summaryrefslogtreecommitdiff
path: root/net/sctp/transport.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/sctp/transport.c')
-rw-r--r--net/sctp/transport.c222
1 files changed, 189 insertions, 33 deletions
diff --git a/net/sctp/transport.c b/net/sctp/transport.c
index 806af58f4375..0c56d9673cc1 100644
--- a/net/sctp/transport.c
+++ b/net/sctp/transport.c
@@ -8,7 +8,7 @@
*
* This file is part of the SCTP kernel implementation
*
- * This module provides the abstraction for an SCTP tranport representing
+ * This module provides the abstraction for an SCTP transport representing
* a remote transport address. For local transport addresses, we just use
* union sctp_addr.
*
@@ -37,10 +37,10 @@
/* 1st Level Abstractions. */
/* Initialize a new transport from provided memory. */
-static struct sctp_transport *sctp_transport_init(struct net *net,
- struct sctp_transport *peer,
- const union sctp_addr *addr,
- gfp_t gfp)
+static void sctp_transport_init(struct net *net,
+ struct sctp_transport *peer,
+ const union sctp_addr *addr,
+ gfp_t gfp)
{
/* Copy in the address. */
peer->af_specific = sctp_get_af_specific(addr->sa.sa_family);
@@ -75,6 +75,7 @@ static struct sctp_transport *sctp_transport_init(struct net *net,
timer_setup(&peer->T3_rtx_timer, sctp_generate_t3_rtx_event, 0);
timer_setup(&peer->hb_timer, sctp_generate_heartbeat_event, 0);
timer_setup(&peer->reconf_timer, sctp_generate_reconf_event, 0);
+ timer_setup(&peer->probe_timer, sctp_generate_probe_event, 0);
timer_setup(&peer->proto_unreach_timer,
sctp_generate_proto_unreach_event, 0);
@@ -82,8 +83,6 @@ static struct sctp_transport *sctp_transport_init(struct net *net,
get_random_bytes(&peer->hb_nonce, sizeof(peer->hb_nonce));
refcount_set(&peer->refcnt, 1);
-
- return peer;
}
/* Allocate and initialize a new transport. */
@@ -95,20 +94,13 @@ struct sctp_transport *sctp_transport_new(struct net *net,
transport = kzalloc(sizeof(*transport), gfp);
if (!transport)
- goto fail;
+ return NULL;
- if (!sctp_transport_init(net, transport, addr, gfp))
- goto fail_init;
+ sctp_transport_init(net, transport, addr, gfp);
SCTP_DBG_OBJCNT_INC(transport);
return transport;
-
-fail_init:
- kfree(transport);
-
-fail:
- return NULL;
}
/* This transport is no longer needed. Free up if possible, or
@@ -116,24 +108,29 @@ fail:
*/
void sctp_transport_free(struct sctp_transport *transport)
{
+ transport->dead = 1;
+
/* Try to delete the heartbeat timer. */
- if (del_timer(&transport->hb_timer))
+ if (timer_delete(&transport->hb_timer))
sctp_transport_put(transport);
/* Delete the T3_rtx timer if it's active.
* There is no point in not doing this now and letting
* structure hang around in memory since we know
- * the tranport is going away.
+ * the transport is going away.
*/
- if (del_timer(&transport->T3_rtx_timer))
+ if (timer_delete(&transport->T3_rtx_timer))
+ sctp_transport_put(transport);
+
+ if (timer_delete(&transport->reconf_timer))
sctp_transport_put(transport);
- if (del_timer(&transport->reconf_timer))
+ if (timer_delete(&transport->probe_timer))
sctp_transport_put(transport);
/* Delete the ICMP proto unreachable timer if it's active. */
- if (del_timer(&transport->proto_unreach_timer))
- sctp_association_put(transport->asoc);
+ if (timer_delete(&transport->proto_unreach_timer))
+ sctp_transport_put(transport);
sctp_transport_put(transport);
}
@@ -192,10 +189,8 @@ void sctp_transport_reset_hb_timer(struct sctp_transport *transport)
/* When a data chunk is sent, reset the heartbeat interval. */
expires = jiffies + sctp_transport_timeout(transport);
- if ((time_before(transport->hb_timer.expires, expires) ||
- !timer_pending(&transport->hb_timer)) &&
- !mod_timer(&transport->hb_timer,
- expires + prandom_u32_max(transport->rto)))
+ if (!mod_timer(&transport->hb_timer,
+ expires + get_random_u32_below(transport->rto)))
sctp_transport_hold(transport);
}
@@ -207,6 +202,20 @@ void sctp_transport_reset_reconf_timer(struct sctp_transport *transport)
sctp_transport_hold(transport);
}
+void sctp_transport_reset_probe_timer(struct sctp_transport *transport)
+{
+ if (!mod_timer(&transport->probe_timer,
+ jiffies + transport->probe_interval))
+ sctp_transport_hold(transport);
+}
+
+void sctp_transport_reset_raise_timer(struct sctp_transport *transport)
+{
+ if (!mod_timer(&transport->probe_timer,
+ jiffies + transport->probe_interval * 30))
+ sctp_transport_hold(transport);
+}
+
/* This transport has been assigned to an association.
* Initialize fields from the association or from the sock itself.
* Register the reference count in the association.
@@ -222,7 +231,7 @@ void sctp_transport_set_owner(struct sctp_transport *transport,
void sctp_transport_pmtu(struct sctp_transport *transport, struct sock *sk)
{
/* If we don't have a fresh route, look one up */
- if (!transport->dst || transport->dst->obsolete) {
+ if (!transport->dst || READ_ONCE(transport->dst->obsolete)) {
sctp_transport_dst_release(transport);
transport->af_specific->get_dst(transport, &transport->saddr,
&transport->fl, sk);
@@ -241,12 +250,150 @@ void sctp_transport_pmtu(struct sctp_transport *transport, struct sock *sk)
transport->pathmtu = sctp_dst_mtu(transport->dst);
else
transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT;
+
+ sctp_transport_pl_update(transport);
+}
+
+void sctp_transport_pl_send(struct sctp_transport *t)
+{
+ if (t->pl.probe_count < SCTP_MAX_PROBES)
+ goto out;
+
+ t->pl.probe_count = 0;
+ if (t->pl.state == SCTP_PL_BASE) {
+ if (t->pl.probe_size == SCTP_BASE_PLPMTU) { /* BASE_PLPMTU Confirmation Failed */
+ t->pl.state = SCTP_PL_ERROR; /* Base -> Error */
+
+ t->pl.pmtu = SCTP_BASE_PLPMTU;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ sctp_assoc_sync_pmtu(t->asoc);
+ }
+ } else if (t->pl.state == SCTP_PL_SEARCH) {
+ if (t->pl.pmtu == t->pl.probe_size) { /* Black Hole Detected */
+ t->pl.state = SCTP_PL_BASE; /* Search -> Base */
+ t->pl.probe_size = SCTP_BASE_PLPMTU;
+ t->pl.probe_high = 0;
+
+ t->pl.pmtu = SCTP_BASE_PLPMTU;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ sctp_assoc_sync_pmtu(t->asoc);
+ } else { /* Normal probe failure. */
+ t->pl.probe_high = t->pl.probe_size;
+ t->pl.probe_size = t->pl.pmtu;
+ }
+ } else if (t->pl.state == SCTP_PL_COMPLETE) {
+ if (t->pl.pmtu == t->pl.probe_size) { /* Black Hole Detected */
+ t->pl.state = SCTP_PL_BASE; /* Search Complete -> Base */
+ t->pl.probe_size = SCTP_BASE_PLPMTU;
+
+ t->pl.pmtu = SCTP_BASE_PLPMTU;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ sctp_assoc_sync_pmtu(t->asoc);
+ }
+ }
+
+out:
+ pr_debug("%s: PLPMTUD: transport: %p, state: %d, pmtu: %d, size: %d, high: %d\n",
+ __func__, t, t->pl.state, t->pl.pmtu, t->pl.probe_size, t->pl.probe_high);
+ t->pl.probe_count++;
+}
+
+bool sctp_transport_pl_recv(struct sctp_transport *t)
+{
+ pr_debug("%s: PLPMTUD: transport: %p, state: %d, pmtu: %d, size: %d, high: %d\n",
+ __func__, t, t->pl.state, t->pl.pmtu, t->pl.probe_size, t->pl.probe_high);
+
+ t->pl.pmtu = t->pl.probe_size;
+ t->pl.probe_count = 0;
+ if (t->pl.state == SCTP_PL_BASE) {
+ t->pl.state = SCTP_PL_SEARCH; /* Base -> Search */
+ t->pl.probe_size += SCTP_PL_BIG_STEP;
+ } else if (t->pl.state == SCTP_PL_ERROR) {
+ t->pl.state = SCTP_PL_SEARCH; /* Error -> Search */
+
+ t->pl.pmtu = t->pl.probe_size;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ sctp_assoc_sync_pmtu(t->asoc);
+ t->pl.probe_size += SCTP_PL_BIG_STEP;
+ } else if (t->pl.state == SCTP_PL_SEARCH) {
+ if (!t->pl.probe_high) {
+ if (t->pl.probe_size < SCTP_MAX_PLPMTU) {
+ t->pl.probe_size = min(t->pl.probe_size + SCTP_PL_BIG_STEP,
+ SCTP_MAX_PLPMTU);
+ return false;
+ }
+ t->pl.probe_high = SCTP_MAX_PLPMTU;
+ }
+ t->pl.probe_size += SCTP_PL_MIN_STEP;
+ if (t->pl.probe_size >= t->pl.probe_high) {
+ t->pl.probe_high = 0;
+ t->pl.state = SCTP_PL_COMPLETE; /* Search -> Search Complete */
+
+ t->pl.probe_size = t->pl.pmtu;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ sctp_assoc_sync_pmtu(t->asoc);
+ sctp_transport_reset_raise_timer(t);
+ }
+ } else if (t->pl.state == SCTP_PL_COMPLETE) {
+ /* Raise probe_size again after 30 * interval in Search Complete */
+ t->pl.state = SCTP_PL_SEARCH; /* Search Complete -> Search */
+ t->pl.probe_size = min(t->pl.probe_size + SCTP_PL_MIN_STEP, SCTP_MAX_PLPMTU);
+ }
+
+ return t->pl.state == SCTP_PL_COMPLETE;
+}
+
+static bool sctp_transport_pl_toobig(struct sctp_transport *t, u32 pmtu)
+{
+ pr_debug("%s: PLPMTUD: transport: %p, state: %d, pmtu: %d, size: %d, ptb: %d\n",
+ __func__, t, t->pl.state, t->pl.pmtu, t->pl.probe_size, pmtu);
+
+ if (pmtu < SCTP_MIN_PLPMTU || pmtu >= t->pl.probe_size)
+ return false;
+
+ if (t->pl.state == SCTP_PL_BASE) {
+ if (pmtu >= SCTP_MIN_PLPMTU && pmtu < SCTP_BASE_PLPMTU) {
+ t->pl.state = SCTP_PL_ERROR; /* Base -> Error */
+
+ t->pl.pmtu = SCTP_BASE_PLPMTU;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ return true;
+ }
+ } else if (t->pl.state == SCTP_PL_SEARCH) {
+ if (pmtu >= SCTP_BASE_PLPMTU && pmtu < t->pl.pmtu) {
+ t->pl.state = SCTP_PL_BASE; /* Search -> Base */
+ t->pl.probe_size = SCTP_BASE_PLPMTU;
+ t->pl.probe_count = 0;
+
+ t->pl.probe_high = 0;
+ t->pl.pmtu = SCTP_BASE_PLPMTU;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ return true;
+ } else if (pmtu > t->pl.pmtu && pmtu < t->pl.probe_size) {
+ t->pl.probe_size = pmtu;
+ t->pl.probe_count = 0;
+ }
+ } else if (t->pl.state == SCTP_PL_COMPLETE) {
+ if (pmtu >= SCTP_BASE_PLPMTU && pmtu < t->pl.pmtu) {
+ t->pl.state = SCTP_PL_BASE; /* Complete -> Base */
+ t->pl.probe_size = SCTP_BASE_PLPMTU;
+ t->pl.probe_count = 0;
+
+ t->pl.probe_high = 0;
+ t->pl.pmtu = SCTP_BASE_PLPMTU;
+ t->pathmtu = t->pl.pmtu + sctp_transport_pl_hlen(t);
+ sctp_transport_reset_probe_timer(t);
+ return true;
+ }
+ }
+
+ return false;
}
bool sctp_transport_update_pmtu(struct sctp_transport *t, u32 pmtu)
{
- struct dst_entry *dst = sctp_transport_dst_check(t);
struct sock *sk = t->asoc->base.sk;
+ struct dst_entry *dst;
bool change = true;
if (unlikely(pmtu < SCTP_DEFAULT_MINSEGMENT)) {
@@ -257,6 +404,10 @@ bool sctp_transport_update_pmtu(struct sctp_transport *t, u32 pmtu)
}
pmtu = SCTP_TRUNC4(pmtu);
+ if (sctp_transport_pl_enabled(t))
+ return sctp_transport_pl_toobig(t, pmtu - sctp_transport_pl_hlen(t));
+
+ dst = sctp_transport_dst_check(t);
if (dst) {
struct sctp_pf *pf = sctp_get_pf_specific(dst->ops->family);
union sctp_addr addr;
@@ -335,6 +486,7 @@ void sctp_transport_update_rto(struct sctp_transport *tp, __u32 rtt)
if (tp->rttvar || tp->srtt) {
struct net *net = tp->asoc->base.net;
+ unsigned int rto_beta, rto_alpha;
/* 6.3.1 C3) When a new RTT measurement R' is made, set
* RTTVAR <- (1 - RTO.Beta) * RTTVAR + RTO.Beta * |SRTT - R'|
* SRTT <- (1 - RTO.Alpha) * SRTT + RTO.Alpha * R'
@@ -346,10 +498,14 @@ void sctp_transport_update_rto(struct sctp_transport *tp, __u32 rtt)
* For example, assuming the default value of RTO.Alpha of
* 1/8, rto_alpha would be expressed as 3.
*/
- tp->rttvar = tp->rttvar - (tp->rttvar >> net->sctp.rto_beta)
- + (((__u32)abs((__s64)tp->srtt - (__s64)rtt)) >> net->sctp.rto_beta);
- tp->srtt = tp->srtt - (tp->srtt >> net->sctp.rto_alpha)
- + (rtt >> net->sctp.rto_alpha);
+ rto_beta = READ_ONCE(net->sctp.rto_beta);
+ if (rto_beta < 32)
+ tp->rttvar = tp->rttvar - (tp->rttvar >> rto_beta)
+ + (((__u32)abs((__s64)tp->srtt - (__s64)rtt)) >> rto_beta);
+ rto_alpha = READ_ONCE(net->sctp.rto_alpha);
+ if (rto_alpha < 32)
+ tp->srtt = tp->srtt - (tp->srtt >> rto_alpha)
+ + (rtt >> rto_alpha);
} else {
/* 6.3.1 C2) When the first RTT measurement R is made, set
* SRTT <- R, RTTVAR <- R/2.
@@ -671,7 +827,7 @@ void sctp_transport_reset(struct sctp_transport *t)
void sctp_transport_immediate_rtx(struct sctp_transport *t)
{
/* Stop pending T3_rtx_timer */
- if (del_timer(&t->T3_rtx_timer))
+ if (timer_delete(&t->T3_rtx_timer))
sctp_transport_put(t);
sctp_retransmit(&t->asoc->outqueue, t, SCTP_RTXR_T3_RTX);