2005-08-10 05:14:34 +02:00
|
|
|
/*
|
|
|
|
* net/dccp/minisocks.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/skbuff.h>
|
|
|
|
#include <linux/timer.h>
|
|
|
|
|
|
|
|
#include <net/sock.h>
|
|
|
|
#include <net/xfrm.h>
|
|
|
|
#include <net/inet_timewait_sock.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
|
|
|
|
2005-08-10 05:45:21 +02:00
|
|
|
struct inet_timewait_death_row dccp_death_row = {
|
|
|
|
.sysctl_max_tw_buckets = NR_FILE * 2,
|
|
|
|
.period = DCCP_TIMEWAIT_LEN / INET_TWDR_TWKILL_SLOTS,
|
|
|
|
.death_lock = SPIN_LOCK_UNLOCKED,
|
|
|
|
.hashinfo = &dccp_hashinfo,
|
|
|
|
.tw_timer = TIMER_INITIALIZER(inet_twdr_hangman, 0,
|
|
|
|
(unsigned long)&dccp_death_row),
|
|
|
|
.twkill_work = __WORK_INITIALIZER(dccp_death_row.twkill_work,
|
|
|
|
inet_twdr_twkill_work,
|
|
|
|
&dccp_death_row),
|
|
|
|
/* Short-time timewait calendar */
|
|
|
|
|
|
|
|
.twcal_hand = -1,
|
|
|
|
.twcal_timer = TIMER_INITIALIZER(inet_twdr_twcal_tick, 0,
|
|
|
|
(unsigned long)&dccp_death_row),
|
|
|
|
};
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_death_row);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
void dccp_time_wait(struct sock *sk, int state, int timeo)
|
|
|
|
{
|
2005-08-10 05:45:21 +02:00
|
|
|
struct inet_timewait_sock *tw = NULL;
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2005-08-10 05:45:21 +02:00
|
|
|
if (dccp_death_row.tw_count < dccp_death_row.sysctl_max_tw_buckets)
|
|
|
|
tw = inet_twsk_alloc(sk, state);
|
|
|
|
|
|
|
|
if (tw != NULL) {
|
|
|
|
const struct inet_connection_sock *icsk = inet_csk(sk);
|
|
|
|
const int rto = (icsk->icsk_rto << 2) - (icsk->icsk_rto >> 1);
|
2005-12-14 08:24:53 +01:00
|
|
|
#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
|
|
|
|
if (tw->tw_family == PF_INET6) {
|
|
|
|
const struct ipv6_pinfo *np = inet6_sk(sk);
|
|
|
|
struct inet6_timewait_sock *tw6;
|
2005-08-10 05:45:21 +02:00
|
|
|
|
2005-12-14 08:24:53 +01:00
|
|
|
tw->tw_ipv6_offset = inet6_tw_offset(sk->sk_prot);
|
|
|
|
tw6 = inet6_twsk((struct sock *)tw);
|
|
|
|
ipv6_addr_copy(&tw6->tw_v6_daddr, &np->daddr);
|
|
|
|
ipv6_addr_copy(&tw6->tw_v6_rcv_saddr, &np->rcv_saddr);
|
|
|
|
tw->tw_ipv6only = np->ipv6only;
|
|
|
|
}
|
|
|
|
#endif
|
2005-08-10 05:45:21 +02:00
|
|
|
/* Linkage updates. */
|
|
|
|
__inet_twsk_hashdance(tw, sk, &dccp_hashinfo);
|
|
|
|
|
|
|
|
/* Get the TIME_WAIT timeout firing. */
|
|
|
|
if (timeo < rto)
|
|
|
|
timeo = rto;
|
|
|
|
|
|
|
|
tw->tw_timeout = DCCP_TIMEWAIT_LEN;
|
|
|
|
if (state == DCCP_TIME_WAIT)
|
|
|
|
timeo = DCCP_TIMEWAIT_LEN;
|
|
|
|
|
|
|
|
inet_twsk_schedule(tw, &dccp_death_row, timeo,
|
|
|
|
DCCP_TIMEWAIT_LEN);
|
|
|
|
inet_twsk_put(tw);
|
|
|
|
} else {
|
|
|
|
/* Sorry, if we're out of memory, just CLOSE this
|
|
|
|
* socket up. We've got bigger problems than
|
|
|
|
* non-graceful socket closings.
|
|
|
|
*/
|
2005-08-14 01:34:54 +02:00
|
|
|
LIMIT_NETDEBUG(KERN_INFO "DCCP: time wait bucket "
|
|
|
|
"table overflow\n");
|
2005-08-10 05:45:21 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
dccp_done(sk);
|
2005-08-10 05:14:34 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
struct sock *dccp_create_openreq_child(struct sock *sk,
|
|
|
|
const struct request_sock *req,
|
|
|
|
const struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
/*
|
|
|
|
* Step 3: Process LISTEN state
|
|
|
|
*
|
|
|
|
* // Generate a new socket and switch to that socket
|
|
|
|
* Set S := new socket for this port pair
|
|
|
|
*/
|
|
|
|
struct sock *newsk = inet_csk_clone(sk, req, GFP_ATOMIC);
|
|
|
|
|
|
|
|
if (newsk != NULL) {
|
|
|
|
const struct dccp_request_sock *dreq = dccp_rsk(req);
|
|
|
|
struct inet_connection_sock *newicsk = inet_csk(sk);
|
|
|
|
struct dccp_sock *newdp = dccp_sk(newsk);
|
2006-03-21 07:50:58 +01:00
|
|
|
struct dccp_minisock *newdmsk = dccp_msk(newsk);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
2005-09-17 01:58:40 +02:00
|
|
|
newdp->dccps_role = DCCP_ROLE_SERVER;
|
2005-09-18 09:17:51 +02:00
|
|
|
newdp->dccps_hc_rx_ackvec = NULL;
|
2005-09-17 01:58:40 +02:00
|
|
|
newdp->dccps_service_list = NULL;
|
|
|
|
newdp->dccps_service = dreq->dreq_service;
|
|
|
|
newicsk->icsk_rto = DCCP_TIMEOUT_INIT;
|
2005-09-09 07:38:35 +02:00
|
|
|
do_gettimeofday(&newdp->dccps_epoch);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
[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
|
|
|
if (dccp_feat_clone(sk, newsk))
|
|
|
|
goto out_free;
|
|
|
|
|
2006-03-21 07:50:58 +01:00
|
|
|
if (newdmsk->dccpms_send_ack_vector) {
|
2005-09-18 09:17:51 +02:00
|
|
|
newdp->dccps_hc_rx_ackvec =
|
2006-03-21 02:15:42 +01:00
|
|
|
dccp_ackvec_alloc(GFP_ATOMIC);
|
2005-09-18 09:17:51 +02:00
|
|
|
if (unlikely(newdp->dccps_hc_rx_ackvec == NULL))
|
2005-08-10 05:14:34 +02:00
|
|
|
goto out_free;
|
|
|
|
}
|
|
|
|
|
2006-03-21 04:21:44 +01:00
|
|
|
newdp->dccps_hc_rx_ccid =
|
2006-03-21 07:50:58 +01:00
|
|
|
ccid_hc_rx_new(newdmsk->dccpms_rx_ccid,
|
2006-03-21 04:21:44 +01:00
|
|
|
newsk, GFP_ATOMIC);
|
|
|
|
newdp->dccps_hc_tx_ccid =
|
2006-03-21 07:50:58 +01:00
|
|
|
ccid_hc_tx_new(newdmsk->dccpms_tx_ccid,
|
2006-03-21 04:21:44 +01:00
|
|
|
newsk, GFP_ATOMIC);
|
|
|
|
if (unlikely(newdp->dccps_hc_rx_ccid == NULL ||
|
|
|
|
newdp->dccps_hc_tx_ccid == NULL)) {
|
2005-09-18 09:17:51 +02:00
|
|
|
dccp_ackvec_free(newdp->dccps_hc_rx_ackvec);
|
2006-03-21 04:21:44 +01:00
|
|
|
ccid_hc_rx_delete(newdp->dccps_hc_rx_ccid, newsk);
|
|
|
|
ccid_hc_tx_delete(newdp->dccps_hc_tx_ccid, newsk);
|
2005-08-10 05:14:34 +02:00
|
|
|
out_free:
|
|
|
|
/* It is still raw copy of parent, so invalidate
|
|
|
|
* destructor and make plain sk_free() */
|
|
|
|
newsk->sk_destruct = NULL;
|
|
|
|
sk_free(newsk);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Step 3: Process LISTEN state
|
|
|
|
*
|
|
|
|
* Choose S.ISS (initial seqno) or set from Init Cookie
|
2005-08-14 01:34:54 +02:00
|
|
|
* Set S.ISR, S.GSR, S.SWL, S.SWH from packet or Init
|
|
|
|
* Cookie
|
2005-08-10 05:14:34 +02:00
|
|
|
*/
|
|
|
|
|
|
|
|
/* See dccp_v4_conn_request */
|
2006-03-21 07:50:58 +01:00
|
|
|
newdmsk->dccpms_sequence_window = req->rcv_wnd;
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
newdp->dccps_gar = newdp->dccps_isr = dreq->dreq_isr;
|
|
|
|
dccp_update_gsr(newsk, dreq->dreq_isr);
|
|
|
|
|
|
|
|
newdp->dccps_iss = dreq->dreq_iss;
|
|
|
|
dccp_update_gss(newsk, dreq->dreq_iss);
|
|
|
|
|
2005-08-21 10:36:45 +02:00
|
|
|
/*
|
|
|
|
* SWL and AWL are initially adjusted so that they are not less than
|
|
|
|
* the initial Sequence Numbers received and sent, respectively:
|
|
|
|
* SWL := max(GSR + 1 - floor(W/4), ISR),
|
|
|
|
* AWL := max(GSS - W' + 1, ISS).
|
|
|
|
* These adjustments MUST be applied only at the beginning of the
|
|
|
|
* connection.
|
|
|
|
*/
|
|
|
|
dccp_set_seqno(&newdp->dccps_swl,
|
|
|
|
max48(newdp->dccps_swl, newdp->dccps_isr));
|
|
|
|
dccp_set_seqno(&newdp->dccps_awl,
|
|
|
|
max48(newdp->dccps_awl, newdp->dccps_iss));
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
dccp_init_xmit_timers(newsk);
|
|
|
|
|
|
|
|
DCCP_INC_STATS_BH(DCCP_MIB_PASSIVEOPENS);
|
|
|
|
}
|
|
|
|
return newsk;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_create_openreq_child);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
/*
|
|
|
|
* Process an incoming packet for RESPOND sockets represented
|
|
|
|
* as an request_sock.
|
|
|
|
*/
|
|
|
|
struct sock *dccp_check_req(struct sock *sk, struct sk_buff *skb,
|
|
|
|
struct request_sock *req,
|
|
|
|
struct request_sock **prev)
|
|
|
|
{
|
|
|
|
struct sock *child = NULL;
|
|
|
|
|
|
|
|
/* Check for retransmitted REQUEST */
|
|
|
|
if (dccp_hdr(skb)->dccph_type == DCCP_PKT_REQUEST) {
|
2005-08-14 01:34:54 +02:00
|
|
|
if (after48(DCCP_SKB_CB(skb)->dccpd_seq,
|
|
|
|
dccp_rsk(req)->dreq_isr)) {
|
2005-08-10 05:14:34 +02:00
|
|
|
struct dccp_request_sock *dreq = dccp_rsk(req);
|
|
|
|
|
|
|
|
dccp_pr_debug("Retransmitted REQUEST\n");
|
|
|
|
/* Send another RESPONSE packet */
|
|
|
|
dccp_set_seqno(&dreq->dreq_iss, dreq->dreq_iss + 1);
|
2005-08-14 01:34:54 +02:00
|
|
|
dccp_set_seqno(&dreq->dreq_isr,
|
|
|
|
DCCP_SKB_CB(skb)->dccpd_seq);
|
2005-08-10 05:14:34 +02:00
|
|
|
req->rsk_ops->rtx_syn_ack(sk, req, NULL);
|
|
|
|
}
|
|
|
|
/* Network Duplicate, discard packet */
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_PACKET_ERROR;
|
|
|
|
|
|
|
|
if (dccp_hdr(skb)->dccph_type != DCCP_PKT_ACK &&
|
|
|
|
dccp_hdr(skb)->dccph_type != DCCP_PKT_DATAACK)
|
|
|
|
goto drop;
|
|
|
|
|
|
|
|
/* Invalid ACK */
|
|
|
|
if (DCCP_SKB_CB(skb)->dccpd_ack_seq != dccp_rsk(req)->dreq_iss) {
|
2005-08-14 01:34:54 +02:00
|
|
|
dccp_pr_debug("Invalid ACK number: ack_seq=%llu, "
|
|
|
|
"dreq_iss=%llu\n",
|
2005-08-10 05:27:14 +02:00
|
|
|
(unsigned long long)
|
|
|
|
DCCP_SKB_CB(skb)->dccpd_ack_seq,
|
|
|
|
(unsigned long long)
|
|
|
|
dccp_rsk(req)->dreq_iss);
|
2005-08-10 05:14:34 +02:00
|
|
|
goto drop;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:16:16 +01:00
|
|
|
child = inet_csk(sk)->icsk_af_ops->syn_recv_sock(sk, skb, req, NULL);
|
2005-08-10 05:14:34 +02:00
|
|
|
if (child == NULL)
|
|
|
|
goto listen_overflow;
|
|
|
|
|
|
|
|
/* FIXME: deal with options */
|
|
|
|
|
|
|
|
inet_csk_reqsk_queue_unlink(sk, req, prev);
|
|
|
|
inet_csk_reqsk_queue_removed(sk, req);
|
|
|
|
inet_csk_reqsk_queue_add(sk, req, child);
|
|
|
|
out:
|
|
|
|
return child;
|
|
|
|
listen_overflow:
|
|
|
|
dccp_pr_debug("listen_overflow!\n");
|
|
|
|
DCCP_SKB_CB(skb)->dccpd_reset_code = DCCP_RESET_CODE_TOO_BUSY;
|
|
|
|
drop:
|
|
|
|
if (dccp_hdr(skb)->dccph_type != DCCP_PKT_RESET)
|
|
|
|
req->rsk_ops->send_reset(skb);
|
|
|
|
|
|
|
|
inet_csk_reqsk_queue_drop(sk, req, prev);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2005-12-14 08:24:16 +01:00
|
|
|
EXPORT_SYMBOL_GPL(dccp_check_req);
|
|
|
|
|
2005-08-10 05:14:34 +02:00
|
|
|
/*
|
|
|
|
* Queue segment on the new socket if the new socket is active,
|
|
|
|
* otherwise we just shortcircuit this and continue with
|
|
|
|
* the new socket.
|
|
|
|
*/
|
|
|
|
int dccp_child_process(struct sock *parent, struct sock *child,
|
|
|
|
struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
const int state = child->sk_state;
|
|
|
|
|
|
|
|
if (!sock_owned_by_user(child)) {
|
2005-08-14 01:34:54 +02:00
|
|
|
ret = dccp_rcv_state_process(child, skb, dccp_hdr(skb),
|
|
|
|
skb->len);
|
2005-08-10 05:14:34 +02:00
|
|
|
|
|
|
|
/* Wakeup parent, send SIGIO */
|
|
|
|
if (state == DCCP_RESPOND && child->sk_state != state)
|
|
|
|
parent->sk_data_ready(parent, 0);
|
|
|
|
} else {
|
|
|
|
/* Alas, it is possible again, because we do lookup
|
|
|
|
* in main socket hash table and lock on listening
|
|
|
|
* socket does not protect us more.
|
|
|
|
*/
|
|
|
|
sk_add_backlog(child, skb);
|
|
|
|
}
|
|
|
|
|
|
|
|
bh_unlock_sock(child);
|
|
|
|
sock_put(child);
|
|
|
|
return ret;
|
|
|
|
}
|
2005-12-14 08:24:16 +01:00
|
|
|
|
|
|
|
EXPORT_SYMBOL_GPL(dccp_child_process);
|