static void dccp_v4_reqsk_send_ack(struct sk_buff *rxskb,
struct request_sock *req)
{
int err;
struct dccp_hdr *rxdh = dccp_hdr(rxskb), *dh;
const u32 dccp_hdr_ack_len = sizeof(struct dccp_hdr) +
sizeof(struct dccp_hdr_ext) +
sizeof(struct dccp_hdr_ack_bits);
struct sk_buff *skb;
if (((struct rtable *)rxskb->dst)->rt_type != RTN_LOCAL)
return;
skb = alloc_skb(dccp_v4_ctl_socket->sk->sk_prot->max_header, GFP_ATOMIC);
if (skb == NULL)
return;
/* Reserve space for headers. */
skb_reserve(skb, dccp_v4_ctl_socket->sk->sk_prot->max_header);
skb->dst = dst_clone(rxskb->dst);
skb->h.raw = skb_push(skb, dccp_hdr_ack_len);
dh = dccp_hdr(skb);
memset(dh, 0, dccp_hdr_ack_len);
/* Build DCCP header and checksum it. */
dh->dccph_type = DCCP_PKT_ACK;
dh->dccph_sport = rxdh->dccph_dport;
dh->dccph_dport = rxdh->dccph_sport;
dh->dccph_doff = dccp_hdr_ack_len / 4;
dh->dccph_x = 1;
dccp_hdr_set_seq(dh, DCCP_SKB_CB(rxskb)->dccpd_ack_seq);
dccp_hdr_set_ack(dccp_hdr_ack_bits(skb),
DCCP_SKB_CB(rxskb)->dccpd_seq);
bh_lock_sock(dccp_v4_ctl_socket->sk);
err = ip_build_and_send_pkt(skb, dccp_v4_ctl_socket->sk,
rxskb->nh.iph->daddr,
rxskb->nh.iph->saddr, NULL);
bh_unlock_sock(dccp_v4_ctl_socket->sk);
if (err == NET_XMIT_CN || err == 0) {
DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS);
}
}
static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
{
struct ipv6hdr *rxip6h;
struct sk_buff *skb;
struct flowi fl;
struct net *net = dev_net(skb_dst(rxskb)->dev);
struct sock *ctl_sk = net->dccp.v6_ctl_sk;
struct dst_entry *dst;
if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
return;
if (!ipv6_unicast_destination(rxskb))
return;
skb = dccp_ctl_make_reset(ctl_sk, rxskb);
if (skb == NULL)
return;
rxip6h = ipv6_hdr(rxskb);
dccp_hdr(skb)->dccph_checksum = dccp_v6_csum_finish(skb, &rxip6h->saddr,
&rxip6h->daddr);
memset(&fl, 0, sizeof(fl));
ipv6_addr_copy(&fl.fl6_dst, &rxip6h->saddr);
ipv6_addr_copy(&fl.fl6_src, &rxip6h->daddr);
fl.proto = IPPROTO_DCCP;
fl.oif = inet6_iif(rxskb);
fl.fl_ip_dport = dccp_hdr(skb)->dccph_dport;
fl.fl_ip_sport = dccp_hdr(skb)->dccph_sport;
security_skb_classify_flow(rxskb, &fl);
/* sk = NULL, but it is safe for now. RST socket required. */
if (!ip6_dst_lookup(ctl_sk, &dst, &fl)) {
if (xfrm_lookup(net, &dst, &fl, NULL, 0) >= 0) {
skb_dst_set(skb, dst);
ip6_xmit(ctl_sk, skb, &fl, NULL, 0);
DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS);
return;
}
}
kfree_skb(skb);
}
static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
{
const struct ipv6hdr *rxip6h;
struct sk_buff *skb;
struct flowi6 fl6;
struct net *net = dev_net(skb_dst(rxskb)->dev);
struct sock *ctl_sk = net->dccp.v6_ctl_sk;
struct dst_entry *dst;
if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
return;
if (!ipv6_unicast_destination(rxskb))
return;
skb = dccp_ctl_make_reset(ctl_sk, rxskb);
if (skb == NULL)
return;
rxip6h = ipv6_hdr(rxskb);
dccp_hdr(skb)->dccph_checksum = dccp_v6_csum_finish(skb, &rxip6h->saddr,
&rxip6h->daddr);
memset(&fl6, 0, sizeof(fl6));
fl6.daddr = rxip6h->saddr;
fl6.saddr = rxip6h->daddr;
fl6.flowi6_proto = IPPROTO_DCCP;
fl6.flowi6_oif = inet6_iif(rxskb);
fl6.fl6_dport = dccp_hdr(skb)->dccph_dport;
fl6.fl6_sport = dccp_hdr(skb)->dccph_sport;
security_skb_classify_flow(rxskb, flowi6_to_flowi(&fl6));
/* sk = NULL, but it is safe for now. RST socket required. */
dst = ip6_dst_lookup_flow(ctl_sk, &fl6, NULL, false);
if (!IS_ERR(dst)) {
skb_dst_set(skb, dst);
ip6_xmit(ctl_sk, skb, &fl6, NULL, 0);
DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS);
return;
}
kfree_skb(skb);
}
static void dccp_v4_ctl_send_reset(const struct sock *sk, struct sk_buff *rxskb)
{
int err;
const struct iphdr *rxiph;
struct sk_buff *skb;
struct dst_entry *dst;
struct net *net = dev_net(skb_dst(rxskb)->dev);
struct sock *ctl_sk = net->dccp.v4_ctl_sk;
/* Never send a reset in response to a reset. */
if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
return;
if (skb_rtable(rxskb)->rt_type != RTN_LOCAL)
return;
dst = dccp_v4_route_skb(net, ctl_sk, rxskb);
if (dst == NULL)
return;
skb = dccp_ctl_make_reset(ctl_sk, rxskb);
if (skb == NULL)
goto out;
rxiph = ip_hdr(rxskb);
dccp_hdr(skb)->dccph_checksum = dccp_v4_csum_finish(skb, rxiph->saddr,
rxiph->daddr);
skb_dst_set(skb, dst_clone(dst));
bh_lock_sock(ctl_sk);
err = ip_build_and_send_pkt(skb, ctl_sk,
rxiph->daddr, rxiph->saddr, NULL);
bh_unlock_sock(ctl_sk);
if (net_xmit_eval(err) == 0) {
DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS);
}
out:
dst_release(dst);
}
int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
{
struct inet_request_sock *ireq;
struct request_sock *req;
struct dccp_request_sock *dreq;
const __be32 service = dccp_hdr_request(skb)->dccph_req_service;
struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
/* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */
if (skb_rtable(skb)->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
return 0; /* discard, don't send a reset here */
if (dccp_bad_service_code(sk, service)) {
dcb->dccpd_reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE;
goto drop;
}
/*
* TW buckets are converted to open requests without
* limitations, they conserve resources and peer is
* evidently real one.
*/
dcb->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY;
if (inet_csk_reqsk_queue_is_full(sk))
goto drop;
/*
* Accept backlog is full. If we have already queued enough
* of warm entries in syn queue, drop request. It is better than
* clogging syn queue with openreqs with exponentially increasing
* timeout.
*/
if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1)
goto drop;
req = inet_reqsk_alloc(&dccp_request_sock_ops, sk, true);
if (req == NULL)
goto drop;
if (dccp_reqsk_init(req, dccp_sk(sk), skb))
goto drop_and_free;
dreq = dccp_rsk(req);
if (dccp_parse_options(sk, dreq, skb))
goto drop_and_free;
if (security_inet_conn_request(sk, skb, req))
goto drop_and_free;
ireq = inet_rsk(req);
sk_rcv_saddr_set(req_to_sk(req), ip_hdr(skb)->daddr);
sk_daddr_set(req_to_sk(req), ip_hdr(skb)->saddr);
ireq->ireq_family = AF_INET;
ireq->ir_iif = sk->sk_bound_dev_if;
/*
* Step 3: Process LISTEN state
*
* Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookie
*
* Setting S.SWL/S.SWH to is deferred to dccp_create_openreq_child().
*/
dreq->dreq_isr = dcb->dccpd_seq;
dreq->dreq_gsr = dreq->dreq_isr;
dreq->dreq_iss = dccp_v4_init_sequence(skb);
dreq->dreq_gss = dreq->dreq_iss;
dreq->dreq_service = service;
if (dccp_v4_send_response(sk, req))
goto drop_and_free;
inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
return 0;
drop_and_free:
reqsk_free(req);
drop:
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
return -1;
}
/*
* This routine is called by the ICMP module when it gets some sort of error
* condition. If err < 0 then the socket should be closed and the error
* returned to the user. If err > 0 it's just the icmp type << 8 | icmp code.
* After adjustment header points to the first 8 bytes of the tcp header. We
* need to find the appropriate port.
*
* The locking strategy used here is very "optimistic". When someone else
* accesses the socket the ICMP is just dropped and for some paths there is no
* check at all. A more general error queue to queue errors for later handling
* is probably better.
*/
static void dccp_v4_err(struct sk_buff *skb, u32 info)
{
const struct iphdr *iph = (struct iphdr *)skb->data;
const u8 offset = iph->ihl << 2;
const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
struct dccp_sock *dp;
struct inet_sock *inet;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct sock *sk;
__u64 seq;
int err;
struct net *net = dev_net(skb->dev);
if (skb->len < offset + sizeof(*dh) ||
skb->len < offset + __dccp_basic_hdr_len(dh)) {
ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
return;
}
sk = __inet_lookup_established(net, &dccp_hashinfo,
iph->daddr, dh->dccph_dport,
iph->saddr, ntohs(dh->dccph_sport),
inet_iif(skb));
if (!sk) {
ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
return;
}
if (sk->sk_state == DCCP_TIME_WAIT) {
inet_twsk_put(inet_twsk(sk));
return;
}
seq = dccp_hdr_seq(dh);
if (sk->sk_state == DCCP_NEW_SYN_RECV)
return dccp_req_err(sk, seq);
bh_lock_sock(sk);
/* If too many ICMPs get dropped on busy
* servers this needs to be solved differently.
*/
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == DCCP_CLOSED)
goto out;
dp = dccp_sk(sk);
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
!between48(seq, dp->dccps_awl, dp->dccps_awh)) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
switch (type) {
case ICMP_REDIRECT:
dccp_do_redirect(skb, sk);
goto out;
case ICMP_SOURCE_QUENCH:
/* Just silently ignore these. */
goto out;
case ICMP_PARAMETERPROB:
err = EPROTO;
break;
case ICMP_DEST_UNREACH:
if (code > NR_ICMP_UNREACH)
goto out;
if (code == ICMP_FRAG_NEEDED) { /* PMTU discovery (RFC1191) */
if (!sock_owned_by_user(sk))
dccp_do_pmtu_discovery(sk, iph, info);
goto out;
}
err = icmp_err_convert[code].errno;
break;
case ICMP_TIME_EXCEEDED:
err = EHOSTUNREACH;
break;
default:
goto out;
}
switch (sk->sk_state) {
case DCCP_REQUESTING:
case DCCP_RESPOND:
if (!sock_owned_by_user(sk)) {
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
sk->sk_error_report(sk);
dccp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
/* If we've already connected we will keep trying
* until we time out, or the user gives up.
*
* rfc1122 4.2.3.9 allows to consider as hard errors
* only PROTO_UNREACH and PORT_UNREACH (well, FRAG_FAILED too,
* but it is obsoleted by pmtu discovery).
*
* Note, that in modern internet, where routing is unreliable
* and in each dark corner broken firewalls sit, sending random
* errors ordered by their masters even this two messages finally lose
* their original sense (even Linux sends invalid PORT_UNREACHs)
*
* Now we are in compliance with RFCs.
* --ANK (980905)
*/
inet = inet_sk(sk);
if (!sock_owned_by_user(sk) && inet->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else /* Only an error on timeout */
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
/**
* dccp_parse_options - Parse DCCP options present in @skb
* @sk: client|server|listening dccp socket (when @dreq != NULL)
* @dreq: request socket to use during connection setup, or NULL
*/
int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
struct sk_buff *skb)
{
struct dccp_sock *dp = dccp_sk(sk);
const struct dccp_hdr *dh = dccp_hdr(skb);
const u8 pkt_type = DCCP_SKB_CB(skb)->dccpd_type;
unsigned char *options = (unsigned char *)dh + dccp_hdr_len(skb);
unsigned char *opt_ptr = options;
const unsigned char *opt_end = (unsigned char *)dh +
(dh->dccph_doff * 4);
struct dccp_options_received *opt_recv = &dp->dccps_options_received;
unsigned char opt, len;
unsigned char *uninitialized_var(value);
u32 elapsed_time;
__be32 opt_val;
int rc;
int mandatory = 0;
memset(opt_recv, 0, sizeof(*opt_recv));
opt = len = 0;
while (opt_ptr != opt_end) {
opt = *opt_ptr++;
len = 0;
value = NULL;
/* Check if this isn't a single byte option */
if (opt > DCCPO_MAX_RESERVED) {
if (opt_ptr == opt_end)
goto out_nonsensical_length;
len = *opt_ptr++;
if (len < 2)
goto out_nonsensical_length;
/*
* Remove the type and len fields, leaving
* just the value size
*/
len -= 2;
value = opt_ptr;
opt_ptr += len;
if (opt_ptr > opt_end)
goto out_nonsensical_length;
}
/*
* CCID-specific options are ignored during connection setup, as
* negotiation may still be in progress (see RFC 4340, 10.3).
* The same applies to Ack Vectors, as these depend on the CCID.
*/
if (dreq != NULL && (opt >= DCCPO_MIN_RX_CCID_SPECIFIC ||
opt == DCCPO_ACK_VECTOR_0 || opt == DCCPO_ACK_VECTOR_1))
goto ignore_option;
switch (opt) {
case DCCPO_PADDING:
break;
case DCCPO_MANDATORY:
if (mandatory)
goto out_invalid_option;
if (pkt_type != DCCP_PKT_DATA)
mandatory = 1;
break;
case DCCPO_NDP_COUNT:
if (len > 6)
goto out_invalid_option;
opt_recv->dccpor_ndp = dccp_decode_value_var(value, len);
dccp_pr_debug("%s opt: NDP count=%llu\n", dccp_role(sk),
(unsigned long long)opt_recv->dccpor_ndp);
break;
case DCCPO_CHANGE_L ... DCCPO_CONFIRM_R:
if (pkt_type == DCCP_PKT_DATA) /* RFC 4340, 6 */
break;
if (len == 0)
goto out_invalid_option;
rc = dccp_feat_parse_options(sk, dreq, mandatory, opt,
*value, value + 1, len - 1);
if (rc)
goto out_featneg_failed;
break;
case DCCPO_TIMESTAMP:
if (len != 4)
goto out_invalid_option;
/*
* RFC 4340 13.1: "The precise time corresponding to
* Timestamp Value zero is not specified". We use
* zero to indicate absence of a meaningful timestamp.
*/
opt_val = get_unaligned((__be32 *)value);
if (unlikely(opt_val == 0)) {
DCCP_WARN("Timestamp with zero value\n");
break;
}
if (dreq != NULL) {
dreq->dreq_timestamp_echo = ntohl(opt_val);
dreq->dreq_timestamp_time = dccp_timestamp();
} else {
opt_recv->dccpor_timestamp =
dp->dccps_timestamp_echo = ntohl(opt_val);
dp->dccps_timestamp_time = dccp_timestamp();
}
dccp_pr_debug("%s rx opt: TIMESTAMP=%u, ackno=%llu\n",
dccp_role(sk), ntohl(opt_val),
(unsigned long long)
DCCP_SKB_CB(skb)->dccpd_ack_seq);
/* schedule an Ack in case this sender is quiescent */
inet_csk_schedule_ack(sk);
break;
case DCCPO_TIMESTAMP_ECHO:
if (len != 4 && len != 6 && len != 8)
goto out_invalid_option;
opt_val = get_unaligned((__be32 *)value);
opt_recv->dccpor_timestamp_echo = ntohl(opt_val);
dccp_pr_debug("%s rx opt: TIMESTAMP_ECHO=%u, len=%d, "
"ackno=%llu", dccp_role(sk),
opt_recv->dccpor_timestamp_echo,
len + 2,
(unsigned long long)
DCCP_SKB_CB(skb)->dccpd_ack_seq);
value += 4;
if (len == 4) { /* no elapsed time included */
dccp_pr_debug_cat("\n");
break;
}
if (len == 6) { /* 2-byte elapsed time */
__be16 opt_val2 = get_unaligned((__be16 *)value);
elapsed_time = ntohs(opt_val2);
} else { /* 4-byte elapsed time */
opt_val = get_unaligned((__be32 *)value);
elapsed_time = ntohl(opt_val);
}
dccp_pr_debug_cat(", ELAPSED_TIME=%u\n", elapsed_time);
/* Give precedence to the biggest ELAPSED_TIME */
if (elapsed_time > opt_recv->dccpor_elapsed_time)
opt_recv->dccpor_elapsed_time = elapsed_time;
break;
case DCCPO_ELAPSED_TIME:
if (dccp_packet_without_ack(skb)) /* RFC 4340, 13.2 */
break;
if (len == 2) {
__be16 opt_val2 = get_unaligned((__be16 *)value);
elapsed_time = ntohs(opt_val2);
} else if (len == 4) {
opt_val = get_unaligned((__be32 *)value);
elapsed_time = ntohl(opt_val);
} else {
goto out_invalid_option;
}
if (elapsed_time > opt_recv->dccpor_elapsed_time)
opt_recv->dccpor_elapsed_time = elapsed_time;
dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n",
dccp_role(sk), elapsed_time);
break;
case DCCPO_MIN_RX_CCID_SPECIFIC ... DCCPO_MAX_RX_CCID_SPECIFIC:
if (ccid_hc_rx_parse_options(dp->dccps_hc_rx_ccid, sk,
pkt_type, opt, value, len))
goto out_invalid_option;
break;
case DCCPO_ACK_VECTOR_0:
case DCCPO_ACK_VECTOR_1:
if (dccp_packet_without_ack(skb)) /* RFC 4340, 11.4 */
break;
/*
* Ack vectors are processed by the TX CCID if it is
* interested. The RX CCID need not parse Ack Vectors,
* since it is only interested in clearing old state.
* Fall through.
*/
case DCCPO_MIN_TX_CCID_SPECIFIC ... DCCPO_MAX_TX_CCID_SPECIFIC:
if (ccid_hc_tx_parse_options(dp->dccps_hc_tx_ccid, sk,
pkt_type, opt, value, len))
goto out_invalid_option;
break;
default:
DCCP_CRIT("DCCP(%p): option %d(len=%d) not "
"implemented, ignoring", sk, opt, len);
break;
}
ignore_option:
if (opt != DCCPO_MANDATORY)
mandatory = 0;
}
/* mandatory was the last byte in option list -> reset connection */
if (mandatory)
goto out_invalid_option;
out_nonsensical_length:
/* RFC 4340, 5.8: ignore option and all remaining option space */
return 0;
out_invalid_option:
DCCP_INC_STATS_BH(DCCP_MIB_INVALIDOPT);
rc = DCCP_RESET_CODE_OPTION_ERROR;
out_featneg_failed:
DCCP_WARN("DCCP(%p): Option %d (len=%d) error=%u\n", sk, opt, len, rc);
DCCP_SKB_CB(skb)->dccpd_reset_code = rc;
DCCP_SKB_CB(skb)->dccpd_reset_data[0] = opt;
DCCP_SKB_CB(skb)->dccpd_reset_data[1] = len > 0 ? value[0] : 0;
DCCP_SKB_CB(skb)->dccpd_reset_data[2] = len > 1 ? value[1] : 0;
return -1;
}
/*
* This routine is called by the ICMP module when it gets some sort of error
* condition. If err < 0 then the socket should be closed and the error
* returned to the user. If err > 0 it's just the icmp type << 8 | icmp code.
* After adjustment header points to the first 8 bytes of the tcp header. We
* need to find the appropriate port.
*
* The locking strategy used here is very "optimistic". When someone else
* accesses the socket the ICMP is just dropped and for some paths there is no
* check at all. A more general error queue to queue errors for later handling
* is probably better.
*/
static void dccp_v4_err(struct sk_buff *skb, u32 info)
{
const struct iphdr *iph = (struct iphdr *)skb->data;
const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data +
(iph->ihl << 2));
struct dccp_sock *dp;
struct inet_sock *inet;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct sock *sk;
__u64 seq;
int err;
if (skb->len < (iph->ihl << 2) + 8) {
ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
return;
}
sk = inet_lookup(&dccp_hashinfo, iph->daddr, dh->dccph_dport,
iph->saddr, dh->dccph_sport, inet_iif(skb));
if (sk == NULL) {
ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
return;
}
if (sk->sk_state == DCCP_TIME_WAIT) {
inet_twsk_put(inet_twsk(sk));
return;
}
bh_lock_sock(sk);
/* If too many ICMPs get dropped on busy
* servers this needs to be solved differently.
*/
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == DCCP_CLOSED)
goto out;
dp = dccp_sk(sk);
seq = dccp_hdr_seq(skb);
if (sk->sk_state != DCCP_LISTEN &&
!between48(seq, dp->dccps_swl, dp->dccps_swh)) {
NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
switch (type) {
case ICMP_SOURCE_QUENCH:
/* Just silently ignore these. */
goto out;
case ICMP_PARAMETERPROB:
err = EPROTO;
break;
case ICMP_DEST_UNREACH:
if (code > NR_ICMP_UNREACH)
goto out;
if (code == ICMP_FRAG_NEEDED) { /* PMTU discovery (RFC1191) */
if (!sock_owned_by_user(sk))
dccp_do_pmtu_discovery(sk, iph, info);
goto out;
}
err = icmp_err_convert[code].errno;
break;
case ICMP_TIME_EXCEEDED:
err = EHOSTUNREACH;
break;
default:
goto out;
}
switch (sk->sk_state) {
struct request_sock *req , **prev;
case DCCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet_csk_search_req(sk, &prev, dh->dccph_dport,
iph->daddr, iph->saddr);
if (!req)
goto out;
/*
* ICMPs are not backlogged, hence we cannot get an established
* socket here.
*/
WARN_ON(req->sk);
if (seq != dccp_rsk(req)->dreq_iss) {
NET_INC_STATS_BH(LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
/*
* Still in RESPOND, just remove it silently.
* There is no good way to pass the error to the newly
* created socket, and POSIX does not want network
* errors returned from accept().
*/
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case DCCP_REQUESTING:
case DCCP_RESPOND:
if (!sock_owned_by_user(sk)) {
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
sk->sk_error_report(sk);
dccp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
/* If we've already connected we will keep trying
* until we time out, or the user gives up.
*
* rfc1122 4.2.3.9 allows to consider as hard errors
* only PROTO_UNREACH and PORT_UNREACH (well, FRAG_FAILED too,
* but it is obsoleted by pmtu discovery).
*
* Note, that in modern internet, where routing is unreliable
* and in each dark corner broken firewalls sit, sending random
* errors ordered by their masters even this two messages finally lose
* their original sense (even Linux sends invalid PORT_UNREACHs)
*
* Now we are in compliance with RFCs.
* --ANK (980905)
*/
inet = inet_sk(sk);
if (!sock_owned_by_user(sk) && inet->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else /* Only an error on timeout */
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
u8 type, u8 code, int offset, __be32 info)
{
const struct ipv6hdr *hdr = (const struct ipv6hdr *)skb->data;
const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
struct dccp_sock *dp;
struct ipv6_pinfo *np;
struct sock *sk;
int err;
__u64 seq;
struct net *net = dev_net(skb->dev);
if (skb->len < offset + sizeof(*dh) ||
skb->len < offset + __dccp_basic_hdr_len(dh)) {
ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev),
ICMP6_MIB_INERRORS);
return;
}
sk = inet6_lookup(net, &dccp_hashinfo,
&hdr->daddr, dh->dccph_dport,
&hdr->saddr, dh->dccph_sport, inet6_iif(skb));
if (sk == NULL) {
ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev),
ICMP6_MIB_INERRORS);
return;
}
if (sk->sk_state == DCCP_TIME_WAIT) {
inet_twsk_put(inet_twsk(sk));
return;
}
bh_lock_sock(sk);
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == DCCP_CLOSED)
goto out;
dp = dccp_sk(sk);
seq = dccp_hdr_seq(dh);
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
!between48(seq, dp->dccps_awl, dp->dccps_awh)) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
np = inet6_sk(sk);
if (type == NDISC_REDIRECT) {
struct dst_entry *dst = __sk_dst_check(sk, np->dst_cookie);
if (dst)
dst->ops->redirect(dst, sk, skb);
goto out;
}
if (type == ICMPV6_PKT_TOOBIG) {
struct dst_entry *dst = NULL;
if (!ip6_sk_accept_pmtu(sk))
goto out;
if (sock_owned_by_user(sk))
goto out;
if ((1 << sk->sk_state) & (DCCPF_LISTEN | DCCPF_CLOSED))
goto out;
dst = inet6_csk_update_pmtu(sk, ntohl(info));
if (!dst)
goto out;
if (inet_csk(sk)->icsk_pmtu_cookie > dst_mtu(dst))
dccp_sync_mss(sk, dst_mtu(dst));
goto out;
}
icmpv6_err_convert(type, code, &err);
/* Might be for an request_sock */
switch (sk->sk_state) {
struct request_sock *req, **prev;
case DCCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet6_csk_search_req(sk, &prev, dh->dccph_dport,
&hdr->daddr, &hdr->saddr,
inet6_iif(skb));
if (req == NULL)
goto out;
/*
* ICMPs are not backlogged, hence we cannot get an established
* socket here.
*/
WARN_ON(req->sk != NULL);
if (!between48(seq, dccp_rsk(req)->dreq_iss,
dccp_rsk(req)->dreq_gss)) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case DCCP_REQUESTING:
case DCCP_RESPOND: /* Cannot happen.
It can, it SYNs are crossed. --ANK */
if (!sock_owned_by_user(sk)) {
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
/*
* Wake people up to see the error
* (see connect in sock.c)
*/
sk->sk_error_report(sk);
dccp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
if (!sock_owned_by_user(sk) && np->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
{
struct request_sock *req;
struct dccp_request_sock *dreq;
struct inet_request_sock *ireq;
struct ipv6_pinfo *np = inet6_sk(sk);
const __be32 service = dccp_hdr_request(skb)->dccph_req_service;
struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
if (skb->protocol == htons(ETH_P_IP))
return dccp_v4_conn_request(sk, skb);
if (!ipv6_unicast_destination(skb))
return 0; /* discard, don't send a reset here */
if (dccp_bad_service_code(sk, service)) {
dcb->dccpd_reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE;
goto drop;
}
/*
* There are no SYN attacks on IPv6, yet...
*/
dcb->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY;
if (inet_csk_reqsk_queue_is_full(sk))
goto drop;
if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1)
goto drop;
req = inet6_reqsk_alloc(&dccp6_request_sock_ops);
if (req == NULL)
goto drop;
if (dccp_reqsk_init(req, dccp_sk(sk), skb))
goto drop_and_free;
dreq = dccp_rsk(req);
if (dccp_parse_options(sk, dreq, skb))
goto drop_and_free;
if (security_inet_conn_request(sk, skb, req))
goto drop_and_free;
ireq = inet_rsk(req);
ireq->ir_v6_rmt_addr = ipv6_hdr(skb)->saddr;
ireq->ir_v6_loc_addr = ipv6_hdr(skb)->daddr;
if (ipv6_opt_accepted(sk, skb) ||
np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo ||
np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) {
atomic_inc(&skb->users);
ireq->pktopts = skb;
}
ireq->ir_iif = sk->sk_bound_dev_if;
/* So that link locals have meaning */
if (!sk->sk_bound_dev_if &&
ipv6_addr_type(&ireq->ir_v6_rmt_addr) & IPV6_ADDR_LINKLOCAL)
ireq->ir_iif = inet6_iif(skb);
/*
* Step 3: Process LISTEN state
*
* Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookie
*
* Setting S.SWL/S.SWH to is deferred to dccp_create_openreq_child().
*/
dreq->dreq_isr = dcb->dccpd_seq;
dreq->dreq_gsr = dreq->dreq_isr;
dreq->dreq_iss = dccp_v6_init_sequence(skb);
dreq->dreq_gss = dreq->dreq_iss;
dreq->dreq_service = service;
if (dccp_v6_send_response(sk, req))
goto drop_and_free;
inet6_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT);
return 0;
drop_and_free:
reqsk_free(req);
drop:
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
return -1;
}
int dccp_parse_options(struct sock *sk, struct sk_buff *skb)
{
struct dccp_sock *dp = dccp_sk(sk);
#ifdef CONFIG_IP_DCCP_DEBUG
const char *debug_prefix = dp->dccps_role == DCCP_ROLE_CLIENT ?
"CLIENT rx opt: " : "server rx opt: ";
#endif
const struct dccp_hdr *dh = dccp_hdr(skb);
const u8 pkt_type = DCCP_SKB_CB(skb)->dccpd_type;
unsigned char *options = (unsigned char *)dh + dccp_hdr_len(skb);
unsigned char *opt_ptr = options;
const unsigned char *opt_end = (unsigned char *)dh +
(dh->dccph_doff * 4);
struct dccp_options_received *opt_recv = &dp->dccps_options_received;
unsigned char opt, len;
unsigned char *value;
u32 elapsed_time;
memset(opt_recv, 0, sizeof(*opt_recv));
while (opt_ptr != opt_end) {
opt = *opt_ptr++;
len = 0;
value = NULL;
/* Check if this isn't a single byte option */
if (opt > DCCPO_MAX_RESERVED) {
if (opt_ptr == opt_end)
goto out_invalid_option;
len = *opt_ptr++;
if (len < 3)
goto out_invalid_option;
/*
* Remove the type and len fields, leaving
* just the value size
*/
len -= 2;
value = opt_ptr;
opt_ptr += len;
if (opt_ptr > opt_end)
goto out_invalid_option;
}
switch (opt) {
case DCCPO_PADDING:
break;
case DCCPO_NDP_COUNT:
if (len > 3)
goto out_invalid_option;
opt_recv->dccpor_ndp = dccp_decode_value_var(value, len);
dccp_pr_debug("%sNDP count=%d\n", debug_prefix,
opt_recv->dccpor_ndp);
break;
case DCCPO_ACK_VECTOR_0:
case DCCPO_ACK_VECTOR_1:
if (pkt_type == DCCP_PKT_DATA)
continue;
if (dp->dccps_options.dccpo_send_ack_vector &&
dccp_ackvec_parse(sk, skb, opt, value, len))
goto out_invalid_option;
break;
case DCCPO_TIMESTAMP:
if (len != 4)
goto out_invalid_option;
opt_recv->dccpor_timestamp = ntohl(*(u32 *)value);
dp->dccps_timestamp_echo = opt_recv->dccpor_timestamp;
dccp_timestamp(sk, &dp->dccps_timestamp_time);
dccp_pr_debug("%sTIMESTAMP=%u, ackno=%llu\n",
debug_prefix, opt_recv->dccpor_timestamp,
(unsigned long long)
DCCP_SKB_CB(skb)->dccpd_ack_seq);
break;
case DCCPO_TIMESTAMP_ECHO:
if (len != 4 && len != 6 && len != 8)
goto out_invalid_option;
opt_recv->dccpor_timestamp_echo = ntohl(*(u32 *)value);
dccp_pr_debug("%sTIMESTAMP_ECHO=%u, len=%d, ackno=%llu, ",
debug_prefix,
opt_recv->dccpor_timestamp_echo,
len + 2,
(unsigned long long)
DCCP_SKB_CB(skb)->dccpd_ack_seq);
if (len == 4)
break;
if (len == 6)
elapsed_time = ntohs(*(u16 *)(value + 4));
else
elapsed_time = ntohl(*(u32 *)(value + 4));
/* Give precedence to the biggest ELAPSED_TIME */
if (elapsed_time > opt_recv->dccpor_elapsed_time)
opt_recv->dccpor_elapsed_time = elapsed_time;
break;
case DCCPO_ELAPSED_TIME:
if (len != 2 && len != 4)
goto out_invalid_option;
if (pkt_type == DCCP_PKT_DATA)
continue;
if (len == 2)
elapsed_time = ntohs(*(u16 *)value);
else
elapsed_time = ntohl(*(u32 *)value);
if (elapsed_time > opt_recv->dccpor_elapsed_time)
opt_recv->dccpor_elapsed_time = elapsed_time;
dccp_pr_debug("%sELAPSED_TIME=%d\n", debug_prefix,
elapsed_time);
break;
/*
* From draft-ietf-dccp-spec-11.txt:
*
* Option numbers 128 through 191 are for
* options sent from the HC-Sender to the
* HC-Receiver; option numbers 192 through 255
* are for options sent from the HC-Receiver to
* the HC-Sender.
*/
case 128 ... 191: {
const u16 idx = value - options;
if (ccid_hc_rx_parse_options(dp->dccps_hc_rx_ccid, sk,
opt, len, idx,
value) != 0)
goto out_invalid_option;
}
break;
case 192 ... 255: {
const u16 idx = value - options;
if (ccid_hc_tx_parse_options(dp->dccps_hc_tx_ccid, sk,
opt, len, idx,
value) != 0)
goto out_invalid_option;
}
break;
default:
pr_info("DCCP(%p): option %d(len=%d) not "
"implemented, ignoring\n",
sk, opt, len);
break;
}
}
return 0;
out_invalid_option:
DCCP_INC_STATS_BH(DCCP_MIB_INVALIDOPT);
DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_OPTION_ERROR;
pr_info("DCCP(%p): invalid option %d, len=%d\n", sk, opt, len);
return -1;
}
static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
u8 type, u8 code, int offset, __be32 info)
{
struct ipv6hdr *hdr = (struct ipv6hdr *)skb->data;
const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
struct dccp_sock *dp;
struct ipv6_pinfo *np;
struct sock *sk;
int err;
__u64 seq;
struct net *net = dev_net(skb->dev);
if (skb->len < offset + sizeof(*dh) ||
skb->len < offset + __dccp_basic_hdr_len(dh)) {
ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev),
ICMP6_MIB_INERRORS);
return;
}
sk = inet6_lookup(net, &dccp_hashinfo,
&hdr->daddr, dh->dccph_dport,
&hdr->saddr, dh->dccph_sport, inet6_iif(skb));
if (sk == NULL) {
ICMP6_INC_STATS_BH(net, __in6_dev_get(skb->dev),
ICMP6_MIB_INERRORS);
return;
}
if (sk->sk_state == DCCP_TIME_WAIT) {
inet_twsk_put(inet_twsk(sk));
return;
}
bh_lock_sock(sk);
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == DCCP_CLOSED)
goto out;
dp = dccp_sk(sk);
seq = dccp_hdr_seq(dh);
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
!between48(seq, dp->dccps_awl, dp->dccps_awh)) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
np = inet6_sk(sk);
if (type == ICMPV6_PKT_TOOBIG) {
struct dst_entry *dst = NULL;
if (sock_owned_by_user(sk))
goto out;
if ((1 << sk->sk_state) & (DCCPF_LISTEN | DCCPF_CLOSED))
goto out;
/* icmp should have updated the destination cache entry */
dst = __sk_dst_check(sk, np->dst_cookie);
if (dst == NULL) {
struct inet_sock *inet = inet_sk(sk);
struct flowi fl;
/* BUGGG_FUTURE: Again, it is not clear how
to handle rthdr case. Ignore this complexity
for now.
*/
memset(&fl, 0, sizeof(fl));
fl.proto = IPPROTO_DCCP;
ipv6_addr_copy(&fl.fl6_dst, &np->daddr);
ipv6_addr_copy(&fl.fl6_src, &np->saddr);
fl.oif = sk->sk_bound_dev_if;
fl.fl_ip_dport = inet->dport;
fl.fl_ip_sport = inet->sport;
security_sk_classify_flow(sk, &fl);
err = ip6_dst_lookup(sk, &dst, &fl);
if (err) {
sk->sk_err_soft = -err;
goto out;
}
err = xfrm_lookup(net, &dst, &fl, sk, 0);
if (err < 0) {
sk->sk_err_soft = -err;
goto out;
}
} else
dst_hold(dst);
if (inet_csk(sk)->icsk_pmtu_cookie > dst_mtu(dst)) {
dccp_sync_mss(sk, dst_mtu(dst));
} /* else let the usual retransmit timer handle it */
dst_release(dst);
goto out;
}
icmpv6_err_convert(type, code, &err);
/* Might be for an request_sock */
switch (sk->sk_state) {
struct request_sock *req, **prev;
case DCCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet6_csk_search_req(sk, &prev, dh->dccph_dport,
&hdr->daddr, &hdr->saddr,
inet6_iif(skb));
if (req == NULL)
goto out;
/*
* ICMPs are not backlogged, hence we cannot get an established
* socket here.
*/
WARN_ON(req->sk != NULL);
if (seq != dccp_rsk(req)->dreq_iss) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case DCCP_REQUESTING:
case DCCP_RESPOND: /* Cannot happen.
It can, it SYNs are crossed. --ANK */
if (!sock_owned_by_user(sk)) {
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
/*
* Wake people up to see the error
* (see connect in sock.c)
*/
sk->sk_error_report(sk);
dccp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
if (!sock_owned_by_user(sk) && np->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
static int __dccp_rcv_established(struct sock *sk, struct sk_buff *skb,
const struct dccp_hdr *dh, const unsigned len)
{
struct dccp_sock *dp = dccp_sk(sk);
switch (dccp_hdr(skb)->dccph_type) {
case DCCP_PKT_DATAACK:
case DCCP_PKT_DATA:
/*
* FIXME: check if sk_receive_queue is full, schedule DATA_DROPPED
* option if it is.
*/
__skb_pull(skb, dh->dccph_doff * 4);
__skb_queue_tail(&sk->sk_receive_queue, skb);
skb_set_owner_r(skb, sk);
sk->sk_data_ready(sk, 0);
return 0;
case DCCP_PKT_ACK:
goto discard;
case DCCP_PKT_RESET:
/*
* Step 9: Process Reset
* If P.type == Reset,
* Tear down connection
* S.state := TIMEWAIT
* Set TIMEWAIT timer
* Drop packet and return
*/
dccp_fin(sk, skb);
dccp_time_wait(sk, DCCP_TIME_WAIT, 0);
return 0;
case DCCP_PKT_CLOSEREQ:
dccp_rcv_closereq(sk, skb);
goto discard;
case DCCP_PKT_CLOSE:
dccp_rcv_close(sk, skb);
return 0;
case DCCP_PKT_REQUEST:
/* Step 7
* or (S.is_server and P.type == Response)
* or (S.is_client and P.type == Request)
* or (S.state >= OPEN and P.type == Request
* and P.seqno >= S.OSR)
* or (S.state >= OPEN and P.type == Response
* and P.seqno >= S.OSR)
* or (S.state == RESPOND and P.type == Data),
* Send Sync packet acknowledging P.seqno
* Drop packet and return
*/
if (dp->dccps_role != DCCP_ROLE_LISTEN)
goto send_sync;
goto check_seq;
case DCCP_PKT_RESPONSE:
if (dp->dccps_role != DCCP_ROLE_CLIENT)
goto send_sync;
check_seq:
if (!before48(DCCP_SKB_CB(skb)->dccpd_seq, dp->dccps_osr)) {
send_sync:
dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq,
DCCP_PKT_SYNC);
}
break;
case DCCP_PKT_SYNC:
dccp_send_sync(sk, DCCP_SKB_CB(skb)->dccpd_seq,
DCCP_PKT_SYNCACK);
/*
* From RFC 4340, sec. 5.7
*
* As with DCCP-Ack packets, DCCP-Sync and DCCP-SyncAck packets
* MAY have non-zero-length application data areas, whose
* contents receivers MUST ignore.
*/
goto discard;
}
DCCP_INC_STATS_BH(DCCP_MIB_INERRS);
discard:
__kfree_skb(skb);
return 0;
}
static void dccp_v4_err(struct sk_buff *skb, u32 info)
{
const struct iphdr *iph = (struct iphdr *)skb->data;
const u8 offset = iph->ihl << 2;
const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
struct dccp_sock *dp;
struct inet_sock *inet;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct sock *sk;
__u64 seq;
int err;
struct net *net = dev_net(skb->dev);
if (skb->len < offset + sizeof(*dh) ||
skb->len < offset + __dccp_basic_hdr_len(dh)) {
ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
return;
}
sk = inet_lookup(net, &dccp_hashinfo,
iph->daddr, dh->dccph_dport,
iph->saddr, dh->dccph_sport, inet_iif(skb));
if (sk == NULL) {
ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
return;
}
if (sk->sk_state == DCCP_TIME_WAIT) {
inet_twsk_put(inet_twsk(sk));
return;
}
bh_lock_sock(sk);
if (sock_owned_by_user(sk))
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
if (sk->sk_state == DCCP_CLOSED)
goto out;
dp = dccp_sk(sk);
seq = dccp_hdr_seq(dh);
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
!between48(seq, dp->dccps_awl, dp->dccps_awh)) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
switch (type) {
case ICMP_SOURCE_QUENCH:
goto out;
case ICMP_PARAMETERPROB:
err = EPROTO;
break;
case ICMP_DEST_UNREACH:
if (code > NR_ICMP_UNREACH)
goto out;
if (code == ICMP_FRAG_NEEDED) {
if (!sock_owned_by_user(sk))
dccp_do_pmtu_discovery(sk, iph, info);
goto out;
}
err = icmp_err_convert[code].errno;
break;
case ICMP_TIME_EXCEEDED:
err = EHOSTUNREACH;
break;
default:
goto out;
}
switch (sk->sk_state) {
struct request_sock *req , **prev;
case DCCP_LISTEN:
if (sock_owned_by_user(sk))
goto out;
req = inet_csk_search_req(sk, &prev, dh->dccph_dport,
iph->daddr, iph->saddr);
if (!req)
goto out;
WARN_ON(req->sk);
if (seq != dccp_rsk(req)->dreq_iss) {
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
goto out;
}
inet_csk_reqsk_queue_drop(sk, req, prev);
goto out;
case DCCP_REQUESTING:
case DCCP_RESPOND:
if (!sock_owned_by_user(sk)) {
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
sk->sk_err = err;
sk->sk_error_report(sk);
dccp_done(sk);
} else
sk->sk_err_soft = err;
goto out;
}
inet = inet_sk(sk);
if (!sock_owned_by_user(sk) && inet->recverr) {
sk->sk_err = err;
sk->sk_error_report(sk);
} else
sk->sk_err_soft = err;
out:
bh_unlock_sock(sk);
sock_put(sk);
}
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 __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;
/* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */
if (((struct rtable *)skb->dst)->rt_flags &
(RTCF_BROADCAST | RTCF_MULTICAST)) {
reset_code = DCCP_RESET_CODE_NO_CONNECTION;
goto drop;
}
if (dccp_bad_service_code(sk, service)) {
reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE;
goto drop;
}
/*
* TW buckets are converted to open requests without
* limitations, they conserve resources and peer is
* evidently real one.
*/
if (inet_csk_reqsk_queue_is_full(sk))
goto drop;
/*
* Accept backlog is full. If we have already queued enough
* of warm entries in syn queue, drop request. It is better than
* clogging syn queue with openreqs with exponentially increasing
* timeout.
*/
if (sk_acceptq_is_full(sk) && inet_csk_reqsk_queue_young(sk) > 1)
goto drop;
req = reqsk_alloc(sk->sk_prot->rsk_prot);
if (req == NULL)
goto drop;
if (dccp_parse_options(sk, skb))
goto drop_and_free;
dccp_openreq_init(req, &dp, skb);
ireq = inet_rsk(req);
ireq->loc_addr = daddr;
ireq->rmt_addr = saddr;
req->rcv_wnd = dccp_feat_default_sequence_window;
ireq->opt = NULL;
/*
* Step 3: Process LISTEN state
*
* Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookie
*
* In fact we defer setting S.GSR, S.SWL, S.SWH to
* dccp_create_openreq_child.
*/
dreq = dccp_rsk(req);
dreq->dreq_isr = dcb->dccpd_seq;
dreq->dreq_iss = dccp_v4_init_sequence(sk, skb);
dreq->dreq_service = service;
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:
reqsk_free(req);
drop:
DCCP_INC_STATS_BH(DCCP_MIB_ATTEMPTFAILS);
dcb->dccpd_reset_code = reset_code;
return -1;
}
int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
struct sk_buff *skb)
{
struct dccp_sock *dp = dccp_sk(sk);
const struct dccp_hdr *dh = dccp_hdr(skb);
const u8 pkt_type = DCCP_SKB_CB(skb)->dccpd_type;
unsigned char *options = (unsigned char *)dh + dccp_hdr_len(skb);
unsigned char *opt_ptr = options;
const unsigned char *opt_end = (unsigned char *)dh +
(dh->dccph_doff * 4);
struct dccp_options_received *opt_recv = &dp->dccps_options_received;
unsigned char opt, len;
unsigned char *uninitialized_var(value);
u32 elapsed_time;
__be32 opt_val;
int rc;
int mandatory = 0;
memset(opt_recv, 0, sizeof(*opt_recv));
opt = len = 0;
while (opt_ptr != opt_end) {
opt = *opt_ptr++;
len = 0;
value = NULL;
if (opt > DCCPO_MAX_RESERVED) {
if (opt_ptr == opt_end)
goto out_nonsensical_length;
len = *opt_ptr++;
if (len < 2)
goto out_nonsensical_length;
len -= 2;
value = opt_ptr;
opt_ptr += len;
if (opt_ptr > opt_end)
goto out_nonsensical_length;
}
if (dreq != NULL && (opt >= DCCPO_MIN_RX_CCID_SPECIFIC ||
opt == DCCPO_ACK_VECTOR_0 || opt == DCCPO_ACK_VECTOR_1))
goto ignore_option;
switch (opt) {
case DCCPO_PADDING:
break;
case DCCPO_MANDATORY:
if (mandatory)
goto out_invalid_option;
if (pkt_type != DCCP_PKT_DATA)
mandatory = 1;
break;
case DCCPO_NDP_COUNT:
if (len > 6)
goto out_invalid_option;
opt_recv->dccpor_ndp = dccp_decode_value_var(value, len);
dccp_pr_debug("%s opt: NDP count=%llu\n", dccp_role(sk),
(unsigned long long)opt_recv->dccpor_ndp);
break;
case DCCPO_CHANGE_L ... DCCPO_CONFIRM_R:
if (pkt_type == DCCP_PKT_DATA)
break;
if (len == 0)
goto out_invalid_option;
rc = dccp_feat_parse_options(sk, dreq, mandatory, opt,
*value, value + 1, len - 1);
if (rc)
goto out_featneg_failed;
break;
case DCCPO_TIMESTAMP:
if (len != 4)
goto out_invalid_option;
opt_val = get_unaligned((__be32 *)value);
if (unlikely(opt_val == 0)) {
DCCP_WARN("Timestamp with zero value\n");
break;
}
if (dreq != NULL) {
dreq->dreq_timestamp_echo = ntohl(opt_val);
dreq->dreq_timestamp_time = dccp_timestamp();
} else {
opt_recv->dccpor_timestamp =
dp->dccps_timestamp_echo = ntohl(opt_val);
dp->dccps_timestamp_time = dccp_timestamp();
}
dccp_pr_debug("%s rx opt: TIMESTAMP=%u, ackno=%llu\n",
dccp_role(sk), ntohl(opt_val),
(unsigned long long)
DCCP_SKB_CB(skb)->dccpd_ack_seq);
inet_csk_schedule_ack(sk);
break;
case DCCPO_TIMESTAMP_ECHO:
if (len != 4 && len != 6 && len != 8)
goto out_invalid_option;
opt_val = get_unaligned((__be32 *)value);
opt_recv->dccpor_timestamp_echo = ntohl(opt_val);
dccp_pr_debug("%s rx opt: TIMESTAMP_ECHO=%u, len=%d, "
"ackno=%llu", dccp_role(sk),
opt_recv->dccpor_timestamp_echo,
len + 2,
(unsigned long long)
DCCP_SKB_CB(skb)->dccpd_ack_seq);
value += 4;
if (len == 4) {
dccp_pr_debug_cat("\n");
break;
}
if (len == 6) {
__be16 opt_val2 = get_unaligned((__be16 *)value);
elapsed_time = ntohs(opt_val2);
} else {
opt_val = get_unaligned((__be32 *)value);
elapsed_time = ntohl(opt_val);
}
dccp_pr_debug_cat(", ELAPSED_TIME=%u\n", elapsed_time);
if (elapsed_time > opt_recv->dccpor_elapsed_time)
opt_recv->dccpor_elapsed_time = elapsed_time;
break;
case DCCPO_ELAPSED_TIME:
if (dccp_packet_without_ack(skb))
break;
if (len == 2) {
__be16 opt_val2 = get_unaligned((__be16 *)value);
elapsed_time = ntohs(opt_val2);
} else if (len == 4) {
opt_val = get_unaligned((__be32 *)value);
elapsed_time = ntohl(opt_val);
} else {
goto out_invalid_option;
}
if (elapsed_time > opt_recv->dccpor_elapsed_time)
opt_recv->dccpor_elapsed_time = elapsed_time;
dccp_pr_debug("%s rx opt: ELAPSED_TIME=%d\n",
dccp_role(sk), elapsed_time);
break;
case DCCPO_MIN_RX_CCID_SPECIFIC ... DCCPO_MAX_RX_CCID_SPECIFIC:
if (ccid_hc_rx_parse_options(dp->dccps_hc_rx_ccid, sk,
pkt_type, opt, value, len))
goto out_invalid_option;
break;
case DCCPO_ACK_VECTOR_0:
case DCCPO_ACK_VECTOR_1:
if (dccp_packet_without_ack(skb))
break;
case DCCPO_MIN_TX_CCID_SPECIFIC ... DCCPO_MAX_TX_CCID_SPECIFIC:
if (ccid_hc_tx_parse_options(dp->dccps_hc_tx_ccid, sk,
pkt_type, opt, value, len))
goto out_invalid_option;
break;
default:
DCCP_CRIT("DCCP(%p): option %d(len=%d) not "
"implemented, ignoring", sk, opt, len);
break;
}
ignore_option:
if (opt != DCCPO_MANDATORY)
mandatory = 0;
}
if (mandatory)
goto out_invalid_option;
out_nonsensical_length:
return 0;
out_invalid_option:
DCCP_INC_STATS_BH(DCCP_MIB_INVALIDOPT);
rc = DCCP_RESET_CODE_OPTION_ERROR;
out_featneg_failed:
DCCP_WARN("DCCP(%p): Option %d (len=%d) error=%u\n", sk, opt, len, rc);
DCCP_SKB_CB(skb)->dccpd_reset_code = rc;
DCCP_SKB_CB(skb)->dccpd_reset_data[0] = opt;
DCCP_SKB_CB(skb)->dccpd_reset_data[1] = len > 0 ? value[0] : 0;
DCCP_SKB_CB(skb)->dccpd_reset_data[2] = len > 1 ? value[1] : 0;
return -1;
}
