-/* SCTP kernel reference Implementation
+/* SCTP kernel implementation
* Copyright (c) 1999-2000 Cisco, Inc.
* Copyright (c) 1999-2001 Motorola, Inc.
* Copyright (c) 2001-2003 International Business Machines Corp.
* Copyright (c) 2001 Intel Corp.
* Copyright (c) 2001 La Monte H.P. Yarroll
*
- * This file is part of the SCTP kernel reference Implementation
+ * This file is part of the SCTP kernel implementation
*
* This module provides the abstraction for an SCTP tranport representing
* a remote transport address. For local transport addresses, we just use
* union sctp_addr.
*
- * The SCTP reference implementation is free software;
+ * This SCTP implementation 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, or (at your option)
* any later version.
*
- * The SCTP reference implementation is distributed in the hope that it
+ * This SCTP implementation is distributed in the hope that it
* will be useful, but WITHOUT ANY WARRANTY; without even the implied
* ************************
* warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
* be incorporated into the next SCTP release.
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/slab.h>
#include <linux/types.h>
+#include <linux/random.h>
#include <net/sctp/sctp.h>
#include <net/sctp/sm.h>
/* Initialize a new transport from provided memory. */
static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer,
const union sctp_addr *addr,
- int gfp)
+ gfp_t gfp)
{
/* Copy in the address. */
peer->ipaddr = *addr;
peer->af_specific = sctp_get_af_specific(addr->sa.sa_family);
- peer->asoc = NULL;
-
- peer->dst = NULL;
memset(&peer->saddr, 0, sizeof(union sctp_addr));
/* From 6.3.1 RTO Calculation:
* given destination transport address, set RTO to the protocol
* parameter 'RTO.Initial'.
*/
- peer->rtt = 0;
- peer->rto = sctp_rto_initial;
- peer->rttvar = 0;
- peer->srtt = 0;
- peer->rto_pending = 0;
+ peer->rto = msecs_to_jiffies(sctp_rto_initial);
peer->last_time_heard = jiffies;
- peer->last_time_used = jiffies;
peer->last_time_ecne_reduced = jiffies;
- peer->init_sent_count = 0;
-
- peer->state = SCTP_ACTIVE;
- peer->hb_allowed = 0;
+ peer->param_flags = SPP_HB_DISABLE |
+ SPP_PMTUD_ENABLE |
+ SPP_SACKDELAY_ENABLE;
/* Initialize the default path max_retrans. */
- peer->max_retrans = sctp_max_retrans_path;
- peer->error_count = 0;
+ peer->pathmaxrxt = sctp_max_retrans_path;
INIT_LIST_HEAD(&peer->transmitted);
INIT_LIST_HEAD(&peer->send_ready);
INIT_LIST_HEAD(&peer->transports);
- /* Set up the retransmission timer. */
- init_timer(&peer->T3_rtx_timer);
- peer->T3_rtx_timer.function = sctp_generate_t3_rtx_event;
- peer->T3_rtx_timer.data = (unsigned long)peer;
+ setup_timer(&peer->T3_rtx_timer, sctp_generate_t3_rtx_event,
+ (unsigned long)peer);
+ setup_timer(&peer->hb_timer, sctp_generate_heartbeat_event,
+ (unsigned long)peer);
+ setup_timer(&peer->proto_unreach_timer,
+ sctp_generate_proto_unreach_event, (unsigned long)peer);
- /* Set up the heartbeat timer. */
- init_timer(&peer->hb_timer);
- peer->hb_timer.function = sctp_generate_heartbeat_event;
- peer->hb_timer.data = (unsigned long)peer;
+ /* Initialize the 64-bit random nonce sent with heartbeat. */
+ get_random_bytes(&peer->hb_nonce, sizeof(peer->hb_nonce));
atomic_set(&peer->refcnt, 1);
- peer->dead = 0;
-
- peer->malloced = 0;
-
- /* Initialize the state information for SFR-CACC */
- peer->cacc.changeover_active = 0;
- peer->cacc.cycling_changeover = 0;
- peer->cacc.next_tsn_at_change = 0;
- peer->cacc.cacc_saw_newack = 0;
return peer;
}
/* Allocate and initialize a new transport. */
-struct sctp_transport *sctp_transport_new(const union sctp_addr *addr, int gfp)
+struct sctp_transport *sctp_transport_new(const union sctp_addr *addr,
+ gfp_t gfp)
{
- struct sctp_transport *transport;
+ struct sctp_transport *transport;
- transport = t_new(struct sctp_transport, gfp);
+ transport = t_new(struct sctp_transport, gfp);
if (!transport)
goto fail;
del_timer(&transport->T3_rtx_timer))
sctp_transport_put(transport);
+ /* Delete the ICMP proto unreachable timer if it's active. */
+ if (timer_pending(&transport->proto_unreach_timer) &&
+ del_timer(&transport->proto_unreach_timer))
+ sctp_association_put(transport->asoc);
sctp_transport_put(transport);
}
if (transport->asoc)
sctp_association_put(transport->asoc);
- sctp_packet_free(&transport->packet);
+ sctp_packet_free(&transport->packet);
dst_release(transport->dst);
kfree(transport);
}
/* Initialize the pmtu of a transport. */
-void sctp_transport_pmtu(struct sctp_transport *transport)
+void sctp_transport_pmtu(struct sctp_transport *transport, struct sock *sk)
{
struct dst_entry *dst;
+ struct flowi fl;
- dst = transport->af_specific->get_dst(NULL, &transport->ipaddr, NULL);
+ dst = transport->af_specific->get_dst(transport->asoc,
+ &transport->ipaddr,
+ &transport->saddr,
+ &fl, sk);
if (dst) {
- transport->pmtu = dst_mtu(dst);
+ transport->pathmtu = dst_mtu(dst);
dst_release(dst);
} else
- transport->pmtu = SCTP_DEFAULT_MAXSEGMENT;
+ transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT;
+}
+
+/* this is a complete rip-off from __sk_dst_check
+ * the cookie is always 0 since this is how it's used in the
+ * pmtu code
+ */
+static struct dst_entry *sctp_transport_dst_check(struct sctp_transport *t)
+{
+ struct dst_entry *dst = t->dst;
+
+ if (dst && dst->obsolete && dst->ops->check(dst, 0) == NULL) {
+ dst_release(t->dst);
+ t->dst = NULL;
+ return NULL;
+ }
+
+ return dst;
+}
+
+void sctp_transport_update_pmtu(struct sctp_transport *t, u32 pmtu)
+{
+ struct dst_entry *dst;
+
+ if (unlikely(pmtu < SCTP_DEFAULT_MINSEGMENT)) {
+ pr_warn("%s: Reported pmtu %d too low, using default minimum of %d\n",
+ __func__, pmtu,
+ SCTP_DEFAULT_MINSEGMENT);
+ /* Use default minimum segment size and disable
+ * pmtu discovery on this transport.
+ */
+ t->pathmtu = SCTP_DEFAULT_MINSEGMENT;
+ } else {
+ t->pathmtu = pmtu;
+ }
+
+ dst = sctp_transport_dst_check(t);
+ if (dst)
+ dst->ops->update_pmtu(dst, pmtu);
}
/* Caches the dst entry and source address for a transport's destination
struct sctp_af *af = transport->af_specific;
union sctp_addr *daddr = &transport->ipaddr;
struct dst_entry *dst;
+ struct flowi fl;
- dst = af->get_dst(asoc, daddr, saddr);
+ dst = af->get_dst(asoc, daddr, saddr, &fl, sctp_opt2sk(opt));
+ transport->dst = dst;
if (saddr)
memcpy(&transport->saddr, saddr, sizeof(union sctp_addr));
else
- af->get_saddr(asoc, dst, daddr, &transport->saddr);
+ af->get_saddr(opt, transport, &fl);
- transport->dst = dst;
+ if ((transport->param_flags & SPP_PMTUD_DISABLE) && transport->pathmtu) {
+ return;
+ }
if (dst) {
- transport->pmtu = dst_mtu(dst);
+ transport->pathmtu = dst_mtu(dst);
/* Initialize sk->sk_rcv_saddr, if the transport is the
* association's active path for getsockname().
- */
- if (asoc && (transport == asoc->peer.active_path))
- af->to_sk_saddr(&transport->saddr, asoc->base.sk);
+ */
+ if (asoc && (!asoc->peer.primary_path ||
+ (transport == asoc->peer.active_path)))
+ opt->pf->af->to_sk_saddr(&transport->saddr,
+ asoc->base.sk);
} else
- transport->pmtu = SCTP_DEFAULT_MAXSEGMENT;
+ transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT;
}
/* Hold a reference to a transport. */
tp->rto_pending = 0;
SCTP_DEBUG_PRINTK("%s: transport: %p, rtt: %d, srtt: %d "
- "rttvar: %d, rto: %d\n", __FUNCTION__,
+ "rttvar: %d, rto: %ld\n", __func__,
tp, rtt, tp->srtt, tp->rttvar, tp->rto);
}
void sctp_transport_raise_cwnd(struct sctp_transport *transport,
__u32 sack_ctsn, __u32 bytes_acked)
{
+ struct sctp_association *asoc = transport->asoc;
__u32 cwnd, ssthresh, flight_size, pba, pmtu;
cwnd = transport->cwnd;
flight_size = transport->flight_size;
+ /* See if we need to exit Fast Recovery first */
+ if (asoc->fast_recovery &&
+ TSN_lte(asoc->fast_recovery_exit, sack_ctsn))
+ asoc->fast_recovery = 0;
+
/* The appropriate cwnd increase algorithm is performed if, and only
- * if the cumulative TSN has advanced and the congestion window is
+ * if the cumulative TSN whould advanced and the congestion window is
* being fully utilized.
*/
- if ((transport->asoc->ctsn_ack_point >= sack_ctsn) ||
+ if (TSN_lte(sack_ctsn, transport->asoc->ctsn_ack_point) ||
(flight_size < cwnd))
return;
ssthresh = transport->ssthresh;
pba = transport->partial_bytes_acked;
- pmtu = transport->asoc->pmtu;
+ pmtu = transport->asoc->pathmtu;
if (cwnd <= ssthresh) {
- /* RFC 2960 7.2.1, sctpimpguide-05 2.14.2 When cwnd is less
- * than or equal to ssthresh an SCTP endpoint MUST use the
- * slow start algorithm to increase cwnd only if the current
- * congestion window is being fully utilized and an incoming
- * SACK advances the Cumulative TSN Ack Point. Only when these
- * two conditions are met can the cwnd be increased otherwise
- * the cwnd MUST not be increased. If these conditions are met
- * then cwnd MUST be increased by at most the lesser of
- * 1) the total size of the previously outstanding DATA
- * chunk(s) acknowledged, and 2) the destination's path MTU.
+ /* RFC 4960 7.2.1
+ * o When cwnd is less than or equal to ssthresh, an SCTP
+ * endpoint MUST use the slow-start algorithm to increase
+ * cwnd only if the current congestion window is being fully
+ * utilized, an incoming SACK advances the Cumulative TSN
+ * Ack Point, and the data sender is not in Fast Recovery.
+ * Only when these three conditions are met can the cwnd be
+ * increased; otherwise, the cwnd MUST not be increased.
+ * If these conditions are met, then cwnd MUST be increased
+ * by, at most, the lesser of 1) the total size of the
+ * previously outstanding DATA chunk(s) acknowledged, and
+ * 2) the destination's path MTU. This upper bound protects
+ * against the ACK-Splitting attack outlined in [SAVAGE99].
*/
+ if (asoc->fast_recovery)
+ return;
+
if (bytes_acked > pmtu)
cwnd += pmtu;
else
SCTP_DEBUG_PRINTK("%s: SLOW START: transport: %p, "
"bytes_acked: %d, cwnd: %d, ssthresh: %d, "
"flight_size: %d, pba: %d\n",
- __FUNCTION__,
+ __func__,
transport, bytes_acked, cwnd,
ssthresh, flight_size, pba);
} else {
SCTP_DEBUG_PRINTK("%s: CONGESTION AVOIDANCE: "
"transport: %p, bytes_acked: %d, cwnd: %d, "
"ssthresh: %d, flight_size: %d, pba: %d\n",
- __FUNCTION__,
+ __func__,
transport, bytes_acked, cwnd,
ssthresh, flight_size, pba);
}
void sctp_transport_lower_cwnd(struct sctp_transport *transport,
sctp_lower_cwnd_t reason)
{
+ struct sctp_association *asoc = transport->asoc;
+
switch (reason) {
case SCTP_LOWER_CWND_T3_RTX:
/* RFC 2960 Section 7.2.3, sctpimpguide
* partial_bytes_acked = 0
*/
transport->ssthresh = max(transport->cwnd/2,
- 4*transport->asoc->pmtu);
- transport->cwnd = transport->asoc->pmtu;
+ 4*asoc->pathmtu);
+ transport->cwnd = asoc->pathmtu;
+
+ /* T3-rtx also clears fast recovery */
+ asoc->fast_recovery = 0;
break;
case SCTP_LOWER_CWND_FAST_RTX:
* destination address(es) to which the missing DATA chunks
* were last sent, according to the formula described in
* Section 7.2.3.
- *
- * RFC 2960 7.2.3, sctpimpguide Upon detection of packet
+ *
+ * RFC 2960 7.2.3, sctpimpguide Upon detection of packet
* losses from SACK (see Section 7.2.4), An endpoint
* should do the following:
* ssthresh = max(cwnd/2, 4*MTU)
* cwnd = ssthresh
* partial_bytes_acked = 0
*/
+ if (asoc->fast_recovery)
+ return;
+
+ /* Mark Fast recovery */
+ asoc->fast_recovery = 1;
+ asoc->fast_recovery_exit = asoc->next_tsn - 1;
+
transport->ssthresh = max(transport->cwnd/2,
- 4*transport->asoc->pmtu);
+ 4*asoc->pathmtu);
transport->cwnd = transport->ssthresh;
break;
* congestion indications more than once every window of
* data (or more loosely more than once every round-trip time).
*/
- if ((jiffies - transport->last_time_ecne_reduced) >
- transport->rtt) {
+ if (time_after(jiffies, transport->last_time_ecne_reduced +
+ transport->rtt)) {
transport->ssthresh = max(transport->cwnd/2,
- 4*transport->asoc->pmtu);
+ 4*asoc->pathmtu);
transport->cwnd = transport->ssthresh;
transport->last_time_ecne_reduced = jiffies;
}
* to be done every RTO interval, we do it every hearbeat
* interval.
*/
- if ((jiffies - transport->last_time_used) > transport->rto)
- transport->cwnd = max(transport->cwnd/2,
- 4*transport->asoc->pmtu);
+ transport->cwnd = max(transport->cwnd/2,
+ 4*asoc->pathmtu);
break;
- };
+ }
transport->partial_bytes_acked = 0;
SCTP_DEBUG_PRINTK("%s: transport: %p reason: %d cwnd: "
- "%d ssthresh: %d\n", __FUNCTION__,
+ "%d ssthresh: %d\n", __func__,
transport, reason,
transport->cwnd, transport->ssthresh);
}
+/* Apply Max.Burst limit to the congestion window:
+ * sctpimpguide-05 2.14.2
+ * D) When the time comes for the sender to
+ * transmit new DATA chunks, the protocol parameter Max.Burst MUST
+ * first be applied to limit how many new DATA chunks may be sent.
+ * The limit is applied by adjusting cwnd as follows:
+ * if ((flightsize+ Max.Burst * MTU) < cwnd)
+ * cwnd = flightsize + Max.Burst * MTU
+ */
+
+void sctp_transport_burst_limited(struct sctp_transport *t)
+{
+ struct sctp_association *asoc = t->asoc;
+ u32 old_cwnd = t->cwnd;
+ u32 max_burst_bytes;
+
+ if (t->burst_limited)
+ return;
+
+ max_burst_bytes = t->flight_size + (asoc->max_burst * asoc->pathmtu);
+ if (max_burst_bytes < old_cwnd) {
+ t->cwnd = max_burst_bytes;
+ t->burst_limited = old_cwnd;
+ }
+}
+
+/* Restore the old cwnd congestion window, after the burst had it's
+ * desired effect.
+ */
+void sctp_transport_burst_reset(struct sctp_transport *t)
+{
+ if (t->burst_limited) {
+ t->cwnd = t->burst_limited;
+ t->burst_limited = 0;
+ }
+}
+
/* What is the next timeout value for this transport? */
unsigned long sctp_transport_timeout(struct sctp_transport *t)
{
unsigned long timeout;
- timeout = t->hb_interval + t->rto + sctp_jitter(t->rto);
+ timeout = t->rto + sctp_jitter(t->rto);
+ if (t->state != SCTP_UNCONFIRMED)
+ timeout += t->hbinterval;
timeout += jiffies;
return timeout;
}
+
+/* Reset transport variables to their initial values */
+void sctp_transport_reset(struct sctp_transport *t)
+{
+ struct sctp_association *asoc = t->asoc;
+
+ /* RFC 2960 (bis), Section 5.2.4
+ * All the congestion control parameters (e.g., cwnd, ssthresh)
+ * related to this peer MUST be reset to their initial values
+ * (see Section 6.2.1)
+ */
+ t->cwnd = min(4*asoc->pathmtu, max_t(__u32, 2*asoc->pathmtu, 4380));
+ t->burst_limited = 0;
+ t->ssthresh = asoc->peer.i.a_rwnd;
+ t->rto = asoc->rto_initial;
+ t->rtt = 0;
+ t->srtt = 0;
+ t->rttvar = 0;
+
+ /* Reset these additional varibles so that we have a clean
+ * slate.
+ */
+ t->partial_bytes_acked = 0;
+ t->flight_size = 0;
+ t->error_count = 0;
+ t->rto_pending = 0;
+ t->hb_sent = 0;
+
+ /* Initialize the state information for SFR-CACC */
+ t->cacc.changeover_active = 0;
+ t->cacc.cycling_changeover = 0;
+ t->cacc.next_tsn_at_change = 0;
+ t->cacc.cacc_saw_newack = 0;
+}
Code Review / linux-3.10.git / blobdiff