X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=net%2Fdccp%2Fipv4.c;h=3baf4c76a89d0311b12ede5242989a878b4b1dec;hb=017487d7d1e905a5bb529f6a2bc8cf8ea14e2307;hp=656e13e38cfb69622e02df1d54aadd08e334bdfc;hpb=db9ace7083dbdcc3d02bdd6a1d26132c80b5b726;p=linux-2.6 diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 656e13e38c..3baf4c76a8 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -19,13 +19,16 @@ #include #include +#include #include +#include #include #include #include "ackvec.h" #include "ccid.h" #include "dccp.h" +#include "feat.h" struct inet_hashinfo __cacheline_aligned dccp_hashinfo = { .lhash_lock = RW_LOCK_UNLOCKED, @@ -37,7 +40,8 @@ EXPORT_SYMBOL_GPL(dccp_hashinfo); static int dccp_v4_get_port(struct sock *sk, const unsigned short snum) { - return inet_csk_get_port(&dccp_hashinfo, sk, snum); + return inet_csk_get_port(&dccp_hashinfo, sk, snum, + inet_csk_bind_conflict); } static void dccp_v4_hash(struct sock *sk) @@ -45,171 +49,14 @@ static void dccp_v4_hash(struct sock *sk) inet_hash(&dccp_hashinfo, sk); } -static void dccp_v4_unhash(struct sock *sk) +void dccp_unhash(struct sock *sk) { inet_unhash(&dccp_hashinfo, sk); } -/* called with local bh disabled */ -static int __dccp_v4_check_established(struct sock *sk, const __u16 lport, - struct inet_timewait_sock **twp) -{ - struct inet_sock *inet = inet_sk(sk); - const u32 daddr = inet->rcv_saddr; - const u32 saddr = inet->daddr; - const int dif = sk->sk_bound_dev_if; - INET_ADDR_COOKIE(acookie, saddr, daddr) - const __u32 ports = INET_COMBINED_PORTS(inet->dport, lport); - unsigned int hash = inet_ehashfn(daddr, lport, saddr, inet->dport); - struct inet_ehash_bucket *head = inet_ehash_bucket(&dccp_hashinfo, hash); - const struct sock *sk2; - const struct hlist_node *node; - struct inet_timewait_sock *tw; - - prefetch(head->chain.first); - write_lock(&head->lock); - - /* Check TIME-WAIT sockets first. */ - sk_for_each(sk2, node, &(head + dccp_hashinfo.ehash_size)->chain) { - tw = inet_twsk(sk2); - - if (INET_TW_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) - goto not_unique; - } - tw = NULL; - - /* And established part... */ - sk_for_each(sk2, node, &head->chain) { - if (INET_MATCH(sk2, hash, acookie, saddr, daddr, ports, dif)) - goto not_unique; - } - - /* Must record num and sport now. Otherwise we will see - * in hash table socket with a funny identity. */ - inet->num = lport; - inet->sport = htons(lport); - sk->sk_hash = hash; - BUG_TRAP(sk_unhashed(sk)); - __sk_add_node(sk, &head->chain); - sock_prot_inc_use(sk->sk_prot); - write_unlock(&head->lock); - - if (twp != NULL) { - *twp = tw; - NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED); - } else if (tw != NULL) { - /* Silly. Should hash-dance instead... */ - inet_twsk_deschedule(tw, &dccp_death_row); - NET_INC_STATS_BH(LINUX_MIB_TIMEWAITRECYCLED); - - inet_twsk_put(tw); - } - - return 0; - -not_unique: - write_unlock(&head->lock); - return -EADDRNOTAVAIL; -} - -/* - * Bind a port for a connect operation and hash it. - */ -static int dccp_v4_hash_connect(struct sock *sk) -{ - const unsigned short snum = inet_sk(sk)->num; - struct inet_bind_hashbucket *head; - struct inet_bind_bucket *tb; - int ret; - - if (snum == 0) { - int low = sysctl_local_port_range[0]; - int high = sysctl_local_port_range[1]; - int remaining = (high - low) + 1; - int rover = net_random() % (high - low) + low; - struct hlist_node *node; - struct inet_timewait_sock *tw = NULL; - - local_bh_disable(); - do { - head = &dccp_hashinfo.bhash[inet_bhashfn(rover, - dccp_hashinfo.bhash_size)]; - spin_lock(&head->lock); - - /* Does not bother with rcv_saddr checks, - * because the established check is already - * unique enough. - */ - inet_bind_bucket_for_each(tb, node, &head->chain) { - if (tb->port == rover) { - BUG_TRAP(!hlist_empty(&tb->owners)); - if (tb->fastreuse >= 0) - goto next_port; - if (!__dccp_v4_check_established(sk, - rover, - &tw)) - goto ok; - goto next_port; - } - } - - tb = inet_bind_bucket_create(dccp_hashinfo.bind_bucket_cachep, - head, rover); - if (tb == NULL) { - spin_unlock(&head->lock); - break; - } - tb->fastreuse = -1; - goto ok; - - next_port: - spin_unlock(&head->lock); - if (++rover > high) - rover = low; - } while (--remaining > 0); - - local_bh_enable(); - - return -EADDRNOTAVAIL; - -ok: - /* All locks still held and bhs disabled */ - inet_bind_hash(sk, tb, rover); - if (sk_unhashed(sk)) { - inet_sk(sk)->sport = htons(rover); - __inet_hash(&dccp_hashinfo, sk, 0); - } - spin_unlock(&head->lock); - - if (tw != NULL) { - inet_twsk_deschedule(tw, &dccp_death_row); - inet_twsk_put(tw); - } - - ret = 0; - goto out; - } - - head = &dccp_hashinfo.bhash[inet_bhashfn(snum, - dccp_hashinfo.bhash_size)]; - tb = inet_csk(sk)->icsk_bind_hash; - spin_lock_bh(&head->lock); - if (sk_head(&tb->owners) == sk && sk->sk_bind_node.next == NULL) { - __inet_hash(&dccp_hashinfo, sk, 0); - spin_unlock_bh(&head->lock); - return 0; - } else { - spin_unlock(&head->lock); - /* No definite answer... Walk to established hash table */ - ret = __dccp_v4_check_established(sk, snum, NULL); -out: - local_bh_enable(); - return ret; - } -} +EXPORT_SYMBOL_GPL(dccp_unhash); -static int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, - int addr_len) +int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { struct inet_sock *inet = inet_sk(sk); struct dccp_sock *dp = dccp_sk(sk); @@ -259,9 +106,9 @@ static int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, inet->dport = usin->sin_port; inet->daddr = daddr; - dp->dccps_ext_header_len = 0; + inet_csk(sk)->icsk_ext_hdr_len = 0; if (inet->opt != NULL) - dp->dccps_ext_header_len = inet->opt->optlen; + inet_csk(sk)->icsk_ext_hdr_len = inet->opt->optlen; /* * Socket identity is still unknown (sport may be zero). * However we set state to DCCP_REQUESTING and not releasing socket @@ -269,11 +116,12 @@ static int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, * complete initialization after this. */ dccp_set_state(sk, DCCP_REQUESTING); - err = dccp_v4_hash_connect(sk); + err = inet_hash_connect(&dccp_death_row, sk); if (err != 0) goto failure; - err = ip_route_newports(&rt, inet->sport, inet->dport, sk); + err = ip_route_newports(&rt, IPPROTO_DCCP, inet->sport, inet->dport, + sk); if (err != 0) goto failure; @@ -287,16 +135,6 @@ static int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, usin->sin_port); dccp_update_gss(sk, dp->dccps_iss); - /* - * SWL and AWL are initially adjusted so that they are not less than - * the initial Sequence Numbers received and sent, respectively: - * SWL := max(GSR + 1 - floor(W/4), ISR), - * AWL := max(GSS - W' + 1, ISS). - * These adjustments MUST be applied only at the beginning of the - * connection. - */ - dccp_set_seqno(&dp->dccps_awl, max48(dp->dccps_awl, dp->dccps_iss)); - inet->id = dp->dccps_iss ^ jiffies; err = dccp_connect(sk); @@ -316,6 +154,8 @@ failure: goto out; } +EXPORT_SYMBOL_GPL(dccp_v4_connect); + /* * This routine does path mtu discovery as defined in RFC1191. */ @@ -354,7 +194,7 @@ static inline void dccp_do_pmtu_discovery(struct sock *sk, mtu = dst_mtu(dst); if (inet->pmtudisc != IP_PMTUDISC_DONT && - dp->dccps_pmtu_cookie > mtu) { + inet_csk(sk)->icsk_pmtu_cookie > mtu) { dccp_sync_mss(sk, mtu); /* @@ -606,32 +446,17 @@ out: sock_put(sk); } -int dccp_v4_send_reset(struct sock *sk, enum dccp_reset_codes code) +/* This routine computes an IPv4 DCCP checksum. */ +void dccp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb) { - struct sk_buff *skb; - /* - * FIXME: what if rebuild_header fails? - * Should we be doing a rebuild_header here? - */ - int err = inet_sk_rebuild_header(sk); - - if (err != 0) - return err; - - skb = dccp_make_reset(sk, sk->sk_dst_cache, code); - if (skb != NULL) { - const struct inet_sock *inet = inet_sk(sk); - - memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt)); - err = ip_build_and_send_pkt(skb, sk, - inet->saddr, inet->daddr, NULL); - if (err == NET_XMIT_CN) - err = 0; - } + const struct inet_sock *inet = inet_sk(sk); + struct dccp_hdr *dh = dccp_hdr(skb); - return err; + dh->dccph_checksum = dccp_v4_checksum(skb, inet->saddr, inet->daddr); } +EXPORT_SYMBOL_GPL(dccp_v4_send_check); + static inline u64 dccp_v4_init_sequence(const struct sock *sk, const struct sk_buff *skb) { @@ -641,28 +466,17 @@ static inline u64 dccp_v4_init_sequence(const struct sock *sk, dccp_hdr(skb)->dccph_sport); } -static inline int dccp_bad_service_code(const struct sock *sk, - const __u32 service) -{ - const struct dccp_sock *dp = dccp_sk(sk); - - if (dp->dccps_service == service) - return 0; - return !dccp_list_has_service(dp->dccps_service_list, service); -} - int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) { struct inet_request_sock *ireq; struct dccp_sock dp; struct request_sock *req; struct dccp_request_sock *dreq; - const __u32 saddr = skb->nh.iph->saddr; - const __u32 daddr = skb->nh.iph->daddr; - const __u32 service = dccp_hdr_request(skb)->dccph_req_service; + const __be32 saddr = skb->nh.iph->saddr; + const __be32 daddr = skb->nh.iph->daddr; + const __be32 service = dccp_hdr_request(skb)->dccph_req_service; struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); __u8 reset_code = DCCP_RESET_CODE_TOO_BUSY; - struct dst_entry *dst = NULL; /* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */ if (((struct rtable *)skb->dst)->rt_flags & @@ -696,14 +510,14 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (req == NULL) goto drop; - /* FIXME: process options */ + if (dccp_parse_options(sk, skb)) + goto drop; dccp_openreq_init(req, &dp, skb); ireq = inet_rsk(req); ireq->loc_addr = daddr; ireq->rmt_addr = saddr; - /* FIXME: Merge Aristeu's option parsing code when ready */ req->rcv_wnd = 100; /* Fake, option parsing will get the right value */ ireq->opt = NULL; @@ -721,23 +535,22 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) dreq->dreq_iss = dccp_v4_init_sequence(sk, skb); dreq->dreq_service = service; - if (dccp_v4_send_response(sk, req, dst)) + if (dccp_v4_send_response(sk, req, NULL)) goto drop_and_free; inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT); return 0; drop_and_free: - /* - * FIXME: should be reqsk_free after implementing req->rsk_ops - */ - __reqsk_free(req); + reqsk_free(req); drop: DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS); dcb->dccpd_reset_code = reset_code; return -1; } +EXPORT_SYMBOL_GPL(dccp_v4_conn_request); + /* * The three way handshake has completed - we got a valid ACK or DATAACK - * now create the new socket. @@ -792,6 +605,8 @@ exit: return NULL; } +EXPORT_SYMBOL_GPL(dccp_v4_request_recv_sock); + static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) { const struct dccp_hdr *dh = dccp_hdr(skb); @@ -821,8 +636,8 @@ static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) return sk; } -int dccp_v4_checksum(const struct sk_buff *skb, const u32 saddr, - const u32 daddr) +int dccp_v4_checksum(const struct sk_buff *skb, const __be32 saddr, + const __be32 daddr) { const struct dccp_hdr* dh = dccp_hdr(skb); int checksum_len; @@ -842,7 +657,7 @@ int dccp_v4_checksum(const struct sk_buff *skb, const u32 saddr, } static int dccp_v4_verify_checksum(struct sk_buff *skb, - const u32 saddr, const u32 daddr) + const __be32 saddr, const __be32 daddr) { struct dccp_hdr *dh = dccp_hdr(skb); int checksum_len; @@ -1011,7 +826,9 @@ discard: return 0; } -static inline int dccp_invalid_packet(struct sk_buff *skb) +EXPORT_SYMBOL_GPL(dccp_v4_do_rcv); + +int dccp_invalid_packet(struct sk_buff *skb) { const struct dccp_hdr *dh; @@ -1065,29 +882,30 @@ static inline int dccp_invalid_packet(struct sk_buff *skb) return 1; } - /* If the header checksum is incorrect, drop packet and return */ - if (dccp_v4_verify_checksum(skb, skb->nh.iph->saddr, - skb->nh.iph->daddr) < 0) { - LIMIT_NETDEBUG(KERN_WARNING "DCCP: header checksum is " - "incorrect\n"); - return 1; - } - return 0; } +EXPORT_SYMBOL_GPL(dccp_invalid_packet); + /* this is called when real data arrives */ int dccp_v4_rcv(struct sk_buff *skb) { const struct dccp_hdr *dh; struct sock *sk; - int rc; /* Step 1: Check header basics: */ if (dccp_invalid_packet(skb)) goto discard_it; + /* If the header checksum is incorrect, drop packet and return */ + if (dccp_v4_verify_checksum(skb, skb->nh.iph->saddr, + skb->nh.iph->daddr) < 0) { + LIMIT_NETDEBUG(KERN_WARNING "%s: incorrect header checksum\n", + __FUNCTION__); + goto discard_it; + } + dh = dccp_hdr(skb); DCCP_SKB_CB(skb)->dccpd_seq = dccp_hdr_seq(skb); @@ -1143,28 +961,11 @@ int dccp_v4_rcv(struct sk_buff *skb) goto do_time_wait; } - if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) { - dccp_pr_debug("xfrm4_policy_check failed\n"); + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) goto discard_and_relse; - } - - if (sk_filter(sk, skb, 0)) { - dccp_pr_debug("sk_filter failed\n"); - goto discard_and_relse; - } - - skb->dev = NULL; - - bh_lock_sock(sk); - rc = 0; - if (!sock_owned_by_user(sk)) - rc = dccp_v4_do_rcv(sk, skb); - else - sk_add_backlog(sk, skb); - bh_unlock_sock(sk); + nf_reset(skb); - sock_put(sk); - return rc; + return sk_receive_skb(sk, skb); no_dccp_socket: if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) @@ -1194,21 +995,28 @@ do_time_wait: goto no_dccp_socket; } -static int dccp_v4_init_sock(struct sock *sk) +struct inet_connection_sock_af_ops dccp_ipv4_af_ops = { + .queue_xmit = ip_queue_xmit, + .send_check = dccp_v4_send_check, + .rebuild_header = inet_sk_rebuild_header, + .conn_request = dccp_v4_conn_request, + .syn_recv_sock = dccp_v4_request_recv_sock, + .net_header_len = sizeof(struct iphdr), + .setsockopt = ip_setsockopt, + .getsockopt = ip_getsockopt, + .addr2sockaddr = inet_csk_addr2sockaddr, + .sockaddr_len = sizeof(struct sockaddr_in), +}; + +int dccp_v4_init_sock(struct sock *sk) { struct dccp_sock *dp = dccp_sk(sk); + struct inet_connection_sock *icsk = inet_csk(sk); static int dccp_ctl_socket_init = 1; dccp_options_init(&dp->dccps_options); do_gettimeofday(&dp->dccps_epoch); - if (dp->dccps_options.dccpo_send_ack_vector) { - dp->dccps_hc_rx_ackvec = dccp_ackvec_alloc(DCCP_MAX_ACKVEC_LEN, - GFP_KERNEL); - if (dp->dccps_hc_rx_ackvec == NULL) - return -ENOMEM; - } - /* * FIXME: We're hardcoding the CCID, and doing this at this point makes * the listening (master) sock get CCID control blocks, which is not @@ -1217,14 +1025,26 @@ static int dccp_v4_init_sock(struct sock *sk) * setsockopt(CCIDs-I-want/accept). -acme */ if (likely(!dccp_ctl_socket_init)) { - dp->dccps_hc_rx_ccid = ccid_init(dp->dccps_options.dccpo_rx_ccid, - sk); - dp->dccps_hc_tx_ccid = ccid_init(dp->dccps_options.dccpo_tx_ccid, - sk); - if (dp->dccps_hc_rx_ccid == NULL || - dp->dccps_hc_tx_ccid == NULL) { - ccid_exit(dp->dccps_hc_rx_ccid, sk); - ccid_exit(dp->dccps_hc_tx_ccid, sk); + int rc = dccp_feat_init(sk); + + if (rc) + return rc; + + if (dp->dccps_options.dccpo_send_ack_vector) { + dp->dccps_hc_rx_ackvec = dccp_ackvec_alloc(GFP_KERNEL); + if (dp->dccps_hc_rx_ackvec == NULL) + return -ENOMEM; + } + dp->dccps_hc_rx_ccid = + ccid_hc_rx_new(dp->dccps_options.dccpo_rx_ccid, + sk, GFP_KERNEL); + dp->dccps_hc_tx_ccid = + ccid_hc_tx_new(dp->dccps_options.dccpo_tx_ccid, + sk, GFP_KERNEL); + if (unlikely(dp->dccps_hc_rx_ccid == NULL || + dp->dccps_hc_tx_ccid == NULL)) { + ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk); + ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); if (dp->dccps_options.dccpo_send_ack_vector) { dccp_ackvec_free(dp->dccps_hc_rx_ackvec); dp->dccps_hc_rx_ackvec = NULL; @@ -1232,21 +1052,30 @@ static int dccp_v4_init_sock(struct sock *sk) dp->dccps_hc_rx_ccid = dp->dccps_hc_tx_ccid = NULL; return -ENOMEM; } - } else + } else { + /* control socket doesn't need feat nego */ + INIT_LIST_HEAD(&dp->dccps_options.dccpo_pending); + INIT_LIST_HEAD(&dp->dccps_options.dccpo_conf); dccp_ctl_socket_init = 0; + } dccp_init_xmit_timers(sk); - inet_csk(sk)->icsk_rto = DCCP_TIMEOUT_INIT; + icsk->icsk_rto = DCCP_TIMEOUT_INIT; sk->sk_state = DCCP_CLOSED; sk->sk_write_space = dccp_write_space; + icsk->icsk_af_ops = &dccp_ipv4_af_ops; + icsk->icsk_sync_mss = dccp_sync_mss; dp->dccps_mss_cache = 536; dp->dccps_role = DCCP_ROLE_UNDEFINED; dp->dccps_service = DCCP_SERVICE_INVALID_VALUE; + dp->dccps_l_ack_ratio = dp->dccps_r_ack_ratio = 1; return 0; } -static int dccp_v4_destroy_sock(struct sock *sk) +EXPORT_SYMBOL_GPL(dccp_v4_init_sock); + +int dccp_v4_destroy_sock(struct sock *sk) { struct dccp_sock *dp = dccp_sk(sk); @@ -1266,19 +1095,22 @@ static int dccp_v4_destroy_sock(struct sock *sk) kfree(dp->dccps_service_list); dp->dccps_service_list = NULL; - ccid_hc_rx_exit(dp->dccps_hc_rx_ccid, sk); - ccid_hc_tx_exit(dp->dccps_hc_tx_ccid, sk); if (dp->dccps_options.dccpo_send_ack_vector) { dccp_ackvec_free(dp->dccps_hc_rx_ackvec); dp->dccps_hc_rx_ackvec = NULL; } - ccid_exit(dp->dccps_hc_rx_ccid, sk); - ccid_exit(dp->dccps_hc_tx_ccid, sk); + ccid_hc_rx_delete(dp->dccps_hc_rx_ccid, sk); + ccid_hc_tx_delete(dp->dccps_hc_tx_ccid, sk); dp->dccps_hc_rx_ccid = dp->dccps_hc_tx_ccid = NULL; + /* clean up feature negotiation state */ + dccp_feat_clean(sk); + return 0; } +EXPORT_SYMBOL_GPL(dccp_v4_destroy_sock); + static void dccp_v4_reqsk_destructor(struct request_sock *req) { kfree(inet_rsk(req)->opt); @@ -1293,7 +1125,11 @@ static struct request_sock_ops dccp_request_sock_ops = { .send_reset = dccp_v4_ctl_send_reset, }; -struct proto dccp_v4_prot = { +static struct timewait_sock_ops dccp_timewait_sock_ops = { + .twsk_obj_size = sizeof(struct inet_timewait_sock), +}; + +struct proto dccp_prot = { .name = "DCCP", .owner = THIS_MODULE, .close = dccp_close, @@ -1307,7 +1143,7 @@ struct proto dccp_v4_prot = { .recvmsg = dccp_recvmsg, .backlog_rcv = dccp_v4_do_rcv, .hash = dccp_v4_hash, - .unhash = dccp_v4_unhash, + .unhash = dccp_unhash, .accept = inet_csk_accept, .get_port = dccp_v4_get_port, .shutdown = dccp_shutdown, @@ -1316,5 +1152,7 @@ struct proto dccp_v4_prot = { .max_header = MAX_DCCP_HEADER, .obj_size = sizeof(struct dccp_sock), .rsk_prot = &dccp_request_sock_ops, - .twsk_obj_size = sizeof(struct inet_timewait_sock), + .twsk_prot = &dccp_timewait_sock_ops, }; + +EXPORT_SYMBOL_GPL(dccp_prot);