_MIN_RCVBUF; /* Update system visible IP port range */ static void set_local_port_range(struct net *net, int range[2]) { + bool same_parity = !((range[0] ^ range[1]) & 1); + write_seqlock(&net->ipv4.ip_local_ports.lock); + if (same_parity && !net->ipv4.ip_local_ports.warned) { + net->ipv4.ip_local_ports.warned = true; + pr_err_ratelimited("ip_local_port_range: prefer different parity for start/end values.\n"); + } net->ipv4.ip_local_ports.range[0] = range[0]; net->ipv4.ip_local_ports.range[1] = range[1]; write_sequnlock(&net->ipv4.ip_local_ports.lock); @@ -522,7 +530,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_tcp_wmem), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &one, + .extra1 = &min_sndbuf, }, { .procname = "tcp_notsent_lowat", @@ -537,7 +545,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_tcp_rmem), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &one, + .extra1 = &min_rcvbuf, }, { .procname = "tcp_app_win", @@ -702,7 +710,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &zero, + .extra1 = &one, .extra2 = &gso_max_segs, }, { @@ -750,7 +758,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_udp_rmem_min), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &one + .extra1 = &min_rcvbuf, }, { .procname = "udp_wmem_min", @@ -758,7 +766,7 @@ static struct ctl_table ipv4_table[] = { .maxlen = sizeof(sysctl_udp_wmem_min), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &one + .extra1 = &min_sndbuf, }, { } }; @@ -821,6 +829,13 @@ static struct ctl_table ipv4_net_table[] = { .proc_handler = proc_dointvec }, { + .procname = "tcp_ecn_fallback", + .data = &init_net.ipv4.sysctl_tcp_ecn_fallback, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, + { .procname = "ip_local_port_range", .maxlen = sizeof(init_net.ipv4.ip_local_ports.range), .data = &init_net.ipv4.ip_local_ports.range, diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 59c8a027721b..65f791f74845 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -252,6 +252,7 @@ #include <linux/types.h> #include <linux/fcntl.h> #include <linux/poll.h> +#include <linux/inet_diag.h> #include <linux/init.h> #include <linux/fs.h> #include <linux/skbuff.h> @@ -401,6 +402,7 @@ void tcp_init_sock(struct sock *sk) tp->snd_ssthresh = TCP_INFINITE_SSTHRESH; tp->snd_cwnd_clamp = ~0; tp->mss_cache = TCP_MSS_DEFAULT; + u64_stats_init(&tp->syncp); tp->reordering = sysctl_tcp_reordering; tcp_enable_early_retrans(tp); @@ -520,8 +522,10 @@ unsigned int tcp_poll(struct file *file, struct socket *sock, poll_table *wait) /* Race breaker. If space is freed after * wspace test but before the flags are set, - * IO signal will be lost. + * IO signal will be lost. Memory barrier + * pairs with the input side. */ + smp_mb__after_atomic(); if (sk_stream_is_writeable(sk)) mask |= POLLOUT | POLLWRNORM; } @@ -691,8 +695,9 @@ static int tcp_splice_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, struct tcp_splice_state *tss = rd_desc->arg.data; int ret; - ret = skb_splice_bits(skb, offset, tss->pipe, min(rd_desc->count, len), - tss->flags); + ret = skb_splice_bits(skb, skb->sk, offset, tss->pipe, + min(rd_desc->count, len), tss->flags, + skb_socket_splice); if (ret > 0) rd_desc->count -= ret; return ret; @@ -805,16 +810,28 @@ ssize_t tcp_splice_read(struct socket *sock, loff_t *ppos, } EXPORT_SYMBOL(tcp_splice_read); -struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp) +struct sk_buff *sk_stream_alloc_skb(struct sock *sk, int size, gfp_t gfp, + bool force_schedule) { struct sk_buff *skb; /* The TCP header must be at least 32-bit aligned. */ size = ALIGN(size, 4); + if (unlikely(tcp_under_memory_pressure(sk))) + sk_mem_reclaim_partial(sk); + skb = alloc_skb_fclone(size + sk->sk_prot->max_header, gfp); - if (skb) { - if (sk_wmem_schedule(sk, skb->truesize)) { + if (likely(skb)) { + bool mem_scheduled; + + if (force_schedule) { + mem_scheduled = true; + sk_forced_mem_schedule(sk, skb->truesize); + } else { + mem_scheduled = sk_wmem_schedule(sk, skb->truesize); + } + if (likely(mem_scheduled)) { skb_reserve(skb, sk->sk_prot->max_header); /* * Make sure that we have exactly size bytes @@ -904,7 +921,8 @@ new_segment: if (!sk_stream_memory_free(sk)) goto wait_for_sndbuf; - skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation); + skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation, + skb_queue_empty(&sk->sk_write_queue)); if (!skb) goto wait_for_memory; @@ -983,6 +1001,9 @@ do_error: if (copied) goto out; out_err: + /* make sure we wake any epoll edge trigger waiter */ + if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && err == -EAGAIN)) + sk->sk_write_space(sk); return sk_stream_error(sk, flags, err); } @@ -1140,7 +1161,8 @@ new_segment: skb = sk_stream_alloc_skb(sk, select_size(sk, sg), - sk->sk_allocation); + sk->sk_allocation, + skb_queue_empty(&sk->sk_write_queue)); if (!skb) goto wait_for_memory; @@ -1271,6 +1293,9 @@ do_error: goto out; out_err: err = sk_stream_error(sk, flags, err); + /* make sure we wake any epoll edge trigger waiter */ + if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 && err == -EAGAIN)) + sk->sk_write_space(sk); release_sock(sk); return err; } @@ -2479,6 +2504,13 @@ static int do_tcp_setsockopt(struct sock *sk, int level, icsk->icsk_syn_retries = val; break; + case TCP_SAVE_SYN: + if (val < 0 || val > 1) + err = -EINVAL; + else + tp->save_syn = val; + break; + case TCP_LINGER2: if (val < 0) tp->linger2 = -1; @@ -2590,11 +2622,12 @@ EXPORT_SYMBOL(compat_tcp_setsockopt); #endif /* Return information about state of tcp endpoint in API format. */ -void tcp_get_info(const struct sock *sk, struct tcp_info *info) +void tcp_get_info(struct sock *sk, struct tcp_info *info) { const struct tcp_sock *tp = tcp_sk(sk); const struct inet_connection_sock *icsk = inet_csk(sk); u32 now = tcp_time_stamp; + unsigned int start; u32 rate; memset(info, 0, sizeof(*info)); @@ -2661,6 +2694,14 @@ void tcp_get_info(const struct sock *sk, struct tcp_info *info) rate = READ_ONCE(sk->sk_max_pacing_rate); info->tcpi_max_pacing_rate = rate != ~0U ? rate : ~0ULL; + + do { + start = u64_stats_fetch_begin_irq(&tp->syncp); + info->tcpi_bytes_acked = tp->bytes_acked; + info->tcpi_bytes_received = tp->bytes_received; + } while (u64_stats_fetch_retry_irq(&tp->syncp, start)); + info->tcpi_segs_out = tp->segs_out; + info->tcpi_segs_in = tp->segs_in; } EXPORT_SYMBOL_GPL(tcp_get_info); @@ -2732,6 +2773,26 @@ static int do_tcp_getsockopt(struct sock *sk, int level, return -EFAULT; return 0; } + case TCP_CC_INFO: { + const struct tcp_congestion_ops *ca_ops; + union tcp_cc_info info; + size_t sz = 0; + int attr; + + if (get_user(len, optlen)) + return -EFAULT; + + ca_ops = icsk->icsk_ca_ops; + if (ca_ops && ca_ops->get_info) + sz = ca_ops->get_info(sk, ~0U, &attr, &info); + + len = min_t(unsigned int, len, sz); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &info, len)) + return -EFAULT; + return 0; + } case TCP_QUICKACK: val = !icsk->icsk_ack.pingpong; break; @@ -2790,6 +2851,42 @@ static int do_tcp_getsockopt(struct sock *sk, int level, case TCP_NOTSENT_LOWAT: val = tp->notsent_lowat; break; + case TCP_SAVE_SYN: + val = tp->save_syn; + break; + case TCP_SAVED_SYN: { + if (get_user(len, optlen)) + return -EFAULT; + + lock_sock(sk); + if (tp->saved_syn) { + if (len < tp->saved_syn[0]) { + if (put_user(tp->saved_syn[0], optlen)) { + release_sock(sk); + return -EFAULT; + } + release_sock(sk); + return -EINVAL; + } + len = tp->saved_syn[0]; + if (put_user(len, optlen)) { + release_sock(sk); + return -EFAULT; + } + if (copy_to_user(optval, tp->saved_syn + 1, len)) { + release_sock(sk); + return -EFAULT; + } + tcp_saved_syn_free(tp); + release_sock(sk); + } else { + release_sock(sk); + len = 0; + if (put_user(len, optlen)) + return -EFAULT; + } + return 0; + } default: return -ENOPROTOOPT; } @@ -2994,11 +3091,12 @@ __setup("thash_entries=", set_thash_entries); static void __init tcp_init_mem(void) { - unsigned long limit = nr_free_buffer_pages() / 8; + unsigned long limit = nr_free_buffer_pages() / 16; + limit = max(limit, 128UL); - sysctl_tcp_mem[0] = limit / 4 * 3; - sysctl_tcp_mem[1] = limit; - sysctl_tcp_mem[2] = sysctl_tcp_mem[0] * 2; + sysctl_tcp_mem[0] = limit / 4 * 3; /* 4.68 % */ + sysctl_tcp_mem[1] = limit; /* 6.25 % */ + sysctl_tcp_mem[2] = sysctl_tcp_mem[0] * 2; /* 9.37 % */ } void __init tcp_init(void) diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 7a5ae50c80c8..84be008c945c 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -187,6 +187,7 @@ static void tcp_reinit_congestion_control(struct sock *sk, tcp_cleanup_congestion_control(sk); icsk->icsk_ca_ops = ca; + icsk->icsk_ca_setsockopt = 1; if (sk->sk_state != TCP_CLOSE && icsk->icsk_ca_ops->init) icsk->icsk_ca_ops->init(sk); @@ -335,8 +336,10 @@ int tcp_set_congestion_control(struct sock *sk, const char *name) rcu_read_lock(); ca = __tcp_ca_find_autoload(name); /* No change asking for existing value */ - if (ca == icsk->icsk_ca_ops) + if (ca == icsk->icsk_ca_ops) { + icsk->icsk_ca_setsockopt = 1; goto out; + } if (!ca) err = -ENOENT; else if (!((ca->flags & TCP_CONG_NON_RESTRICTED) || diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp_dctcp.c index 4376016f7fa5..4c41c1287197 100644 --- a/net/ipv4/tcp_dctcp.c +++ b/net/ipv4/tcp_dctcp.c @@ -277,7 +277,8 @@ static void dctcp_cwnd_event(struct sock *sk, enum tcp_ca_event ev) } } -static int dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) +static size_t dctcp_get_info(struct sock *sk, u32 ext, int *attr, + union tcp_cc_info *info) { const struct dctcp *ca = inet_csk_ca(sk); @@ -286,18 +287,17 @@ static int dctcp_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) */ if (ext & (1 << (INET_DIAG_DCTCPINFO - 1)) || ext & (1 << (INET_DIAG_VEGASINFO - 1))) { - struct tcp_dctcp_info info; - - memset(&info, 0, sizeof(info)); + memset(info, 0, sizeof(struct tcp_dctcp_info)); if (inet_csk(sk)->icsk_ca_ops != &dctcp_reno) { - info.dctcp_enabled = 1; - info.dctcp_ce_state = (u16) ca->ce_state; - info.dctcp_alpha = ca->dctcp_alpha; - info.dctcp_ab_ecn = ca->acked_bytes_ecn; - info.dctcp_ab_tot = ca->acked_bytes_total; + info->dctcp.dctcp_enabled = 1; + info->dctcp.dctcp_ce_state = (u16) ca->ce_state; + info->dctcp.dctcp_alpha = ca->dctcp_alpha; + info->dctcp.dctcp_ab_ecn = ca->acked_bytes_ecn; + info->dctcp.dctcp_ab_tot = ca->acked_bytes_total; } - return nla_put(skb, INET_DIAG_DCTCPINFO, sizeof(info), &info); + *attr = INET_DIAG_DCTCPINFO; + return sizeof(*info); } return 0; } diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp_fastopen.c index e3d87aca6be8..46b087a27503 100644 --- a/net/ipv4/tcp_fastopen.c +++ b/net/ipv4/tcp_fastopen.c @@ -206,6 +206,11 @@ static bool tcp_fastopen_create_child(struct sock *sk, skb_set_owner_r(skb2, child); __skb_queue_tail(&child->sk_receive_queue, skb2); tp->syn_data_acked = 1; + + /* u64_stats_update_begin(&tp->syncp) not needed here, + * as we certainly are not changing upper 32bit value (0) + */ + tp->bytes_received = end_seq - TCP_SKB_CB(skb)->seq - 1; } else { end_seq = TCP_SKB_CB(skb)->seq + 1; } diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp_illinois.c index 67476f085e48..f71002e4db0b 100644 --- a/net/ipv4/tcp_illinois.c +++ b/net/ipv4/tcp_illinois.c @@ -300,24 +300,25 @@ static u32 tcp_illinois_ssthresh(struct sock *sk) } /* Extract info for Tcp socket info provided via netlink. */ -static int tcp_illinois_info(struct sock *sk, u32 ext, struct sk_buff *skb) +static size_t tcp_illinois_info(struct sock *sk, u32 ext, int *attr, + union tcp_cc_info *info) { const struct illinois *ca = inet_csk_ca(sk); if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) { - struct tcpvegas_info info = { - .tcpv_enabled = 1, - .tcpv_rttcnt = ca->cnt_rtt, - .tcpv_minrtt = ca->base_rtt, - }; + info->vegas.tcpv_enabled = 1; + info->vegas.tcpv_rttcnt = ca->cnt_rtt; + info->vegas.tcpv_minrtt = ca->base_rtt; + info->vegas.tcpv_rtt = 0; - if (info.tcpv_rttcnt > 0) { + if (info->vegas.tcpv_rttcnt > 0) { u64 t = ca->sum_rtt; - do_div(t, info.tcpv_rttcnt); - info.tcpv_rtt = t; + do_div(t, info->vegas.tcpv_rttcnt); + info->vegas.tcpv_rtt = t; } - return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); + *attr = INET_DIAG_VEGASINFO; + return sizeof(struct tcpvegas_info); } return 0; } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a7ef679dd3ea..15c4536188a4 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -359,7 +359,7 @@ static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb) /* Check #1 */ if (tp->rcv_ssthresh < tp->window_clamp && (int)tp->rcv_ssthresh < tcp_space(sk) && - !sk_under_memory_pressure(sk)) { + !tcp_under_memory_pressure(sk)) { int incr; /* Check #2. Increase window, if skb with such overhead @@ -446,7 +446,7 @@ static void tcp_clamp_window(struct sock *sk) if (sk->sk_rcvbuf < sysctl_tcp_rmem[2] && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && - !sk_under_memory_pressure(sk) && + !tcp_under_memory_pressure(sk) && sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) { sk->sk_rcvbuf = min(atomic_read(&sk->sk_rmem_alloc), sysctl_tcp_rmem[2]); @@ -1130,7 +1130,12 @@ static bool tcp_check_dsack(struct sock *sk, const struct sk_buff *ack_skb, struct tcp_sacktag_state { int reord; int fack_count; - long rtt_us; /* RTT measured by SACKing never-retransmitted data */ + /* Timestamps for earliest and latest never-retransmitted segment + * that was SACKed. RTO needs the earliest RTT to stay conservative, + * but congestion control should still get an accurate delay signal. + */ + struct skb_mstamp first_sackt; + struct skb_mstamp last_sackt; int flag; }; @@ -1233,14 +1238,9 @@ static u8 tcp_sacktag_one(struct sock *sk, state->reord); if (!after(end_seq, tp->high_seq)) state->flag |= FLAG_ORIG_SACK_ACKED; - /* Pick the earliest sequence sacked for RTT */ - if (state->rtt_us < 0) { - struct skb_mstamp now; - - skb_mstamp_get(&now); - state->rtt_us = skb_mstamp_us_delta(&now, - xmit_time); - } + if (state->first_sackt.v64 == 0) + state->first_sackt = *xmit_time; + state->last_sackt = *xmit_time; } if (sacked & TCPCB_LOST) { @@ -1634,7 +1634,7 @@ static int tcp_sack_cache_ok(const struct tcp_sock *tp, const struct tcp_sack_bl static int tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, - u32 prior_snd_una, long *sack_rtt_us) + u32 prior_snd_una, struct tcp_sacktag_state *state) { struct tcp_sock *tp = tcp_sk(sk); const unsigned char *ptr = (skb_transport_header(ack_skb) + @@ -1642,7 +1642,6 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, struct tcp_sack_block_wire *sp_wire = (struct tcp_sack_block_wire *)(ptr+2); struct tcp_sack_block sp[TCP_NUM_SACKS]; struct tcp_sack_block *cache; - struct tcp_sacktag_state state; struct sk_buff *skb; int num_sacks = min(TCP_NUM_SACKS, (ptr[1] - TCPOLEN_SACK_BASE) >> 3); int used_sacks; @@ -1650,9 +1649,8 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, int i, j; int first_sack_index; - state.flag = 0; - state.reord = tp->packets_out; - state.rtt_us = -1L; + state->flag = 0; + state->reord = tp->packets_out; if (!tp->sacked_out) { if (WARN_ON(tp->fackets_out)) @@ -1663,7 +1661,7 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, found_dup_sack = tcp_check_dsack(sk, ack_skb, sp_wire, num_sacks, prior_snd_una); if (found_dup_sack) - state.flag |= FLAG_DSACKING_ACK; + state->flag |= FLAG_DSACKING_ACK; /* Eliminate too old ACKs, but take into * account more or less fresh ones, they can @@ -1728,7 +1726,7 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, } skb = tcp_write_queue_head(sk); - state.fack_count = 0; + state->fack_count = 0; i = 0; if (!tp->sacked_out) { @@ -1762,10 +1760,10 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, /* Head todo? */ if (before(start_seq, cache->start_seq)) { - skb = tcp_sacktag_skip(skb, sk, &state, + skb = tcp_sacktag_skip(skb, sk, state, start_seq); skb = tcp_sacktag_walk(skb, sk, next_dup, - &state, + state, start_seq, cache->start_seq, dup_sack); @@ -1776,7 +1774,7 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, goto advance_sp; skb = tcp_maybe_skipping_dsack(skb, sk, next_dup, - &state, + state, cache->end_seq); /* ...tail remains todo... */ @@ -1785,12 +1783,12 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, skb = tcp_highest_sack(sk); if (!skb) break; - state.fack_count = tp->fackets_out; + state->fack_count = tp->fackets_out; cache++; goto walk; } - skb = tcp_sacktag_skip(skb, sk, &state, cache->end_seq); + skb = tcp_sacktag_skip(skb, sk, state, cache->end_seq); /* Check overlap against next cached too (past this one already) */ cache++; continue; @@ -1800,12 +1798,12 @@ tcp_sacktag_write_queue(struct sock *sk, const struct sk_buff *ack_skb, skb = tcp_highest_sack(sk); if (!skb) break; - state.fack_count = tp->fackets_out; + state->fack_count = tp->fackets_out; } - skb = tcp_sacktag_skip(skb, sk, &state, start_seq); + skb = tcp_sacktag_skip(skb, sk, state, start_seq); walk: - skb = tcp_sacktag_walk(skb, sk, next_dup, &state, + skb = tcp_sacktag_walk(skb, sk, next_dup, state, start_seq, end_seq, dup_sack); advance_sp: @@ -1820,14 +1818,12 @@ advance_sp: for (j = 0; j < used_sacks; j++) tp->recv_sack_cache[i++] = sp[j]; - tcp_mark_lost_retrans(sk); - - tcp_verify_left_out(tp); - - if ((state.reord < tp->fackets_out) && + if ((state->reord < tp->fackets_out) && ((inet_csk(sk)->icsk_ca_state != TCP_CA_Loss) || tp->undo_marker)) - tcp_update_reordering(sk, tp->fackets_out - state.reord, 0); + tcp_update_reordering(sk, tp->fackets_out - state->reord, 0); + tcp_mark_lost_retrans(sk); + tcp_verify_left_out(tp); out: #if FASTRETRANS_DEBUG > 0 @@ -1836,8 +1832,7 @@ out: WARN_ON((int)tp->retrans_out < 0); WARN_ON((int)tcp_packets_in_flight(tp) < 0); #endif - *sack_rtt_us = state.rtt_us; - return state.flag; + return state->flag; } /* Limits sacked_out so that sum with lost_out isn't ever larger than @@ -2700,16 +2695,21 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack) struct tcp_sock *tp = tcp_sk(sk); bool recovered = !before(tp->snd_una, tp->high_seq); + if ((flag & FLAG_SND_UNA_ADVANCED) && + tcp_try_undo_loss(sk, false)) + return; + if (tp->frto) { /* F-RTO RFC5682 sec 3.1 (sack enhanced version). */ /* Step 3.b. A timeout is spurious if not all data are * lost, i.e., never-retransmitted data are (s)acked. */ - if (tcp_try_undo_loss(sk, flag & FLAG_ORIG_SACK_ACKED)) + if ((flag & FLAG_ORIG_SACK_ACKED) && + tcp_try_undo_loss(sk, true)) return; - if (after(tp->snd_nxt, tp->high_seq) && - (flag & FLAG_DATA_SACKED || is_dupack)) { - tp->frto = 0; /* Loss was real: 2nd part of step 3.a */ + if (after(tp->snd_nxt, tp->high_seq)) { + if (flag & FLAG_DATA_SACKED || is_dupack) + tp->frto = 0; /* Step 3.a. loss was real */ } else if (flag & FLAG_SND_UNA_ADVANCED && !recovered) { tp->high_seq = tp->snd_nxt; __tcp_push_pending_frames(sk, tcp_current_mss(sk), @@ -2734,8 +2734,6 @@ static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack) else if (flag & FLAG_SND_UNA_ADVANCED) tcp_reset_reno_sack(tp); } - if (tcp_try_undo_loss(sk, false)) - return; tcp_xmit_retransmit_queue(sk); } @@ -3054,7 +3052,8 @@ static void tcp_ack_tstamp(struct sock *sk, struct sk_buff *skb, * arrived at the other end. */ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, - u32 prior_snd_una, long sack_rtt_us) + u32 prior_snd_una, + struct tcp_sacktag_state *sack) { const struct inet_connection_sock *icsk = inet_csk(sk); struct skb_mstamp first_ackt, last_ackt, now; @@ -3062,8 +3061,9 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, u32 prior_sacked = tp->sacked_out; u32 reord = tp->packets_out; bool fully_acked = true; - long ca_seq_rtt_us = -1L; + long sack_rtt_us = -1L; long seq_rtt_us = -1L; + long ca_rtt_us = -1L; struct sk_buff *skb; u32 pkts_acked = 0; bool rtt_update; @@ -3152,15 +3152,16 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, skb_mstamp_get(&now); if (likely(first_ackt.v64)) { seq_rtt_us = skb_mstamp_us_delta(&now, &first_ackt); - ca_seq_rtt_us = skb_mstamp_us_delta(&now, &last_ackt); + ca_rtt_us = skb_mstamp_us_delta(&now, &last_ackt); + } + if (sack->first_sackt.v64) { + sack_rtt_us = skb_mstamp_us_delta(&now, &sack->first_sackt); + ca_rtt_us = skb_mstamp_us_delta(&now, &sack->last_sackt); } rtt_update = tcp_ack_update_rtt(sk, flag, seq_rtt_us, sack_rtt_us); if (flag & FLAG_ACKED) { - const struct tcp_congestion_ops *ca_ops - = inet_csk(sk)->icsk_ca_ops; - tcp_rearm_rto(sk); if (unlikely(icsk->icsk_mtup.probe_size && !after(tp->mtu_probe.probe_seq_end, tp->snd_una))) { @@ -3183,11 +3184,6 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, tp->fackets_out -= min(pkts_acked, tp->fackets_out); - if (ca_ops->pkts_acked) { - long rtt_us = min_t(ulong, ca_seq_rtt_us, sack_rtt_us); - ca_ops->pkts_acked(sk, pkts_acked, rtt_us); - } - } else if (skb && rtt_update && sack_rtt_us >= 0 && sack_rtt_us > skb_mstamp_us_delta(&now, &skb->skb_mstamp)) { /* Do not re-arm RTO if the sack RTT is measured from data sent @@ -3197,6 +3193,9 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets, tcp_rearm_rto(sk); } + if (icsk->icsk_ca_ops->pkts_acked) + icsk->icsk_ca_ops->pkts_acked(sk, pkts_acked, ca_rtt_us); + #if FASTRETRANS_DEBUG > 0 WARN_ON((int)tp->sacked_out < 0); WARN_ON((int)tp->lost_out < 0); @@ -3237,7 +3236,7 @@ static void tcp_ack_probe(struct sock *sk) * This function is not for random using! */ } else { - unsigned long when = inet_csk_rto_backoff(icsk, TCP_RTO_MAX); + unsigned long when = tcp_probe0_when(sk, TCP_RTO_MAX); inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0, when, TCP_RTO_MAX); @@ -3280,6 +3279,28 @@ static inline bool tcp_may_update_window(const struct tcp_sock *tp, (ack_seq == tp->snd_wl1 && nwin > tp->snd_wnd); } +/* If we update tp->snd_una, also update tp->bytes_acked */ +static void tcp_snd_una_update(struct tcp_sock *tp, u32 ack) +{ + u32 delta = ack - tp->snd_una; + + u64_stats_update_begin(&tp->syncp); + tp->bytes_acked += delta; + u64_stats_update_end(&tp->syncp); + tp->snd_una = ack; +} + +/* If we update tp->rcv_nxt, also update tp->bytes_received */ +static void tcp_rcv_nxt_update(struct tcp_sock *tp, u32 seq) +{ + u32 delta = seq - tp->rcv_nxt; + + u64_stats_update_begin(&tp->syncp); + tp->bytes_received += delta; + u64_stats_update_end(&tp->syncp); + tp->rcv_nxt = seq; +} + /* Update our send window. * * Window update algorithm, described in RFC793/RFC1122 (used in linux-2.2 @@ -3315,7 +3336,7 @@ static int tcp_ack_update_window(struct sock *sk, const struct sk_buff *skb, u32 } } - tp->snd_una = ack; + tcp_snd_una_update(tp, ack); return flag; } @@ -3443,6 +3464,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) { struct inet_connection_sock *icsk = inet_csk(sk); struct tcp_sock *tp = tcp_sk(sk); + struct tcp_sacktag_state sack_state; u32 prior_snd_una = tp->snd_una; u32 ack_seq = TCP_SKB_CB(skb)->seq; u32 ack = TCP_SKB_CB(skb)->ack_seq; @@ -3451,7 +3473,8 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) int prior_packets = tp->packets_out; const int prior_unsacked = tp->packets_out - tp->sacked_out; int acked = 0; /* Number of packets newly acked */ - long sack_rtt_us = -1L; + + sack_state.first_sackt.v64 = 0; /* We very likely will need to access write queue head. */ prefetchw(sk->sk_write_queue.next); @@ -3497,7 +3520,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) * Note, we use the fact that SND.UNA>=SND.WL2. */ tcp_update_wl(tp, ack_seq); - tp->snd_una = ack; + tcp_snd_una_update(tp, ack); flag |= FLAG_WIN_UPDATE; tcp_in_ack_event(sk, CA_ACK_WIN_UPDATE); @@ -3515,7 +3538,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) if (TCP_SKB_CB(skb)->sacked) flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una, - &sack_rtt_us); + &sack_state); if (tcp_ecn_rcv_ecn_echo(tp, tcp_hdr(skb))) { flag |= FLAG_ECE; @@ -3540,7 +3563,7 @@ static int tcp_ack(struct sock *sk, const struct sk_buff *skb, int flag) /* See if we can take anything off of the retransmit queue. */ acked = tp->packets_out; flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una, - sack_rtt_us); + &sack_state); acked -= tp->packets_out; /* Advance cwnd if state allows */ @@ -3592,7 +3615,7 @@ old_ack: */ if (TCP_SKB_CB(skb)->sacked) { flag |= tcp_sacktag_write_queue(sk, skb, prior_snd_una, - &sack_rtt_us); + &sack_state); tcp_fastretrans_alert(sk, acked, prior_unsacked, is_dupack, flag); } @@ -4236,7 +4259,7 @@ static void tcp_ofo_queue(struct sock *sk) tail = skb_peek_tail(&sk->sk_receive_queue); eaten = tail && tcp_try_coalesce(sk, tail, skb, &fragstolen); - tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); if (!eaten) __skb_queue_tail(&sk->sk_receive_queue, skb); if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) @@ -4404,7 +4427,7 @@ static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, int __skb_pull(skb, hdrlen); eaten = (tail && tcp_try_coalesce(sk, tail, skb, fragstolen)) ? 1 : 0; - tcp_sk(sk)->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + tcp_rcv_nxt_update(tcp_sk(sk), TCP_SKB_CB(skb)->end_seq); if (!eaten) { __skb_queue_tail(&sk->sk_receive_queue, skb); skb_set_owner_r(skb, sk); @@ -4491,13 +4514,15 @@ static void tcp_data_queue(struct sock *sk, struct sk_buff *skb) if (eaten <= 0) { queue_and_out: - if (eaten < 0 && - tcp_try_rmem_schedule(sk, skb, skb->truesize)) - goto drop; - + if (eaten < 0) { + if (skb_queue_len(&sk->sk_receive_queue) == 0) + sk_forced_mem_schedule(sk, skb->truesize); + else if (tcp_try_rmem_schedule(sk, skb, skb->truesize)) + goto drop; + } eaten = tcp_queue_rcv(sk, skb, 0, &fragstolen); } - tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); if (skb->len) tcp_event_data_recv(sk, skb); if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN) @@ -4765,7 +4790,7 @@ static int tcp_prune_queue(struct sock *sk) if (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf) tcp_clamp_window(sk); - else if (sk_under_memory_pressure(sk)) + else if (tcp_under_memory_pressure(sk)) tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U * tp->advmss); tcp_collapse_ofo_queue(sk); @@ -4809,7 +4834,7 @@ static bool tcp_should_expand_sndbuf(const struct sock *sk) return false; /* If we are under global TCP memory pressure, do not expand. */ - if (sk_under_memory_pressure(sk)) + if (tcp_under_memory_pressure(sk)) return false; /* If we are under soft global TCP memory pressure, do not expand. */ @@ -4845,6 +4870,8 @@ static void tcp_check_space(struct sock *sk) { if (sock_flag(sk, SOCK_QUEUE_SHRUNK)) { sock_reset_flag(sk, SOCK_QUEUE_SHRUNK); + /* pairs with tcp_poll() */ + smp_mb__after_atomic(); if (sk->sk_socket && test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) tcp_new_space(sk); @@ -5243,7 +5270,7 @@ void tcp_rcv_established(struct sock *sk, struct sk_buff *skb, tcp_rcv_rtt_measure_ts(sk, skb); __skb_pull(skb, tcp_header_len); - tp->rcv_nxt = TCP_SKB_CB(skb)->end_seq; + tcp_rcv_nxt_update(tp, TCP_SKB_CB(skb)->end_seq); NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPHPHITSTOUSER); eaten = 1; } @@ -6042,6 +6069,23 @@ static bool tcp_syn_flood_action(struct sock *sk, return want_cookie; } +static void tcp_reqsk_record_syn(const struct sock *sk, + struct request_sock *req, + const struct sk_buff *skb) +{ + if (tcp_sk(sk)->save_syn) { + u32 len = skb_network_header_len(skb) + tcp_hdrlen(skb); + u32 *copy; + + copy = kmalloc(len + sizeof(u32), GFP_ATOMIC); + if (copy) { + copy[0] = len; + memcpy(©[1], skb_network_header(skb), len); + req->saved_syn = copy; + } + } +} + int tcp_conn_request(struct request_sock_ops *rsk_ops, const struct tcp_request_sock_ops *af_ops, struct sock *sk, struct sk_buff *skb) @@ -6174,6 +6218,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops, tcp_rsk(req)->tfo_listener = false; af_ops->queue_hash_add(sk, req, TCP_TIMEOUT_INIT); } + tcp_reqsk_record_syn(sk, req, skb); return 0; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 3571f2be4470..d7d4c2b79cf2 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1348,7 +1348,8 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) req = inet_csk_search_req(sk, th->source, iph->saddr, iph->daddr); if (req) { nsk = tcp_check_req(sk, skb, req, false); - reqsk_put(req); + if (!nsk) + reqsk_put(req); return nsk; } @@ -1399,7 +1400,7 @@ int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb) return 0; } - if (skb->len < tcp_hdrlen(skb) || tcp_checksum_complete(skb)) + if (tcp_checksum_complete(skb)) goto csum_err; if (sk->sk_state == TCP_LISTEN) { @@ -1625,6 +1626,7 @@ process: skb->dev = NULL; bh_lock_sock_nested(sk); + tcp_sk(sk)->segs_in += max_t(u16, 1, skb_shinfo(skb)->gso_segs); ret = 0; if (!sock_owned_by_user(sk)) { if (!tcp_prequeue(sk, skb)) @@ -1645,7 +1647,7 @@ no_tcp_socket: if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) goto discard_it; - if (skb->len < (th->doff << 2) || tcp_checksum_complete(skb)) { + if (tcp_checksum_complete(skb)) { csum_error: TCP_INC_STATS_BH(net, TCP_MIB_CSUMERRORS); bad_packet: @@ -1669,10 +1671,6 @@ do_time_wait: goto discard_it; } - if (skb->len < (th->doff << 2)) { - inet_twsk_put(inet_twsk(sk)); - goto bad_packet; - } if (tcp_checksum_complete(skb)) { inet_twsk_put(inet_twsk(sk)); goto csum_error; @@ -1801,6 +1799,7 @@ void tcp_v4_destroy_sock(struct sock *sk) /* If socket is aborted during connect operation */ tcp_free_fastopen_req(tp); + tcp_saved_syn_free(tp); sk_sockets_allocated_dec(sk); sock_release_memcg(sk); @@ -2409,12 +2408,15 @@ static int __net_init tcp_sk_init(struct net *net) goto fail; *per_cpu_ptr(net->ipv4.tcp_sk, cpu) = sk; } + net->ipv4.sysctl_tcp_ecn = 2; + net->ipv4.sysctl_tcp_ecn_fallback = 1; + net->ipv4.sysctl_tcp_base_mss = TCP_BASE_MSS; net->ipv4.sysctl_tcp_probe_threshold = TCP_PROBE_THRESHOLD; net->ipv4.sysctl_tcp_probe_interval = TCP_PROBE_INTERVAL; - return 0; + return 0; fail: tcp_sk_exit(net); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 63d6311b5365..4bc00cb79e60 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -300,7 +300,7 @@ void tcp_time_wait(struct sock *sk, int state, int timeo) tw->tw_v6_daddr = sk->sk_v6_daddr; tw->tw_v6_rcv_saddr = sk->sk_v6_rcv_saddr; tw->tw_tclass = np->tclass; - tw->tw_flowlabel = np->flow_label >> 12; + tw->tw_flowlabel = be32_to_cpu(np->flow_label & IPV6_FLOWLABEL_MASK); tw->tw_ipv6only = sk->sk_ipv6only; } #endif @@ -420,7 +420,10 @@ void tcp_ca_openreq_child(struct sock *sk, const struct dst_entry *dst) rcu_read_unlock(); } - if (!ca_got_dst && !try_module_get(icsk->icsk_ca_ops->owner)) + /* If no valid choice made yet, assign current system default ca. */ + if (!ca_got_dst && + (!icsk->icsk_ca_setsockopt || + !try_module_get(icsk->icsk_ca_ops->owner))) tcp_assign_congestion_control(sk); tcp_set_ca_state(sk, TCP_CA_Open); @@ -448,6 +451,7 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->rcv_wup = newtp->copied_seq = newtp->rcv_nxt = treq->rcv_isn + 1; + newtp->segs_in = 0; newtp->snd_sml = newtp->snd_una = newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1; @@ -536,6 +540,9 @@ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, newtp->fastopen_rsk = NULL; newtp->syn_data_acked = 0; + newtp->saved_syn = req->saved_syn; + req->saved_syn = NULL; + TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_PASSIVEOPENS); } return newsk; @@ -755,10 +762,11 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb, if (!child) goto listen_overflow; - inet_csk_reqsk_queue_unlink(sk, req); - inet_csk_reqsk_queue_removed(sk, req); - + inet_csk_reqsk_queue_drop(sk, req); inet_csk_reqsk_queue_add(sk, req, child); + /* Warning: caller must not call reqsk_put(req); + * child stole last reference on it. + */ return child; listen_overflow: diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 8c8d7e06b72f..eeb59befaf06 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -50,8 +50,8 @@ int sysctl_tcp_retrans_collapse __read_mostly = 1; */ int sysctl_tcp_workaround_signed_windows __read_mostly = 0; -/* Default TSQ limit of two TSO segments */ -int sysctl_tcp_limit_output_bytes __read_mostly = 131072; +/* Default TSQ limit of four TSO segments */ +int sysctl_tcp_limit_output_bytes __read_mostly = 262144; /* This limits the percentage of the congestion window which we * will allow a single TSO frame to consume. Building TSO frames @@ -350,6 +350,15 @@ static void tcp_ecn_send_syn(struct sock *sk, struct sk_buff *skb) } } +static void tcp_ecn_clear_syn(struct sock *sk, struct sk_buff *skb) +{ + if (sock_net(sk)->ipv4.sysctl_tcp_ecn_fallback) + /* tp->ecn_flags are cleared at a later point in time when + * SYN ACK is ultimatively being received. + */ + TCP_SKB_CB(skb)->tcp_flags &= ~(TCPHDR_ECE | TCPHDR_CWR); +} + static void tcp_ecn_make_synack(const struct request_sock *req, struct tcphdr *th, struct sock *sk) @@ -1018,6 +1027,7 @@ static int tcp_transmit_skb(struct sock *sk, struct sk_buff *skb, int clone_it, TCP_ADD_STATS(sock_net(sk), TCP_MIB_OUTSEGS, tcp_skb_pcount(skb)); + tp->segs_out += tcp_skb_pcount(skb); /* OK, its time to fill skb_shinfo(skb)->gso_segs */ skb_shinfo(skb)->gso_segs = tcp_skb_pcount(skb); @@ -1163,7 +1173,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len, return -ENOMEM; /* Get a new skb... force flag on. */ - buff = sk_stream_alloc_skb(sk, nsize, gfp); + buff = sk_stream_alloc_skb(sk, nsize, gfp, true); if (!buff) return -ENOMEM; /* We'll just try again later. */ @@ -1722,7 +1732,7 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len, if (skb->len != skb->data_len) return tcp_fragment(sk, skb, len, mss_now, gfp); - buff = sk_stream_alloc_skb(sk, 0, gfp); + buff = sk_stream_alloc_skb(sk, 0, gfp, true); if (unlikely(!buff)) return -ENOMEM; @@ -1941,7 +1951,7 @@ static int tcp_mtu_probe(struct sock *sk) } /* We're allowed to probe. Build it now. */ - nskb = sk_stream_alloc_skb(sk, probe_size, GFP_ATOMIC); + nskb = sk_stream_alloc_skb(sk, probe_size, GFP_ATOMIC, false); if (!nskb) return -1; sk->sk_wmem_queued += nskb->truesize; @@ -2078,7 +2088,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, if (unlikely(!tcp_snd_wnd_test(tp, skb, mss_now))) break; - if (tso_segs == 1 || !max_segs) { + if (tso_segs == 1) { if (unlikely(!tcp_nagle_test(tp, skb, mss_now, (tcp_skb_is_last(sk, skb) ? nonagle : TCP_NAGLE_PUSH)))) @@ -2091,7 +2101,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle, } limit = mss_now; - if (tso_segs > 1 && max_segs && !tcp_urg_mode(tp)) + if (tso_segs > 1 && !tcp_urg_mode(tp)) limit = tcp_mss_split_point(sk, skb, mss_now, min_t(unsigned int, cwnd_quota, @@ -2392,7 +2402,7 @@ u32 __tcp_select_window(struct sock *sk) if (free_space < (full_space >> 1)) { icsk->icsk_ack.quick = 0; - if (sk_under_memory_pressure(sk)) + if (tcp_under_memory_pressure(sk)) tp->rcv_ssthresh = min(tp->rcv_ssthresh, 4U * tp->advmss); @@ -2615,6 +2625,10 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) } } + /* RFC3168, section 6.1.1.1. ECN fallback */ + if ((TCP_SKB_CB(skb)->tcp_flags & TCPHDR_SYN_ECN) == TCPHDR_SYN_ECN) + tcp_ecn_clear_syn(sk, skb); + tcp_retrans_try_collapse(sk, skb, cur_mss); /* Make a copy, if the first transmission SKB clone we made @@ -2812,39 +2826,67 @@ begin_fwd: } } -/* Send a fin. The caller locks the socket for us. This cannot be - * allowed to fail queueing a FIN frame under any circumstances. +/* We allow to exceed memory limits for FIN packets to expedite + * connection tear down and (memory) recovery. + * Otherwise tcp_send_fin() could be tempted to either delay FIN + * or even be forced to close flow without any FIN. + * In general, we want to allow one skb per socket to avoid hangs + * with edge trigger epoll() + */ +void sk_forced_mem_schedule(struct sock *sk, int size) +{ + int amt, status; + + if (size <= sk->sk_forward_alloc) + return; + amt = sk_mem_pages(size); + sk->sk_forward_alloc += amt * SK_MEM_QUANTUM; + sk_memory_allocated_add(sk, amt, &status); +} + +/* Send a FIN. The caller locks the socket for us. + * We should try to send a FIN packet really hard, but eventually give up. */ void tcp_send_fin(struct sock *sk) { + struct sk_buff *skb, *tskb = tcp_write_queue_tail(sk); struct tcp_sock *tp = tcp_sk(sk); - struct sk_buff *skb = tcp_write_queue_tail(sk); - int mss_now; - /* Optimization, tack on the FIN if we have a queue of - * unsent frames. But be careful about outgoing SACKS - * and IP options. + /* Optimization, tack on the FIN if we have one skb in write queue and + * this skb was not yet sent, or we are under memory pressure. + * Note: in the latter case, FIN packet will be sent after a timeout, + * as TCP stack thinks it has already been transmitted. */ - mss_now = tcp_current_mss(sk); - - if (tcp_send_head(sk)) { - TCP_SKB_CB(skb)->tcp_flags |= TCPHDR_FIN; - TCP_SKB_CB(skb)->end_seq++; + if (tskb && (tcp_send_head(sk) || tcp_under_memory_pressure(sk))) { +coalesce: + TCP_SKB_CB(tskb)->tcp_flags |= TCPHDR_FIN; + TCP_SKB_CB(tskb)->end_seq++; tp->write_seq++; + if (!tcp_send_head(sk)) { + /* This means tskb was already sent. + * Pretend we included the FIN on previous transmit. + * We need to set tp->snd_nxt to the value it would have + * if FIN had been sent. This is because retransmit path + * does not change tp->snd_nxt. + */ + tp->snd_nxt++; + return; + } } else { - /* Socket is locked, keep trying until memory is available. */ - for (;;) { - skb = sk_stream_alloc_skb(sk, 0, sk->sk_allocation); - if (skb) - break; - yield(); + skb = alloc_skb_fclone(MAX_TCP_HEADER, sk->sk_allocation); + if (unlikely(!skb)) { + if (tskb) + goto coalesce; + return; } + skb_reserve(skb, MAX_TCP_HEADER); + sk_forced_mem_schedule(sk, skb->truesize); /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */ tcp_init_nondata_skb(skb, tp->write_seq, TCPHDR_ACK | TCPHDR_FIN); tcp_queue_skb(sk, skb); } - __tcp_push_pending_frames(sk, mss_now, TCP_NAGLE_OFF); + __tcp_push_pending_frames(sk, tcp_current_mss(sk), TCP_NAGLE_OFF); } /* We get here when a process closes a file descriptor (either due to @@ -3149,7 +3191,7 @@ static int tcp_send_syn_data(struct sock *sk, struct sk_buff *syn) /* limit to order-0 allocations */ space = min_t(size_t, space, SKB_MAX_HEAD(MAX_TCP_HEADER)); - syn_data = sk_stream_alloc_skb(sk, space, sk->sk_allocation); + syn_data = sk_stream_alloc_skb(sk, space, sk->sk_allocation, false); if (!syn_data) goto fallback; syn_data->ip_summed = CHECKSUM_PARTIAL; @@ -3215,7 +3257,7 @@ int tcp_connect(struct sock *sk) return 0; } - buff = sk_stream_alloc_skb(sk, 0, sk->sk_allocation); + buff = sk_stream_alloc_skb(sk, 0, sk->sk_allocation, true); if (unlikely(!buff)) return -ENOBUFS; @@ -3356,7 +3398,7 @@ EXPORT_SYMBOL_GPL(tcp_send_ack); * one is with SEG.SEQ=SND.UNA to deliver urgent pointer, another is * out-of-date with SND.UNA-1 to probe window. */ -static int tcp_xmit_probe_skb(struct sock *sk, int urgent) +static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; @@ -3374,6 +3416,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent) */ tcp_init_nondata_skb(skb, tp->snd_una - !urgent, TCPHDR_ACK); skb_mstamp_get(&skb->skb_mstamp); + NET_INC_STATS_BH(sock_net(sk), mib); return tcp_transmit_skb(sk, skb, 0, GFP_ATOMIC); } @@ -3381,12 +3424,12 @@ void tcp_send_window_probe(struct sock *sk) { if (sk->sk_state == TCP_ESTABLISHED) { tcp_sk(sk)->snd_wl1 = tcp_sk(sk)->rcv_nxt - 1; - tcp_xmit_probe_skb(sk, 0); + tcp_xmit_probe_skb(sk, 0, LINUX_MIB_TCPWINPROBE); } } /* Initiate keepalive or window probe from timer. */ -int tcp_write_wakeup(struct sock *sk) +int tcp_write_wakeup(struct sock *sk, int mib) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; @@ -3423,8 +3466,8 @@ int tcp_write_wakeup(struct sock *sk) return err; } else { if (between(tp->snd_up, tp->snd_una + 1, tp->snd_una + 0xFFFF)) - tcp_xmit_probe_skb(sk, 1); - return tcp_xmit_probe_skb(sk, 0); + tcp_xmit_probe_skb(sk, 1, mib); + return tcp_xmit_probe_skb(sk, 0, mib); } } @@ -3438,7 +3481,7 @@ void tcp_send_probe0(struct sock *sk) unsigned long probe_max; int err; - err = tcp_write_wakeup(sk); + err = tcp_write_wakeup(sk, LINUX_MIB_TCPWINPROBE); if (tp->packets_out || !tcp_send_head(sk)) { /* Cancel probe timer, if it is not required. */ @@ -3464,7 +3507,7 @@ void tcp_send_probe0(struct sock *sk) probe_max = TCP_RESOURCE_PROBE_INTERVAL; } inet_csk_reset_xmit_timer(sk, ICSK_TIME_PROBE0, - inet_csk_rto_backoff(icsk, probe_max), + tcp_probe0_when(sk, probe_max), TCP_RTO_MAX); } diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 8c65dc147d8b..5b752f58a900 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -247,7 +247,7 @@ void tcp_delack_timer_handler(struct sock *sk) } out: - if (sk_under_memory_pressure(sk)) + if (tcp_under_memory_pressure(sk)) sk_mem_reclaim(sk); } @@ -616,7 +616,7 @@ static void tcp_keepalive_timer (unsigned long data) tcp_write_err(sk); goto out; } - if (tcp_write_wakeup(sk) <= 0) { + if (tcp_write_wakeup(sk, LINUX_MIB_TCPKEEPALIVE) <= 0) { icsk->icsk_probes_out++; elapsed = keepalive_intvl_when(tp); } else { diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp_vegas.c index c71a1b8f7bde..a6cea1d5e20d 100644 --- a/net/ipv4/tcp_vegas.c +++ b/net/ipv4/tcp_vegas.c @@ -286,18 +286,19 @@ static void tcp_vegas_cong_avoid(struct sock *sk, u32 ack, u32 acked) } /* Extract info for Tcp socket info provided via netlink. */ -int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb) +size_t tcp_vegas_get_info(struct sock *sk, u32 ext, int *attr, + union tcp_cc_info *info) { const struct vegas *ca = inet_csk_ca(sk); + if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) { - struct tcpvegas_info info = { - .tcpv_enabled = ca->doing_vegas_now, - .tcpv_rttcnt = ca->cntRTT, - .tcpv_rtt = ca->baseRTT, - .tcpv_minrtt = ca->minRTT, - }; - - return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); + info->vegas.tcpv_enabled = ca->doing_vegas_now, + info->vegas.tcpv_rttcnt = ca->cntRTT, + info->vegas.tcpv_rtt = ca->baseRTT, + info->vegas.tcpv_minrtt = ca->minRTT, + + *attr = INET_DIAG_VEGASINFO; + return sizeof(struct tcpvegas_info); } return 0; } diff --git a/net/ipv4/tcp_vegas.h b/net/ipv4/tcp_vegas.h index e8a6b33cc61d..ef9da5306c68 100644 --- a/net/ipv4/tcp_vegas.h +++ b/net/ipv4/tcp_vegas.h @@ -19,6 +19,7 @@ void tcp_vegas_init(struct sock *sk); void tcp_vegas_state(struct sock *sk, u8 ca_state); void tcp_vegas_pkts_acked(struct sock *sk, u32 cnt, s32 rtt_us); void tcp_vegas_cwnd_event(struct sock *sk, enum tcp_ca_event event); -int tcp_vegas_get_info(struct sock *sk, u32 ext, struct sk_buff *skb); +size_t tcp_vegas_get_info(struct sock *sk, u32 ext, int *attr, + union tcp_cc_info *info); #endif /* __TCP_VEGAS_H */ diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp_westwood.c index b3c57cceb990..c10732e39837 100644 --- a/net/ipv4/tcp_westwood.c +++ b/net/ipv4/tcp_westwood.c @@ -256,18 +256,19 @@ static void tcp_westwood_event(struct sock *sk, enum tcp_ca_event event) } /* Extract info for Tcp socket info provided via netlink. */ -static int tcp_westwood_info(struct sock *sk, u32 ext, struct sk_buff *skb) +static size_t tcp_westwood_info(struct sock *sk, u32 ext, int *attr, + union tcp_cc_info *info) { const struct westwood *ca = inet_csk_ca(sk); if (ext & (1 << (INET_DIAG_VEGASINFO - 1))) { - struct tcpvegas_info info = { - .tcpv_enabled = 1, - .tcpv_rtt = jiffies_to_usecs(ca->rtt), - .tcpv_minrtt = jiffies_to_usecs(ca->rtt_min), - }; + info->vegas.tcpv_enabled = 1; + info->vegas.tcpv_rttcnt = 0; + info->vegas.tcpv_rtt = jiffies_to_usecs(ca->rtt), + info->vegas.tcpv_minrtt = jiffies_to_usecs(ca->rtt_min), - return nla_put(skb, INET_DIAG_VEGASINFO, sizeof(info), &info); + *attr = INET_DIAG_VEGASINFO; + return sizeof(struct tcpvegas_info); } return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index d10b7e0112eb..83aa604f9273 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -90,6 +90,7 @@ #include <linux/socket.h> #include <linux/sockios.h> #include <linux/igmp.h> +#include <linux/inetdevice.h> #include <linux/in.h> #include <linux/errno.h> #include <linux/timer.h> @@ -1345,10 +1346,8 @@ csum_copy_err: } unlock_sock_fast(sk, slow); - if (noblock) - return -EAGAIN; - - /* starting over for a new packet */ + /* starting over for a new packet, but check if we need to yield */ + cond_resched(); msg->msg_flags &= ~MSG_TRUNC; goto try_again; } @@ -1962,6 +1961,7 @@ void udp_v4_early_demux(struct sk_buff *skb) struct sock *sk; struct dst_entry *dst; int dif = skb->dev->ifindex; + int ours; /* validate the packet */ if (!pskb_may_pull(skb, skb_transport_offset(skb) + sizeof(struct udphdr))) @@ -1971,14 +1971,24 @@ void udp_v4_early_demux(struct sk_buff *skb) uh = udp_hdr(skb); if (skb->pkt_type == PACKET_BROADCAST || - skb->pkt_type == PACKET_MULTICAST) + skb->pkt_type == PACKET_MULTICAST) { + struct in_device *in_dev = __in_dev_get_rcu(skb->dev); + + if (!in_dev) + return; + + ours = ip_check_mc_rcu(in_dev, iph->daddr, iph->saddr, + iph->protocol); + if (!ours) + return; sk = __udp4_lib_mcast_demux_lookup(net, uh->dest, iph->daddr, uh->source, iph->saddr, dif); - else if (skb->pkt_type == PACKET_HOST) + } else if (skb->pkt_type == PACKET_HOST) { sk = __udp4_lib_demux_lookup(net, uh->dest, iph->daddr, uh->source, iph->saddr, dif); - else + } else { return; + } if (!sk) return; diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 6bb98cc193c9..933ea903f7b8 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -15,12 +15,10 @@ int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg, struct socket *sock = NULL; struct sockaddr_in udp_addr; - err = sock_create_kern(AF_INET, SOCK_DGRAM, 0, &sock); + err = sock_create_kern(net, AF_INET, SOCK_DGRAM, 0, &sock); if (err < 0) goto error; - sk_change_net(sock->sk, net); - udp_addr.sin_family = AF_INET; udp_addr.sin_addr = cfg->local_ip; udp_addr.sin_port = cfg->local_udp_port; @@ -47,7 +45,7 @@ int udp_sock_create4(struct net *net, struct udp_port_cfg *cfg, error: if (sock) { kernel_sock_shutdown(sock, SHUT_RDWR); - sk_release_kernel(sock->sk); + sock_release(sock); } *sockp = NULL; return err; @@ -101,7 +99,7 @@ void udp_tunnel_sock_release(struct socket *sock) { rcu_assign_sk_user_data(sock->sk, NULL); kernel_sock_shutdown(sock, SHUT_RDWR); - sk_release_kernel(sock->sk); + sock_release(sock); } EXPORT_SYMBOL_GPL(udp_tunnel_sock_release); |