* be incorporated into the next SCTP release.
*/
-#include <linux/config.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/wait.h>
#include <linux/time.h>
#include <linux/ip.h>
+#include <linux/capability.h>
#include <linux/fcntl.h>
#include <linux/poll.h>
#include <linux/init.h>
struct sctp_association *, sctp_socket_type_t);
static char *sctp_hmac_alg = SCTP_COOKIE_HMAC_ALG;
-extern kmem_cache_t *sctp_bucket_cachep;
+extern struct kmem_cache *sctp_bucket_cachep;
/* Get the sndbuf space available at the time on the association. */
static inline int sctp_wspace(struct sctp_association *asoc)
sizeof(struct sk_buff) +
sizeof(struct sctp_chunk);
- sk->sk_wmem_queued += SCTP_DATA_SNDSIZE(chunk) +
- sizeof(struct sk_buff) +
- sizeof(struct sctp_chunk);
-
atomic_add(sizeof(struct sctp_chunk), &sk->sk_wmem_alloc);
}
return -EINVAL;
/* Is this a valid SCTP address? */
- if (!af->addr_valid(addr, sctp_sk(sk)))
+ if (!af->addr_valid(addr, sctp_sk(sk), NULL))
return -EINVAL;
if (!sctp_sk(sk)->pf->send_verify(sctp_sk(sk), (addr)))
struct sctp_transport *transport;
union sctp_addr *laddr = (union sctp_addr *)addr;
- laddr->v4.sin_port = ntohs(laddr->v4.sin_port);
addr_asoc = sctp_endpoint_lookup_assoc(sctp_sk(sk)->ep,
- (union sctp_addr *)addr,
+ laddr,
&transport);
- laddr->v4.sin_port = htons(laddr->v4.sin_port);
if (!addr_asoc)
return NULL;
sctp_write_lock(&ep->base.addr_lock);
/* Use GFP_ATOMIC since BHs are disabled. */
- addr->v4.sin_port = ntohs(addr->v4.sin_port);
- ret = sctp_add_bind_addr(bp, addr, GFP_ATOMIC);
- addr->v4.sin_port = htons(addr->v4.sin_port);
+ ret = sctp_add_bind_addr(bp, addr, 1, GFP_ATOMIC);
sctp_write_unlock(&ep->base.addr_lock);
sctp_local_bh_enable();
/* ADDIP Section 4.1.1 Congestion Control of ASCONF Chunks
*
- * R1) One and only one ASCONF Chunk MAY be in transit and unacknowledged
+ * R1) One and only one ASCONF Chunk MAY be in transit and unacknowledged
* at any one time. If a sender, after sending an ASCONF chunk, decides
- * it needs to transfer another ASCONF Chunk, it MUST wait until the
+ * it needs to transfer another ASCONF Chunk, it MUST wait until the
* ASCONF-ACK Chunk returns from the previous ASCONF Chunk before sending a
- * subsequent ASCONF. Note this restriction binds each side, so at any
- * time two ASCONF may be in-transit on any given association (one sent
+ * subsequent ASCONF. Note this restriction binds each side, so at any
+ * time two ASCONF may be in-transit on any given association (one sent
* from each endpoint).
*/
static int sctp_send_asconf(struct sctp_association *asoc,
/* If there is an outstanding ASCONF chunk, queue it for later
* transmission.
- */
+ */
if (asoc->addip_last_asconf) {
list_add_tail(&chunk->list, &asoc->addip_chunk_list);
- goto out;
+ goto out;
}
/* Hold the chunk until an ASCONF_ACK is received. */
goto err_bindx_add;
}
- retval = sctp_do_bind(sk, (union sctp_addr *)sa_addr,
+ retval = sctp_do_bind(sk, (union sctp_addr *)sa_addr,
af->sockaddr_len);
addr_buf += af->sockaddr_len;
* associations that are part of the endpoint indicating that a list of local
* addresses are added to the endpoint.
*
- * If any of the addresses is already in the bind address list of the
+ * If any of the addresses is already in the bind address list of the
* association, we do not send the chunk for that association. But it will not
* affect other associations.
*
* Only sctp_setsockopt_bindx() is supposed to call this function.
*/
-static int sctp_send_asconf_add_ip(struct sock *sk,
+static int sctp_send_asconf_add_ip(struct sock *sk,
struct sockaddr *addrs,
int addrcnt)
{
struct sctp_chunk *chunk;
struct sctp_sockaddr_entry *laddr;
union sctp_addr *addr;
+ union sctp_addr saveaddr;
void *addr_buf;
struct sctp_af *af;
struct list_head *pos;
continue;
/* Check if any address in the packed array of addresses is
- * in the bind address list of the association. If so,
- * do not send the asconf chunk to its peer, but continue with
+ * in the bind address list of the association. If so,
+ * do not send the asconf chunk to its peer, but continue with
* other associations.
*/
addr_buf = addrs;
}
retval = sctp_send_asconf(asoc, chunk);
+ if (retval)
+ goto out;
- /* FIXME: After sending the add address ASCONF chunk, we
- * cannot append the address to the association's binding
- * address list, because the new address may be used as the
- * source of a message sent to the peer before the ASCONF
- * chunk is received by the peer. So we should wait until
- * ASCONF_ACK is received.
+ /* Add the new addresses to the bind address list with
+ * use_as_src set to 0.
*/
+ sctp_local_bh_disable();
+ sctp_write_lock(&asoc->base.addr_lock);
+ addr_buf = addrs;
+ for (i = 0; i < addrcnt; i++) {
+ addr = (union sctp_addr *)addr_buf;
+ af = sctp_get_af_specific(addr->v4.sin_family);
+ memcpy(&saveaddr, addr, af->sockaddr_len);
+ retval = sctp_add_bind_addr(bp, &saveaddr, 0,
+ GFP_ATOMIC);
+ addr_buf += af->sockaddr_len;
+ }
+ sctp_write_unlock(&asoc->base.addr_lock);
+ sctp_local_bh_enable();
}
out:
int cnt;
struct sctp_bind_addr *bp = &ep->base.bind_addr;
int retval = 0;
- union sctp_addr saveaddr;
void *addr_buf;
- struct sockaddr *sa_addr;
+ union sctp_addr *sa_addr;
struct sctp_af *af;
SCTP_DEBUG_PRINTK("sctp_bindx_rem (sk: %p, addrs: %p, addrcnt: %d)\n",
goto err_bindx_rem;
}
- /* The list may contain either IPv4 or IPv6 address;
- * determine the address length to copy the address to
- * saveaddr.
- */
- sa_addr = (struct sockaddr *)addr_buf;
- af = sctp_get_af_specific(sa_addr->sa_family);
+ sa_addr = (union sctp_addr *)addr_buf;
+ af = sctp_get_af_specific(sa_addr->sa.sa_family);
if (!af) {
retval = -EINVAL;
goto err_bindx_rem;
}
- memcpy(&saveaddr, sa_addr, af->sockaddr_len);
- saveaddr.v4.sin_port = ntohs(saveaddr.v4.sin_port);
- if (saveaddr.v4.sin_port != bp->port) {
+
+ if (!af->addr_valid(sa_addr, sp, NULL)) {
+ retval = -EADDRNOTAVAIL;
+ goto err_bindx_rem;
+ }
+
+ if (sa_addr->v4.sin_port != htons(bp->port)) {
retval = -EINVAL;
goto err_bindx_rem;
}
sctp_local_bh_disable();
sctp_write_lock(&ep->base.addr_lock);
- retval = sctp_del_bind_addr(bp, &saveaddr);
+ retval = sctp_del_bind_addr(bp, sa_addr);
sctp_write_unlock(&ep->base.addr_lock);
sctp_local_bh_enable();
* the associations that are part of the endpoint indicating that a list of
* local addresses are removed from the endpoint.
*
- * If any of the addresses is already in the bind address list of the
+ * If any of the addresses is already in the bind address list of the
* association, we do not send the chunk for that association. But it will not
* affect other associations.
*
struct sctp_sock *sp;
struct sctp_endpoint *ep;
struct sctp_association *asoc;
+ struct sctp_transport *transport;
struct sctp_bind_addr *bp;
struct sctp_chunk *chunk;
union sctp_addr *laddr;
void *addr_buf;
struct sctp_af *af;
- struct list_head *pos;
+ struct list_head *pos, *pos1;
+ struct sctp_sockaddr_entry *saddr;
int i;
int retval = 0;
continue;
/* Check if any address in the packed array of addresses is
- * not present in the bind address list of the association.
+ * not present in the bind address list of the association.
* If so, do not send the asconf chunk to its peer, but
* continue with other associations.
*/
goto out;
}
- retval = sctp_send_asconf(asoc, chunk);
+ /* Reset use_as_src flag for the addresses in the bind address
+ * list that are to be deleted.
+ */
+ sctp_local_bh_disable();
+ sctp_write_lock(&asoc->base.addr_lock);
+ addr_buf = addrs;
+ for (i = 0; i < addrcnt; i++) {
+ laddr = (union sctp_addr *)addr_buf;
+ af = sctp_get_af_specific(laddr->v4.sin_family);
+ list_for_each(pos1, &bp->address_list) {
+ saddr = list_entry(pos1,
+ struct sctp_sockaddr_entry,
+ list);
+ if (sctp_cmp_addr_exact(&saddr->a, laddr))
+ saddr->use_as_src = 0;
+ }
+ addr_buf += af->sockaddr_len;
+ }
+ sctp_write_unlock(&asoc->base.addr_lock);
+ sctp_local_bh_enable();
- /* FIXME: After sending the delete address ASCONF chunk, we
- * cannot remove the addresses from the association's bind
- * address list, because there maybe some packet send to
- * the delete addresses, so we should wait until ASCONF_ACK
- * packet is received.
+ /* Update the route and saddr entries for all the transports
+ * as some of the addresses in the bind address list are
+ * about to be deleted and cannot be used as source addresses.
*/
+ list_for_each(pos1, &asoc->peer.transport_addr_list) {
+ transport = list_entry(pos1, struct sctp_transport,
+ transports);
+ dst_release(transport->dst);
+ sctp_transport_route(transport, NULL,
+ sctp_sk(asoc->base.sk));
+ }
+
+ retval = sctp_send_asconf(asoc, chunk);
}
out:
return retval;
* addrs is a pointer to an array of one or more socket addresses. Each
* address is contained in its appropriate structure (i.e. struct
* sockaddr_in or struct sockaddr_in6) the family of the address type
- * must be used to distengish the address length (note that this
+ * must be used to distinguish the address length (note that this
* representation is termed a "packed array" of addresses). The caller
* specifies the number of addresses in the array with addrcnt.
*
return -EFAULT;
/* Alloc space for the address array in kernel memory. */
- kaddrs = (struct sockaddr *)kmalloc(addrs_size, GFP_KERNEL);
+ kaddrs = kmalloc(addrs_size, GFP_KERNEL);
if (unlikely(!kaddrs))
return -ENOMEM;
return -EFAULT;
}
- /* Walk through the addrs buffer and count the number of addresses. */
+ /* Walk through the addrs buffer and count the number of addresses. */
addr_buf = kaddrs;
while (walk_size < addrs_size) {
sa_addr = (struct sockaddr *)addr_buf;
/* If the address family is not supported or if this address
* causes the address buffer to overflow return EINVAL.
- */
+ */
if (!af || (walk_size + af->sockaddr_len) > addrs_size) {
kfree(kaddrs);
return -EINVAL;
default:
err = -EINVAL;
break;
- };
+ }
out:
kfree(kaddrs);
int err = 0;
int addrcnt = 0;
int walk_size = 0;
- struct sockaddr *sa_addr;
+ union sctp_addr *sa_addr;
void *addr_buf;
+ unsigned short port;
sp = sctp_sk(sk);
ep = sp->ep;
/* Walk through the addrs buffer and count the number of addresses. */
addr_buf = kaddrs;
while (walk_size < addrs_size) {
- sa_addr = (struct sockaddr *)addr_buf;
- af = sctp_get_af_specific(sa_addr->sa_family);
+ sa_addr = (union sctp_addr *)addr_buf;
+ af = sctp_get_af_specific(sa_addr->sa.sa_family);
+ port = ntohs(sa_addr->v4.sin_port);
/* If the address family is not supported or if this address
* causes the address buffer to overflow return EINVAL.
goto out_free;
}
- err = sctp_verify_addr(sk, (union sctp_addr *)sa_addr,
- af->sockaddr_len);
+ err = sctp_verify_addr(sk, sa_addr, af->sockaddr_len);
if (err)
goto out_free;
+ /* Make sure the destination port is correctly set
+ * in all addresses.
+ */
+ if (asoc && asoc->peer.port && asoc->peer.port != port)
+ goto out_free;
+
memcpy(&to, sa_addr, af->sockaddr_len);
- to.v4.sin_port = ntohs(to.v4.sin_port);
/* Check if there already is a matching association on the
* endpoint (other than the one created here).
*/
- asoc2 = sctp_endpoint_lookup_assoc(ep, &to, &transport);
+ asoc2 = sctp_endpoint_lookup_assoc(ep, sa_addr, &transport);
if (asoc2 && asoc2 != asoc) {
if (asoc2->state >= SCTP_STATE_ESTABLISHED)
err = -EISCONN;
* make sure that there is no peeled-off association matching
* the peer address even on another socket.
*/
- if (sctp_endpoint_is_peeled_off(ep, &to)) {
+ if (sctp_endpoint_is_peeled_off(ep, sa_addr)) {
err = -EADDRNOTAVAIL;
goto out_free;
}
}
} else {
/*
- * If an unprivileged user inherits a 1-many
- * style socket with open associations on a
- * privileged port, it MAY be permitted to
- * accept new associations, but it SHOULD NOT
+ * If an unprivileged user inherits a 1-many
+ * style socket with open associations on a
+ * privileged port, it MAY be permitted to
+ * accept new associations, but it SHOULD NOT
* be permitted to open new associations.
*/
if (ep->base.bind_addr.port < PROT_SOCK &&
}
}
- scope = sctp_scope(&to);
+ scope = sctp_scope(sa_addr);
asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL);
if (!asoc) {
err = -ENOMEM;
}
/* Prime the peer's transport structures. */
- transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL,
+ transport = sctp_assoc_add_peer(asoc, sa_addr, GFP_KERNEL,
SCTP_UNKNOWN);
if (!transport) {
err = -ENOMEM;
inet_sk(sk)->dport = htons(asoc->peer.port);
af = sctp_get_af_specific(to.sa.sa_family);
af->to_sk_daddr(&to, sk);
+ sk->sk_err = 0;
timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK);
err = sctp_wait_for_connect(asoc, &timeo);
out_free:
SCTP_DEBUG_PRINTK("About to exit __sctp_connect() free asoc: %p"
- " kaddrs: %p err: %d\n",
- asoc, kaddrs, err);
+ " kaddrs: %p err: %d\n",
+ asoc, kaddrs, err);
if (asoc)
sctp_association_free(asoc);
return err;
return -EFAULT;
/* Alloc space for the address array in kernel memory. */
- kaddrs = (struct sockaddr *)kmalloc(addrs_size, GFP_KERNEL);
+ kaddrs = kmalloc(addrs_size, GFP_KERNEL);
if (unlikely(!kaddrs))
return -ENOMEM;
ep = sctp_sk(sk)->ep;
- /* Walk all associations on a socket, not on an endpoint. */
+ /* Walk all associations on an endpoint. */
list_for_each_safe(pos, temp, &ep->asocs) {
asoc = list_entry(pos, struct sctp_association, asocs);
if (sctp_state(asoc, CLOSED)) {
sctp_unhash_established(asoc);
sctp_association_free(asoc);
+ continue;
+ }
+ }
- } else if (sock_flag(sk, SOCK_LINGER) &&
- !sk->sk_lingertime)
- sctp_primitive_ABORT(asoc, NULL);
- else
- sctp_primitive_SHUTDOWN(asoc, NULL);
+ if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) {
+ struct sctp_chunk *chunk;
+
+ chunk = sctp_make_abort_user(asoc, NULL, 0);
+ if (chunk)
+ sctp_primitive_ABORT(asoc, chunk);
} else
sctp_primitive_SHUTDOWN(asoc, NULL);
}
if (msg_namelen > sizeof(to))
msg_namelen = sizeof(to);
memcpy(&to, msg->msg_name, msg_namelen);
- SCTP_DEBUG_PRINTK("Just memcpy'd. msg_name is "
- "0x%x:%u.\n",
- to.v4.sin_addr.s_addr, to.v4.sin_port);
-
- to.v4.sin_port = ntohs(to.v4.sin_port);
msg_name = msg->msg_name;
}
* length messages when SCTP_EOF|SCTP_ABORT is not set.
* If SCTP_ABORT is set, the message length could be non zero with
* the msg_iov set to the user abort reason.
- */
+ */
if (((sinfo_flags & SCTP_EOF) && (msg_len > 0)) ||
(!(sinfo_flags & (SCTP_EOF|SCTP_ABORT)) && (msg_len == 0))) {
err = -EINVAL;
goto out_unlock;
}
if (sinfo_flags & SCTP_ABORT) {
+ struct sctp_chunk *chunk;
+
+ chunk = sctp_make_abort_user(asoc, msg, msg_len);
+ if (!chunk) {
+ err = -ENOMEM;
+ goto out_unlock;
+ }
+
SCTP_DEBUG_PRINTK("Aborting association: %p\n", asoc);
- sctp_primitive_ABORT(asoc, msg);
+ sctp_primitive_ABORT(asoc, chunk);
err = 0;
goto out_unlock;
}
= sinit->sinit_max_attempts;
}
if (sinit->sinit_max_init_timeo) {
- asoc->max_init_timeo =
+ asoc->max_init_timeo =
msecs_to_jiffies(sinit->sinit_max_init_timeo);
}
}
* address's parameters:
*
* struct sctp_paddrparams {
- * sctp_assoc_t spp_assoc_id;
- * struct sockaddr_storage spp_address;
- * uint32_t spp_hbinterval;
- * uint16_t spp_pathmaxrxt;
- * };
- *
- * spp_assoc_id - (UDP style socket) This is filled in the application,
- * and identifies the association for this query.
+ * sctp_assoc_t spp_assoc_id;
+ * struct sockaddr_storage spp_address;
+ * uint32_t spp_hbinterval;
+ * uint16_t spp_pathmaxrxt;
+ * uint32_t spp_pathmtu;
+ * uint32_t spp_sackdelay;
+ * uint32_t spp_flags;
+ * };
+ *
+ * spp_assoc_id - (one-to-many style socket) This is filled in the
+ * application, and identifies the association for
+ * this query.
* spp_address - This specifies which address is of interest.
* spp_hbinterval - This contains the value of the heartbeat interval,
- * in milliseconds. A value of 0, when modifying the
- * parameter, specifies that the heartbeat on this
- * address should be disabled. A value of UINT32_MAX
- * (4294967295), when modifying the parameter,
- * specifies that a heartbeat should be sent
- * immediately to the peer address, and the current
- * interval should remain unchanged.
+ * in milliseconds. If a value of zero
+ * is present in this field then no changes are to
+ * be made to this parameter.
* spp_pathmaxrxt - This contains the maximum number of
* retransmissions before this address shall be
- * considered unreachable.
+ * considered unreachable. If a value of zero
+ * is present in this field then no changes are to
+ * be made to this parameter.
+ * spp_pathmtu - When Path MTU discovery is disabled the value
+ * specified here will be the "fixed" path mtu.
+ * Note that if the spp_address field is empty
+ * then all associations on this address will
+ * have this fixed path mtu set upon them.
+ *
+ * spp_sackdelay - When delayed sack is enabled, this value specifies
+ * the number of milliseconds that sacks will be delayed
+ * for. This value will apply to all addresses of an
+ * association if the spp_address field is empty. Note
+ * also, that if delayed sack is enabled and this
+ * value is set to 0, no change is made to the last
+ * recorded delayed sack timer value.
+ *
+ * spp_flags - These flags are used to control various features
+ * on an association. The flag field may contain
+ * zero or more of the following options.
+ *
+ * SPP_HB_ENABLE - Enable heartbeats on the
+ * specified address. Note that if the address
+ * field is empty all addresses for the association
+ * have heartbeats enabled upon them.
+ *
+ * SPP_HB_DISABLE - Disable heartbeats on the
+ * speicifed address. Note that if the address
+ * field is empty all addresses for the association
+ * will have their heartbeats disabled. Note also
+ * that SPP_HB_ENABLE and SPP_HB_DISABLE are
+ * mutually exclusive, only one of these two should
+ * be specified. Enabling both fields will have
+ * undetermined results.
+ *
+ * SPP_HB_DEMAND - Request a user initiated heartbeat
+ * to be made immediately.
+ *
+ * SPP_HB_TIME_IS_ZERO - Specify's that the time for
+ * heartbeat delayis to be set to the value of 0
+ * milliseconds.
+ *
+ * SPP_PMTUD_ENABLE - This field will enable PMTU
+ * discovery upon the specified address. Note that
+ * if the address feild is empty then all addresses
+ * on the association are effected.
+ *
+ * SPP_PMTUD_DISABLE - This field will disable PMTU
+ * discovery upon the specified address. Note that
+ * if the address feild is empty then all addresses
+ * on the association are effected. Not also that
+ * SPP_PMTUD_ENABLE and SPP_PMTUD_DISABLE are mutually
+ * exclusive. Enabling both will have undetermined
+ * results.
+ *
+ * SPP_SACKDELAY_ENABLE - Setting this flag turns
+ * on delayed sack. The time specified in spp_sackdelay
+ * is used to specify the sack delay for this address. Note
+ * that if spp_address is empty then all addresses will
+ * enable delayed sack and take on the sack delay
+ * value specified in spp_sackdelay.
+ * SPP_SACKDELAY_DISABLE - Setting this flag turns
+ * off delayed sack. If the spp_address field is blank then
+ * delayed sack is disabled for the entire association. Note
+ * also that this field is mutually exclusive to
+ * SPP_SACKDELAY_ENABLE, setting both will have undefined
+ * results.
*/
+static int sctp_apply_peer_addr_params(struct sctp_paddrparams *params,
+ struct sctp_transport *trans,
+ struct sctp_association *asoc,
+ struct sctp_sock *sp,
+ int hb_change,
+ int pmtud_change,
+ int sackdelay_change)
+{
+ int error;
+
+ if (params->spp_flags & SPP_HB_DEMAND && trans) {
+ error = sctp_primitive_REQUESTHEARTBEAT (trans->asoc, trans);
+ if (error)
+ return error;
+ }
+
+ /* Note that unless the spp_flag is set to SPP_HB_ENABLE the value of
+ * this field is ignored. Note also that a value of zero indicates
+ * the current setting should be left unchanged.
+ */
+ if (params->spp_flags & SPP_HB_ENABLE) {
+
+ /* Re-zero the interval if the SPP_HB_TIME_IS_ZERO is
+ * set. This lets us use 0 value when this flag
+ * is set.
+ */
+ if (params->spp_flags & SPP_HB_TIME_IS_ZERO)
+ params->spp_hbinterval = 0;
+
+ if (params->spp_hbinterval ||
+ (params->spp_flags & SPP_HB_TIME_IS_ZERO)) {
+ if (trans) {
+ trans->hbinterval =
+ msecs_to_jiffies(params->spp_hbinterval);
+ } else if (asoc) {
+ asoc->hbinterval =
+ msecs_to_jiffies(params->spp_hbinterval);
+ } else {
+ sp->hbinterval = params->spp_hbinterval;
+ }
+ }
+ }
+
+ if (hb_change) {
+ if (trans) {
+ trans->param_flags =
+ (trans->param_flags & ~SPP_HB) | hb_change;
+ } else if (asoc) {
+ asoc->param_flags =
+ (asoc->param_flags & ~SPP_HB) | hb_change;
+ } else {
+ sp->param_flags =
+ (sp->param_flags & ~SPP_HB) | hb_change;
+ }
+ }
+
+ /* When Path MTU discovery is disabled the value specified here will
+ * be the "fixed" path mtu (i.e. the value of the spp_flags field must
+ * include the flag SPP_PMTUD_DISABLE for this field to have any
+ * effect).
+ */
+ if ((params->spp_flags & SPP_PMTUD_DISABLE) && params->spp_pathmtu) {
+ if (trans) {
+ trans->pathmtu = params->spp_pathmtu;
+ sctp_assoc_sync_pmtu(asoc);
+ } else if (asoc) {
+ asoc->pathmtu = params->spp_pathmtu;
+ sctp_frag_point(sp, params->spp_pathmtu);
+ } else {
+ sp->pathmtu = params->spp_pathmtu;
+ }
+ }
+
+ if (pmtud_change) {
+ if (trans) {
+ int update = (trans->param_flags & SPP_PMTUD_DISABLE) &&
+ (params->spp_flags & SPP_PMTUD_ENABLE);
+ trans->param_flags =
+ (trans->param_flags & ~SPP_PMTUD) | pmtud_change;
+ if (update) {
+ sctp_transport_pmtu(trans);
+ sctp_assoc_sync_pmtu(asoc);
+ }
+ } else if (asoc) {
+ asoc->param_flags =
+ (asoc->param_flags & ~SPP_PMTUD) | pmtud_change;
+ } else {
+ sp->param_flags =
+ (sp->param_flags & ~SPP_PMTUD) | pmtud_change;
+ }
+ }
+
+ /* Note that unless the spp_flag is set to SPP_SACKDELAY_ENABLE the
+ * value of this field is ignored. Note also that a value of zero
+ * indicates the current setting should be left unchanged.
+ */
+ if ((params->spp_flags & SPP_SACKDELAY_ENABLE) && params->spp_sackdelay) {
+ if (trans) {
+ trans->sackdelay =
+ msecs_to_jiffies(params->spp_sackdelay);
+ } else if (asoc) {
+ asoc->sackdelay =
+ msecs_to_jiffies(params->spp_sackdelay);
+ } else {
+ sp->sackdelay = params->spp_sackdelay;
+ }
+ }
+
+ if (sackdelay_change) {
+ if (trans) {
+ trans->param_flags =
+ (trans->param_flags & ~SPP_SACKDELAY) |
+ sackdelay_change;
+ } else if (asoc) {
+ asoc->param_flags =
+ (asoc->param_flags & ~SPP_SACKDELAY) |
+ sackdelay_change;
+ } else {
+ sp->param_flags =
+ (sp->param_flags & ~SPP_SACKDELAY) |
+ sackdelay_change;
+ }
+ }
+
+ /* Note that unless the spp_flag is set to SPP_PMTUD_ENABLE the value
+ * of this field is ignored. Note also that a value of zero
+ * indicates the current setting should be left unchanged.
+ */
+ if ((params->spp_flags & SPP_PMTUD_ENABLE) && params->spp_pathmaxrxt) {
+ if (trans) {
+ trans->pathmaxrxt = params->spp_pathmaxrxt;
+ } else if (asoc) {
+ asoc->pathmaxrxt = params->spp_pathmaxrxt;
+ } else {
+ sp->pathmaxrxt = params->spp_pathmaxrxt;
+ }
+ }
+
+ return 0;
+}
+
static int sctp_setsockopt_peer_addr_params(struct sock *sk,
char __user *optval, int optlen)
{
- struct sctp_paddrparams params;
- struct sctp_transport *trans;
+ struct sctp_paddrparams params;
+ struct sctp_transport *trans = NULL;
+ struct sctp_association *asoc = NULL;
+ struct sctp_sock *sp = sctp_sk(sk);
int error;
+ int hb_change, pmtud_change, sackdelay_change;
if (optlen != sizeof(struct sctp_paddrparams))
- return -EINVAL;
+ return - EINVAL;
+
if (copy_from_user(¶ms, optval, optlen))
return -EFAULT;
- /*
- * API 7. Socket Options (setting the default value for the endpoint)
- * All options that support specific settings on an association by
- * filling in either an association id variable or a sockaddr_storage
- * SHOULD also support setting of the same value for the entire endpoint
- * (i.e. future associations). To accomplish this the following logic is
- * used when setting one of these options:
-
- * c) If neither the sockaddr_storage or association identification is
- * set i.e. the sockaddr_storage is set to all 0's (INADDR_ANY) and
- * the association identification is 0, the settings are a default
- * and to be applied to the endpoint (all future associations).
- */
+ /* Validate flags and value parameters. */
+ hb_change = params.spp_flags & SPP_HB;
+ pmtud_change = params.spp_flags & SPP_PMTUD;
+ sackdelay_change = params.spp_flags & SPP_SACKDELAY;
+
+ if (hb_change == SPP_HB ||
+ pmtud_change == SPP_PMTUD ||
+ sackdelay_change == SPP_SACKDELAY ||
+ params.spp_sackdelay > 500 ||
+ (params.spp_pathmtu
+ && params.spp_pathmtu < SCTP_DEFAULT_MINSEGMENT))
+ return -EINVAL;
- /* update default value for endpoint (all future associations) */
- if (!params.spp_assoc_id &&
- sctp_is_any(( union sctp_addr *)¶ms.spp_address)) {
- /* Manual heartbeat on an endpoint is invalid. */
- if (0xffffffff == params.spp_hbinterval)
+ /* If an address other than INADDR_ANY is specified, and
+ * no transport is found, then the request is invalid.
+ */
+ if (!sctp_is_any(( union sctp_addr *)¶ms.spp_address)) {
+ trans = sctp_addr_id2transport(sk, ¶ms.spp_address,
+ params.spp_assoc_id);
+ if (!trans)
return -EINVAL;
- else if (params.spp_hbinterval)
- sctp_sk(sk)->paddrparam.spp_hbinterval =
- params.spp_hbinterval;
- if (params.spp_pathmaxrxt)
- sctp_sk(sk)->paddrparam.spp_pathmaxrxt =
- params.spp_pathmaxrxt;
- return 0;
}
- trans = sctp_addr_id2transport(sk, ¶ms.spp_address,
- params.spp_assoc_id);
- if (!trans)
+ /* Get association, if assoc_id != 0 and the socket is a one
+ * to many style socket, and an association was not found, then
+ * the id was invalid.
+ */
+ asoc = sctp_id2assoc(sk, params.spp_assoc_id);
+ if (!asoc && params.spp_assoc_id && sctp_style(sk, UDP))
return -EINVAL;
- /* Applications can enable or disable heartbeats for any peer address
- * of an association, modify an address's heartbeat interval, force a
- * heartbeat to be sent immediately, and adjust the address's maximum
- * number of retransmissions sent before an address is considered
- * unreachable.
- *
- * The value of the heartbeat interval, in milliseconds. A value of
- * UINT32_MAX (4294967295), when modifying the parameter, specifies
- * that a heartbeat should be sent immediately to the peer address,
- * and the current interval should remain unchanged.
+ /* Heartbeat demand can only be sent on a transport or
+ * association, but not a socket.
*/
- if (0xffffffff == params.spp_hbinterval) {
- error = sctp_primitive_REQUESTHEARTBEAT (trans->asoc, trans);
- if (error)
- return error;
- } else {
- /* The value of the heartbeat interval, in milliseconds. A value of 0,
- * when modifying the parameter, specifies that the heartbeat on this
- * address should be disabled.
+ if (params.spp_flags & SPP_HB_DEMAND && !trans && !asoc)
+ return -EINVAL;
+
+ /* Process parameters. */
+ error = sctp_apply_peer_addr_params(¶ms, trans, asoc, sp,
+ hb_change, pmtud_change,
+ sackdelay_change);
+
+ if (error)
+ return error;
+
+ /* If changes are for association, also apply parameters to each
+ * transport.
*/
- if (params.spp_hbinterval) {
- trans->hb_allowed = 1;
- trans->hb_interval =
- msecs_to_jiffies(params.spp_hbinterval);
- } else
- trans->hb_allowed = 0;
+ if (!trans && asoc) {
+ struct list_head *pos;
+
+ list_for_each(pos, &asoc->peer.transport_addr_list) {
+ trans = list_entry(pos, struct sctp_transport,
+ transports);
+ sctp_apply_peer_addr_params(¶ms, trans, asoc, sp,
+ hb_change, pmtud_change,
+ sackdelay_change);
+ }
}
- /* spp_pathmaxrxt contains the maximum number of retransmissions
- * before this address shall be considered unreachable.
+ return 0;
+}
+
+/* 7.1.23. Delayed Ack Timer (SCTP_DELAYED_ACK_TIME)
+ *
+ * This options will get or set the delayed ack timer. The time is set
+ * in milliseconds. If the assoc_id is 0, then this sets or gets the
+ * endpoints default delayed ack timer value. If the assoc_id field is
+ * non-zero, then the set or get effects the specified association.
+ *
+ * struct sctp_assoc_value {
+ * sctp_assoc_t assoc_id;
+ * uint32_t assoc_value;
+ * };
+ *
+ * assoc_id - This parameter, indicates which association the
+ * user is preforming an action upon. Note that if
+ * this field's value is zero then the endpoints
+ * default value is changed (effecting future
+ * associations only).
+ *
+ * assoc_value - This parameter contains the number of milliseconds
+ * that the user is requesting the delayed ACK timer
+ * be set to. Note that this value is defined in
+ * the standard to be between 200 and 500 milliseconds.
+ *
+ * Note: a value of zero will leave the value alone,
+ * but disable SACK delay. A non-zero value will also
+ * enable SACK delay.
+ */
+
+static int sctp_setsockopt_delayed_ack_time(struct sock *sk,
+ char __user *optval, int optlen)
+{
+ struct sctp_assoc_value params;
+ struct sctp_transport *trans = NULL;
+ struct sctp_association *asoc = NULL;
+ struct sctp_sock *sp = sctp_sk(sk);
+
+ if (optlen != sizeof(struct sctp_assoc_value))
+ return - EINVAL;
+
+ if (copy_from_user(¶ms, optval, optlen))
+ return -EFAULT;
+
+ /* Validate value parameter. */
+ if (params.assoc_value > 500)
+ return -EINVAL;
+
+ /* Get association, if assoc_id != 0 and the socket is a one
+ * to many style socket, and an association was not found, then
+ * the id was invalid.
*/
- if (params.spp_pathmaxrxt)
- trans->max_retrans = params.spp_pathmaxrxt;
+ asoc = sctp_id2assoc(sk, params.assoc_id);
+ if (!asoc && params.assoc_id && sctp_style(sk, UDP))
+ return -EINVAL;
+
+ if (params.assoc_value) {
+ if (asoc) {
+ asoc->sackdelay =
+ msecs_to_jiffies(params.assoc_value);
+ asoc->param_flags =
+ (asoc->param_flags & ~SPP_SACKDELAY) |
+ SPP_SACKDELAY_ENABLE;
+ } else {
+ sp->sackdelay = params.assoc_value;
+ sp->param_flags =
+ (sp->param_flags & ~SPP_SACKDELAY) |
+ SPP_SACKDELAY_ENABLE;
+ }
+ } else {
+ if (asoc) {
+ asoc->param_flags =
+ (asoc->param_flags & ~SPP_SACKDELAY) |
+ SPP_SACKDELAY_DISABLE;
+ } else {
+ sp->param_flags =
+ (sp->param_flags & ~SPP_SACKDELAY) |
+ SPP_SACKDELAY_DISABLE;
+ }
+ }
+
+ /* If change is for association, also apply to each transport. */
+ if (asoc) {
+ struct list_head *pos;
+
+ list_for_each(pos, &asoc->peer.transport_addr_list) {
+ trans = list_entry(pos, struct sctp_transport,
+ transports);
+ if (params.assoc_value) {
+ trans->sackdelay =
+ msecs_to_jiffies(params.assoc_value);
+ trans->param_flags =
+ (trans->param_flags & ~SPP_SACKDELAY) |
+ SPP_SACKDELAY_ENABLE;
+ } else {
+ trans->param_flags =
+ (trans->param_flags & ~SPP_SACKDELAY) |
+ SPP_SACKDELAY_DISABLE;
+ }
+ }
+ }
return 0;
}
return -EFAULT;
if (sinit.sinit_num_ostreams)
- sp->initmsg.sinit_num_ostreams = sinit.sinit_num_ostreams;
+ sp->initmsg.sinit_num_ostreams = sinit.sinit_num_ostreams;
if (sinit.sinit_max_instreams)
- sp->initmsg.sinit_max_instreams = sinit.sinit_max_instreams;
+ sp->initmsg.sinit_max_instreams = sinit.sinit_max_instreams;
if (sinit.sinit_max_attempts)
- sp->initmsg.sinit_max_attempts = sinit.sinit_max_attempts;
+ sp->initmsg.sinit_max_attempts = sinit.sinit_max_attempts;
if (sinit.sinit_max_init_timeo)
- sp->initmsg.sinit_max_init_timeo = sinit.sinit_max_init_timeo;
+ sp->initmsg.sinit_max_init_timeo = sinit.sinit_max_init_timeo;
return 0;
}
if (asoc) {
if (rtoinfo.srto_initial != 0)
- asoc->rto_initial =
+ asoc->rto_initial =
msecs_to_jiffies(rtoinfo.srto_initial);
if (rtoinfo.srto_max != 0)
asoc->rto_max = msecs_to_jiffies(rtoinfo.srto_max);
*
* 7.1.2 SCTP_ASSOCINFO
*
- * This option is used to tune the the maximum retransmission attempts
+ * This option is used to tune the maximum retransmission attempts
* of the association.
* Returns an error if the new association retransmission value is
* greater than the sum of the retransmission value of the peer.
/* Set the values to the specific association */
if (asoc) {
- if (assocparams.sasoc_asocmaxrxt != 0)
+ if (assocparams.sasoc_asocmaxrxt != 0) {
+ __u32 path_sum = 0;
+ int paths = 0;
+ struct list_head *pos;
+ struct sctp_transport *peer_addr;
+
+ list_for_each(pos, &asoc->peer.transport_addr_list) {
+ peer_addr = list_entry(pos,
+ struct sctp_transport,
+ transports);
+ path_sum += peer_addr->pathmaxrxt;
+ paths++;
+ }
+
+ /* Only validate asocmaxrxt if we have more then
+ * one path/transport. We do this because path
+ * retransmissions are only counted when we have more
+ * then one path.
+ */
+ if (paths > 1 &&
+ assocparams.sasoc_asocmaxrxt > path_sum)
+ return -EINVAL;
+
asoc->max_retrans = assocparams.sasoc_asocmaxrxt;
+ }
+
if (assocparams.sasoc_cookie_life != 0) {
asoc->cookie_life.tv_sec =
assocparams.sasoc_cookie_life / 1000;
/* Update the frag_point of the existing associations. */
list_for_each(pos, &(sp->ep->asocs)) {
asoc = list_entry(pos, struct sctp_association, asocs);
- asoc->frag_point = sctp_frag_point(sp, asoc->pmtu);
+ asoc->frag_point = sctp_frag_point(sp, asoc->pathmtu);
}
return 0;
return -EFAULT;
asoc = sctp_id2assoc(sk, prim.sspp_assoc_id);
- if (!asoc)
+ if (!asoc)
return -EINVAL;
if (!asoc->peer.asconf_capable)
return err;
}
-static int sctp_setsockopt_adaption_layer(struct sock *sk, char __user *optval,
+static int sctp_setsockopt_adaptation_layer(struct sock *sk, char __user *optval,
int optlen)
{
- struct sctp_setadaption adaption;
+ struct sctp_setadaptation adaptation;
+
+ if (optlen != sizeof(struct sctp_setadaptation))
+ return -EINVAL;
+ if (copy_from_user(&adaptation, optval, optlen))
+ return -EFAULT;
+
+ sctp_sk(sk)->adaptation_ind = adaptation.ssb_adaptation_ind;
+
+ return 0;
+}
+
+/*
+ * 7.1.29. Set or Get the default context (SCTP_CONTEXT)
+ *
+ * The context field in the sctp_sndrcvinfo structure is normally only
+ * used when a failed message is retrieved holding the value that was
+ * sent down on the actual send call. This option allows the setting of
+ * a default context on an association basis that will be received on
+ * reading messages from the peer. This is especially helpful in the
+ * one-2-many model for an application to keep some reference to an
+ * internal state machine that is processing messages on the
+ * association. Note that the setting of this value only effects
+ * received messages from the peer and does not effect the value that is
+ * saved with outbound messages.
+ */
+static int sctp_setsockopt_context(struct sock *sk, char __user *optval,
+ int optlen)
+{
+ struct sctp_assoc_value params;
+ struct sctp_sock *sp;
+ struct sctp_association *asoc;
+
+ if (optlen != sizeof(struct sctp_assoc_value))
+ return -EINVAL;
+ if (copy_from_user(¶ms, optval, optlen))
+ return -EFAULT;
+
+ sp = sctp_sk(sk);
+
+ if (params.assoc_id != 0) {
+ asoc = sctp_id2assoc(sk, params.assoc_id);
+ if (!asoc)
+ return -EINVAL;
+ asoc->default_rcv_context = params.assoc_value;
+ } else {
+ sp->default_rcv_context = params.assoc_value;
+ }
+
+ return 0;
+}
+
+/*
+ * 7.1.24. Get or set fragmented interleave (SCTP_FRAGMENT_INTERLEAVE)
+ *
+ * This options will at a minimum specify if the implementation is doing
+ * fragmented interleave. Fragmented interleave, for a one to many
+ * socket, is when subsequent calls to receive a message may return
+ * parts of messages from different associations. Some implementations
+ * may allow you to turn this value on or off. If so, when turned off,
+ * no fragment interleave will occur (which will cause a head of line
+ * blocking amongst multiple associations sharing the same one to many
+ * socket). When this option is turned on, then each receive call may
+ * come from a different association (thus the user must receive data
+ * with the extended calls (e.g. sctp_recvmsg) to keep track of which
+ * association each receive belongs to.
+ *
+ * This option takes a boolean value. A non-zero value indicates that
+ * fragmented interleave is on. A value of zero indicates that
+ * fragmented interleave is off.
+ *
+ * Note that it is important that an implementation that allows this
+ * option to be turned on, have it off by default. Otherwise an unaware
+ * application using the one to many model may become confused and act
+ * incorrectly.
+ */
+static int sctp_setsockopt_fragment_interleave(struct sock *sk,
+ char __user *optval,
+ int optlen)
+{
+ int val;
- if (optlen != sizeof(struct sctp_setadaption))
+ if (optlen != sizeof(int))
return -EINVAL;
- if (copy_from_user(&adaption, optval, optlen))
+ if (get_user(val, (int __user *)optval))
return -EFAULT;
- sctp_sk(sk)->adaption_ind = adaption.ssb_adaption_ind;
+ sctp_sk(sk)->frag_interleave = (val == 0) ? 0 : 1;
+
+ return 0;
+}
+
+/*
+ * 7.1.25. Set or Get the sctp partial delivery point
+ * (SCTP_PARTIAL_DELIVERY_POINT)
+ * This option will set or get the SCTP partial delivery point. This
+ * point is the size of a message where the partial delivery API will be
+ * invoked to help free up rwnd space for the peer. Setting this to a
+ * lower value will cause partial delivery's to happen more often. The
+ * calls argument is an integer that sets or gets the partial delivery
+ * point.
+ */
+static int sctp_setsockopt_partial_delivery_point(struct sock *sk,
+ char __user *optval,
+ int optlen)
+{
+ u32 val;
+
+ if (optlen != sizeof(u32))
+ return -EINVAL;
+ if (get_user(val, (int __user *)optval))
+ return -EFAULT;
+
+ sctp_sk(sk)->pd_point = val;
+
+ return 0; /* is this the right error code? */
+}
+
+/*
+ * 7.1.28. Set or Get the maximum burst (SCTP_MAX_BURST)
+ *
+ * This option will allow a user to change the maximum burst of packets
+ * that can be emitted by this association. Note that the default value
+ * is 4, and some implementations may restrict this setting so that it
+ * can only be lowered.
+ *
+ * NOTE: This text doesn't seem right. Do this on a socket basis with
+ * future associations inheriting the socket value.
+ */
+static int sctp_setsockopt_maxburst(struct sock *sk,
+ char __user *optval,
+ int optlen)
+{
+ int val;
+
+ if (optlen != sizeof(int))
+ return -EINVAL;
+ if (get_user(val, (int __user *)optval))
+ return -EFAULT;
+
+ if (val < 0)
+ return -EINVAL;
+
+ sctp_sk(sk)->max_burst = val;
return 0;
}
retval = sctp_setsockopt_peer_addr_params(sk, optval, optlen);
break;
+ case SCTP_DELAYED_ACK_TIME:
+ retval = sctp_setsockopt_delayed_ack_time(sk, optval, optlen);
+ break;
+ case SCTP_PARTIAL_DELIVERY_POINT:
+ retval = sctp_setsockopt_partial_delivery_point(sk, optval, optlen);
+ break;
+
case SCTP_INITMSG:
retval = sctp_setsockopt_initmsg(sk, optval, optlen);
break;
case SCTP_MAXSEG:
retval = sctp_setsockopt_maxseg(sk, optval, optlen);
break;
- case SCTP_ADAPTION_LAYER:
- retval = sctp_setsockopt_adaption_layer(sk, optval, optlen);
+ case SCTP_ADAPTATION_LAYER:
+ retval = sctp_setsockopt_adaptation_layer(sk, optval, optlen);
+ break;
+ case SCTP_CONTEXT:
+ retval = sctp_setsockopt_context(sk, optval, optlen);
+ break;
+ case SCTP_FRAGMENT_INTERLEAVE:
+ retval = sctp_setsockopt_fragment_interleave(sk, optval, optlen);
+ break;
+ case SCTP_MAX_BURST:
+ retval = sctp_setsockopt_maxburst(sk, optval, optlen);
break;
-
default:
retval = -ENOPROTOOPT;
break;
- };
+ }
sctp_release_sock(sk);
goto out;
}
- timeo = sock_rcvtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK);
+ timeo = sock_rcvtimeo(sk, flags & O_NONBLOCK);
error = sctp_wait_for_accept(sk, timeo);
if (error)
out:
sctp_release_sock(sk);
- *err = error;
+ *err = error;
return newsk;
}
sp->default_context = 0;
sp->default_timetolive = 0;
+ sp->default_rcv_context = 0;
+ sp->max_burst = sctp_max_burst;
+
/* Initialize default setup parameters. These parameters
* can be modified with the SCTP_INITMSG socket option or
* overridden by the SCTP_INIT CMSG.
sp->initmsg.sinit_num_ostreams = sctp_max_outstreams;
sp->initmsg.sinit_max_instreams = sctp_max_instreams;
sp->initmsg.sinit_max_attempts = sctp_max_retrans_init;
- sp->initmsg.sinit_max_init_timeo = jiffies_to_msecs(sctp_rto_max);
+ sp->initmsg.sinit_max_init_timeo = sctp_rto_max;
/* Initialize default RTO related parameters. These parameters can
* be modified for with the SCTP_RTOINFO socket option.
*/
- sp->rtoinfo.srto_initial = jiffies_to_msecs(sctp_rto_initial);
- sp->rtoinfo.srto_max = jiffies_to_msecs(sctp_rto_max);
- sp->rtoinfo.srto_min = jiffies_to_msecs(sctp_rto_min);
+ sp->rtoinfo.srto_initial = sctp_rto_initial;
+ sp->rtoinfo.srto_max = sctp_rto_max;
+ sp->rtoinfo.srto_min = sctp_rto_min;
/* Initialize default association related parameters. These parameters
* can be modified with the SCTP_ASSOCINFO socket option.
sp->assocparams.sasoc_number_peer_destinations = 0;
sp->assocparams.sasoc_peer_rwnd = 0;
sp->assocparams.sasoc_local_rwnd = 0;
- sp->assocparams.sasoc_cookie_life =
- jiffies_to_msecs(sctp_valid_cookie_life);
+ sp->assocparams.sasoc_cookie_life = sctp_valid_cookie_life;
/* Initialize default event subscriptions. By default, all the
- * options are off.
+ * options are off.
*/
memset(&sp->subscribe, 0, sizeof(struct sctp_event_subscribe));
/* Default Peer Address Parameters. These defaults can
* be modified via SCTP_PEER_ADDR_PARAMS
*/
- sp->paddrparam.spp_hbinterval = jiffies_to_msecs(sctp_hb_interval);
- sp->paddrparam.spp_pathmaxrxt = sctp_max_retrans_path;
+ sp->hbinterval = sctp_hb_interval;
+ sp->pathmaxrxt = sctp_max_retrans_path;
+ sp->pathmtu = 0; // allow default discovery
+ sp->sackdelay = sctp_sack_timeout;
+ sp->param_flags = SPP_HB_ENABLE |
+ SPP_PMTUD_ENABLE |
+ SPP_SACKDELAY_ENABLE;
/* If enabled no SCTP message fragmentation will be performed.
* Configure through SCTP_DISABLE_FRAGMENTS socket option.
*/
sp->disable_fragments = 0;
- /* Turn on/off any Nagle-like algorithm. */
- sp->nodelay = 1;
+ /* Enable Nagle algorithm by default. */
+ sp->nodelay = 0;
/* Enable by default. */
sp->v4mapped = 1;
/* User specified fragmentation limit. */
sp->user_frag = 0;
- sp->adaption_ind = 0;
+ sp->adaptation_ind = 0;
sp->pf = sctp_get_pf_specific(sk->sk_family);
/* Control variables for partial data delivery. */
- sp->pd_mode = 0;
+ atomic_set(&sp->pd_mode, 0);
skb_queue_head_init(&sp->pd_lobby);
+ sp->frag_interleave = 0;
/* Create a per socket endpoint structure. Even if we
* change the data structure relationships, this may still
status.sstat_outstrms = asoc->c.sinit_num_ostreams;
status.sstat_fragmentation_point = asoc->frag_point;
status.sstat_primary.spinfo_assoc_id = sctp_assoc2id(transport->asoc);
- memcpy(&status.sstat_primary.spinfo_address,
- &(transport->ipaddr), sizeof(union sctp_addr));
+ memcpy(&status.sstat_primary.spinfo_address, &transport->ipaddr,
+ transport->af_specific->sockaddr_len);
/* Map ipv4 address into v4-mapped-on-v6 address. */
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk),
(union sctp_addr *)&status.sstat_primary.spinfo_address);
status.sstat_primary.spinfo_cwnd = transport->cwnd;
status.sstat_primary.spinfo_srtt = transport->srtt;
status.sstat_primary.spinfo_rto = jiffies_to_msecs(transport->rto);
- status.sstat_primary.spinfo_mtu = transport->pmtu;
+ status.sstat_primary.spinfo_mtu = transport->pathmtu;
if (status.sstat_primary.spinfo_state == SCTP_UNKNOWN)
status.sstat_primary.spinfo_state = SCTP_ACTIVE;
pinfo.spinfo_cwnd = transport->cwnd;
pinfo.spinfo_srtt = transport->srtt;
pinfo.spinfo_rto = jiffies_to_msecs(transport->rto);
- pinfo.spinfo_mtu = transport->pmtu;
+ pinfo.spinfo_mtu = transport->pathmtu;
if (pinfo.spinfo_state == SCTP_UNKNOWN)
pinfo.spinfo_state = SCTP_ACTIVE;
{
struct sock *sk = asoc->base.sk;
struct socket *sock;
+ struct inet_sock *inetsk;
int err = 0;
/* An association cannot be branched off from an already peeled-off
* asoc to the newsk.
*/
sctp_sock_migrate(sk, sock->sk, asoc, SCTP_SOCKET_UDP_HIGH_BANDWIDTH);
+
+ /* Make peeled-off sockets more like 1-1 accepted sockets.
+ * Set the daddr and initialize id to something more random
+ */
+ inetsk = inet_sk(sock->sk);
+ inetsk->daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr;
+ inetsk->id = asoc->next_tsn ^ jiffies;
+
*sockp = sock;
return err;
* address's parameters:
*
* struct sctp_paddrparams {
- * sctp_assoc_t spp_assoc_id;
- * struct sockaddr_storage spp_address;
- * uint32_t spp_hbinterval;
- * uint16_t spp_pathmaxrxt;
- * };
- *
- * spp_assoc_id - (UDP style socket) This is filled in the application,
- * and identifies the association for this query.
+ * sctp_assoc_t spp_assoc_id;
+ * struct sockaddr_storage spp_address;
+ * uint32_t spp_hbinterval;
+ * uint16_t spp_pathmaxrxt;
+ * uint32_t spp_pathmtu;
+ * uint32_t spp_sackdelay;
+ * uint32_t spp_flags;
+ * };
+ *
+ * spp_assoc_id - (one-to-many style socket) This is filled in the
+ * application, and identifies the association for
+ * this query.
* spp_address - This specifies which address is of interest.
* spp_hbinterval - This contains the value of the heartbeat interval,
- * in milliseconds. A value of 0, when modifying the
- * parameter, specifies that the heartbeat on this
- * address should be disabled. A value of UINT32_MAX
- * (4294967295), when modifying the parameter,
- * specifies that a heartbeat should be sent
- * immediately to the peer address, and the current
- * interval should remain unchanged.
+ * in milliseconds. If a value of zero
+ * is present in this field then no changes are to
+ * be made to this parameter.
* spp_pathmaxrxt - This contains the maximum number of
* retransmissions before this address shall be
- * considered unreachable.
+ * considered unreachable. If a value of zero
+ * is present in this field then no changes are to
+ * be made to this parameter.
+ * spp_pathmtu - When Path MTU discovery is disabled the value
+ * specified here will be the "fixed" path mtu.
+ * Note that if the spp_address field is empty
+ * then all associations on this address will
+ * have this fixed path mtu set upon them.
+ *
+ * spp_sackdelay - When delayed sack is enabled, this value specifies
+ * the number of milliseconds that sacks will be delayed
+ * for. This value will apply to all addresses of an
+ * association if the spp_address field is empty. Note
+ * also, that if delayed sack is enabled and this
+ * value is set to 0, no change is made to the last
+ * recorded delayed sack timer value.
+ *
+ * spp_flags - These flags are used to control various features
+ * on an association. The flag field may contain
+ * zero or more of the following options.
+ *
+ * SPP_HB_ENABLE - Enable heartbeats on the
+ * specified address. Note that if the address
+ * field is empty all addresses for the association
+ * have heartbeats enabled upon them.
+ *
+ * SPP_HB_DISABLE - Disable heartbeats on the
+ * speicifed address. Note that if the address
+ * field is empty all addresses for the association
+ * will have their heartbeats disabled. Note also
+ * that SPP_HB_ENABLE and SPP_HB_DISABLE are
+ * mutually exclusive, only one of these two should
+ * be specified. Enabling both fields will have
+ * undetermined results.
+ *
+ * SPP_HB_DEMAND - Request a user initiated heartbeat
+ * to be made immediately.
+ *
+ * SPP_PMTUD_ENABLE - This field will enable PMTU
+ * discovery upon the specified address. Note that
+ * if the address feild is empty then all addresses
+ * on the association are effected.
+ *
+ * SPP_PMTUD_DISABLE - This field will disable PMTU
+ * discovery upon the specified address. Note that
+ * if the address feild is empty then all addresses
+ * on the association are effected. Not also that
+ * SPP_PMTUD_ENABLE and SPP_PMTUD_DISABLE are mutually
+ * exclusive. Enabling both will have undetermined
+ * results.
+ *
+ * SPP_SACKDELAY_ENABLE - Setting this flag turns
+ * on delayed sack. The time specified in spp_sackdelay
+ * is used to specify the sack delay for this address. Note
+ * that if spp_address is empty then all addresses will
+ * enable delayed sack and take on the sack delay
+ * value specified in spp_sackdelay.
+ * SPP_SACKDELAY_DISABLE - Setting this flag turns
+ * off delayed sack. If the spp_address field is blank then
+ * delayed sack is disabled for the entire association. Note
+ * also that this field is mutually exclusive to
+ * SPP_SACKDELAY_ENABLE, setting both will have undefined
+ * results.
*/
static int sctp_getsockopt_peer_addr_params(struct sock *sk, int len,
- char __user *optval, int __user *optlen)
+ char __user *optval, int __user *optlen)
{
- struct sctp_paddrparams params;
- struct sctp_transport *trans;
+ struct sctp_paddrparams params;
+ struct sctp_transport *trans = NULL;
+ struct sctp_association *asoc = NULL;
+ struct sctp_sock *sp = sctp_sk(sk);
if (len != sizeof(struct sctp_paddrparams))
return -EINVAL;
+
if (copy_from_user(¶ms, optval, len))
return -EFAULT;
- /* If no association id is specified retrieve the default value
- * for the endpoint that will be used for all future associations
+ /* If an address other than INADDR_ANY is specified, and
+ * no transport is found, then the request is invalid.
*/
- if (!params.spp_assoc_id &&
- sctp_is_any(( union sctp_addr *)¶ms.spp_address)) {
- params.spp_hbinterval = sctp_sk(sk)->paddrparam.spp_hbinterval;
- params.spp_pathmaxrxt = sctp_sk(sk)->paddrparam.spp_pathmaxrxt;
-
- goto done;
+ if (!sctp_is_any(( union sctp_addr *)¶ms.spp_address)) {
+ trans = sctp_addr_id2transport(sk, ¶ms.spp_address,
+ params.spp_assoc_id);
+ if (!trans) {
+ SCTP_DEBUG_PRINTK("Failed no transport\n");
+ return -EINVAL;
+ }
}
- trans = sctp_addr_id2transport(sk, ¶ms.spp_address,
- params.spp_assoc_id);
- if (!trans)
+ /* Get association, if assoc_id != 0 and the socket is a one
+ * to many style socket, and an association was not found, then
+ * the id was invalid.
+ */
+ asoc = sctp_id2assoc(sk, params.spp_assoc_id);
+ if (!asoc && params.spp_assoc_id && sctp_style(sk, UDP)) {
+ SCTP_DEBUG_PRINTK("Failed no association\n");
return -EINVAL;
+ }
- /* The value of the heartbeat interval, in milliseconds. A value of 0,
- * when modifying the parameter, specifies that the heartbeat on this
- * address should be disabled.
- */
- if (!trans->hb_allowed)
- params.spp_hbinterval = 0;
- else
- params.spp_hbinterval = jiffies_to_msecs(trans->hb_interval);
+ if (trans) {
+ /* Fetch transport values. */
+ params.spp_hbinterval = jiffies_to_msecs(trans->hbinterval);
+ params.spp_pathmtu = trans->pathmtu;
+ params.spp_pathmaxrxt = trans->pathmaxrxt;
+ params.spp_sackdelay = jiffies_to_msecs(trans->sackdelay);
+
+ /*draft-11 doesn't say what to return in spp_flags*/
+ params.spp_flags = trans->param_flags;
+ } else if (asoc) {
+ /* Fetch association values. */
+ params.spp_hbinterval = jiffies_to_msecs(asoc->hbinterval);
+ params.spp_pathmtu = asoc->pathmtu;
+ params.spp_pathmaxrxt = asoc->pathmaxrxt;
+ params.spp_sackdelay = jiffies_to_msecs(asoc->sackdelay);
+
+ /*draft-11 doesn't say what to return in spp_flags*/
+ params.spp_flags = asoc->param_flags;
+ } else {
+ /* Fetch socket values. */
+ params.spp_hbinterval = sp->hbinterval;
+ params.spp_pathmtu = sp->pathmtu;
+ params.spp_sackdelay = sp->sackdelay;
+ params.spp_pathmaxrxt = sp->pathmaxrxt;
+
+ /*draft-11 doesn't say what to return in spp_flags*/
+ params.spp_flags = sp->param_flags;
+ }
+
+ if (copy_to_user(optval, ¶ms, len))
+ return -EFAULT;
+
+ if (put_user(len, optlen))
+ return -EFAULT;
+
+ return 0;
+}
+
+/* 7.1.23. Delayed Ack Timer (SCTP_DELAYED_ACK_TIME)
+ *
+ * This options will get or set the delayed ack timer. The time is set
+ * in milliseconds. If the assoc_id is 0, then this sets or gets the
+ * endpoints default delayed ack timer value. If the assoc_id field is
+ * non-zero, then the set or get effects the specified association.
+ *
+ * struct sctp_assoc_value {
+ * sctp_assoc_t assoc_id;
+ * uint32_t assoc_value;
+ * };
+ *
+ * assoc_id - This parameter, indicates which association the
+ * user is preforming an action upon. Note that if
+ * this field's value is zero then the endpoints
+ * default value is changed (effecting future
+ * associations only).
+ *
+ * assoc_value - This parameter contains the number of milliseconds
+ * that the user is requesting the delayed ACK timer
+ * be set to. Note that this value is defined in
+ * the standard to be between 200 and 500 milliseconds.
+ *
+ * Note: a value of zero will leave the value alone,
+ * but disable SACK delay. A non-zero value will also
+ * enable SACK delay.
+ */
+static int sctp_getsockopt_delayed_ack_time(struct sock *sk, int len,
+ char __user *optval,
+ int __user *optlen)
+{
+ struct sctp_assoc_value params;
+ struct sctp_association *asoc = NULL;
+ struct sctp_sock *sp = sctp_sk(sk);
+
+ if (len != sizeof(struct sctp_assoc_value))
+ return - EINVAL;
- /* spp_pathmaxrxt contains the maximum number of retransmissions
- * before this address shall be considered unreachable.
+ if (copy_from_user(¶ms, optval, len))
+ return -EFAULT;
+
+ /* Get association, if assoc_id != 0 and the socket is a one
+ * to many style socket, and an association was not found, then
+ * the id was invalid.
*/
- params.spp_pathmaxrxt = trans->max_retrans;
+ asoc = sctp_id2assoc(sk, params.assoc_id);
+ if (!asoc && params.assoc_id && sctp_style(sk, UDP))
+ return -EINVAL;
+
+ if (asoc) {
+ /* Fetch association values. */
+ if (asoc->param_flags & SPP_SACKDELAY_ENABLE)
+ params.assoc_value = jiffies_to_msecs(
+ asoc->sackdelay);
+ else
+ params.assoc_value = 0;
+ } else {
+ /* Fetch socket values. */
+ if (sp->param_flags & SPP_SACKDELAY_ENABLE)
+ params.assoc_value = sp->sackdelay;
+ else
+ params.assoc_value = 0;
+ }
-done:
if (copy_to_user(optval, ¶ms, len))
return -EFAULT;
return cnt;
}
-/*
+/*
* Old API for getting list of peer addresses. Does not work for 32-bit
* programs running on a 64-bit kernel
*/
memcpy(&temp, &from->ipaddr, sizeof(temp));
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp);
addrlen = sctp_get_af_specific(sk->sk_family)->sockaddr_len;
- temp.v4.sin_port = htons(temp.v4.sin_port);
if (copy_to_user(to, &temp, addrlen))
return -EFAULT;
to += addrlen ;
return -EINVAL;
to = optval + offsetof(struct sctp_getaddrs,addrs);
- space_left = len - sizeof(struct sctp_getaddrs) -
+ space_left = len - sizeof(struct sctp_getaddrs) -
offsetof(struct sctp_getaddrs,addrs);
list_for_each(pos, &asoc->peer.transport_addr_list) {
memcpy(&temp, &from->ipaddr, sizeof(temp));
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp);
addrlen = sctp_get_af_specific(sk->sk_family)->sockaddr_len;
- if(space_left < addrlen)
+ if (space_left < addrlen)
return -ENOMEM;
- temp.v4.sin_port = htons(temp.v4.sin_port);
if (copy_to_user(to, &temp, addrlen))
return -EFAULT;
to += addrlen;
sctp_assoc_t id;
struct sctp_bind_addr *bp;
struct sctp_association *asoc;
- struct list_head *pos;
+ struct list_head *pos, *temp;
struct sctp_sockaddr_entry *addr;
rwlock_t *addr_lock;
- unsigned long flags;
int cnt = 0;
if (len != sizeof(sctp_assoc_t))
addr = list_entry(bp->address_list.next,
struct sctp_sockaddr_entry, list);
if (sctp_is_any(&addr->a)) {
- sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags);
- list_for_each(pos, &sctp_local_addr_list) {
+ list_for_each_safe(pos, temp, &sctp_local_addr_list) {
addr = list_entry(pos,
struct sctp_sockaddr_entry,
list);
- if ((PF_INET == sk->sk_family) &&
- (AF_INET6 == addr->a.sa.sa_family))
+ if ((PF_INET == sk->sk_family) &&
+ (AF_INET6 == addr->a.sa.sa_family))
continue;
cnt++;
}
- sctp_spin_unlock_irqrestore(&sctp_local_addr_lock,
- flags);
} else {
cnt = 1;
}
/* Helper function that copies local addresses to user and returns the number
* of addresses copied.
*/
-static int sctp_copy_laddrs_to_user_old(struct sock *sk, __u16 port, int max_addrs,
- void __user *to)
+static int sctp_copy_laddrs_old(struct sock *sk, __u16 port,
+ int max_addrs, void *to,
+ int *bytes_copied)
{
- struct list_head *pos;
+ struct list_head *pos, *next;
struct sctp_sockaddr_entry *addr;
- unsigned long flags;
union sctp_addr temp;
int cnt = 0;
int addrlen;
- sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags);
- list_for_each(pos, &sctp_local_addr_list) {
+ list_for_each_safe(pos, next, &sctp_local_addr_list) {
addr = list_entry(pos, struct sctp_sockaddr_entry, list);
- if ((PF_INET == sk->sk_family) &&
+ if ((PF_INET == sk->sk_family) &&
(AF_INET6 == addr->a.sa.sa_family))
continue;
memcpy(&temp, &addr->a, sizeof(temp));
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk),
&temp);
addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len;
- temp.v4.sin_port = htons(port);
- if (copy_to_user(to, &temp, addrlen)) {
- sctp_spin_unlock_irqrestore(&sctp_local_addr_lock,
- flags);
- return -EFAULT;
- }
+ memcpy(to, &temp, addrlen);
+
to += addrlen;
+ *bytes_copied += addrlen;
cnt ++;
if (cnt >= max_addrs) break;
}
- sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags);
return cnt;
}
-static int sctp_copy_laddrs_to_user(struct sock *sk, __u16 port,
- void * __user *to, size_t space_left)
+static int sctp_copy_laddrs(struct sock *sk, __u16 port, void *to,
+ size_t space_left, int *bytes_copied)
{
- struct list_head *pos;
+ struct list_head *pos, *next;
struct sctp_sockaddr_entry *addr;
- unsigned long flags;
union sctp_addr temp;
int cnt = 0;
int addrlen;
- sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags);
- list_for_each(pos, &sctp_local_addr_list) {
+ list_for_each_safe(pos, next, &sctp_local_addr_list) {
addr = list_entry(pos, struct sctp_sockaddr_entry, list);
- if ((PF_INET == sk->sk_family) &&
+ if ((PF_INET == sk->sk_family) &&
(AF_INET6 == addr->a.sa.sa_family))
continue;
memcpy(&temp, &addr->a, sizeof(temp));
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk),
&temp);
addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len;
- if(space_left<addrlen)
+ if (space_left < addrlen)
return -ENOMEM;
- temp.v4.sin_port = htons(port);
- if (copy_to_user(*to, &temp, addrlen)) {
- sctp_spin_unlock_irqrestore(&sctp_local_addr_lock,
- flags);
- return -EFAULT;
- }
- *to += addrlen;
+ memcpy(to, &temp, addrlen);
+
+ to += addrlen;
cnt ++;
space_left -= addrlen;
+ bytes_copied += addrlen;
}
- sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags);
return cnt;
}
int addrlen;
rwlock_t *addr_lock;
int err = 0;
+ void *addrs;
+ void *buf;
+ int bytes_copied = 0;
if (len != sizeof(struct sctp_getaddrs_old))
return -EINVAL;
to = getaddrs.addrs;
+ /* Allocate space for a local instance of packed array to hold all
+ * the data. We store addresses here first and then put write them
+ * to the user in one shot.
+ */
+ addrs = kmalloc(sizeof(union sctp_addr) * getaddrs.addr_num,
+ GFP_KERNEL);
+ if (!addrs)
+ return -ENOMEM;
+
sctp_read_lock(addr_lock);
/* If the endpoint is bound to 0.0.0.0 or ::0, get the valid
addr = list_entry(bp->address_list.next,
struct sctp_sockaddr_entry, list);
if (sctp_is_any(&addr->a)) {
- cnt = sctp_copy_laddrs_to_user_old(sk, bp->port,
- getaddrs.addr_num,
- to);
- if (cnt < 0) {
- err = cnt;
- goto unlock;
- }
- goto copy_getaddrs;
+ cnt = sctp_copy_laddrs_old(sk, bp->port,
+ getaddrs.addr_num,
+ addrs, &bytes_copied);
+ goto copy_getaddrs;
}
}
+ buf = addrs;
list_for_each(pos, &bp->address_list) {
addr = list_entry(pos, struct sctp_sockaddr_entry, list);
memcpy(&temp, &addr->a, sizeof(temp));
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp);
addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len;
- temp.v4.sin_port = htons(temp.v4.sin_port);
- if (copy_to_user(to, &temp, addrlen)) {
- err = -EFAULT;
- goto unlock;
- }
- to += addrlen;
+ memcpy(buf, &temp, addrlen);
+ buf += addrlen;
+ bytes_copied += addrlen;
cnt ++;
if (cnt >= getaddrs.addr_num) break;
}
copy_getaddrs:
+ sctp_read_unlock(addr_lock);
+
+ /* copy the entire address list into the user provided space */
+ if (copy_to_user(to, addrs, bytes_copied)) {
+ err = -EFAULT;
+ goto error;
+ }
+
+ /* copy the leading structure back to user */
getaddrs.addr_num = cnt;
if (copy_to_user(optval, &getaddrs, sizeof(struct sctp_getaddrs_old)))
err = -EFAULT;
-unlock:
- sctp_read_unlock(addr_lock);
+error:
+ kfree(addrs);
return err;
}
rwlock_t *addr_lock;
int err = 0;
size_t space_left;
- int bytes_copied;
+ int bytes_copied = 0;
+ void *addrs;
+ void *buf;
if (len <= sizeof(struct sctp_getaddrs))
return -EINVAL;
to = optval + offsetof(struct sctp_getaddrs,addrs);
space_left = len - sizeof(struct sctp_getaddrs) -
offsetof(struct sctp_getaddrs,addrs);
+ addrs = kmalloc(space_left, GFP_KERNEL);
+ if (!addrs)
+ return -ENOMEM;
sctp_read_lock(addr_lock);
addr = list_entry(bp->address_list.next,
struct sctp_sockaddr_entry, list);
if (sctp_is_any(&addr->a)) {
- cnt = sctp_copy_laddrs_to_user(sk, bp->port,
- &to, space_left);
+ cnt = sctp_copy_laddrs(sk, bp->port, addrs,
+ space_left, &bytes_copied);
if (cnt < 0) {
err = cnt;
- goto unlock;
+ goto error;
}
- goto copy_getaddrs;
+ goto copy_getaddrs;
}
}
+ buf = addrs;
list_for_each(pos, &bp->address_list) {
addr = list_entry(pos, struct sctp_sockaddr_entry, list);
memcpy(&temp, &addr->a, sizeof(temp));
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, &temp);
addrlen = sctp_get_af_specific(temp.sa.sa_family)->sockaddr_len;
- if(space_left < addrlen)
- return -ENOMEM; /*fixme: right error?*/
- temp.v4.sin_port = htons(temp.v4.sin_port);
- if (copy_to_user(to, &temp, addrlen)) {
- err = -EFAULT;
- goto unlock;
+ if (space_left < addrlen) {
+ err = -ENOMEM; /*fixme: right error?*/
+ goto error;
}
- to += addrlen;
+ memcpy(buf, &temp, addrlen);
+ buf += addrlen;
+ bytes_copied += addrlen;
cnt ++;
space_left -= addrlen;
}
copy_getaddrs:
+ sctp_read_unlock(addr_lock);
+
+ if (copy_to_user(to, addrs, bytes_copied)) {
+ err = -EFAULT;
+ goto error;
+ }
if (put_user(cnt, &((struct sctp_getaddrs __user *)optval)->addr_num))
return -EFAULT;
- bytes_copied = ((char __user *)to) - optval;
if (put_user(bytes_copied, optlen))
return -EFAULT;
-unlock:
- sctp_read_unlock(addr_lock);
+error:
+ kfree(addrs);
return err;
}
if (!asoc->peer.primary_path)
return -ENOTCONN;
-
- asoc->peer.primary_path->ipaddr.v4.sin_port =
- htons(asoc->peer.primary_path->ipaddr.v4.sin_port);
+
memcpy(&prim.ssp_addr, &asoc->peer.primary_path->ipaddr,
- sizeof(union sctp_addr));
- asoc->peer.primary_path->ipaddr.v4.sin_port =
- ntohs(asoc->peer.primary_path->ipaddr.v4.sin_port);
+ asoc->peer.primary_path->af_specific->sockaddr_len);
sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp,
(union sctp_addr *)&prim.ssp_addr);
}
/*
- * 7.1.11 Set Adaption Layer Indicator (SCTP_ADAPTION_LAYER)
+ * 7.1.11 Set Adaptation Layer Indicator (SCTP_ADAPTATION_LAYER)
*
- * Requests that the local endpoint set the specified Adaption Layer
+ * Requests that the local endpoint set the specified Adaptation Layer
* Indication parameter for all future INIT and INIT-ACK exchanges.
*/
-static int sctp_getsockopt_adaption_layer(struct sock *sk, int len,
+static int sctp_getsockopt_adaptation_layer(struct sock *sk, int len,
char __user *optval, int __user *optlen)
{
- struct sctp_setadaption adaption;
+ struct sctp_setadaptation adaptation;
- if (len != sizeof(struct sctp_setadaption))
+ if (len != sizeof(struct sctp_setadaptation))
return -EINVAL;
- adaption.ssb_adaption_ind = sctp_sk(sk)->adaption_ind;
- if (copy_to_user(optval, &adaption, len))
+ adaptation.ssb_adaptation_ind = sctp_sk(sk)->adaptation_ind;
+ if (copy_to_user(optval, &adaptation, len))
return -EFAULT;
return 0;
*
* 7.1.2 SCTP_ASSOCINFO
*
- * This option is used to tune the the maximum retransmission attempts
+ * This option is used to tune the maximum retransmission attempts
* of the association.
* Returns an error if the new association retransmission value is
* greater than the sum of the retransmission value of the peer.
}
/*
+ * 7.1.29. Set or Get the default context (SCTP_CONTEXT)
+ * (chapter and verse is quoted at sctp_setsockopt_context())
+ */
+static int sctp_getsockopt_context(struct sock *sk, int len,
+ char __user *optval, int __user *optlen)
+{
+ struct sctp_assoc_value params;
+ struct sctp_sock *sp;
+ struct sctp_association *asoc;
+
+ if (len != sizeof(struct sctp_assoc_value))
+ return -EINVAL;
+
+ if (copy_from_user(¶ms, optval, len))
+ return -EFAULT;
+
+ sp = sctp_sk(sk);
+
+ if (params.assoc_id != 0) {
+ asoc = sctp_id2assoc(sk, params.assoc_id);
+ if (!asoc)
+ return -EINVAL;
+ params.assoc_value = asoc->default_rcv_context;
+ } else {
+ params.assoc_value = sp->default_rcv_context;
+ }
+
+ if (put_user(len, optlen))
+ return -EFAULT;
+ if (copy_to_user(optval, ¶ms, len))
+ return -EFAULT;
+
+ return 0;
+}
+
+/*
* 7.1.17 Set the maximum fragrmentation size (SCTP_MAXSEG)
*
* This socket option specifies the maximum size to put in any outgoing
return 0;
}
+/*
+ * 7.1.24. Get or set fragmented interleave (SCTP_FRAGMENT_INTERLEAVE)
+ * (chapter and verse is quoted at sctp_setsockopt_fragment_interleave())
+ */
+static int sctp_getsockopt_fragment_interleave(struct sock *sk, int len,
+ char __user *optval, int __user *optlen)
+{
+ int val;
+
+ if (len < sizeof(int))
+ return -EINVAL;
+
+ len = sizeof(int);
+
+ val = sctp_sk(sk)->frag_interleave;
+ if (put_user(len, optlen))
+ return -EFAULT;
+ if (copy_to_user(optval, &val, len))
+ return -EFAULT;
+
+ return 0;
+}
+
+/*
+ * 7.1.25. Set or Get the sctp partial delivery point
+ * (chapter and verse is quoted at sctp_setsockopt_partial_delivery_point())
+ */
+static int sctp_getsockopt_partial_delivery_point(struct sock *sk, int len,
+ char __user *optval,
+ int __user *optlen)
+{
+ u32 val;
+
+ if (len < sizeof(u32))
+ return -EINVAL;
+
+ len = sizeof(u32);
+
+ val = sctp_sk(sk)->pd_point;
+ if (put_user(len, optlen))
+ return -EFAULT;
+ if (copy_to_user(optval, &val, len))
+ return -EFAULT;
+
+ return -ENOTSUPP;
+}
+
+/*
+ * 7.1.28. Set or Get the maximum burst (SCTP_MAX_BURST)
+ * (chapter and verse is quoted at sctp_setsockopt_maxburst())
+ */
+static int sctp_getsockopt_maxburst(struct sock *sk, int len,
+ char __user *optval,
+ int __user *optlen)
+{
+ int val;
+
+ if (len < sizeof(int))
+ return -EINVAL;
+
+ len = sizeof(int);
+
+ val = sctp_sk(sk)->max_burst;
+ if (put_user(len, optlen))
+ return -EFAULT;
+ if (copy_to_user(optval, &val, len))
+ return -EFAULT;
+
+ return -ENOTSUPP;
+}
+
SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname,
char __user *optval, int __user *optlen)
{
retval = sctp_getsockopt_peer_addr_params(sk, len, optval,
optlen);
break;
+ case SCTP_DELAYED_ACK_TIME:
+ retval = sctp_getsockopt_delayed_ack_time(sk, len, optval,
+ optlen);
+ break;
case SCTP_INITMSG:
retval = sctp_getsockopt_initmsg(sk, len, optval, optlen);
break;
retval = sctp_getsockopt_peer_addr_info(sk, len, optval,
optlen);
break;
- case SCTP_ADAPTION_LAYER:
- retval = sctp_getsockopt_adaption_layer(sk, len, optval,
+ case SCTP_ADAPTATION_LAYER:
+ retval = sctp_getsockopt_adaptation_layer(sk, len, optval,
optlen);
break;
+ case SCTP_CONTEXT:
+ retval = sctp_getsockopt_context(sk, len, optval, optlen);
+ break;
+ case SCTP_FRAGMENT_INTERLEAVE:
+ retval = sctp_getsockopt_fragment_interleave(sk, len, optval,
+ optlen);
+ break;
+ case SCTP_PARTIAL_DELIVERY_POINT:
+ retval = sctp_getsockopt_partial_delivery_point(sk, len, optval,
+ optlen);
+ break;
+ case SCTP_MAX_BURST:
+ retval = sctp_getsockopt_maxburst(sk, len, optval, optlen);
+ break;
default:
retval = -ENOPROTOOPT;
break;
- };
+ }
sctp_release_sock(sk);
return retval;
unsigned short snum;
int ret;
- /* NOTE: Remember to put this back to net order. */
- addr->v4.sin_port = ntohs(addr->v4.sin_port);
- snum = addr->v4.sin_port;
+ snum = ntohs(addr->v4.sin_port);
SCTP_DEBUG_PRINTK("sctp_get_port() begins, snum=%d\n", snum);
sctp_local_bh_disable();
struct hlist_node *node;
SCTP_DEBUG_PRINTK("sctp_get_port() found a possible match\n");
- if (pp->fastreuse && sk->sk_reuse)
+ if (pp->fastreuse && sk->sk_reuse &&
+ sk->sk_state != SCTP_SS_LISTENING)
goto success;
/* Run through the list of sockets bound to the port
struct sctp_endpoint *ep2;
ep2 = sctp_sk(sk2)->ep;
- if (reuse && sk2->sk_reuse)
+ if (reuse && sk2->sk_reuse &&
+ sk2->sk_state != SCTP_SS_LISTENING)
continue;
if (sctp_bind_addr_match(&ep2->base.bind_addr, addr,
* if sk->sk_reuse is too (that is, if the caller requested
* SO_REUSEADDR on this socket -sk-).
*/
- if (hlist_empty(&pp->owner))
- pp->fastreuse = sk->sk_reuse ? 1 : 0;
- else if (pp->fastreuse && !sk->sk_reuse)
+ if (hlist_empty(&pp->owner)) {
+ if (sk->sk_reuse && sk->sk_state != SCTP_SS_LISTENING)
+ pp->fastreuse = 1;
+ else
+ pp->fastreuse = 0;
+ } else if (pp->fastreuse &&
+ (!sk->sk_reuse || sk->sk_state == SCTP_SS_LISTENING))
pp->fastreuse = 0;
/* We are set, so fill up all the data in the hash table
* sockets FIXME: Blurry, NPI (ipg).
*/
success:
- inet_sk(sk)->num = snum;
if (!sctp_sk(sk)->bind_hash) {
+ inet_sk(sk)->num = snum;
sk_add_bind_node(sk, &pp->owner);
sctp_sk(sk)->bind_hash = pp;
}
fail:
sctp_local_bh_enable();
- addr->v4.sin_port = htons(addr->v4.sin_port);
return ret;
}
if (!backlog) {
if (sctp_sstate(sk, CLOSED))
return 0;
-
+
sctp_unhash_endpoint(ep);
sk->sk_state = SCTP_SS_CLOSED;
}
/* Return if we are already listening. */
if (sctp_sstate(sk, LISTENING))
return 0;
-
+
/*
* If a bind() or sctp_bindx() is not called prior to a listen()
* call that allows new associations to be accepted, the system
* This is not currently spelled out in the SCTP sockets
* extensions draft, but follows the practice as seen in TCP
* sockets.
+ *
+ * Additionally, turn off fastreuse flag since we are not listening
*/
+ sk->sk_state = SCTP_SS_LISTENING;
if (!ep->base.bind_addr.port) {
if (sctp_autobind(sk))
return -EAGAIN;
- }
- sk->sk_state = SCTP_SS_LISTENING;
+ } else
+ sctp_sk(sk)->bind_hash->fastreuse = 0;
+
sctp_hash_endpoint(ep);
return 0;
}
if (!backlog) {
if (sctp_sstate(sk, CLOSED))
return 0;
-
+
sctp_unhash_endpoint(ep);
sk->sk_state = SCTP_SS_CLOSED;
}
* extensions draft, but follows the practice as seen in TCP
* sockets.
*/
+ sk->sk_state = SCTP_SS_LISTENING;
if (!ep->base.bind_addr.port) {
if (sctp_autobind(sk))
return -EAGAIN;
- }
- sk->sk_state = SCTP_SS_LISTENING;
+ } else
+ sctp_sk(sk)->bind_hash->fastreuse = 0;
+
sk->sk_max_ack_backlog = backlog;
sctp_hash_endpoint(ep);
return 0;
int sctp_inet_listen(struct socket *sock, int backlog)
{
struct sock *sk = sock->sk;
- struct crypto_tfm *tfm=NULL;
+ struct crypto_hash *tfm = NULL;
int err = -EINVAL;
if (unlikely(backlog < 0))
/* Allocate HMAC for generating cookie. */
if (sctp_hmac_alg) {
- tfm = sctp_crypto_alloc_tfm(sctp_hmac_alg, 0);
- if (!tfm) {
+ tfm = crypto_alloc_hash(sctp_hmac_alg, 0, CRYPTO_ALG_ASYNC);
+ if (IS_ERR(tfm)) {
+ if (net_ratelimit()) {
+ printk(KERN_INFO
+ "SCTP: failed to load transform for %s: %ld\n",
+ sctp_hmac_alg, PTR_ERR(tfm));
+ }
err = -ENOSYS;
goto out;
}
break;
default:
break;
- };
+ }
+
if (err)
goto cleanup;
sctp_release_sock(sk);
return err;
cleanup:
- sctp_crypto_free_tfm(tfm);
+ crypto_free_hash(tfm);
goto out;
}
* tcp_poll(). Note that, based on these implementations, we don't
* lock the socket in this function, even though it seems that,
* ideally, locking or some other mechanisms can be used to ensure
- * the integrity of the counters (sndbuf and wmem_queued) used
+ * the integrity of the counters (sndbuf and wmem_alloc) used
* in this place. We assume that we don't need locks either until proven
* otherwise.
*
*/
if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))
return (!list_empty(&sp->ep->asocs)) ?
- (POLLIN | POLLRDNORM) : 0;
+ (POLLIN | POLLRDNORM) : 0;
mask = 0;
/* Is there any exceptional events? */
if (sk->sk_err || !skb_queue_empty(&sk->sk_error_queue))
mask |= POLLERR;
+ if (sk->sk_shutdown & RCV_SHUTDOWN)
+ mask |= POLLRDHUP;
if (sk->sk_shutdown == SHUTDOWN_MASK)
mask |= POLLHUP;
{
struct sctp_bind_bucket *pp;
- pp = kmem_cache_alloc(sctp_bucket_cachep, SLAB_ATOMIC);
+ pp = kmem_cache_alloc(sctp_bucket_cachep, GFP_ATOMIC);
SCTP_DBG_OBJCNT_INC(bind_bucket);
if (pp) {
pp->port = snum;
/* Caller must hold hashbucket lock for this tb with local BH disabled */
static void sctp_bucket_destroy(struct sctp_bind_bucket *pp)
{
- if (hlist_empty(&pp->owner)) {
+ if (pp && hlist_empty(&pp->owner)) {
if (pp->next)
pp->next->pprev = pp->pprev;
*(pp->pprev) = pp->next;
{
union sctp_addr autoaddr;
struct sctp_af *af;
- unsigned short port;
+ __be16 port;
/* Initialize a local sockaddr structure to INADDR_ANY. */
af = sctp_sk(sk)->pf->af;
default:
return -EINVAL;
- };
+ }
}
return 0;
}
struct sk_buff *skb;
long timeo;
- /* Caller is allowed not to check sk->sk_err before calling. */
- error = sock_error(sk);
- if (error)
- goto no_packet;
-
timeo = sock_rcvtimeo(sk, noblock);
SCTP_DEBUG_PRINTK("Timeout: timeo: %ld, MAX: %ld.\n",
if (skb)
return skb;
+ /* Caller is allowed not to check sk->sk_err before calling. */
+ error = sock_error(sk);
+ if (error)
+ goto no_packet;
+
if (sk->sk_shutdown & RCV_SHUTDOWN)
break;
sizeof(struct sk_buff) +
sizeof(struct sctp_chunk);
- sk->sk_wmem_queued -= SCTP_DATA_SNDSIZE(chunk) +
- sizeof(struct sk_buff) +
- sizeof(struct sctp_chunk);
-
atomic_sub(sizeof(struct sctp_chunk), &sk->sk_wmem_alloc);
sock_wfree(skb);
sctp_association_put(asoc);
}
+/* Do accounting for the receive space on the socket.
+ * Accounting for the association is done in ulpevent.c
+ * We set this as a destructor for the cloned data skbs so that
+ * accounting is done at the correct time.
+ */
+void sctp_sock_rfree(struct sk_buff *skb)
+{
+ struct sock *sk = skb->sk;
+ struct sctp_ulpevent *event = sctp_skb2event(skb);
+
+ atomic_sub(event->rmem_len, &sk->sk_rmem_alloc);
+}
+
+
/* Helper function to wait for space in the sndbuf. */
static int sctp_wait_for_sndbuf(struct sctp_association *asoc, long *timeo_p,
size_t msg_len)
DEFINE_WAIT(wait);
SCTP_DEBUG_PRINTK("wait_for_sndbuf: asoc=%p, timeo=%ld, msg_len=%zu\n",
- asoc, (long)(*timeo_p), msg_len);
+ asoc, (long)(*timeo_p), msg_len);
/* Increment the association's refcnt. */
sctp_association_hold(asoc);
*/
sctp_release_sock(sk);
current_timeo = schedule_timeout(current_timeo);
+ BUG_ON(sk != asoc->base.sk);
sctp_lock_sock(sk);
*timeo_p = current_timeo;
/* Is there any sndbuf space available on the socket?
*
- * Note that wmem_queued is the sum of the send buffers on all of the
+ * Note that sk_wmem_alloc is the sum of the send buffers on all of the
* associations on the same socket. For a UDP-style socket with
* multiple associations, it is possible for it to be "unwriteable"
* prematurely. I assume that this is acceptable because
{
int amt = 0;
- amt = sk->sk_sndbuf - sk->sk_wmem_queued;
+ amt = sk->sk_sndbuf - atomic_read(&sk->sk_wmem_alloc);
if (amt < 0)
amt = 0;
return amt;
return err;
do_error:
- if (asoc->init_err_counter + 1 >= asoc->max_init_attempts)
+ if (asoc->init_err_counter + 1 > asoc->max_init_attempts)
err = -ETIMEDOUT;
else
err = -ECONNREFUSED;
finish_wait(sk->sk_sleep, &wait);
}
+static void sctp_sock_rfree_frag(struct sk_buff *skb)
+{
+ struct sk_buff *frag;
+
+ if (!skb->data_len)
+ goto done;
+
+ /* Don't forget the fragments. */
+ for (frag = skb_shinfo(skb)->frag_list; frag; frag = frag->next)
+ sctp_sock_rfree_frag(frag);
+
+done:
+ sctp_sock_rfree(skb);
+}
+
+static void sctp_skb_set_owner_r_frag(struct sk_buff *skb, struct sock *sk)
+{
+ struct sk_buff *frag;
+
+ if (!skb->data_len)
+ goto done;
+
+ /* Don't forget the fragments. */
+ for (frag = skb_shinfo(skb)->frag_list; frag; frag = frag->next)
+ sctp_skb_set_owner_r_frag(frag, sk);
+
+done:
+ sctp_skb_set_owner_r(skb, sk);
+}
+
/* Populate the fields of the newsk from the oldsk and migrate the assoc
* and its messages to the newsk.
*/
/* Copy the bind_addr list from the original endpoint to the new
* endpoint so that we can handle restarts properly
*/
+ if (PF_INET6 == assoc->base.sk->sk_family)
+ flags = SCTP_ADDR6_ALLOWED;
if (assoc->peer.ipv4_address)
flags |= SCTP_ADDR4_PEERSUPP;
if (assoc->peer.ipv6_address)
sctp_skb_for_each(skb, &oldsk->sk_receive_queue, tmp) {
event = sctp_skb2event(skb);
if (event->asoc == assoc) {
+ sctp_sock_rfree_frag(skb);
__skb_unlink(skb, &oldsk->sk_receive_queue);
__skb_queue_tail(&newsk->sk_receive_queue, skb);
+ sctp_skb_set_owner_r_frag(skb, newsk);
}
}
* 3) Peeling off non-partial delivery; move pd_lobby to receive_queue.
*/
skb_queue_head_init(&newsp->pd_lobby);
- sctp_sk(newsk)->pd_mode = assoc->ulpq.pd_mode;
+ atomic_set(&sctp_sk(newsk)->pd_mode, assoc->ulpq.pd_mode);
- if (sctp_sk(oldsk)->pd_mode) {
+ if (atomic_read(&sctp_sk(oldsk)->pd_mode)) {
struct sk_buff_head *queue;
/* Decide which queue to move pd_lobby skbs to. */
sctp_skb_for_each(skb, &oldsp->pd_lobby, tmp) {
event = sctp_skb2event(skb);
if (event->asoc == assoc) {
+ sctp_sock_rfree_frag(skb);
__skb_unlink(skb, &oldsp->pd_lobby);
__skb_queue_tail(queue, skb);
+ sctp_skb_set_owner_r_frag(skb, newsk);
}
}
* delivery to finish.
*/
if (assoc->ulpq.pd_mode)
- sctp_clear_pd(oldsk);
+ sctp_clear_pd(oldsk, NULL);
+
+ }
+ sctp_skb_for_each(skb, &assoc->ulpq.reasm, tmp) {
+ sctp_sock_rfree_frag(skb);
+ sctp_skb_set_owner_r_frag(skb, newsk);
+ }
+
+ sctp_skb_for_each(skb, &assoc->ulpq.lobby, tmp) {
+ sctp_sock_rfree_frag(skb);
+ sctp_skb_set_owner_r_frag(skb, newsk);
}
/* Set the type of socket to indicate that it is peeled off from the
*/
newsp->type = type;
- /* Migrate the association to the new socket. */
+ /* Mark the new socket "in-use" by the user so that any packets
+ * that may arrive on the association after we've moved it are
+ * queued to the backlog. This prevents a potential race between
+ * backlog processing on the old socket and new-packet processing
+ * on the new socket.
+ */
+ sctp_lock_sock(newsk);
sctp_assoc_migrate(assoc, newsk);
/* If the association on the newsk is already closed before accept()
newsk->sk_shutdown |= RCV_SHUTDOWN;
newsk->sk_state = SCTP_SS_ESTABLISHED;
+ sctp_release_sock(newsk);
}
/* This proto struct describes the ULP interface for SCTP. */
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff