2005-08-10 05:14:34 +02:00
|
|
|
/*
|
|
|
|
* net/dccp/ipv4.c
|
|
|
|
*
|
|
|
|
* An implementation of the DCCP protocol
|
|
|
|
* Arnaldo Carvalho de Melo <acme@conectiva.com.br>
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License
|
|
|
|
* as published by the Free Software Foundation; either version
|
|
|
|
* 2 of the License, or (at your option) any later version.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/dccp.h>
|
|
|
|
#include <linux/icmp.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/skbuff.h>
|
|
|
|
#include <linux/random.h>
|
|
|
|
|
|
|
|
#include <net/icmp.h>
|
2006-03-21 06:25:11 +01:00
|
|
|
#include <net/inet_common.h>
|
2005-08-10 05:14:34 +02:00
|
|
|
#include <net/inet_hashtables.h>
|
2005-12-27 05:43:12 +01:00
|
|
|
#include <net/inet_sock.h>
|
2006-03-21 06:25:11 +01:00
|
|
|
#include <net/protocol.h>
|
2005-08-10 05:14:34 +02:00
|
|
|
#include <net/sock.h>
|
2005-12-14 08:25:19 +01:00
|
|
|
#include <net/timewait_sock.h>
|
2005-08-10 05:14:34 +02:00
|
|
|
#include <net/tcp_states.h>
|
|
|
|
#include <net/xfrm.h>
|
|
|
|
|
2005-09-18 09:17:51 +02:00
|
|
|
#include "ackvec.h"
|
2005-08-10 05:14:34 +02:00
|
|
|
#include "ccid.h"
|
|
|
|
#include "dccp.h"
|
[DCCP]: Initial feature negotiation implementation
Still needs more work, but boots and doesn't crashes, even
does some negotiation!
18:38:52.174934 127.0.0.1.43458 > 127.0.0.1.5001: request <change_l ack_ratio 2, change_r ccid 2, change_l ccid 2>
18:38:52.218526 127.0.0.1.5001 > 127.0.0.1.43458: response <nop, nop, change_l ack_ratio 2, confirm_r ccid 2 2, confirm_l ccid 2 2, confirm_r ack_ratio 2>
18:38:52.185398 127.0.0.1.43458 > 127.0.0.1.5001: <nop, confirm_r ack_ratio 2, ack_vector0 0x00, elapsed_time 212>
:-)
Signed-off-by: Andrea Bittau <a.bittau@cs.ucl.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2006-03-21 02:43:56 +01:00
|
|
|
#include "feat.h"
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2006-03-21 07:00:37 +01:00
|
|
|
/*
|
2008-04-14 11:38:45 +02:00
|
|
|
* The per-net dccp.v4_ctl_sk socket is used for responding to
|
2006-03-21 07:00:37 +01:00
|
|
|
* the Out-of-the-blue (OOTB) packets. A control sock will be created
|
|
|
|
* for this socket at the initialization time.
|
|
|
|
*/
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
|
|
|
struct inet_sock *inet = inet_sk(sk);
|
|
|
|
struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
const struct sockaddr_in *usin = (struct sockaddr_in *)uaddr;
|
|
|
|
struct rtable *rt;
|
2006-09-27 06:27:15 +02:00
|
|
|
__be32 daddr, nexthop;
|
2005-08-10 05:14:34 +02:00
|
|
|
int tmp;
|
|
|
|
int err;
|
|
|
|
|
|
|
|
dp->dccps_role = DCCP_ROLE_CLIENT;
|
|
|
|
|
|
|
|
if (addr_len < sizeof(struct sockaddr_in))
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (usin->sin_family != AF_INET)
|
|
|
|
return -EAFNOSUPPORT;
|
|
|
|
|
|
|
|
nexthop = daddr = usin->sin_addr.s_addr;
|
|
|
|
if (inet->opt != NULL && inet->opt->srr) {
|
|
|
|
if (daddr == 0)
|
|
|
|
return -EINVAL;
|
|
|
|
nexthop = inet->opt->faddr;
|
|
|
|
}
|
|
|
|
|
|
|
|
tmp = ip_route_connect(&rt, nexthop, inet->saddr,
|
|
|
|
RT_CONN_FLAGS(sk), sk->sk_bound_dev_if,
|
|
|
|
IPPROTO_DCCP,
|
2007-02-08 11:09:21 +01:00
|
|
|
inet->sport, usin->sin_port, sk, 1);
|
2005-08-10 05:14:34 +02:00
|
|
|
if (tmp < 0)
|
|
|
|
return tmp;
|
|
|
|
|
|
|
|
if (rt->rt_flags & (RTCF_MULTICAST | RTCF_BROADCAST)) {
|
|
|
|
ip_rt_put(rt);
|
|
|
|
return -ENETUNREACH;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (inet->opt == NULL || !inet->opt->srr)
|
|
|
|
daddr = rt->rt_dst;
|
|
|
|
|
|
|
|
if (inet->saddr == 0)
|
|
|
|
inet->saddr = rt->rt_src;
|
|
|
|
inet->rcv_saddr = inet->saddr;
|
|
|
|
|
|
|
|
inet->dport = usin->sin_port;
|
|
|
|
inet->daddr = daddr;
|
|
|
|
|
2005-12-14 08:26:10 +01:00
|
|
|
inet_csk(sk)->icsk_ext_hdr_len = 0;
|
2005-08-10 05:14:34 +02:00
|
|
|
if (inet->opt != NULL)
|
2005-12-14 08:26:10 +01:00
|
|
|
inet_csk(sk)->icsk_ext_hdr_len = inet->opt->optlen;
|
2005-08-10 05:14:34 +02:00
|
|
|
/*
|
|
|
|
* Socket identity is still unknown (sport may be zero).
|
|
|
|
* However we set state to DCCP_REQUESTING and not releasing socket
|
|
|
|
* lock select source port, enter ourselves into the hash tables and
|
|
|
|
* complete initialization after this.
|
|
|
|
*/
|
|
|
|
dccp_set_state(sk, DCCP_REQUESTING);
|
2005-12-14 08:25:31 +01:00
|
|
|
err = inet_hash_connect(&dccp_death_row, sk);
|
2005-08-10 05:14:34 +02:00
|
|
|
if (err != 0)
|
|
|
|
goto failure;
|
|
|
|
|
2006-02-01 02:35:35 +01:00
|
|
|
err = ip_route_newports(&rt, IPPROTO_DCCP, inet->sport, inet->dport,
|
2007-02-09 15:24:38 +01:00
|
|
|
sk);
|
2005-08-10 05:14:34 +02:00
|
|
|
if (err != 0)
|
|
|
|
goto failure;
|
|
|
|
|
|
|
|
/* OK, now commit destination to socket. */
|
|
|
|
sk_setup_caps(sk, &rt->u.dst);
|
|
|
|
|
2006-11-13 16:34:38 +01:00
|
|
|
dp->dccps_iss = secure_dccp_sequence_number(inet->saddr, inet->daddr,
|
|
|
|
inet->sport, inet->dport);
|
2005-08-10 05:14:34 +02:00
|
|
|
inet->id = dp->dccps_iss ^ jiffies;
|
|
|
|
|
|
|
|
err = dccp_connect(sk);
|
|
|
|
rt = NULL;
|
|
|
|
if (err != 0)
|
|
|
|
goto failure;
|
|
|
|
out:
|
|
|
|
return err;
|
|
|
|
failure:
|
2005-08-14 01:34:54 +02:00
|
|
|
/*
|
|
|
|
* This unhashes the socket and releases the local port, if necessary.
|
|
|
|
*/
|
2005-08-10 05:14:34 +02:00
|
|
|
dccp_set_state(sk, DCCP_CLOSED);
|
|
|
|
ip_rt_put(rt);
|
|
|
|
sk->sk_route_caps = 0;
|
|
|
|
inet->dport = 0;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_v4_connect);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
/*
|
|
|
|
* This routine does path mtu discovery as defined in RFC1191.
|
|
|
|
*/
|
|
|
|
static inline void dccp_do_pmtu_discovery(struct sock *sk,
|
|
|
|
const struct iphdr *iph,
|
|
|
|
u32 mtu)
|
|
|
|
{
|
|
|
|
struct dst_entry *dst;
|
|
|
|
const struct inet_sock *inet = inet_sk(sk);
|
|
|
|
const struct dccp_sock *dp = dccp_sk(sk);
|
|
|
|
|
|
|
|
/* We are not interested in DCCP_LISTEN and request_socks (RESPONSEs
|
|
|
|
* send out by Linux are always < 576bytes so they should go through
|
|
|
|
* unfragmented).
|
|
|
|
*/
|
|
|
|
if (sk->sk_state == DCCP_LISTEN)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* We don't check in the destentry if pmtu discovery is forbidden
|
|
|
|
* on this route. We just assume that no packet_to_big packets
|
|
|
|
* are send back when pmtu discovery is not active.
|
2007-02-09 15:24:38 +01:00
|
|
|
* There is a small race when the user changes this flag in the
|
2005-08-10 05:14:34 +02:00
|
|
|
* route, but I think that's acceptable.
|
|
|
|
*/
|
|
|
|
if ((dst = __sk_dst_check(sk, 0)) == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
dst->ops->update_pmtu(dst, mtu);
|
|
|
|
|
|
|
|
/* Something is about to be wrong... Remember soft error
|
|
|
|
* for the case, if this connection will not able to recover.
|
|
|
|
*/
|
|
|
|
if (mtu < dst_mtu(dst) && ip_dont_fragment(sk, dst))
|
|
|
|
sk->sk_err_soft = EMSGSIZE;
|
|
|
|
|
|
|
|
mtu = dst_mtu(dst);
|
|
|
|
|
|
|
|
if (inet->pmtudisc != IP_PMTUDISC_DONT &&
|
2005-12-14 08:26:10 +01:00
|
|
|
inet_csk(sk)->icsk_pmtu_cookie > mtu) {
|
2005-08-10 05:14:34 +02:00
|
|
|
dccp_sync_mss(sk, mtu);
|
|
|
|
|
|
|
|
/*
|
2006-10-25 01:17:51 +02:00
|
|
|
* From RFC 4340, sec. 14.1:
|
2005-08-10 05:14:34 +02:00
|
|
|
*
|
2005-08-14 01:34:54 +02:00
|
|
|
* DCCP-Sync packets are the best choice for upward
|
|
|
|
* probing, since DCCP-Sync probes do not risk application
|
|
|
|
* data loss.
|
2005-08-10 05:14:34 +02:00
|
|
|
*/
|
2005-08-17 08:10:59 +02:00
|
|
|
dccp_send_sync(sk, dp->dccps_gsr, DCCP_PKT_SYNC);
|
2005-08-10 05:14:34 +02:00
|
|
|
} /* else let the usual retransmit timer handle it */
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
2006-03-21 06:25:11 +01:00
|
|
|
static void dccp_v4_err(struct sk_buff *skb, u32 info)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
|
|
|
const struct iphdr *iph = (struct iphdr *)skb->data;
|
2008-07-26 12:59:10 +02:00
|
|
|
const u8 offset = iph->ihl << 2;
|
|
|
|
const struct dccp_hdr *dh = (struct dccp_hdr *)(skb->data + offset);
|
2005-08-10 05:14:34 +02:00
|
|
|
struct dccp_sock *dp;
|
|
|
|
struct inet_sock *inet;
|
2007-03-13 18:43:18 +01:00
|
|
|
const int type = icmp_hdr(skb)->type;
|
|
|
|
const int code = icmp_hdr(skb)->code;
|
2005-08-10 05:14:34 +02:00
|
|
|
struct sock *sk;
|
|
|
|
__u64 seq;
|
|
|
|
int err;
|
2008-07-15 08:01:40 +02:00
|
|
|
struct net *net = dev_net(skb->dev);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2008-07-26 12:59:10 +02:00
|
|
|
if (skb->len < offset + sizeof(*dh) ||
|
|
|
|
skb->len < offset + __dccp_basic_hdr_len(dh)) {
|
2008-07-15 08:03:00 +02:00
|
|
|
ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
|
2005-08-10 05:14:34 +02:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2008-07-15 08:01:40 +02:00
|
|
|
sk = inet_lookup(net, &dccp_hashinfo,
|
2008-04-14 07:30:43 +02:00
|
|
|
iph->daddr, dh->dccph_dport,
|
|
|
|
iph->saddr, dh->dccph_sport, inet_iif(skb));
|
2005-08-10 05:14:34 +02:00
|
|
|
if (sk == NULL) {
|
2008-07-15 08:03:00 +02:00
|
|
|
ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
|
2005-08-10 05:14:34 +02:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (sk->sk_state == DCCP_TIME_WAIT) {
|
2006-10-11 04:41:46 +02:00
|
|
|
inet_twsk_put(inet_twsk(sk));
|
2005-08-10 05:14:34 +02:00
|
|
|
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))
|
2008-07-17 05:31:16 +02:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_LOCKDROPPEDICMPS);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
if (sk->sk_state == DCCP_CLOSED)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
dp = dccp_sk(sk);
|
2007-10-24 14:12:09 +02:00
|
|
|
seq = dccp_hdr_seq(dh);
|
2007-10-24 14:18:06 +02:00
|
|
|
if ((1 << sk->sk_state) & ~(DCCPF_REQUESTING | DCCPF_LISTEN) &&
|
2008-07-26 12:59:10 +02:00
|
|
|
!between48(seq, dp->dccps_awl, dp->dccps_awh)) {
|
2008-07-17 05:31:16 +02:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
|
2005-08-10 05:14:34 +02:00
|
|
|
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.
|
|
|
|
*/
|
2008-07-26 06:43:18 +02:00
|
|
|
WARN_ON(req->sk);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
if (seq != dccp_rsk(req)->dreq_iss) {
|
2008-07-17 05:31:16 +02:00
|
|
|
NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
|
2005-08-10 05:14:34 +02:00
|
|
|
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);
|
|
|
|
}
|
|
|
|
|
2006-11-15 06:28:51 +01:00
|
|
|
static inline __sum16 dccp_v4_csum_finish(struct sk_buff *skb,
|
2006-11-10 20:43:06 +01:00
|
|
|
__be32 src, __be32 dst)
|
|
|
|
{
|
|
|
|
return csum_tcpudp_magic(src, dst, skb->len, IPPROTO_DCCP, skb->csum);
|
|
|
|
}
|
|
|
|
|
|
|
|
void dccp_v4_send_check(struct sock *sk, int unused, struct sk_buff *skb)
|
2005-12-14 08:16:16 +01:00
|
|
|
{
|
|
|
|
const struct inet_sock *inet = inet_sk(sk);
|
|
|
|
struct dccp_hdr *dh = dccp_hdr(skb);
|
|
|
|
|
2006-11-10 20:43:06 +01:00
|
|
|
dccp_csum_outgoing(skb);
|
|
|
|
dh->dccph_checksum = dccp_v4_csum_finish(skb, inet->saddr, inet->daddr);
|
2005-12-14 08:16:16 +01:00
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_v4_send_check);
|
|
|
|
|
2006-11-13 16:31:50 +01:00
|
|
|
static inline u64 dccp_v4_init_sequence(const struct sk_buff *skb)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
2007-04-21 07:47:35 +02:00
|
|
|
return secure_dccp_sequence_number(ip_hdr(skb)->daddr,
|
|
|
|
ip_hdr(skb)->saddr,
|
2005-08-10 05:14:34 +02:00
|
|
|
dccp_hdr(skb)->dccph_dport,
|
|
|
|
dccp_hdr(skb)->dccph_sport);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The three way handshake has completed - we got a valid ACK or DATAACK -
|
|
|
|
* now create the new socket.
|
|
|
|
*
|
|
|
|
* This is the equivalent of TCP's tcp_v4_syn_recv_sock
|
|
|
|
*/
|
|
|
|
struct sock *dccp_v4_request_recv_sock(struct sock *sk, struct sk_buff *skb,
|
|
|
|
struct request_sock *req,
|
|
|
|
struct dst_entry *dst)
|
|
|
|
{
|
|
|
|
struct inet_request_sock *ireq;
|
|
|
|
struct inet_sock *newinet;
|
|
|
|
struct sock *newsk;
|
|
|
|
|
|
|
|
if (sk_acceptq_is_full(sk))
|
|
|
|
goto exit_overflow;
|
|
|
|
|
|
|
|
if (dst == NULL && (dst = inet_csk_route_req(sk, req)) == NULL)
|
|
|
|
goto exit;
|
|
|
|
|
|
|
|
newsk = dccp_create_openreq_child(sk, req, skb);
|
|
|
|
if (newsk == NULL)
|
|
|
|
goto exit;
|
|
|
|
|
|
|
|
sk_setup_caps(newsk, dst);
|
|
|
|
|
|
|
|
newinet = inet_sk(newsk);
|
|
|
|
ireq = inet_rsk(req);
|
|
|
|
newinet->daddr = ireq->rmt_addr;
|
|
|
|
newinet->rcv_saddr = ireq->loc_addr;
|
|
|
|
newinet->saddr = ireq->loc_addr;
|
|
|
|
newinet->opt = ireq->opt;
|
|
|
|
ireq->opt = NULL;
|
|
|
|
newinet->mc_index = inet_iif(skb);
|
2007-04-21 07:47:35 +02:00
|
|
|
newinet->mc_ttl = ip_hdr(skb)->ttl;
|
2005-08-10 05:14:34 +02:00
|
|
|
newinet->id = jiffies;
|
|
|
|
|
|
|
|
dccp_sync_mss(newsk, dst_mtu(dst));
|
|
|
|
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 13:06:04 +01:00
|
|
|
__inet_hash_nolisten(newsk);
|
|
|
|
__inet_inherit_port(sk, newsk);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
return newsk;
|
|
|
|
|
|
|
|
exit_overflow:
|
2008-07-17 05:31:16 +02:00
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENOVERFLOWS);
|
2005-08-10 05:14:34 +02:00
|
|
|
exit:
|
2008-07-17 05:31:16 +02:00
|
|
|
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_LISTENDROPS);
|
2005-08-10 05:14:34 +02:00
|
|
|
dst_release(dst);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_v4_request_recv_sock);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
static struct sock *dccp_v4_hnd_req(struct sock *sk, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
const struct dccp_hdr *dh = dccp_hdr(skb);
|
2007-04-21 07:47:35 +02:00
|
|
|
const struct iphdr *iph = ip_hdr(skb);
|
2005-08-10 05:14:34 +02:00
|
|
|
struct sock *nsk;
|
|
|
|
struct request_sock **prev;
|
|
|
|
/* Find possible connection requests. */
|
|
|
|
struct request_sock *req = inet_csk_search_req(sk, &prev,
|
|
|
|
dh->dccph_sport,
|
|
|
|
iph->saddr, iph->daddr);
|
|
|
|
if (req != NULL)
|
|
|
|
return dccp_check_req(sk, skb, req, prev);
|
|
|
|
|
2008-04-14 07:30:43 +02:00
|
|
|
nsk = inet_lookup_established(sock_net(sk), &dccp_hashinfo,
|
2006-08-10 00:47:12 +02:00
|
|
|
iph->saddr, dh->dccph_sport,
|
|
|
|
iph->daddr, dh->dccph_dport,
|
|
|
|
inet_iif(skb));
|
2005-08-10 05:14:34 +02:00
|
|
|
if (nsk != NULL) {
|
|
|
|
if (nsk->sk_state != DCCP_TIME_WAIT) {
|
|
|
|
bh_lock_sock(nsk);
|
|
|
|
return nsk;
|
|
|
|
}
|
2006-10-11 04:41:46 +02:00
|
|
|
inet_twsk_put(inet_twsk(nsk));
|
2005-08-10 05:14:34 +02:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return sk;
|
|
|
|
}
|
|
|
|
|
2008-04-14 07:30:19 +02:00
|
|
|
static struct dst_entry* dccp_v4_route_skb(struct net *net, struct sock *sk,
|
2005-08-10 05:14:34 +02:00
|
|
|
struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
struct rtable *rt;
|
2008-03-06 03:30:47 +01:00
|
|
|
struct flowi fl = { .oif = skb->rtable->rt_iif,
|
2005-08-10 05:14:34 +02:00
|
|
|
.nl_u = { .ip4_u =
|
2007-04-21 07:47:35 +02:00
|
|
|
{ .daddr = ip_hdr(skb)->saddr,
|
|
|
|
.saddr = ip_hdr(skb)->daddr,
|
2005-08-10 05:14:34 +02:00
|
|
|
.tos = RT_CONN_FLAGS(sk) } },
|
|
|
|
.proto = sk->sk_protocol,
|
|
|
|
.uli_u = { .ports =
|
|
|
|
{ .sport = dccp_hdr(skb)->dccph_dport,
|
2005-08-14 01:34:54 +02:00
|
|
|
.dport = dccp_hdr(skb)->dccph_sport }
|
2006-12-10 19:01:18 +01:00
|
|
|
}
|
2005-08-14 01:34:54 +02:00
|
|
|
};
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2006-08-05 08:12:42 +02:00
|
|
|
security_skb_classify_flow(skb, &fl);
|
2008-04-14 07:30:19 +02:00
|
|
|
if (ip_route_output_flow(net, &rt, &fl, sk, 0)) {
|
2008-07-17 05:20:11 +02:00
|
|
|
IP_INC_STATS_BH(net, IPSTATS_MIB_OUTNOROUTES);
|
2005-08-10 05:14:34 +02:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
return &rt->u.dst;
|
|
|
|
}
|
|
|
|
|
2008-02-29 20:43:03 +01:00
|
|
|
static int dccp_v4_send_response(struct sock *sk, struct request_sock *req)
|
2006-11-10 15:52:36 +01:00
|
|
|
{
|
|
|
|
int err = -1;
|
|
|
|
struct sk_buff *skb;
|
2008-02-29 20:43:03 +01:00
|
|
|
struct dst_entry *dst;
|
2006-11-10 15:52:36 +01:00
|
|
|
|
2008-02-29 20:43:03 +01:00
|
|
|
dst = inet_csk_route_req(sk, req);
|
|
|
|
if (dst == NULL)
|
2006-11-10 15:52:36 +01:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
skb = dccp_make_response(sk, dst, req);
|
|
|
|
if (skb != NULL) {
|
|
|
|
const struct inet_request_sock *ireq = inet_rsk(req);
|
|
|
|
struct dccp_hdr *dh = dccp_hdr(skb);
|
|
|
|
|
2006-11-10 20:43:06 +01:00
|
|
|
dh->dccph_checksum = dccp_v4_csum_finish(skb, ireq->loc_addr,
|
|
|
|
ireq->rmt_addr);
|
2006-11-10 15:52:36 +01:00
|
|
|
err = ip_build_and_send_pkt(skb, sk, ireq->loc_addr,
|
|
|
|
ireq->rmt_addr,
|
|
|
|
ireq->opt);
|
2006-11-14 14:21:36 +01:00
|
|
|
err = net_xmit_eval(err);
|
2006-11-10 15:52:36 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
out:
|
|
|
|
dst_release(dst);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2006-11-15 04:07:45 +01:00
|
|
|
static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
|
|
|
int err;
|
2007-04-21 07:47:35 +02:00
|
|
|
const struct iphdr *rxiph;
|
2005-08-10 05:14:34 +02:00
|
|
|
struct sk_buff *skb;
|
|
|
|
struct dst_entry *dst;
|
2008-04-14 07:29:59 +02:00
|
|
|
struct net *net = dev_net(rxskb->dst->dev);
|
|
|
|
struct sock *ctl_sk = net->dccp.v4_ctl_sk;
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
/* Never send a reset in response to a reset. */
|
[DCCP]: Factor out common code for generating Resets
This factors code common to dccp_v{4,6}_ctl_send_reset into a separate function,
and adds support for filling in the Data 1 ... Data 3 fields from RFC 4340, 5.6.
It is useful to have this separate, since the following Reset codes will always
be generated from the control socket rather than via dccp_send_reset:
* Code 3, "No Connection", cf. 8.3.1;
* Code 4, "Packet Error" (identification for Data 1 added);
* Code 5, "Option Error" (identification for Data 1..3 added, will be used later);
* Code 6, "Mandatory Error" (same as Option Error);
* Code 7, "Connection Refused" (what on Earth is the difference to "No Connection"?);
* Code 8, "Bad Service Code";
* Code 9, "Too Busy";
* Code 10, "Bad Init Cookie" (not used).
Code 0 is not recommended by the RFC, the following codes would be used in
dccp_send_reset() instead, since they all relate to an established DCCP connection:
* Code 1, "Closed";
* Code 2, "Aborted";
* Code 11, "Aggression Penalty" (12.3).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2007-09-26 19:35:19 +02:00
|
|
|
if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET)
|
2005-08-10 05:14:34 +02:00
|
|
|
return;
|
|
|
|
|
2008-03-06 03:30:47 +01:00
|
|
|
if (rxskb->rtable->rt_type != RTN_LOCAL)
|
2005-08-10 05:14:34 +02:00
|
|
|
return;
|
|
|
|
|
2008-04-14 07:30:19 +02:00
|
|
|
dst = dccp_v4_route_skb(net, ctl_sk, rxskb);
|
2005-08-10 05:14:34 +02:00
|
|
|
if (dst == NULL)
|
|
|
|
return;
|
|
|
|
|
2008-04-14 07:29:37 +02:00
|
|
|
skb = dccp_ctl_make_reset(ctl_sk, rxskb);
|
2005-08-10 05:14:34 +02:00
|
|
|
if (skb == NULL)
|
|
|
|
goto out;
|
|
|
|
|
2007-04-21 07:47:35 +02:00
|
|
|
rxiph = ip_hdr(rxskb);
|
[DCCP]: Factor out common code for generating Resets
This factors code common to dccp_v{4,6}_ctl_send_reset into a separate function,
and adds support for filling in the Data 1 ... Data 3 fields from RFC 4340, 5.6.
It is useful to have this separate, since the following Reset codes will always
be generated from the control socket rather than via dccp_send_reset:
* Code 3, "No Connection", cf. 8.3.1;
* Code 4, "Packet Error" (identification for Data 1 added);
* Code 5, "Option Error" (identification for Data 1..3 added, will be used later);
* Code 6, "Mandatory Error" (same as Option Error);
* Code 7, "Connection Refused" (what on Earth is the difference to "No Connection"?);
* Code 8, "Bad Service Code";
* Code 9, "Too Busy";
* Code 10, "Bad Init Cookie" (not used).
Code 0 is not recommended by the RFC, the following codes would be used in
dccp_send_reset() instead, since they all relate to an established DCCP connection:
* Code 1, "Closed";
* Code 2, "Aborted";
* Code 11, "Aggression Penalty" (12.3).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
2007-09-26 19:35:19 +02:00
|
|
|
dccp_hdr(skb)->dccph_checksum = dccp_v4_csum_finish(skb, rxiph->saddr,
|
|
|
|
rxiph->daddr);
|
|
|
|
skb->dst = dst_clone(dst);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2008-04-14 07:29:37 +02:00
|
|
|
bh_lock_sock(ctl_sk);
|
|
|
|
err = ip_build_and_send_pkt(skb, ctl_sk,
|
2007-04-21 07:47:35 +02:00
|
|
|
rxiph->daddr, rxiph->saddr, NULL);
|
2008-04-14 07:29:37 +02:00
|
|
|
bh_unlock_sock(ctl_sk);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2006-11-14 14:21:36 +01:00
|
|
|
if (net_xmit_eval(err) == 0) {
|
2005-08-10 05:14:34 +02:00
|
|
|
DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS);
|
|
|
|
DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS);
|
|
|
|
}
|
|
|
|
out:
|
|
|
|
dst_release(dst);
|
|
|
|
}
|
|
|
|
|
2006-11-10 15:52:36 +01:00
|
|
|
static void dccp_v4_reqsk_destructor(struct request_sock *req)
|
|
|
|
{
|
|
|
|
kfree(inet_rsk(req)->opt);
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct request_sock_ops dccp_request_sock_ops __read_mostly = {
|
|
|
|
.family = PF_INET,
|
|
|
|
.obj_size = sizeof(struct dccp_request_sock),
|
|
|
|
.rtx_syn_ack = dccp_v4_send_response,
|
|
|
|
.send_ack = dccp_reqsk_send_ack,
|
|
|
|
.destructor = dccp_v4_reqsk_destructor,
|
|
|
|
.send_reset = dccp_v4_ctl_send_reset,
|
|
|
|
};
|
|
|
|
|
|
|
|
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;
|
2006-12-10 19:01:18 +01:00
|
|
|
const __be32 service = dccp_hdr_request(skb)->dccph_req_service;
|
2006-11-10 15:52:36 +01:00
|
|
|
struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb);
|
|
|
|
|
|
|
|
/* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */
|
2008-03-06 03:30:47 +01:00
|
|
|
if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST))
|
[DCCP]: Twice the wrong reset code in receiving connection-Requests
This fixes two bugs in processing of connection-Requests in
v{4,6}_conn_request:
1. Due to using the variable `reset_code', the Reset code generated
internally by dccp_parse_options() is overwritten with the
initialised value ("Too Busy") of reset_code, which is not what is
intended.
2. When receiving a connection-Request on a multicast or broadcast
address, no Reset should be generated, to avoid storms of such
packets. Instead of jumping to the `drop' label, the
v{4,6}_conn_request functions now return 0. Below is why in my
understanding this is correct:
When the conn_request function returns < 0, then the caller,
dccp_rcv_state_process(), returns 1. In all instances where
dccp_rcv_state_process is called (dccp_v4_do_rcv, dccp_v6_do_rcv,
and dccp_child_process), a return value of != 0 from
dccp_rcv_state_process() means that a Reset is generated.
If on the other hand the conn_request function returns 0, the
packet is discarded and no Reset is generated.
Note: There may be a related problem when sending the Response, due to
the following.
if (dccp_v6_send_response(sk, req, NULL))
goto drop_and_free;
/* ... */
drop_and_free:
return -1;
In this case, if send_response fails due to transmission errors, the
next thing that is generated is a Reset with a code "Too Busy". I
haven't been able to conjure up such a condition, but it might be good
to change the behaviour here also (not done by this patch).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 23:52:28 +02:00
|
|
|
return 0; /* discard, don't send a reset here */
|
2006-11-10 15:52:36 +01:00
|
|
|
|
|
|
|
if (dccp_bad_service_code(sk, service)) {
|
[DCCP]: Twice the wrong reset code in receiving connection-Requests
This fixes two bugs in processing of connection-Requests in
v{4,6}_conn_request:
1. Due to using the variable `reset_code', the Reset code generated
internally by dccp_parse_options() is overwritten with the
initialised value ("Too Busy") of reset_code, which is not what is
intended.
2. When receiving a connection-Request on a multicast or broadcast
address, no Reset should be generated, to avoid storms of such
packets. Instead of jumping to the `drop' label, the
v{4,6}_conn_request functions now return 0. Below is why in my
understanding this is correct:
When the conn_request function returns < 0, then the caller,
dccp_rcv_state_process(), returns 1. In all instances where
dccp_rcv_state_process is called (dccp_v4_do_rcv, dccp_v6_do_rcv,
and dccp_child_process), a return value of != 0 from
dccp_rcv_state_process() means that a Reset is generated.
If on the other hand the conn_request function returns 0, the
packet is discarded and no Reset is generated.
Note: There may be a related problem when sending the Response, due to
the following.
if (dccp_v6_send_response(sk, req, NULL))
goto drop_and_free;
/* ... */
drop_and_free:
return -1;
In this case, if send_response fails due to transmission errors, the
next thing that is generated is a Reset with a code "Too Busy". I
haven't been able to conjure up such a condition, but it might be good
to change the behaviour here also (not done by this patch).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 23:52:28 +02:00
|
|
|
dcb->dccpd_reset_code = DCCP_RESET_CODE_BAD_SERVICE_CODE;
|
2006-11-10 15:52:36 +01:00
|
|
|
goto drop;
|
2006-12-10 19:01:18 +01:00
|
|
|
}
|
2006-11-10 15:52:36 +01:00
|
|
|
/*
|
|
|
|
* TW buckets are converted to open requests without
|
|
|
|
* limitations, they conserve resources and peer is
|
|
|
|
* evidently real one.
|
|
|
|
*/
|
[DCCP]: Twice the wrong reset code in receiving connection-Requests
This fixes two bugs in processing of connection-Requests in
v{4,6}_conn_request:
1. Due to using the variable `reset_code', the Reset code generated
internally by dccp_parse_options() is overwritten with the
initialised value ("Too Busy") of reset_code, which is not what is
intended.
2. When receiving a connection-Request on a multicast or broadcast
address, no Reset should be generated, to avoid storms of such
packets. Instead of jumping to the `drop' label, the
v{4,6}_conn_request functions now return 0. Below is why in my
understanding this is correct:
When the conn_request function returns < 0, then the caller,
dccp_rcv_state_process(), returns 1. In all instances where
dccp_rcv_state_process is called (dccp_v4_do_rcv, dccp_v6_do_rcv,
and dccp_child_process), a return value of != 0 from
dccp_rcv_state_process() means that a Reset is generated.
If on the other hand the conn_request function returns 0, the
packet is discarded and no Reset is generated.
Note: There may be a related problem when sending the Response, due to
the following.
if (dccp_v6_send_response(sk, req, NULL))
goto drop_and_free;
/* ... */
drop_and_free:
return -1;
In this case, if send_response fails due to transmission errors, the
next thing that is generated is a Reset with a code "Too Busy". I
haven't been able to conjure up such a condition, but it might be good
to change the behaviour here also (not done by this patch).
Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 23:52:28 +02:00
|
|
|
dcb->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY;
|
2006-11-10 15:52:36 +01:00
|
|
|
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;
|
|
|
|
|
2008-06-10 21:39:35 +02:00
|
|
|
req = inet_reqsk_alloc(&dccp_request_sock_ops);
|
2006-11-10 15:52:36 +01:00
|
|
|
if (req == NULL)
|
|
|
|
goto drop;
|
|
|
|
|
2008-11-05 08:55:49 +01:00
|
|
|
if (dccp_reqsk_init(req, dccp_sk(sk), skb))
|
|
|
|
goto drop_and_free;
|
2006-11-10 15:52:36 +01:00
|
|
|
|
2007-12-13 15:29:24 +01:00
|
|
|
dreq = dccp_rsk(req);
|
|
|
|
if (dccp_parse_options(sk, dreq, skb))
|
|
|
|
goto drop_and_free;
|
|
|
|
|
2006-11-10 15:52:36 +01:00
|
|
|
if (security_inet_conn_request(sk, skb, req))
|
|
|
|
goto drop_and_free;
|
|
|
|
|
|
|
|
ireq = inet_rsk(req);
|
2007-04-21 07:47:35 +02:00
|
|
|
ireq->loc_addr = ip_hdr(skb)->daddr;
|
|
|
|
ireq->rmt_addr = ip_hdr(skb)->saddr;
|
2006-11-10 15:52:36 +01:00
|
|
|
|
2006-12-10 19:01:18 +01:00
|
|
|
/*
|
2006-11-10 15:52:36 +01:00
|
|
|
* 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->dreq_isr = dcb->dccpd_seq;
|
2006-11-13 16:31:50 +01:00
|
|
|
dreq->dreq_iss = dccp_v4_init_sequence(skb);
|
2006-11-10 15:52:36 +01:00
|
|
|
dreq->dreq_service = service;
|
|
|
|
|
2008-02-29 20:43:03 +01:00
|
|
|
if (dccp_v4_send_response(sk, req))
|
2006-11-10 15:52:36 +01:00
|
|
|
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;
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_v4_conn_request);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
int dccp_v4_do_rcv(struct sock *sk, struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
struct dccp_hdr *dh = dccp_hdr(skb);
|
|
|
|
|
|
|
|
if (sk->sk_state == DCCP_OPEN) { /* Fast path */
|
|
|
|
if (dccp_rcv_established(sk, skb, dh, skb->len))
|
|
|
|
goto reset;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Step 3: Process LISTEN state
|
2006-11-10 19:29:14 +01:00
|
|
|
* If P.type == Request or P contains a valid Init Cookie option,
|
|
|
|
* (* Must scan the packet's options to check for Init
|
|
|
|
* Cookies. Only Init Cookies are processed here,
|
|
|
|
* however; other options are processed in Step 8. This
|
|
|
|
* scan need only be performed if the endpoint uses Init
|
|
|
|
* Cookies *)
|
|
|
|
* (* Generate a new socket and switch to that socket *)
|
|
|
|
* Set S := new socket for this port pair
|
|
|
|
* S.state = RESPOND
|
|
|
|
* Choose S.ISS (initial seqno) or set from Init Cookies
|
|
|
|
* Initialize S.GAR := S.ISS
|
|
|
|
* Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init Cookies
|
|
|
|
* Continue with S.state == RESPOND
|
|
|
|
* (* A Response packet will be generated in Step 11 *)
|
|
|
|
* Otherwise,
|
|
|
|
* Generate Reset(No Connection) unless P.type == Reset
|
|
|
|
* Drop packet and return
|
2005-08-10 05:14:34 +02:00
|
|
|
*
|
2005-08-14 01:34:54 +02:00
|
|
|
* NOTE: the check for the packet types is done in
|
|
|
|
* dccp_rcv_state_process
|
2005-08-10 05:14:34 +02:00
|
|
|
*/
|
|
|
|
if (sk->sk_state == DCCP_LISTEN) {
|
|
|
|
struct sock *nsk = dccp_v4_hnd_req(sk, skb);
|
|
|
|
|
|
|
|
if (nsk == NULL)
|
|
|
|
goto discard;
|
|
|
|
|
|
|
|
if (nsk != sk) {
|
|
|
|
if (dccp_child_process(sk, nsk, skb))
|
|
|
|
goto reset;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (dccp_rcv_state_process(sk, skb, dh, skb->len))
|
|
|
|
goto reset;
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
reset:
|
2006-11-15 04:07:45 +01:00
|
|
|
dccp_v4_ctl_send_reset(sk, skb);
|
2005-08-10 05:14:34 +02:00
|
|
|
discard:
|
|
|
|
kfree_skb(skb);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_v4_do_rcv);
|
|
|
|
|
2006-11-14 15:57:34 +01:00
|
|
|
/**
|
|
|
|
* dccp_invalid_packet - check for malformed packets
|
|
|
|
* Implements RFC 4340, 8.5: Step 1: Check header basics
|
|
|
|
* Packets that fail these checks are ignored and do not receive Resets.
|
|
|
|
*/
|
2005-12-14 08:24:16 +01:00
|
|
|
int dccp_invalid_packet(struct sk_buff *skb)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
|
|
|
const struct dccp_hdr *dh;
|
2006-11-10 20:43:06 +01:00
|
|
|
unsigned int cscov;
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
if (skb->pkt_type != PACKET_HOST)
|
|
|
|
return 1;
|
|
|
|
|
2006-11-14 15:57:34 +01:00
|
|
|
/* If the packet is shorter than 12 bytes, drop packet and return */
|
2005-08-10 05:14:34 +02:00
|
|
|
if (!pskb_may_pull(skb, sizeof(struct dccp_hdr))) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("pskb_may_pull failed\n");
|
2005-08-10 05:14:34 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
dh = dccp_hdr(skb);
|
|
|
|
|
2006-11-14 15:57:34 +01:00
|
|
|
/* If P.type is not understood, drop packet and return */
|
2005-08-10 05:14:34 +02:00
|
|
|
if (dh->dccph_type >= DCCP_PKT_INVALID) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("invalid packet type\n");
|
2005-08-10 05:14:34 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2006-11-14 15:57:34 +01:00
|
|
|
* If P.Data Offset is too small for packet type, drop packet and return
|
2005-08-10 05:14:34 +02:00
|
|
|
*/
|
|
|
|
if (dh->dccph_doff < dccp_hdr_len(skb) / sizeof(u32)) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("P.Data Offset(%u) too small\n", dh->dccph_doff);
|
2005-08-10 05:14:34 +02:00
|
|
|
return 1;
|
|
|
|
}
|
2006-11-14 15:57:34 +01:00
|
|
|
/*
|
|
|
|
* If P.Data Offset is too too large for packet, drop packet and return
|
|
|
|
*/
|
2005-08-10 05:14:34 +02:00
|
|
|
if (!pskb_may_pull(skb, dh->dccph_doff * sizeof(u32))) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("P.Data Offset(%u) too large\n", dh->dccph_doff);
|
2005-08-10 05:14:34 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If P.type is not Data, Ack, or DataAck and P.X == 0 (the packet
|
|
|
|
* has short sequence numbers), drop packet and return
|
|
|
|
*/
|
2008-05-27 15:22:38 +02:00
|
|
|
if ((dh->dccph_type < DCCP_PKT_DATA ||
|
|
|
|
dh->dccph_type > DCCP_PKT_DATAACK) && dh->dccph_x == 0) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("P.type (%s) not Data || [Data]Ack, while P.X == 0\n",
|
|
|
|
dccp_packet_name(dh->dccph_type));
|
2005-08-10 05:14:34 +02:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
2006-11-10 20:43:06 +01:00
|
|
|
/*
|
|
|
|
* If P.CsCov is too large for the packet size, drop packet and return.
|
|
|
|
* This must come _before_ checksumming (not as RFC 4340 suggests).
|
|
|
|
*/
|
|
|
|
cscov = dccp_csum_coverage(skb);
|
|
|
|
if (cscov > skb->len) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("P.CsCov %u exceeds packet length %d\n",
|
|
|
|
dh->dccph_cscov, skb->len);
|
2006-11-10 20:43:06 +01:00
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* If header checksum is incorrect, drop packet and return.
|
|
|
|
* (This step is completed in the AF-dependent functions.) */
|
|
|
|
skb->csum = skb_checksum(skb, 0, cscov, 0);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_invalid_packet);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
/* this is called when real data arrives */
|
2006-03-21 06:25:11 +01:00
|
|
|
static int dccp_v4_rcv(struct sk_buff *skb)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
|
|
|
const struct dccp_hdr *dh;
|
2007-04-21 07:47:35 +02:00
|
|
|
const struct iphdr *iph;
|
2005-08-10 05:14:34 +02:00
|
|
|
struct sock *sk;
|
2006-11-10 20:43:06 +01:00
|
|
|
int min_cov;
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2006-11-10 20:43:06 +01:00
|
|
|
/* Step 1: Check header basics */
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
if (dccp_invalid_packet(skb))
|
|
|
|
goto discard_it;
|
|
|
|
|
2007-04-21 07:47:35 +02:00
|
|
|
iph = ip_hdr(skb);
|
2006-11-10 20:43:06 +01:00
|
|
|
/* Step 1: If header checksum is incorrect, drop packet and return */
|
2007-04-21 07:47:35 +02:00
|
|
|
if (dccp_v4_csum_finish(skb, iph->saddr, iph->daddr)) {
|
2006-11-20 21:39:23 +01:00
|
|
|
DCCP_WARN("dropped packet with invalid checksum\n");
|
2005-12-14 08:24:16 +01:00
|
|
|
goto discard_it;
|
|
|
|
}
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
dh = dccp_hdr(skb);
|
|
|
|
|
2007-10-24 14:12:09 +02:00
|
|
|
DCCP_SKB_CB(skb)->dccpd_seq = dccp_hdr_seq(dh);
|
2005-08-10 05:14:34 +02:00
|
|
|
DCCP_SKB_CB(skb)->dccpd_type = dh->dccph_type;
|
|
|
|
|
2008-10-31 08:54:56 +01:00
|
|
|
dccp_pr_debug("%8.8s src=%pI4@%-5d dst=%pI4@%-5d seq=%llu",
|
2005-08-10 05:14:34 +02:00
|
|
|
dccp_packet_name(dh->dccph_type),
|
2008-10-31 08:54:56 +01:00
|
|
|
&iph->saddr, ntohs(dh->dccph_sport),
|
|
|
|
&iph->daddr, ntohs(dh->dccph_dport),
|
2005-08-10 05:27:14 +02:00
|
|
|
(unsigned long long) DCCP_SKB_CB(skb)->dccpd_seq);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
if (dccp_packet_without_ack(skb)) {
|
|
|
|
DCCP_SKB_CB(skb)->dccpd_ack_seq = DCCP_PKT_WITHOUT_ACK_SEQ;
|
|
|
|
dccp_pr_debug_cat("\n");
|
|
|
|
} else {
|
|
|
|
DCCP_SKB_CB(skb)->dccpd_ack_seq = dccp_hdr_ack_seq(skb);
|
2006-11-10 14:46:34 +01:00
|
|
|
dccp_pr_debug_cat(", ack=%llu\n", (unsigned long long)
|
2005-08-10 05:27:14 +02:00
|
|
|
DCCP_SKB_CB(skb)->dccpd_ack_seq);
|
2005-08-10 05:14:34 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Step 2:
|
2006-12-10 19:01:18 +01:00
|
|
|
* Look up flow ID in table and get corresponding socket */
|
2008-10-07 20:41:57 +02:00
|
|
|
sk = __inet_lookup_skb(&dccp_hashinfo, skb,
|
|
|
|
dh->dccph_sport, dh->dccph_dport);
|
2006-12-10 19:01:18 +01:00
|
|
|
/*
|
2005-08-10 05:14:34 +02:00
|
|
|
* Step 2:
|
2006-12-10 19:01:18 +01:00
|
|
|
* If no socket ...
|
2005-08-10 05:14:34 +02:00
|
|
|
*/
|
|
|
|
if (sk == NULL) {
|
|
|
|
dccp_pr_debug("failed to look up flow ID in table and "
|
|
|
|
"get corresponding socket\n");
|
|
|
|
goto no_dccp_socket;
|
|
|
|
}
|
|
|
|
|
2006-12-10 19:01:18 +01:00
|
|
|
/*
|
2005-08-10 05:14:34 +02:00
|
|
|
* Step 2:
|
2006-12-10 19:01:18 +01:00
|
|
|
* ... or S.state == TIMEWAIT,
|
2005-08-10 05:14:34 +02:00
|
|
|
* Generate Reset(No Connection) unless P.type == Reset
|
|
|
|
* Drop packet and return
|
|
|
|
*/
|
|
|
|
if (sk->sk_state == DCCP_TIME_WAIT) {
|
2006-11-10 14:46:34 +01:00
|
|
|
dccp_pr_debug("sk->sk_state == DCCP_TIME_WAIT: do_time_wait\n");
|
|
|
|
inet_twsk_put(inet_twsk(sk));
|
|
|
|
goto no_dccp_socket;
|
2005-08-10 05:14:34 +02:00
|
|
|
}
|
|
|
|
|
2006-11-10 20:43:06 +01:00
|
|
|
/*
|
|
|
|
* RFC 4340, sec. 9.2.1: Minimum Checksum Coverage
|
2006-12-10 19:01:18 +01:00
|
|
|
* o if MinCsCov = 0, only packets with CsCov = 0 are accepted
|
|
|
|
* o if MinCsCov > 0, also accept packets with CsCov >= MinCsCov
|
2006-11-10 20:43:06 +01:00
|
|
|
*/
|
|
|
|
min_cov = dccp_sk(sk)->dccps_pcrlen;
|
|
|
|
if (dh->dccph_cscov && (min_cov == 0 || dh->dccph_cscov < min_cov)) {
|
|
|
|
dccp_pr_debug("Packet CsCov %d does not satisfy MinCsCov %d\n",
|
|
|
|
dh->dccph_cscov, min_cov);
|
|
|
|
/* FIXME: "Such packets SHOULD be reported using Data Dropped
|
|
|
|
* options (Section 11.7) with Drop Code 0, Protocol
|
|
|
|
* Constraints." */
|
|
|
|
goto discard_and_relse;
|
|
|
|
}
|
|
|
|
|
2005-12-27 05:42:22 +01:00
|
|
|
if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
|
2005-08-10 05:14:34 +02:00
|
|
|
goto discard_and_relse;
|
2006-01-07 08:06:30 +01:00
|
|
|
nf_reset(skb);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2006-11-16 17:06:06 +01:00
|
|
|
return sk_receive_skb(sk, skb, 1);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
no_dccp_socket:
|
|
|
|
if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
|
|
|
|
goto discard_it;
|
|
|
|
/*
|
|
|
|
* Step 2:
|
2006-12-10 19:01:18 +01:00
|
|
|
* If no socket ...
|
2005-08-10 05:14:34 +02:00
|
|
|
* Generate Reset(No Connection) unless P.type == Reset
|
|
|
|
* Drop packet and return
|
|
|
|
*/
|
|
|
|
if (dh->dccph_type != DCCP_PKT_RESET) {
|
2005-08-14 01:34:54 +02:00
|
|
|
DCCP_SKB_CB(skb)->dccpd_reset_code =
|
|
|
|
DCCP_RESET_CODE_NO_CONNECTION;
|
2006-11-15 04:07:45 +01:00
|
|
|
dccp_v4_ctl_send_reset(sk, skb);
|
2005-08-10 05:14:34 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
discard_it:
|
|
|
|
kfree_skb(skb);
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
discard_and_relse:
|
|
|
|
sock_put(sk);
|
|
|
|
goto discard_it;
|
|
|
|
}
|
|
|
|
|
2006-03-21 06:25:11 +01:00
|
|
|
static struct inet_connection_sock_af_ops dccp_ipv4_af_ops = {
|
2006-03-21 07:48:35 +01:00
|
|
|
.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),
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 13:06:04 +01:00
|
|
|
.bind_conflict = inet_csk_bind_conflict,
|
2006-03-21 07:45:21 +01:00
|
|
|
#ifdef CONFIG_COMPAT
|
2006-03-21 07:48:35 +01:00
|
|
|
.compat_setsockopt = compat_ip_setsockopt,
|
|
|
|
.compat_getsockopt = compat_ip_getsockopt,
|
2006-03-21 07:45:21 +01:00
|
|
|
#endif
|
2005-12-14 08:16:16 +01:00
|
|
|
};
|
|
|
|
|
2006-03-21 06:23:15 +01:00
|
|
|
static int dccp_v4_init_sock(struct sock *sk)
|
2005-08-10 05:14:34 +02:00
|
|
|
{
|
2006-03-21 07:00:37 +01:00
|
|
|
static __u8 dccp_v4_ctl_sock_initialized;
|
|
|
|
int err = dccp_init_sock(sk, dccp_v4_ctl_sock_initialized);
|
2005-12-14 08:24:16 +01:00
|
|
|
|
2006-03-21 07:00:37 +01:00
|
|
|
if (err == 0) {
|
|
|
|
if (unlikely(!dccp_v4_ctl_sock_initialized))
|
|
|
|
dccp_v4_ctl_sock_initialized = 1;
|
2006-03-21 06:23:15 +01:00
|
|
|
inet_csk(sk)->icsk_af_ops = &dccp_ipv4_af_ops;
|
2006-03-21 07:00:37 +01:00
|
|
|
}
|
|
|
|
|
2006-03-21 06:23:15 +01:00
|
|
|
return err;
|
2005-08-10 05:14:34 +02:00
|
|
|
}
|
|
|
|
|
2005-12-14 08:25:19 +01:00
|
|
|
static struct timewait_sock_ops dccp_timewait_sock_ops = {
|
|
|
|
.twsk_obj_size = sizeof(struct inet_timewait_sock),
|
|
|
|
};
|
|
|
|
|
2006-03-21 06:58:29 +01:00
|
|
|
static struct proto dccp_v4_prot = {
|
2005-08-10 05:14:34 +02:00
|
|
|
.name = "DCCP",
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.close = dccp_close,
|
|
|
|
.connect = dccp_v4_connect,
|
|
|
|
.disconnect = dccp_disconnect,
|
|
|
|
.ioctl = dccp_ioctl,
|
|
|
|
.init = dccp_v4_init_sock,
|
|
|
|
.setsockopt = dccp_setsockopt,
|
|
|
|
.getsockopt = dccp_getsockopt,
|
|
|
|
.sendmsg = dccp_sendmsg,
|
|
|
|
.recvmsg = dccp_recvmsg,
|
|
|
|
.backlog_rcv = dccp_v4_do_rcv,
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 13:06:04 +01:00
|
|
|
.hash = inet_hash,
|
|
|
|
.unhash = inet_unhash,
|
2005-08-10 05:14:34 +02:00
|
|
|
.accept = inet_csk_accept,
|
[SOCK] proto: Add hashinfo member to struct proto
This way we can remove TCP and DCCP specific versions of
sk->sk_prot->get_port: both v4 and v6 use inet_csk_get_port
sk->sk_prot->hash: inet_hash is directly used, only v6 need
a specific version to deal with mapped sockets
sk->sk_prot->unhash: both v4 and v6 use inet_hash directly
struct inet_connection_sock_af_ops also gets a new member, bind_conflict, so
that inet_csk_get_port can find the per family routine.
Now only the lookup routines receive as a parameter a struct inet_hashtable.
With this we further reuse code, reducing the difference among INET transport
protocols.
Eventually work has to be done on UDP and SCTP to make them share this
infrastructure and get as a bonus inet_diag interfaces so that iproute can be
used with these protocols.
net-2.6/net/ipv4/inet_hashtables.c:
struct proto | +8
struct inet_connection_sock_af_ops | +8
2 structs changed
__inet_hash_nolisten | +18
__inet_hash | -210
inet_put_port | +8
inet_bind_bucket_create | +1
__inet_hash_connect | -8
5 functions changed, 27 bytes added, 218 bytes removed, diff: -191
net-2.6/net/core/sock.c:
proto_seq_show | +3
1 function changed, 3 bytes added, diff: +3
net-2.6/net/ipv4/inet_connection_sock.c:
inet_csk_get_port | +15
1 function changed, 15 bytes added, diff: +15
net-2.6/net/ipv4/tcp.c:
tcp_set_state | -7
1 function changed, 7 bytes removed, diff: -7
net-2.6/net/ipv4/tcp_ipv4.c:
tcp_v4_get_port | -31
tcp_v4_hash | -48
tcp_v4_destroy_sock | -7
tcp_v4_syn_recv_sock | -2
tcp_unhash | -179
5 functions changed, 267 bytes removed, diff: -267
net-2.6/net/ipv6/inet6_hashtables.c:
__inet6_hash | +8
1 function changed, 8 bytes added, diff: +8
net-2.6/net/ipv4/inet_hashtables.c:
inet_unhash | +190
inet_hash | +242
2 functions changed, 432 bytes added, diff: +432
vmlinux:
16 functions changed, 485 bytes added, 492 bytes removed, diff: -7
/home/acme/git/net-2.6/net/ipv6/tcp_ipv6.c:
tcp_v6_get_port | -31
tcp_v6_hash | -7
tcp_v6_syn_recv_sock | -9
3 functions changed, 47 bytes removed, diff: -47
/home/acme/git/net-2.6/net/dccp/proto.c:
dccp_destroy_sock | -7
dccp_unhash | -179
dccp_hash | -49
dccp_set_state | -7
dccp_done | +1
5 functions changed, 1 bytes added, 242 bytes removed, diff: -241
/home/acme/git/net-2.6/net/dccp/ipv4.c:
dccp_v4_get_port | -31
dccp_v4_request_recv_sock | -2
2 functions changed, 33 bytes removed, diff: -33
/home/acme/git/net-2.6/net/dccp/ipv6.c:
dccp_v6_get_port | -31
dccp_v6_hash | -7
dccp_v6_request_recv_sock | +5
3 functions changed, 5 bytes added, 38 bytes removed, diff: -33
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2008-02-03 13:06:04 +01:00
|
|
|
.get_port = inet_csk_get_port,
|
2005-08-10 05:14:34 +02:00
|
|
|
.shutdown = dccp_shutdown,
|
2006-03-21 06:23:15 +01:00
|
|
|
.destroy = dccp_destroy_sock,
|
2005-08-10 05:14:34 +02:00
|
|
|
.orphan_count = &dccp_orphan_count,
|
|
|
|
.max_header = MAX_DCCP_HEADER,
|
|
|
|
.obj_size = sizeof(struct dccp_sock),
|
|
|
|
.rsk_prot = &dccp_request_sock_ops,
|
2005-12-14 08:25:19 +01:00
|
|
|
.twsk_prot = &dccp_timewait_sock_ops,
|
2008-03-23 00:50:58 +01:00
|
|
|
.h.hashinfo = &dccp_hashinfo,
|
2006-03-21 07:48:35 +01:00
|
|
|
#ifdef CONFIG_COMPAT
|
|
|
|
.compat_setsockopt = compat_dccp_setsockopt,
|
|
|
|
.compat_getsockopt = compat_dccp_getsockopt,
|
|
|
|
#endif
|
2005-08-10 05:14:34 +02:00
|
|
|
};
|
2005-12-14 08:25:19 +01:00
|
|
|
|
2006-03-21 06:25:11 +01:00
|
|
|
static struct net_protocol dccp_v4_protocol = {
|
|
|
|
.handler = dccp_v4_rcv,
|
|
|
|
.err_handler = dccp_v4_err,
|
|
|
|
.no_policy = 1,
|
2008-04-14 07:31:05 +02:00
|
|
|
.netns_ok = 1,
|
2006-03-21 06:25:11 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
static const struct proto_ops inet_dccp_ops = {
|
2006-03-21 07:48:35 +01:00
|
|
|
.family = PF_INET,
|
|
|
|
.owner = THIS_MODULE,
|
|
|
|
.release = inet_release,
|
|
|
|
.bind = inet_bind,
|
|
|
|
.connect = inet_stream_connect,
|
|
|
|
.socketpair = sock_no_socketpair,
|
|
|
|
.accept = inet_accept,
|
|
|
|
.getname = inet_getname,
|
2006-03-21 06:25:11 +01:00
|
|
|
/* FIXME: work on tcp_poll to rename it to inet_csk_poll */
|
2006-03-21 07:48:35 +01:00
|
|
|
.poll = dccp_poll,
|
|
|
|
.ioctl = inet_ioctl,
|
2006-03-21 06:25:11 +01:00
|
|
|
/* FIXME: work on inet_listen to rename it to sock_common_listen */
|
2006-03-21 07:48:35 +01:00
|
|
|
.listen = inet_dccp_listen,
|
|
|
|
.shutdown = inet_shutdown,
|
|
|
|
.setsockopt = sock_common_setsockopt,
|
|
|
|
.getsockopt = sock_common_getsockopt,
|
|
|
|
.sendmsg = inet_sendmsg,
|
|
|
|
.recvmsg = sock_common_recvmsg,
|
|
|
|
.mmap = sock_no_mmap,
|
|
|
|
.sendpage = sock_no_sendpage,
|
2006-03-21 07:45:21 +01:00
|
|
|
#ifdef CONFIG_COMPAT
|
2006-03-21 07:48:35 +01:00
|
|
|
.compat_setsockopt = compat_sock_common_setsockopt,
|
|
|
|
.compat_getsockopt = compat_sock_common_getsockopt,
|
2006-03-21 07:45:21 +01:00
|
|
|
#endif
|
2006-03-21 06:25:11 +01:00
|
|
|
};
|
|
|
|
|
|
|
|
static struct inet_protosw dccp_v4_protosw = {
|
|
|
|
.type = SOCK_DCCP,
|
|
|
|
.protocol = IPPROTO_DCCP,
|
|
|
|
.prot = &dccp_v4_prot,
|
|
|
|
.ops = &inet_dccp_ops,
|
|
|
|
.capability = -1,
|
|
|
|
.no_check = 0,
|
|
|
|
.flags = INET_PROTOSW_ICSK,
|
|
|
|
};
|
|
|
|
|
2008-04-14 07:29:13 +02:00
|
|
|
static int dccp_v4_init_net(struct net *net)
|
|
|
|
{
|
2008-04-14 07:29:59 +02:00
|
|
|
int err;
|
|
|
|
|
|
|
|
err = inet_ctl_sock_create(&net->dccp.v4_ctl_sk, PF_INET,
|
|
|
|
SOCK_DCCP, IPPROTO_DCCP, net);
|
|
|
|
return err;
|
2008-04-14 07:29:13 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static void dccp_v4_exit_net(struct net *net)
|
|
|
|
{
|
2008-04-14 07:29:59 +02:00
|
|
|
inet_ctl_sock_destroy(net->dccp.v4_ctl_sk);
|
2008-04-14 07:29:13 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
static struct pernet_operations dccp_v4_ops = {
|
|
|
|
.init = dccp_v4_init_net,
|
|
|
|
.exit = dccp_v4_exit_net,
|
|
|
|
};
|
|
|
|
|
2006-03-21 06:25:11 +01:00
|
|
|
static int __init dccp_v4_init(void)
|
|
|
|
{
|
|
|
|
int err = proto_register(&dccp_v4_prot, 1);
|
|
|
|
|
|
|
|
if (err != 0)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
err = inet_add_protocol(&dccp_v4_protocol, IPPROTO_DCCP);
|
|
|
|
if (err != 0)
|
|
|
|
goto out_proto_unregister;
|
|
|
|
|
|
|
|
inet_register_protosw(&dccp_v4_protosw);
|
|
|
|
|
2008-04-14 07:29:13 +02:00
|
|
|
err = register_pernet_subsys(&dccp_v4_ops);
|
|
|
|
if (err)
|
|
|
|
goto out_destroy_ctl_sock;
|
2006-03-21 06:25:11 +01:00
|
|
|
out:
|
|
|
|
return err;
|
2008-04-14 07:29:13 +02:00
|
|
|
out_destroy_ctl_sock:
|
2006-03-21 06:25:11 +01:00
|
|
|
inet_unregister_protosw(&dccp_v4_protosw);
|
|
|
|
inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP);
|
|
|
|
out_proto_unregister:
|
|
|
|
proto_unregister(&dccp_v4_prot);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __exit dccp_v4_exit(void)
|
|
|
|
{
|
2008-04-14 07:29:13 +02:00
|
|
|
unregister_pernet_subsys(&dccp_v4_ops);
|
2006-03-21 06:25:11 +01:00
|
|
|
inet_unregister_protosw(&dccp_v4_protosw);
|
|
|
|
inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP);
|
|
|
|
proto_unregister(&dccp_v4_prot);
|
|
|
|
}
|
|
|
|
|
|
|
|
module_init(dccp_v4_init);
|
|
|
|
module_exit(dccp_v4_exit);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* __stringify doesn't likes enums, so use SOCK_DCCP (6) and IPPROTO_DCCP (33)
|
|
|
|
* values directly, Also cover the case where the protocol is not specified,
|
|
|
|
* i.e. net-pf-PF_INET-proto-0-type-SOCK_DCCP
|
|
|
|
*/
|
2007-10-22 01:45:03 +02:00
|
|
|
MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 33, 6);
|
|
|
|
MODULE_ALIAS_NET_PF_PROTO_TYPE(PF_INET, 0, 6);
|
2006-03-21 06:25:11 +01:00
|
|
|
MODULE_LICENSE("GPL");
|
|
|
|
MODULE_AUTHOR("Arnaldo Carvalho de Melo <acme@mandriva.com>");
|
|
|
|
MODULE_DESCRIPTION("DCCP - Datagram Congestion Controlled Protocol");
|