diff options
Diffstat (limited to 'net/sctp')
| -rw-r--r-- | net/sctp/Kconfig | 89 | ||||
| -rw-r--r-- | net/sctp/Makefile | 17 | ||||
| -rw-r--r-- | net/sctp/associola.c | 1205 | ||||
| -rw-r--r-- | net/sctp/bind_addr.c | 417 | ||||
| -rw-r--r-- | net/sctp/chunk.c | 309 | ||||
| -rw-r--r-- | net/sctp/command.c | 81 | ||||
| -rw-r--r-- | net/sctp/crc32c.c | 220 | ||||
| -rw-r--r-- | net/sctp/debug.c | 191 | ||||
| -rw-r--r-- | net/sctp/endpointola.c | 389 | ||||
| -rw-r--r-- | net/sctp/input.c | 913 | ||||
| -rw-r--r-- | net/sctp/inqueue.c | 204 | ||||
| -rw-r--r-- | net/sctp/ipv6.c | 1013 | ||||
| -rw-r--r-- | net/sctp/objcnt.c | 140 | ||||
| -rw-r--r-- | net/sctp/output.c | 646 | ||||
| -rw-r--r-- | net/sctp/outqueue.c | 1734 | ||||
| -rw-r--r-- | net/sctp/primitive.c | 219 | ||||
| -rw-r--r-- | net/sctp/proc.c | 288 | ||||
| -rw-r--r-- | net/sctp/protocol.c | 1240 | ||||
| -rw-r--r-- | net/sctp/sm_make_chunk.c | 2766 | ||||
| -rw-r--r-- | net/sctp/sm_sideeffect.c | 1395 | ||||
| -rw-r--r-- | net/sctp/sm_statefuns.c | 5238 | ||||
| -rw-r--r-- | net/sctp/sm_statetable.c | 1004 | ||||
| -rw-r--r-- | net/sctp/socket.c | 4797 | ||||
| -rw-r--r-- | net/sctp/ssnmap.c | 131 | ||||
| -rw-r--r-- | net/sctp/sysctl.c | 251 | ||||
| -rw-r--r-- | net/sctp/transport.c | 514 | ||||
| -rw-r--r-- | net/sctp/tsnmap.c | 417 | ||||
| -rw-r--r-- | net/sctp/ulpevent.c | 942 | ||||
| -rw-r--r-- | net/sctp/ulpqueue.c | 864 |
29 files changed, 27634 insertions, 0 deletions
diff --git a/net/sctp/Kconfig b/net/sctp/Kconfig new file mode 100644 index 000000000000..9cba49e2ad43 --- /dev/null +++ b/net/sctp/Kconfig @@ -0,0 +1,89 @@ +# +# SCTP configuration +# + +menu "SCTP Configuration (EXPERIMENTAL)" + depends on INET && EXPERIMENTAL + +config IP_SCTP + tristate "The SCTP Protocol (EXPERIMENTAL)" + depends on IPV6 || IPV6=n + select CRYPTO if SCTP_HMAC_SHA1 || SCTP_HMAC_MD5 + select CRYPTO_HMAC if SCTP_HMAC_SHA1 || SCTP_HMAC_MD5 + select CRYPTO_SHA1 if SCTP_HMAC_SHA1 + select CRYPTO_MD5 if SCTP_HMAC_MD5 + ---help--- + Stream Control Transmission Protocol + + From RFC 2960 <http://www.ietf.org/rfc/rfc2960.txt>. + + "SCTP is a reliable transport protocol operating on top of a + connectionless packet network such as IP. It offers the following + services to its users: + + -- acknowledged error-free non-duplicated transfer of user data, + -- data fragmentation to conform to discovered path MTU size, + -- sequenced delivery of user messages within multiple streams, + with an option for order-of-arrival delivery of individual user + messages, + -- optional bundling of multiple user messages into a single SCTP + packet, and + -- network-level fault tolerance through supporting of multi- + homing at either or both ends of an association." + + To compile this protocol support as a module, choose M here: the + module will be called sctp. + + If in doubt, say N. + +config SCTP_DBG_MSG + bool "SCTP: Debug messages" + depends on IP_SCTP + help + If you say Y, this will enable verbose debugging messages. + + If unsure, say N. However, if you are running into problems, use + this option to gather detailed trace information + +config SCTP_DBG_OBJCNT + bool "SCTP: Debug object counts" + depends on IP_SCTP + help + If you say Y, this will enable debugging support for counting the + type of objects that are currently allocated. This is useful for + identifying memory leaks. If the /proc filesystem is enabled this + debug information can be viewed by + 'cat /proc/net/sctp/sctp_dbg_objcnt' + + If unsure, say N + +choice + prompt "SCTP: Cookie HMAC Algorithm" + depends on IP_SCTP + default SCTP_HMAC_MD5 + help + HMAC algorithm to be used during association initialization. It + is strongly recommended to use HMAC-SHA1 or HMAC-MD5. See + configuration for Cryptographic API and enable those algorithms + to make usable by SCTP. + +config SCTP_HMAC_NONE + bool "None" + help + Choosing this disables the use of an HMAC during association + establishment. It is advised to use either HMAC-MD5 or HMAC-SHA1. + +config SCTP_HMAC_SHA1 + bool "HMAC-SHA1" + help + Enable the use of HMAC-SHA1 during association establishment. It + is advised to use either HMAC-MD5 or HMAC-SHA1. + +config SCTP_HMAC_MD5 + bool "HMAC-MD5" + help + Enable the use of HMAC-MD5 during association establishment. It is + advised to use either HMAC-MD5 or HMAC-SHA1. + +endchoice +endmenu diff --git a/net/sctp/Makefile b/net/sctp/Makefile new file mode 100644 index 000000000000..70c828bbe444 --- /dev/null +++ b/net/sctp/Makefile @@ -0,0 +1,17 @@ +# +# Makefile for SCTP support code. +# + +obj-$(CONFIG_IP_SCTP) += sctp.o + +sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \ + protocol.o endpointola.o associola.o \ + transport.o chunk.o sm_make_chunk.o ulpevent.o \ + inqueue.o outqueue.o ulpqueue.o command.o \ + tsnmap.o bind_addr.o socket.o primitive.o \ + output.o input.o debug.o ssnmap.o proc.o crc32c.o + +sctp-$(CONFIG_SCTP_DBG_OBJCNT) += objcnt.o +sctp-$(CONFIG_SYSCTL) += sysctl.o + +sctp-$(subst m,y,$(CONFIG_IPV6)) += ipv6.o diff --git a/net/sctp/associola.c b/net/sctp/associola.c new file mode 100644 index 000000000000..663843d97a92 --- /dev/null +++ b/net/sctp/associola.c @@ -0,0 +1,1205 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * This file is part of the SCTP kernel reference Implementation + * + * This module provides the abstraction for an SCTP association. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@us.ibm.com> + * Xingang Guo <xingang.guo@intel.com> + * Hui Huang <hui.huang@nokia.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Daisy Chang <daisyc@us.ibm.com> + * Ryan Layer <rmlayer@us.ibm.com> + * Kevin Gao <kevin.gao@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/fcntl.h> +#include <linux/poll.h> +#include <linux/init.h> +#include <linux/sched.h> + +#include <linux/slab.h> +#include <linux/in.h> +#include <net/ipv6.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Forward declarations for internal functions. */ +static void sctp_assoc_bh_rcv(struct sctp_association *asoc); + + +/* 1st Level Abstractions. */ + +/* Initialize a new association from provided memory. */ +static struct sctp_association *sctp_association_init(struct sctp_association *asoc, + const struct sctp_endpoint *ep, + const struct sock *sk, + sctp_scope_t scope, + int gfp) +{ + struct sctp_sock *sp; + int i; + + /* Retrieve the SCTP per socket area. */ + sp = sctp_sk((struct sock *)sk); + + /* Init all variables to a known value. */ + memset(asoc, 0, sizeof(struct sctp_association)); + + /* Discarding const is appropriate here. */ + asoc->ep = (struct sctp_endpoint *)ep; + sctp_endpoint_hold(asoc->ep); + + /* Hold the sock. */ + asoc->base.sk = (struct sock *)sk; + sock_hold(asoc->base.sk); + + /* Initialize the common base substructure. */ + asoc->base.type = SCTP_EP_TYPE_ASSOCIATION; + + /* Initialize the object handling fields. */ + atomic_set(&asoc->base.refcnt, 1); + asoc->base.dead = 0; + asoc->base.malloced = 0; + + /* Initialize the bind addr area. */ + sctp_bind_addr_init(&asoc->base.bind_addr, ep->base.bind_addr.port); + rwlock_init(&asoc->base.addr_lock); + + asoc->state = SCTP_STATE_CLOSED; + + /* Set these values from the socket values, a conversion between + * millsecons to seconds/microseconds must also be done. + */ + asoc->cookie_life.tv_sec = sp->assocparams.sasoc_cookie_life / 1000; + asoc->cookie_life.tv_usec = (sp->assocparams.sasoc_cookie_life % 1000) + * 1000; + asoc->pmtu = 0; + asoc->frag_point = 0; + + /* Set the association max_retrans and RTO values from the + * socket values. + */ + asoc->max_retrans = sp->assocparams.sasoc_asocmaxrxt; + asoc->rto_initial = msecs_to_jiffies(sp->rtoinfo.srto_initial); + asoc->rto_max = msecs_to_jiffies(sp->rtoinfo.srto_max); + asoc->rto_min = msecs_to_jiffies(sp->rtoinfo.srto_min); + + asoc->overall_error_count = 0; + + /* Initialize the maximum mumber of new data packets that can be sent + * in a burst. + */ + asoc->max_burst = sctp_max_burst; + + /* Copy things from the endpoint. */ + for (i = SCTP_EVENT_TIMEOUT_NONE; i < SCTP_NUM_TIMEOUT_TYPES; ++i) { + asoc->timeouts[i] = ep->timeouts[i]; + init_timer(&asoc->timers[i]); + asoc->timers[i].function = sctp_timer_events[i]; + asoc->timers[i].data = (unsigned long) asoc; + } + + /* Pull default initialization values from the sock options. + * Note: This assumes that the values have already been + * validated in the sock. + */ + asoc->c.sinit_max_instreams = sp->initmsg.sinit_max_instreams; + asoc->c.sinit_num_ostreams = sp->initmsg.sinit_num_ostreams; + asoc->max_init_attempts = sp->initmsg.sinit_max_attempts; + + asoc->max_init_timeo = + msecs_to_jiffies(sp->initmsg.sinit_max_init_timeo); + + /* Allocate storage for the ssnmap after the inbound and outbound + * streams have been negotiated during Init. + */ + asoc->ssnmap = NULL; + + /* Set the local window size for receive. + * This is also the rcvbuf space per association. + * RFC 6 - A SCTP receiver MUST be able to receive a minimum of + * 1500 bytes in one SCTP packet. + */ + if (sk->sk_rcvbuf < SCTP_DEFAULT_MINWINDOW) + asoc->rwnd = SCTP_DEFAULT_MINWINDOW; + else + asoc->rwnd = sk->sk_rcvbuf; + + asoc->a_rwnd = asoc->rwnd; + + asoc->rwnd_over = 0; + + /* Use my own max window until I learn something better. */ + asoc->peer.rwnd = SCTP_DEFAULT_MAXWINDOW; + + /* Set the sndbuf size for transmit. */ + asoc->sndbuf_used = 0; + + init_waitqueue_head(&asoc->wait); + + asoc->c.my_vtag = sctp_generate_tag(ep); + asoc->peer.i.init_tag = 0; /* INIT needs a vtag of 0. */ + asoc->c.peer_vtag = 0; + asoc->c.my_ttag = 0; + asoc->c.peer_ttag = 0; + asoc->c.my_port = ep->base.bind_addr.port; + + asoc->c.initial_tsn = sctp_generate_tsn(ep); + + asoc->next_tsn = asoc->c.initial_tsn; + + asoc->ctsn_ack_point = asoc->next_tsn - 1; + asoc->adv_peer_ack_point = asoc->ctsn_ack_point; + asoc->highest_sacked = asoc->ctsn_ack_point; + asoc->last_cwr_tsn = asoc->ctsn_ack_point; + asoc->unack_data = 0; + + SCTP_DEBUG_PRINTK("myctsnap for %s INIT as 0x%x.\n", + asoc->ep->debug_name, + asoc->ctsn_ack_point); + + /* ADDIP Section 4.1 Asconf Chunk Procedures + * + * When an endpoint has an ASCONF signaled change to be sent to the + * remote endpoint it should do the following: + * ... + * A2) a serial number should be assigned to the chunk. The serial + * number SHOULD be a monotonically increasing number. The serial + * numbers SHOULD be initialized at the start of the + * association to the same value as the initial TSN. + */ + asoc->addip_serial = asoc->c.initial_tsn; + + skb_queue_head_init(&asoc->addip_chunks); + + /* Make an empty list of remote transport addresses. */ + INIT_LIST_HEAD(&asoc->peer.transport_addr_list); + + /* RFC 2960 5.1 Normal Establishment of an Association + * + * After the reception of the first data chunk in an + * association the endpoint must immediately respond with a + * sack to acknowledge the data chunk. Subsequent + * acknowledgements should be done as described in Section + * 6.2. + * + * [We implement this by telling a new association that it + * already received one packet.] + */ + asoc->peer.sack_needed = 1; + + /* Assume that the peer recongizes ASCONF until reported otherwise + * via an ERROR chunk. + */ + asoc->peer.asconf_capable = 1; + + /* Create an input queue. */ + sctp_inq_init(&asoc->base.inqueue); + sctp_inq_set_th_handler(&asoc->base.inqueue, + (void (*)(void *))sctp_assoc_bh_rcv, + asoc); + + /* Create an output queue. */ + sctp_outq_init(asoc, &asoc->outqueue); + + if (!sctp_ulpq_init(&asoc->ulpq, asoc)) + goto fail_init; + + /* Set up the tsn tracking. */ + sctp_tsnmap_init(&asoc->peer.tsn_map, SCTP_TSN_MAP_SIZE, 0); + + asoc->need_ecne = 0; + + asoc->assoc_id = 0; + + /* Assume that peer would support both address types unless we are + * told otherwise. + */ + asoc->peer.ipv4_address = 1; + asoc->peer.ipv6_address = 1; + INIT_LIST_HEAD(&asoc->asocs); + + asoc->autoclose = sp->autoclose; + + asoc->default_stream = sp->default_stream; + asoc->default_ppid = sp->default_ppid; + asoc->default_flags = sp->default_flags; + asoc->default_context = sp->default_context; + asoc->default_timetolive = sp->default_timetolive; + + return asoc; + +fail_init: + sctp_endpoint_put(asoc->ep); + sock_put(asoc->base.sk); + return NULL; +} + +/* Allocate and initialize a new association */ +struct sctp_association *sctp_association_new(const struct sctp_endpoint *ep, + const struct sock *sk, + sctp_scope_t scope, int gfp) +{ + struct sctp_association *asoc; + + asoc = t_new(struct sctp_association, gfp); + if (!asoc) + goto fail; + + if (!sctp_association_init(asoc, ep, sk, scope, gfp)) + goto fail_init; + + asoc->base.malloced = 1; + SCTP_DBG_OBJCNT_INC(assoc); + + return asoc; + +fail_init: + kfree(asoc); +fail: + return NULL; +} + +/* Free this association if possible. There may still be users, so + * the actual deallocation may be delayed. + */ +void sctp_association_free(struct sctp_association *asoc) +{ + struct sock *sk = asoc->base.sk; + struct sctp_transport *transport; + struct list_head *pos, *temp; + int i; + + list_del(&asoc->asocs); + + /* Decrement the backlog value for a TCP-style listening socket. */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) + sk->sk_ack_backlog--; + + /* Mark as dead, so other users can know this structure is + * going away. + */ + asoc->base.dead = 1; + + /* Dispose of any data lying around in the outqueue. */ + sctp_outq_free(&asoc->outqueue); + + /* Dispose of any pending messages for the upper layer. */ + sctp_ulpq_free(&asoc->ulpq); + + /* Dispose of any pending chunks on the inqueue. */ + sctp_inq_free(&asoc->base.inqueue); + + /* Free ssnmap storage. */ + sctp_ssnmap_free(asoc->ssnmap); + + /* Clean up the bound address list. */ + sctp_bind_addr_free(&asoc->base.bind_addr); + + /* Do we need to go through all of our timers and + * delete them? To be safe we will try to delete all, but we + * should be able to go through and make a guess based + * on our state. + */ + for (i = SCTP_EVENT_TIMEOUT_NONE; i < SCTP_NUM_TIMEOUT_TYPES; ++i) { + if (timer_pending(&asoc->timers[i]) && + del_timer(&asoc->timers[i])) + sctp_association_put(asoc); + } + + /* Free peer's cached cookie. */ + if (asoc->peer.cookie) { + kfree(asoc->peer.cookie); + } + + /* Release the transport structures. */ + list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + list_del(pos); + sctp_transport_free(transport); + } + + /* Free any cached ASCONF_ACK chunk. */ + if (asoc->addip_last_asconf_ack) + sctp_chunk_free(asoc->addip_last_asconf_ack); + + /* Free any cached ASCONF chunk. */ + if (asoc->addip_last_asconf) + sctp_chunk_free(asoc->addip_last_asconf); + + sctp_association_put(asoc); +} + +/* Cleanup and free up an association. */ +static void sctp_association_destroy(struct sctp_association *asoc) +{ + SCTP_ASSERT(asoc->base.dead, "Assoc is not dead", return); + + sctp_endpoint_put(asoc->ep); + sock_put(asoc->base.sk); + + if (asoc->assoc_id != 0) { + spin_lock_bh(&sctp_assocs_id_lock); + idr_remove(&sctp_assocs_id, asoc->assoc_id); + spin_unlock_bh(&sctp_assocs_id_lock); + } + + if (asoc->base.malloced) { + kfree(asoc); + SCTP_DBG_OBJCNT_DEC(assoc); + } +} + +/* Change the primary destination address for the peer. */ +void sctp_assoc_set_primary(struct sctp_association *asoc, + struct sctp_transport *transport) +{ + asoc->peer.primary_path = transport; + + /* Set a default msg_name for events. */ + memcpy(&asoc->peer.primary_addr, &transport->ipaddr, + sizeof(union sctp_addr)); + + /* If the primary path is changing, assume that the + * user wants to use this new path. + */ + if (transport->active) + asoc->peer.active_path = transport; + + /* + * SFR-CACC algorithm: + * Upon the receipt of a request to change the primary + * destination address, on the data structure for the new + * primary destination, the sender MUST do the following: + * + * 1) If CHANGEOVER_ACTIVE is set, then there was a switch + * to this destination address earlier. The sender MUST set + * CYCLING_CHANGEOVER to indicate that this switch is a + * double switch to the same destination address. + */ + if (transport->cacc.changeover_active) + transport->cacc.cycling_changeover = 1; + + /* 2) The sender MUST set CHANGEOVER_ACTIVE to indicate that + * a changeover has occurred. + */ + transport->cacc.changeover_active = 1; + + /* 3) The sender MUST store the next TSN to be sent in + * next_tsn_at_change. + */ + transport->cacc.next_tsn_at_change = asoc->next_tsn; +} + +/* Add a transport address to an association. */ +struct sctp_transport *sctp_assoc_add_peer(struct sctp_association *asoc, + const union sctp_addr *addr, + int gfp) +{ + struct sctp_transport *peer; + struct sctp_sock *sp; + unsigned short port; + + sp = sctp_sk(asoc->base.sk); + + /* AF_INET and AF_INET6 share common port field. */ + port = addr->v4.sin_port; + + /* Set the port if it has not been set yet. */ + if (0 == asoc->peer.port) + asoc->peer.port = port; + + /* Check to see if this is a duplicate. */ + peer = sctp_assoc_lookup_paddr(asoc, addr); + if (peer) + return peer; + + peer = sctp_transport_new(addr, gfp); + if (!peer) + return NULL; + + sctp_transport_set_owner(peer, asoc); + + /* Initialize the pmtu of the transport. */ + sctp_transport_pmtu(peer); + + /* If this is the first transport addr on this association, + * initialize the association PMTU to the peer's PMTU. + * If not and the current association PMTU is higher than the new + * peer's PMTU, reset the association PMTU to the new peer's PMTU. + */ + if (asoc->pmtu) + asoc->pmtu = min_t(int, peer->pmtu, asoc->pmtu); + else + asoc->pmtu = peer->pmtu; + + SCTP_DEBUG_PRINTK("sctp_assoc_add_peer:association %p PMTU set to " + "%d\n", asoc, asoc->pmtu); + + asoc->frag_point = sctp_frag_point(sp, asoc->pmtu); + + /* The asoc->peer.port might not be meaningful yet, but + * initialize the packet structure anyway. + */ + sctp_packet_init(&peer->packet, peer, asoc->base.bind_addr.port, + asoc->peer.port); + + /* 7.2.1 Slow-Start + * + * o The initial cwnd before DATA transmission or after a sufficiently + * long idle period MUST be set to + * min(4*MTU, max(2*MTU, 4380 bytes)) + * + * o The initial value of ssthresh MAY be arbitrarily high + * (for example, implementations MAY use the size of the + * receiver advertised window). + */ + peer->cwnd = min(4*asoc->pmtu, max_t(__u32, 2*asoc->pmtu, 4380)); + + /* At this point, we may not have the receiver's advertised window, + * so initialize ssthresh to the default value and it will be set + * later when we process the INIT. + */ + peer->ssthresh = SCTP_DEFAULT_MAXWINDOW; + + peer->partial_bytes_acked = 0; + peer->flight_size = 0; + + /* By default, enable heartbeat for peer address. */ + peer->hb_allowed = 1; + + /* Initialize the peer's heartbeat interval based on the + * sock configured value. + */ + peer->hb_interval = msecs_to_jiffies(sp->paddrparam.spp_hbinterval); + + /* Set the path max_retrans. */ + peer->max_retrans = sp->paddrparam.spp_pathmaxrxt; + + /* Set the transport's RTO.initial value */ + peer->rto = asoc->rto_initial; + + /* Attach the remote transport to our asoc. */ + list_add_tail(&peer->transports, &asoc->peer.transport_addr_list); + + /* If we do not yet have a primary path, set one. */ + if (!asoc->peer.primary_path) { + sctp_assoc_set_primary(asoc, peer); + asoc->peer.retran_path = peer; + } + + if (asoc->peer.active_path == asoc->peer.retran_path) + asoc->peer.retran_path = peer; + + return peer; +} + +/* Delete a transport address from an association. */ +void sctp_assoc_del_peer(struct sctp_association *asoc, + const union sctp_addr *addr) +{ + struct list_head *pos; + struct list_head *temp; + struct sctp_transport *peer = NULL; + struct sctp_transport *transport; + + list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + if (sctp_cmp_addr_exact(addr, &transport->ipaddr)) { + peer = transport; + list_del(pos); + break; + } + } + + /* The address we want delete is not in the association. */ + if (!peer) + return; + + /* Get the first transport of asoc. */ + pos = asoc->peer.transport_addr_list.next; + transport = list_entry(pos, struct sctp_transport, transports); + + /* Update any entries that match the peer to be deleted. */ + if (asoc->peer.primary_path == peer) + sctp_assoc_set_primary(asoc, transport); + if (asoc->peer.active_path == peer) + asoc->peer.active_path = transport; + if (asoc->peer.retran_path == peer) + asoc->peer.retran_path = transport; + if (asoc->peer.last_data_from == peer) + asoc->peer.last_data_from = transport; + + sctp_transport_free(peer); +} + +/* Lookup a transport by address. */ +struct sctp_transport *sctp_assoc_lookup_paddr( + const struct sctp_association *asoc, + const union sctp_addr *address) +{ + struct sctp_transport *t; + struct list_head *pos; + + /* Cycle through all transports searching for a peer address. */ + + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + if (sctp_cmp_addr_exact(address, &t->ipaddr)) + return t; + } + + return NULL; +} + +/* Engage in transport control operations. + * Mark the transport up or down and send a notification to the user. + * Select and update the new active and retran paths. + */ +void sctp_assoc_control_transport(struct sctp_association *asoc, + struct sctp_transport *transport, + sctp_transport_cmd_t command, + sctp_sn_error_t error) +{ + struct sctp_transport *t = NULL; + struct sctp_transport *first; + struct sctp_transport *second; + struct sctp_ulpevent *event; + struct list_head *pos; + int spc_state = 0; + + /* Record the transition on the transport. */ + switch (command) { + case SCTP_TRANSPORT_UP: + transport->active = SCTP_ACTIVE; + spc_state = SCTP_ADDR_AVAILABLE; + break; + + case SCTP_TRANSPORT_DOWN: + transport->active = SCTP_INACTIVE; + spc_state = SCTP_ADDR_UNREACHABLE; + break; + + default: + return; + }; + + /* Generate and send a SCTP_PEER_ADDR_CHANGE notification to the + * user. + */ + event = sctp_ulpevent_make_peer_addr_change(asoc, + (struct sockaddr_storage *) &transport->ipaddr, + 0, spc_state, error, GFP_ATOMIC); + if (event) + sctp_ulpq_tail_event(&asoc->ulpq, event); + + /* Select new active and retran paths. */ + + /* Look for the two most recently used active transports. + * + * This code produces the wrong ordering whenever jiffies + * rolls over, but we still get usable transports, so we don't + * worry about it. + */ + first = NULL; second = NULL; + + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + + if (!t->active) + continue; + if (!first || t->last_time_heard > first->last_time_heard) { + second = first; + first = t; + } + if (!second || t->last_time_heard > second->last_time_heard) + second = t; + } + + /* RFC 2960 6.4 Multi-Homed SCTP Endpoints + * + * By default, an endpoint should always transmit to the + * primary path, unless the SCTP user explicitly specifies the + * destination transport address (and possibly source + * transport address) to use. + * + * [If the primary is active but not most recent, bump the most + * recently used transport.] + */ + if (asoc->peer.primary_path->active && + first != asoc->peer.primary_path) { + second = first; + first = asoc->peer.primary_path; + } + + /* If we failed to find a usable transport, just camp on the + * primary, even if it is inactive. + */ + if (!first) { + first = asoc->peer.primary_path; + second = asoc->peer.primary_path; + } + + /* Set the active and retran transports. */ + asoc->peer.active_path = first; + asoc->peer.retran_path = second; +} + +/* Hold a reference to an association. */ +void sctp_association_hold(struct sctp_association *asoc) +{ + atomic_inc(&asoc->base.refcnt); +} + +/* Release a reference to an association and cleanup + * if there are no more references. + */ +void sctp_association_put(struct sctp_association *asoc) +{ + if (atomic_dec_and_test(&asoc->base.refcnt)) + sctp_association_destroy(asoc); +} + +/* Allocate the next TSN, Transmission Sequence Number, for the given + * association. + */ +__u32 sctp_association_get_next_tsn(struct sctp_association *asoc) +{ + /* From Section 1.6 Serial Number Arithmetic: + * Transmission Sequence Numbers wrap around when they reach + * 2**32 - 1. That is, the next TSN a DATA chunk MUST use + * after transmitting TSN = 2*32 - 1 is TSN = 0. + */ + __u32 retval = asoc->next_tsn; + asoc->next_tsn++; + asoc->unack_data++; + + return retval; +} + +/* Compare two addresses to see if they match. Wildcard addresses + * only match themselves. + */ +int sctp_cmp_addr_exact(const union sctp_addr *ss1, + const union sctp_addr *ss2) +{ + struct sctp_af *af; + + af = sctp_get_af_specific(ss1->sa.sa_family); + if (unlikely(!af)) + return 0; + + return af->cmp_addr(ss1, ss2); +} + +/* Return an ecne chunk to get prepended to a packet. + * Note: We are sly and return a shared, prealloced chunk. FIXME: + * No we don't, but we could/should. + */ +struct sctp_chunk *sctp_get_ecne_prepend(struct sctp_association *asoc) +{ + struct sctp_chunk *chunk; + + /* Send ECNE if needed. + * Not being able to allocate a chunk here is not deadly. + */ + if (asoc->need_ecne) + chunk = sctp_make_ecne(asoc, asoc->last_ecne_tsn); + else + chunk = NULL; + + return chunk; +} + +/* + * Find which transport this TSN was sent on. + */ +struct sctp_transport *sctp_assoc_lookup_tsn(struct sctp_association *asoc, + __u32 tsn) +{ + struct sctp_transport *active; + struct sctp_transport *match; + struct list_head *entry, *pos; + struct sctp_transport *transport; + struct sctp_chunk *chunk; + __u32 key = htonl(tsn); + + match = NULL; + + /* + * FIXME: In general, find a more efficient data structure for + * searching. + */ + + /* + * The general strategy is to search each transport's transmitted + * list. Return which transport this TSN lives on. + * + * Let's be hopeful and check the active_path first. + * Another optimization would be to know if there is only one + * outbound path and not have to look for the TSN at all. + * + */ + + active = asoc->peer.active_path; + + list_for_each(entry, &active->transmitted) { + chunk = list_entry(entry, struct sctp_chunk, transmitted_list); + + if (key == chunk->subh.data_hdr->tsn) { + match = active; + goto out; + } + } + + /* If not found, go search all the other transports. */ + list_for_each(pos, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + + if (transport == active) + break; + list_for_each(entry, &transport->transmitted) { + chunk = list_entry(entry, struct sctp_chunk, + transmitted_list); + if (key == chunk->subh.data_hdr->tsn) { + match = transport; + goto out; + } + } + } +out: + return match; +} + +/* Is this the association we are looking for? */ +struct sctp_transport *sctp_assoc_is_match(struct sctp_association *asoc, + const union sctp_addr *laddr, + const union sctp_addr *paddr) +{ + struct sctp_transport *transport; + + sctp_read_lock(&asoc->base.addr_lock); + + if ((asoc->base.bind_addr.port == laddr->v4.sin_port) && + (asoc->peer.port == paddr->v4.sin_port)) { + transport = sctp_assoc_lookup_paddr(asoc, paddr); + if (!transport) + goto out; + + if (sctp_bind_addr_match(&asoc->base.bind_addr, laddr, + sctp_sk(asoc->base.sk))) + goto out; + } + transport = NULL; + +out: + sctp_read_unlock(&asoc->base.addr_lock); + return transport; +} + +/* Do delayed input processing. This is scheduled by sctp_rcv(). */ +static void sctp_assoc_bh_rcv(struct sctp_association *asoc) +{ + struct sctp_endpoint *ep; + struct sctp_chunk *chunk; + struct sock *sk; + struct sctp_inq *inqueue; + int state; + sctp_subtype_t subtype; + int error = 0; + + /* The association should be held so we should be safe. */ + ep = asoc->ep; + sk = asoc->base.sk; + + inqueue = &asoc->base.inqueue; + sctp_association_hold(asoc); + while (NULL != (chunk = sctp_inq_pop(inqueue))) { + state = asoc->state; + subtype = SCTP_ST_CHUNK(chunk->chunk_hdr->type); + + /* Remember where the last DATA chunk came from so we + * know where to send the SACK. + */ + if (sctp_chunk_is_data(chunk)) + asoc->peer.last_data_from = chunk->transport; + else + SCTP_INC_STATS(SCTP_MIB_INCTRLCHUNKS); + + if (chunk->transport) + chunk->transport->last_time_heard = jiffies; + + /* Run through the state machine. */ + error = sctp_do_sm(SCTP_EVENT_T_CHUNK, subtype, + state, ep, asoc, chunk, GFP_ATOMIC); + + /* Check to see if the association is freed in response to + * the incoming chunk. If so, get out of the while loop. + */ + if (asoc->base.dead) + break; + + /* If there is an error on chunk, discard this packet. */ + if (error && chunk) + chunk->pdiscard = 1; + } + sctp_association_put(asoc); +} + +/* This routine moves an association from its old sk to a new sk. */ +void sctp_assoc_migrate(struct sctp_association *assoc, struct sock *newsk) +{ + struct sctp_sock *newsp = sctp_sk(newsk); + struct sock *oldsk = assoc->base.sk; + + /* Delete the association from the old endpoint's list of + * associations. + */ + list_del_init(&assoc->asocs); + + /* Decrement the backlog value for a TCP-style socket. */ + if (sctp_style(oldsk, TCP)) + oldsk->sk_ack_backlog--; + + /* Release references to the old endpoint and the sock. */ + sctp_endpoint_put(assoc->ep); + sock_put(assoc->base.sk); + + /* Get a reference to the new endpoint. */ + assoc->ep = newsp->ep; + sctp_endpoint_hold(assoc->ep); + + /* Get a reference to the new sock. */ + assoc->base.sk = newsk; + sock_hold(assoc->base.sk); + + /* Add the association to the new endpoint's list of associations. */ + sctp_endpoint_add_asoc(newsp->ep, assoc); +} + +/* Update an association (possibly from unexpected COOKIE-ECHO processing). */ +void sctp_assoc_update(struct sctp_association *asoc, + struct sctp_association *new) +{ + struct sctp_transport *trans; + struct list_head *pos, *temp; + + /* Copy in new parameters of peer. */ + asoc->c = new->c; + asoc->peer.rwnd = new->peer.rwnd; + asoc->peer.sack_needed = new->peer.sack_needed; + asoc->peer.i = new->peer.i; + sctp_tsnmap_init(&asoc->peer.tsn_map, SCTP_TSN_MAP_SIZE, + asoc->peer.i.initial_tsn); + + /* Remove any peer addresses not present in the new association. */ + list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { + trans = list_entry(pos, struct sctp_transport, transports); + if (!sctp_assoc_lookup_paddr(new, &trans->ipaddr)) + sctp_assoc_del_peer(asoc, &trans->ipaddr); + } + + /* If the case is A (association restart), use + * initial_tsn as next_tsn. If the case is B, use + * current next_tsn in case data sent to peer + * has been discarded and needs retransmission. + */ + if (asoc->state >= SCTP_STATE_ESTABLISHED) { + asoc->next_tsn = new->next_tsn; + asoc->ctsn_ack_point = new->ctsn_ack_point; + asoc->adv_peer_ack_point = new->adv_peer_ack_point; + + /* Reinitialize SSN for both local streams + * and peer's streams. + */ + sctp_ssnmap_clear(asoc->ssnmap); + + } else { + /* Add any peer addresses from the new association. */ + list_for_each(pos, &new->peer.transport_addr_list) { + trans = list_entry(pos, struct sctp_transport, + transports); + if (!sctp_assoc_lookup_paddr(asoc, &trans->ipaddr)) + sctp_assoc_add_peer(asoc, &trans->ipaddr, + GFP_ATOMIC); + } + + asoc->ctsn_ack_point = asoc->next_tsn - 1; + asoc->adv_peer_ack_point = asoc->ctsn_ack_point; + if (!asoc->ssnmap) { + /* Move the ssnmap. */ + asoc->ssnmap = new->ssnmap; + new->ssnmap = NULL; + } + } +} + +/* Update the retran path for sending a retransmitted packet. + * Round-robin through the active transports, else round-robin + * through the inactive transports as this is the next best thing + * we can try. + */ +void sctp_assoc_update_retran_path(struct sctp_association *asoc) +{ + struct sctp_transport *t, *next; + struct list_head *head = &asoc->peer.transport_addr_list; + struct list_head *pos; + + /* Find the next transport in a round-robin fashion. */ + t = asoc->peer.retran_path; + pos = &t->transports; + next = NULL; + + while (1) { + /* Skip the head. */ + if (pos->next == head) + pos = head->next; + else + pos = pos->next; + + t = list_entry(pos, struct sctp_transport, transports); + + /* Try to find an active transport. */ + + if (t->active) { + break; + } else { + /* Keep track of the next transport in case + * we don't find any active transport. + */ + if (!next) + next = t; + } + + /* We have exhausted the list, but didn't find any + * other active transports. If so, use the next + * transport. + */ + if (t == asoc->peer.retran_path) { + t = next; + break; + } + } + + asoc->peer.retran_path = t; +} + +/* Choose the transport for sending a SHUTDOWN packet. */ +struct sctp_transport *sctp_assoc_choose_shutdown_transport( + struct sctp_association *asoc) +{ + /* If this is the first time SHUTDOWN is sent, use the active path, + * else use the retran path. If the last SHUTDOWN was sent over the + * retran path, update the retran path and use it. + */ + if (!asoc->shutdown_last_sent_to) + return asoc->peer.active_path; + else { + if (asoc->shutdown_last_sent_to == asoc->peer.retran_path) + sctp_assoc_update_retran_path(asoc); + return asoc->peer.retran_path; + } + +} + +/* Update the association's pmtu and frag_point by going through all the + * transports. This routine is called when a transport's PMTU has changed. + */ +void sctp_assoc_sync_pmtu(struct sctp_association *asoc) +{ + struct sctp_transport *t; + struct list_head *pos; + __u32 pmtu = 0; + + if (!asoc) + return; + + /* Get the lowest pmtu of all the transports. */ + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + if (!pmtu || (t->pmtu < pmtu)) + pmtu = t->pmtu; + } + + if (pmtu) { + struct sctp_sock *sp = sctp_sk(asoc->base.sk); + asoc->pmtu = pmtu; + asoc->frag_point = sctp_frag_point(sp, pmtu); + } + + SCTP_DEBUG_PRINTK("%s: asoc:%p, pmtu:%d, frag_point:%d\n", + __FUNCTION__, asoc, asoc->pmtu, asoc->frag_point); +} + +/* Should we send a SACK to update our peer? */ +static inline int sctp_peer_needs_update(struct sctp_association *asoc) +{ + switch (asoc->state) { + case SCTP_STATE_ESTABLISHED: + case SCTP_STATE_SHUTDOWN_PENDING: + case SCTP_STATE_SHUTDOWN_RECEIVED: + case SCTP_STATE_SHUTDOWN_SENT: + if ((asoc->rwnd > asoc->a_rwnd) && + ((asoc->rwnd - asoc->a_rwnd) >= + min_t(__u32, (asoc->base.sk->sk_rcvbuf >> 1), asoc->pmtu))) + return 1; + break; + default: + break; + } + return 0; +} + +/* Increase asoc's rwnd by len and send any window update SACK if needed. */ +void sctp_assoc_rwnd_increase(struct sctp_association *asoc, unsigned len) +{ + struct sctp_chunk *sack; + struct timer_list *timer; + + if (asoc->rwnd_over) { + if (asoc->rwnd_over >= len) { + asoc->rwnd_over -= len; + } else { + asoc->rwnd += (len - asoc->rwnd_over); + asoc->rwnd_over = 0; + } + } else { + asoc->rwnd += len; + } + + SCTP_DEBUG_PRINTK("%s: asoc %p rwnd increased by %d to (%u, %u) " + "- %u\n", __FUNCTION__, asoc, len, asoc->rwnd, + asoc->rwnd_over, asoc->a_rwnd); + + /* Send a window update SACK if the rwnd has increased by at least the + * minimum of the association's PMTU and half of the receive buffer. + * The algorithm used is similar to the one described in + * Section 4.2.3.3 of RFC 1122. + */ + if (sctp_peer_needs_update(asoc)) { + asoc->a_rwnd = asoc->rwnd; + SCTP_DEBUG_PRINTK("%s: Sending window update SACK- asoc: %p " + "rwnd: %u a_rwnd: %u\n", __FUNCTION__, + asoc, asoc->rwnd, asoc->a_rwnd); + sack = sctp_make_sack(asoc); + if (!sack) + return; + + asoc->peer.sack_needed = 0; + + sctp_outq_tail(&asoc->outqueue, sack); + + /* Stop the SACK timer. */ + timer = &asoc->timers[SCTP_EVENT_TIMEOUT_SACK]; + if (timer_pending(timer) && del_timer(timer)) + sctp_association_put(asoc); + } +} + +/* Decrease asoc's rwnd by len. */ +void sctp_assoc_rwnd_decrease(struct sctp_association *asoc, unsigned len) +{ + SCTP_ASSERT(asoc->rwnd, "rwnd zero", return); + SCTP_ASSERT(!asoc->rwnd_over, "rwnd_over not zero", return); + if (asoc->rwnd >= len) { + asoc->rwnd -= len; + } else { + asoc->rwnd_over = len - asoc->rwnd; + asoc->rwnd = 0; + } + SCTP_DEBUG_PRINTK("%s: asoc %p rwnd decreased by %d to (%u, %u)\n", + __FUNCTION__, asoc, len, asoc->rwnd, + asoc->rwnd_over); +} + +/* Build the bind address list for the association based on info from the + * local endpoint and the remote peer. + */ +int sctp_assoc_set_bind_addr_from_ep(struct sctp_association *asoc, int gfp) +{ + sctp_scope_t scope; + int flags; + + /* Use scoping rules to determine the subset of addresses from + * the endpoint. + */ + scope = sctp_scope(&asoc->peer.active_path->ipaddr); + flags = (PF_INET6 == asoc->base.sk->sk_family) ? SCTP_ADDR6_ALLOWED : 0; + if (asoc->peer.ipv4_address) + flags |= SCTP_ADDR4_PEERSUPP; + if (asoc->peer.ipv6_address) + flags |= SCTP_ADDR6_PEERSUPP; + + return sctp_bind_addr_copy(&asoc->base.bind_addr, + &asoc->ep->base.bind_addr, + scope, gfp, flags); +} + +/* Build the association's bind address list from the cookie. */ +int sctp_assoc_set_bind_addr_from_cookie(struct sctp_association *asoc, + struct sctp_cookie *cookie, int gfp) +{ + int var_size2 = ntohs(cookie->peer_init->chunk_hdr.length); + int var_size3 = cookie->raw_addr_list_len; + __u8 *raw = (__u8 *)cookie->peer_init + var_size2; + + return sctp_raw_to_bind_addrs(&asoc->base.bind_addr, raw, var_size3, + asoc->ep->base.bind_addr.port, gfp); +} + +/* Lookup laddr in the bind address list of an association. */ +int sctp_assoc_lookup_laddr(struct sctp_association *asoc, + const union sctp_addr *laddr) +{ + int found; + + sctp_read_lock(&asoc->base.addr_lock); + if ((asoc->base.bind_addr.port == ntohs(laddr->v4.sin_port)) && + sctp_bind_addr_match(&asoc->base.bind_addr, laddr, + sctp_sk(asoc->base.sk))) { + found = 1; + goto out; + } + + found = 0; +out: + sctp_read_unlock(&asoc->base.addr_lock); + return found; +} diff --git a/net/sctp/bind_addr.c b/net/sctp/bind_addr.c new file mode 100644 index 000000000000..f90eadfb60a2 --- /dev/null +++ b/net/sctp/bind_addr.c @@ -0,0 +1,417 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2003 + * Copyright (c) Cisco 1999,2000 + * Copyright (c) Motorola 1999,2000,2001 + * Copyright (c) La Monte H.P. Yarroll 2001 + * + * This file is part of the SCTP kernel reference implementation. + * + * A collection class to handle the storage of transport addresses. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@us.ibm.com> + * Daisy Chang <daisyc@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/in.h> +#include <net/sock.h> +#include <net/ipv6.h> +#include <net/if_inet6.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Forward declarations for internal helpers. */ +static int sctp_copy_one_addr(struct sctp_bind_addr *, union sctp_addr *, + sctp_scope_t scope, int gfp, int flags); +static void sctp_bind_addr_clean(struct sctp_bind_addr *); + +/* First Level Abstractions. */ + +/* Copy 'src' to 'dest' taking 'scope' into account. Omit addresses + * in 'src' which have a broader scope than 'scope'. + */ +int sctp_bind_addr_copy(struct sctp_bind_addr *dest, + const struct sctp_bind_addr *src, + sctp_scope_t scope, int gfp, int flags) +{ + struct sctp_sockaddr_entry *addr; + struct list_head *pos; + int error = 0; + + /* All addresses share the same port. */ + dest->port = src->port; + + /* Extract the addresses which are relevant for this scope. */ + list_for_each(pos, &src->address_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + error = sctp_copy_one_addr(dest, &addr->a, scope, + gfp, flags); + if (error < 0) + goto out; + } + + /* If there are no addresses matching the scope and + * this is global scope, try to get a link scope address, with + * the assumption that we must be sitting behind a NAT. + */ + if (list_empty(&dest->address_list) && (SCTP_SCOPE_GLOBAL == scope)) { + list_for_each(pos, &src->address_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, + list); + error = sctp_copy_one_addr(dest, &addr->a, + SCTP_SCOPE_LINK, gfp, + flags); + if (error < 0) + goto out; + } + } + +out: + if (error) + sctp_bind_addr_clean(dest); + + return error; +} + +/* Initialize the SCTP_bind_addr structure for either an endpoint or + * an association. + */ +void sctp_bind_addr_init(struct sctp_bind_addr *bp, __u16 port) +{ + bp->malloced = 0; + + INIT_LIST_HEAD(&bp->address_list); + bp->port = port; +} + +/* Dispose of the address list. */ +static void sctp_bind_addr_clean(struct sctp_bind_addr *bp) +{ + struct sctp_sockaddr_entry *addr; + struct list_head *pos, *temp; + + /* Empty the bind address list. */ + list_for_each_safe(pos, temp, &bp->address_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + list_del(pos); + kfree(addr); + SCTP_DBG_OBJCNT_DEC(addr); + } +} + +/* Dispose of an SCTP_bind_addr structure */ +void sctp_bind_addr_free(struct sctp_bind_addr *bp) +{ + /* Empty the bind address list. */ + sctp_bind_addr_clean(bp); + + if (bp->malloced) { + kfree(bp); + SCTP_DBG_OBJCNT_DEC(bind_addr); + } +} + +/* Add an address to the bind address list in the SCTP_bind_addr structure. */ +int sctp_add_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *new, + int gfp) +{ + struct sctp_sockaddr_entry *addr; + + /* Add the address to the bind address list. */ + addr = t_new(struct sctp_sockaddr_entry, gfp); + if (!addr) + return -ENOMEM; + + memcpy(&addr->a, new, sizeof(*new)); + + /* Fix up the port if it has not yet been set. + * Both v4 and v6 have the port at the same offset. + */ + if (!addr->a.v4.sin_port) + addr->a.v4.sin_port = bp->port; + + INIT_LIST_HEAD(&addr->list); + list_add_tail(&addr->list, &bp->address_list); + SCTP_DBG_OBJCNT_INC(addr); + + return 0; +} + +/* Delete an address from the bind address list in the SCTP_bind_addr + * structure. + */ +int sctp_del_bind_addr(struct sctp_bind_addr *bp, union sctp_addr *del_addr) +{ + struct list_head *pos, *temp; + struct sctp_sockaddr_entry *addr; + + list_for_each_safe(pos, temp, &bp->address_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + if (sctp_cmp_addr_exact(&addr->a, del_addr)) { + /* Found the exact match. */ + list_del(pos); + kfree(addr); + SCTP_DBG_OBJCNT_DEC(addr); + + return 0; + } + } + + return -EINVAL; +} + +/* Create a network byte-order representation of all the addresses + * formated as SCTP parameters. + * + * The second argument is the return value for the length. + */ +union sctp_params sctp_bind_addrs_to_raw(const struct sctp_bind_addr *bp, + int *addrs_len, int gfp) +{ + union sctp_params addrparms; + union sctp_params retval; + int addrparms_len; + union sctp_addr_param rawaddr; + int len; + struct sctp_sockaddr_entry *addr; + struct list_head *pos; + struct sctp_af *af; + + addrparms_len = 0; + len = 0; + + /* Allocate enough memory at once. */ + list_for_each(pos, &bp->address_list) { + len += sizeof(union sctp_addr_param); + } + + /* Don't even bother embedding an address if there + * is only one. + */ + if (len == sizeof(union sctp_addr_param)) { + retval.v = NULL; + goto end_raw; + } + + retval.v = kmalloc(len, gfp); + if (!retval.v) + goto end_raw; + + addrparms = retval; + + list_for_each(pos, &bp->address_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + af = sctp_get_af_specific(addr->a.v4.sin_family); + len = af->to_addr_param(&addr->a, &rawaddr); + memcpy(addrparms.v, &rawaddr, len); + addrparms.v += len; + addrparms_len += len; + } + +end_raw: + *addrs_len = addrparms_len; + return retval; +} + +/* + * Create an address list out of the raw address list format (IPv4 and IPv6 + * address parameters). + */ +int sctp_raw_to_bind_addrs(struct sctp_bind_addr *bp, __u8 *raw_addr_list, + int addrs_len, __u16 port, int gfp) +{ + union sctp_addr_param *rawaddr; + struct sctp_paramhdr *param; + union sctp_addr addr; + int retval = 0; + int len; + struct sctp_af *af; + + /* Convert the raw address to standard address format */ + while (addrs_len) { + param = (struct sctp_paramhdr *)raw_addr_list; + rawaddr = (union sctp_addr_param *)raw_addr_list; + + af = sctp_get_af_specific(param_type2af(param->type)); + if (unlikely(!af)) { + retval = -EINVAL; + sctp_bind_addr_clean(bp); + break; + } + + af->from_addr_param(&addr, rawaddr, port, 0); + retval = sctp_add_bind_addr(bp, &addr, gfp); + if (retval) { + /* Can't finish building the list, clean up. */ + sctp_bind_addr_clean(bp); + break; + } + + len = ntohs(param->length); + addrs_len -= len; + raw_addr_list += len; + } + + return retval; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* Does this contain a specified address? Allow wildcarding. */ +int sctp_bind_addr_match(struct sctp_bind_addr *bp, + const union sctp_addr *addr, + struct sctp_sock *opt) +{ + struct sctp_sockaddr_entry *laddr; + struct list_head *pos; + + list_for_each(pos, &bp->address_list) { + laddr = list_entry(pos, struct sctp_sockaddr_entry, list); + if (opt->pf->cmp_addr(&laddr->a, addr, opt)) + return 1; + } + + return 0; +} + +/* Find the first address in the bind address list that is not present in + * the addrs packed array. + */ +union sctp_addr *sctp_find_unmatch_addr(struct sctp_bind_addr *bp, + const union sctp_addr *addrs, + int addrcnt, + struct sctp_sock *opt) +{ + struct sctp_sockaddr_entry *laddr; + union sctp_addr *addr; + void *addr_buf; + struct sctp_af *af; + struct list_head *pos; + int i; + + list_for_each(pos, &bp->address_list) { + laddr = list_entry(pos, struct sctp_sockaddr_entry, list); + + addr_buf = (union sctp_addr *)addrs; + for (i = 0; i < addrcnt; i++) { + addr = (union sctp_addr *)addr_buf; + af = sctp_get_af_specific(addr->v4.sin_family); + if (!af) + return NULL; + + if (opt->pf->cmp_addr(&laddr->a, addr, opt)) + break; + + addr_buf += af->sockaddr_len; + } + if (i == addrcnt) + return &laddr->a; + } + + return NULL; +} + +/* Copy out addresses from the global local address list. */ +static int sctp_copy_one_addr(struct sctp_bind_addr *dest, + union sctp_addr *addr, + sctp_scope_t scope, int gfp, int flags) +{ + int error = 0; + + if (sctp_is_any(addr)) { + error = sctp_copy_local_addr_list(dest, scope, gfp, flags); + } else if (sctp_in_scope(addr, scope)) { + /* Now that the address is in scope, check to see if + * the address type is supported by local sock as + * well as the remote peer. + */ + if ((((AF_INET == addr->sa.sa_family) && + (flags & SCTP_ADDR4_PEERSUPP))) || + (((AF_INET6 == addr->sa.sa_family) && + (flags & SCTP_ADDR6_ALLOWED) && + (flags & SCTP_ADDR6_PEERSUPP)))) + error = sctp_add_bind_addr(dest, addr, gfp); + } + + return error; +} + +/* Is this a wildcard address? */ +int sctp_is_any(const union sctp_addr *addr) +{ + struct sctp_af *af = sctp_get_af_specific(addr->sa.sa_family); + if (!af) + return 0; + return af->is_any(addr); +} + +/* Is 'addr' valid for 'scope'? */ +int sctp_in_scope(const union sctp_addr *addr, sctp_scope_t scope) +{ + sctp_scope_t addr_scope = sctp_scope(addr); + + /* The unusable SCTP addresses will not be considered with + * any defined scopes. + */ + if (SCTP_SCOPE_UNUSABLE == addr_scope) + return 0; + /* + * For INIT and INIT-ACK address list, let L be the level of + * of requested destination address, sender and receiver + * SHOULD include all of its addresses with level greater + * than or equal to L. + */ + if (addr_scope <= scope) + return 1; + + return 0; +} + +/******************************************************************** + * 3rd Level Abstractions + ********************************************************************/ + +/* What is the scope of 'addr'? */ +sctp_scope_t sctp_scope(const union sctp_addr *addr) +{ + struct sctp_af *af; + + af = sctp_get_af_specific(addr->sa.sa_family); + if (!af) + return SCTP_SCOPE_UNUSABLE; + + return af->scope((union sctp_addr *)addr); +} diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c new file mode 100644 index 000000000000..0c2ab7885058 --- /dev/null +++ b/net/sctp/chunk.c @@ -0,0 +1,309 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2003, 2004 + * + * This file is part of the SCTP kernel reference Implementation + * + * This file contains the code relating the the chunk abstraction. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Jon Grimm <jgrimm@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/net.h> +#include <linux/inet.h> +#include <linux/skbuff.h> +#include <net/sock.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* This file is mostly in anticipation of future work, but initially + * populate with fragment tracking for an outbound message. + */ + +/* Initialize datamsg from memory. */ +static void sctp_datamsg_init(struct sctp_datamsg *msg) +{ + atomic_set(&msg->refcnt, 1); + msg->send_failed = 0; + msg->send_error = 0; + msg->can_abandon = 0; + msg->expires_at = 0; + INIT_LIST_HEAD(&msg->chunks); +} + +/* Allocate and initialize datamsg. */ +SCTP_STATIC struct sctp_datamsg *sctp_datamsg_new(int gfp) +{ + struct sctp_datamsg *msg; + msg = kmalloc(sizeof(struct sctp_datamsg), gfp); + if (msg) + sctp_datamsg_init(msg); + SCTP_DBG_OBJCNT_INC(datamsg); + return msg; +} + +/* Final destructruction of datamsg memory. */ +static void sctp_datamsg_destroy(struct sctp_datamsg *msg) +{ + struct list_head *pos, *temp; + struct sctp_chunk *chunk; + struct sctp_sock *sp; + struct sctp_ulpevent *ev; + struct sctp_association *asoc = NULL; + int error = 0, notify; + + /* If we failed, we may need to notify. */ + notify = msg->send_failed ? -1 : 0; + + /* Release all references. */ + list_for_each_safe(pos, temp, &msg->chunks) { + list_del_init(pos); + chunk = list_entry(pos, struct sctp_chunk, frag_list); + /* Check whether we _really_ need to notify. */ + if (notify < 0) { + asoc = chunk->asoc; + if (msg->send_error) + error = msg->send_error; + else + error = asoc->outqueue.error; + + sp = sctp_sk(asoc->base.sk); + notify = sctp_ulpevent_type_enabled(SCTP_SEND_FAILED, + &sp->subscribe); + } + + /* Generate a SEND FAILED event only if enabled. */ + if (notify > 0) { + int sent; + if (chunk->has_tsn) + sent = SCTP_DATA_SENT; + else + sent = SCTP_DATA_UNSENT; + + ev = sctp_ulpevent_make_send_failed(asoc, chunk, sent, + error, GFP_ATOMIC); + if (ev) + sctp_ulpq_tail_event(&asoc->ulpq, ev); + } + + sctp_chunk_put(chunk); + } + + SCTP_DBG_OBJCNT_DEC(datamsg); + kfree(msg); +} + +/* Hold a reference. */ +static void sctp_datamsg_hold(struct sctp_datamsg *msg) +{ + atomic_inc(&msg->refcnt); +} + +/* Release a reference. */ +void sctp_datamsg_put(struct sctp_datamsg *msg) +{ + if (atomic_dec_and_test(&msg->refcnt)) + sctp_datamsg_destroy(msg); +} + +/* Free a message. Really just give up a reference, the + * really free happens in sctp_datamsg_destroy(). + */ +void sctp_datamsg_free(struct sctp_datamsg *msg) +{ + sctp_datamsg_put(msg); +} + +/* Hold on to all the fragments until all chunks have been sent. */ +void sctp_datamsg_track(struct sctp_chunk *chunk) +{ + sctp_chunk_hold(chunk); +} + +/* Assign a chunk to this datamsg. */ +static void sctp_datamsg_assign(struct sctp_datamsg *msg, struct sctp_chunk *chunk) +{ + sctp_datamsg_hold(msg); + chunk->msg = msg; +} + + +/* A data chunk can have a maximum payload of (2^16 - 20). Break + * down any such message into smaller chunks. Opportunistically, fragment + * the chunks down to the current MTU constraints. We may get refragmented + * later if the PMTU changes, but it is _much better_ to fragment immediately + * with a reasonable guess than always doing our fragmentation on the + * soft-interrupt. + */ +struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *asoc, + struct sctp_sndrcvinfo *sinfo, + struct msghdr *msgh, int msg_len) +{ + int max, whole, i, offset, over, err; + int len, first_len; + struct sctp_chunk *chunk; + struct sctp_datamsg *msg; + struct list_head *pos, *temp; + __u8 frag; + + msg = sctp_datamsg_new(GFP_KERNEL); + if (!msg) + return NULL; + + /* Note: Calculate this outside of the loop, so that all fragments + * have the same expiration. + */ + if (sinfo->sinfo_timetolive) { + /* sinfo_timetolive is in milliseconds */ + msg->expires_at = jiffies + + msecs_to_jiffies(sinfo->sinfo_timetolive); + msg->can_abandon = 1; + SCTP_DEBUG_PRINTK("%s: msg:%p expires_at: %ld jiffies:%ld\n", + __FUNCTION__, msg, msg->expires_at, jiffies); + } + + max = asoc->frag_point; + + whole = 0; + first_len = max; + + /* Encourage Cookie-ECHO bundling. */ + if (asoc->state < SCTP_STATE_COOKIE_ECHOED) { + whole = msg_len / (max - SCTP_ARBITRARY_COOKIE_ECHO_LEN); + + /* Account for the DATA to be bundled with the COOKIE-ECHO. */ + if (whole) { + first_len = max - SCTP_ARBITRARY_COOKIE_ECHO_LEN; + msg_len -= first_len; + whole = 1; + } + } + + /* How many full sized? How many bytes leftover? */ + whole += msg_len / max; + over = msg_len % max; + offset = 0; + + if ((whole > 1) || (whole && over)) + SCTP_INC_STATS_USER(SCTP_MIB_FRAGUSRMSGS); + + /* Create chunks for all the full sized DATA chunks. */ + for (i=0, len=first_len; i < whole; i++) { + frag = SCTP_DATA_MIDDLE_FRAG; + + if (0 == i) + frag |= SCTP_DATA_FIRST_FRAG; + + if ((i == (whole - 1)) && !over) + frag |= SCTP_DATA_LAST_FRAG; + + chunk = sctp_make_datafrag_empty(asoc, sinfo, len, frag, 0); + + if (!chunk) + goto errout; + err = sctp_user_addto_chunk(chunk, offset, len, msgh->msg_iov); + if (err < 0) + goto errout; + + offset += len; + + /* Put the chunk->skb back into the form expected by send. */ + __skb_pull(chunk->skb, (__u8 *)chunk->chunk_hdr + - (__u8 *)chunk->skb->data); + + sctp_datamsg_assign(msg, chunk); + list_add_tail(&chunk->frag_list, &msg->chunks); + + /* The first chunk, the first chunk was likely short + * to allow bundling, so reset to full size. + */ + if (0 == i) + len = max; + } + + /* .. now the leftover bytes. */ + if (over) { + if (!whole) + frag = SCTP_DATA_NOT_FRAG; + else + frag = SCTP_DATA_LAST_FRAG; + + chunk = sctp_make_datafrag_empty(asoc, sinfo, over, frag, 0); + + if (!chunk) + goto errout; + + err = sctp_user_addto_chunk(chunk, offset, over,msgh->msg_iov); + + /* Put the chunk->skb back into the form expected by send. */ + __skb_pull(chunk->skb, (__u8 *)chunk->chunk_hdr + - (__u8 *)chunk->skb->data); + if (err < 0) + goto errout; + + sctp_datamsg_assign(msg, chunk); + list_add_tail(&chunk->frag_list, &msg->chunks); + } + + return msg; + +errout: + list_for_each_safe(pos, temp, &msg->chunks) { + list_del_init(pos); + chunk = list_entry(pos, struct sctp_chunk, frag_list); + sctp_chunk_free(chunk); + } + sctp_datamsg_free(msg); + return NULL; +} + +/* Check whether this message has expired. */ +int sctp_chunk_abandoned(struct sctp_chunk *chunk) +{ + struct sctp_datamsg *msg = chunk->msg; + + if (!msg->can_abandon) + return 0; + + if (time_after(jiffies, msg->expires_at)) + return 1; + + return 0; +} + +/* This chunk (and consequently entire message) has failed in its sending. */ +void sctp_chunk_fail(struct sctp_chunk *chunk, int error) +{ + chunk->msg->send_failed = 1; + chunk->msg->send_error = error; +} diff --git a/net/sctp/command.c b/net/sctp/command.c new file mode 100644 index 000000000000..3ff804757f4a --- /dev/null +++ b/net/sctp/command.c @@ -0,0 +1,81 @@ +/* SCTP kernel reference Implementation Copyright (C) 1999-2001 + * Cisco, Motorola, and IBM + * Copyright 2001 La Monte H.P. Yarroll + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions manipulate sctp command sequences. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Initialize a block of memory as a command sequence. */ +int sctp_init_cmd_seq(sctp_cmd_seq_t *seq) +{ + memset(seq, 0, sizeof(sctp_cmd_seq_t)); + return 1; /* We always succeed. */ +} + +/* Add a command to a sctp_cmd_seq_t. + * Return 0 if the command sequence is full. + */ +int sctp_add_cmd(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj) +{ + if (seq->next_free_slot >= SCTP_MAX_NUM_COMMANDS) + goto fail; + + seq->cmds[seq->next_free_slot].verb = verb; + seq->cmds[seq->next_free_slot++].obj = obj; + + return 1; + +fail: + return 0; +} + +/* Return the next command structure in a sctp_cmd_seq. + * Returns NULL at the end of the sequence. + */ +sctp_cmd_t *sctp_next_cmd(sctp_cmd_seq_t *seq) +{ + sctp_cmd_t *retval = NULL; + + if (seq->next_cmd < seq->next_free_slot) + retval = &seq->cmds[seq->next_cmd++]; + + return retval; +} + diff --git a/net/sctp/crc32c.c b/net/sctp/crc32c.c new file mode 100644 index 000000000000..31f05ec8e1d3 --- /dev/null +++ b/net/sctp/crc32c.c @@ -0,0 +1,220 @@ +/* SCTP kernel reference Implementation + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001-2003 International Business Machines, Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * SCTP Checksum functions + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Dinakaran Joseph + * Jon Grimm <jgrimm@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +/* The following code has been taken directly from + * draft-ietf-tsvwg-sctpcsum-03.txt + * + * The code has now been modified specifically for SCTP knowledge. + */ + +#include <linux/types.h> +#include <net/sctp/sctp.h> + +#define CRC32C_POLY 0x1EDC6F41 +#define CRC32C(c,d) (c=(c>>8)^crc_c[(c^(d))&0xFF]) +/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ +/* Copyright 2001, D. Otis. Use this program, code or tables */ +/* extracted from it, as desired without restriction. */ +/* */ +/* 32 Bit Reflected CRC table generation for SCTP. */ +/* To accommodate serial byte data being shifted out least */ +/* significant bit first, the table's 32 bit words are reflected */ +/* which flips both byte and bit MS and LS positions. The CRC */ +/* is calculated MS bits first from the perspective of the serial*/ +/* stream. The x^32 term is implied and the x^0 term may also */ +/* be shown as +1. The polynomial code used is 0x1EDC6F41. */ +/* Castagnoli93 */ +/* x^32+x^28+x^27+x^26+x^25+x^23+x^22+x^20+x^19+x^18+x^14+x^13+ */ +/* x^11+x^10+x^9+x^8+x^6+x^0 */ +/* Guy Castagnoli Stefan Braeuer and Martin Herrman */ +/* "Optimization of Cyclic Redundancy-Check Codes */ +/* with 24 and 32 Parity Bits", */ +/* IEEE Transactions on Communications, Vol.41, No.6, June 1993 */ +/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */ +static const __u32 crc_c[256] = { + 0x00000000, 0xF26B8303, 0xE13B70F7, 0x1350F3F4, + 0xC79A971F, 0x35F1141C, 0x26A1E7E8, 0xD4CA64EB, + 0x8AD958CF, 0x78B2DBCC, 0x6BE22838, 0x9989AB3B, + 0x4D43CFD0, 0xBF284CD3, 0xAC78BF27, 0x5E133C24, + 0x105EC76F, 0xE235446C, 0xF165B798, 0x030E349B, + 0xD7C45070, 0x25AFD373, 0x36FF2087, 0xC494A384, + 0x9A879FA0, 0x68EC1CA3, 0x7BBCEF57, 0x89D76C54, + 0x5D1D08BF, 0xAF768BBC, 0xBC267848, 0x4E4DFB4B, + 0x20BD8EDE, 0xD2D60DDD, 0xC186FE29, 0x33ED7D2A, + 0xE72719C1, 0x154C9AC2, 0x061C6936, 0xF477EA35, + 0xAA64D611, 0x580F5512, 0x4B5FA6E6, 0xB93425E5, + 0x6DFE410E, 0x9F95C20D, 0x8CC531F9, 0x7EAEB2FA, + 0x30E349B1, 0xC288CAB2, 0xD1D83946, 0x23B3BA45, + 0xF779DEAE, 0x05125DAD, 0x1642AE59, 0xE4292D5A, + 0xBA3A117E, 0x4851927D, 0x5B016189, 0xA96AE28A, + 0x7DA08661, 0x8FCB0562, 0x9C9BF696, 0x6EF07595, + 0x417B1DBC, 0xB3109EBF, 0xA0406D4B, 0x522BEE48, + 0x86E18AA3, 0x748A09A0, 0x67DAFA54, 0x95B17957, + 0xCBA24573, 0x39C9C670, 0x2A993584, 0xD8F2B687, + 0x0C38D26C, 0xFE53516F, 0xED03A29B, 0x1F682198, + 0x5125DAD3, 0xA34E59D0, 0xB01EAA24, 0x42752927, + 0x96BF4DCC, 0x64D4CECF, 0x77843D3B, 0x85EFBE38, + 0xDBFC821C, 0x2997011F, 0x3AC7F2EB, 0xC8AC71E8, + 0x1C661503, 0xEE0D9600, 0xFD5D65F4, 0x0F36E6F7, + 0x61C69362, 0x93AD1061, 0x80FDE395, 0x72966096, + 0xA65C047D, 0x5437877E, 0x4767748A, 0xB50CF789, + 0xEB1FCBAD, 0x197448AE, 0x0A24BB5A, 0xF84F3859, + 0x2C855CB2, 0xDEEEDFB1, 0xCDBE2C45, 0x3FD5AF46, + 0x7198540D, 0x83F3D70E, 0x90A324FA, 0x62C8A7F9, + 0xB602C312, 0x44694011, 0x5739B3E5, 0xA55230E6, + 0xFB410CC2, 0x092A8FC1, 0x1A7A7C35, 0xE811FF36, + 0x3CDB9BDD, 0xCEB018DE, 0xDDE0EB2A, 0x2F8B6829, + 0x82F63B78, 0x709DB87B, 0x63CD4B8F, 0x91A6C88C, + 0x456CAC67, 0xB7072F64, 0xA457DC90, 0x563C5F93, + 0x082F63B7, 0xFA44E0B4, 0xE9141340, 0x1B7F9043, + 0xCFB5F4A8, 0x3DDE77AB, 0x2E8E845F, 0xDCE5075C, + 0x92A8FC17, 0x60C37F14, 0x73938CE0, 0x81F80FE3, + 0x55326B08, 0xA759E80B, 0xB4091BFF, 0x466298FC, + 0x1871A4D8, 0xEA1A27DB, 0xF94AD42F, 0x0B21572C, + 0xDFEB33C7, 0x2D80B0C4, 0x3ED04330, 0xCCBBC033, + 0xA24BB5A6, 0x502036A5, 0x4370C551, 0xB11B4652, + 0x65D122B9, 0x97BAA1BA, 0x84EA524E, 0x7681D14D, + 0x2892ED69, 0xDAF96E6A, 0xC9A99D9E, 0x3BC21E9D, + 0xEF087A76, 0x1D63F975, 0x0E330A81, 0xFC588982, + 0xB21572C9, 0x407EF1CA, 0x532E023E, 0xA145813D, + 0x758FE5D6, 0x87E466D5, 0x94B49521, 0x66DF1622, + 0x38CC2A06, 0xCAA7A905, 0xD9F75AF1, 0x2B9CD9F2, + 0xFF56BD19, 0x0D3D3E1A, 0x1E6DCDEE, 0xEC064EED, + 0xC38D26C4, 0x31E6A5C7, 0x22B65633, 0xD0DDD530, + 0x0417B1DB, 0xF67C32D8, 0xE52CC12C, 0x1747422F, + 0x49547E0B, 0xBB3FFD08, 0xA86F0EFC, 0x5A048DFF, + 0x8ECEE914, 0x7CA56A17, 0x6FF599E3, 0x9D9E1AE0, + 0xD3D3E1AB, 0x21B862A8, 0x32E8915C, 0xC083125F, + 0x144976B4, 0xE622F5B7, 0xF5720643, 0x07198540, + 0x590AB964, 0xAB613A67, 0xB831C993, 0x4A5A4A90, + 0x9E902E7B, 0x6CFBAD78, 0x7FAB5E8C, 0x8DC0DD8F, + 0xE330A81A, 0x115B2B19, 0x020BD8ED, 0xF0605BEE, + 0x24AA3F05, 0xD6C1BC06, 0xC5914FF2, 0x37FACCF1, + 0x69E9F0D5, 0x9B8273D6, 0x88D28022, 0x7AB90321, + 0xAE7367CA, 0x5C18E4C9, 0x4F48173D, 0xBD23943E, + 0xF36E6F75, 0x0105EC76, 0x12551F82, 0xE03E9C81, + 0x34F4F86A, 0xC69F7B69, 0xD5CF889D, 0x27A40B9E, + 0x79B737BA, 0x8BDCB4B9, 0x988C474D, 0x6AE7C44E, + 0xBE2DA0A5, 0x4C4623A6, 0x5F16D052, 0xAD7D5351, +}; + +__u32 sctp_start_cksum(__u8 *buffer, __u16 length) +{ + __u32 crc32 = ~(__u32) 0; + __u32 i; + + /* Optimize this routine to be SCTP specific, knowing how + * to skip the checksum field of the SCTP header. + */ + + /* Calculate CRC up to the checksum. */ + for (i = 0; i < (sizeof(struct sctphdr) - sizeof(__u32)); i++) + CRC32C(crc32, buffer[i]); + + /* Skip checksum field of the header. */ + for (i = 0; i < sizeof(__u32); i++) + CRC32C(crc32, 0); + + /* Calculate the rest of the CRC. */ + for (i = sizeof(struct sctphdr); i < length ; i++) + CRC32C(crc32, buffer[i]); + + return crc32; +} + +__u32 sctp_update_cksum(__u8 *buffer, __u16 length, __u32 crc32) +{ + __u32 i; + + for (i = 0; i < length ; i++) + CRC32C(crc32, buffer[i]); + + return crc32; +} + +__u32 sctp_update_copy_cksum(__u8 *to, __u8 *from, __u16 length, __u32 crc32) +{ + __u32 i; + __u32 *_to = (__u32 *)to; + __u32 *_from = (__u32 *)from; + + for (i = 0; i < (length/4); i++) { + _to[i] = _from[i]; + CRC32C(crc32, from[i*4]); + CRC32C(crc32, from[i*4+1]); + CRC32C(crc32, from[i*4+2]); + CRC32C(crc32, from[i*4+3]); + } + + return crc32; +} + +__u32 sctp_end_cksum(__u32 crc32) +{ + __u32 result; + __u8 byte0, byte1, byte2, byte3; + + result = ~crc32; + + /* result now holds the negated polynomial remainder; + * since the table and algorithm is "reflected" [williams95]. + * That is, result has the same value as if we mapped the message + * to a polyomial, computed the host-bit-order polynomial + * remainder, performed final negation, then did an end-for-end + * bit-reversal. + * Note that a 32-bit bit-reversal is identical to four inplace + * 8-bit reversals followed by an end-for-end byteswap. + * In other words, the bytes of each bit are in the right order, + * but the bytes have been byteswapped. So we now do an explicit + * byteswap. On a little-endian machine, this byteswap and + * the final ntohl cancel out and could be elided. + */ + byte0 = result & 0xff; + byte1 = (result>>8) & 0xff; + byte2 = (result>>16) & 0xff; + byte3 = (result>>24) & 0xff; + + crc32 = ((byte0 << 24) | + (byte1 << 16) | + (byte2 << 8) | + byte3); + return crc32; +} diff --git a/net/sctp/debug.c b/net/sctp/debug.c new file mode 100644 index 000000000000..aa8340373af7 --- /dev/null +++ b/net/sctp/debug.c @@ -0,0 +1,191 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * This file is part of the implementation of the add-IP extension, + * based on <draft-ietf-tsvwg-addip-sctp-02.txt> June 29, 2001, + * for the SCTP kernel reference Implementation. + * + * This file converts numerical ID value to alphabetical names for SCTP + * terms such as chunk type, parameter time, event type, etc. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Xingang Guo <xingang.guo@intel.com> + * Jon Grimm <jgrimm@us.ibm.com> + * Daisy Chang <daisyc@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <net/sctp/sctp.h> + +#if SCTP_DEBUG +int sctp_debug_flag = 1; /* Initially enable DEBUG */ +#endif /* SCTP_DEBUG */ + +/* These are printable forms of Chunk ID's from section 3.1. */ +static const char *sctp_cid_tbl[SCTP_NUM_BASE_CHUNK_TYPES] = { + "DATA", + "INIT", + "INIT_ACK", + "SACK", + "HEARTBEAT", + "HEARTBEAT_ACK", + "ABORT", + "SHUTDOWN", + "SHUTDOWN_ACK", + "ERROR", + "COOKIE_ECHO", + "COOKIE_ACK", + "ECN_ECNE", + "ECN_CWR", + "SHUTDOWN_COMPLETE", +}; + +/* Lookup "chunk type" debug name. */ +const char *sctp_cname(const sctp_subtype_t cid) +{ + if (cid.chunk < 0) + return "illegal chunk id"; + if (cid.chunk <= SCTP_CID_BASE_MAX) + return sctp_cid_tbl[cid.chunk]; + + switch (cid.chunk) { + case SCTP_CID_ASCONF: + return "ASCONF"; + + case SCTP_CID_ASCONF_ACK: + return "ASCONF_ACK"; + + case SCTP_CID_FWD_TSN: + return "FWD_TSN"; + + default: + return "unknown chunk"; + }; + return "unknown chunk"; +} + +/* These are printable forms of the states. */ +const char *sctp_state_tbl[SCTP_STATE_NUM_STATES] = { + "STATE_EMPTY", + "STATE_CLOSED", + "STATE_COOKIE_WAIT", + "STATE_COOKIE_ECHOED", + "STATE_ESTABLISHED", + "STATE_SHUTDOWN_PENDING", + "STATE_SHUTDOWN_SENT", + "STATE_SHUTDOWN_RECEIVED", + "STATE_SHUTDOWN_ACK_SENT", +}; + +/* Events that could change the state of an association. */ +const char *sctp_evttype_tbl[] = { + "EVENT_T_unknown", + "EVENT_T_CHUNK", + "EVENT_T_TIMEOUT", + "EVENT_T_OTHER", + "EVENT_T_PRIMITIVE" +}; + +/* Return value of a state function */ +const char *sctp_status_tbl[] = { + "DISPOSITION_DISCARD", + "DISPOSITION_CONSUME", + "DISPOSITION_NOMEM", + "DISPOSITION_DELETE_TCB", + "DISPOSITION_ABORT", + "DISPOSITION_VIOLATION", + "DISPOSITION_NOT_IMPL", + "DISPOSITION_ERROR", + "DISPOSITION_BUG" +}; + +/* Printable forms of primitives */ +static const char *sctp_primitive_tbl[SCTP_NUM_PRIMITIVE_TYPES] = { + "PRIMITIVE_ASSOCIATE", + "PRIMITIVE_SHUTDOWN", + "PRIMITIVE_ABORT", + "PRIMITIVE_SEND", + "PRIMITIVE_REQUESTHEARTBEAT", +}; + +/* Lookup primitive debug name. */ +const char *sctp_pname(const sctp_subtype_t id) +{ + if (id.primitive < 0) + return "illegal primitive"; + if (id.primitive <= SCTP_EVENT_PRIMITIVE_MAX) + return sctp_primitive_tbl[id.primitive]; + return "unknown_primitive"; +} + +static const char *sctp_other_tbl[] = { + "NO_PENDING_TSN", + "ICMP_PROTO_UNREACH", +}; + +/* Lookup "other" debug name. */ +const char *sctp_oname(const sctp_subtype_t id) +{ + if (id.other < 0) + return "illegal 'other' event"; + if (id.other <= SCTP_EVENT_OTHER_MAX) + return sctp_other_tbl[id.other]; + return "unknown 'other' event"; +} + +static const char *sctp_timer_tbl[] = { + "TIMEOUT_NONE", + "TIMEOUT_T1_COOKIE", + "TIMEOUT_T1_INIT", + "TIMEOUT_T2_SHUTDOWN", + "TIMEOUT_T3_RTX", + "TIMEOUT_T4_RTO", + "TIMEOUT_T5_SHUTDOWN_GUARD", + "TIMEOUT_HEARTBEAT", + "TIMEOUT_SACK", + "TIMEOUT_AUTOCLOSE", +}; + +/* Lookup timer debug name. */ +const char *sctp_tname(const sctp_subtype_t id) +{ + if (id.timeout < 0) + return "illegal 'timer' event"; + if (id.timeout <= SCTP_EVENT_TIMEOUT_MAX) + return sctp_timer_tbl[id.timeout]; + return "unknown_timer"; +} diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c new file mode 100644 index 000000000000..544b75077dbd --- /dev/null +++ b/net/sctp/endpointola.c @@ -0,0 +1,389 @@ +/* SCTP kernel reference Implementation + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001-2002 International Business Machines, Corp. + * Copyright (c) 2001 Intel Corp. + * Copyright (c) 2001 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * This file is part of the SCTP kernel reference Implementation + * + * This abstraction represents an SCTP endpoint. + * + * This file is part of the implementation of the add-IP extension, + * based on <draft-ietf-tsvwg-addip-sctp-02.txt> June 29, 2001, + * for the SCTP kernel reference Implementation. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@austin.ibm.com> + * Daisy Chang <daisyc@us.ibm.com> + * Dajiang Zhang <dajiang.zhang@nokia.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/in.h> +#include <linux/random.h> /* get_random_bytes() */ +#include <linux/crypto.h> +#include <net/sock.h> +#include <net/ipv6.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Forward declarations for internal helpers. */ +static void sctp_endpoint_bh_rcv(struct sctp_endpoint *ep); + +/* + * Initialize the base fields of the endpoint structure. + */ +static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep, + struct sock *sk, int gfp) +{ + struct sctp_sock *sp = sctp_sk(sk); + memset(ep, 0, sizeof(struct sctp_endpoint)); + + /* Initialize the base structure. */ + /* What type of endpoint are we? */ + ep->base.type = SCTP_EP_TYPE_SOCKET; + + /* Initialize the basic object fields. */ + atomic_set(&ep->base.refcnt, 1); + ep->base.dead = 0; + ep->base.malloced = 1; + + /* Create an input queue. */ + sctp_inq_init(&ep->base.inqueue); + + /* Set its top-half handler */ + sctp_inq_set_th_handler(&ep->base.inqueue, + (void (*)(void *))sctp_endpoint_bh_rcv, ep); + + /* Initialize the bind addr area */ + sctp_bind_addr_init(&ep->base.bind_addr, 0); + rwlock_init(&ep->base.addr_lock); + + /* Remember who we are attached to. */ + ep->base.sk = sk; + sock_hold(ep->base.sk); + + /* Create the lists of associations. */ + INIT_LIST_HEAD(&ep->asocs); + + /* Set up the base timeout information. */ + ep->timeouts[SCTP_EVENT_TIMEOUT_NONE] = 0; + ep->timeouts[SCTP_EVENT_TIMEOUT_T1_COOKIE] = + SCTP_DEFAULT_TIMEOUT_T1_COOKIE; + ep->timeouts[SCTP_EVENT_TIMEOUT_T1_INIT] = + SCTP_DEFAULT_TIMEOUT_T1_INIT; + ep->timeouts[SCTP_EVENT_TIMEOUT_T2_SHUTDOWN] = + msecs_to_jiffies(sp->rtoinfo.srto_initial); + ep->timeouts[SCTP_EVENT_TIMEOUT_T3_RTX] = 0; + ep->timeouts[SCTP_EVENT_TIMEOUT_T4_RTO] = 0; + + /* sctpimpguide-05 Section 2.12.2 + * If the 'T5-shutdown-guard' timer is used, it SHOULD be set to the + * recommended value of 5 times 'RTO.Max'. + */ + ep->timeouts[SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD] + = 5 * msecs_to_jiffies(sp->rtoinfo.srto_max); + + ep->timeouts[SCTP_EVENT_TIMEOUT_HEARTBEAT] = + SCTP_DEFAULT_TIMEOUT_HEARTBEAT; + ep->timeouts[SCTP_EVENT_TIMEOUT_SACK] = + SCTP_DEFAULT_TIMEOUT_SACK; + ep->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = + sp->autoclose * HZ; + + /* Use SCTP specific send buffer space queues. */ + sk->sk_write_space = sctp_write_space; + sock_set_flag(sk, SOCK_USE_WRITE_QUEUE); + + /* Initialize the secret key used with cookie. */ + get_random_bytes(&ep->secret_key[0], SCTP_SECRET_SIZE); + ep->last_key = ep->current_key = 0; + ep->key_changed_at = jiffies; + + ep->debug_name = "unnamedEndpoint"; + return ep; +} + +/* Create a sctp_endpoint with all that boring stuff initialized. + * Returns NULL if there isn't enough memory. + */ +struct sctp_endpoint *sctp_endpoint_new(struct sock *sk, int gfp) +{ + struct sctp_endpoint *ep; + + /* Build a local endpoint. */ + ep = t_new(struct sctp_endpoint, gfp); + if (!ep) + goto fail; + if (!sctp_endpoint_init(ep, sk, gfp)) + goto fail_init; + ep->base.malloced = 1; + SCTP_DBG_OBJCNT_INC(ep); + return ep; + +fail_init: + kfree(ep); +fail: + return NULL; +} + +/* Add an association to an endpoint. */ +void sctp_endpoint_add_asoc(struct sctp_endpoint *ep, + struct sctp_association *asoc) +{ + struct sock *sk = ep->base.sk; + + /* Now just add it to our list of asocs */ + list_add_tail(&asoc->asocs, &ep->asocs); + + /* Increment the backlog value for a TCP-style listening socket. */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) + sk->sk_ack_backlog++; +} + +/* Free the endpoint structure. Delay cleanup until + * all users have released their reference count on this structure. + */ +void sctp_endpoint_free(struct sctp_endpoint *ep) +{ + ep->base.dead = 1; + sctp_endpoint_put(ep); +} + +/* Final destructor for endpoint. */ +static void sctp_endpoint_destroy(struct sctp_endpoint *ep) +{ + SCTP_ASSERT(ep->base.dead, "Endpoint is not dead", return); + + ep->base.sk->sk_state = SCTP_SS_CLOSED; + + /* Unlink this endpoint, so we can't find it again! */ + sctp_unhash_endpoint(ep); + + /* Free up the HMAC transform. */ + if (sctp_sk(ep->base.sk)->hmac) + sctp_crypto_free_tfm(sctp_sk(ep->base.sk)->hmac); + + /* Cleanup. */ + sctp_inq_free(&ep->base.inqueue); + sctp_bind_addr_free(&ep->base.bind_addr); + + /* Remove and free the port */ + if (sctp_sk(ep->base.sk)->bind_hash) + sctp_put_port(ep->base.sk); + + /* Give up our hold on the sock. */ + if (ep->base.sk) + sock_put(ep->base.sk); + + /* Finally, free up our memory. */ + if (ep->base.malloced) { + kfree(ep); + SCTP_DBG_OBJCNT_DEC(ep); + } +} + +/* Hold a reference to an endpoint. */ +void sctp_endpoint_hold(struct sctp_endpoint *ep) +{ + atomic_inc(&ep->base.refcnt); +} + +/* Release a reference to an endpoint and clean up if there are + * no more references. + */ +void sctp_endpoint_put(struct sctp_endpoint *ep) +{ + if (atomic_dec_and_test(&ep->base.refcnt)) + sctp_endpoint_destroy(ep); +} + +/* Is this the endpoint we are looking for? */ +struct sctp_endpoint *sctp_endpoint_is_match(struct sctp_endpoint *ep, + const union sctp_addr *laddr) +{ + struct sctp_endpoint *retval; + + sctp_read_lock(&ep->base.addr_lock); + if (ep->base.bind_addr.port == laddr->v4.sin_port) { + if (sctp_bind_addr_match(&ep->base.bind_addr, laddr, + sctp_sk(ep->base.sk))) { + retval = ep; + goto out; + } + } + + retval = NULL; + +out: + sctp_read_unlock(&ep->base.addr_lock); + return retval; +} + +/* Find the association that goes with this chunk. + * We do a linear search of the associations for this endpoint. + * We return the matching transport address too. + */ +static struct sctp_association *__sctp_endpoint_lookup_assoc( + const struct sctp_endpoint *ep, + const union sctp_addr *paddr, + struct sctp_transport **transport) +{ + int rport; + struct sctp_association *asoc; + struct list_head *pos; + + rport = paddr->v4.sin_port; + + list_for_each(pos, &ep->asocs) { + asoc = list_entry(pos, struct sctp_association, asocs); + if (rport == asoc->peer.port) { + sctp_read_lock(&asoc->base.addr_lock); + *transport = sctp_assoc_lookup_paddr(asoc, paddr); + sctp_read_unlock(&asoc->base.addr_lock); + + if (*transport) + return asoc; + } + } + + *transport = NULL; + return NULL; +} + +/* Lookup association on an endpoint based on a peer address. BH-safe. */ +struct sctp_association *sctp_endpoint_lookup_assoc( + const struct sctp_endpoint *ep, + const union sctp_addr *paddr, + struct sctp_transport **transport) +{ + struct sctp_association *asoc; + + sctp_local_bh_disable(); + asoc = __sctp_endpoint_lookup_assoc(ep, paddr, transport); + sctp_local_bh_enable(); + + return asoc; +} + +/* Look for any peeled off association from the endpoint that matches the + * given peer address. + */ +int sctp_endpoint_is_peeled_off(struct sctp_endpoint *ep, + const union sctp_addr *paddr) +{ + struct list_head *pos; + struct sctp_sockaddr_entry *addr; + struct sctp_bind_addr *bp; + + sctp_read_lock(&ep->base.addr_lock); + bp = &ep->base.bind_addr; + list_for_each(pos, &bp->address_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + if (sctp_has_association(&addr->a, paddr)) { + sctp_read_unlock(&ep->base.addr_lock); + return 1; + } + } + sctp_read_unlock(&ep->base.addr_lock); + + return 0; +} + +/* Do delayed input processing. This is scheduled by sctp_rcv(). + * This may be called on BH or task time. + */ +static void sctp_endpoint_bh_rcv(struct sctp_endpoint *ep) +{ + struct sctp_association *asoc; + struct sock *sk; + struct sctp_transport *transport; + struct sctp_chunk *chunk; + struct sctp_inq *inqueue; + sctp_subtype_t subtype; + sctp_state_t state; + int error = 0; + + if (ep->base.dead) + return; + + asoc = NULL; + inqueue = &ep->base.inqueue; + sk = ep->base.sk; + + while (NULL != (chunk = sctp_inq_pop(inqueue))) { + subtype = SCTP_ST_CHUNK(chunk->chunk_hdr->type); + + /* We might have grown an association since last we + * looked, so try again. + * + * This happens when we've just processed our + * COOKIE-ECHO chunk. + */ + if (NULL == chunk->asoc) { + asoc = sctp_endpoint_lookup_assoc(ep, + sctp_source(chunk), + &transport); + chunk->asoc = asoc; + chunk->transport = transport; + } + + state = asoc ? asoc->state : SCTP_STATE_CLOSED; + + /* Remember where the last DATA chunk came from so we + * know where to send the SACK. + */ + if (asoc && sctp_chunk_is_data(chunk)) + asoc->peer.last_data_from = chunk->transport; + else + SCTP_INC_STATS(SCTP_MIB_INCTRLCHUNKS); + + if (chunk->transport) + chunk->transport->last_time_heard = jiffies; + + error = sctp_do_sm(SCTP_EVENT_T_CHUNK, subtype, state, + ep, asoc, chunk, GFP_ATOMIC); + + if (error && chunk) + chunk->pdiscard = 1; + + /* Check to see if the endpoint is freed in response to + * the incoming chunk. If so, get out of the while loop. + */ + if (!sctp_sk(sk)->ep) + break; + } +} diff --git a/net/sctp/input.c b/net/sctp/input.c new file mode 100644 index 000000000000..b719a77d66b4 --- /dev/null +++ b/net/sctp/input.c @@ -0,0 +1,913 @@ +/* SCTP kernel reference 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 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions handle all input from the IP layer into SCTP. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Xingang Guo <xingang.guo@intel.com> + * Jon Grimm <jgrimm@us.ibm.com> + * Hui Huang <hui.huang@nokia.com> + * Daisy Chang <daisyc@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/list.h> /* For struct list_head */ +#include <linux/socket.h> +#include <linux/ip.h> +#include <linux/time.h> /* For struct timeval */ +#include <net/ip.h> +#include <net/icmp.h> +#include <net/snmp.h> +#include <net/sock.h> +#include <net/xfrm.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Forward declarations for internal helpers. */ +static int sctp_rcv_ootb(struct sk_buff *); +static struct sctp_association *__sctp_rcv_lookup(struct sk_buff *skb, + const union sctp_addr *laddr, + const union sctp_addr *paddr, + struct sctp_transport **transportp); +static struct sctp_endpoint *__sctp_rcv_lookup_endpoint(const union sctp_addr *laddr); +static struct sctp_association *__sctp_lookup_association( + const union sctp_addr *local, + const union sctp_addr *peer, + struct sctp_transport **pt); + + +/* Calculate the SCTP checksum of an SCTP packet. */ +static inline int sctp_rcv_checksum(struct sk_buff *skb) +{ + struct sctphdr *sh; + __u32 cmp, val; + struct sk_buff *list = skb_shinfo(skb)->frag_list; + + sh = (struct sctphdr *) skb->h.raw; + cmp = ntohl(sh->checksum); + + val = sctp_start_cksum((__u8 *)sh, skb_headlen(skb)); + + for (; list; list = list->next) + val = sctp_update_cksum((__u8 *)list->data, skb_headlen(list), + val); + + val = sctp_end_cksum(val); + + if (val != cmp) { + /* CRC failure, dump it. */ + SCTP_INC_STATS_BH(SCTP_MIB_CHECKSUMERRORS); + return -1; + } + return 0; +} + +/* The free routine for skbuffs that sctp receives */ +static void sctp_rfree(struct sk_buff *skb) +{ + atomic_sub(sizeof(struct sctp_chunk),&skb->sk->sk_rmem_alloc); + sock_rfree(skb); +} + +/* The ownership wrapper routine to do receive buffer accounting */ +static void sctp_rcv_set_owner_r(struct sk_buff *skb, struct sock *sk) +{ + skb_set_owner_r(skb,sk); + skb->destructor = sctp_rfree; + atomic_add(sizeof(struct sctp_chunk),&sk->sk_rmem_alloc); +} + +/* + * This is the routine which IP calls when receiving an SCTP packet. + */ +int sctp_rcv(struct sk_buff *skb) +{ + struct sock *sk; + struct sctp_association *asoc; + struct sctp_endpoint *ep = NULL; + struct sctp_ep_common *rcvr; + struct sctp_transport *transport = NULL; + struct sctp_chunk *chunk; + struct sctphdr *sh; + union sctp_addr src; + union sctp_addr dest; + int family; + struct sctp_af *af; + int ret = 0; + + if (skb->pkt_type!=PACKET_HOST) + goto discard_it; + + SCTP_INC_STATS_BH(SCTP_MIB_INSCTPPACKS); + + sh = (struct sctphdr *) skb->h.raw; + + /* Pull up the IP and SCTP headers. */ + __skb_pull(skb, skb->h.raw - skb->data); + if (skb->len < sizeof(struct sctphdr)) + goto discard_it; + if (sctp_rcv_checksum(skb) < 0) + goto discard_it; + + skb_pull(skb, sizeof(struct sctphdr)); + + /* Make sure we at least have chunk headers worth of data left. */ + if (skb->len < sizeof(struct sctp_chunkhdr)) + goto discard_it; + + family = ipver2af(skb->nh.iph->version); + af = sctp_get_af_specific(family); + if (unlikely(!af)) + goto discard_it; + + /* Initialize local addresses for lookups. */ + af->from_skb(&src, skb, 1); + af->from_skb(&dest, skb, 0); + + /* If the packet is to or from a non-unicast address, + * silently discard the packet. + * + * This is not clearly defined in the RFC except in section + * 8.4 - OOTB handling. However, based on the book "Stream Control + * Transmission Protocol" 2.1, "It is important to note that the + * IP address of an SCTP transport address must be a routable + * unicast address. In other words, IP multicast addresses and + * IP broadcast addresses cannot be used in an SCTP transport + * address." + */ + if (!af->addr_valid(&src, NULL) || !af->addr_valid(&dest, NULL)) + goto discard_it; + + asoc = __sctp_rcv_lookup(skb, &src, &dest, &transport); + + /* + * RFC 2960, 8.4 - Handle "Out of the blue" Packets. + * An SCTP packet is called an "out of the blue" (OOTB) + * packet if it is correctly formed, i.e., passed the + * receiver's checksum check, but the receiver is not + * able to identify the association to which this + * packet belongs. + */ + if (!asoc) { + ep = __sctp_rcv_lookup_endpoint(&dest); + if (sctp_rcv_ootb(skb)) { + SCTP_INC_STATS_BH(SCTP_MIB_OUTOFBLUES); + goto discard_release; + } + } + + /* Retrieve the common input handling substructure. */ + rcvr = asoc ? &asoc->base : &ep->base; + sk = rcvr->sk; + + if ((sk) && (atomic_read(&sk->sk_rmem_alloc) >= sk->sk_rcvbuf)) { + goto discard_release; + } + + + /* SCTP seems to always need a timestamp right now (FIXME) */ + if (skb->stamp.tv_sec == 0) { + do_gettimeofday(&skb->stamp); + sock_enable_timestamp(sk); + } + + if (!xfrm_policy_check(sk, XFRM_POLICY_IN, skb, family)) + goto discard_release; + + ret = sk_filter(sk, skb, 1); + if (ret) + goto discard_release; + + /* Create an SCTP packet structure. */ + chunk = sctp_chunkify(skb, asoc, sk); + if (!chunk) { + ret = -ENOMEM; + goto discard_release; + } + + sctp_rcv_set_owner_r(skb,sk); + + /* Remember what endpoint is to handle this packet. */ + chunk->rcvr = rcvr; + + /* Remember the SCTP header. */ + chunk->sctp_hdr = sh; + + /* Set the source and destination addresses of the incoming chunk. */ + sctp_init_addrs(chunk, &src, &dest); + + /* Remember where we came from. */ + chunk->transport = transport; + + /* Acquire access to the sock lock. Note: We are safe from other + * bottom halves on this lock, but a user may be in the lock too, + * so check if it is busy. + */ + sctp_bh_lock_sock(sk); + + if (sock_owned_by_user(sk)) + sk_add_backlog(sk, (struct sk_buff *) chunk); + else + sctp_backlog_rcv(sk, (struct sk_buff *) chunk); + + /* Release the sock and any reference counts we took in the + * lookup calls. + */ + sctp_bh_unlock_sock(sk); + if (asoc) + sctp_association_put(asoc); + else + sctp_endpoint_put(ep); + sock_put(sk); + return ret; + +discard_it: + kfree_skb(skb); + return ret; + +discard_release: + /* Release any structures we may be holding. */ + if (asoc) { + sock_put(asoc->base.sk); + sctp_association_put(asoc); + } else { + sock_put(ep->base.sk); + sctp_endpoint_put(ep); + } + + goto discard_it; +} + +/* Handle second half of inbound skb processing. If the sock was busy, + * we may have need to delay processing until later when the sock is + * released (on the backlog). If not busy, we call this routine + * directly from the bottom half. + */ +int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb) +{ + struct sctp_chunk *chunk; + struct sctp_inq *inqueue; + + /* One day chunk will live inside the skb, but for + * now this works. + */ + chunk = (struct sctp_chunk *) skb; + inqueue = &chunk->rcvr->inqueue; + + sctp_inq_push(inqueue, chunk); + return 0; +} + +/* Handle icmp frag needed error. */ +void sctp_icmp_frag_needed(struct sock *sk, struct sctp_association *asoc, + struct sctp_transport *t, __u32 pmtu) +{ + if (unlikely(pmtu < SCTP_DEFAULT_MINSEGMENT)) { + printk(KERN_WARNING "%s: Reported pmtu %d too low, " + "using default minimum of %d\n", __FUNCTION__, pmtu, + SCTP_DEFAULT_MINSEGMENT); + pmtu = SCTP_DEFAULT_MINSEGMENT; + } + + if (!sock_owned_by_user(sk) && t && (t->pmtu != pmtu)) { + t->pmtu = pmtu; + sctp_assoc_sync_pmtu(asoc); + sctp_retransmit(&asoc->outqueue, t, SCTP_RTXR_PMTUD); + } +} + +/* + * SCTP Implementer's Guide, 2.37 ICMP handling procedures + * + * ICMP8) If the ICMP code is a "Unrecognized next header type encountered" + * or a "Protocol Unreachable" treat this message as an abort + * with the T bit set. + * + * This function sends an event to the state machine, which will abort the + * association. + * + */ +void sctp_icmp_proto_unreachable(struct sock *sk, + struct sctp_endpoint *ep, + struct sctp_association *asoc, + struct sctp_transport *t) +{ + SCTP_DEBUG_PRINTK("%s\n", __FUNCTION__); + + sctp_do_sm(SCTP_EVENT_T_OTHER, + SCTP_ST_OTHER(SCTP_EVENT_ICMP_PROTO_UNREACH), + asoc->state, asoc->ep, asoc, NULL, + GFP_ATOMIC); + +} + +/* Common lookup code for icmp/icmpv6 error handler. */ +struct sock *sctp_err_lookup(int family, struct sk_buff *skb, + struct sctphdr *sctphdr, + struct sctp_endpoint **epp, + struct sctp_association **app, + struct sctp_transport **tpp) +{ + union sctp_addr saddr; + union sctp_addr daddr; + struct sctp_af *af; + struct sock *sk = NULL; + struct sctp_endpoint *ep = NULL; + struct sctp_association *asoc = NULL; + struct sctp_transport *transport = NULL; + + *app = NULL; *epp = NULL; *tpp = NULL; + + af = sctp_get_af_specific(family); + if (unlikely(!af)) { + return NULL; + } + + /* Initialize local addresses for lookups. */ + af->from_skb(&saddr, skb, 1); + af->from_skb(&daddr, skb, 0); + + /* Look for an association that matches the incoming ICMP error + * packet. + */ + asoc = __sctp_lookup_association(&saddr, &daddr, &transport); + if (!asoc) { + /* If there is no matching association, see if it matches any + * endpoint. This may happen for an ICMP error generated in + * response to an INIT_ACK. + */ + ep = __sctp_rcv_lookup_endpoint(&daddr); + if (!ep) { + return NULL; + } + } + + if (asoc) { + sk = asoc->base.sk; + + if (ntohl(sctphdr->vtag) != asoc->c.peer_vtag) { + ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); + goto out; + } + } else + sk = ep->base.sk; + + sctp_bh_lock_sock(sk); + + /* If too many ICMPs get dropped on busy + * servers this needs to be solved differently. + */ + if (sock_owned_by_user(sk)) + NET_INC_STATS_BH(LINUX_MIB_LOCKDROPPEDICMPS); + + *epp = ep; + *app = asoc; + *tpp = transport; + return sk; + +out: + sock_put(sk); + if (asoc) + sctp_association_put(asoc); + if (ep) + sctp_endpoint_put(ep); + return NULL; +} + +/* Common cleanup code for icmp/icmpv6 error handler. */ +void sctp_err_finish(struct sock *sk, struct sctp_endpoint *ep, + struct sctp_association *asoc) +{ + sctp_bh_unlock_sock(sk); + sock_put(sk); + if (asoc) + sctp_association_put(asoc); + if (ep) + sctp_endpoint_put(ep); +} + +/* + * This routine is called by the ICMP module when it gets some + * sort of error condition. If err < 0 then the socket should + * be closed and the error returned to the user. If err > 0 + * it's just the icmp type << 8 | icmp code. After adjustment + * header points to the first 8 bytes of the sctp header. We need + * to find the appropriate port. + * + * The locking strategy used here is very "optimistic". When + * someone else accesses the socket the ICMP is just dropped + * and for some paths there is no check at all. + * A more general error queue to queue errors for later handling + * is probably better. + * + */ +void sctp_v4_err(struct sk_buff *skb, __u32 info) +{ + struct iphdr *iph = (struct iphdr *)skb->data; + struct sctphdr *sh = (struct sctphdr *)(skb->data + (iph->ihl <<2)); + int type = skb->h.icmph->type; + int code = skb->h.icmph->code; + struct sock *sk; + struct sctp_endpoint *ep; + struct sctp_association *asoc; + struct sctp_transport *transport; + struct inet_sock *inet; + char *saveip, *savesctp; + int err; + + if (skb->len < ((iph->ihl << 2) + 8)) { + ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); + return; + } + + /* Fix up skb to look at the embedded net header. */ + saveip = skb->nh.raw; + savesctp = skb->h.raw; + skb->nh.iph = iph; + skb->h.raw = (char *)sh; + sk = sctp_err_lookup(AF_INET, skb, sh, &ep, &asoc, &transport); + /* Put back, the original pointers. */ + skb->nh.raw = saveip; + skb->h.raw = savesctp; + if (!sk) { + ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); + return; + } + /* Warning: The sock lock is held. Remember to call + * sctp_err_finish! + */ + + switch (type) { + case ICMP_PARAMETERPROB: + err = EPROTO; + break; + case ICMP_DEST_UNREACH: + if (code > NR_ICMP_UNREACH) + goto out_unlock; + + /* PMTU discovery (RFC1191) */ + if (ICMP_FRAG_NEEDED == code) { + sctp_icmp_frag_needed(sk, asoc, transport, info); + goto out_unlock; + } + else { + if (ICMP_PROT_UNREACH == code) { + sctp_icmp_proto_unreachable(sk, ep, asoc, + transport); + goto out_unlock; + } + } + err = icmp_err_convert[code].errno; + break; + case ICMP_TIME_EXCEEDED: + /* Ignore any time exceeded errors due to fragment reassembly + * timeouts. + */ + if (ICMP_EXC_FRAGTIME == code) + goto out_unlock; + + err = EHOSTUNREACH; + break; + default: + goto out_unlock; + } + + inet = inet_sk(sk); + if (!sock_owned_by_user(sk) && inet->recverr) { + sk->sk_err = err; + sk->sk_error_report(sk); + } else { /* Only an error on timeout */ + sk->sk_err_soft = err; + } + +out_unlock: + sctp_err_finish(sk, ep, asoc); +} + +/* + * RFC 2960, 8.4 - Handle "Out of the blue" Packets. + * + * This function scans all the chunks in the OOTB packet to determine if + * the packet should be discarded right away. If a response might be needed + * for this packet, or, if further processing is possible, the packet will + * be queued to a proper inqueue for the next phase of handling. + * + * Output: + * Return 0 - If further processing is needed. + * Return 1 - If the packet can be discarded right away. + */ +int sctp_rcv_ootb(struct sk_buff *skb) +{ + sctp_chunkhdr_t *ch; + __u8 *ch_end; + sctp_errhdr_t *err; + + ch = (sctp_chunkhdr_t *) skb->data; + ch_end = ((__u8 *) ch) + WORD_ROUND(ntohs(ch->length)); + + /* Scan through all the chunks in the packet. */ + while (ch_end > (__u8 *)ch && ch_end < skb->tail) { + + /* RFC 8.4, 2) If the OOTB packet contains an ABORT chunk, the + * receiver MUST silently discard the OOTB packet and take no + * further action. + */ + if (SCTP_CID_ABORT == ch->type) + goto discard; + + /* RFC 8.4, 6) If the packet contains a SHUTDOWN COMPLETE + * chunk, the receiver should silently discard the packet + * and take no further action. + */ + if (SCTP_CID_SHUTDOWN_COMPLETE == ch->type) + goto discard; + + /* RFC 8.4, 7) If the packet contains a "Stale cookie" ERROR + * or a COOKIE ACK the SCTP Packet should be silently + * discarded. + */ + if (SCTP_CID_COOKIE_ACK == ch->type) + goto discard; + + if (SCTP_CID_ERROR == ch->type) { + sctp_walk_errors(err, ch) { + if (SCTP_ERROR_STALE_COOKIE == err->cause) + goto discard; + } + } + + ch = (sctp_chunkhdr_t *) ch_end; + ch_end = ((__u8 *) ch) + WORD_ROUND(ntohs(ch->length)); + } + + return 0; + +discard: + return 1; +} + +/* Insert endpoint into the hash table. */ +static void __sctp_hash_endpoint(struct sctp_endpoint *ep) +{ + struct sctp_ep_common **epp; + struct sctp_ep_common *epb; + struct sctp_hashbucket *head; + + epb = &ep->base; + + epb->hashent = sctp_ep_hashfn(epb->bind_addr.port); + head = &sctp_ep_hashtable[epb->hashent]; + + sctp_write_lock(&head->lock); + epp = &head->chain; + epb->next = *epp; + if (epb->next) + (*epp)->pprev = &epb->next; + *epp = epb; + epb->pprev = epp; + sctp_write_unlock(&head->lock); +} + +/* Add an endpoint to the hash. Local BH-safe. */ +void sctp_hash_endpoint(struct sctp_endpoint *ep) +{ + sctp_local_bh_disable(); + __sctp_hash_endpoint(ep); + sctp_local_bh_enable(); +} + +/* Remove endpoint from the hash table. */ +static void __sctp_unhash_endpoint(struct sctp_endpoint *ep) +{ + struct sctp_hashbucket *head; + struct sctp_ep_common *epb; + + epb = &ep->base; + + epb->hashent = sctp_ep_hashfn(epb->bind_addr.port); + + head = &sctp_ep_hashtable[epb->hashent]; + + sctp_write_lock(&head->lock); + + if (epb->pprev) { + if (epb->next) + epb->next->pprev = epb->pprev; + *epb->pprev = epb->next; + epb->pprev = NULL; + } + + sctp_write_unlock(&head->lock); +} + +/* Remove endpoint from the hash. Local BH-safe. */ +void sctp_unhash_endpoint(struct sctp_endpoint *ep) +{ + sctp_local_bh_disable(); + __sctp_unhash_endpoint(ep); + sctp_local_bh_enable(); +} + +/* Look up an endpoint. */ +static struct sctp_endpoint *__sctp_rcv_lookup_endpoint(const union sctp_addr *laddr) +{ + struct sctp_hashbucket *head; + struct sctp_ep_common *epb; + struct sctp_endpoint *ep; + int hash; + + hash = sctp_ep_hashfn(laddr->v4.sin_port); + head = &sctp_ep_hashtable[hash]; + read_lock(&head->lock); + for (epb = head->chain; epb; epb = epb->next) { + ep = sctp_ep(epb); + if (sctp_endpoint_is_match(ep, laddr)) + goto hit; + } + + ep = sctp_sk((sctp_get_ctl_sock()))->ep; + epb = &ep->base; + +hit: + sctp_endpoint_hold(ep); + sock_hold(epb->sk); + read_unlock(&head->lock); + return ep; +} + +/* Insert association into the hash table. */ +static void __sctp_hash_established(struct sctp_association *asoc) +{ + struct sctp_ep_common **epp; + struct sctp_ep_common *epb; + struct sctp_hashbucket *head; + + epb = &asoc->base; + + /* Calculate which chain this entry will belong to. */ + epb->hashent = sctp_assoc_hashfn(epb->bind_addr.port, asoc->peer.port); + + head = &sctp_assoc_hashtable[epb->hashent]; + + sctp_write_lock(&head->lock); + epp = &head->chain; + epb->next = *epp; + if (epb->next) + (*epp)->pprev = &epb->next; + *epp = epb; + epb->pprev = epp; + sctp_write_unlock(&head->lock); +} + +/* Add an association to the hash. Local BH-safe. */ +void sctp_hash_established(struct sctp_association *asoc) +{ + sctp_local_bh_disable(); + __sctp_hash_established(asoc); + sctp_local_bh_enable(); +} + +/* Remove association from the hash table. */ +static void __sctp_unhash_established(struct sctp_association *asoc) +{ + struct sctp_hashbucket *head; + struct sctp_ep_common *epb; + + epb = &asoc->base; + + epb->hashent = sctp_assoc_hashfn(epb->bind_addr.port, + asoc->peer.port); + + head = &sctp_assoc_hashtable[epb->hashent]; + + sctp_write_lock(&head->lock); + + if (epb->pprev) { + if (epb->next) + epb->next->pprev = epb->pprev; + *epb->pprev = epb->next; + epb->pprev = NULL; + } + + sctp_write_unlock(&head->lock); +} + +/* Remove association from the hash table. Local BH-safe. */ +void sctp_unhash_established(struct sctp_association *asoc) +{ + sctp_local_bh_disable(); + __sctp_unhash_established(asoc); + sctp_local_bh_enable(); +} + +/* Look up an association. */ +static struct sctp_association *__sctp_lookup_association( + const union sctp_addr *local, + const union sctp_addr *peer, + struct sctp_transport **pt) +{ + struct sctp_hashbucket *head; + struct sctp_ep_common *epb; + struct sctp_association *asoc; + struct sctp_transport *transport; + int hash; + + /* Optimize here for direct hit, only listening connections can + * have wildcards anyways. + */ + hash = sctp_assoc_hashfn(local->v4.sin_port, peer->v4.sin_port); + head = &sctp_assoc_hashtable[hash]; + read_lock(&head->lock); + for (epb = head->chain; epb; epb = epb->next) { + asoc = sctp_assoc(epb); + transport = sctp_assoc_is_match(asoc, local, peer); + if (transport) + goto hit; + } + + read_unlock(&head->lock); + + return NULL; + +hit: + *pt = transport; + sctp_association_hold(asoc); + sock_hold(epb->sk); + read_unlock(&head->lock); + return asoc; +} + +/* Look up an association. BH-safe. */ +SCTP_STATIC +struct sctp_association *sctp_lookup_association(const union sctp_addr *laddr, + const union sctp_addr *paddr, + struct sctp_transport **transportp) +{ + struct sctp_association *asoc; + + sctp_local_bh_disable(); + asoc = __sctp_lookup_association(laddr, paddr, transportp); + sctp_local_bh_enable(); + + return asoc; +} + +/* Is there an association matching the given local and peer addresses? */ +int sctp_has_association(const union sctp_addr *laddr, + const union sctp_addr *paddr) +{ + struct sctp_association *asoc; + struct sctp_transport *transport; + + if ((asoc = sctp_lookup_association(laddr, paddr, &transport))) { + sock_put(asoc->base.sk); + sctp_association_put(asoc); + return 1; + } + + return 0; +} + +/* + * SCTP Implementors Guide, 2.18 Handling of address + * parameters within the INIT or INIT-ACK. + * + * D) When searching for a matching TCB upon reception of an INIT + * or INIT-ACK chunk the receiver SHOULD use not only the + * source address of the packet (containing the INIT or + * INIT-ACK) but the receiver SHOULD also use all valid + * address parameters contained within the chunk. + * + * 2.18.3 Solution description + * + * This new text clearly specifies to an implementor the need + * to look within the INIT or INIT-ACK. Any implementation that + * does not do this, may not be able to establish associations + * in certain circumstances. + * + */ +static struct sctp_association *__sctp_rcv_init_lookup(struct sk_buff *skb, + const union sctp_addr *laddr, struct sctp_transport **transportp) +{ + struct sctp_association *asoc; + union sctp_addr addr; + union sctp_addr *paddr = &addr; + struct sctphdr *sh = (struct sctphdr *) skb->h.raw; + sctp_chunkhdr_t *ch; + union sctp_params params; + sctp_init_chunk_t *init; + struct sctp_transport *transport; + struct sctp_af *af; + + ch = (sctp_chunkhdr_t *) skb->data; + + /* If this is INIT/INIT-ACK look inside the chunk too. */ + switch (ch->type) { + case SCTP_CID_INIT: + case SCTP_CID_INIT_ACK: + break; + default: + return NULL; + } + + /* The code below will attempt to walk the chunk and extract + * parameter information. Before we do that, we need to verify + * that the chunk length doesn't cause overflow. Otherwise, we'll + * walk off the end. + */ + if (WORD_ROUND(ntohs(ch->length)) > skb->len) + return NULL; + + /* + * This code will NOT touch anything inside the chunk--it is + * strictly READ-ONLY. + * + * RFC 2960 3 SCTP packet Format + * + * Multiple chunks can be bundled into one SCTP packet up to + * the MTU size, except for the INIT, INIT ACK, and SHUTDOWN + * COMPLETE chunks. These chunks MUST NOT be bundled with any + * other chunk in a packet. See Section 6.10 for more details + * on chunk bundling. + */ + + /* Find the start of the TLVs and the end of the chunk. This is + * the region we search for address parameters. + */ + init = (sctp_init_chunk_t *)skb->data; + + /* Walk the parameters looking for embedded addresses. */ + sctp_walk_params(params, init, init_hdr.params) { + + /* Note: Ignoring hostname addresses. */ + af = sctp_get_af_specific(param_type2af(params.p->type)); + if (!af) + continue; + + af->from_addr_param(paddr, params.addr, ntohs(sh->source), 0); + + asoc = __sctp_lookup_association(laddr, paddr, &transport); + if (asoc) + return asoc; + } + + return NULL; +} + +/* Lookup an association for an inbound skb. */ +static struct sctp_association *__sctp_rcv_lookup(struct sk_buff *skb, + const union sctp_addr *paddr, + const union sctp_addr *laddr, + struct sctp_transport **transportp) +{ + struct sctp_association *asoc; + + asoc = __sctp_lookup_association(laddr, paddr, transportp); + + /* Further lookup for INIT/INIT-ACK packets. + * SCTP Implementors Guide, 2.18 Handling of address + * parameters within the INIT or INIT-ACK. + */ + if (!asoc) + asoc = __sctp_rcv_init_lookup(skb, laddr, transportp); + + return asoc; +} diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c new file mode 100644 index 000000000000..cedf4351556c --- /dev/null +++ b/net/sctp/inqueue.c @@ -0,0 +1,204 @@ +/* SCTP kernel reference Implementation + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2002 International Business Machines, Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions are the methods for accessing the SCTP inqueue. + * + * An SCTP inqueue is a queue into which you push SCTP packets + * (which might be bundles or fragments of chunks) and out of which you + * pop SCTP whole chunks. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> +#include <linux/interrupt.h> + +/* Initialize an SCTP inqueue. */ +void sctp_inq_init(struct sctp_inq *queue) +{ + skb_queue_head_init(&queue->in); + queue->in_progress = NULL; + + /* Create a task for delivering data. */ + INIT_WORK(&queue->immediate, NULL, NULL); + + queue->malloced = 0; +} + +/* Release the memory associated with an SCTP inqueue. */ +void sctp_inq_free(struct sctp_inq *queue) +{ + struct sctp_chunk *chunk; + + /* Empty the queue. */ + while ((chunk = (struct sctp_chunk *) skb_dequeue(&queue->in)) != NULL) + sctp_chunk_free(chunk); + + /* If there is a packet which is currently being worked on, + * free it as well. + */ + if (queue->in_progress) + sctp_chunk_free(queue->in_progress); + + if (queue->malloced) { + /* Dump the master memory segment. */ + kfree(queue); + } +} + +/* Put a new packet in an SCTP inqueue. + * We assume that packet->sctp_hdr is set and in host byte order. + */ +void sctp_inq_push(struct sctp_inq *q, struct sctp_chunk *packet) +{ + /* Directly call the packet handling routine. */ + + /* We are now calling this either from the soft interrupt + * or from the backlog processing. + * Eventually, we should clean up inqueue to not rely + * on the BH related data structures. + */ + skb_queue_tail(&(q->in), (struct sk_buff *) packet); + q->immediate.func(q->immediate.data); +} + +/* Extract a chunk from an SCTP inqueue. + * + * WARNING: If you need to put the chunk on another queue, you need to + * make a shallow copy (clone) of it. + */ +struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) +{ + struct sctp_chunk *chunk; + sctp_chunkhdr_t *ch = NULL; + + /* The assumption is that we are safe to process the chunks + * at this time. + */ + + if ((chunk = queue->in_progress)) { + /* There is a packet that we have been working on. + * Any post processing work to do before we move on? + */ + if (chunk->singleton || + chunk->end_of_packet || + chunk->pdiscard) { + sctp_chunk_free(chunk); + chunk = queue->in_progress = NULL; + } else { + /* Nothing to do. Next chunk in the packet, please. */ + ch = (sctp_chunkhdr_t *) chunk->chunk_end; + + /* Force chunk->skb->data to chunk->chunk_end. */ + skb_pull(chunk->skb, + chunk->chunk_end - chunk->skb->data); + } + } + + /* Do we need to take the next packet out of the queue to process? */ + if (!chunk) { + /* Is the queue empty? */ + if (skb_queue_empty(&queue->in)) + return NULL; + + chunk = queue->in_progress = + (struct sctp_chunk *) skb_dequeue(&queue->in); + + /* This is the first chunk in the packet. */ + chunk->singleton = 1; + ch = (sctp_chunkhdr_t *) chunk->skb->data; + } + + chunk->chunk_hdr = ch; + chunk->chunk_end = ((__u8 *)ch) + WORD_ROUND(ntohs(ch->length)); + /* In the unlikely case of an IP reassembly, the skb could be + * non-linear. If so, update chunk_end so that it doesn't go past + * the skb->tail. + */ + if (unlikely(skb_is_nonlinear(chunk->skb))) { + if (chunk->chunk_end > chunk->skb->tail) + chunk->chunk_end = chunk->skb->tail; + } + skb_pull(chunk->skb, sizeof(sctp_chunkhdr_t)); + chunk->subh.v = NULL; /* Subheader is no longer valid. */ + + if (chunk->chunk_end < chunk->skb->tail) { + /* This is not a singleton */ + chunk->singleton = 0; + } else if (chunk->chunk_end > chunk->skb->tail) { + /* RFC 2960, Section 6.10 Bundling + * + * Partial chunks MUST NOT be placed in an SCTP packet. + * If the receiver detects a partial chunk, it MUST drop + * the chunk. + * + * Since the end of the chunk is past the end of our buffer + * (which contains the whole packet, we can freely discard + * the whole packet. + */ + sctp_chunk_free(chunk); + chunk = queue->in_progress = NULL; + + return NULL; + } else { + /* We are at the end of the packet, so mark the chunk + * in case we need to send a SACK. + */ + chunk->end_of_packet = 1; + } + + SCTP_DEBUG_PRINTK("+++sctp_inq_pop+++ chunk %p[%s]," + " length %d, skb->len %d\n",chunk, + sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), + ntohs(chunk->chunk_hdr->length), chunk->skb->len); + return chunk; +} + +/* Set a top-half handler. + * + * Originally, we the top-half handler was scheduled as a BH. We now + * call the handler directly in sctp_inq_push() at a time that + * we know we are lock safe. + * The intent is that this routine will pull stuff out of the + * inqueue and process it. + */ +void sctp_inq_set_th_handler(struct sctp_inq *q, + void (*callback)(void *), void *arg) +{ + INIT_WORK(&q->immediate, callback, arg); +} + diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c new file mode 100644 index 000000000000..e42c74e3ec1e --- /dev/null +++ b/net/sctp/ipv6.c @@ -0,0 +1,1013 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2002, 2004 + * Copyright (c) 2001 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * Copyright (c) 2002-2003 Intel Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * SCTP over IPv6. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Le Yanqun <yanqun.le@nokia.com> + * Hui Huang <hui.huang@nokia.com> + * La Monte H.P. Yarroll <piggy@acm.org> + * Sridhar Samudrala <sri@us.ibm.com> + * Jon Grimm <jgrimm@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * + * Based on: + * linux/net/ipv6/tcp_ipv6.c + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/module.h> +#include <linux/errno.h> +#include <linux/types.h> +#include <linux/socket.h> +#include <linux/sockios.h> +#include <linux/net.h> +#include <linux/sched.h> +#include <linux/in.h> +#include <linux/in6.h> +#include <linux/netdevice.h> +#include <linux/init.h> +#include <linux/ipsec.h> + +#include <linux/ipv6.h> +#include <linux/icmpv6.h> +#include <linux/random.h> +#include <linux/seq_file.h> + +#include <net/protocol.h> +#include <net/tcp.h> +#include <net/ndisc.h> +#include <net/ipv6.h> +#include <net/transp_v6.h> +#include <net/addrconf.h> +#include <net/ip6_route.h> +#include <net/inet_common.h> +#include <net/inet_ecn.h> +#include <net/sctp/sctp.h> + +#include <asm/uaccess.h> + +extern int sctp_inetaddr_event(struct notifier_block *, unsigned long, void *); +static struct notifier_block sctp_inet6addr_notifier = { + .notifier_call = sctp_inetaddr_event, +}; + +/* ICMP error handler. */ +SCTP_STATIC void sctp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, + int type, int code, int offset, __u32 info) +{ + struct inet6_dev *idev; + struct ipv6hdr *iph = (struct ipv6hdr *)skb->data; + struct sctphdr *sh = (struct sctphdr *)(skb->data + offset); + struct sock *sk; + struct sctp_endpoint *ep; + struct sctp_association *asoc; + struct sctp_transport *transport; + struct ipv6_pinfo *np; + char *saveip, *savesctp; + int err; + + idev = in6_dev_get(skb->dev); + + /* Fix up skb to look at the embedded net header. */ + saveip = skb->nh.raw; + savesctp = skb->h.raw; + skb->nh.ipv6h = iph; + skb->h.raw = (char *)sh; + sk = sctp_err_lookup(AF_INET6, skb, sh, &ep, &asoc, &transport); + /* Put back, the original pointers. */ + skb->nh.raw = saveip; + skb->h.raw = savesctp; + if (!sk) { + ICMP6_INC_STATS_BH(idev, ICMP6_MIB_INERRORS); + goto out; + } + + /* Warning: The sock lock is held. Remember to call + * sctp_err_finish! + */ + + switch (type) { + case ICMPV6_PKT_TOOBIG: + sctp_icmp_frag_needed(sk, asoc, transport, ntohl(info)); + goto out_unlock; + case ICMPV6_PARAMPROB: + if (ICMPV6_UNK_NEXTHDR == code) { + sctp_icmp_proto_unreachable(sk, ep, asoc, transport); + goto out_unlock; + } + break; + default: + break; + } + + np = inet6_sk(sk); + icmpv6_err_convert(type, code, &err); + if (!sock_owned_by_user(sk) && np->recverr) { + sk->sk_err = err; + sk->sk_error_report(sk); + } else { /* Only an error on timeout */ + sk->sk_err_soft = err; + } + +out_unlock: + sctp_err_finish(sk, ep, asoc); +out: + if (likely(idev != NULL)) + in6_dev_put(idev); +} + +/* Based on tcp_v6_xmit() in tcp_ipv6.c. */ +static int sctp_v6_xmit(struct sk_buff *skb, struct sctp_transport *transport, + int ipfragok) +{ + struct sock *sk = skb->sk; + struct ipv6_pinfo *np = inet6_sk(sk); + struct flowi fl; + + memset(&fl, 0, sizeof(fl)); + + fl.proto = sk->sk_protocol; + + /* Fill in the dest address from the route entry passed with the skb + * and the source address from the transport. + */ + ipv6_addr_copy(&fl.fl6_dst, &transport->ipaddr.v6.sin6_addr); + ipv6_addr_copy(&fl.fl6_src, &transport->saddr.v6.sin6_addr); + + fl.fl6_flowlabel = np->flow_label; + IP6_ECN_flow_xmit(sk, fl.fl6_flowlabel); + if (ipv6_addr_type(&fl.fl6_src) & IPV6_ADDR_LINKLOCAL) + fl.oif = transport->saddr.v6.sin6_scope_id; + else + fl.oif = sk->sk_bound_dev_if; + fl.fl_ip_sport = inet_sk(sk)->sport; + fl.fl_ip_dport = transport->ipaddr.v6.sin6_port; + + if (np->opt && np->opt->srcrt) { + struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt; + ipv6_addr_copy(&fl.fl6_dst, rt0->addr); + } + + SCTP_DEBUG_PRINTK("%s: skb:%p, len:%d, " + "src:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x " + "dst:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + __FUNCTION__, skb, skb->len, + NIP6(fl.fl6_src), NIP6(fl.fl6_dst)); + + SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS); + + return ip6_xmit(sk, skb, &fl, np->opt, ipfragok); +} + +/* Returns the dst cache entry for the given source and destination ip + * addresses. + */ +static struct dst_entry *sctp_v6_get_dst(struct sctp_association *asoc, + union sctp_addr *daddr, + union sctp_addr *saddr) +{ + struct dst_entry *dst; + struct flowi fl; + + memset(&fl, 0, sizeof(fl)); + ipv6_addr_copy(&fl.fl6_dst, &daddr->v6.sin6_addr); + if (ipv6_addr_type(&daddr->v6.sin6_addr) & IPV6_ADDR_LINKLOCAL) + fl.oif = daddr->v6.sin6_scope_id; + + + SCTP_DEBUG_PRINTK("%s: DST=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", + __FUNCTION__, NIP6(fl.fl6_dst)); + + if (saddr) { + ipv6_addr_copy(&fl.fl6_src, &saddr->v6.sin6_addr); + SCTP_DEBUG_PRINTK( + "SRC=%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x - ", + NIP6(fl.fl6_src)); + } + + dst = ip6_route_output(NULL, &fl); + if (dst) { + struct rt6_info *rt; + rt = (struct rt6_info *)dst; + SCTP_DEBUG_PRINTK( + "rt6_dst:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x " + "rt6_src:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + NIP6(rt->rt6i_dst.addr), NIP6(rt->rt6i_src.addr)); + } else { + SCTP_DEBUG_PRINTK("NO ROUTE\n"); + } + + return dst; +} + +/* Returns the number of consecutive initial bits that match in the 2 ipv6 + * addresses. + */ +static inline int sctp_v6_addr_match_len(union sctp_addr *s1, + union sctp_addr *s2) +{ + struct in6_addr *a1 = &s1->v6.sin6_addr; + struct in6_addr *a2 = &s2->v6.sin6_addr; + int i, j; + + for (i = 0; i < 4 ; i++) { + __u32 a1xora2; + + a1xora2 = a1->s6_addr32[i] ^ a2->s6_addr32[i]; + + if ((j = fls(ntohl(a1xora2)))) + return (i * 32 + 32 - j); + } + + return (i*32); +} + +/* Fills in the source address(saddr) based on the destination address(daddr) + * and asoc's bind address list. + */ +static void sctp_v6_get_saddr(struct sctp_association *asoc, + struct dst_entry *dst, + union sctp_addr *daddr, + union sctp_addr *saddr) +{ + struct sctp_bind_addr *bp; + rwlock_t *addr_lock; + struct sctp_sockaddr_entry *laddr; + struct list_head *pos; + sctp_scope_t scope; + union sctp_addr *baddr = NULL; + __u8 matchlen = 0; + __u8 bmatchlen; + + SCTP_DEBUG_PRINTK("%s: asoc:%p dst:%p " + "daddr:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", + __FUNCTION__, asoc, dst, NIP6(daddr->v6.sin6_addr)); + + if (!asoc) { + ipv6_get_saddr(dst, &daddr->v6.sin6_addr,&saddr->v6.sin6_addr); + SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " + "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + NIP6(saddr->v6.sin6_addr)); + return; + } + + scope = sctp_scope(daddr); + + bp = &asoc->base.bind_addr; + addr_lock = &asoc->base.addr_lock; + + /* Go through the bind address list and find the best source address + * that matches the scope of the destination address. + */ + sctp_read_lock(addr_lock); + list_for_each(pos, &bp->address_list) { + laddr = list_entry(pos, struct sctp_sockaddr_entry, list); + if ((laddr->a.sa.sa_family == AF_INET6) && + (scope <= sctp_scope(&laddr->a))) { + bmatchlen = sctp_v6_addr_match_len(daddr, &laddr->a); + if (!baddr || (matchlen < bmatchlen)) { + baddr = &laddr->a; + matchlen = bmatchlen; + } + } + } + + if (baddr) { + memcpy(saddr, baddr, sizeof(union sctp_addr)); + SCTP_DEBUG_PRINTK("saddr: " + "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + NIP6(saddr->v6.sin6_addr)); + } else { + printk(KERN_ERR "%s: asoc:%p Could not find a valid source " + "address for the " + "dest:%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", + __FUNCTION__, asoc, NIP6(daddr->v6.sin6_addr)); + } + + sctp_read_unlock(addr_lock); +} + +/* Make a copy of all potential local addresses. */ +static void sctp_v6_copy_addrlist(struct list_head *addrlist, + struct net_device *dev) +{ + struct inet6_dev *in6_dev; + struct inet6_ifaddr *ifp; + struct sctp_sockaddr_entry *addr; + + read_lock(&addrconf_lock); + if ((in6_dev = __in6_dev_get(dev)) == NULL) { + read_unlock(&addrconf_lock); + return; + } + + read_lock(&in6_dev->lock); + for (ifp = in6_dev->addr_list; ifp; ifp = ifp->if_next) { + /* Add the address to the local list. */ + addr = t_new(struct sctp_sockaddr_entry, GFP_ATOMIC); + if (addr) { + addr->a.v6.sin6_family = AF_INET6; + addr->a.v6.sin6_port = 0; + addr->a.v6.sin6_addr = ifp->addr; + addr->a.v6.sin6_scope_id = dev->ifindex; + INIT_LIST_HEAD(&addr->list); + list_add_tail(&addr->list, addrlist); + } + } + + read_unlock(&in6_dev->lock); + read_unlock(&addrconf_lock); +} + +/* Initialize a sockaddr_storage from in incoming skb. */ +static void sctp_v6_from_skb(union sctp_addr *addr,struct sk_buff *skb, + int is_saddr) +{ + void *from; + __u16 *port; + struct sctphdr *sh; + + port = &addr->v6.sin6_port; + addr->v6.sin6_family = AF_INET6; + addr->v6.sin6_flowinfo = 0; /* FIXME */ + addr->v6.sin6_scope_id = ((struct inet6_skb_parm *)skb->cb)->iif; + + sh = (struct sctphdr *) skb->h.raw; + if (is_saddr) { + *port = ntohs(sh->source); + from = &skb->nh.ipv6h->saddr; + } else { + *port = ntohs(sh->dest); + from = &skb->nh.ipv6h->daddr; + } + ipv6_addr_copy(&addr->v6.sin6_addr, from); +} + +/* Initialize an sctp_addr from a socket. */ +static void sctp_v6_from_sk(union sctp_addr *addr, struct sock *sk) +{ + addr->v6.sin6_family = AF_INET6; + addr->v6.sin6_port = inet_sk(sk)->num; + addr->v6.sin6_addr = inet6_sk(sk)->rcv_saddr; +} + +/* Initialize sk->sk_rcv_saddr from sctp_addr. */ +static void sctp_v6_to_sk_saddr(union sctp_addr *addr, struct sock *sk) +{ + if (addr->sa.sa_family == AF_INET && sctp_sk(sk)->v4mapped) { + inet6_sk(sk)->rcv_saddr.s6_addr32[0] = 0; + inet6_sk(sk)->rcv_saddr.s6_addr32[1] = 0; + inet6_sk(sk)->rcv_saddr.s6_addr32[2] = htonl(0x0000ffff); + inet6_sk(sk)->rcv_saddr.s6_addr32[3] = + addr->v4.sin_addr.s_addr; + } else { + inet6_sk(sk)->rcv_saddr = addr->v6.sin6_addr; + } +} + +/* Initialize sk->sk_daddr from sctp_addr. */ +static void sctp_v6_to_sk_daddr(union sctp_addr *addr, struct sock *sk) +{ + if (addr->sa.sa_family == AF_INET && sctp_sk(sk)->v4mapped) { + inet6_sk(sk)->daddr.s6_addr32[0] = 0; + inet6_sk(sk)->daddr.s6_addr32[1] = 0; + inet6_sk(sk)->daddr.s6_addr32[2] = htonl(0x0000ffff); + inet6_sk(sk)->daddr.s6_addr32[3] = addr->v4.sin_addr.s_addr; + } else { + inet6_sk(sk)->daddr = addr->v6.sin6_addr; + } +} + +/* Initialize a sctp_addr from an address parameter. */ +static void sctp_v6_from_addr_param(union sctp_addr *addr, + union sctp_addr_param *param, + __u16 port, int iif) +{ + addr->v6.sin6_family = AF_INET6; + addr->v6.sin6_port = port; + addr->v6.sin6_flowinfo = 0; /* BUG */ + ipv6_addr_copy(&addr->v6.sin6_addr, ¶m->v6.addr); + addr->v6.sin6_scope_id = iif; +} + +/* Initialize an address parameter from a sctp_addr and return the length + * of the address parameter. + */ +static int sctp_v6_to_addr_param(const union sctp_addr *addr, + union sctp_addr_param *param) +{ + int length = sizeof(sctp_ipv6addr_param_t); + + param->v6.param_hdr.type = SCTP_PARAM_IPV6_ADDRESS; + param->v6.param_hdr.length = ntohs(length); + ipv6_addr_copy(¶m->v6.addr, &addr->v6.sin6_addr); + + return length; +} + +/* Initialize a sctp_addr from a dst_entry. */ +static void sctp_v6_dst_saddr(union sctp_addr *addr, struct dst_entry *dst, + unsigned short port) +{ + struct rt6_info *rt = (struct rt6_info *)dst; + addr->sa.sa_family = AF_INET6; + addr->v6.sin6_port = port; + ipv6_addr_copy(&addr->v6.sin6_addr, &rt->rt6i_src.addr); +} + +/* Compare addresses exactly. + * v4-mapped-v6 is also in consideration. + */ +static int sctp_v6_cmp_addr(const union sctp_addr *addr1, + const union sctp_addr *addr2) +{ + if (addr1->sa.sa_family != addr2->sa.sa_family) { + if (addr1->sa.sa_family == AF_INET && + addr2->sa.sa_family == AF_INET6 && + IPV6_ADDR_MAPPED == ipv6_addr_type(&addr2->v6.sin6_addr)) { + if (addr2->v6.sin6_port == addr1->v4.sin_port && + addr2->v6.sin6_addr.s6_addr32[3] == + addr1->v4.sin_addr.s_addr) + return 1; + } + if (addr2->sa.sa_family == AF_INET && + addr1->sa.sa_family == AF_INET6 && + IPV6_ADDR_MAPPED == ipv6_addr_type(&addr1->v6.sin6_addr)) { + if (addr1->v6.sin6_port == addr2->v4.sin_port && + addr1->v6.sin6_addr.s6_addr32[3] == + addr2->v4.sin_addr.s_addr) + return 1; + } + return 0; + } + if (!ipv6_addr_equal(&addr1->v6.sin6_addr, &addr2->v6.sin6_addr)) + return 0; + /* If this is a linklocal address, compare the scope_id. */ + if (ipv6_addr_type(&addr1->v6.sin6_addr) & IPV6_ADDR_LINKLOCAL) { + if (addr1->v6.sin6_scope_id && addr2->v6.sin6_scope_id && + (addr1->v6.sin6_scope_id != addr2->v6.sin6_scope_id)) { + return 0; + } + } + + return 1; +} + +/* Initialize addr struct to INADDR_ANY. */ +static void sctp_v6_inaddr_any(union sctp_addr *addr, unsigned short port) +{ + memset(addr, 0x00, sizeof(union sctp_addr)); + addr->v6.sin6_family = AF_INET6; + addr->v6.sin6_port = port; +} + +/* Is this a wildcard address? */ +static int sctp_v6_is_any(const union sctp_addr *addr) +{ + int type; + type = ipv6_addr_type((struct in6_addr *)&addr->v6.sin6_addr); + return IPV6_ADDR_ANY == type; +} + +/* Should this be available for binding? */ +static int sctp_v6_available(union sctp_addr *addr, struct sctp_sock *sp) +{ + int type; + struct in6_addr *in6 = (struct in6_addr *)&addr->v6.sin6_addr; + + type = ipv6_addr_type(in6); + if (IPV6_ADDR_ANY == type) + return 1; + if (type == IPV6_ADDR_MAPPED) { + if (sp && !sp->v4mapped) + return 0; + if (sp && ipv6_only_sock(sctp_opt2sk(sp))) + return 0; + sctp_v6_map_v4(addr); + return sctp_get_af_specific(AF_INET)->available(addr, sp); + } + if (!(type & IPV6_ADDR_UNICAST)) + return 0; + + return ipv6_chk_addr(in6, NULL, 0); +} + +/* This function checks if the address is a valid address to be used for + * SCTP. + * + * Output: + * Return 0 - If the address is a non-unicast or an illegal address. + * Return 1 - If the address is a unicast. + */ +static int sctp_v6_addr_valid(union sctp_addr *addr, struct sctp_sock *sp) +{ + int ret = ipv6_addr_type(&addr->v6.sin6_addr); + + /* Support v4-mapped-v6 address. */ + if (ret == IPV6_ADDR_MAPPED) { + /* Note: This routine is used in input, so v4-mapped-v6 + * are disallowed here when there is no sctp_sock. + */ + if (!sp || !sp->v4mapped) + return 0; + if (sp && ipv6_only_sock(sctp_opt2sk(sp))) + return 0; + sctp_v6_map_v4(addr); + return sctp_get_af_specific(AF_INET)->addr_valid(addr, sp); + } + + /* Is this a non-unicast address */ + if (!(ret & IPV6_ADDR_UNICAST)) + return 0; + + return 1; +} + +/* What is the scope of 'addr'? */ +static sctp_scope_t sctp_v6_scope(union sctp_addr *addr) +{ + int v6scope; + sctp_scope_t retval; + + /* The IPv6 scope is really a set of bit fields. + * See IFA_* in <net/if_inet6.h>. Map to a generic SCTP scope. + */ + + v6scope = ipv6_addr_scope(&addr->v6.sin6_addr); + switch (v6scope) { + case IFA_HOST: + retval = SCTP_SCOPE_LOOPBACK; + break; + case IFA_LINK: + retval = SCTP_SCOPE_LINK; + break; + case IFA_SITE: + retval = SCTP_SCOPE_PRIVATE; + break; + default: + retval = SCTP_SCOPE_GLOBAL; + break; + }; + + return retval; +} + +/* Create and initialize a new sk for the socket to be returned by accept(). */ +static struct sock *sctp_v6_create_accept_sk(struct sock *sk, + struct sctp_association *asoc) +{ + struct inet_sock *inet = inet_sk(sk); + struct sock *newsk; + struct inet_sock *newinet; + struct ipv6_pinfo *newnp, *np = inet6_sk(sk); + struct sctp6_sock *newsctp6sk; + + newsk = sk_alloc(PF_INET6, GFP_KERNEL, sk->sk_prot, 1); + if (!newsk) + goto out; + + sock_init_data(NULL, newsk); + + newsk->sk_type = SOCK_STREAM; + + newsk->sk_prot = sk->sk_prot; + newsk->sk_no_check = sk->sk_no_check; + newsk->sk_reuse = sk->sk_reuse; + + newsk->sk_destruct = inet_sock_destruct; + newsk->sk_family = PF_INET6; + newsk->sk_protocol = IPPROTO_SCTP; + newsk->sk_backlog_rcv = sk->sk_prot->backlog_rcv; + newsk->sk_shutdown = sk->sk_shutdown; + sock_reset_flag(sk, SOCK_ZAPPED); + + newsctp6sk = (struct sctp6_sock *)newsk; + inet_sk(newsk)->pinet6 = &newsctp6sk->inet6; + + newinet = inet_sk(newsk); + newnp = inet6_sk(newsk); + + memcpy(newnp, np, sizeof(struct ipv6_pinfo)); + + /* Initialize sk's sport, dport, rcv_saddr and daddr for getsockname() + * and getpeername(). + */ + newinet->sport = inet->sport; + newnp->saddr = np->saddr; + newnp->rcv_saddr = np->rcv_saddr; + newinet->dport = htons(asoc->peer.port); + sctp_v6_to_sk_daddr(&asoc->peer.primary_addr, newsk); + + /* Init the ipv4 part of the socket since we can have sockets + * using v6 API for ipv4. + */ + newinet->uc_ttl = -1; + newinet->mc_loop = 1; + newinet->mc_ttl = 1; + newinet->mc_index = 0; + newinet->mc_list = NULL; + + if (ipv4_config.no_pmtu_disc) + newinet->pmtudisc = IP_PMTUDISC_DONT; + else + newinet->pmtudisc = IP_PMTUDISC_WANT; + +#ifdef INET_REFCNT_DEBUG + atomic_inc(&inet6_sock_nr); + atomic_inc(&inet_sock_nr); +#endif + + if (newsk->sk_prot->init(newsk)) { + sk_common_release(newsk); + newsk = NULL; + } + +out: + return newsk; +} + +/* Map v4 address to mapped v6 address */ +static void sctp_v6_addr_v4map(struct sctp_sock *sp, union sctp_addr *addr) +{ + if (sp->v4mapped && AF_INET == addr->sa.sa_family) + sctp_v4_map_v6(addr); +} + +/* Where did this skb come from? */ +static int sctp_v6_skb_iif(const struct sk_buff *skb) +{ + struct inet6_skb_parm *opt = (struct inet6_skb_parm *) skb->cb; + return opt->iif; +} + +/* Was this packet marked by Explicit Congestion Notification? */ +static int sctp_v6_is_ce(const struct sk_buff *skb) +{ + return *((__u32 *)(skb->nh.ipv6h)) & htonl(1<<20); +} + +/* Dump the v6 addr to the seq file. */ +static void sctp_v6_seq_dump_addr(struct seq_file *seq, union sctp_addr *addr) +{ + seq_printf(seq, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x ", + NIP6(addr->v6.sin6_addr)); +} + +/* Initialize a PF_INET6 socket msg_name. */ +static void sctp_inet6_msgname(char *msgname, int *addr_len) +{ + struct sockaddr_in6 *sin6; + + sin6 = (struct sockaddr_in6 *)msgname; + sin6->sin6_family = AF_INET6; + sin6->sin6_flowinfo = 0; + sin6->sin6_scope_id = 0; /*FIXME */ + *addr_len = sizeof(struct sockaddr_in6); +} + +/* Initialize a PF_INET msgname from a ulpevent. */ +static void sctp_inet6_event_msgname(struct sctp_ulpevent *event, + char *msgname, int *addrlen) +{ + struct sockaddr_in6 *sin6, *sin6from; + + if (msgname) { + union sctp_addr *addr; + struct sctp_association *asoc; + + asoc = event->asoc; + sctp_inet6_msgname(msgname, addrlen); + sin6 = (struct sockaddr_in6 *)msgname; + sin6->sin6_port = htons(asoc->peer.port); + addr = &asoc->peer.primary_addr; + + /* Note: If we go to a common v6 format, this code + * will change. + */ + + /* Map ipv4 address into v4-mapped-on-v6 address. */ + if (sctp_sk(asoc->base.sk)->v4mapped && + AF_INET == addr->sa.sa_family) { + sctp_v4_map_v6((union sctp_addr *)sin6); + sin6->sin6_addr.s6_addr32[3] = + addr->v4.sin_addr.s_addr; + return; + } + + sin6from = &asoc->peer.primary_addr.v6; + ipv6_addr_copy(&sin6->sin6_addr, &sin6from->sin6_addr); + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL) + sin6->sin6_scope_id = sin6from->sin6_scope_id; + } +} + +/* Initialize a msg_name from an inbound skb. */ +static void sctp_inet6_skb_msgname(struct sk_buff *skb, char *msgname, + int *addr_len) +{ + struct sctphdr *sh; + struct sockaddr_in6 *sin6; + + if (msgname) { + sctp_inet6_msgname(msgname, addr_len); + sin6 = (struct sockaddr_in6 *)msgname; + sh = (struct sctphdr *)skb->h.raw; + sin6->sin6_port = sh->source; + + /* Map ipv4 address into v4-mapped-on-v6 address. */ + if (sctp_sk(skb->sk)->v4mapped && + skb->nh.iph->version == 4) { + sctp_v4_map_v6((union sctp_addr *)sin6); + sin6->sin6_addr.s6_addr32[3] = skb->nh.iph->saddr; + return; + } + + /* Otherwise, just copy the v6 address. */ + ipv6_addr_copy(&sin6->sin6_addr, &skb->nh.ipv6h->saddr); + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL) { + struct sctp_ulpevent *ev = sctp_skb2event(skb); + sin6->sin6_scope_id = ev->iif; + } + } +} + +/* Do we support this AF? */ +static int sctp_inet6_af_supported(sa_family_t family, struct sctp_sock *sp) +{ + switch (family) { + case AF_INET6: + return 1; + /* v4-mapped-v6 addresses */ + case AF_INET: + if (!__ipv6_only_sock(sctp_opt2sk(sp)) && sp->v4mapped) + return 1; + default: + return 0; + } +} + +/* Address matching with wildcards allowed. This extra level + * of indirection lets us choose whether a PF_INET6 should + * disallow any v4 addresses if we so choose. + */ +static int sctp_inet6_cmp_addr(const union sctp_addr *addr1, + const union sctp_addr *addr2, + struct sctp_sock *opt) +{ + struct sctp_af *af1, *af2; + + af1 = sctp_get_af_specific(addr1->sa.sa_family); + af2 = sctp_get_af_specific(addr2->sa.sa_family); + + if (!af1 || !af2) + return 0; + /* Today, wildcard AF_INET/AF_INET6. */ + if (sctp_is_any(addr1) || sctp_is_any(addr2)) + return 1; + + if (addr1->sa.sa_family != addr2->sa.sa_family) + return 0; + + return af1->cmp_addr(addr1, addr2); +} + +/* Verify that the provided sockaddr looks bindable. Common verification, + * has already been taken care of. + */ +static int sctp_inet6_bind_verify(struct sctp_sock *opt, union sctp_addr *addr) +{ + struct sctp_af *af; + + /* ASSERT: address family has already been verified. */ + if (addr->sa.sa_family != AF_INET6) + af = sctp_get_af_specific(addr->sa.sa_family); + else { + struct sock *sk; + int type = ipv6_addr_type(&addr->v6.sin6_addr); + sk = sctp_opt2sk(opt); + if (type & IPV6_ADDR_LINKLOCAL) { + /* Note: Behavior similar to af_inet6.c: + * 1) Overrides previous bound_dev_if + * 2) Destructive even if bind isn't successful. + */ + + if (addr->v6.sin6_scope_id) + sk->sk_bound_dev_if = addr->v6.sin6_scope_id; + if (!sk->sk_bound_dev_if) + return 0; + } + af = opt->pf->af; + } + return af->available(addr, opt); +} + +/* Verify that the provided sockaddr looks bindable. Common verification, + * has already been taken care of. + */ +static int sctp_inet6_send_verify(struct sctp_sock *opt, union sctp_addr *addr) +{ + struct sctp_af *af = NULL; + + /* ASSERT: address family has already been verified. */ + if (addr->sa.sa_family != AF_INET6) + af = sctp_get_af_specific(addr->sa.sa_family); + else { + struct sock *sk; + int type = ipv6_addr_type(&addr->v6.sin6_addr); + sk = sctp_opt2sk(opt); + if (type & IPV6_ADDR_LINKLOCAL) { + /* Note: Behavior similar to af_inet6.c: + * 1) Overrides previous bound_dev_if + * 2) Destructive even if bind isn't successful. + */ + + if (addr->v6.sin6_scope_id) + sk->sk_bound_dev_if = addr->v6.sin6_scope_id; + if (!sk->sk_bound_dev_if) + return 0; + } + af = opt->pf->af; + } + + return af != NULL; +} + +/* Fill in Supported Address Type information for INIT and INIT-ACK + * chunks. Note: In the future, we may want to look at sock options + * to determine whether a PF_INET6 socket really wants to have IPV4 + * addresses. + * Returns number of addresses supported. + */ +static int sctp_inet6_supported_addrs(const struct sctp_sock *opt, + __u16 *types) +{ + types[0] = SCTP_PARAM_IPV4_ADDRESS; + types[1] = SCTP_PARAM_IPV6_ADDRESS; + return 2; +} + +static struct proto_ops inet6_seqpacket_ops = { + .family = PF_INET6, + .owner = THIS_MODULE, + .release = inet6_release, + .bind = inet6_bind, + .connect = inet_dgram_connect, + .socketpair = sock_no_socketpair, + .accept = inet_accept, + .getname = inet6_getname, + .poll = sctp_poll, + .ioctl = inet6_ioctl, + .listen = sctp_inet_listen, + .shutdown = inet_shutdown, + .setsockopt = sock_common_setsockopt, + .getsockopt = sock_common_getsockopt, + .sendmsg = inet_sendmsg, + .recvmsg = sock_common_recvmsg, + .mmap = sock_no_mmap, +}; + +static struct inet_protosw sctpv6_seqpacket_protosw = { + .type = SOCK_SEQPACKET, + .protocol = IPPROTO_SCTP, + .prot = &sctpv6_prot, + .ops = &inet6_seqpacket_ops, + .capability = -1, + .no_check = 0, + .flags = SCTP_PROTOSW_FLAG +}; +static struct inet_protosw sctpv6_stream_protosw = { + .type = SOCK_STREAM, + .protocol = IPPROTO_SCTP, + .prot = &sctpv6_prot, + .ops = &inet6_seqpacket_ops, + .capability = -1, + .no_check = 0, + .flags = SCTP_PROTOSW_FLAG, +}; + +static int sctp6_rcv(struct sk_buff **pskb, unsigned int *nhoffp) +{ + return sctp_rcv(*pskb) ? -1 : 0; +} + +static struct inet6_protocol sctpv6_protocol = { + .handler = sctp6_rcv, + .err_handler = sctp_v6_err, + .flags = INET6_PROTO_NOPOLICY | INET6_PROTO_FINAL, +}; + +static struct sctp_af sctp_ipv6_specific = { + .sctp_xmit = sctp_v6_xmit, + .setsockopt = ipv6_setsockopt, + .getsockopt = ipv6_getsockopt, + .get_dst = sctp_v6_get_dst, + .get_saddr = sctp_v6_get_saddr, + .copy_addrlist = sctp_v6_copy_addrlist, + .from_skb = sctp_v6_from_skb, + .from_sk = sctp_v6_from_sk, + .to_sk_saddr = sctp_v6_to_sk_saddr, + .to_sk_daddr = sctp_v6_to_sk_daddr, + .from_addr_param = sctp_v6_from_addr_param, + .to_addr_param = sctp_v6_to_addr_param, + .dst_saddr = sctp_v6_dst_saddr, + .cmp_addr = sctp_v6_cmp_addr, + .scope = sctp_v6_scope, + .addr_valid = sctp_v6_addr_valid, + .inaddr_any = sctp_v6_inaddr_any, + .is_any = sctp_v6_is_any, + .available = sctp_v6_available, + .skb_iif = sctp_v6_skb_iif, + .is_ce = sctp_v6_is_ce, + .seq_dump_addr = sctp_v6_seq_dump_addr, + .net_header_len = sizeof(struct ipv6hdr), + .sockaddr_len = sizeof(struct sockaddr_in6), + .sa_family = AF_INET6, +}; + +static struct sctp_pf sctp_pf_inet6_specific = { + .event_msgname = sctp_inet6_event_msgname, + .skb_msgname = sctp_inet6_skb_msgname, + .af_supported = sctp_inet6_af_supported, + .cmp_addr = sctp_inet6_cmp_addr, + .bind_verify = sctp_inet6_bind_verify, + .send_verify = sctp_inet6_send_verify, + .supported_addrs = sctp_inet6_supported_addrs, + .create_accept_sk = sctp_v6_create_accept_sk, + .addr_v4map = sctp_v6_addr_v4map, + .af = &sctp_ipv6_specific, +}; + +/* Initialize IPv6 support and register with inet6 stack. */ +int sctp_v6_init(void) +{ + int rc = proto_register(&sctpv6_prot, 1); + + if (rc) + goto out; + /* Register inet6 protocol. */ + rc = -EAGAIN; + if (inet6_add_protocol(&sctpv6_protocol, IPPROTO_SCTP) < 0) + goto out_unregister_sctp_proto; + + /* Add SCTPv6(UDP and TCP style) to inetsw6 linked list. */ + inet6_register_protosw(&sctpv6_seqpacket_protosw); + inet6_register_protosw(&sctpv6_stream_protosw); + + /* Register the SCTP specific PF_INET6 functions. */ + sctp_register_pf(&sctp_pf_inet6_specific, PF_INET6); + + /* Register the SCTP specific AF_INET6 functions. */ + sctp_register_af(&sctp_ipv6_specific); + + /* Register notifier for inet6 address additions/deletions. */ + register_inet6addr_notifier(&sctp_inet6addr_notifier); + rc = 0; +out: + return rc; +out_unregister_sctp_proto: + proto_unregister(&sctpv6_prot); + goto out; +} + +/* IPv6 specific exit support. */ +void sctp_v6_exit(void) +{ + list_del(&sctp_ipv6_specific.list); + inet6_del_protocol(&sctpv6_protocol, IPPROTO_SCTP); + inet6_unregister_protosw(&sctpv6_seqpacket_protosw); + inet6_unregister_protosw(&sctpv6_stream_protosw); + unregister_inet6addr_notifier(&sctp_inet6addr_notifier); + proto_unregister(&sctpv6_prot); +} diff --git a/net/sctp/objcnt.c b/net/sctp/objcnt.c new file mode 100644 index 000000000000..0781e5d509fd --- /dev/null +++ b/net/sctp/objcnt.c @@ -0,0 +1,140 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * + * This file is part of the SCTP kernel reference Implementation + * + * Support for memory object debugging. This allows one to monitor the + * object allocations/deallocations for types instrumented for this + * via the proc fs. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Jon Grimm <jgrimm@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/kernel.h> +#include <net/sctp/sctp.h> + +/* + * Global counters to count raw object allocation counts. + * To add new counters, choose a unique suffix for the variable + * name as the helper macros key off this suffix to make + * life easier for the programmer. + */ + +SCTP_DBG_OBJCNT(sock); +SCTP_DBG_OBJCNT(ep); +SCTP_DBG_OBJCNT(transport); +SCTP_DBG_OBJCNT(assoc); +SCTP_DBG_OBJCNT(bind_addr); +SCTP_DBG_OBJCNT(bind_bucket); +SCTP_DBG_OBJCNT(chunk); +SCTP_DBG_OBJCNT(addr); +SCTP_DBG_OBJCNT(ssnmap); +SCTP_DBG_OBJCNT(datamsg); + +/* An array to make it easy to pretty print the debug information + * to the proc fs. + */ +static sctp_dbg_objcnt_entry_t sctp_dbg_objcnt[] = { + SCTP_DBG_OBJCNT_ENTRY(sock), + SCTP_DBG_OBJCNT_ENTRY(ep), + SCTP_DBG_OBJCNT_ENTRY(assoc), + SCTP_DBG_OBJCNT_ENTRY(transport), + SCTP_DBG_OBJCNT_ENTRY(chunk), + SCTP_DBG_OBJCNT_ENTRY(bind_addr), + SCTP_DBG_OBJCNT_ENTRY(bind_bucket), + SCTP_DBG_OBJCNT_ENTRY(addr), + SCTP_DBG_OBJCNT_ENTRY(ssnmap), + SCTP_DBG_OBJCNT_ENTRY(datamsg), +}; + +/* Callback from procfs to read out objcount information. + * Walk through the entries in the sctp_dbg_objcnt array, dumping + * the raw object counts for each monitored type. + * + * This code was modified from similar code in route.c + */ +static int sctp_dbg_objcnt_read(char *buffer, char **start, off_t offset, + int length, int *eof, void *data) +{ + int len = 0; + off_t pos = 0; + int entries; + int i; + char temp[128]; + + /* How many entries? */ + entries = ARRAY_SIZE(sctp_dbg_objcnt); + + /* Walk the entries and print out the debug information + * for proc fs. + */ + for (i = 0; i < entries; i++) { + pos += 128; + + /* Skip ahead. */ + if (pos <= offset) { + len = 0; + continue; + } + /* Print out each entry. */ + sprintf(temp, "%s: %d", + sctp_dbg_objcnt[i].label, + atomic_read(sctp_dbg_objcnt[i].counter)); + + sprintf(buffer + len, "%-127s\n", temp); + len += 128; + if (pos >= offset+length) + goto done; + } + +done: + *start = buffer + len - (pos - offset); + len = pos - offset; + if (len > length) + len = length; + + return len; +} + +/* Initialize the objcount in the proc filesystem. */ +void sctp_dbg_objcnt_init(void) +{ + create_proc_read_entry("sctp_dbg_objcnt", 0, proc_net_sctp, + sctp_dbg_objcnt_read, NULL); +} + +/* Cleanup the objcount entry in the proc filesystem. */ +void sctp_dbg_objcnt_exit(void) +{ + remove_proc_entry("sctp_dbg_objcnt", proc_net_sctp); +} + + diff --git a/net/sctp/output.c b/net/sctp/output.c new file mode 100644 index 000000000000..9013f64f5219 --- /dev/null +++ b/net/sctp/output.c @@ -0,0 +1,646 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions handle output processing. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@austin.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/wait.h> +#include <linux/time.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/init.h> +#include <net/inet_ecn.h> +#include <net/icmp.h> + +#ifndef TEST_FRAME +#include <net/tcp.h> +#endif /* TEST_FRAME (not defined) */ + +#include <linux/socket.h> /* for sa_family_t */ +#include <net/sock.h> + +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Forward declarations for private helpers. */ +static sctp_xmit_t sctp_packet_append_data(struct sctp_packet *packet, + struct sctp_chunk *chunk); + +/* Config a packet. + * This appears to be a followup set of initializations. + */ +struct sctp_packet *sctp_packet_config(struct sctp_packet *packet, + __u32 vtag, int ecn_capable) +{ + struct sctp_chunk *chunk = NULL; + + SCTP_DEBUG_PRINTK("%s: packet:%p vtag:0x%x\n", __FUNCTION__, + packet, vtag); + + packet->vtag = vtag; + packet->has_cookie_echo = 0; + packet->has_sack = 0; + packet->ipfragok = 0; + + if (ecn_capable && sctp_packet_empty(packet)) { + chunk = sctp_get_ecne_prepend(packet->transport->asoc); + + /* If there a is a prepend chunk stick it on the list before + * any other chunks get appended. + */ + if (chunk) + sctp_packet_append_chunk(packet, chunk); + } + + return packet; +} + +/* Initialize the packet structure. */ +struct sctp_packet *sctp_packet_init(struct sctp_packet *packet, + struct sctp_transport *transport, + __u16 sport, __u16 dport) +{ + struct sctp_association *asoc = transport->asoc; + size_t overhead; + + SCTP_DEBUG_PRINTK("%s: packet:%p transport:%p\n", __FUNCTION__, + packet, transport); + + packet->transport = transport; + packet->source_port = sport; + packet->destination_port = dport; + skb_queue_head_init(&packet->chunks); + if (asoc) { + struct sctp_sock *sp = sctp_sk(asoc->base.sk); + overhead = sp->pf->af->net_header_len; + } else { + overhead = sizeof(struct ipv6hdr); + } + overhead += sizeof(struct sctphdr); + packet->overhead = overhead; + packet->size = overhead; + packet->vtag = 0; + packet->has_cookie_echo = 0; + packet->has_sack = 0; + packet->ipfragok = 0; + packet->malloced = 0; + return packet; +} + +/* Free a packet. */ +void sctp_packet_free(struct sctp_packet *packet) +{ + struct sctp_chunk *chunk; + + SCTP_DEBUG_PRINTK("%s: packet:%p\n", __FUNCTION__, packet); + + while ((chunk = (struct sctp_chunk *)__skb_dequeue(&packet->chunks)) != NULL) + sctp_chunk_free(chunk); + + if (packet->malloced) + kfree(packet); +} + +/* This routine tries to append the chunk to the offered packet. If adding + * the chunk causes the packet to exceed the path MTU and COOKIE_ECHO chunk + * is not present in the packet, it transmits the input packet. + * Data can be bundled with a packet containing a COOKIE_ECHO chunk as long + * as it can fit in the packet, but any more data that does not fit in this + * packet can be sent only after receiving the COOKIE_ACK. + */ +sctp_xmit_t sctp_packet_transmit_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk) +{ + sctp_xmit_t retval; + int error = 0; + + SCTP_DEBUG_PRINTK("%s: packet:%p chunk:%p\n", __FUNCTION__, + packet, chunk); + + switch ((retval = (sctp_packet_append_chunk(packet, chunk)))) { + case SCTP_XMIT_PMTU_FULL: + if (!packet->has_cookie_echo) { + error = sctp_packet_transmit(packet); + if (error < 0) + chunk->skb->sk->sk_err = -error; + + /* If we have an empty packet, then we can NOT ever + * return PMTU_FULL. + */ + retval = sctp_packet_append_chunk(packet, chunk); + } + break; + + case SCTP_XMIT_RWND_FULL: + case SCTP_XMIT_OK: + case SCTP_XMIT_NAGLE_DELAY: + break; + }; + + return retval; +} + +/* Try to bundle a SACK with the packet. */ +static sctp_xmit_t sctp_packet_bundle_sack(struct sctp_packet *pkt, + struct sctp_chunk *chunk) +{ + sctp_xmit_t retval = SCTP_XMIT_OK; + + /* If sending DATA and haven't aleady bundled a SACK, try to + * bundle one in to the packet. + */ + if (sctp_chunk_is_data(chunk) && !pkt->has_sack && + !pkt->has_cookie_echo) { + struct sctp_association *asoc; + asoc = pkt->transport->asoc; + + if (asoc->a_rwnd > asoc->rwnd) { + struct sctp_chunk *sack; + asoc->a_rwnd = asoc->rwnd; + sack = sctp_make_sack(asoc); + if (sack) { + struct timer_list *timer; + retval = sctp_packet_append_chunk(pkt, sack); + asoc->peer.sack_needed = 0; + timer = &asoc->timers[SCTP_EVENT_TIMEOUT_SACK]; + if (timer_pending(timer) && del_timer(timer)) + sctp_association_put(asoc); + } + } + } + return retval; +} + +/* Append a chunk to the offered packet reporting back any inability to do + * so. + */ +sctp_xmit_t sctp_packet_append_chunk(struct sctp_packet *packet, + struct sctp_chunk *chunk) +{ + sctp_xmit_t retval = SCTP_XMIT_OK; + __u16 chunk_len = WORD_ROUND(ntohs(chunk->chunk_hdr->length)); + size_t psize; + size_t pmtu; + int too_big; + + SCTP_DEBUG_PRINTK("%s: packet:%p chunk:%p\n", __FUNCTION__, packet, + chunk); + + retval = sctp_packet_bundle_sack(packet, chunk); + psize = packet->size; + + if (retval != SCTP_XMIT_OK) + goto finish; + + pmtu = ((packet->transport->asoc) ? + (packet->transport->asoc->pmtu) : + (packet->transport->pmtu)); + + too_big = (psize + chunk_len > pmtu); + + /* Decide if we need to fragment or resubmit later. */ + if (too_big) { + /* Both control chunks and data chunks with TSNs are + * non-fragmentable. + */ + if (sctp_packet_empty(packet) || !sctp_chunk_is_data(chunk)) { + /* We no longer do re-fragmentation. + * Just fragment at the IP layer, if we + * actually hit this condition + */ + packet->ipfragok = 1; + goto append; + + } else { + retval = SCTP_XMIT_PMTU_FULL; + goto finish; + } + } + +append: + /* We believe that this chunk is OK to add to the packet (as + * long as we have the cwnd for it). + */ + + /* DATA is a special case since we must examine both rwnd and cwnd + * before we send DATA. + */ + if (sctp_chunk_is_data(chunk)) { + retval = sctp_packet_append_data(packet, chunk); + /* Disallow SACK bundling after DATA. */ + packet->has_sack = 1; + if (SCTP_XMIT_OK != retval) + goto finish; + } else if (SCTP_CID_COOKIE_ECHO == chunk->chunk_hdr->type) + packet->has_cookie_echo = 1; + else if (SCTP_CID_SACK == chunk->chunk_hdr->type) + packet->has_sack = 1; + + /* It is OK to send this chunk. */ + __skb_queue_tail(&packet->chunks, (struct sk_buff *)chunk); + packet->size += chunk_len; + chunk->transport = packet->transport; +finish: + return retval; +} + +/* All packets are sent to the network through this function from + * sctp_outq_tail(). + * + * The return value is a normal kernel error return value. + */ +int sctp_packet_transmit(struct sctp_packet *packet) +{ + struct sctp_transport *tp = packet->transport; + struct sctp_association *asoc = tp->asoc; + struct sctphdr *sh; + __u32 crc32; + struct sk_buff *nskb; + struct sctp_chunk *chunk; + struct sock *sk; + int err = 0; + int padding; /* How much padding do we need? */ + __u8 has_data = 0; + struct dst_entry *dst; + + SCTP_DEBUG_PRINTK("%s: packet:%p\n", __FUNCTION__, packet); + + /* Do NOT generate a chunkless packet. */ + chunk = (struct sctp_chunk *)skb_peek(&packet->chunks); + if (unlikely(!chunk)) + return err; + + /* Set up convenience variables... */ + sk = chunk->skb->sk; + + /* Allocate the new skb. */ + nskb = dev_alloc_skb(packet->size); + if (!nskb) + goto nomem; + + /* Make sure the outbound skb has enough header room reserved. */ + skb_reserve(nskb, packet->overhead); + + /* Set the owning socket so that we know where to get the + * destination IP address. + */ + skb_set_owner_w(nskb, sk); + + /* Build the SCTP header. */ + sh = (struct sctphdr *)skb_push(nskb, sizeof(struct sctphdr)); + sh->source = htons(packet->source_port); + sh->dest = htons(packet->destination_port); + + /* From 6.8 Adler-32 Checksum Calculation: + * After the packet is constructed (containing the SCTP common + * header and one or more control or DATA chunks), the + * transmitter shall: + * + * 1) Fill in the proper Verification Tag in the SCTP common + * header and initialize the checksum field to 0's. + */ + sh->vtag = htonl(packet->vtag); + sh->checksum = 0; + + /* 2) Calculate the Adler-32 checksum of the whole packet, + * including the SCTP common header and all the + * chunks. + * + * Note: Adler-32 is no longer applicable, as has been replaced + * by CRC32-C as described in <draft-ietf-tsvwg-sctpcsum-02.txt>. + */ + crc32 = sctp_start_cksum((__u8 *)sh, sizeof(struct sctphdr)); + + /** + * 6.10 Bundling + * + * An endpoint bundles chunks by simply including multiple + * chunks in one outbound SCTP packet. ... + */ + + /** + * 3.2 Chunk Field Descriptions + * + * The total length of a chunk (including Type, Length and + * Value fields) MUST be a multiple of 4 bytes. If the length + * of the chunk is not a multiple of 4 bytes, the sender MUST + * pad the chunk with all zero bytes and this padding is not + * included in the chunk length field. The sender should + * never pad with more than 3 bytes. + * + * [This whole comment explains WORD_ROUND() below.] + */ + SCTP_DEBUG_PRINTK("***sctp_transmit_packet***\n"); + while ((chunk = (struct sctp_chunk *)__skb_dequeue(&packet->chunks)) != NULL) { + if (sctp_chunk_is_data(chunk)) { + + if (!chunk->has_tsn) { + sctp_chunk_assign_ssn(chunk); + sctp_chunk_assign_tsn(chunk); + + /* 6.3.1 C4) When data is in flight and when allowed + * by rule C5, a new RTT measurement MUST be made each + * round trip. Furthermore, new RTT measurements + * SHOULD be made no more than once per round-trip + * for a given destination transport address. + */ + + if (!tp->rto_pending) { + chunk->rtt_in_progress = 1; + tp->rto_pending = 1; + } + } else + chunk->resent = 1; + + chunk->sent_at = jiffies; + has_data = 1; + } + + padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len; + if (padding) + memset(skb_put(chunk->skb, padding), 0, padding); + + crc32 = sctp_update_copy_cksum(skb_put(nskb, chunk->skb->len), + chunk->skb->data, + chunk->skb->len, crc32); + + SCTP_DEBUG_PRINTK("%s %p[%s] %s 0x%x, %s %d, %s %d, %s %d\n", + "*** Chunk", chunk, + sctp_cname(SCTP_ST_CHUNK( + chunk->chunk_hdr->type)), + chunk->has_tsn ? "TSN" : "No TSN", + chunk->has_tsn ? + ntohl(chunk->subh.data_hdr->tsn) : 0, + "length", ntohs(chunk->chunk_hdr->length), + "chunk->skb->len", chunk->skb->len, + "rtt_in_progress", chunk->rtt_in_progress); + + /* + * If this is a control chunk, this is our last + * reference. Free data chunks after they've been + * acknowledged or have failed. + */ + if (!sctp_chunk_is_data(chunk)) + sctp_chunk_free(chunk); + } + + /* Perform final transformation on checksum. */ + crc32 = sctp_end_cksum(crc32); + + /* 3) Put the resultant value into the checksum field in the + * common header, and leave the rest of the bits unchanged. + */ + sh->checksum = htonl(crc32); + + /* IP layer ECN support + * From RFC 2481 + * "The ECN-Capable Transport (ECT) bit would be set by the + * data sender to indicate that the end-points of the + * transport protocol are ECN-capable." + * + * Now setting the ECT bit all the time, as it should not cause + * any problems protocol-wise even if our peer ignores it. + * + * Note: The works for IPv6 layer checks this bit too later + * in transmission. See IP6_ECN_flow_xmit(). + */ + INET_ECN_xmit(nskb->sk); + + /* Set up the IP options. */ + /* BUG: not implemented + * For v4 this all lives somewhere in sk->sk_opt... + */ + + /* Dump that on IP! */ + if (asoc && asoc->peer.last_sent_to != tp) { + /* Considering the multiple CPU scenario, this is a + * "correcter" place for last_sent_to. --xguo + */ + asoc->peer.last_sent_to = tp; + } + + if (has_data) { + struct timer_list *timer; + unsigned long timeout; + + tp->last_time_used = jiffies; + + /* Restart the AUTOCLOSE timer when sending data. */ + if (sctp_state(asoc, ESTABLISHED) && asoc->autoclose) { + timer = &asoc->timers[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; + timeout = asoc->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE]; + + if (!mod_timer(timer, jiffies + timeout)) + sctp_association_hold(asoc); + } + } + + dst = tp->dst; + /* The 'obsolete' field of dst is set to 2 when a dst is freed. */ + if (!dst || (dst->obsolete > 1)) { + dst_release(dst); + sctp_transport_route(tp, NULL, sctp_sk(sk)); + sctp_assoc_sync_pmtu(asoc); + } + + nskb->dst = dst_clone(tp->dst); + if (!nskb->dst) + goto no_route; + + SCTP_DEBUG_PRINTK("***sctp_transmit_packet*** skb len %d\n", + nskb->len); + + (*tp->af_specific->sctp_xmit)(nskb, tp, packet->ipfragok); + +out: + packet->size = packet->overhead; + return err; +no_route: + kfree_skb(nskb); + IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); + + /* FIXME: Returning the 'err' will effect all the associations + * associated with a socket, although only one of the paths of the + * association is unreachable. + * The real failure of a transport or association can be passed on + * to the user via notifications. So setting this error may not be + * required. + */ + /* err = -EHOSTUNREACH; */ +err: + /* Control chunks are unreliable so just drop them. DATA chunks + * will get resent or dropped later. + */ + + while ((chunk = (struct sctp_chunk *)__skb_dequeue(&packet->chunks)) != NULL) { + if (!sctp_chunk_is_data(chunk)) + sctp_chunk_free(chunk); + } + goto out; +nomem: + err = -ENOMEM; + goto err; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* This private function handles the specifics of appending DATA chunks. */ +static sctp_xmit_t sctp_packet_append_data(struct sctp_packet *packet, + struct sctp_chunk *chunk) +{ + sctp_xmit_t retval = SCTP_XMIT_OK; + size_t datasize, rwnd, inflight; + struct sctp_transport *transport = packet->transport; + __u32 max_burst_bytes; + struct sctp_association *asoc = transport->asoc; + struct sctp_sock *sp = sctp_sk(asoc->base.sk); + struct sctp_outq *q = &asoc->outqueue; + + /* RFC 2960 6.1 Transmission of DATA Chunks + * + * A) At any given time, the data sender MUST NOT transmit new data to + * any destination transport address if its peer's rwnd indicates + * that the peer has no buffer space (i.e. rwnd is 0, see Section + * 6.2.1). However, regardless of the value of rwnd (including if it + * is 0), the data sender can always have one DATA chunk in flight to + * the receiver if allowed by cwnd (see rule B below). This rule + * allows the sender to probe for a change in rwnd that the sender + * missed due to the SACK having been lost in transit from the data + * receiver to the data sender. + */ + + rwnd = asoc->peer.rwnd; + inflight = asoc->outqueue.outstanding_bytes; + + datasize = sctp_data_size(chunk); + + if (datasize > rwnd) { + if (inflight > 0) { + /* We have (at least) one data chunk in flight, + * so we can't fall back to rule 6.1 B). + */ + retval = SCTP_XMIT_RWND_FULL; + goto finish; + } + } + + /* 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 + */ + max_burst_bytes = asoc->max_burst * asoc->pmtu; + if ((transport->flight_size + max_burst_bytes) < transport->cwnd) { + transport->cwnd = transport->flight_size + max_burst_bytes; + SCTP_DEBUG_PRINTK("%s: cwnd limited by max_burst: " + "transport: %p, cwnd: %d, " + "ssthresh: %d, flight_size: %d, " + "pba: %d\n", + __FUNCTION__, transport, + transport->cwnd, + transport->ssthresh, + transport->flight_size, + transport->partial_bytes_acked); + } + + /* RFC 2960 6.1 Transmission of DATA Chunks + * + * B) At any given time, the sender MUST NOT transmit new data + * to a given transport address if it has cwnd or more bytes + * of data outstanding to that transport address. + */ + /* RFC 7.2.4 & the Implementers Guide 2.8. + * + * 3) ... + * When a Fast Retransmit is being performed the sender SHOULD + * ignore the value of cwnd and SHOULD NOT delay retransmission. + */ + if (!chunk->fast_retransmit) + if (transport->flight_size >= transport->cwnd) { + retval = SCTP_XMIT_RWND_FULL; + goto finish; + } + + /* Nagle's algorithm to solve small-packet problem: + * Inhibit the sending of new chunks when new outgoing data arrives + * if any previously transmitted data on the connection remains + * unacknowledged. + */ + if (!sp->nodelay && sctp_packet_empty(packet) && + q->outstanding_bytes && sctp_state(asoc, ESTABLISHED)) { + unsigned len = datasize + q->out_qlen; + + /* Check whether this chunk and all the rest of pending + * data will fit or delay in hopes of bundling a full + * sized packet. + */ + if (len < asoc->pmtu - packet->overhead) { + retval = SCTP_XMIT_NAGLE_DELAY; + goto finish; + } + } + + /* Keep track of how many bytes are in flight over this transport. */ + transport->flight_size += datasize; + + /* Keep track of how many bytes are in flight to the receiver. */ + asoc->outqueue.outstanding_bytes += datasize; + + /* Update our view of the receiver's rwnd. */ + if (datasize < rwnd) + rwnd -= datasize; + else + rwnd = 0; + + asoc->peer.rwnd = rwnd; + /* Has been accepted for transmission. */ + if (!asoc->peer.prsctp_capable) + chunk->msg->can_abandon = 0; + +finish: + return retval; +} diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c new file mode 100644 index 000000000000..1b2d4adc4ddb --- /dev/null +++ b/net/sctp/outqueue.c @@ -0,0 +1,1734 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001-2003 Intel Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions implement the sctp_outq class. The outqueue handles + * bundling and queueing of outgoing SCTP chunks. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Perry Melange <pmelange@null.cc.uic.edu> + * Xingang Guo <xingang.guo@intel.com> + * Hui Huang <hui.huang@nokia.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Jon Grimm <jgrimm@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/list.h> /* For struct list_head */ +#include <linux/socket.h> +#include <linux/ip.h> +#include <net/sock.h> /* For skb_set_owner_w */ + +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Declare internal functions here. */ +static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn); +static void sctp_check_transmitted(struct sctp_outq *q, + struct list_head *transmitted_queue, + struct sctp_transport *transport, + struct sctp_sackhdr *sack, + __u32 highest_new_tsn); + +static void sctp_mark_missing(struct sctp_outq *q, + struct list_head *transmitted_queue, + struct sctp_transport *transport, + __u32 highest_new_tsn, + int count_of_newacks); + +static void sctp_generate_fwdtsn(struct sctp_outq *q, __u32 sack_ctsn); + +/* Add data to the front of the queue. */ +static inline void sctp_outq_head_data(struct sctp_outq *q, + struct sctp_chunk *ch) +{ + __skb_queue_head(&q->out, (struct sk_buff *)ch); + q->out_qlen += ch->skb->len; + return; +} + +/* Take data from the front of the queue. */ +static inline struct sctp_chunk *sctp_outq_dequeue_data(struct sctp_outq *q) +{ + struct sctp_chunk *ch; + ch = (struct sctp_chunk *)__skb_dequeue(&q->out); + if (ch) + q->out_qlen -= ch->skb->len; + return ch; +} +/* Add data chunk to the end of the queue. */ +static inline void sctp_outq_tail_data(struct sctp_outq *q, + struct sctp_chunk *ch) +{ + __skb_queue_tail(&q->out, (struct sk_buff *)ch); + q->out_qlen += ch->skb->len; + return; +} + +/* + * SFR-CACC algorithm: + * D) If count_of_newacks is greater than or equal to 2 + * and t was not sent to the current primary then the + * sender MUST NOT increment missing report count for t. + */ +static inline int sctp_cacc_skip_3_1_d(struct sctp_transport *primary, + struct sctp_transport *transport, + int count_of_newacks) +{ + if (count_of_newacks >=2 && transport != primary) + return 1; + return 0; +} + +/* + * SFR-CACC algorithm: + * F) If count_of_newacks is less than 2, let d be the + * destination to which t was sent. If cacc_saw_newack + * is 0 for destination d, then the sender MUST NOT + * increment missing report count for t. + */ +static inline int sctp_cacc_skip_3_1_f(struct sctp_transport *transport, + int count_of_newacks) +{ + if (count_of_newacks < 2 && !transport->cacc.cacc_saw_newack) + return 1; + return 0; +} + +/* + * SFR-CACC algorithm: + * 3.1) If CYCLING_CHANGEOVER is 0, the sender SHOULD + * execute steps C, D, F. + * + * C has been implemented in sctp_outq_sack + */ +static inline int sctp_cacc_skip_3_1(struct sctp_transport *primary, + struct sctp_transport *transport, + int count_of_newacks) +{ + if (!primary->cacc.cycling_changeover) { + if (sctp_cacc_skip_3_1_d(primary, transport, count_of_newacks)) + return 1; + if (sctp_cacc_skip_3_1_f(transport, count_of_newacks)) + return 1; + return 0; + } + return 0; +} + +/* + * SFR-CACC algorithm: + * 3.2) Else if CYCLING_CHANGEOVER is 1, and t is less + * than next_tsn_at_change of the current primary, then + * the sender MUST NOT increment missing report count + * for t. + */ +static inline int sctp_cacc_skip_3_2(struct sctp_transport *primary, __u32 tsn) +{ + if (primary->cacc.cycling_changeover && + TSN_lt(tsn, primary->cacc.next_tsn_at_change)) + return 1; + return 0; +} + +/* + * SFR-CACC algorithm: + * 3) If the missing report count for TSN t is to be + * incremented according to [RFC2960] and + * [SCTP_STEWART-2002], and CHANGEOVER_ACTIVE is set, + * then the sender MUST futher execute steps 3.1 and + * 3.2 to determine if the missing report count for + * TSN t SHOULD NOT be incremented. + * + * 3.3) If 3.1 and 3.2 do not dictate that the missing + * report count for t should not be incremented, then + * the sender SOULD increment missing report count for + * t (according to [RFC2960] and [SCTP_STEWART_2002]). + */ +static inline int sctp_cacc_skip(struct sctp_transport *primary, + struct sctp_transport *transport, + int count_of_newacks, + __u32 tsn) +{ + if (primary->cacc.changeover_active && + (sctp_cacc_skip_3_1(primary, transport, count_of_newacks) + || sctp_cacc_skip_3_2(primary, tsn))) + return 1; + return 0; +} + +/* Initialize an existing sctp_outq. This does the boring stuff. + * You still need to define handlers if you really want to DO + * something with this structure... + */ +void sctp_outq_init(struct sctp_association *asoc, struct sctp_outq *q) +{ + q->asoc = asoc; + skb_queue_head_init(&q->out); + skb_queue_head_init(&q->control); + INIT_LIST_HEAD(&q->retransmit); + INIT_LIST_HEAD(&q->sacked); + INIT_LIST_HEAD(&q->abandoned); + + q->outstanding_bytes = 0; + q->empty = 1; + q->cork = 0; + + q->malloced = 0; + q->out_qlen = 0; +} + +/* Free the outqueue structure and any related pending chunks. + */ +void sctp_outq_teardown(struct sctp_outq *q) +{ + struct sctp_transport *transport; + struct list_head *lchunk, *pos, *temp; + struct sctp_chunk *chunk; + + /* Throw away unacknowledged chunks. */ + list_for_each(pos, &q->asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + while ((lchunk = sctp_list_dequeue(&transport->transmitted)) != NULL) { + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + /* Mark as part of a failed message. */ + sctp_chunk_fail(chunk, q->error); + sctp_chunk_free(chunk); + } + } + + /* Throw away chunks that have been gap ACKed. */ + list_for_each_safe(lchunk, temp, &q->sacked) { + list_del_init(lchunk); + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + sctp_chunk_fail(chunk, q->error); + sctp_chunk_free(chunk); + } + + /* Throw away any chunks in the retransmit queue. */ + list_for_each_safe(lchunk, temp, &q->retransmit) { + list_del_init(lchunk); + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + sctp_chunk_fail(chunk, q->error); + sctp_chunk_free(chunk); + } + + /* Throw away any chunks that are in the abandoned queue. */ + list_for_each_safe(lchunk, temp, &q->abandoned) { + list_del_init(lchunk); + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + sctp_chunk_fail(chunk, q->error); + sctp_chunk_free(chunk); + } + + /* Throw away any leftover data chunks. */ + while ((chunk = sctp_outq_dequeue_data(q)) != NULL) { + + /* Mark as send failure. */ + sctp_chunk_fail(chunk, q->error); + sctp_chunk_free(chunk); + } + + q->error = 0; + + /* Throw away any leftover control chunks. */ + while ((chunk = (struct sctp_chunk *) skb_dequeue(&q->control)) != NULL) + sctp_chunk_free(chunk); +} + +/* Free the outqueue structure and any related pending chunks. */ +void sctp_outq_free(struct sctp_outq *q) +{ + /* Throw away leftover chunks. */ + sctp_outq_teardown(q); + + /* If we were kmalloc()'d, free the memory. */ + if (q->malloced) + kfree(q); +} + +/* Put a new chunk in an sctp_outq. */ +int sctp_outq_tail(struct sctp_outq *q, struct sctp_chunk *chunk) +{ + int error = 0; + + SCTP_DEBUG_PRINTK("sctp_outq_tail(%p, %p[%s])\n", + q, chunk, chunk && chunk->chunk_hdr ? + sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)) + : "Illegal Chunk"); + + /* If it is data, queue it up, otherwise, send it + * immediately. + */ + if (SCTP_CID_DATA == chunk->chunk_hdr->type) { + /* Is it OK to queue data chunks? */ + /* From 9. Termination of Association + * + * When either endpoint performs a shutdown, the + * association on each peer will stop accepting new + * data from its user and only deliver data in queue + * at the time of sending or receiving the SHUTDOWN + * chunk. + */ + switch (q->asoc->state) { + case SCTP_STATE_EMPTY: + case SCTP_STATE_CLOSED: + case SCTP_STATE_SHUTDOWN_PENDING: + case SCTP_STATE_SHUTDOWN_SENT: + case SCTP_STATE_SHUTDOWN_RECEIVED: + case SCTP_STATE_SHUTDOWN_ACK_SENT: + /* Cannot send after transport endpoint shutdown */ + error = -ESHUTDOWN; + break; + + default: + SCTP_DEBUG_PRINTK("outqueueing (%p, %p[%s])\n", + q, chunk, chunk && chunk->chunk_hdr ? + sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)) + : "Illegal Chunk"); + + sctp_outq_tail_data(q, chunk); + if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) + SCTP_INC_STATS(SCTP_MIB_OUTUNORDERCHUNKS); + else + SCTP_INC_STATS(SCTP_MIB_OUTORDERCHUNKS); + q->empty = 0; + break; + }; + } else { + __skb_queue_tail(&q->control, (struct sk_buff *) chunk); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + } + + if (error < 0) + return error; + + if (!q->cork) + error = sctp_outq_flush(q, 0); + + return error; +} + +/* Insert a chunk into the sorted list based on the TSNs. The retransmit list + * and the abandoned list are in ascending order. + */ +static void sctp_insert_list(struct list_head *head, struct list_head *new) +{ + struct list_head *pos; + struct sctp_chunk *nchunk, *lchunk; + __u32 ntsn, ltsn; + int done = 0; + + nchunk = list_entry(new, struct sctp_chunk, transmitted_list); + ntsn = ntohl(nchunk->subh.data_hdr->tsn); + + list_for_each(pos, head) { + lchunk = list_entry(pos, struct sctp_chunk, transmitted_list); + ltsn = ntohl(lchunk->subh.data_hdr->tsn); + if (TSN_lt(ntsn, ltsn)) { + list_add(new, pos->prev); + done = 1; + break; + } + } + if (!done) + list_add_tail(new, head); +} + +/* Mark all the eligible packets on a transport for retransmission. */ +void sctp_retransmit_mark(struct sctp_outq *q, + struct sctp_transport *transport, + __u8 fast_retransmit) +{ + struct list_head *lchunk, *ltemp; + struct sctp_chunk *chunk; + + /* Walk through the specified transmitted queue. */ + list_for_each_safe(lchunk, ltemp, &transport->transmitted) { + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + + /* If the chunk is abandoned, move it to abandoned list. */ + if (sctp_chunk_abandoned(chunk)) { + list_del_init(lchunk); + sctp_insert_list(&q->abandoned, lchunk); + continue; + } + + /* If we are doing retransmission due to a fast retransmit, + * only the chunk's that are marked for fast retransmit + * should be added to the retransmit queue. If we are doing + * retransmission due to a timeout or pmtu discovery, only the + * chunks that are not yet acked should be added to the + * retransmit queue. + */ + if ((fast_retransmit && chunk->fast_retransmit) || + (!fast_retransmit && !chunk->tsn_gap_acked)) { + /* RFC 2960 6.2.1 Processing a Received SACK + * + * C) Any time a DATA chunk is marked for + * retransmission (via either T3-rtx timer expiration + * (Section 6.3.3) or via fast retransmit + * (Section 7.2.4)), add the data size of those + * chunks to the rwnd. + */ + q->asoc->peer.rwnd += sctp_data_size(chunk); + q->outstanding_bytes -= sctp_data_size(chunk); + transport->flight_size -= sctp_data_size(chunk); + + /* sctpimpguide-05 Section 2.8.2 + * M5) If a T3-rtx timer expires, the + * 'TSN.Missing.Report' of all affected TSNs is set + * to 0. + */ + chunk->tsn_missing_report = 0; + + /* If a chunk that is being used for RTT measurement + * has to be retransmitted, we cannot use this chunk + * anymore for RTT measurements. Reset rto_pending so + * that a new RTT measurement is started when a new + * data chunk is sent. + */ + if (chunk->rtt_in_progress) { + chunk->rtt_in_progress = 0; + transport->rto_pending = 0; + } + + /* Move the chunk to the retransmit queue. The chunks + * on the retransmit queue are always kept in order. + */ + list_del_init(lchunk); + sctp_insert_list(&q->retransmit, lchunk); + } + } + + SCTP_DEBUG_PRINTK("%s: transport: %p, fast_retransmit: %d, " + "cwnd: %d, ssthresh: %d, flight_size: %d, " + "pba: %d\n", __FUNCTION__, + transport, fast_retransmit, + transport->cwnd, transport->ssthresh, + transport->flight_size, + transport->partial_bytes_acked); + +} + +/* Mark all the eligible packets on a transport for retransmission and force + * one packet out. + */ +void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport, + sctp_retransmit_reason_t reason) +{ + int error = 0; + __u8 fast_retransmit = 0; + + switch(reason) { + case SCTP_RTXR_T3_RTX: + sctp_transport_lower_cwnd(transport, SCTP_LOWER_CWND_T3_RTX); + /* Update the retran path if the T3-rtx timer has expired for + * the current retran path. + */ + if (transport == transport->asoc->peer.retran_path) + sctp_assoc_update_retran_path(transport->asoc); + break; + case SCTP_RTXR_FAST_RTX: + sctp_transport_lower_cwnd(transport, SCTP_LOWER_CWND_FAST_RTX); + fast_retransmit = 1; + break; + case SCTP_RTXR_PMTUD: + default: + break; + } + + sctp_retransmit_mark(q, transport, fast_retransmit); + + /* PR-SCTP A5) Any time the T3-rtx timer expires, on any destination, + * the sender SHOULD try to advance the "Advanced.Peer.Ack.Point" by + * following the procedures outlined in C1 - C5. + */ + sctp_generate_fwdtsn(q, q->asoc->ctsn_ack_point); + + error = sctp_outq_flush(q, /* rtx_timeout */ 1); + + if (error) + q->asoc->base.sk->sk_err = -error; +} + +/* + * Transmit DATA chunks on the retransmit queue. Upon return from + * sctp_outq_flush_rtx() the packet 'pkt' may contain chunks which + * need to be transmitted by the caller. + * We assume that pkt->transport has already been set. + * + * The return value is a normal kernel error return value. + */ +static int sctp_outq_flush_rtx(struct sctp_outq *q, struct sctp_packet *pkt, + int rtx_timeout, int *start_timer) +{ + struct list_head *lqueue; + struct list_head *lchunk, *lchunk1; + struct sctp_transport *transport = pkt->transport; + sctp_xmit_t status; + struct sctp_chunk *chunk, *chunk1; + struct sctp_association *asoc; + int error = 0; + + asoc = q->asoc; + lqueue = &q->retransmit; + + /* RFC 2960 6.3.3 Handle T3-rtx Expiration + * + * E3) Determine how many of the earliest (i.e., lowest TSN) + * outstanding DATA chunks for the address for which the + * T3-rtx has expired will fit into a single packet, subject + * to the MTU constraint for the path corresponding to the + * destination transport address to which the retransmission + * is being sent (this may be different from the address for + * which the timer expires [see Section 6.4]). Call this value + * K. Bundle and retransmit those K DATA chunks in a single + * packet to the destination endpoint. + * + * [Just to be painfully clear, if we are retransmitting + * because a timeout just happened, we should send only ONE + * packet of retransmitted data.] + */ + lchunk = sctp_list_dequeue(lqueue); + + while (lchunk) { + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + + /* Make sure that Gap Acked TSNs are not retransmitted. A + * simple approach is just to move such TSNs out of the + * way and into a 'transmitted' queue and skip to the + * next chunk. + */ + if (chunk->tsn_gap_acked) { + list_add_tail(lchunk, &transport->transmitted); + lchunk = sctp_list_dequeue(lqueue); + continue; + } + + /* Attempt to append this chunk to the packet. */ + status = sctp_packet_append_chunk(pkt, chunk); + + switch (status) { + case SCTP_XMIT_PMTU_FULL: + /* Send this packet. */ + if ((error = sctp_packet_transmit(pkt)) == 0) + *start_timer = 1; + + /* If we are retransmitting, we should only + * send a single packet. + */ + if (rtx_timeout) { + list_add(lchunk, lqueue); + lchunk = NULL; + } + + /* Bundle lchunk in the next round. */ + break; + + case SCTP_XMIT_RWND_FULL: + /* Send this packet. */ + if ((error = sctp_packet_transmit(pkt)) == 0) + *start_timer = 1; + + /* Stop sending DATA as there is no more room + * at the receiver. + */ + list_add(lchunk, lqueue); + lchunk = NULL; + break; + + case SCTP_XMIT_NAGLE_DELAY: + /* Send this packet. */ + if ((error = sctp_packet_transmit(pkt)) == 0) + *start_timer = 1; + + /* Stop sending DATA because of nagle delay. */ + list_add(lchunk, lqueue); + lchunk = NULL; + break; + + default: + /* The append was successful, so add this chunk to + * the transmitted list. + */ + list_add_tail(lchunk, &transport->transmitted); + + /* Mark the chunk as ineligible for fast retransmit + * after it is retransmitted. + */ + chunk->fast_retransmit = 0; + + *start_timer = 1; + q->empty = 0; + + /* Retrieve a new chunk to bundle. */ + lchunk = sctp_list_dequeue(lqueue); + break; + }; + + /* If we are here due to a retransmit timeout or a fast + * retransmit and if there are any chunks left in the retransmit + * queue that could not fit in the PMTU sized packet, they need * to be marked as ineligible for a subsequent fast retransmit. + */ + if (rtx_timeout && !lchunk) { + list_for_each(lchunk1, lqueue) { + chunk1 = list_entry(lchunk1, struct sctp_chunk, + transmitted_list); + chunk1->fast_retransmit = 0; + } + } + } + + return error; +} + +/* Cork the outqueue so queued chunks are really queued. */ +int sctp_outq_uncork(struct sctp_outq *q) +{ + int error = 0; + if (q->cork) { + q->cork = 0; + error = sctp_outq_flush(q, 0); + } + return error; +} + +/* + * Try to flush an outqueue. + * + * Description: Send everything in q which we legally can, subject to + * congestion limitations. + * * Note: This function can be called from multiple contexts so appropriate + * locking concerns must be made. Today we use the sock lock to protect + * this function. + */ +int sctp_outq_flush(struct sctp_outq *q, int rtx_timeout) +{ + struct sctp_packet *packet; + struct sctp_packet singleton; + struct sctp_association *asoc = q->asoc; + __u16 sport = asoc->base.bind_addr.port; + __u16 dport = asoc->peer.port; + __u32 vtag = asoc->peer.i.init_tag; + struct sk_buff_head *queue; + struct sctp_transport *transport = NULL; + struct sctp_transport *new_transport; + struct sctp_chunk *chunk; + sctp_xmit_t status; + int error = 0; + int start_timer = 0; + + /* These transports have chunks to send. */ + struct list_head transport_list; + struct list_head *ltransport; + + INIT_LIST_HEAD(&transport_list); + packet = NULL; + + /* + * 6.10 Bundling + * ... + * When bundling control chunks with DATA chunks, an + * endpoint MUST place control chunks first in the outbound + * SCTP packet. The transmitter MUST transmit DATA chunks + * within a SCTP packet in increasing order of TSN. + * ... + */ + + queue = &q->control; + while ((chunk = (struct sctp_chunk *)skb_dequeue(queue)) != NULL) { + /* Pick the right transport to use. */ + new_transport = chunk->transport; + + if (!new_transport) { + new_transport = asoc->peer.active_path; + } else if (!new_transport->active) { + /* If the chunk is Heartbeat or Heartbeat Ack, + * send it to chunk->transport, even if it's + * inactive. + * + * 3.3.6 Heartbeat Acknowledgement: + * ... + * A HEARTBEAT ACK is always sent to the source IP + * address of the IP datagram containing the + * HEARTBEAT chunk to which this ack is responding. + * ... + */ + if (chunk->chunk_hdr->type != SCTP_CID_HEARTBEAT && + chunk->chunk_hdr->type != SCTP_CID_HEARTBEAT_ACK) + new_transport = asoc->peer.active_path; + } + + /* Are we switching transports? + * Take care of transport locks. + */ + if (new_transport != transport) { + transport = new_transport; + if (list_empty(&transport->send_ready)) { + list_add_tail(&transport->send_ready, + &transport_list); + } + packet = &transport->packet; + sctp_packet_config(packet, vtag, + asoc->peer.ecn_capable); + } + + switch (chunk->chunk_hdr->type) { + /* + * 6.10 Bundling + * ... + * An endpoint MUST NOT bundle INIT, INIT ACK or SHUTDOWN + * COMPLETE with any other chunks. [Send them immediately.] + */ + case SCTP_CID_INIT: + case SCTP_CID_INIT_ACK: + case SCTP_CID_SHUTDOWN_COMPLETE: + sctp_packet_init(&singleton, transport, sport, dport); + sctp_packet_config(&singleton, vtag, 0); + sctp_packet_append_chunk(&singleton, chunk); + error = sctp_packet_transmit(&singleton); + if (error < 0) + return error; + break; + + case SCTP_CID_ABORT: + case SCTP_CID_SACK: + case SCTP_CID_HEARTBEAT: + case SCTP_CID_HEARTBEAT_ACK: + case SCTP_CID_SHUTDOWN: + case SCTP_CID_SHUTDOWN_ACK: + case SCTP_CID_ERROR: + case SCTP_CID_COOKIE_ECHO: + case SCTP_CID_COOKIE_ACK: + case SCTP_CID_ECN_ECNE: + case SCTP_CID_ECN_CWR: + case SCTP_CID_ASCONF: + case SCTP_CID_ASCONF_ACK: + case SCTP_CID_FWD_TSN: + sctp_packet_transmit_chunk(packet, chunk); + break; + + default: + /* We built a chunk with an illegal type! */ + BUG(); + }; + } + + /* Is it OK to send data chunks? */ + switch (asoc->state) { + case SCTP_STATE_COOKIE_ECHOED: + /* Only allow bundling when this packet has a COOKIE-ECHO + * chunk. + */ + if (!packet || !packet->has_cookie_echo) + break; + + /* fallthru */ + case SCTP_STATE_ESTABLISHED: + case SCTP_STATE_SHUTDOWN_PENDING: + case SCTP_STATE_SHUTDOWN_RECEIVED: + /* + * RFC 2960 6.1 Transmission of DATA Chunks + * + * C) When the time comes for the sender to transmit, + * before sending new DATA chunks, the sender MUST + * first transmit any outstanding DATA chunks which + * are marked for retransmission (limited by the + * current cwnd). + */ + if (!list_empty(&q->retransmit)) { + if (transport == asoc->peer.retran_path) + goto retran; + + /* Switch transports & prepare the packet. */ + + transport = asoc->peer.retran_path; + + if (list_empty(&transport->send_ready)) { + list_add_tail(&transport->send_ready, + &transport_list); + } + + packet = &transport->packet; + sctp_packet_config(packet, vtag, + asoc->peer.ecn_capable); + retran: + error = sctp_outq_flush_rtx(q, packet, + rtx_timeout, &start_timer); + + if (start_timer) + sctp_transport_reset_timers(transport); + + /* This can happen on COOKIE-ECHO resend. Only + * one chunk can get bundled with a COOKIE-ECHO. + */ + if (packet->has_cookie_echo) + goto sctp_flush_out; + + /* Don't send new data if there is still data + * waiting to retransmit. + */ + if (!list_empty(&q->retransmit)) + goto sctp_flush_out; + } + + /* Finally, transmit new packets. */ + start_timer = 0; + queue = &q->out; + + while ((chunk = sctp_outq_dequeue_data(q)) != NULL) { + /* RFC 2960 6.5 Every DATA chunk MUST carry a valid + * stream identifier. + */ + if (chunk->sinfo.sinfo_stream >= + asoc->c.sinit_num_ostreams) { + + /* Mark as failed send. */ + sctp_chunk_fail(chunk, SCTP_ERROR_INV_STRM); + sctp_chunk_free(chunk); + continue; + } + + /* Has this chunk expired? */ + if (sctp_chunk_abandoned(chunk)) { + sctp_chunk_fail(chunk, 0); + sctp_chunk_free(chunk); + continue; + } + + /* If there is a specified transport, use it. + * Otherwise, we want to use the active path. + */ + new_transport = chunk->transport; + if (!new_transport || !new_transport->active) + new_transport = asoc->peer.active_path; + + /* Change packets if necessary. */ + if (new_transport != transport) { + transport = new_transport; + + /* Schedule to have this transport's + * packet flushed. + */ + if (list_empty(&transport->send_ready)) { + list_add_tail(&transport->send_ready, + &transport_list); + } + + packet = &transport->packet; + sctp_packet_config(packet, vtag, + asoc->peer.ecn_capable); + } + + SCTP_DEBUG_PRINTK("sctp_outq_flush(%p, %p[%s]), ", + q, chunk, + chunk && chunk->chunk_hdr ? + sctp_cname(SCTP_ST_CHUNK( + chunk->chunk_hdr->type)) + : "Illegal Chunk"); + + SCTP_DEBUG_PRINTK("TX TSN 0x%x skb->head " + "%p skb->users %d.\n", + ntohl(chunk->subh.data_hdr->tsn), + chunk->skb ?chunk->skb->head : NULL, + chunk->skb ? + atomic_read(&chunk->skb->users) : -1); + + /* Add the chunk to the packet. */ + status = sctp_packet_transmit_chunk(packet, chunk); + + switch (status) { + case SCTP_XMIT_PMTU_FULL: + case SCTP_XMIT_RWND_FULL: + case SCTP_XMIT_NAGLE_DELAY: + /* We could not append this chunk, so put + * the chunk back on the output queue. + */ + SCTP_DEBUG_PRINTK("sctp_outq_flush: could " + "not transmit TSN: 0x%x, status: %d\n", + ntohl(chunk->subh.data_hdr->tsn), + status); + sctp_outq_head_data(q, chunk); + goto sctp_flush_out; + break; + + case SCTP_XMIT_OK: + break; + + default: + BUG(); + } + + /* BUG: We assume that the sctp_packet_transmit() + * call below will succeed all the time and add the + * chunk to the transmitted list and restart the + * timers. + * It is possible that the call can fail under OOM + * conditions. + * + * Is this really a problem? Won't this behave + * like a lost TSN? + */ + list_add_tail(&chunk->transmitted_list, + &transport->transmitted); + + sctp_transport_reset_timers(transport); + + q->empty = 0; + + /* Only let one DATA chunk get bundled with a + * COOKIE-ECHO chunk. + */ + if (packet->has_cookie_echo) + goto sctp_flush_out; + } + break; + + default: + /* Do nothing. */ + break; + } + +sctp_flush_out: + + /* Before returning, examine all the transports touched in + * this call. Right now, we bluntly force clear all the + * transports. Things might change after we implement Nagle. + * But such an examination is still required. + * + * --xguo + */ + while ((ltransport = sctp_list_dequeue(&transport_list)) != NULL ) { + struct sctp_transport *t = list_entry(ltransport, + struct sctp_transport, + send_ready); + packet = &t->packet; + if (!sctp_packet_empty(packet)) + error = sctp_packet_transmit(packet); + } + + return error; +} + +/* Update unack_data based on the incoming SACK chunk */ +static void sctp_sack_update_unack_data(struct sctp_association *assoc, + struct sctp_sackhdr *sack) +{ + sctp_sack_variable_t *frags; + __u16 unack_data; + int i; + + unack_data = assoc->next_tsn - assoc->ctsn_ack_point - 1; + + frags = sack->variable; + for (i = 0; i < ntohs(sack->num_gap_ack_blocks); i++) { + unack_data -= ((ntohs(frags[i].gab.end) - + ntohs(frags[i].gab.start) + 1)); + } + + assoc->unack_data = unack_data; +} + +/* Return the highest new tsn that is acknowledged by the given SACK chunk. */ +static __u32 sctp_highest_new_tsn(struct sctp_sackhdr *sack, + struct sctp_association *asoc) +{ + struct list_head *ltransport, *lchunk; + struct sctp_transport *transport; + struct sctp_chunk *chunk; + __u32 highest_new_tsn, tsn; + struct list_head *transport_list = &asoc->peer.transport_addr_list; + + highest_new_tsn = ntohl(sack->cum_tsn_ack); + + list_for_each(ltransport, transport_list) { + transport = list_entry(ltransport, struct sctp_transport, + transports); + list_for_each(lchunk, &transport->transmitted) { + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + tsn = ntohl(chunk->subh.data_hdr->tsn); + + if (!chunk->tsn_gap_acked && + TSN_lt(highest_new_tsn, tsn) && + sctp_acked(sack, tsn)) + highest_new_tsn = tsn; + } + } + + return highest_new_tsn; +} + +/* This is where we REALLY process a SACK. + * + * Process the SACK against the outqueue. Mostly, this just frees + * things off the transmitted queue. + */ +int sctp_outq_sack(struct sctp_outq *q, struct sctp_sackhdr *sack) +{ + struct sctp_association *asoc = q->asoc; + struct sctp_transport *transport; + struct sctp_chunk *tchunk = NULL; + struct list_head *lchunk, *transport_list, *pos, *temp; + sctp_sack_variable_t *frags = sack->variable; + __u32 sack_ctsn, ctsn, tsn; + __u32 highest_tsn, highest_new_tsn; + __u32 sack_a_rwnd; + unsigned outstanding; + struct sctp_transport *primary = asoc->peer.primary_path; + int count_of_newacks = 0; + + /* Grab the association's destination address list. */ + transport_list = &asoc->peer.transport_addr_list; + + sack_ctsn = ntohl(sack->cum_tsn_ack); + + /* + * SFR-CACC algorithm: + * On receipt of a SACK the sender SHOULD execute the + * following statements. + * + * 1) If the cumulative ack in the SACK passes next tsn_at_change + * on the current primary, the CHANGEOVER_ACTIVE flag SHOULD be + * cleared. The CYCLING_CHANGEOVER flag SHOULD also be cleared for + * all destinations. + */ + if (TSN_lte(primary->cacc.next_tsn_at_change, sack_ctsn)) { + primary->cacc.changeover_active = 0; + list_for_each(pos, transport_list) { + transport = list_entry(pos, struct sctp_transport, + transports); + transport->cacc.cycling_changeover = 0; + } + } + + /* + * SFR-CACC algorithm: + * 2) If the SACK contains gap acks and the flag CHANGEOVER_ACTIVE + * is set the receiver of the SACK MUST take the following actions: + * + * A) Initialize the cacc_saw_newack to 0 for all destination + * addresses. + */ + if (sack->num_gap_ack_blocks > 0 && + primary->cacc.changeover_active) { + list_for_each(pos, transport_list) { + transport = list_entry(pos, struct sctp_transport, + transports); + transport->cacc.cacc_saw_newack = 0; + } + } + + /* Get the highest TSN in the sack. */ + highest_tsn = sack_ctsn; + if (sack->num_gap_ack_blocks) + highest_tsn += + ntohs(frags[ntohs(sack->num_gap_ack_blocks) - 1].gab.end); + + if (TSN_lt(asoc->highest_sacked, highest_tsn)) { + highest_new_tsn = highest_tsn; + asoc->highest_sacked = highest_tsn; + } else { + highest_new_tsn = sctp_highest_new_tsn(sack, asoc); + } + + /* Run through the retransmit queue. Credit bytes received + * and free those chunks that we can. + */ + sctp_check_transmitted(q, &q->retransmit, NULL, sack, highest_new_tsn); + sctp_mark_missing(q, &q->retransmit, NULL, highest_new_tsn, 0); + + /* Run through the transmitted queue. + * Credit bytes received and free those chunks which we can. + * + * This is a MASSIVE candidate for optimization. + */ + list_for_each(pos, transport_list) { + transport = list_entry(pos, struct sctp_transport, + transports); + sctp_check_transmitted(q, &transport->transmitted, + transport, sack, highest_new_tsn); + /* + * SFR-CACC algorithm: + * C) Let count_of_newacks be the number of + * destinations for which cacc_saw_newack is set. + */ + if (transport->cacc.cacc_saw_newack) + count_of_newacks ++; + } + + list_for_each(pos, transport_list) { + transport = list_entry(pos, struct sctp_transport, + transports); + sctp_mark_missing(q, &transport->transmitted, transport, + highest_new_tsn, count_of_newacks); + } + + /* Move the Cumulative TSN Ack Point if appropriate. */ + if (TSN_lt(asoc->ctsn_ack_point, sack_ctsn)) + asoc->ctsn_ack_point = sack_ctsn; + + /* Update unack_data field in the assoc. */ + sctp_sack_update_unack_data(asoc, sack); + + ctsn = asoc->ctsn_ack_point; + + /* Throw away stuff rotting on the sack queue. */ + list_for_each_safe(lchunk, temp, &q->sacked) { + tchunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + tsn = ntohl(tchunk->subh.data_hdr->tsn); + if (TSN_lte(tsn, ctsn)) + sctp_chunk_free(tchunk); + } + + /* ii) Set rwnd equal to the newly received a_rwnd minus the + * number of bytes still outstanding after processing the + * Cumulative TSN Ack and the Gap Ack Blocks. + */ + + sack_a_rwnd = ntohl(sack->a_rwnd); + outstanding = q->outstanding_bytes; + + if (outstanding < sack_a_rwnd) + sack_a_rwnd -= outstanding; + else + sack_a_rwnd = 0; + + asoc->peer.rwnd = sack_a_rwnd; + + sctp_generate_fwdtsn(q, sack_ctsn); + + SCTP_DEBUG_PRINTK("%s: sack Cumulative TSN Ack is 0x%x.\n", + __FUNCTION__, sack_ctsn); + SCTP_DEBUG_PRINTK("%s: Cumulative TSN Ack of association, " + "%p is 0x%x. Adv peer ack point: 0x%x\n", + __FUNCTION__, asoc, ctsn, asoc->adv_peer_ack_point); + + /* See if all chunks are acked. + * Make sure the empty queue handler will get run later. + */ + q->empty = skb_queue_empty(&q->out) && skb_queue_empty(&q->control) && + list_empty(&q->retransmit); + if (!q->empty) + goto finish; + + list_for_each(pos, transport_list) { + transport = list_entry(pos, struct sctp_transport, + transports); + q->empty = q->empty && list_empty(&transport->transmitted); + if (!q->empty) + goto finish; + } + + SCTP_DEBUG_PRINTK("sack queue is empty.\n"); +finish: + return q->empty; +} + +/* Is the outqueue empty? */ +int sctp_outq_is_empty(const struct sctp_outq *q) +{ + return q->empty; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* Go through a transport's transmitted list or the association's retransmit + * list and move chunks that are acked by the Cumulative TSN Ack to q->sacked. + * The retransmit list will not have an associated transport. + * + * I added coherent debug information output. --xguo + * + * Instead of printing 'sacked' or 'kept' for each TSN on the + * transmitted_queue, we print a range: SACKED: TSN1-TSN2, TSN3, TSN4-TSN5. + * KEPT TSN6-TSN7, etc. + */ +static void sctp_check_transmitted(struct sctp_outq *q, + struct list_head *transmitted_queue, + struct sctp_transport *transport, + struct sctp_sackhdr *sack, + __u32 highest_new_tsn_in_sack) +{ + struct list_head *lchunk; + struct sctp_chunk *tchunk; + struct list_head tlist; + __u32 tsn; + __u32 sack_ctsn; + __u32 rtt; + __u8 restart_timer = 0; + int bytes_acked = 0; + + /* These state variables are for coherent debug output. --xguo */ + +#if SCTP_DEBUG + __u32 dbg_ack_tsn = 0; /* An ACKed TSN range starts here... */ + __u32 dbg_last_ack_tsn = 0; /* ...and finishes here. */ + __u32 dbg_kept_tsn = 0; /* An un-ACKed range starts here... */ + __u32 dbg_last_kept_tsn = 0; /* ...and finishes here. */ + + /* 0 : The last TSN was ACKed. + * 1 : The last TSN was NOT ACKed (i.e. KEPT). + * -1: We need to initialize. + */ + int dbg_prt_state = -1; +#endif /* SCTP_DEBUG */ + + sack_ctsn = ntohl(sack->cum_tsn_ack); + + INIT_LIST_HEAD(&tlist); + + /* The while loop will skip empty transmitted queues. */ + while (NULL != (lchunk = sctp_list_dequeue(transmitted_queue))) { + tchunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + + if (sctp_chunk_abandoned(tchunk)) { + /* Move the chunk to abandoned list. */ + sctp_insert_list(&q->abandoned, lchunk); + continue; + } + + tsn = ntohl(tchunk->subh.data_hdr->tsn); + if (sctp_acked(sack, tsn)) { + /* If this queue is the retransmit queue, the + * retransmit timer has already reclaimed + * the outstanding bytes for this chunk, so only + * count bytes associated with a transport. + */ + if (transport) { + /* If this chunk is being used for RTT + * measurement, calculate the RTT and update + * the RTO using this value. + * + * 6.3.1 C5) Karn's algorithm: RTT measurements + * MUST NOT be made using packets that were + * retransmitted (and thus for which it is + * ambiguous whether the reply was for the + * first instance of the packet or a later + * instance). + */ + if (!tchunk->tsn_gap_acked && + !tchunk->resent && + tchunk->rtt_in_progress) { + rtt = jiffies - tchunk->sent_at; + sctp_transport_update_rto(transport, + rtt); + } + } + if (TSN_lte(tsn, sack_ctsn)) { + /* RFC 2960 6.3.2 Retransmission Timer Rules + * + * R3) Whenever a SACK is received + * that acknowledges the DATA chunk + * with the earliest outstanding TSN + * for that address, restart T3-rtx + * timer for that address with its + * current RTO. + */ + restart_timer = 1; + + if (!tchunk->tsn_gap_acked) { + tchunk->tsn_gap_acked = 1; + bytes_acked += sctp_data_size(tchunk); + /* + * SFR-CACC algorithm: + * 2) If the SACK contains gap acks + * and the flag CHANGEOVER_ACTIVE is + * set the receiver of the SACK MUST + * take the following action: + * + * B) For each TSN t being acked that + * has not been acked in any SACK so + * far, set cacc_saw_newack to 1 for + * the destination that the TSN was + * sent to. + */ + if (transport && + sack->num_gap_ack_blocks && + q->asoc->peer.primary_path->cacc. + changeover_active) + transport->cacc.cacc_saw_newack + = 1; + } + + list_add_tail(&tchunk->transmitted_list, + &q->sacked); + } else { + /* RFC2960 7.2.4, sctpimpguide-05 2.8.2 + * M2) Each time a SACK arrives reporting + * 'Stray DATA chunk(s)' record the highest TSN + * reported as newly acknowledged, call this + * value 'HighestTSNinSack'. A newly + * acknowledged DATA chunk is one not + * previously acknowledged in a SACK. + * + * When the SCTP sender of data receives a SACK + * chunk that acknowledges, for the first time, + * the receipt of a DATA chunk, all the still + * unacknowledged DATA chunks whose TSN is + * older than that newly acknowledged DATA + * chunk, are qualified as 'Stray DATA chunks'. + */ + if (!tchunk->tsn_gap_acked) { + tchunk->tsn_gap_acked = 1; + bytes_acked += sctp_data_size(tchunk); + } + list_add_tail(lchunk, &tlist); + } + +#if SCTP_DEBUG + switch (dbg_prt_state) { + case 0: /* last TSN was ACKed */ + if (dbg_last_ack_tsn + 1 == tsn) { + /* This TSN belongs to the + * current ACK range. + */ + break; + } + + if (dbg_last_ack_tsn != dbg_ack_tsn) { + /* Display the end of the + * current range. + */ + SCTP_DEBUG_PRINTK("-%08x", + dbg_last_ack_tsn); + } + + /* Start a new range. */ + SCTP_DEBUG_PRINTK(",%08x", tsn); + dbg_ack_tsn = tsn; + break; + + case 1: /* The last TSN was NOT ACKed. */ + if (dbg_last_kept_tsn != dbg_kept_tsn) { + /* Display the end of current range. */ + SCTP_DEBUG_PRINTK("-%08x", + dbg_last_kept_tsn); + } + + SCTP_DEBUG_PRINTK("\n"); + + /* FALL THROUGH... */ + default: + /* This is the first-ever TSN we examined. */ + /* Start a new range of ACK-ed TSNs. */ + SCTP_DEBUG_PRINTK("ACKed: %08x", tsn); + dbg_prt_state = 0; + dbg_ack_tsn = tsn; + }; + + dbg_last_ack_tsn = tsn; +#endif /* SCTP_DEBUG */ + + } else { + if (tchunk->tsn_gap_acked) { + SCTP_DEBUG_PRINTK("%s: Receiver reneged on " + "data TSN: 0x%x\n", + __FUNCTION__, + tsn); + tchunk->tsn_gap_acked = 0; + + bytes_acked -= sctp_data_size(tchunk); + + /* RFC 2960 6.3.2 Retransmission Timer Rules + * + * R4) Whenever a SACK is received missing a + * TSN that was previously acknowledged via a + * Gap Ack Block, start T3-rtx for the + * destination address to which the DATA + * chunk was originally + * transmitted if it is not already running. + */ + restart_timer = 1; + } + + list_add_tail(lchunk, &tlist); + +#if SCTP_DEBUG + /* See the above comments on ACK-ed TSNs. */ + switch (dbg_prt_state) { + case 1: + if (dbg_last_kept_tsn + 1 == tsn) + break; + + if (dbg_last_kept_tsn != dbg_kept_tsn) + SCTP_DEBUG_PRINTK("-%08x", + dbg_last_kept_tsn); + + SCTP_DEBUG_PRINTK(",%08x", tsn); + dbg_kept_tsn = tsn; + break; + + case 0: + if (dbg_last_ack_tsn != dbg_ack_tsn) + SCTP_DEBUG_PRINTK("-%08x", + dbg_last_ack_tsn); + SCTP_DEBUG_PRINTK("\n"); + + /* FALL THROUGH... */ + default: + SCTP_DEBUG_PRINTK("KEPT: %08x",tsn); + dbg_prt_state = 1; + dbg_kept_tsn = tsn; + }; + + dbg_last_kept_tsn = tsn; +#endif /* SCTP_DEBUG */ + } + } + +#if SCTP_DEBUG + /* Finish off the last range, displaying its ending TSN. */ + switch (dbg_prt_state) { + case 0: + if (dbg_last_ack_tsn != dbg_ack_tsn) { + SCTP_DEBUG_PRINTK("-%08x\n", dbg_last_ack_tsn); + } else { + SCTP_DEBUG_PRINTK("\n"); + } + break; + + case 1: + if (dbg_last_kept_tsn != dbg_kept_tsn) { + SCTP_DEBUG_PRINTK("-%08x\n", dbg_last_kept_tsn); + } else { + SCTP_DEBUG_PRINTK("\n"); + } + }; +#endif /* SCTP_DEBUG */ + if (transport) { + if (bytes_acked) { + /* 8.2. When an outstanding TSN is acknowledged, + * the endpoint shall clear the error counter of + * the destination transport address to which the + * DATA chunk was last sent. + * The association's overall error counter is + * also cleared. + */ + transport->error_count = 0; + transport->asoc->overall_error_count = 0; + + /* Mark the destination transport address as + * active if it is not so marked. + */ + if (!transport->active) { + sctp_assoc_control_transport( + transport->asoc, + transport, + SCTP_TRANSPORT_UP, + SCTP_RECEIVED_SACK); + } + + sctp_transport_raise_cwnd(transport, sack_ctsn, + bytes_acked); + + transport->flight_size -= bytes_acked; + q->outstanding_bytes -= bytes_acked; + } else { + /* RFC 2960 6.1, sctpimpguide-06 2.15.2 + * When a sender is doing zero window probing, it + * should not timeout the association if it continues + * to receive new packets from the receiver. The + * reason is that the receiver MAY keep its window + * closed for an indefinite time. + * A sender is doing zero window probing when the + * receiver's advertised window is zero, and there is + * only one data chunk in flight to the receiver. + */ + if (!q->asoc->peer.rwnd && + !list_empty(&tlist) && + (sack_ctsn+2 == q->asoc->next_tsn)) { + SCTP_DEBUG_PRINTK("%s: SACK received for zero " + "window probe: %u\n", + __FUNCTION__, sack_ctsn); + q->asoc->overall_error_count = 0; + transport->error_count = 0; + } + } + + /* RFC 2960 6.3.2 Retransmission Timer Rules + * + * R2) Whenever all outstanding data sent to an address have + * been acknowledged, turn off the T3-rtx timer of that + * address. + */ + if (!transport->flight_size) { + if (timer_pending(&transport->T3_rtx_timer) && + del_timer(&transport->T3_rtx_timer)) { + sctp_transport_put(transport); + } + } else if (restart_timer) { + if (!mod_timer(&transport->T3_rtx_timer, + jiffies + transport->rto)) + sctp_transport_hold(transport); + } + } + + list_splice(&tlist, transmitted_queue); +} + +/* Mark chunks as missing and consequently may get retransmitted. */ +static void sctp_mark_missing(struct sctp_outq *q, + struct list_head *transmitted_queue, + struct sctp_transport *transport, + __u32 highest_new_tsn_in_sack, + int count_of_newacks) +{ + struct sctp_chunk *chunk; + struct list_head *pos; + __u32 tsn; + char do_fast_retransmit = 0; + struct sctp_transport *primary = q->asoc->peer.primary_path; + + list_for_each(pos, transmitted_queue) { + + chunk = list_entry(pos, struct sctp_chunk, transmitted_list); + tsn = ntohl(chunk->subh.data_hdr->tsn); + + /* RFC 2960 7.2.4, sctpimpguide-05 2.8.2 M3) Examine all + * 'Unacknowledged TSN's', if the TSN number of an + * 'Unacknowledged TSN' is smaller than the 'HighestTSNinSack' + * value, increment the 'TSN.Missing.Report' count on that + * chunk if it has NOT been fast retransmitted or marked for + * fast retransmit already. + */ + if (!chunk->fast_retransmit && + !chunk->tsn_gap_acked && + TSN_lt(tsn, highest_new_tsn_in_sack)) { + + /* SFR-CACC may require us to skip marking + * this chunk as missing. + */ + if (!transport || !sctp_cacc_skip(primary, transport, + count_of_newacks, tsn)) { + chunk->tsn_missing_report++; + + SCTP_DEBUG_PRINTK( + "%s: TSN 0x%x missing counter: %d\n", + __FUNCTION__, tsn, + chunk->tsn_missing_report); + } + } + /* + * M4) If any DATA chunk is found to have a + * 'TSN.Missing.Report' + * value larger than or equal to 4, mark that chunk for + * retransmission and start the fast retransmit procedure. + */ + + if (chunk->tsn_missing_report >= 4) { + chunk->fast_retransmit = 1; + do_fast_retransmit = 1; + } + } + + if (transport) { + if (do_fast_retransmit) + sctp_retransmit(q, transport, SCTP_RTXR_FAST_RTX); + + SCTP_DEBUG_PRINTK("%s: transport: %p, cwnd: %d, " + "ssthresh: %d, flight_size: %d, pba: %d\n", + __FUNCTION__, transport, transport->cwnd, + transport->ssthresh, transport->flight_size, + transport->partial_bytes_acked); + } +} + +/* Is the given TSN acked by this packet? */ +static int sctp_acked(struct sctp_sackhdr *sack, __u32 tsn) +{ + int i; + sctp_sack_variable_t *frags; + __u16 gap; + __u32 ctsn = ntohl(sack->cum_tsn_ack); + + if (TSN_lte(tsn, ctsn)) + goto pass; + + /* 3.3.4 Selective Acknowledgement (SACK) (3): + * + * Gap Ack Blocks: + * These fields contain the Gap Ack Blocks. They are repeated + * for each Gap Ack Block up to the number of Gap Ack Blocks + * defined in the Number of Gap Ack Blocks field. All DATA + * chunks with TSNs greater than or equal to (Cumulative TSN + * Ack + Gap Ack Block Start) and less than or equal to + * (Cumulative TSN Ack + Gap Ack Block End) of each Gap Ack + * Block are assumed to have been received correctly. + */ + + frags = sack->variable; + gap = tsn - ctsn; + for (i = 0; i < ntohs(sack->num_gap_ack_blocks); ++i) { + if (TSN_lte(ntohs(frags[i].gab.start), gap) && + TSN_lte(gap, ntohs(frags[i].gab.end))) + goto pass; + } + + return 0; +pass: + return 1; +} + +static inline int sctp_get_skip_pos(struct sctp_fwdtsn_skip *skiplist, + int nskips, __u16 stream) +{ + int i; + + for (i = 0; i < nskips; i++) { + if (skiplist[i].stream == stream) + return i; + } + return i; +} + +/* Create and add a fwdtsn chunk to the outq's control queue if needed. */ +static void sctp_generate_fwdtsn(struct sctp_outq *q, __u32 ctsn) +{ + struct sctp_association *asoc = q->asoc; + struct sctp_chunk *ftsn_chunk = NULL; + struct sctp_fwdtsn_skip ftsn_skip_arr[10]; + int nskips = 0; + int skip_pos = 0; + __u32 tsn; + struct sctp_chunk *chunk; + struct list_head *lchunk, *temp; + + /* PR-SCTP C1) Let SackCumAck be the Cumulative TSN ACK carried in the + * received SACK. + * + * If (Advanced.Peer.Ack.Point < SackCumAck), then update + * Advanced.Peer.Ack.Point to be equal to SackCumAck. + */ + if (TSN_lt(asoc->adv_peer_ack_point, ctsn)) + asoc->adv_peer_ack_point = ctsn; + + /* PR-SCTP C2) Try to further advance the "Advanced.Peer.Ack.Point" + * locally, that is, to move "Advanced.Peer.Ack.Point" up as long as + * the chunk next in the out-queue space is marked as "abandoned" as + * shown in the following example: + * + * Assuming that a SACK arrived with the Cumulative TSN ACK 102 + * and the Advanced.Peer.Ack.Point is updated to this value: + * + * out-queue at the end of ==> out-queue after Adv.Ack.Point + * normal SACK processing local advancement + * ... ... + * Adv.Ack.Pt-> 102 acked 102 acked + * 103 abandoned 103 abandoned + * 104 abandoned Adv.Ack.P-> 104 abandoned + * 105 105 + * 106 acked 106 acked + * ... ... + * + * In this example, the data sender successfully advanced the + * "Advanced.Peer.Ack.Point" from 102 to 104 locally. + */ + list_for_each_safe(lchunk, temp, &q->abandoned) { + chunk = list_entry(lchunk, struct sctp_chunk, + transmitted_list); + tsn = ntohl(chunk->subh.data_hdr->tsn); + + /* Remove any chunks in the abandoned queue that are acked by + * the ctsn. + */ + if (TSN_lte(tsn, ctsn)) { + list_del_init(lchunk); + if (!chunk->tsn_gap_acked) { + chunk->transport->flight_size -= + sctp_data_size(chunk); + q->outstanding_bytes -= sctp_data_size(chunk); + } + sctp_chunk_free(chunk); + } else { + if (TSN_lte(tsn, asoc->adv_peer_ack_point+1)) { + asoc->adv_peer_ack_point = tsn; + if (chunk->chunk_hdr->flags & + SCTP_DATA_UNORDERED) + continue; + skip_pos = sctp_get_skip_pos(&ftsn_skip_arr[0], + nskips, + chunk->subh.data_hdr->stream); + ftsn_skip_arr[skip_pos].stream = + chunk->subh.data_hdr->stream; + ftsn_skip_arr[skip_pos].ssn = + chunk->subh.data_hdr->ssn; + if (skip_pos == nskips) + nskips++; + if (nskips == 10) + break; + } else + break; + } + } + + /* PR-SCTP C3) If, after step C1 and C2, the "Advanced.Peer.Ack.Point" + * is greater than the Cumulative TSN ACK carried in the received + * SACK, the data sender MUST send the data receiver a FORWARD TSN + * chunk containing the latest value of the + * "Advanced.Peer.Ack.Point". + * + * C4) For each "abandoned" TSN the sender of the FORWARD TSN SHOULD + * list each stream and sequence number in the forwarded TSN. This + * information will enable the receiver to easily find any + * stranded TSN's waiting on stream reorder queues. Each stream + * SHOULD only be reported once; this means that if multiple + * abandoned messages occur in the same stream then only the + * highest abandoned stream sequence number is reported. If the + * total size of the FORWARD TSN does NOT fit in a single MTU then + * the sender of the FORWARD TSN SHOULD lower the + * Advanced.Peer.Ack.Point to the last TSN that will fit in a + * single MTU. + */ + if (asoc->adv_peer_ack_point > ctsn) + ftsn_chunk = sctp_make_fwdtsn(asoc, asoc->adv_peer_ack_point, + nskips, &ftsn_skip_arr[0]); + + if (ftsn_chunk) { + __skb_queue_tail(&q->control, (struct sk_buff *)ftsn_chunk); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + } +} diff --git a/net/sctp/primitive.c b/net/sctp/primitive.c new file mode 100644 index 000000000000..3a7ebfcc1fdb --- /dev/null +++ b/net/sctp/primitive.c @@ -0,0 +1,219 @@ +/* SCTP kernel reference Implementation + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions implement the SCTP primitive functions from Section 10. + * + * Note that the descriptions from the specification are USER level + * functions--this file is the functions which populate the struct proto + * for SCTP which is the BOTTOM of the sockets interface. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Narasimha Budihal <narasimha@refcode.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Ardelle Fan <ardelle.fan@intel.com> + * Kevin Gao <kevin.gao@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/list.h> /* For struct list_head */ +#include <linux/socket.h> +#include <linux/ip.h> +#include <linux/time.h> /* For struct timeval */ +#include <net/sock.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +#define DECLARE_PRIMITIVE(name) \ +/* This is called in the code as sctp_primitive_ ## name. */ \ +int sctp_primitive_ ## name(struct sctp_association *asoc, \ + void *arg) { \ + int error = 0; \ + sctp_event_t event_type; sctp_subtype_t subtype; \ + sctp_state_t state; \ + struct sctp_endpoint *ep; \ + \ + event_type = SCTP_EVENT_T_PRIMITIVE; \ + subtype = SCTP_ST_PRIMITIVE(SCTP_PRIMITIVE_ ## name); \ + state = asoc ? asoc->state : SCTP_STATE_CLOSED; \ + ep = asoc ? asoc->ep : NULL; \ + \ + error = sctp_do_sm(event_type, subtype, state, ep, asoc, \ + arg, GFP_KERNEL); \ + return error; \ +} + +/* 10.1 ULP-to-SCTP + * B) Associate + * + * Format: ASSOCIATE(local SCTP instance name, destination transport addr, + * outbound stream count) + * -> association id [,destination transport addr list] [,outbound stream + * count] + * + * This primitive allows the upper layer to initiate an association to a + * specific peer endpoint. + * + * This version assumes that asoc is fully populated with the initial + * parameters. We then return a traditional kernel indicator of + * success or failure. + */ + +/* This is called in the code as sctp_primitive_ASSOCIATE. */ + +DECLARE_PRIMITIVE(ASSOCIATE) + +/* 10.1 ULP-to-SCTP + * C) Shutdown + * + * Format: SHUTDOWN(association id) + * -> result + * + * Gracefully closes an association. Any locally queued user data + * will be delivered to the peer. The association will be terminated only + * after the peer acknowledges all the SCTP packets sent. A success code + * will be returned on successful termination of the association. If + * attempting to terminate the association results in a failure, an error + * code shall be returned. + */ + +DECLARE_PRIMITIVE(SHUTDOWN); + +/* 10.1 ULP-to-SCTP + * C) Abort + * + * Format: Abort(association id [, cause code]) + * -> result + * + * Ungracefully closes an association. Any locally queued user data + * will be discarded and an ABORT chunk is sent to the peer. A success + * code will be returned on successful abortion of the association. If + * attempting to abort the association results in a failure, an error + * code shall be returned. + */ + +DECLARE_PRIMITIVE(ABORT); + +/* 10.1 ULP-to-SCTP + * E) Send + * + * Format: SEND(association id, buffer address, byte count [,context] + * [,stream id] [,life time] [,destination transport address] + * [,unorder flag] [,no-bundle flag] [,payload protocol-id] ) + * -> result + * + * This is the main method to send user data via SCTP. + * + * Mandatory attributes: + * + * o association id - local handle to the SCTP association + * + * o buffer address - the location where the user message to be + * transmitted is stored; + * + * o byte count - The size of the user data in number of bytes; + * + * Optional attributes: + * + * o context - an optional 32 bit integer that will be carried in the + * sending failure notification to the ULP if the transportation of + * this User Message fails. + * + * o stream id - to indicate which stream to send the data on. If not + * specified, stream 0 will be used. + * + * o life time - specifies the life time of the user data. The user data + * will not be sent by SCTP after the life time expires. This + * parameter can be used to avoid efforts to transmit stale + * user messages. SCTP notifies the ULP if the data cannot be + * initiated to transport (i.e. sent to the destination via SCTP's + * send primitive) within the life time variable. However, the + * user data will be transmitted if SCTP has attempted to transmit a + * chunk before the life time expired. + * + * o destination transport address - specified as one of the destination + * transport addresses of the peer endpoint to which this packet + * should be sent. Whenever possible, SCTP should use this destination + * transport address for sending the packets, instead of the current + * primary path. + * + * o unorder flag - this flag, if present, indicates that the user + * would like the data delivered in an unordered fashion to the peer + * (i.e., the U flag is set to 1 on all DATA chunks carrying this + * message). + * + * o no-bundle flag - instructs SCTP not to bundle this user data with + * other outbound DATA chunks. SCTP MAY still bundle even when + * this flag is present, when faced with network congestion. + * + * o payload protocol-id - A 32 bit unsigned integer that is to be + * passed to the peer indicating the type of payload protocol data + * being transmitted. This value is passed as opaque data by SCTP. + */ + +DECLARE_PRIMITIVE(SEND); + +/* 10.1 ULP-to-SCTP + * J) Request Heartbeat + * + * Format: REQUESTHEARTBEAT(association id, destination transport address) + * + * -> result + * + * Instructs the local endpoint to perform a HeartBeat on the specified + * destination transport address of the given association. The returned + * result should indicate whether the transmission of the HEARTBEAT + * chunk to the destination address is successful. + * + * Mandatory attributes: + * + * o association id - local handle to the SCTP association + * + * o destination transport address - the transport address of the + * association on which a heartbeat should be issued. + */ + +DECLARE_PRIMITIVE(REQUESTHEARTBEAT); + +/* ADDIP +* 3.1.1 Address Configuration Change Chunk (ASCONF) +* +* This chunk is used to communicate to the remote endpoint one of the +* configuration change requests that MUST be acknowledged. The +* information carried in the ASCONF Chunk uses the form of a +* Type-Length-Value (TLV), as described in "3.2.1 Optional/ +* Variable-length Parameter Format" in RFC2960 [5], forall variable +* parameters. +*/ + +DECLARE_PRIMITIVE(ASCONF); diff --git a/net/sctp/proc.c b/net/sctp/proc.c new file mode 100644 index 000000000000..e42fd8c2916b --- /dev/null +++ b/net/sctp/proc.c @@ -0,0 +1,288 @@ +/* SCTP kernel reference Implementation + * Copyright (c) 2003 International Business Machines, Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/seq_file.h> +#include <linux/init.h> +#include <net/sctp/sctp.h> + +static struct snmp_mib sctp_snmp_list[] = { + SNMP_MIB_ITEM("SctpCurrEstab", SCTP_MIB_CURRESTAB), + SNMP_MIB_ITEM("SctpActiveEstabs", SCTP_MIB_ACTIVEESTABS), + SNMP_MIB_ITEM("SctpPassiveEstabs", SCTP_MIB_PASSIVEESTABS), + SNMP_MIB_ITEM("SctpAborteds", SCTP_MIB_ABORTEDS), + SNMP_MIB_ITEM("SctpShutdowns", SCTP_MIB_SHUTDOWNS), + SNMP_MIB_ITEM("SctpOutOfBlues", SCTP_MIB_OUTOFBLUES), + SNMP_MIB_ITEM("SctpChecksumErrors", SCTP_MIB_CHECKSUMERRORS), + SNMP_MIB_ITEM("SctpOutCtrlChunks", SCTP_MIB_OUTCTRLCHUNKS), + SNMP_MIB_ITEM("SctpOutOrderChunks", SCTP_MIB_OUTORDERCHUNKS), + SNMP_MIB_ITEM("SctpOutUnorderChunks", SCTP_MIB_OUTUNORDERCHUNKS), + SNMP_MIB_ITEM("SctpInCtrlChunks", SCTP_MIB_INCTRLCHUNKS), + SNMP_MIB_ITEM("SctpInOrderChunks", SCTP_MIB_INORDERCHUNKS), + SNMP_MIB_ITEM("SctpInUnorderChunks", SCTP_MIB_INUNORDERCHUNKS), + SNMP_MIB_ITEM("SctpFragUsrMsgs", SCTP_MIB_FRAGUSRMSGS), + SNMP_MIB_ITEM("SctpReasmUsrMsgs", SCTP_MIB_REASMUSRMSGS), + SNMP_MIB_ITEM("SctpOutSCTPPacks", SCTP_MIB_OUTSCTPPACKS), + SNMP_MIB_ITEM("SctpInSCTPPacks", SCTP_MIB_INSCTPPACKS), +}; + +/* Return the current value of a particular entry in the mib by adding its + * per cpu counters. + */ +static unsigned long +fold_field(void *mib[], int nr) +{ + unsigned long res = 0; + int i; + + for (i = 0; i < NR_CPUS; i++) { + if (!cpu_possible(i)) + continue; + res += + *((unsigned long *) (((void *) per_cpu_ptr(mib[0], i)) + + sizeof (unsigned long) * nr)); + res += + *((unsigned long *) (((void *) per_cpu_ptr(mib[1], i)) + + sizeof (unsigned long) * nr)); + } + return res; +} + +/* Display sctp snmp mib statistics(/proc/net/sctp/snmp). */ +static int sctp_snmp_seq_show(struct seq_file *seq, void *v) +{ + int i; + + for (i = 0; sctp_snmp_list[i].name != NULL; i++) + seq_printf(seq, "%-32s\t%ld\n", sctp_snmp_list[i].name, + fold_field((void **)sctp_statistics, + sctp_snmp_list[i].entry)); + + return 0; +} + +/* Initialize the seq file operations for 'snmp' object. */ +static int sctp_snmp_seq_open(struct inode *inode, struct file *file) +{ + return single_open(file, sctp_snmp_seq_show, NULL); +} + +static struct file_operations sctp_snmp_seq_fops = { + .owner = THIS_MODULE, + .open = sctp_snmp_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/* Set up the proc fs entry for 'snmp' object. */ +int __init sctp_snmp_proc_init(void) +{ + struct proc_dir_entry *p; + + p = create_proc_entry("snmp", S_IRUGO, proc_net_sctp); + if (!p) + return -ENOMEM; + + p->proc_fops = &sctp_snmp_seq_fops; + + return 0; +} + +/* Cleanup the proc fs entry for 'snmp' object. */ +void sctp_snmp_proc_exit(void) +{ + remove_proc_entry("snmp", proc_net_sctp); +} + +/* Dump local addresses of an association/endpoint. */ +static void sctp_seq_dump_local_addrs(struct seq_file *seq, struct sctp_ep_common *epb) +{ + struct list_head *pos; + struct sctp_sockaddr_entry *laddr; + union sctp_addr *addr; + struct sctp_af *af; + + list_for_each(pos, &epb->bind_addr.address_list) { + laddr = list_entry(pos, struct sctp_sockaddr_entry, list); + addr = (union sctp_addr *)&laddr->a; + af = sctp_get_af_specific(addr->sa.sa_family); + af->seq_dump_addr(seq, addr); + } +} + +/* Dump remote addresses of an association. */ +static void sctp_seq_dump_remote_addrs(struct seq_file *seq, struct sctp_association *assoc) +{ + struct list_head *pos; + struct sctp_transport *transport; + union sctp_addr *addr; + struct sctp_af *af; + + list_for_each(pos, &assoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + addr = (union sctp_addr *)&transport->ipaddr; + af = sctp_get_af_specific(addr->sa.sa_family); + af->seq_dump_addr(seq, addr); + } +} + +/* Display sctp endpoints (/proc/net/sctp/eps). */ +static int sctp_eps_seq_show(struct seq_file *seq, void *v) +{ + struct sctp_hashbucket *head; + struct sctp_ep_common *epb; + struct sctp_endpoint *ep; + struct sock *sk; + int hash; + + seq_printf(seq, " ENDPT SOCK STY SST HBKT LPORT LADDRS\n"); + for (hash = 0; hash < sctp_ep_hashsize; hash++) { + head = &sctp_ep_hashtable[hash]; + read_lock(&head->lock); + for (epb = head->chain; epb; epb = epb->next) { + ep = sctp_ep(epb); + sk = epb->sk; + seq_printf(seq, "%8p %8p %-3d %-3d %-4d %-5d ", ep, sk, + sctp_sk(sk)->type, sk->sk_state, hash, + epb->bind_addr.port); + sctp_seq_dump_local_addrs(seq, epb); + seq_printf(seq, "\n"); + } + read_unlock(&head->lock); + } + + return 0; +} + +/* Initialize the seq file operations for 'eps' object. */ +static int sctp_eps_seq_open(struct inode *inode, struct file *file) +{ + return single_open(file, sctp_eps_seq_show, NULL); +} + +static struct file_operations sctp_eps_seq_fops = { + .open = sctp_eps_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/* Set up the proc fs entry for 'eps' object. */ +int __init sctp_eps_proc_init(void) +{ + struct proc_dir_entry *p; + + p = create_proc_entry("eps", S_IRUGO, proc_net_sctp); + if (!p) + return -ENOMEM; + + p->proc_fops = &sctp_eps_seq_fops; + + return 0; +} + +/* Cleanup the proc fs entry for 'eps' object. */ +void sctp_eps_proc_exit(void) +{ + remove_proc_entry("eps", proc_net_sctp); +} + +/* Display sctp associations (/proc/net/sctp/assocs). */ +static int sctp_assocs_seq_show(struct seq_file *seq, void *v) +{ + struct sctp_hashbucket *head; + struct sctp_ep_common *epb; + struct sctp_association *assoc; + struct sock *sk; + int hash; + + seq_printf(seq, " ASSOC SOCK STY SST ST HBKT LPORT RPORT " + "LADDRS <-> RADDRS\n"); + for (hash = 0; hash < sctp_assoc_hashsize; hash++) { + head = &sctp_assoc_hashtable[hash]; + read_lock(&head->lock); + for (epb = head->chain; epb; epb = epb->next) { + assoc = sctp_assoc(epb); + sk = epb->sk; + seq_printf(seq, + "%8p %8p %-3d %-3d %-2d %-4d %-5d %-5d ", + assoc, sk, sctp_sk(sk)->type, sk->sk_state, + assoc->state, hash, epb->bind_addr.port, + assoc->peer.port); + sctp_seq_dump_local_addrs(seq, epb); + seq_printf(seq, "<-> "); + sctp_seq_dump_remote_addrs(seq, assoc); + seq_printf(seq, "\n"); + } + read_unlock(&head->lock); + } + + return 0; +} + +/* Initialize the seq file operations for 'assocs' object. */ +static int sctp_assocs_seq_open(struct inode *inode, struct file *file) +{ + return single_open(file, sctp_assocs_seq_show, NULL); +} + +static struct file_operations sctp_assocs_seq_fops = { + .open = sctp_assocs_seq_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +/* Set up the proc fs entry for 'assocs' object. */ +int __init sctp_assocs_proc_init(void) +{ + struct proc_dir_entry *p; + + p = create_proc_entry("assocs", S_IRUGO, proc_net_sctp); + if (!p) + return -ENOMEM; + + p->proc_fops = &sctp_assocs_seq_fops; + + return 0; +} + +/* Cleanup the proc fs entry for 'assocs' object. */ +void sctp_assocs_proc_exit(void) +{ + remove_proc_entry("assocs", proc_net_sctp); +} diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c new file mode 100644 index 000000000000..b9813cf3d91c --- /dev/null +++ b/net/sctp/protocol.c @@ -0,0 +1,1240 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * Copyright (c) 2001 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * This file is part of the SCTP kernel reference Implementation + * + * Initialization/cleanup for SCTP protocol support. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Daisy Chang <daisyc@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/module.h> +#include <linux/init.h> +#include <linux/netdevice.h> +#include <linux/inetdevice.h> +#include <linux/seq_file.h> +#include <net/protocol.h> +#include <net/ip.h> +#include <net/ipv6.h> +#include <net/sctp/sctp.h> +#include <net/addrconf.h> +#include <net/inet_common.h> +#include <net/inet_ecn.h> + +/* Global data structures. */ +struct sctp_globals sctp_globals; +struct proc_dir_entry *proc_net_sctp; +DEFINE_SNMP_STAT(struct sctp_mib, sctp_statistics); + +struct idr sctp_assocs_id; +DEFINE_SPINLOCK(sctp_assocs_id_lock); + +/* This is the global socket data structure used for responding to + * the Out-of-the-blue (OOTB) packets. A control sock will be created + * for this socket at the initialization time. + */ +static struct socket *sctp_ctl_socket; + +static struct sctp_pf *sctp_pf_inet6_specific; +static struct sctp_pf *sctp_pf_inet_specific; +static struct sctp_af *sctp_af_v4_specific; +static struct sctp_af *sctp_af_v6_specific; + +kmem_cache_t *sctp_chunk_cachep; +kmem_cache_t *sctp_bucket_cachep; + +extern int sctp_snmp_proc_init(void); +extern int sctp_snmp_proc_exit(void); +extern int sctp_eps_proc_init(void); +extern int sctp_eps_proc_exit(void); +extern int sctp_assocs_proc_init(void); +extern int sctp_assocs_proc_exit(void); + +/* Return the address of the control sock. */ +struct sock *sctp_get_ctl_sock(void) +{ + return sctp_ctl_socket->sk; +} + +/* Set up the proc fs entry for the SCTP protocol. */ +static __init int sctp_proc_init(void) +{ + if (!proc_net_sctp) { + struct proc_dir_entry *ent; + ent = proc_mkdir("net/sctp", NULL); + if (ent) { + ent->owner = THIS_MODULE; + proc_net_sctp = ent; + } else + goto out_nomem; + } + + if (sctp_snmp_proc_init()) + goto out_nomem; + if (sctp_eps_proc_init()) + goto out_nomem; + if (sctp_assocs_proc_init()) + goto out_nomem; + + return 0; + +out_nomem: + return -ENOMEM; +} + +/* Clean up the proc fs entry for the SCTP protocol. + * Note: Do not make this __exit as it is used in the init error + * path. + */ +static void sctp_proc_exit(void) +{ + sctp_snmp_proc_exit(); + sctp_eps_proc_exit(); + sctp_assocs_proc_exit(); + + if (proc_net_sctp) { + proc_net_sctp = NULL; + remove_proc_entry("net/sctp", NULL); + } +} + +/* Private helper to extract ipv4 address and stash them in + * the protocol structure. + */ +static void sctp_v4_copy_addrlist(struct list_head *addrlist, + struct net_device *dev) +{ + struct in_device *in_dev; + struct in_ifaddr *ifa; + struct sctp_sockaddr_entry *addr; + + rcu_read_lock(); + if ((in_dev = __in_dev_get(dev)) == NULL) { + rcu_read_unlock(); + return; + } + + for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) { + /* Add the address to the local list. */ + addr = t_new(struct sctp_sockaddr_entry, GFP_ATOMIC); + if (addr) { + addr->a.v4.sin_family = AF_INET; + addr->a.v4.sin_port = 0; + addr->a.v4.sin_addr.s_addr = ifa->ifa_local; + list_add_tail(&addr->list, addrlist); + } + } + + rcu_read_unlock(); +} + +/* Extract our IP addresses from the system and stash them in the + * protocol structure. + */ +static void __sctp_get_local_addr_list(void) +{ + struct net_device *dev; + struct list_head *pos; + struct sctp_af *af; + + read_lock(&dev_base_lock); + for (dev = dev_base; dev; dev = dev->next) { + __list_for_each(pos, &sctp_address_families) { + af = list_entry(pos, struct sctp_af, list); + af->copy_addrlist(&sctp_local_addr_list, dev); + } + } + read_unlock(&dev_base_lock); +} + +static void sctp_get_local_addr_list(void) +{ + unsigned long flags; + + sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags); + __sctp_get_local_addr_list(); + sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags); +} + +/* Free the existing local addresses. */ +static void __sctp_free_local_addr_list(void) +{ + struct sctp_sockaddr_entry *addr; + struct list_head *pos, *temp; + + list_for_each_safe(pos, temp, &sctp_local_addr_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + list_del(pos); + kfree(addr); + } +} + +/* Free the existing local addresses. */ +static void sctp_free_local_addr_list(void) +{ + unsigned long flags; + + sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags); + __sctp_free_local_addr_list(); + sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags); +} + +/* Copy the local addresses which are valid for 'scope' into 'bp'. */ +int sctp_copy_local_addr_list(struct sctp_bind_addr *bp, sctp_scope_t scope, + int gfp, int copy_flags) +{ + struct sctp_sockaddr_entry *addr; + int error = 0; + struct list_head *pos; + unsigned long flags; + + sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags); + list_for_each(pos, &sctp_local_addr_list) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + if (sctp_in_scope(&addr->a, scope)) { + /* Now that the address is in scope, check to see if + * the address type is really supported by the local + * sock as well as the remote peer. + */ + if ((((AF_INET == addr->a.sa.sa_family) && + (copy_flags & SCTP_ADDR4_PEERSUPP))) || + (((AF_INET6 == addr->a.sa.sa_family) && + (copy_flags & SCTP_ADDR6_ALLOWED) && + (copy_flags & SCTP_ADDR6_PEERSUPP)))) { + error = sctp_add_bind_addr(bp, &addr->a, + GFP_ATOMIC); + if (error) + goto end_copy; + } + } + } + +end_copy: + sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags); + return error; +} + +/* Initialize a sctp_addr from in incoming skb. */ +static void sctp_v4_from_skb(union sctp_addr *addr, struct sk_buff *skb, + int is_saddr) +{ + void *from; + __u16 *port; + struct sctphdr *sh; + + port = &addr->v4.sin_port; + addr->v4.sin_family = AF_INET; + + sh = (struct sctphdr *) skb->h.raw; + if (is_saddr) { + *port = ntohs(sh->source); + from = &skb->nh.iph->saddr; + } else { + *port = ntohs(sh->dest); + from = &skb->nh.iph->daddr; + } + memcpy(&addr->v4.sin_addr.s_addr, from, sizeof(struct in_addr)); +} + +/* Initialize an sctp_addr from a socket. */ +static void sctp_v4_from_sk(union sctp_addr *addr, struct sock *sk) +{ + addr->v4.sin_family = AF_INET; + addr->v4.sin_port = inet_sk(sk)->num; + addr->v4.sin_addr.s_addr = inet_sk(sk)->rcv_saddr; +} + +/* Initialize sk->sk_rcv_saddr from sctp_addr. */ +static void sctp_v4_to_sk_saddr(union sctp_addr *addr, struct sock *sk) +{ + inet_sk(sk)->rcv_saddr = addr->v4.sin_addr.s_addr; +} + +/* Initialize sk->sk_daddr from sctp_addr. */ +static void sctp_v4_to_sk_daddr(union sctp_addr *addr, struct sock *sk) +{ + inet_sk(sk)->daddr = addr->v4.sin_addr.s_addr; +} + +/* Initialize a sctp_addr from an address parameter. */ +static void sctp_v4_from_addr_param(union sctp_addr *addr, + union sctp_addr_param *param, + __u16 port, int iif) +{ + addr->v4.sin_family = AF_INET; + addr->v4.sin_port = port; + addr->v4.sin_addr.s_addr = param->v4.addr.s_addr; +} + +/* Initialize an address parameter from a sctp_addr and return the length + * of the address parameter. + */ +static int sctp_v4_to_addr_param(const union sctp_addr *addr, + union sctp_addr_param *param) +{ + int length = sizeof(sctp_ipv4addr_param_t); + + param->v4.param_hdr.type = SCTP_PARAM_IPV4_ADDRESS; + param->v4.param_hdr.length = ntohs(length); + param->v4.addr.s_addr = addr->v4.sin_addr.s_addr; + + return length; +} + +/* Initialize a sctp_addr from a dst_entry. */ +static void sctp_v4_dst_saddr(union sctp_addr *saddr, struct dst_entry *dst, + unsigned short port) +{ + struct rtable *rt = (struct rtable *)dst; + saddr->v4.sin_family = AF_INET; + saddr->v4.sin_port = port; + saddr->v4.sin_addr.s_addr = rt->rt_src; +} + +/* Compare two addresses exactly. */ +static int sctp_v4_cmp_addr(const union sctp_addr *addr1, + const union sctp_addr *addr2) +{ + if (addr1->sa.sa_family != addr2->sa.sa_family) + return 0; + if (addr1->v4.sin_port != addr2->v4.sin_port) + return 0; + if (addr1->v4.sin_addr.s_addr != addr2->v4.sin_addr.s_addr) + return 0; + + return 1; +} + +/* Initialize addr struct to INADDR_ANY. */ +static void sctp_v4_inaddr_any(union sctp_addr *addr, unsigned short port) +{ + addr->v4.sin_family = AF_INET; + addr->v4.sin_addr.s_addr = INADDR_ANY; + addr->v4.sin_port = port; +} + +/* Is this a wildcard address? */ +static int sctp_v4_is_any(const union sctp_addr *addr) +{ + return INADDR_ANY == addr->v4.sin_addr.s_addr; +} + +/* This function checks if the address is a valid address to be used for + * SCTP binding. + * + * Output: + * Return 0 - If the address is a non-unicast or an illegal address. + * Return 1 - If the address is a unicast. + */ +static int sctp_v4_addr_valid(union sctp_addr *addr, struct sctp_sock *sp) +{ + /* Is this a non-unicast address or a unusable SCTP address? */ + if (IS_IPV4_UNUSABLE_ADDRESS(&addr->v4.sin_addr.s_addr)) + return 0; + + return 1; +} + +/* Should this be available for binding? */ +static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp) +{ + int ret = inet_addr_type(addr->v4.sin_addr.s_addr); + + /* FIXME: ip_nonlocal_bind sysctl support. */ + + if (addr->v4.sin_addr.s_addr != INADDR_ANY && ret != RTN_LOCAL) + return 0; + return 1; +} + +/* Checking the loopback, private and other address scopes as defined in + * RFC 1918. The IPv4 scoping is based on the draft for SCTP IPv4 + * scoping <draft-stewart-tsvwg-sctp-ipv4-00.txt>. + * + * Level 0 - unusable SCTP addresses + * Level 1 - loopback address + * Level 2 - link-local addresses + * Level 3 - private addresses. + * Level 4 - global addresses + * For INIT and INIT-ACK address list, let L be the level of + * of requested destination address, sender and receiver + * SHOULD include all of its addresses with level greater + * than or equal to L. + */ +static sctp_scope_t sctp_v4_scope(union sctp_addr *addr) +{ + sctp_scope_t retval; + + /* Should IPv4 scoping be a sysctl configurable option + * so users can turn it off (default on) for certain + * unconventional networking environments? + */ + + /* Check for unusable SCTP addresses. */ + if (IS_IPV4_UNUSABLE_ADDRESS(&addr->v4.sin_addr.s_addr)) { + retval = SCTP_SCOPE_UNUSABLE; + } else if (LOOPBACK(addr->v4.sin_addr.s_addr)) { + retval = SCTP_SCOPE_LOOPBACK; + } else if (IS_IPV4_LINK_ADDRESS(&addr->v4.sin_addr.s_addr)) { + retval = SCTP_SCOPE_LINK; + } else if (IS_IPV4_PRIVATE_ADDRESS(&addr->v4.sin_addr.s_addr)) { + retval = SCTP_SCOPE_PRIVATE; + } else { + retval = SCTP_SCOPE_GLOBAL; + } + + return retval; +} + +/* Returns a valid dst cache entry for the given source and destination ip + * addresses. If an association is passed, trys to get a dst entry with a + * source address that matches an address in the bind address list. + */ +static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, + union sctp_addr *daddr, + union sctp_addr *saddr) +{ + struct rtable *rt; + struct flowi fl; + struct sctp_bind_addr *bp; + rwlock_t *addr_lock; + struct sctp_sockaddr_entry *laddr; + struct list_head *pos; + struct dst_entry *dst = NULL; + union sctp_addr dst_saddr; + + memset(&fl, 0x0, sizeof(struct flowi)); + fl.fl4_dst = daddr->v4.sin_addr.s_addr; + fl.proto = IPPROTO_SCTP; + if (asoc) { + fl.fl4_tos = RT_CONN_FLAGS(asoc->base.sk); + fl.oif = asoc->base.sk->sk_bound_dev_if; + } + if (saddr) + fl.fl4_src = saddr->v4.sin_addr.s_addr; + + SCTP_DEBUG_PRINTK("%s: DST:%u.%u.%u.%u, SRC:%u.%u.%u.%u - ", + __FUNCTION__, NIPQUAD(fl.fl4_dst), + NIPQUAD(fl.fl4_src)); + + if (!ip_route_output_key(&rt, &fl)) { + dst = &rt->u.dst; + } + + /* If there is no association or if a source address is passed, no + * more validation is required. + */ + if (!asoc || saddr) + goto out; + + bp = &asoc->base.bind_addr; + addr_lock = &asoc->base.addr_lock; + + if (dst) { + /* Walk through the bind address list and look for a bind + * address that matches the source address of the returned dst. + */ + sctp_read_lock(addr_lock); + list_for_each(pos, &bp->address_list) { + laddr = list_entry(pos, struct sctp_sockaddr_entry, + list); + sctp_v4_dst_saddr(&dst_saddr, dst, bp->port); + if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a)) + goto out_unlock; + } + sctp_read_unlock(addr_lock); + + /* None of the bound addresses match the source address of the + * dst. So release it. + */ + dst_release(dst); + dst = NULL; + } + + /* Walk through the bind address list and try to get a dst that + * matches a bind address as the source address. + */ + sctp_read_lock(addr_lock); + list_for_each(pos, &bp->address_list) { + laddr = list_entry(pos, struct sctp_sockaddr_entry, list); + + if (AF_INET == laddr->a.sa.sa_family) { + fl.fl4_src = laddr->a.v4.sin_addr.s_addr; + if (!ip_route_output_key(&rt, &fl)) { + dst = &rt->u.dst; + goto out_unlock; + } + } + } + +out_unlock: + sctp_read_unlock(addr_lock); +out: + if (dst) + SCTP_DEBUG_PRINTK("rt_dst:%u.%u.%u.%u, rt_src:%u.%u.%u.%u\n", + NIPQUAD(rt->rt_dst), NIPQUAD(rt->rt_src)); + else + SCTP_DEBUG_PRINTK("NO ROUTE\n"); + + return dst; +} + +/* For v4, the source address is cached in the route entry(dst). So no need + * to cache it separately and hence this is an empty routine. + */ +static void sctp_v4_get_saddr(struct sctp_association *asoc, + struct dst_entry *dst, + union sctp_addr *daddr, + union sctp_addr *saddr) +{ + struct rtable *rt = (struct rtable *)dst; + + if (rt) { + saddr->v4.sin_family = AF_INET; + saddr->v4.sin_port = asoc->base.bind_addr.port; + saddr->v4.sin_addr.s_addr = rt->rt_src; + } +} + +/* What interface did this skb arrive on? */ +static int sctp_v4_skb_iif(const struct sk_buff *skb) +{ + return ((struct rtable *)skb->dst)->rt_iif; +} + +/* Was this packet marked by Explicit Congestion Notification? */ +static int sctp_v4_is_ce(const struct sk_buff *skb) +{ + return INET_ECN_is_ce(skb->nh.iph->tos); +} + +/* Create and initialize a new sk for the socket returned by accept(). */ +static struct sock *sctp_v4_create_accept_sk(struct sock *sk, + struct sctp_association *asoc) +{ + struct inet_sock *inet = inet_sk(sk); + struct inet_sock *newinet; + struct sock *newsk = sk_alloc(PF_INET, GFP_KERNEL, sk->sk_prot, 1); + + if (!newsk) + goto out; + + sock_init_data(NULL, newsk); + + newsk->sk_type = SOCK_STREAM; + + newsk->sk_no_check = sk->sk_no_check; + newsk->sk_reuse = sk->sk_reuse; + newsk->sk_shutdown = sk->sk_shutdown; + + newsk->sk_destruct = inet_sock_destruct; + newsk->sk_family = PF_INET; + newsk->sk_protocol = IPPROTO_SCTP; + newsk->sk_backlog_rcv = sk->sk_prot->backlog_rcv; + sock_reset_flag(newsk, SOCK_ZAPPED); + + newinet = inet_sk(newsk); + + /* Initialize sk's sport, dport, rcv_saddr and daddr for + * getsockname() and getpeername() + */ + newinet->sport = inet->sport; + newinet->saddr = inet->saddr; + newinet->rcv_saddr = inet->rcv_saddr; + newinet->dport = htons(asoc->peer.port); + newinet->daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; + newinet->pmtudisc = inet->pmtudisc; + newinet->id = 0; + + newinet->uc_ttl = -1; + newinet->mc_loop = 1; + newinet->mc_ttl = 1; + newinet->mc_index = 0; + newinet->mc_list = NULL; + +#ifdef INET_REFCNT_DEBUG + atomic_inc(&inet_sock_nr); +#endif + + if (newsk->sk_prot->init(newsk)) { + sk_common_release(newsk); + newsk = NULL; + } + +out: + return newsk; +} + +/* Map address, empty for v4 family */ +static void sctp_v4_addr_v4map(struct sctp_sock *sp, union sctp_addr *addr) +{ + /* Empty */ +} + +/* Dump the v4 addr to the seq file. */ +static void sctp_v4_seq_dump_addr(struct seq_file *seq, union sctp_addr *addr) +{ + seq_printf(seq, "%d.%d.%d.%d ", NIPQUAD(addr->v4.sin_addr)); +} + +/* Event handler for inet address addition/deletion events. + * Basically, whenever there is an event, we re-build our local address list. + */ +int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev, + void *ptr) +{ + unsigned long flags; + + sctp_spin_lock_irqsave(&sctp_local_addr_lock, flags); + __sctp_free_local_addr_list(); + __sctp_get_local_addr_list(); + sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags); + + return NOTIFY_DONE; +} + +/* + * Initialize the control inode/socket with a control endpoint data + * structure. This endpoint is reserved exclusively for the OOTB processing. + */ +static int sctp_ctl_sock_init(void) +{ + int err; + sa_family_t family; + + if (sctp_get_pf_specific(PF_INET6)) + family = PF_INET6; + else + family = PF_INET; + + err = sock_create_kern(family, SOCK_SEQPACKET, IPPROTO_SCTP, + &sctp_ctl_socket); + if (err < 0) { + printk(KERN_ERR + "SCTP: Failed to create the SCTP control socket.\n"); + return err; + } + sctp_ctl_socket->sk->sk_allocation = GFP_ATOMIC; + inet_sk(sctp_ctl_socket->sk)->uc_ttl = -1; + + return 0; +} + +/* Register address family specific functions. */ +int sctp_register_af(struct sctp_af *af) +{ + switch (af->sa_family) { + case AF_INET: + if (sctp_af_v4_specific) + return 0; + sctp_af_v4_specific = af; + break; + case AF_INET6: + if (sctp_af_v6_specific) + return 0; + sctp_af_v6_specific = af; + break; + default: + return 0; + } + + INIT_LIST_HEAD(&af->list); + list_add_tail(&af->list, &sctp_address_families); + return 1; +} + +/* Get the table of functions for manipulating a particular address + * family. + */ +struct sctp_af *sctp_get_af_specific(sa_family_t family) +{ + switch (family) { + case AF_INET: + return sctp_af_v4_specific; + case AF_INET6: + return sctp_af_v6_specific; + default: + return NULL; + } +} + +/* Common code to initialize a AF_INET msg_name. */ +static void sctp_inet_msgname(char *msgname, int *addr_len) +{ + struct sockaddr_in *sin; + + sin = (struct sockaddr_in *)msgname; + *addr_len = sizeof(struct sockaddr_in); + sin->sin_family = AF_INET; + memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); +} + +/* Copy the primary address of the peer primary address as the msg_name. */ +static void sctp_inet_event_msgname(struct sctp_ulpevent *event, char *msgname, + int *addr_len) +{ + struct sockaddr_in *sin, *sinfrom; + + if (msgname) { + struct sctp_association *asoc; + + asoc = event->asoc; + sctp_inet_msgname(msgname, addr_len); + sin = (struct sockaddr_in *)msgname; + sinfrom = &asoc->peer.primary_addr.v4; + sin->sin_port = htons(asoc->peer.port); + sin->sin_addr.s_addr = sinfrom->sin_addr.s_addr; + } +} + +/* Initialize and copy out a msgname from an inbound skb. */ +static void sctp_inet_skb_msgname(struct sk_buff *skb, char *msgname, int *len) +{ + struct sctphdr *sh; + struct sockaddr_in *sin; + + if (msgname) { + sctp_inet_msgname(msgname, len); + sin = (struct sockaddr_in *)msgname; + sh = (struct sctphdr *)skb->h.raw; + sin->sin_port = sh->source; + sin->sin_addr.s_addr = skb->nh.iph->saddr; + } +} + +/* Do we support this AF? */ +static int sctp_inet_af_supported(sa_family_t family, struct sctp_sock *sp) +{ + /* PF_INET only supports AF_INET addresses. */ + return (AF_INET == family); +} + +/* Address matching with wildcards allowed. */ +static int sctp_inet_cmp_addr(const union sctp_addr *addr1, + const union sctp_addr *addr2, + struct sctp_sock *opt) +{ + /* PF_INET only supports AF_INET addresses. */ + if (addr1->sa.sa_family != addr2->sa.sa_family) + return 0; + if (INADDR_ANY == addr1->v4.sin_addr.s_addr || + INADDR_ANY == addr2->v4.sin_addr.s_addr) + return 1; + if (addr1->v4.sin_addr.s_addr == addr2->v4.sin_addr.s_addr) + return 1; + + return 0; +} + +/* Verify that provided sockaddr looks bindable. Common verification has + * already been taken care of. + */ +static int sctp_inet_bind_verify(struct sctp_sock *opt, union sctp_addr *addr) +{ + return sctp_v4_available(addr, opt); +} + +/* Verify that sockaddr looks sendable. Common verification has already + * been taken care of. + */ +static int sctp_inet_send_verify(struct sctp_sock *opt, union sctp_addr *addr) +{ + return 1; +} + +/* Fill in Supported Address Type information for INIT and INIT-ACK + * chunks. Returns number of addresses supported. + */ +static int sctp_inet_supported_addrs(const struct sctp_sock *opt, + __u16 *types) +{ + types[0] = SCTP_PARAM_IPV4_ADDRESS; + return 1; +} + +/* Wrapper routine that calls the ip transmit routine. */ +static inline int sctp_v4_xmit(struct sk_buff *skb, + struct sctp_transport *transport, int ipfragok) +{ + SCTP_DEBUG_PRINTK("%s: skb:%p, len:%d, " + "src:%u.%u.%u.%u, dst:%u.%u.%u.%u\n", + __FUNCTION__, skb, skb->len, + NIPQUAD(((struct rtable *)skb->dst)->rt_src), + NIPQUAD(((struct rtable *)skb->dst)->rt_dst)); + + SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS); + return ip_queue_xmit(skb, ipfragok); +} + +static struct sctp_af sctp_ipv4_specific; + +static struct sctp_pf sctp_pf_inet = { + .event_msgname = sctp_inet_event_msgname, + .skb_msgname = sctp_inet_skb_msgname, + .af_supported = sctp_inet_af_supported, + .cmp_addr = sctp_inet_cmp_addr, + .bind_verify = sctp_inet_bind_verify, + .send_verify = sctp_inet_send_verify, + .supported_addrs = sctp_inet_supported_addrs, + .create_accept_sk = sctp_v4_create_accept_sk, + .addr_v4map = sctp_v4_addr_v4map, + .af = &sctp_ipv4_specific, +}; + +/* Notifier for inetaddr addition/deletion events. */ +static struct notifier_block sctp_inetaddr_notifier = { + .notifier_call = sctp_inetaddr_event, +}; + +/* Socket operations. */ +static struct proto_ops inet_seqpacket_ops = { + .family = PF_INET, + .owner = THIS_MODULE, + .release = inet_release, /* Needs to be wrapped... */ + .bind = inet_bind, + .connect = inet_dgram_connect, + .socketpair = sock_no_socketpair, + .accept = inet_accept, + .getname = inet_getname, /* Semantics are different. */ + .poll = sctp_poll, + .ioctl = inet_ioctl, + .listen = sctp_inet_listen, + .shutdown = inet_shutdown, /* Looks harmless. */ + .setsockopt = sock_common_setsockopt, /* IP_SOL IP_OPTION is a problem. */ + .getsockopt = sock_common_getsockopt, + .sendmsg = inet_sendmsg, + .recvmsg = sock_common_recvmsg, + .mmap = sock_no_mmap, + .sendpage = sock_no_sendpage, +}; + +/* Registration with AF_INET family. */ +static struct inet_protosw sctp_seqpacket_protosw = { + .type = SOCK_SEQPACKET, + .protocol = IPPROTO_SCTP, + .prot = &sctp_prot, + .ops = &inet_seqpacket_ops, + .capability = -1, + .no_check = 0, + .flags = SCTP_PROTOSW_FLAG +}; +static struct inet_protosw sctp_stream_protosw = { + .type = SOCK_STREAM, + .protocol = IPPROTO_SCTP, + .prot = &sctp_prot, + .ops = &inet_seqpacket_ops, + .capability = -1, + .no_check = 0, + .flags = SCTP_PROTOSW_FLAG +}; + +/* Register with IP layer. */ +static struct net_protocol sctp_protocol = { + .handler = sctp_rcv, + .err_handler = sctp_v4_err, + .no_policy = 1, +}; + +/* IPv4 address related functions. */ +static struct sctp_af sctp_ipv4_specific = { + .sctp_xmit = sctp_v4_xmit, + .setsockopt = ip_setsockopt, + .getsockopt = ip_getsockopt, + .get_dst = sctp_v4_get_dst, + .get_saddr = sctp_v4_get_saddr, + .copy_addrlist = sctp_v4_copy_addrlist, + .from_skb = sctp_v4_from_skb, + .from_sk = sctp_v4_from_sk, + .to_sk_saddr = sctp_v4_to_sk_saddr, + .to_sk_daddr = sctp_v4_to_sk_daddr, + .from_addr_param= sctp_v4_from_addr_param, + .to_addr_param = sctp_v4_to_addr_param, + .dst_saddr = sctp_v4_dst_saddr, + .cmp_addr = sctp_v4_cmp_addr, + .addr_valid = sctp_v4_addr_valid, + .inaddr_any = sctp_v4_inaddr_any, + .is_any = sctp_v4_is_any, + .available = sctp_v4_available, + .scope = sctp_v4_scope, + .skb_iif = sctp_v4_skb_iif, + .is_ce = sctp_v4_is_ce, + .seq_dump_addr = sctp_v4_seq_dump_addr, + .net_header_len = sizeof(struct iphdr), + .sockaddr_len = sizeof(struct sockaddr_in), + .sa_family = AF_INET, +}; + +struct sctp_pf *sctp_get_pf_specific(sa_family_t family) { + + switch (family) { + case PF_INET: + return sctp_pf_inet_specific; + case PF_INET6: + return sctp_pf_inet6_specific; + default: + return NULL; + } +} + +/* Register the PF specific function table. */ +int sctp_register_pf(struct sctp_pf *pf, sa_family_t family) +{ + switch (family) { + case PF_INET: + if (sctp_pf_inet_specific) + return 0; + sctp_pf_inet_specific = pf; + break; + case PF_INET6: + if (sctp_pf_inet6_specific) + return 0; + sctp_pf_inet6_specific = pf; + break; + default: + return 0; + } + return 1; +} + +static int __init init_sctp_mibs(void) +{ + sctp_statistics[0] = alloc_percpu(struct sctp_mib); + if (!sctp_statistics[0]) + return -ENOMEM; + sctp_statistics[1] = alloc_percpu(struct sctp_mib); + if (!sctp_statistics[1]) { + free_percpu(sctp_statistics[0]); + return -ENOMEM; + } + return 0; + +} + +static void cleanup_sctp_mibs(void) +{ + free_percpu(sctp_statistics[0]); + free_percpu(sctp_statistics[1]); +} + +/* Initialize the universe into something sensible. */ +SCTP_STATIC __init int sctp_init(void) +{ + int i; + int status = -EINVAL; + unsigned long goal; + int order; + + /* SCTP_DEBUG sanity check. */ + if (!sctp_sanity_check()) + goto out; + + status = proto_register(&sctp_prot, 1); + if (status) + goto out; + + /* Add SCTP to inet_protos hash table. */ + status = -EAGAIN; + if (inet_add_protocol(&sctp_protocol, IPPROTO_SCTP) < 0) + goto err_add_protocol; + + /* Add SCTP(TCP and UDP style) to inetsw linked list. */ + inet_register_protosw(&sctp_seqpacket_protosw); + inet_register_protosw(&sctp_stream_protosw); + + /* Allocate a cache pools. */ + status = -ENOBUFS; + sctp_bucket_cachep = kmem_cache_create("sctp_bind_bucket", + sizeof(struct sctp_bind_bucket), + 0, SLAB_HWCACHE_ALIGN, + NULL, NULL); + + if (!sctp_bucket_cachep) + goto err_bucket_cachep; + + sctp_chunk_cachep = kmem_cache_create("sctp_chunk", + sizeof(struct sctp_chunk), + 0, SLAB_HWCACHE_ALIGN, + NULL, NULL); + if (!sctp_chunk_cachep) + goto err_chunk_cachep; + + /* Allocate and initialise sctp mibs. */ + status = init_sctp_mibs(); + if (status) + goto err_init_mibs; + + /* Initialize proc fs directory. */ + status = sctp_proc_init(); + if (status) + goto err_init_proc; + + /* Initialize object count debugging. */ + sctp_dbg_objcnt_init(); + + /* Initialize the SCTP specific PF functions. */ + sctp_register_pf(&sctp_pf_inet, PF_INET); + /* + * 14. Suggested SCTP Protocol Parameter Values + */ + /* The following protocol parameters are RECOMMENDED: */ + /* RTO.Initial - 3 seconds */ + sctp_rto_initial = SCTP_RTO_INITIAL; + /* RTO.Min - 1 second */ + sctp_rto_min = SCTP_RTO_MIN; + /* RTO.Max - 60 seconds */ + sctp_rto_max = SCTP_RTO_MAX; + /* RTO.Alpha - 1/8 */ + sctp_rto_alpha = SCTP_RTO_ALPHA; + /* RTO.Beta - 1/4 */ + sctp_rto_beta = SCTP_RTO_BETA; + + /* Valid.Cookie.Life - 60 seconds */ + sctp_valid_cookie_life = 60 * HZ; + + /* Whether Cookie Preservative is enabled(1) or not(0) */ + sctp_cookie_preserve_enable = 1; + + /* Max.Burst - 4 */ + sctp_max_burst = SCTP_MAX_BURST; + + /* Association.Max.Retrans - 10 attempts + * Path.Max.Retrans - 5 attempts (per destination address) + * Max.Init.Retransmits - 8 attempts + */ + sctp_max_retrans_association = 10; + sctp_max_retrans_path = 5; + sctp_max_retrans_init = 8; + + /* HB.interval - 30 seconds */ + sctp_hb_interval = 30 * HZ; + + /* Implementation specific variables. */ + + /* Initialize default stream count setup information. */ + sctp_max_instreams = SCTP_DEFAULT_INSTREAMS; + sctp_max_outstreams = SCTP_DEFAULT_OUTSTREAMS; + + /* Initialize handle used for association ids. */ + idr_init(&sctp_assocs_id); + + /* Size and allocate the association hash table. + * The methodology is similar to that of the tcp hash tables. + */ + if (num_physpages >= (128 * 1024)) + goal = num_physpages >> (22 - PAGE_SHIFT); + else + goal = num_physpages >> (24 - PAGE_SHIFT); + + for (order = 0; (1UL << order) < goal; order++) + ; + + do { + sctp_assoc_hashsize = (1UL << order) * PAGE_SIZE / + sizeof(struct sctp_hashbucket); + if ((sctp_assoc_hashsize > (64 * 1024)) && order > 0) + continue; + sctp_assoc_hashtable = (struct sctp_hashbucket *) + __get_free_pages(GFP_ATOMIC, order); + } while (!sctp_assoc_hashtable && --order > 0); + if (!sctp_assoc_hashtable) { + printk(KERN_ERR "SCTP: Failed association hash alloc.\n"); + status = -ENOMEM; + goto err_ahash_alloc; + } + for (i = 0; i < sctp_assoc_hashsize; i++) { + rwlock_init(&sctp_assoc_hashtable[i].lock); + sctp_assoc_hashtable[i].chain = NULL; + } + + /* Allocate and initialize the endpoint hash table. */ + sctp_ep_hashsize = 64; + sctp_ep_hashtable = (struct sctp_hashbucket *) + kmalloc(64 * sizeof(struct sctp_hashbucket), GFP_KERNEL); + if (!sctp_ep_hashtable) { + printk(KERN_ERR "SCTP: Failed endpoint_hash alloc.\n"); + status = -ENOMEM; + goto err_ehash_alloc; + } + for (i = 0; i < sctp_ep_hashsize; i++) { + rwlock_init(&sctp_ep_hashtable[i].lock); + sctp_ep_hashtable[i].chain = NULL; + } + + /* Allocate and initialize the SCTP port hash table. */ + do { + sctp_port_hashsize = (1UL << order) * PAGE_SIZE / + sizeof(struct sctp_bind_hashbucket); + if ((sctp_port_hashsize > (64 * 1024)) && order > 0) + continue; + sctp_port_hashtable = (struct sctp_bind_hashbucket *) + __get_free_pages(GFP_ATOMIC, order); + } while (!sctp_port_hashtable && --order > 0); + if (!sctp_port_hashtable) { + printk(KERN_ERR "SCTP: Failed bind hash alloc."); + status = -ENOMEM; + goto err_bhash_alloc; + } + for (i = 0; i < sctp_port_hashsize; i++) { + spin_lock_init(&sctp_port_hashtable[i].lock); + sctp_port_hashtable[i].chain = NULL; + } + + spin_lock_init(&sctp_port_alloc_lock); + sctp_port_rover = sysctl_local_port_range[0] - 1; + + printk(KERN_INFO "SCTP: Hash tables configured " + "(established %d bind %d)\n", + sctp_assoc_hashsize, sctp_port_hashsize); + + /* Disable ADDIP by default. */ + sctp_addip_enable = 0; + + /* Enable PR-SCTP by default. */ + sctp_prsctp_enable = 1; + + sctp_sysctl_register(); + + INIT_LIST_HEAD(&sctp_address_families); + sctp_register_af(&sctp_ipv4_specific); + + status = sctp_v6_init(); + if (status) + goto err_v6_init; + + /* Initialize the control inode/socket for handling OOTB packets. */ + if ((status = sctp_ctl_sock_init())) { + printk (KERN_ERR + "SCTP: Failed to initialize the SCTP control sock.\n"); + goto err_ctl_sock_init; + } + + /* Initialize the local address list. */ + INIT_LIST_HEAD(&sctp_local_addr_list); + spin_lock_init(&sctp_local_addr_lock); + + /* Register notifier for inet address additions/deletions. */ + register_inetaddr_notifier(&sctp_inetaddr_notifier); + + sctp_get_local_addr_list(); + + __unsafe(THIS_MODULE); + status = 0; +out: + return status; +err_add_protocol: + proto_unregister(&sctp_prot); +err_ctl_sock_init: + sctp_v6_exit(); +err_v6_init: + sctp_sysctl_unregister(); + list_del(&sctp_ipv4_specific.list); + free_pages((unsigned long)sctp_port_hashtable, + get_order(sctp_port_hashsize * + sizeof(struct sctp_bind_hashbucket))); +err_bhash_alloc: + kfree(sctp_ep_hashtable); +err_ehash_alloc: + free_pages((unsigned long)sctp_assoc_hashtable, + get_order(sctp_assoc_hashsize * + sizeof(struct sctp_hashbucket))); +err_ahash_alloc: + sctp_dbg_objcnt_exit(); +err_init_proc: + sctp_proc_exit(); + cleanup_sctp_mibs(); +err_init_mibs: + kmem_cache_destroy(sctp_chunk_cachep); +err_chunk_cachep: + kmem_cache_destroy(sctp_bucket_cachep); +err_bucket_cachep: + inet_del_protocol(&sctp_protocol, IPPROTO_SCTP); + inet_unregister_protosw(&sctp_seqpacket_protosw); + inet_unregister_protosw(&sctp_stream_protosw); + goto out; +} + +/* Exit handler for the SCTP protocol. */ +SCTP_STATIC __exit void sctp_exit(void) +{ + /* BUG. This should probably do something useful like clean + * up all the remaining associations and all that memory. + */ + + /* Unregister notifier for inet address additions/deletions. */ + unregister_inetaddr_notifier(&sctp_inetaddr_notifier); + + /* Free the local address list. */ + sctp_free_local_addr_list(); + + /* Free the control endpoint. */ + sock_release(sctp_ctl_socket); + + sctp_v6_exit(); + sctp_sysctl_unregister(); + list_del(&sctp_ipv4_specific.list); + + free_pages((unsigned long)sctp_assoc_hashtable, + get_order(sctp_assoc_hashsize * + sizeof(struct sctp_hashbucket))); + kfree(sctp_ep_hashtable); + free_pages((unsigned long)sctp_port_hashtable, + get_order(sctp_port_hashsize * + sizeof(struct sctp_bind_hashbucket))); + + kmem_cache_destroy(sctp_chunk_cachep); + kmem_cache_destroy(sctp_bucket_cachep); + + sctp_dbg_objcnt_exit(); + sctp_proc_exit(); + cleanup_sctp_mibs(); + + inet_del_protocol(&sctp_protocol, IPPROTO_SCTP); + inet_unregister_protosw(&sctp_seqpacket_protosw); + inet_unregister_protosw(&sctp_stream_protosw); + proto_unregister(&sctp_prot); +} + +module_init(sctp_init); +module_exit(sctp_exit); + +MODULE_AUTHOR("Linux Kernel SCTP developers <lksctp-developers@lists.sourceforge.net>"); +MODULE_DESCRIPTION("Support for the SCTP protocol (RFC2960)"); +MODULE_LICENSE("GPL"); diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c new file mode 100644 index 000000000000..1db12cc18cf7 --- /dev/null +++ b/net/sctp/sm_make_chunk.c @@ -0,0 +1,2766 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001-2002 Intel Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions work with the state functions in sctp_sm_statefuns.c + * to implement the state operations. These functions implement the + * steps which require modifying existing data structures. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * C. Robin <chris@hundredacre.ac.uk> + * Jon Grimm <jgrimm@us.ibm.com> + * Xingang Guo <xingang.guo@intel.com> + * Dajiang Zhang <dajiang.zhang@nokia.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Daisy Chang <daisyc@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * Kevin Gao <kevin.gao@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/net.h> +#include <linux/inet.h> +#include <asm/scatterlist.h> +#include <linux/crypto.h> +#include <net/sock.h> + +#include <linux/skbuff.h> +#include <linux/random.h> /* for get_random_bytes */ +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +extern kmem_cache_t *sctp_chunk_cachep; + +SCTP_STATIC +struct sctp_chunk *sctp_make_chunk(const struct sctp_association *asoc, + __u8 type, __u8 flags, int paylen); +static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const struct sctp_chunk *init_chunk, + int *cookie_len, + const __u8 *raw_addrs, int addrs_len); +static int sctp_process_param(struct sctp_association *asoc, + union sctp_params param, + const union sctp_addr *peer_addr, + int gfp); + +/* What was the inbound interface for this chunk? */ +int sctp_chunk_iif(const struct sctp_chunk *chunk) +{ + struct sctp_af *af; + int iif = 0; + + af = sctp_get_af_specific(ipver2af(chunk->skb->nh.iph->version)); + if (af) + iif = af->skb_iif(chunk->skb); + + return iif; +} + +/* RFC 2960 3.3.2 Initiation (INIT) (1) + * + * Note 2: The ECN capable field is reserved for future use of + * Explicit Congestion Notification. + */ +static const struct sctp_paramhdr ecap_param = { + SCTP_PARAM_ECN_CAPABLE, + __constant_htons(sizeof(struct sctp_paramhdr)), +}; +static const struct sctp_paramhdr prsctp_param = { + SCTP_PARAM_FWD_TSN_SUPPORT, + __constant_htons(sizeof(struct sctp_paramhdr)), +}; + +/* A helper to initialize to initialize an op error inside a + * provided chunk, as most cause codes will be embedded inside an + * abort chunk. + */ +void sctp_init_cause(struct sctp_chunk *chunk, __u16 cause_code, + const void *payload, size_t paylen) +{ + sctp_errhdr_t err; + int padlen; + __u16 len; + + /* Cause code constants are now defined in network order. */ + err.cause = cause_code; + len = sizeof(sctp_errhdr_t) + paylen; + padlen = len % 4; + err.length = htons(len); + len += padlen; + sctp_addto_chunk(chunk, sizeof(sctp_errhdr_t), &err); + chunk->subh.err_hdr = sctp_addto_chunk(chunk, paylen, payload); +} + +/* 3.3.2 Initiation (INIT) (1) + * + * This chunk is used to initiate a SCTP association between two + * endpoints. The format of the INIT chunk is shown below: + * + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 1 | Chunk Flags | Chunk Length | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Initiate Tag | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Advertised Receiver Window Credit (a_rwnd) | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Number of Outbound Streams | Number of Inbound Streams | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Initial TSN | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * \ \ + * / Optional/Variable-Length Parameters / + * \ \ + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * + * The INIT chunk contains the following parameters. Unless otherwise + * noted, each parameter MUST only be included once in the INIT chunk. + * + * Fixed Parameters Status + * ---------------------------------------------- + * Initiate Tag Mandatory + * Advertised Receiver Window Credit Mandatory + * Number of Outbound Streams Mandatory + * Number of Inbound Streams Mandatory + * Initial TSN Mandatory + * + * Variable Parameters Status Type Value + * ------------------------------------------------------------- + * IPv4 Address (Note 1) Optional 5 + * IPv6 Address (Note 1) Optional 6 + * Cookie Preservative Optional 9 + * Reserved for ECN Capable (Note 2) Optional 32768 (0x8000) + * Host Name Address (Note 3) Optional 11 + * Supported Address Types (Note 4) Optional 12 + */ +struct sctp_chunk *sctp_make_init(const struct sctp_association *asoc, + const struct sctp_bind_addr *bp, + int gfp, int vparam_len) +{ + sctp_inithdr_t init; + union sctp_params addrs; + size_t chunksize; + struct sctp_chunk *retval = NULL; + int num_types, addrs_len = 0; + struct sctp_sock *sp; + sctp_supported_addrs_param_t sat; + __u16 types[2]; + sctp_adaption_ind_param_t aiparam; + + /* RFC 2960 3.3.2 Initiation (INIT) (1) + * + * Note 1: The INIT chunks can contain multiple addresses that + * can be IPv4 and/or IPv6 in any combination. + */ + retval = NULL; + + /* Convert the provided bind address list to raw format. */ + addrs = sctp_bind_addrs_to_raw(bp, &addrs_len, gfp); + + init.init_tag = htonl(asoc->c.my_vtag); + init.a_rwnd = htonl(asoc->rwnd); + init.num_outbound_streams = htons(asoc->c.sinit_num_ostreams); + init.num_inbound_streams = htons(asoc->c.sinit_max_instreams); + init.initial_tsn = htonl(asoc->c.initial_tsn); + + /* How many address types are needed? */ + sp = sctp_sk(asoc->base.sk); + num_types = sp->pf->supported_addrs(sp, types); + + chunksize = sizeof(init) + addrs_len + SCTP_SAT_LEN(num_types); + chunksize += sizeof(ecap_param); + if (sctp_prsctp_enable) + chunksize += sizeof(prsctp_param); + chunksize += sizeof(aiparam); + chunksize += vparam_len; + + /* RFC 2960 3.3.2 Initiation (INIT) (1) + * + * Note 3: An INIT chunk MUST NOT contain more than one Host + * Name address parameter. Moreover, the sender of the INIT + * MUST NOT combine any other address types with the Host Name + * address in the INIT. The receiver of INIT MUST ignore any + * other address types if the Host Name address parameter is + * present in the received INIT chunk. + * + * PLEASE DO NOT FIXME [This version does not support Host Name.] + */ + + retval = sctp_make_chunk(asoc, SCTP_CID_INIT, 0, chunksize); + if (!retval) + goto nodata; + + retval->subh.init_hdr = + sctp_addto_chunk(retval, sizeof(init), &init); + retval->param_hdr.v = + sctp_addto_chunk(retval, addrs_len, addrs.v); + + /* RFC 2960 3.3.2 Initiation (INIT) (1) + * + * Note 4: This parameter, when present, specifies all the + * address types the sending endpoint can support. The absence + * of this parameter indicates that the sending endpoint can + * support any address type. + */ + sat.param_hdr.type = SCTP_PARAM_SUPPORTED_ADDRESS_TYPES; + sat.param_hdr.length = htons(SCTP_SAT_LEN(num_types)); + sctp_addto_chunk(retval, sizeof(sat), &sat); + sctp_addto_chunk(retval, num_types * sizeof(__u16), &types); + + sctp_addto_chunk(retval, sizeof(ecap_param), &ecap_param); + if (sctp_prsctp_enable) + sctp_addto_chunk(retval, sizeof(prsctp_param), &prsctp_param); + aiparam.param_hdr.type = SCTP_PARAM_ADAPTION_LAYER_IND; + aiparam.param_hdr.length = htons(sizeof(aiparam)); + aiparam.adaption_ind = htonl(sp->adaption_ind); + sctp_addto_chunk(retval, sizeof(aiparam), &aiparam); +nodata: + if (addrs.v) + kfree(addrs.v); + return retval; +} + +struct sctp_chunk *sctp_make_init_ack(const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + int gfp, int unkparam_len) +{ + sctp_inithdr_t initack; + struct sctp_chunk *retval; + union sctp_params addrs; + int addrs_len; + sctp_cookie_param_t *cookie; + int cookie_len; + size_t chunksize; + sctp_adaption_ind_param_t aiparam; + + retval = NULL; + + /* Note: there may be no addresses to embed. */ + addrs = sctp_bind_addrs_to_raw(&asoc->base.bind_addr, &addrs_len, gfp); + + initack.init_tag = htonl(asoc->c.my_vtag); + initack.a_rwnd = htonl(asoc->rwnd); + initack.num_outbound_streams = htons(asoc->c.sinit_num_ostreams); + initack.num_inbound_streams = htons(asoc->c.sinit_max_instreams); + initack.initial_tsn = htonl(asoc->c.initial_tsn); + + /* FIXME: We really ought to build the cookie right + * into the packet instead of allocating more fresh memory. + */ + cookie = sctp_pack_cookie(asoc->ep, asoc, chunk, &cookie_len, + addrs.v, addrs_len); + if (!cookie) + goto nomem_cookie; + + /* Calculate the total size of allocation, include the reserved + * space for reporting unknown parameters if it is specified. + */ + chunksize = sizeof(initack) + addrs_len + cookie_len + unkparam_len; + + /* Tell peer that we'll do ECN only if peer advertised such cap. */ + if (asoc->peer.ecn_capable) + chunksize += sizeof(ecap_param); + + /* Tell peer that we'll do PR-SCTP only if peer advertised. */ + if (asoc->peer.prsctp_capable) + chunksize += sizeof(prsctp_param); + + chunksize += sizeof(aiparam); + + /* Now allocate and fill out the chunk. */ + retval = sctp_make_chunk(asoc, SCTP_CID_INIT_ACK, 0, chunksize); + if (!retval) + goto nomem_chunk; + + /* Per the advice in RFC 2960 6.4, send this reply to + * the source of the INIT packet. + */ + retval->transport = chunk->transport; + retval->subh.init_hdr = + sctp_addto_chunk(retval, sizeof(initack), &initack); + retval->param_hdr.v = sctp_addto_chunk(retval, addrs_len, addrs.v); + sctp_addto_chunk(retval, cookie_len, cookie); + if (asoc->peer.ecn_capable) + sctp_addto_chunk(retval, sizeof(ecap_param), &ecap_param); + if (asoc->peer.prsctp_capable) + sctp_addto_chunk(retval, sizeof(prsctp_param), &prsctp_param); + + aiparam.param_hdr.type = SCTP_PARAM_ADAPTION_LAYER_IND; + aiparam.param_hdr.length = htons(sizeof(aiparam)); + aiparam.adaption_ind = htonl(sctp_sk(asoc->base.sk)->adaption_ind); + sctp_addto_chunk(retval, sizeof(aiparam), &aiparam); + + /* We need to remove the const qualifier at this point. */ + retval->asoc = (struct sctp_association *) asoc; + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it received the DATA or control chunk + * to which it is replying. + * + * [INIT ACK back to where the INIT came from.] + */ + if (chunk) + retval->transport = chunk->transport; + +nomem_chunk: + kfree(cookie); +nomem_cookie: + if (addrs.v) + kfree(addrs.v); + return retval; +} + +/* 3.3.11 Cookie Echo (COOKIE ECHO) (10): + * + * This chunk is used only during the initialization of an association. + * It is sent by the initiator of an association to its peer to complete + * the initialization process. This chunk MUST precede any DATA chunk + * sent within the association, but MAY be bundled with one or more DATA + * chunks in the same packet. + * + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 10 |Chunk Flags | Length | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * / Cookie / + * \ \ + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * Chunk Flags: 8 bit + * + * Set to zero on transmit and ignored on receipt. + * + * Length: 16 bits (unsigned integer) + * + * Set to the size of the chunk in bytes, including the 4 bytes of + * the chunk header and the size of the Cookie. + * + * Cookie: variable size + * + * This field must contain the exact cookie received in the + * State Cookie parameter from the previous INIT ACK. + * + * An implementation SHOULD make the cookie as small as possible + * to insure interoperability. + */ +struct sctp_chunk *sctp_make_cookie_echo(const struct sctp_association *asoc, + const struct sctp_chunk *chunk) +{ + struct sctp_chunk *retval; + void *cookie; + int cookie_len; + + cookie = asoc->peer.cookie; + cookie_len = asoc->peer.cookie_len; + + /* Build a cookie echo chunk. */ + retval = sctp_make_chunk(asoc, SCTP_CID_COOKIE_ECHO, 0, cookie_len); + if (!retval) + goto nodata; + retval->subh.cookie_hdr = + sctp_addto_chunk(retval, cookie_len, cookie); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [COOKIE ECHO back to where the INIT ACK came from.] + */ + if (chunk) + retval->transport = chunk->transport; + +nodata: + return retval; +} + +/* 3.3.12 Cookie Acknowledgement (COOKIE ACK) (11): + * + * This chunk is used only during the initialization of an + * association. It is used to acknowledge the receipt of a COOKIE + * ECHO chunk. This chunk MUST precede any DATA or SACK chunk sent + * within the association, but MAY be bundled with one or more DATA + * chunks or SACK chunk in the same SCTP packet. + * + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 11 |Chunk Flags | Length = 4 | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * Chunk Flags: 8 bits + * + * Set to zero on transmit and ignored on receipt. + */ +struct sctp_chunk *sctp_make_cookie_ack(const struct sctp_association *asoc, + const struct sctp_chunk *chunk) +{ + struct sctp_chunk *retval; + + retval = sctp_make_chunk(asoc, SCTP_CID_COOKIE_ACK, 0, 0); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [COOKIE ACK back to where the COOKIE ECHO came from.] + */ + if (retval && chunk) + retval->transport = chunk->transport; + + return retval; +} + +/* + * Appendix A: Explicit Congestion Notification: + * CWR: + * + * RFC 2481 details a specific bit for a sender to send in the header of + * its next outbound TCP segment to indicate to its peer that it has + * reduced its congestion window. This is termed the CWR bit. For + * SCTP the same indication is made by including the CWR chunk. + * This chunk contains one data element, i.e. the TSN number that + * was sent in the ECNE chunk. This element represents the lowest + * TSN number in the datagram that was originally marked with the + * CE bit. + * + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Chunk Type=13 | Flags=00000000| Chunk Length = 8 | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Lowest TSN Number | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * Note: The CWR is considered a Control chunk. + */ +struct sctp_chunk *sctp_make_cwr(const struct sctp_association *asoc, + const __u32 lowest_tsn, + const struct sctp_chunk *chunk) +{ + struct sctp_chunk *retval; + sctp_cwrhdr_t cwr; + + cwr.lowest_tsn = htonl(lowest_tsn); + retval = sctp_make_chunk(asoc, SCTP_CID_ECN_CWR, 0, + sizeof(sctp_cwrhdr_t)); + + if (!retval) + goto nodata; + + retval->subh.ecn_cwr_hdr = + sctp_addto_chunk(retval, sizeof(cwr), &cwr); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [Report a reduced congestion window back to where the ECNE + * came from.] + */ + if (chunk) + retval->transport = chunk->transport; + +nodata: + return retval; +} + +/* Make an ECNE chunk. This is a congestion experienced report. */ +struct sctp_chunk *sctp_make_ecne(const struct sctp_association *asoc, + const __u32 lowest_tsn) +{ + struct sctp_chunk *retval; + sctp_ecnehdr_t ecne; + + ecne.lowest_tsn = htonl(lowest_tsn); + retval = sctp_make_chunk(asoc, SCTP_CID_ECN_ECNE, 0, + sizeof(sctp_ecnehdr_t)); + if (!retval) + goto nodata; + retval->subh.ecne_hdr = + sctp_addto_chunk(retval, sizeof(ecne), &ecne); + +nodata: + return retval; +} + +/* Make a DATA chunk for the given association from the provided + * parameters. However, do not populate the data payload. + */ +struct sctp_chunk *sctp_make_datafrag_empty(struct sctp_association *asoc, + const struct sctp_sndrcvinfo *sinfo, + int data_len, __u8 flags, __u16 ssn) +{ + struct sctp_chunk *retval; + struct sctp_datahdr dp; + int chunk_len; + + /* We assign the TSN as LATE as possible, not here when + * creating the chunk. + */ + dp.tsn = 0; + dp.stream = htons(sinfo->sinfo_stream); + dp.ppid = sinfo->sinfo_ppid; + + /* Set the flags for an unordered send. */ + if (sinfo->sinfo_flags & MSG_UNORDERED) { + flags |= SCTP_DATA_UNORDERED; + dp.ssn = 0; + } else + dp.ssn = htons(ssn); + + chunk_len = sizeof(dp) + data_len; + retval = sctp_make_chunk(asoc, SCTP_CID_DATA, flags, chunk_len); + if (!retval) + goto nodata; + + retval->subh.data_hdr = sctp_addto_chunk(retval, sizeof(dp), &dp); + memcpy(&retval->sinfo, sinfo, sizeof(struct sctp_sndrcvinfo)); + +nodata: + return retval; +} + +/* Create a selective ackowledgement (SACK) for the given + * association. This reports on which TSN's we've seen to date, + * including duplicates and gaps. + */ +struct sctp_chunk *sctp_make_sack(const struct sctp_association *asoc) +{ + struct sctp_chunk *retval; + struct sctp_sackhdr sack; + int len; + __u32 ctsn; + __u16 num_gabs, num_dup_tsns; + struct sctp_tsnmap *map = (struct sctp_tsnmap *)&asoc->peer.tsn_map; + + ctsn = sctp_tsnmap_get_ctsn(map); + SCTP_DEBUG_PRINTK("sackCTSNAck sent: 0x%x.\n", ctsn); + + /* How much room is needed in the chunk? */ + num_gabs = sctp_tsnmap_num_gabs(map); + num_dup_tsns = sctp_tsnmap_num_dups(map); + + /* Initialize the SACK header. */ + sack.cum_tsn_ack = htonl(ctsn); + sack.a_rwnd = htonl(asoc->a_rwnd); + sack.num_gap_ack_blocks = htons(num_gabs); + sack.num_dup_tsns = htons(num_dup_tsns); + + len = sizeof(sack) + + sizeof(struct sctp_gap_ack_block) * num_gabs + + sizeof(__u32) * num_dup_tsns; + + /* Create the chunk. */ + retval = sctp_make_chunk(asoc, SCTP_CID_SACK, 0, len); + if (!retval) + goto nodata; + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, etc.) to the same destination transport + * address from which it received the DATA or control chunk to + * which it is replying. This rule should also be followed if + * the endpoint is bundling DATA chunks together with the + * reply chunk. + * + * However, when acknowledging multiple DATA chunks received + * in packets from different source addresses in a single + * SACK, the SACK chunk may be transmitted to one of the + * destination transport addresses from which the DATA or + * control chunks being acknowledged were received. + * + * [BUG: We do not implement the following paragraph. + * Perhaps we should remember the last transport we used for a + * SACK and avoid that (if possible) if we have seen any + * duplicates. --piggy] + * + * When a receiver of a duplicate DATA chunk sends a SACK to a + * multi- homed endpoint it MAY be beneficial to vary the + * destination address and not use the source address of the + * DATA chunk. The reason being that receiving a duplicate + * from a multi-homed endpoint might indicate that the return + * path (as specified in the source address of the DATA chunk) + * for the SACK is broken. + * + * [Send to the address from which we last received a DATA chunk.] + */ + retval->transport = asoc->peer.last_data_from; + + retval->subh.sack_hdr = + sctp_addto_chunk(retval, sizeof(sack), &sack); + + /* Add the gap ack block information. */ + if (num_gabs) + sctp_addto_chunk(retval, sizeof(__u32) * num_gabs, + sctp_tsnmap_get_gabs(map)); + + /* Add the duplicate TSN information. */ + if (num_dup_tsns) + sctp_addto_chunk(retval, sizeof(__u32) * num_dup_tsns, + sctp_tsnmap_get_dups(map)); + +nodata: + return retval; +} + +/* Make a SHUTDOWN chunk. */ +struct sctp_chunk *sctp_make_shutdown(const struct sctp_association *asoc, + const struct sctp_chunk *chunk) +{ + struct sctp_chunk *retval; + sctp_shutdownhdr_t shut; + __u32 ctsn; + + ctsn = sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map); + shut.cum_tsn_ack = htonl(ctsn); + + retval = sctp_make_chunk(asoc, SCTP_CID_SHUTDOWN, 0, + sizeof(sctp_shutdownhdr_t)); + if (!retval) + goto nodata; + + retval->subh.shutdown_hdr = + sctp_addto_chunk(retval, sizeof(shut), &shut); + + if (chunk) + retval->transport = chunk->transport; +nodata: + return retval; +} + +struct sctp_chunk *sctp_make_shutdown_ack(const struct sctp_association *asoc, + const struct sctp_chunk *chunk) +{ + struct sctp_chunk *retval; + + retval = sctp_make_chunk(asoc, SCTP_CID_SHUTDOWN_ACK, 0, 0); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [ACK back to where the SHUTDOWN came from.] + */ + if (retval && chunk) + retval->transport = chunk->transport; + + return retval; +} + +struct sctp_chunk *sctp_make_shutdown_complete( + const struct sctp_association *asoc, + const struct sctp_chunk *chunk) +{ + struct sctp_chunk *retval; + __u8 flags = 0; + + /* Maybe set the T-bit if we have no association. */ + flags |= asoc ? 0 : SCTP_CHUNK_FLAG_T; + + retval = sctp_make_chunk(asoc, SCTP_CID_SHUTDOWN_COMPLETE, flags, 0); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [Report SHUTDOWN COMPLETE back to where the SHUTDOWN ACK + * came from.] + */ + if (retval && chunk) + retval->transport = chunk->transport; + + return retval; +} + +/* Create an ABORT. Note that we set the T bit if we have no + * association. + */ +struct sctp_chunk *sctp_make_abort(const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + const size_t hint) +{ + struct sctp_chunk *retval; + __u8 flags = 0; + + /* Maybe set the T-bit if we have no association. */ + flags |= asoc ? 0 : SCTP_CHUNK_FLAG_T; + + retval = sctp_make_chunk(asoc, SCTP_CID_ABORT, flags, hint); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [ABORT back to where the offender came from.] + */ + if (retval && chunk) + retval->transport = chunk->transport; + + return retval; +} + +/* Helper to create ABORT with a NO_USER_DATA error. */ +struct sctp_chunk *sctp_make_abort_no_data( + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, __u32 tsn) +{ + struct sctp_chunk *retval; + __u32 payload; + + retval = sctp_make_abort(asoc, chunk, sizeof(sctp_errhdr_t) + + sizeof(tsn)); + + if (!retval) + goto no_mem; + + /* Put the tsn back into network byte order. */ + payload = htonl(tsn); + sctp_init_cause(retval, SCTP_ERROR_NO_DATA, (const void *)&payload, + sizeof(payload)); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [ABORT back to where the offender came from.] + */ + if (chunk) + retval->transport = chunk->transport; + +no_mem: + return retval; +} + +/* Helper to create ABORT with a SCTP_ERROR_USER_ABORT error. */ +struct sctp_chunk *sctp_make_abort_user(const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + const struct msghdr *msg) +{ + struct sctp_chunk *retval; + void *payload = NULL, *payoff; + size_t paylen = 0; + struct iovec *iov = NULL; + int iovlen = 0; + + if (msg) { + iov = msg->msg_iov; + iovlen = msg->msg_iovlen; + paylen = get_user_iov_size(iov, iovlen); + } + + retval = sctp_make_abort(asoc, chunk, sizeof(sctp_errhdr_t) + paylen); + if (!retval) + goto err_chunk; + + if (paylen) { + /* Put the msg_iov together into payload. */ + payload = kmalloc(paylen, GFP_ATOMIC); + if (!payload) + goto err_payload; + payoff = payload; + + for (; iovlen > 0; --iovlen) { + if (copy_from_user(payoff, iov->iov_base,iov->iov_len)) + goto err_copy; + payoff += iov->iov_len; + iov++; + } + } + + sctp_init_cause(retval, SCTP_ERROR_USER_ABORT, payload, paylen); + + if (paylen) + kfree(payload); + + return retval; + +err_copy: + kfree(payload); +err_payload: + sctp_chunk_free(retval); + retval = NULL; +err_chunk: + return retval; +} + +/* Make an ABORT chunk with a PROTOCOL VIOLATION cause code. */ +struct sctp_chunk *sctp_make_abort_violation( + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + const __u8 *payload, + const size_t paylen) +{ + struct sctp_chunk *retval; + struct sctp_paramhdr phdr; + + retval = sctp_make_abort(asoc, chunk, sizeof(sctp_errhdr_t) + paylen + + sizeof(sctp_chunkhdr_t)); + if (!retval) + goto end; + + sctp_init_cause(retval, SCTP_ERROR_PROTO_VIOLATION, payload, paylen); + + phdr.type = htons(chunk->chunk_hdr->type); + phdr.length = chunk->chunk_hdr->length; + sctp_addto_chunk(retval, sizeof(sctp_paramhdr_t), &phdr); + +end: + return retval; +} + +/* Make a HEARTBEAT chunk. */ +struct sctp_chunk *sctp_make_heartbeat(const struct sctp_association *asoc, + const struct sctp_transport *transport, + const void *payload, const size_t paylen) +{ + struct sctp_chunk *retval = sctp_make_chunk(asoc, SCTP_CID_HEARTBEAT, + 0, paylen); + + if (!retval) + goto nodata; + + /* Cast away the 'const', as this is just telling the chunk + * what transport it belongs to. + */ + retval->transport = (struct sctp_transport *) transport; + retval->subh.hbs_hdr = sctp_addto_chunk(retval, paylen, payload); + +nodata: + return retval; +} + +struct sctp_chunk *sctp_make_heartbeat_ack(const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + const void *payload, const size_t paylen) +{ + struct sctp_chunk *retval; + + retval = sctp_make_chunk(asoc, SCTP_CID_HEARTBEAT_ACK, 0, paylen); + if (!retval) + goto nodata; + + retval->subh.hbs_hdr = sctp_addto_chunk(retval, paylen, payload); + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, * etc.) to the same destination transport + * address from which it * received the DATA or control chunk + * to which it is replying. + * + * [HBACK back to where the HEARTBEAT came from.] + */ + if (chunk) + retval->transport = chunk->transport; + +nodata: + return retval; +} + +/* Create an Operation Error chunk with the specified space reserved. + * This routine can be used for containing multiple causes in the chunk. + */ +static struct sctp_chunk *sctp_make_op_error_space( + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + size_t size) +{ + struct sctp_chunk *retval; + + retval = sctp_make_chunk(asoc, SCTP_CID_ERROR, 0, + sizeof(sctp_errhdr_t) + size); + if (!retval) + goto nodata; + + /* RFC 2960 6.4 Multi-homed SCTP Endpoints + * + * An endpoint SHOULD transmit reply chunks (e.g., SACK, + * HEARTBEAT ACK, etc.) to the same destination transport + * address from which it received the DATA or control chunk + * to which it is replying. + * + */ + if (chunk) + retval->transport = chunk->transport; + +nodata: + return retval; +} + +/* Create an Operation Error chunk. */ +struct sctp_chunk *sctp_make_op_error(const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + __u16 cause_code, const void *payload, + size_t paylen) +{ + struct sctp_chunk *retval; + + retval = sctp_make_op_error_space(asoc, chunk, paylen); + if (!retval) + goto nodata; + + sctp_init_cause(retval, cause_code, payload, paylen); + +nodata: + return retval; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* Turn an skb into a chunk. + * FIXME: Eventually move the structure directly inside the skb->cb[]. + */ +struct sctp_chunk *sctp_chunkify(struct sk_buff *skb, + const struct sctp_association *asoc, + struct sock *sk) +{ + struct sctp_chunk *retval; + + retval = kmem_cache_alloc(sctp_chunk_cachep, SLAB_ATOMIC); + + if (!retval) + goto nodata; + memset(retval, 0, sizeof(struct sctp_chunk)); + + if (!sk) { + SCTP_DEBUG_PRINTK("chunkifying skb %p w/o an sk\n", skb); + } + + retval->skb = skb; + retval->asoc = (struct sctp_association *)asoc; + retval->resent = 0; + retval->has_tsn = 0; + retval->has_ssn = 0; + retval->rtt_in_progress = 0; + retval->sent_at = 0; + retval->singleton = 1; + retval->end_of_packet = 0; + retval->ecn_ce_done = 0; + retval->pdiscard = 0; + + /* sctpimpguide-05.txt Section 2.8.2 + * M1) Each time a new DATA chunk is transmitted + * set the 'TSN.Missing.Report' count for that TSN to 0. The + * 'TSN.Missing.Report' count will be used to determine missing chunks + * and when to fast retransmit. + */ + retval->tsn_missing_report = 0; + retval->tsn_gap_acked = 0; + retval->fast_retransmit = 0; + + /* If this is a fragmented message, track all fragments + * of the message (for SEND_FAILED). + */ + retval->msg = NULL; + + /* Polish the bead hole. */ + INIT_LIST_HEAD(&retval->transmitted_list); + INIT_LIST_HEAD(&retval->frag_list); + SCTP_DBG_OBJCNT_INC(chunk); + atomic_set(&retval->refcnt, 1); + +nodata: + return retval; +} + +/* Set chunk->source and dest based on the IP header in chunk->skb. */ +void sctp_init_addrs(struct sctp_chunk *chunk, union sctp_addr *src, + union sctp_addr *dest) +{ + memcpy(&chunk->source, src, sizeof(union sctp_addr)); + memcpy(&chunk->dest, dest, sizeof(union sctp_addr)); +} + +/* Extract the source address from a chunk. */ +const union sctp_addr *sctp_source(const struct sctp_chunk *chunk) +{ + /* If we have a known transport, use that. */ + if (chunk->transport) { + return &chunk->transport->ipaddr; + } else { + /* Otherwise, extract it from the IP header. */ + return &chunk->source; + } +} + +/* Create a new chunk, setting the type and flags headers from the + * arguments, reserving enough space for a 'paylen' byte payload. + */ +SCTP_STATIC +struct sctp_chunk *sctp_make_chunk(const struct sctp_association *asoc, + __u8 type, __u8 flags, int paylen) +{ + struct sctp_chunk *retval; + sctp_chunkhdr_t *chunk_hdr; + struct sk_buff *skb; + struct sock *sk; + + /* No need to allocate LL here, as this is only a chunk. */ + skb = alloc_skb(WORD_ROUND(sizeof(sctp_chunkhdr_t) + paylen), + GFP_ATOMIC); + if (!skb) + goto nodata; + + /* Make room for the chunk header. */ + chunk_hdr = (sctp_chunkhdr_t *)skb_put(skb, sizeof(sctp_chunkhdr_t)); + chunk_hdr->type = type; + chunk_hdr->flags = flags; + chunk_hdr->length = htons(sizeof(sctp_chunkhdr_t)); + + sk = asoc ? asoc->base.sk : NULL; + retval = sctp_chunkify(skb, asoc, sk); + if (!retval) { + kfree_skb(skb); + goto nodata; + } + + retval->chunk_hdr = chunk_hdr; + retval->chunk_end = ((__u8 *)chunk_hdr) + sizeof(struct sctp_chunkhdr); + + /* Set the skb to the belonging sock for accounting. */ + skb->sk = sk; + + return retval; +nodata: + return NULL; +} + + +/* Release the memory occupied by a chunk. */ +static void sctp_chunk_destroy(struct sctp_chunk *chunk) +{ + /* Free the chunk skb data and the SCTP_chunk stub itself. */ + dev_kfree_skb(chunk->skb); + + SCTP_DBG_OBJCNT_DEC(chunk); + kmem_cache_free(sctp_chunk_cachep, chunk); +} + +/* Possibly, free the chunk. */ +void sctp_chunk_free(struct sctp_chunk *chunk) +{ + /* Make sure that we are not on any list. */ + skb_unlink((struct sk_buff *) chunk); + list_del_init(&chunk->transmitted_list); + + /* Release our reference on the message tracker. */ + if (chunk->msg) + sctp_datamsg_put(chunk->msg); + + sctp_chunk_put(chunk); +} + +/* Grab a reference to the chunk. */ +void sctp_chunk_hold(struct sctp_chunk *ch) +{ + atomic_inc(&ch->refcnt); +} + +/* Release a reference to the chunk. */ +void sctp_chunk_put(struct sctp_chunk *ch) +{ + if (atomic_dec_and_test(&ch->refcnt)) + sctp_chunk_destroy(ch); +} + +/* Append bytes to the end of a chunk. Will panic if chunk is not big + * enough. + */ +void *sctp_addto_chunk(struct sctp_chunk *chunk, int len, const void *data) +{ + void *target; + void *padding; + int chunklen = ntohs(chunk->chunk_hdr->length); + int padlen = chunklen % 4; + + padding = skb_put(chunk->skb, padlen); + target = skb_put(chunk->skb, len); + + memset(padding, 0, padlen); + memcpy(target, data, len); + + /* Adjust the chunk length field. */ + chunk->chunk_hdr->length = htons(chunklen + padlen + len); + chunk->chunk_end = chunk->skb->tail; + + return target; +} + +/* Append bytes from user space to the end of a chunk. Will panic if + * chunk is not big enough. + * Returns a kernel err value. + */ +int sctp_user_addto_chunk(struct sctp_chunk *chunk, int off, int len, + struct iovec *data) +{ + __u8 *target; + int err = 0; + + /* Make room in chunk for data. */ + target = skb_put(chunk->skb, len); + + /* Copy data (whole iovec) into chunk */ + if ((err = memcpy_fromiovecend(target, data, off, len))) + goto out; + + /* Adjust the chunk length field. */ + chunk->chunk_hdr->length = + htons(ntohs(chunk->chunk_hdr->length) + len); + chunk->chunk_end = chunk->skb->tail; + +out: + return err; +} + +/* Helper function to assign a TSN if needed. This assumes that both + * the data_hdr and association have already been assigned. + */ +void sctp_chunk_assign_ssn(struct sctp_chunk *chunk) +{ + __u16 ssn; + __u16 sid; + + if (chunk->has_ssn) + return; + + /* This is the last possible instant to assign a SSN. */ + if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) { + ssn = 0; + } else { + sid = htons(chunk->subh.data_hdr->stream); + if (chunk->chunk_hdr->flags & SCTP_DATA_LAST_FRAG) + ssn = sctp_ssn_next(&chunk->asoc->ssnmap->out, sid); + else + ssn = sctp_ssn_peek(&chunk->asoc->ssnmap->out, sid); + ssn = htons(ssn); + } + + chunk->subh.data_hdr->ssn = ssn; + chunk->has_ssn = 1; +} + +/* Helper function to assign a TSN if needed. This assumes that both + * the data_hdr and association have already been assigned. + */ +void sctp_chunk_assign_tsn(struct sctp_chunk *chunk) +{ + if (!chunk->has_tsn) { + /* This is the last possible instant to + * assign a TSN. + */ + chunk->subh.data_hdr->tsn = + htonl(sctp_association_get_next_tsn(chunk->asoc)); + chunk->has_tsn = 1; + } +} + +/* Create a CLOSED association to use with an incoming packet. */ +struct sctp_association *sctp_make_temp_asoc(const struct sctp_endpoint *ep, + struct sctp_chunk *chunk, int gfp) +{ + struct sctp_association *asoc; + struct sk_buff *skb; + sctp_scope_t scope; + struct sctp_af *af; + + /* Create the bare association. */ + scope = sctp_scope(sctp_source(chunk)); + asoc = sctp_association_new(ep, ep->base.sk, scope, gfp); + if (!asoc) + goto nodata; + asoc->temp = 1; + skb = chunk->skb; + /* Create an entry for the source address of the packet. */ + af = sctp_get_af_specific(ipver2af(skb->nh.iph->version)); + if (unlikely(!af)) + goto fail; + af->from_skb(&asoc->c.peer_addr, skb, 1); +nodata: + return asoc; + +fail: + sctp_association_free(asoc); + return NULL; +} + +/* Build a cookie representing asoc. + * This INCLUDES the param header needed to put the cookie in the INIT ACK. + */ +static sctp_cookie_param_t *sctp_pack_cookie(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const struct sctp_chunk *init_chunk, + int *cookie_len, + const __u8 *raw_addrs, int addrs_len) +{ + sctp_cookie_param_t *retval; + struct sctp_signed_cookie *cookie; + struct scatterlist sg; + int headersize, bodysize; + unsigned int keylen; + char *key; + + headersize = sizeof(sctp_paramhdr_t) + SCTP_SECRET_SIZE; + bodysize = sizeof(struct sctp_cookie) + + ntohs(init_chunk->chunk_hdr->length) + addrs_len; + + /* Pad out the cookie to a multiple to make the signature + * functions simpler to write. + */ + if (bodysize % SCTP_COOKIE_MULTIPLE) + bodysize += SCTP_COOKIE_MULTIPLE + - (bodysize % SCTP_COOKIE_MULTIPLE); + *cookie_len = headersize + bodysize; + + retval = (sctp_cookie_param_t *)kmalloc(*cookie_len, GFP_ATOMIC); + + if (!retval) { + *cookie_len = 0; + goto nodata; + } + + /* Clear this memory since we are sending this data structure + * out on the network. + */ + memset(retval, 0x00, *cookie_len); + cookie = (struct sctp_signed_cookie *) retval->body; + + /* Set up the parameter header. */ + retval->p.type = SCTP_PARAM_STATE_COOKIE; + retval->p.length = htons(*cookie_len); + + /* Copy the cookie part of the association itself. */ + cookie->c = asoc->c; + /* Save the raw address list length in the cookie. */ + cookie->c.raw_addr_list_len = addrs_len; + + /* Remember PR-SCTP capability. */ + cookie->c.prsctp_capable = asoc->peer.prsctp_capable; + + /* Save adaption indication in the cookie. */ + cookie->c.adaption_ind = asoc->peer.adaption_ind; + + /* Set an expiration time for the cookie. */ + do_gettimeofday(&cookie->c.expiration); + TIMEVAL_ADD(asoc->cookie_life, cookie->c.expiration); + + /* Copy the peer's init packet. */ + memcpy(&cookie->c.peer_init[0], init_chunk->chunk_hdr, + ntohs(init_chunk->chunk_hdr->length)); + + /* Copy the raw local address list of the association. */ + memcpy((__u8 *)&cookie->c.peer_init[0] + + ntohs(init_chunk->chunk_hdr->length), raw_addrs, addrs_len); + + if (sctp_sk(ep->base.sk)->hmac) { + /* Sign the message. */ + sg.page = virt_to_page(&cookie->c); + sg.offset = (unsigned long)(&cookie->c) % PAGE_SIZE; + sg.length = bodysize; + keylen = SCTP_SECRET_SIZE; + key = (char *)ep->secret_key[ep->current_key]; + + sctp_crypto_hmac(sctp_sk(ep->base.sk)->hmac, key, &keylen, + &sg, 1, cookie->signature); + } + +nodata: + return retval; +} + +/* Unpack the cookie from COOKIE ECHO chunk, recreating the association. */ +struct sctp_association *sctp_unpack_cookie( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, int gfp, + int *error, struct sctp_chunk **errp) +{ + struct sctp_association *retval = NULL; + struct sctp_signed_cookie *cookie; + struct sctp_cookie *bear_cookie; + int headersize, bodysize, fixed_size; + __u8 digest[SCTP_SIGNATURE_SIZE]; + struct scatterlist sg; + unsigned int keylen, len; + char *key; + sctp_scope_t scope; + struct sk_buff *skb = chunk->skb; + + headersize = sizeof(sctp_chunkhdr_t) + SCTP_SECRET_SIZE; + bodysize = ntohs(chunk->chunk_hdr->length) - headersize; + fixed_size = headersize + sizeof(struct sctp_cookie); + + /* Verify that the chunk looks like it even has a cookie. + * There must be enough room for our cookie and our peer's + * INIT chunk. + */ + len = ntohs(chunk->chunk_hdr->length); + if (len < fixed_size + sizeof(struct sctp_chunkhdr)) + goto malformed; + + /* Verify that the cookie has been padded out. */ + if (bodysize % SCTP_COOKIE_MULTIPLE) + goto malformed; + + /* Process the cookie. */ + cookie = chunk->subh.cookie_hdr; + bear_cookie = &cookie->c; + + if (!sctp_sk(ep->base.sk)->hmac) + goto no_hmac; + + /* Check the signature. */ + keylen = SCTP_SECRET_SIZE; + sg.page = virt_to_page(bear_cookie); + sg.offset = (unsigned long)(bear_cookie) % PAGE_SIZE; + sg.length = bodysize; + key = (char *)ep->secret_key[ep->current_key]; + + memset(digest, 0x00, sizeof(digest)); + sctp_crypto_hmac(sctp_sk(ep->base.sk)->hmac, key, &keylen, &sg, + 1, digest); + + if (memcmp(digest, cookie->signature, SCTP_SIGNATURE_SIZE)) { + /* Try the previous key. */ + key = (char *)ep->secret_key[ep->last_key]; + memset(digest, 0x00, sizeof(digest)); + sctp_crypto_hmac(sctp_sk(ep->base.sk)->hmac, key, &keylen, + &sg, 1, digest); + + if (memcmp(digest, cookie->signature, SCTP_SIGNATURE_SIZE)) { + /* Yikes! Still bad signature! */ + *error = -SCTP_IERROR_BAD_SIG; + goto fail; + } + } + +no_hmac: + /* IG Section 2.35.2: + * 3) Compare the port numbers and the verification tag contained + * within the COOKIE ECHO chunk to the actual port numbers and the + * verification tag within the SCTP common header of the received + * packet. If these values do not match the packet MUST be silently + * discarded, + */ + if (ntohl(chunk->sctp_hdr->vtag) != bear_cookie->my_vtag) { + *error = -SCTP_IERROR_BAD_TAG; + goto fail; + } + + if (ntohs(chunk->sctp_hdr->source) != bear_cookie->peer_addr.v4.sin_port || + ntohs(chunk->sctp_hdr->dest) != bear_cookie->my_port) { + *error = -SCTP_IERROR_BAD_PORTS; + goto fail; + } + + /* Check to see if the cookie is stale. If there is already + * an association, there is no need to check cookie's expiration + * for init collision case of lost COOKIE ACK. + */ + if (!asoc && tv_lt(bear_cookie->expiration, skb->stamp)) { + __u16 len; + /* + * Section 3.3.10.3 Stale Cookie Error (3) + * + * Cause of error + * --------------- + * Stale Cookie Error: Indicates the receipt of a valid State + * Cookie that has expired. + */ + len = ntohs(chunk->chunk_hdr->length); + *errp = sctp_make_op_error_space(asoc, chunk, len); + if (*errp) { + suseconds_t usecs = (skb->stamp.tv_sec - + bear_cookie->expiration.tv_sec) * 1000000L + + skb->stamp.tv_usec - + bear_cookie->expiration.tv_usec; + + usecs = htonl(usecs); + sctp_init_cause(*errp, SCTP_ERROR_STALE_COOKIE, + &usecs, sizeof(usecs)); + *error = -SCTP_IERROR_STALE_COOKIE; + } else + *error = -SCTP_IERROR_NOMEM; + + goto fail; + } + + /* Make a new base association. */ + scope = sctp_scope(sctp_source(chunk)); + retval = sctp_association_new(ep, ep->base.sk, scope, gfp); + if (!retval) { + *error = -SCTP_IERROR_NOMEM; + goto fail; + } + + /* Set up our peer's port number. */ + retval->peer.port = ntohs(chunk->sctp_hdr->source); + + /* Populate the association from the cookie. */ + memcpy(&retval->c, bear_cookie, sizeof(*bear_cookie)); + + if (sctp_assoc_set_bind_addr_from_cookie(retval, bear_cookie, + GFP_ATOMIC) < 0) { + *error = -SCTP_IERROR_NOMEM; + goto fail; + } + + /* Also, add the destination address. */ + if (list_empty(&retval->base.bind_addr.address_list)) { + sctp_add_bind_addr(&retval->base.bind_addr, &chunk->dest, + GFP_ATOMIC); + } + + retval->next_tsn = retval->c.initial_tsn; + retval->ctsn_ack_point = retval->next_tsn - 1; + retval->addip_serial = retval->c.initial_tsn; + retval->adv_peer_ack_point = retval->ctsn_ack_point; + retval->peer.prsctp_capable = retval->c.prsctp_capable; + retval->peer.adaption_ind = retval->c.adaption_ind; + + /* The INIT stuff will be done by the side effects. */ + return retval; + +fail: + if (retval) + sctp_association_free(retval); + + return NULL; + +malformed: + /* Yikes! The packet is either corrupt or deliberately + * malformed. + */ + *error = -SCTP_IERROR_MALFORMED; + goto fail; +} + +/******************************************************************** + * 3rd Level Abstractions + ********************************************************************/ + +struct __sctp_missing { + __u32 num_missing; + __u16 type; +} __attribute__((packed)); + +/* + * Report a missing mandatory parameter. + */ +static int sctp_process_missing_param(const struct sctp_association *asoc, + sctp_param_t paramtype, + struct sctp_chunk *chunk, + struct sctp_chunk **errp) +{ + struct __sctp_missing report; + __u16 len; + + len = WORD_ROUND(sizeof(report)); + + /* Make an ERROR chunk, preparing enough room for + * returning multiple unknown parameters. + */ + if (!*errp) + *errp = sctp_make_op_error_space(asoc, chunk, len); + + if (*errp) { + report.num_missing = htonl(1); + report.type = paramtype; + sctp_init_cause(*errp, SCTP_ERROR_INV_PARAM, + &report, sizeof(report)); + } + + /* Stop processing this chunk. */ + return 0; +} + +/* Report an Invalid Mandatory Parameter. */ +static int sctp_process_inv_mandatory(const struct sctp_association *asoc, + struct sctp_chunk *chunk, + struct sctp_chunk **errp) +{ + /* Invalid Mandatory Parameter Error has no payload. */ + + if (!*errp) + *errp = sctp_make_op_error_space(asoc, chunk, 0); + + if (*errp) + sctp_init_cause(*errp, SCTP_ERROR_INV_PARAM, NULL, 0); + + /* Stop processing this chunk. */ + return 0; +} + +static int sctp_process_inv_paramlength(const struct sctp_association *asoc, + struct sctp_paramhdr *param, + const struct sctp_chunk *chunk, + struct sctp_chunk **errp) +{ + char error[] = "The following parameter had invalid length:"; + size_t payload_len = WORD_ROUND(sizeof(error)) + + sizeof(sctp_paramhdr_t); + + + /* Create an error chunk and fill it in with our payload. */ + if (!*errp) + *errp = sctp_make_op_error_space(asoc, chunk, payload_len); + + if (*errp) { + sctp_init_cause(*errp, SCTP_ERROR_PROTO_VIOLATION, error, + sizeof(error)); + sctp_addto_chunk(*errp, sizeof(sctp_paramhdr_t), param); + } + + return 0; +} + + +/* Do not attempt to handle the HOST_NAME parm. However, do + * send back an indicator to the peer. + */ +static int sctp_process_hn_param(const struct sctp_association *asoc, + union sctp_params param, + struct sctp_chunk *chunk, + struct sctp_chunk **errp) +{ + __u16 len = ntohs(param.p->length); + + /* Make an ERROR chunk. */ + if (!*errp) + *errp = sctp_make_op_error_space(asoc, chunk, len); + + if (*errp) + sctp_init_cause(*errp, SCTP_ERROR_DNS_FAILED, + param.v, len); + + /* Stop processing this chunk. */ + return 0; +} + +/* RFC 3.2.1 & the Implementers Guide 2.2. + * + * The Parameter Types are encoded such that the + * highest-order two bits specify the action that must be + * taken if the processing endpoint does not recognize the + * Parameter Type. + * + * 00 - Stop processing this SCTP chunk and discard it, + * do not process any further chunks within it. + * + * 01 - Stop processing this SCTP chunk and discard it, + * do not process any further chunks within it, and report + * the unrecognized parameter in an 'Unrecognized + * Parameter Type' (in either an ERROR or in the INIT ACK). + * + * 10 - Skip this parameter and continue processing. + * + * 11 - Skip this parameter and continue processing but + * report the unrecognized parameter in an + * 'Unrecognized Parameter Type' (in either an ERROR or in + * the INIT ACK). + * + * Return value: + * 0 - discard the chunk + * 1 - continue with the chunk + */ +static int sctp_process_unk_param(const struct sctp_association *asoc, + union sctp_params param, + struct sctp_chunk *chunk, + struct sctp_chunk **errp) +{ + int retval = 1; + + switch (param.p->type & SCTP_PARAM_ACTION_MASK) { + case SCTP_PARAM_ACTION_DISCARD: + retval = 0; + break; + case SCTP_PARAM_ACTION_DISCARD_ERR: + retval = 0; + /* Make an ERROR chunk, preparing enough room for + * returning multiple unknown parameters. + */ + if (NULL == *errp) + *errp = sctp_make_op_error_space(asoc, chunk, + ntohs(chunk->chunk_hdr->length)); + + if (*errp) + sctp_init_cause(*errp, SCTP_ERROR_UNKNOWN_PARAM, + param.v, + WORD_ROUND(ntohs(param.p->length))); + + break; + case SCTP_PARAM_ACTION_SKIP: + break; + case SCTP_PARAM_ACTION_SKIP_ERR: + /* Make an ERROR chunk, preparing enough room for + * returning multiple unknown parameters. + */ + if (NULL == *errp) + *errp = sctp_make_op_error_space(asoc, chunk, + ntohs(chunk->chunk_hdr->length)); + + if (*errp) { + sctp_init_cause(*errp, SCTP_ERROR_UNKNOWN_PARAM, + param.v, + WORD_ROUND(ntohs(param.p->length))); + } else { + /* If there is no memory for generating the ERROR + * report as specified, an ABORT will be triggered + * to the peer and the association won't be + * established. + */ + retval = 0; + } + + break; + default: + break; + } + + return retval; +} + +/* Find unrecognized parameters in the chunk. + * Return values: + * 0 - discard the chunk + * 1 - continue with the chunk + */ +static int sctp_verify_param(const struct sctp_association *asoc, + union sctp_params param, + sctp_cid_t cid, + struct sctp_chunk *chunk, + struct sctp_chunk **err_chunk) +{ + int retval = 1; + + /* FIXME - This routine is not looking at each parameter per the + * chunk type, i.e., unrecognized parameters should be further + * identified based on the chunk id. + */ + + switch (param.p->type) { + case SCTP_PARAM_IPV4_ADDRESS: + case SCTP_PARAM_IPV6_ADDRESS: + case SCTP_PARAM_COOKIE_PRESERVATIVE: + case SCTP_PARAM_SUPPORTED_ADDRESS_TYPES: + case SCTP_PARAM_STATE_COOKIE: + case SCTP_PARAM_HEARTBEAT_INFO: + case SCTP_PARAM_UNRECOGNIZED_PARAMETERS: + case SCTP_PARAM_ECN_CAPABLE: + case SCTP_PARAM_ADAPTION_LAYER_IND: + break; + + case SCTP_PARAM_HOST_NAME_ADDRESS: + /* Tell the peer, we won't support this param. */ + return sctp_process_hn_param(asoc, param, chunk, err_chunk); + case SCTP_PARAM_FWD_TSN_SUPPORT: + if (sctp_prsctp_enable) + break; + /* Fall Through */ + default: + SCTP_DEBUG_PRINTK("Unrecognized param: %d for chunk %d.\n", + ntohs(param.p->type), cid); + return sctp_process_unk_param(asoc, param, chunk, err_chunk); + + break; + } + return retval; +} + +/* Verify the INIT packet before we process it. */ +int sctp_verify_init(const struct sctp_association *asoc, + sctp_cid_t cid, + sctp_init_chunk_t *peer_init, + struct sctp_chunk *chunk, + struct sctp_chunk **errp) +{ + union sctp_params param; + int has_cookie = 0; + + /* Verify stream values are non-zero. */ + if ((0 == peer_init->init_hdr.num_outbound_streams) || + (0 == peer_init->init_hdr.num_inbound_streams)) { + + sctp_process_inv_mandatory(asoc, chunk, errp); + return 0; + } + + /* Check for missing mandatory parameters. */ + sctp_walk_params(param, peer_init, init_hdr.params) { + + if (SCTP_PARAM_STATE_COOKIE == param.p->type) + has_cookie = 1; + + } /* for (loop through all parameters) */ + + /* There is a possibility that a parameter length was bad and + * in that case we would have stoped walking the parameters. + * The current param.p would point at the bad one. + * Current consensus on the mailing list is to generate a PROTOCOL + * VIOLATION error. We build the ERROR chunk here and let the normal + * error handling code build and send the packet. + */ + if (param.v < (void*)chunk->chunk_end - sizeof(sctp_paramhdr_t)) { + sctp_process_inv_paramlength(asoc, param.p, chunk, errp); + return 0; + } + + /* The only missing mandatory param possible today is + * the state cookie for an INIT-ACK chunk. + */ + if ((SCTP_CID_INIT_ACK == cid) && !has_cookie) { + sctp_process_missing_param(asoc, SCTP_PARAM_STATE_COOKIE, + chunk, errp); + return 0; + } + + /* Find unrecognized parameters. */ + + sctp_walk_params(param, peer_init, init_hdr.params) { + + if (!sctp_verify_param(asoc, param, cid, chunk, errp)) { + if (SCTP_PARAM_HOST_NAME_ADDRESS == param.p->type) + return 0; + else + return 1; + } + + } /* for (loop through all parameters) */ + + return 1; +} + +/* Unpack the parameters in an INIT packet into an association. + * Returns 0 on failure, else success. + * FIXME: This is an association method. + */ +int sctp_process_init(struct sctp_association *asoc, sctp_cid_t cid, + const union sctp_addr *peer_addr, + sctp_init_chunk_t *peer_init, int gfp) +{ + union sctp_params param; + struct sctp_transport *transport; + struct list_head *pos, *temp; + char *cookie; + + /* We must include the address that the INIT packet came from. + * This is the only address that matters for an INIT packet. + * When processing a COOKIE ECHO, we retrieve the from address + * of the INIT from the cookie. + */ + + /* This implementation defaults to making the first transport + * added as the primary transport. The source address seems to + * be a a better choice than any of the embedded addresses. + */ + if (peer_addr) + if(!sctp_assoc_add_peer(asoc, peer_addr, gfp)) + goto nomem; + + /* Process the initialization parameters. */ + + sctp_walk_params(param, peer_init, init_hdr.params) { + + if (!sctp_process_param(asoc, param, peer_addr, gfp)) + goto clean_up; + } + + /* The fixed INIT headers are always in network byte + * order. + */ + asoc->peer.i.init_tag = + ntohl(peer_init->init_hdr.init_tag); + asoc->peer.i.a_rwnd = + ntohl(peer_init->init_hdr.a_rwnd); + asoc->peer.i.num_outbound_streams = + ntohs(peer_init->init_hdr.num_outbound_streams); + asoc->peer.i.num_inbound_streams = + ntohs(peer_init->init_hdr.num_inbound_streams); + asoc->peer.i.initial_tsn = + ntohl(peer_init->init_hdr.initial_tsn); + + /* Apply the upper bounds for output streams based on peer's + * number of inbound streams. + */ + if (asoc->c.sinit_num_ostreams > + ntohs(peer_init->init_hdr.num_inbound_streams)) { + asoc->c.sinit_num_ostreams = + ntohs(peer_init->init_hdr.num_inbound_streams); + } + + if (asoc->c.sinit_max_instreams > + ntohs(peer_init->init_hdr.num_outbound_streams)) { + asoc->c.sinit_max_instreams = + ntohs(peer_init->init_hdr.num_outbound_streams); + } + + /* Copy Initiation tag from INIT to VT_peer in cookie. */ + asoc->c.peer_vtag = asoc->peer.i.init_tag; + + /* Peer Rwnd : Current calculated value of the peer's rwnd. */ + asoc->peer.rwnd = asoc->peer.i.a_rwnd; + + /* Copy cookie in case we need to resend COOKIE-ECHO. */ + cookie = asoc->peer.cookie; + if (cookie) { + asoc->peer.cookie = kmalloc(asoc->peer.cookie_len, gfp); + if (!asoc->peer.cookie) + goto clean_up; + memcpy(asoc->peer.cookie, cookie, asoc->peer.cookie_len); + } + + /* RFC 2960 7.2.1 The initial value of ssthresh MAY be arbitrarily + * high (for example, implementations MAY use the size of the receiver + * advertised window). + */ + list_for_each(pos, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + transport->ssthresh = asoc->peer.i.a_rwnd; + } + + /* Set up the TSN tracking pieces. */ + sctp_tsnmap_init(&asoc->peer.tsn_map, SCTP_TSN_MAP_SIZE, + asoc->peer.i.initial_tsn); + + /* RFC 2960 6.5 Stream Identifier and Stream Sequence Number + * + * The stream sequence number in all the streams shall start + * from 0 when the association is established. Also, when the + * stream sequence number reaches the value 65535 the next + * stream sequence number shall be set to 0. + */ + + /* Allocate storage for the negotiated streams if it is not a temporary * association. + */ + if (!asoc->temp) { + int assoc_id; + int error; + + asoc->ssnmap = sctp_ssnmap_new(asoc->c.sinit_max_instreams, + asoc->c.sinit_num_ostreams, gfp); + if (!asoc->ssnmap) + goto clean_up; + + retry: + if (unlikely(!idr_pre_get(&sctp_assocs_id, gfp))) + goto clean_up; + spin_lock_bh(&sctp_assocs_id_lock); + error = idr_get_new_above(&sctp_assocs_id, (void *)asoc, 1, + &assoc_id); + spin_unlock_bh(&sctp_assocs_id_lock); + if (error == -EAGAIN) + goto retry; + else if (error) + goto clean_up; + + asoc->assoc_id = (sctp_assoc_t) assoc_id; + } + + /* ADDIP Section 4.1 ASCONF Chunk Procedures + * + * When an endpoint has an ASCONF signaled change to be sent to the + * remote endpoint it should do the following: + * ... + * A2) A serial number should be assigned to the Chunk. The serial + * number should be a monotonically increasing number. All serial + * numbers are defined to be initialized at the start of the + * association to the same value as the Initial TSN. + */ + asoc->peer.addip_serial = asoc->peer.i.initial_tsn - 1; + return 1; + +clean_up: + /* Release the transport structures. */ + list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, transports); + list_del_init(pos); + sctp_transport_free(transport); + } +nomem: + return 0; +} + + +/* Update asoc with the option described in param. + * + * RFC2960 3.3.2.1 Optional/Variable Length Parameters in INIT + * + * asoc is the association to update. + * param is the variable length parameter to use for update. + * cid tells us if this is an INIT, INIT ACK or COOKIE ECHO. + * If the current packet is an INIT we want to minimize the amount of + * work we do. In particular, we should not build transport + * structures for the addresses. + */ +static int sctp_process_param(struct sctp_association *asoc, + union sctp_params param, + const union sctp_addr *peer_addr, + int gfp) +{ + union sctp_addr addr; + int i; + __u16 sat; + int retval = 1; + sctp_scope_t scope; + time_t stale; + struct sctp_af *af; + + /* We maintain all INIT parameters in network byte order all the + * time. This allows us to not worry about whether the parameters + * came from a fresh INIT, and INIT ACK, or were stored in a cookie. + */ + switch (param.p->type) { + case SCTP_PARAM_IPV6_ADDRESS: + if (PF_INET6 != asoc->base.sk->sk_family) + break; + /* Fall through. */ + case SCTP_PARAM_IPV4_ADDRESS: + af = sctp_get_af_specific(param_type2af(param.p->type)); + af->from_addr_param(&addr, param.addr, asoc->peer.port, 0); + scope = sctp_scope(peer_addr); + if (sctp_in_scope(&addr, scope)) + if (!sctp_assoc_add_peer(asoc, &addr, gfp)) + return 0; + break; + + case SCTP_PARAM_COOKIE_PRESERVATIVE: + if (!sctp_cookie_preserve_enable) + break; + + stale = ntohl(param.life->lifespan_increment); + + /* Suggested Cookie Life span increment's unit is msec, + * (1/1000sec). + */ + asoc->cookie_life.tv_sec += stale / 1000; + asoc->cookie_life.tv_usec += (stale % 1000) * 1000; + break; + + case SCTP_PARAM_HOST_NAME_ADDRESS: + SCTP_DEBUG_PRINTK("unimplemented SCTP_HOST_NAME_ADDRESS\n"); + break; + + case SCTP_PARAM_SUPPORTED_ADDRESS_TYPES: + /* Turn off the default values first so we'll know which + * ones are really set by the peer. + */ + asoc->peer.ipv4_address = 0; + asoc->peer.ipv6_address = 0; + + /* Cycle through address types; avoid divide by 0. */ + sat = ntohs(param.p->length) - sizeof(sctp_paramhdr_t); + if (sat) + sat /= sizeof(__u16); + + for (i = 0; i < sat; ++i) { + switch (param.sat->types[i]) { + case SCTP_PARAM_IPV4_ADDRESS: + asoc->peer.ipv4_address = 1; + break; + + case SCTP_PARAM_IPV6_ADDRESS: + asoc->peer.ipv6_address = 1; + break; + + case SCTP_PARAM_HOST_NAME_ADDRESS: + asoc->peer.hostname_address = 1; + break; + + default: /* Just ignore anything else. */ + break; + }; + } + break; + + case SCTP_PARAM_STATE_COOKIE: + asoc->peer.cookie_len = + ntohs(param.p->length) - sizeof(sctp_paramhdr_t); + asoc->peer.cookie = param.cookie->body; + break; + + case SCTP_PARAM_HEARTBEAT_INFO: + /* Would be odd to receive, but it causes no problems. */ + break; + + case SCTP_PARAM_UNRECOGNIZED_PARAMETERS: + /* Rejected during verify stage. */ + break; + + case SCTP_PARAM_ECN_CAPABLE: + asoc->peer.ecn_capable = 1; + break; + + case SCTP_PARAM_ADAPTION_LAYER_IND: + asoc->peer.adaption_ind = param.aind->adaption_ind; + break; + + case SCTP_PARAM_FWD_TSN_SUPPORT: + if (sctp_prsctp_enable) { + asoc->peer.prsctp_capable = 1; + break; + } + /* Fall Through */ + default: + /* Any unrecognized parameters should have been caught + * and handled by sctp_verify_param() which should be + * called prior to this routine. Simply log the error + * here. + */ + SCTP_DEBUG_PRINTK("Ignoring param: %d for association %p.\n", + ntohs(param.p->type), asoc); + break; + }; + + return retval; +} + +/* Select a new verification tag. */ +__u32 sctp_generate_tag(const struct sctp_endpoint *ep) +{ + /* I believe that this random number generator complies with RFC1750. + * A tag of 0 is reserved for special cases (e.g. INIT). + */ + __u32 x; + + do { + get_random_bytes(&x, sizeof(__u32)); + } while (x == 0); + + return x; +} + +/* Select an initial TSN to send during startup. */ +__u32 sctp_generate_tsn(const struct sctp_endpoint *ep) +{ + __u32 retval; + + get_random_bytes(&retval, sizeof(__u32)); + return retval; +} + +/* + * ADDIP 3.1.1 Address Configuration Change Chunk (ASCONF) + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 0xC1 | Chunk Flags | Chunk Length | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Serial Number | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Address Parameter | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF Parameter #1 | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * \ \ + * / .... / + * \ \ + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF Parameter #N | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * Address Parameter and other parameter will not be wrapped in this function + */ +static struct sctp_chunk *sctp_make_asconf(struct sctp_association *asoc, + union sctp_addr *addr, + int vparam_len) +{ + sctp_addiphdr_t asconf; + struct sctp_chunk *retval; + int length = sizeof(asconf) + vparam_len; + union sctp_addr_param addrparam; + int addrlen; + struct sctp_af *af = sctp_get_af_specific(addr->v4.sin_family); + + addrlen = af->to_addr_param(addr, &addrparam); + if (!addrlen) + return NULL; + length += addrlen; + + /* Create the chunk. */ + retval = sctp_make_chunk(asoc, SCTP_CID_ASCONF, 0, length); + if (!retval) + return NULL; + + asconf.serial = htonl(asoc->addip_serial++); + + retval->subh.addip_hdr = + sctp_addto_chunk(retval, sizeof(asconf), &asconf); + retval->param_hdr.v = + sctp_addto_chunk(retval, addrlen, &addrparam); + + return retval; +} + +/* ADDIP + * 3.2.1 Add IP Address + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 0xC001 | Length = Variable | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF-Request Correlation ID | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Address Parameter | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * 3.2.2 Delete IP Address + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 0xC002 | Length = Variable | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF-Request Correlation ID | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Address Parameter | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + */ +struct sctp_chunk *sctp_make_asconf_update_ip(struct sctp_association *asoc, + union sctp_addr *laddr, + struct sockaddr *addrs, + int addrcnt, + __u16 flags) +{ + sctp_addip_param_t param; + struct sctp_chunk *retval; + union sctp_addr_param addr_param; + union sctp_addr *addr; + void *addr_buf; + struct sctp_af *af; + int paramlen = sizeof(param); + int addr_param_len = 0; + int totallen = 0; + int i; + + /* Get total length of all the address parameters. */ + addr_buf = addrs; + for (i = 0; i < addrcnt; i++) { + addr = (union sctp_addr *)addr_buf; + af = sctp_get_af_specific(addr->v4.sin_family); + addr_param_len = af->to_addr_param(addr, &addr_param); + + totallen += paramlen; + totallen += addr_param_len; + + addr_buf += af->sockaddr_len; + } + + /* Create an asconf chunk with the required length. */ + retval = sctp_make_asconf(asoc, laddr, totallen); + if (!retval) + return NULL; + + /* Add the address parameters to the asconf chunk. */ + addr_buf = addrs; + for (i = 0; i < addrcnt; i++) { + addr = (union sctp_addr *)addr_buf; + af = sctp_get_af_specific(addr->v4.sin_family); + addr_param_len = af->to_addr_param(addr, &addr_param); + param.param_hdr.type = flags; + param.param_hdr.length = htons(paramlen + addr_param_len); + param.crr_id = i; + + sctp_addto_chunk(retval, paramlen, ¶m); + sctp_addto_chunk(retval, addr_param_len, &addr_param); + + addr_buf += af->sockaddr_len; + } + return retval; +} + +/* ADDIP + * 3.2.4 Set Primary IP Address + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type =0xC004 | Length = Variable | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF-Request Correlation ID | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Address Parameter | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * Create an ASCONF chunk with Set Primary IP address parameter. + */ +struct sctp_chunk *sctp_make_asconf_set_prim(struct sctp_association *asoc, + union sctp_addr *addr) +{ + sctp_addip_param_t param; + struct sctp_chunk *retval; + int len = sizeof(param); + union sctp_addr_param addrparam; + int addrlen; + struct sctp_af *af = sctp_get_af_specific(addr->v4.sin_family); + + addrlen = af->to_addr_param(addr, &addrparam); + if (!addrlen) + return NULL; + len += addrlen; + + /* Create the chunk and make asconf header. */ + retval = sctp_make_asconf(asoc, addr, len); + if (!retval) + return NULL; + + param.param_hdr.type = SCTP_PARAM_SET_PRIMARY; + param.param_hdr.length = htons(len); + param.crr_id = 0; + + sctp_addto_chunk(retval, sizeof(param), ¶m); + sctp_addto_chunk(retval, addrlen, &addrparam); + + return retval; +} + +/* ADDIP 3.1.2 Address Configuration Acknowledgement Chunk (ASCONF-ACK) + * 0 1 2 3 + * 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Type = 0x80 | Chunk Flags | Chunk Length | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | Serial Number | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF Parameter Response#1 | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * \ \ + * / .... / + * \ \ + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * | ASCONF Parameter Response#N | + * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + * + * Create an ASCONF_ACK chunk with enough space for the parameter responses. + */ +static struct sctp_chunk *sctp_make_asconf_ack(const struct sctp_association *asoc, + __u32 serial, int vparam_len) +{ + sctp_addiphdr_t asconf; + struct sctp_chunk *retval; + int length = sizeof(asconf) + vparam_len; + + /* Create the chunk. */ + retval = sctp_make_chunk(asoc, SCTP_CID_ASCONF_ACK, 0, length); + if (!retval) + return NULL; + + asconf.serial = htonl(serial); + + retval->subh.addip_hdr = + sctp_addto_chunk(retval, sizeof(asconf), &asconf); + + return retval; +} + +/* Add response parameters to an ASCONF_ACK chunk. */ +static void sctp_add_asconf_response(struct sctp_chunk *chunk, __u32 crr_id, + __u16 err_code, sctp_addip_param_t *asconf_param) +{ + sctp_addip_param_t ack_param; + sctp_errhdr_t err_param; + int asconf_param_len = 0; + int err_param_len = 0; + __u16 response_type; + + if (SCTP_ERROR_NO_ERROR == err_code) { + response_type = SCTP_PARAM_SUCCESS_REPORT; + } else { + response_type = SCTP_PARAM_ERR_CAUSE; + err_param_len = sizeof(err_param); + if (asconf_param) + asconf_param_len = + ntohs(asconf_param->param_hdr.length); + } + + /* Add Success Indication or Error Cause Indication parameter. */ + ack_param.param_hdr.type = response_type; + ack_param.param_hdr.length = htons(sizeof(ack_param) + + err_param_len + + asconf_param_len); + ack_param.crr_id = crr_id; + sctp_addto_chunk(chunk, sizeof(ack_param), &ack_param); + + if (SCTP_ERROR_NO_ERROR == err_code) + return; + + /* Add Error Cause parameter. */ + err_param.cause = err_code; + err_param.length = htons(err_param_len + asconf_param_len); + sctp_addto_chunk(chunk, err_param_len, &err_param); + + /* Add the failed TLV copied from ASCONF chunk. */ + if (asconf_param) + sctp_addto_chunk(chunk, asconf_param_len, asconf_param); +} + +/* Process a asconf parameter. */ +static __u16 sctp_process_asconf_param(struct sctp_association *asoc, + struct sctp_chunk *asconf, + sctp_addip_param_t *asconf_param) +{ + struct sctp_transport *peer; + struct sctp_af *af; + union sctp_addr addr; + struct list_head *pos; + union sctp_addr_param *addr_param; + + addr_param = (union sctp_addr_param *) + ((void *)asconf_param + sizeof(sctp_addip_param_t)); + + af = sctp_get_af_specific(param_type2af(addr_param->v4.param_hdr.type)); + if (unlikely(!af)) + return SCTP_ERROR_INV_PARAM; + + af->from_addr_param(&addr, addr_param, asoc->peer.port, 0); + switch (asconf_param->param_hdr.type) { + case SCTP_PARAM_ADD_IP: + /* ADDIP 4.3 D9) If an endpoint receives an ADD IP address + * request and does not have the local resources to add this + * new address to the association, it MUST return an Error + * Cause TLV set to the new error code 'Operation Refused + * Due to Resource Shortage'. + */ + + peer = sctp_assoc_add_peer(asoc, &addr, GFP_ATOMIC); + if (!peer) + return SCTP_ERROR_RSRC_LOW; + + /* Start the heartbeat timer. */ + if (!mod_timer(&peer->hb_timer, sctp_transport_timeout(peer))) + sctp_transport_hold(peer); + break; + case SCTP_PARAM_DEL_IP: + /* ADDIP 4.3 D7) If a request is received to delete the + * last remaining IP address of a peer endpoint, the receiver + * MUST send an Error Cause TLV with the error cause set to the + * new error code 'Request to Delete Last Remaining IP Address'. + */ + pos = asoc->peer.transport_addr_list.next; + if (pos->next == &asoc->peer.transport_addr_list) + return SCTP_ERROR_DEL_LAST_IP; + + /* ADDIP 4.3 D8) If a request is received to delete an IP + * address which is also the source address of the IP packet + * which contained the ASCONF chunk, the receiver MUST reject + * this request. To reject the request the receiver MUST send + * an Error Cause TLV set to the new error code 'Request to + * Delete Source IP Address' + */ + if (sctp_cmp_addr_exact(sctp_source(asconf), &addr)) + return SCTP_ERROR_DEL_SRC_IP; + + sctp_assoc_del_peer(asoc, &addr); + break; + case SCTP_PARAM_SET_PRIMARY: + peer = sctp_assoc_lookup_paddr(asoc, &addr); + if (!peer) + return SCTP_ERROR_INV_PARAM; + + sctp_assoc_set_primary(asoc, peer); + break; + default: + return SCTP_ERROR_INV_PARAM; + break; + } + + return SCTP_ERROR_NO_ERROR; +} + +/* Process an incoming ASCONF chunk with the next expected serial no. and + * return an ASCONF_ACK chunk to be sent in response. + */ +struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc, + struct sctp_chunk *asconf) +{ + sctp_addiphdr_t *hdr; + union sctp_addr_param *addr_param; + sctp_addip_param_t *asconf_param; + struct sctp_chunk *asconf_ack; + + __u16 err_code; + int length = 0; + int chunk_len = asconf->skb->len; + __u32 serial; + int all_param_pass = 1; + + hdr = (sctp_addiphdr_t *)asconf->skb->data; + serial = ntohl(hdr->serial); + + /* Skip the addiphdr and store a pointer to address parameter. */ + length = sizeof(sctp_addiphdr_t); + addr_param = (union sctp_addr_param *)(asconf->skb->data + length); + chunk_len -= length; + + /* Skip the address parameter and store a pointer to the first + * asconf paramter. + */ + length = ntohs(addr_param->v4.param_hdr.length); + asconf_param = (sctp_addip_param_t *)((void *)addr_param + length); + chunk_len -= length; + + /* create an ASCONF_ACK chunk. + * Based on the definitions of parameters, we know that the size of + * ASCONF_ACK parameters are less than or equal to the twice of ASCONF + * paramters. + */ + asconf_ack = sctp_make_asconf_ack(asoc, serial, chunk_len * 2); + if (!asconf_ack) + goto done; + + /* Process the TLVs contained within the ASCONF chunk. */ + while (chunk_len > 0) { + err_code = sctp_process_asconf_param(asoc, asconf, + asconf_param); + /* ADDIP 4.1 A7) + * If an error response is received for a TLV parameter, + * all TLVs with no response before the failed TLV are + * considered successful if not reported. All TLVs after + * the failed response are considered unsuccessful unless + * a specific success indication is present for the parameter. + */ + if (SCTP_ERROR_NO_ERROR != err_code) + all_param_pass = 0; + + if (!all_param_pass) + sctp_add_asconf_response(asconf_ack, + asconf_param->crr_id, err_code, + asconf_param); + + /* ADDIP 4.3 D11) When an endpoint receiving an ASCONF to add + * an IP address sends an 'Out of Resource' in its response, it + * MUST also fail any subsequent add or delete requests bundled + * in the ASCONF. + */ + if (SCTP_ERROR_RSRC_LOW == err_code) + goto done; + + /* Move to the next ASCONF param. */ + length = ntohs(asconf_param->param_hdr.length); + asconf_param = (sctp_addip_param_t *)((void *)asconf_param + + length); + chunk_len -= length; + } + +done: + asoc->peer.addip_serial++; + + /* If we are sending a new ASCONF_ACK hold a reference to it in assoc + * after freeing the reference to old asconf ack if any. + */ + if (asconf_ack) { + if (asoc->addip_last_asconf_ack) + sctp_chunk_free(asoc->addip_last_asconf_ack); + + sctp_chunk_hold(asconf_ack); + asoc->addip_last_asconf_ack = asconf_ack; + } + + return asconf_ack; +} + +/* Process a asconf parameter that is successfully acked. */ +static int sctp_asconf_param_success(struct sctp_association *asoc, + sctp_addip_param_t *asconf_param) +{ + struct sctp_af *af; + union sctp_addr addr; + struct sctp_bind_addr *bp = &asoc->base.bind_addr; + union sctp_addr_param *addr_param; + struct list_head *pos; + struct sctp_transport *transport; + int retval = 0; + + addr_param = (union sctp_addr_param *) + ((void *)asconf_param + sizeof(sctp_addip_param_t)); + + /* We have checked the packet before, so we do not check again. */ + af = sctp_get_af_specific(param_type2af(addr_param->v4.param_hdr.type)); + af->from_addr_param(&addr, addr_param, bp->port, 0); + + switch (asconf_param->param_hdr.type) { + case SCTP_PARAM_ADD_IP: + sctp_local_bh_disable(); + sctp_write_lock(&asoc->base.addr_lock); + retval = sctp_add_bind_addr(bp, &addr, GFP_ATOMIC); + sctp_write_unlock(&asoc->base.addr_lock); + sctp_local_bh_enable(); + break; + case SCTP_PARAM_DEL_IP: + sctp_local_bh_disable(); + sctp_write_lock(&asoc->base.addr_lock); + retval = sctp_del_bind_addr(bp, &addr); + sctp_write_unlock(&asoc->base.addr_lock); + sctp_local_bh_enable(); + list_for_each(pos, &asoc->peer.transport_addr_list) { + transport = list_entry(pos, struct sctp_transport, + transports); + sctp_transport_route(transport, NULL, + sctp_sk(asoc->base.sk)); + } + break; + default: + break; + } + + return retval; +} + +/* Get the corresponding ASCONF response error code from the ASCONF_ACK chunk + * for the given asconf parameter. If there is no response for this parameter, + * return the error code based on the third argument 'no_err'. + * ADDIP 4.1 + * A7) If an error response is received for a TLV parameter, all TLVs with no + * response before the failed TLV are considered successful if not reported. + * All TLVs after the failed response are considered unsuccessful unless a + * specific success indication is present for the parameter. + */ +static __u16 sctp_get_asconf_response(struct sctp_chunk *asconf_ack, + sctp_addip_param_t *asconf_param, + int no_err) +{ + sctp_addip_param_t *asconf_ack_param; + sctp_errhdr_t *err_param; + int length; + int asconf_ack_len = asconf_ack->skb->len; + __u16 err_code; + + if (no_err) + err_code = SCTP_ERROR_NO_ERROR; + else + err_code = SCTP_ERROR_REQ_REFUSED; + + /* Skip the addiphdr from the asconf_ack chunk and store a pointer to + * the first asconf_ack parameter. + */ + length = sizeof(sctp_addiphdr_t); + asconf_ack_param = (sctp_addip_param_t *)(asconf_ack->skb->data + + length); + asconf_ack_len -= length; + + while (asconf_ack_len > 0) { + if (asconf_ack_param->crr_id == asconf_param->crr_id) { + switch(asconf_ack_param->param_hdr.type) { + case SCTP_PARAM_SUCCESS_REPORT: + return SCTP_ERROR_NO_ERROR; + case SCTP_PARAM_ERR_CAUSE: + length = sizeof(sctp_addip_param_t); + err_param = (sctp_errhdr_t *) + ((void *)asconf_ack_param + length); + asconf_ack_len -= length; + if (asconf_ack_len > 0) + return err_param->cause; + else + return SCTP_ERROR_INV_PARAM; + break; + default: + return SCTP_ERROR_INV_PARAM; + } + } + + length = ntohs(asconf_ack_param->param_hdr.length); + asconf_ack_param = (sctp_addip_param_t *) + ((void *)asconf_ack_param + length); + asconf_ack_len -= length; + } + + return err_code; +} + +/* Process an incoming ASCONF_ACK chunk against the cached last ASCONF chunk. */ +int sctp_process_asconf_ack(struct sctp_association *asoc, + struct sctp_chunk *asconf_ack) +{ + struct sctp_chunk *asconf = asoc->addip_last_asconf; + union sctp_addr_param *addr_param; + sctp_addip_param_t *asconf_param; + int length = 0; + int asconf_len = asconf->skb->len; + int all_param_pass = 0; + int no_err = 1; + int retval = 0; + __u16 err_code = SCTP_ERROR_NO_ERROR; + + /* Skip the chunkhdr and addiphdr from the last asconf sent and store + * a pointer to address parameter. + */ + length = sizeof(sctp_addip_chunk_t); + addr_param = (union sctp_addr_param *)(asconf->skb->data + length); + asconf_len -= length; + + /* Skip the address parameter in the last asconf sent and store a + * pointer to the first asconf paramter. + */ + length = ntohs(addr_param->v4.param_hdr.length); + asconf_param = (sctp_addip_param_t *)((void *)addr_param + length); + asconf_len -= length; + + /* ADDIP 4.1 + * A8) If there is no response(s) to specific TLV parameter(s), and no + * failures are indicated, then all request(s) are considered + * successful. + */ + if (asconf_ack->skb->len == sizeof(sctp_addiphdr_t)) + all_param_pass = 1; + + /* Process the TLVs contained in the last sent ASCONF chunk. */ + while (asconf_len > 0) { + if (all_param_pass) + err_code = SCTP_ERROR_NO_ERROR; + else { + err_code = sctp_get_asconf_response(asconf_ack, + asconf_param, + no_err); + if (no_err && (SCTP_ERROR_NO_ERROR != err_code)) + no_err = 0; + } + + switch (err_code) { + case SCTP_ERROR_NO_ERROR: + retval = sctp_asconf_param_success(asoc, asconf_param); + break; + + case SCTP_ERROR_RSRC_LOW: + retval = 1; + break; + + case SCTP_ERROR_INV_PARAM: + /* Disable sending this type of asconf parameter in + * future. + */ + asoc->peer.addip_disabled_mask |= + asconf_param->param_hdr.type; + break; + + case SCTP_ERROR_REQ_REFUSED: + case SCTP_ERROR_DEL_LAST_IP: + case SCTP_ERROR_DEL_SRC_IP: + default: + break; + } + + /* Skip the processed asconf parameter and move to the next + * one. + */ + length = ntohs(asconf_param->param_hdr.length); + asconf_param = (sctp_addip_param_t *)((void *)asconf_param + + length); + asconf_len -= length; + } + + /* Free the cached last sent asconf chunk. */ + sctp_chunk_free(asconf); + asoc->addip_last_asconf = NULL; + + /* Send the next asconf chunk from the addip chunk queue. */ + asconf = (struct sctp_chunk *)__skb_dequeue(&asoc->addip_chunks); + if (asconf) { + /* Hold the chunk until an ASCONF_ACK is received. */ + sctp_chunk_hold(asconf); + if (sctp_primitive_ASCONF(asoc, asconf)) + sctp_chunk_free(asconf); + else + asoc->addip_last_asconf = asconf; + } + + return retval; +} + +/* Make a FWD TSN chunk. */ +struct sctp_chunk *sctp_make_fwdtsn(const struct sctp_association *asoc, + __u32 new_cum_tsn, size_t nstreams, + struct sctp_fwdtsn_skip *skiplist) +{ + struct sctp_chunk *retval = NULL; + struct sctp_fwdtsn_chunk *ftsn_chunk; + struct sctp_fwdtsn_hdr ftsn_hdr; + struct sctp_fwdtsn_skip skip; + size_t hint; + int i; + + hint = (nstreams + 1) * sizeof(__u32); + + /* Maybe set the T-bit if we have no association. */ + retval = sctp_make_chunk(asoc, SCTP_CID_FWD_TSN, 0, hint); + + if (!retval) + return NULL; + + ftsn_chunk = (struct sctp_fwdtsn_chunk *)retval->subh.fwdtsn_hdr; + + ftsn_hdr.new_cum_tsn = htonl(new_cum_tsn); + retval->subh.fwdtsn_hdr = + sctp_addto_chunk(retval, sizeof(ftsn_hdr), &ftsn_hdr); + + for (i = 0; i < nstreams; i++) { + skip.stream = skiplist[i].stream; + skip.ssn = skiplist[i].ssn; + sctp_addto_chunk(retval, sizeof(skip), &skip); + } + + return retval; +} diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c new file mode 100644 index 000000000000..f65fa441952f --- /dev/null +++ b/net/sctp/sm_sideeffect.c @@ -0,0 +1,1395 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions work with the state functions in sctp_sm_statefuns.c + * to implement that state operations. These functions implement the + * steps which require modifying existing data structures. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@austin.ibm.com> + * Hui Huang <hui.huang@nokia.com> + * Dajiang Zhang <dajiang.zhang@nokia.com> + * Daisy Chang <daisyc@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/skbuff.h> +#include <linux/types.h> +#include <linux/socket.h> +#include <linux/ip.h> +#include <net/sock.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +static int sctp_cmd_interpreter(sctp_event_t event_type, + sctp_subtype_t subtype, + sctp_state_t state, + struct sctp_endpoint *ep, + struct sctp_association *asoc, + void *event_arg, + sctp_disposition_t status, + sctp_cmd_seq_t *commands, + int gfp); +static int sctp_side_effects(sctp_event_t event_type, sctp_subtype_t subtype, + sctp_state_t state, + struct sctp_endpoint *ep, + struct sctp_association *asoc, + void *event_arg, + sctp_disposition_t status, + sctp_cmd_seq_t *commands, + int gfp); + +/******************************************************************** + * Helper functions + ********************************************************************/ + +/* A helper function for delayed processing of INET ECN CE bit. */ +static void sctp_do_ecn_ce_work(struct sctp_association *asoc, + __u32 lowest_tsn) +{ + /* Save the TSN away for comparison when we receive CWR */ + + asoc->last_ecne_tsn = lowest_tsn; + asoc->need_ecne = 1; +} + +/* Helper function for delayed processing of SCTP ECNE chunk. */ +/* RFC 2960 Appendix A + * + * RFC 2481 details a specific bit for a sender to send in + * the header of its next outbound TCP segment to indicate to + * its peer that it has reduced its congestion window. This + * is termed the CWR bit. For SCTP the same indication is made + * by including the CWR chunk. This chunk contains one data + * element, i.e. the TSN number that was sent in the ECNE chunk. + * This element represents the lowest TSN number in the datagram + * that was originally marked with the CE bit. + */ +static struct sctp_chunk *sctp_do_ecn_ecne_work(struct sctp_association *asoc, + __u32 lowest_tsn, + struct sctp_chunk *chunk) +{ + struct sctp_chunk *repl; + + /* Our previously transmitted packet ran into some congestion + * so we should take action by reducing cwnd and ssthresh + * and then ACK our peer that we we've done so by + * sending a CWR. + */ + + /* First, try to determine if we want to actually lower + * our cwnd variables. Only lower them if the ECNE looks more + * recent than the last response. + */ + if (TSN_lt(asoc->last_cwr_tsn, lowest_tsn)) { + struct sctp_transport *transport; + + /* Find which transport's congestion variables + * need to be adjusted. + */ + transport = sctp_assoc_lookup_tsn(asoc, lowest_tsn); + + /* Update the congestion variables. */ + if (transport) + sctp_transport_lower_cwnd(transport, + SCTP_LOWER_CWND_ECNE); + asoc->last_cwr_tsn = lowest_tsn; + } + + /* Always try to quiet the other end. In case of lost CWR, + * resend last_cwr_tsn. + */ + repl = sctp_make_cwr(asoc, asoc->last_cwr_tsn, chunk); + + /* If we run out of memory, it will look like a lost CWR. We'll + * get back in sync eventually. + */ + return repl; +} + +/* Helper function to do delayed processing of ECN CWR chunk. */ +static void sctp_do_ecn_cwr_work(struct sctp_association *asoc, + __u32 lowest_tsn) +{ + /* Turn off ECNE getting auto-prepended to every outgoing + * packet + */ + asoc->need_ecne = 0; +} + +/* Generate SACK if necessary. We call this at the end of a packet. */ +static int sctp_gen_sack(struct sctp_association *asoc, int force, + sctp_cmd_seq_t *commands) +{ + __u32 ctsn, max_tsn_seen; + struct sctp_chunk *sack; + int error = 0; + + if (force) + asoc->peer.sack_needed = 1; + + ctsn = sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map); + max_tsn_seen = sctp_tsnmap_get_max_tsn_seen(&asoc->peer.tsn_map); + + /* From 12.2 Parameters necessary per association (i.e. the TCB): + * + * Ack State : This flag indicates if the next received packet + * : is to be responded to with a SACK. ... + * : When DATA chunks are out of order, SACK's + * : are not delayed (see Section 6). + * + * [This is actually not mentioned in Section 6, but we + * implement it here anyway. --piggy] + */ + if (max_tsn_seen != ctsn) + asoc->peer.sack_needed = 1; + + /* From 6.2 Acknowledgement on Reception of DATA Chunks: + * + * Section 4.2 of [RFC2581] SHOULD be followed. Specifically, + * an acknowledgement SHOULD be generated for at least every + * second packet (not every second DATA chunk) received, and + * SHOULD be generated within 200 ms of the arrival of any + * unacknowledged DATA chunk. ... + */ + if (!asoc->peer.sack_needed) { + /* We will need a SACK for the next packet. */ + asoc->peer.sack_needed = 1; + goto out; + } else { + if (asoc->a_rwnd > asoc->rwnd) + asoc->a_rwnd = asoc->rwnd; + sack = sctp_make_sack(asoc); + if (!sack) + goto nomem; + + asoc->peer.sack_needed = 0; + + error = sctp_outq_tail(&asoc->outqueue, sack); + + /* Stop the SACK timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_SACK)); + } +out: + return error; +nomem: + error = -ENOMEM; + return error; +} + +/* When the T3-RTX timer expires, it calls this function to create the + * relevant state machine event. + */ +void sctp_generate_t3_rtx_event(unsigned long peer) +{ + int error; + struct sctp_transport *transport = (struct sctp_transport *) peer; + struct sctp_association *asoc = transport->asoc; + + /* Check whether a task is in the sock. */ + + sctp_bh_lock_sock(asoc->base.sk); + if (sock_owned_by_user(asoc->base.sk)) { + SCTP_DEBUG_PRINTK("%s:Sock is busy.\n", __FUNCTION__); + + /* Try again later. */ + if (!mod_timer(&transport->T3_rtx_timer, jiffies + (HZ/20))) + sctp_transport_hold(transport); + goto out_unlock; + } + + /* Is this transport really dead and just waiting around for + * the timer to let go of the reference? + */ + if (transport->dead) + goto out_unlock; + + /* Run through the state machine. */ + error = sctp_do_sm(SCTP_EVENT_T_TIMEOUT, + SCTP_ST_TIMEOUT(SCTP_EVENT_TIMEOUT_T3_RTX), + asoc->state, + asoc->ep, asoc, + transport, GFP_ATOMIC); + + if (error) + asoc->base.sk->sk_err = -error; + +out_unlock: + sctp_bh_unlock_sock(asoc->base.sk); + sctp_transport_put(transport); +} + +/* This is a sa interface for producing timeout events. It works + * for timeouts which use the association as their parameter. + */ +static void sctp_generate_timeout_event(struct sctp_association *asoc, + sctp_event_timeout_t timeout_type) +{ + int error = 0; + + sctp_bh_lock_sock(asoc->base.sk); + if (sock_owned_by_user(asoc->base.sk)) { + SCTP_DEBUG_PRINTK("%s:Sock is busy: timer %d\n", + __FUNCTION__, + timeout_type); + + /* Try again later. */ + if (!mod_timer(&asoc->timers[timeout_type], jiffies + (HZ/20))) + sctp_association_hold(asoc); + goto out_unlock; + } + + /* Is this association really dead and just waiting around for + * the timer to let go of the reference? + */ + if (asoc->base.dead) + goto out_unlock; + + /* Run through the state machine. */ + error = sctp_do_sm(SCTP_EVENT_T_TIMEOUT, + SCTP_ST_TIMEOUT(timeout_type), + asoc->state, asoc->ep, asoc, + (void *)timeout_type, GFP_ATOMIC); + + if (error) + asoc->base.sk->sk_err = -error; + +out_unlock: + sctp_bh_unlock_sock(asoc->base.sk); + sctp_association_put(asoc); +} + +static void sctp_generate_t1_cookie_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *) data; + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T1_COOKIE); +} + +static void sctp_generate_t1_init_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *) data; + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T1_INIT); +} + +static void sctp_generate_t2_shutdown_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *) data; + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T2_SHUTDOWN); +} + +static void sctp_generate_t4_rto_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *) data; + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_T4_RTO); +} + +static void sctp_generate_t5_shutdown_guard_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *)data; + sctp_generate_timeout_event(asoc, + SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD); + +} /* sctp_generate_t5_shutdown_guard_event() */ + +static void sctp_generate_autoclose_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *) data; + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_AUTOCLOSE); +} + +/* Generate a heart beat event. If the sock is busy, reschedule. Make + * sure that the transport is still valid. + */ +void sctp_generate_heartbeat_event(unsigned long data) +{ + int error = 0; + struct sctp_transport *transport = (struct sctp_transport *) data; + struct sctp_association *asoc = transport->asoc; + + sctp_bh_lock_sock(asoc->base.sk); + if (sock_owned_by_user(asoc->base.sk)) { + SCTP_DEBUG_PRINTK("%s:Sock is busy.\n", __FUNCTION__); + + /* Try again later. */ + if (!mod_timer(&transport->hb_timer, jiffies + (HZ/20))) + sctp_transport_hold(transport); + goto out_unlock; + } + + /* Is this structure just waiting around for us to actually + * get destroyed? + */ + if (transport->dead) + goto out_unlock; + + error = sctp_do_sm(SCTP_EVENT_T_TIMEOUT, + SCTP_ST_TIMEOUT(SCTP_EVENT_TIMEOUT_HEARTBEAT), + asoc->state, asoc->ep, asoc, + transport, GFP_ATOMIC); + + if (error) + asoc->base.sk->sk_err = -error; + +out_unlock: + sctp_bh_unlock_sock(asoc->base.sk); + sctp_transport_put(transport); +} + +/* Inject a SACK Timeout event into the state machine. */ +static void sctp_generate_sack_event(unsigned long data) +{ + struct sctp_association *asoc = (struct sctp_association *) data; + sctp_generate_timeout_event(asoc, SCTP_EVENT_TIMEOUT_SACK); +} + +sctp_timer_event_t *sctp_timer_events[SCTP_NUM_TIMEOUT_TYPES] = { + NULL, + sctp_generate_t1_cookie_event, + sctp_generate_t1_init_event, + sctp_generate_t2_shutdown_event, + NULL, + sctp_generate_t4_rto_event, + sctp_generate_t5_shutdown_guard_event, + sctp_generate_heartbeat_event, + sctp_generate_sack_event, + sctp_generate_autoclose_event, +}; + + +/* RFC 2960 8.2 Path Failure Detection + * + * When its peer endpoint is multi-homed, an endpoint should keep a + * error counter for each of the destination transport addresses of the + * peer endpoint. + * + * Each time the T3-rtx timer expires on any address, or when a + * HEARTBEAT sent to an idle address is not acknowledged within a RTO, + * the error counter of that destination address will be incremented. + * When the value in the error counter exceeds the protocol parameter + * 'Path.Max.Retrans' of that destination address, the endpoint should + * mark the destination transport address as inactive, and a + * notification SHOULD be sent to the upper layer. + * + */ +static void sctp_do_8_2_transport_strike(struct sctp_association *asoc, + struct sctp_transport *transport) +{ + /* The check for association's overall error counter exceeding the + * threshold is done in the state function. + */ + asoc->overall_error_count++; + + if (transport->active && + (transport->error_count++ >= transport->max_retrans)) { + SCTP_DEBUG_PRINTK("transport_strike: transport " + "IP:%d.%d.%d.%d failed.\n", + NIPQUAD(transport->ipaddr.v4.sin_addr)); + sctp_assoc_control_transport(asoc, transport, + SCTP_TRANSPORT_DOWN, + SCTP_FAILED_THRESHOLD); + } + + /* E2) For the destination address for which the timer + * expires, set RTO <- RTO * 2 ("back off the timer"). The + * maximum value discussed in rule C7 above (RTO.max) may be + * used to provide an upper bound to this doubling operation. + */ + transport->rto = min((transport->rto * 2), transport->asoc->rto_max); +} + +/* Worker routine to handle INIT command failure. */ +static void sctp_cmd_init_failed(sctp_cmd_seq_t *commands, + struct sctp_association *asoc, + unsigned error) +{ + struct sctp_ulpevent *event; + + event = sctp_ulpevent_make_assoc_change(asoc,0, SCTP_CANT_STR_ASSOC, + (__u16)error, 0, 0, + GFP_ATOMIC); + + if (event) + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(event)); + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + + /* SEND_FAILED sent later when cleaning up the association. */ + asoc->outqueue.error = error; + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); +} + +/* Worker routine to handle SCTP_CMD_ASSOC_FAILED. */ +static void sctp_cmd_assoc_failed(sctp_cmd_seq_t *commands, + struct sctp_association *asoc, + sctp_event_t event_type, + sctp_subtype_t subtype, + struct sctp_chunk *chunk, + unsigned error) +{ + struct sctp_ulpevent *event; + + /* Cancel any partial delivery in progress. */ + sctp_ulpq_abort_pd(&asoc->ulpq, GFP_ATOMIC); + + event = sctp_ulpevent_make_assoc_change(asoc, 0, SCTP_COMM_LOST, + (__u16)error, 0, 0, + GFP_ATOMIC); + if (event) + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(event)); + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + + /* Set sk_err to ECONNRESET on a 1-1 style socket. */ + if (!sctp_style(asoc->base.sk, UDP)) + asoc->base.sk->sk_err = ECONNRESET; + + /* SEND_FAILED sent later when cleaning up the association. */ + asoc->outqueue.error = error; + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); +} + +/* Process an init chunk (may be real INIT/INIT-ACK or an embedded INIT + * inside the cookie. In reality, this is only used for INIT-ACK processing + * since all other cases use "temporary" associations and can do all + * their work in statefuns directly. + */ +static int sctp_cmd_process_init(sctp_cmd_seq_t *commands, + struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_init_chunk_t *peer_init, int gfp) +{ + int error; + + /* We only process the init as a sideeffect in a single + * case. This is when we process the INIT-ACK. If we + * fail during INIT processing (due to malloc problems), + * just return the error and stop processing the stack. + */ + if (!sctp_process_init(asoc, chunk->chunk_hdr->type, + sctp_source(chunk), peer_init, gfp)) + error = -ENOMEM; + else + error = 0; + + return error; +} + +/* Helper function to break out starting up of heartbeat timers. */ +static void sctp_cmd_hb_timers_start(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc) +{ + struct sctp_transport *t; + struct list_head *pos; + + /* Start a heartbeat timer for each transport on the association. + * hold a reference on the transport to make sure none of + * the needed data structures go away. + */ + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + + if (!mod_timer(&t->hb_timer, sctp_transport_timeout(t))) + sctp_transport_hold(t); + } +} + +static void sctp_cmd_hb_timers_stop(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc) +{ + struct sctp_transport *t; + struct list_head *pos; + + /* Stop all heartbeat timers. */ + + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + if (del_timer(&t->hb_timer)) + sctp_transport_put(t); + } +} + +/* Helper function to stop any pending T3-RTX timers */ +static void sctp_cmd_t3_rtx_timers_stop(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc) +{ + struct sctp_transport *t; + struct list_head *pos; + + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + if (timer_pending(&t->T3_rtx_timer) && + del_timer(&t->T3_rtx_timer)) { + sctp_transport_put(t); + } + } +} + + +/* Helper function to update the heartbeat timer. */ +static void sctp_cmd_hb_timer_update(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_transport *t) +{ + /* Update the heartbeat timer. */ + if (!mod_timer(&t->hb_timer, sctp_transport_timeout(t))) + sctp_transport_hold(t); +} + +/* Helper function to handle the reception of an HEARTBEAT ACK. */ +static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_transport *t, + struct sctp_chunk *chunk) +{ + sctp_sender_hb_info_t *hbinfo; + + /* 8.3 Upon the receipt of the HEARTBEAT ACK, the sender of the + * HEARTBEAT should clear the error counter of the destination + * transport address to which the HEARTBEAT was sent. + * The association's overall error count is also cleared. + */ + t->error_count = 0; + t->asoc->overall_error_count = 0; + + /* Mark the destination transport address as active if it is not so + * marked. + */ + if (!t->active) + sctp_assoc_control_transport(asoc, t, SCTP_TRANSPORT_UP, + SCTP_HEARTBEAT_SUCCESS); + + /* The receiver of the HEARTBEAT ACK should also perform an + * RTT measurement for that destination transport address + * using the time value carried in the HEARTBEAT ACK chunk. + */ + hbinfo = (sctp_sender_hb_info_t *) chunk->skb->data; + sctp_transport_update_rto(t, (jiffies - hbinfo->sent_at)); +} + +/* Helper function to do a transport reset at the expiry of the hearbeat + * timer. + */ +static void sctp_cmd_transport_reset(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_transport *t) +{ + sctp_transport_lower_cwnd(t, SCTP_LOWER_CWND_INACTIVE); + + /* Mark one strike against a transport. */ + sctp_do_8_2_transport_strike(asoc, t); +} + +/* Helper function to process the process SACK command. */ +static int sctp_cmd_process_sack(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_sackhdr *sackh) +{ + int err; + + if (sctp_outq_sack(&asoc->outqueue, sackh)) { + /* There are no more TSNs awaiting SACK. */ + err = sctp_do_sm(SCTP_EVENT_T_OTHER, + SCTP_ST_OTHER(SCTP_EVENT_NO_PENDING_TSN), + asoc->state, asoc->ep, asoc, NULL, + GFP_ATOMIC); + } else { + /* Windows may have opened, so we need + * to check if we have DATA to transmit + */ + err = sctp_outq_flush(&asoc->outqueue, 0); + } + + return err; +} + +/* Helper function to set the timeout value for T2-SHUTDOWN timer and to set + * the transport for a shutdown chunk. + */ +static void sctp_cmd_setup_t2(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_chunk *chunk) +{ + struct sctp_transport *t; + + t = sctp_assoc_choose_shutdown_transport(asoc); + asoc->shutdown_last_sent_to = t; + asoc->timeouts[SCTP_EVENT_TIMEOUT_T2_SHUTDOWN] = t->rto; + chunk->transport = t; +} + +/* Helper function to change the state of an association. */ +static void sctp_cmd_new_state(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + sctp_state_t state) +{ + struct sock *sk = asoc->base.sk; + + asoc->state = state; + + if (sctp_style(sk, TCP)) { + /* Change the sk->sk_state of a TCP-style socket that has + * sucessfully completed a connect() call. + */ + if (sctp_state(asoc, ESTABLISHED) && sctp_sstate(sk, CLOSED)) + sk->sk_state = SCTP_SS_ESTABLISHED; + + /* Set the RCV_SHUTDOWN flag when a SHUTDOWN is received. */ + if (sctp_state(asoc, SHUTDOWN_RECEIVED) && + sctp_sstate(sk, ESTABLISHED)) + sk->sk_shutdown |= RCV_SHUTDOWN; + } + + if (sctp_state(asoc, ESTABLISHED) || + sctp_state(asoc, CLOSED) || + sctp_state(asoc, SHUTDOWN_RECEIVED)) { + /* Wake up any processes waiting in the asoc's wait queue in + * sctp_wait_for_connect() or sctp_wait_for_sndbuf(). + */ + if (waitqueue_active(&asoc->wait)) + wake_up_interruptible(&asoc->wait); + + /* Wake up any processes waiting in the sk's sleep queue of + * a TCP-style or UDP-style peeled-off socket in + * sctp_wait_for_accept() or sctp_wait_for_packet(). + * For a UDP-style socket, the waiters are woken up by the + * notifications. + */ + if (!sctp_style(sk, UDP)) + sk->sk_state_change(sk); + } +} + +/* Helper function to delete an association. */ +static void sctp_cmd_delete_tcb(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc) +{ + struct sock *sk = asoc->base.sk; + + /* If it is a non-temporary association belonging to a TCP-style + * listening socket that is not closed, do not free it so that accept() + * can pick it up later. + */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING) && + (!asoc->temp) && (sk->sk_shutdown != SHUTDOWN_MASK)) + return; + + sctp_unhash_established(asoc); + sctp_association_free(asoc); +} + +/* + * ADDIP Section 4.1 ASCONF Chunk Procedures + * A4) Start a T-4 RTO timer, using the RTO value of the selected + * destination address (we use active path instead of primary path just + * because primary path may be inactive. + */ +static void sctp_cmd_setup_t4(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_chunk *chunk) +{ + struct sctp_transport *t; + + t = asoc->peer.active_path; + asoc->timeouts[SCTP_EVENT_TIMEOUT_T4_RTO] = t->rto; + chunk->transport = t; +} + +/* Process an incoming Operation Error Chunk. */ +static void sctp_cmd_process_operr(sctp_cmd_seq_t *cmds, + struct sctp_association *asoc, + struct sctp_chunk *chunk) +{ + struct sctp_operr_chunk *operr_chunk; + struct sctp_errhdr *err_hdr; + + operr_chunk = (struct sctp_operr_chunk *)chunk->chunk_hdr; + err_hdr = &operr_chunk->err_hdr; + + switch (err_hdr->cause) { + case SCTP_ERROR_UNKNOWN_CHUNK: + { + struct sctp_chunkhdr *unk_chunk_hdr; + + unk_chunk_hdr = (struct sctp_chunkhdr *)err_hdr->variable; + switch (unk_chunk_hdr->type) { + /* ADDIP 4.1 A9) If the peer responds to an ASCONF with an + * ERROR chunk reporting that it did not recognized the ASCONF + * chunk type, the sender of the ASCONF MUST NOT send any + * further ASCONF chunks and MUST stop its T-4 timer. + */ + case SCTP_CID_ASCONF: + asoc->peer.asconf_capable = 0; + sctp_add_cmd_sf(cmds, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T4_RTO)); + break; + default: + break; + } + break; + } + default: + break; + } +} + +/* Process variable FWDTSN chunk information. */ +static void sctp_cmd_process_fwdtsn(struct sctp_ulpq *ulpq, + struct sctp_chunk *chunk) +{ + struct sctp_fwdtsn_skip *skip; + /* Walk through all the skipped SSNs */ + sctp_walk_fwdtsn(skip, chunk) { + sctp_ulpq_skip(ulpq, ntohs(skip->stream), ntohs(skip->ssn)); + } + + return; +} + +/* Helper function to remove the association non-primary peer + * transports. + */ +static void sctp_cmd_del_non_primary(struct sctp_association *asoc) +{ + struct sctp_transport *t; + struct list_head *pos; + struct list_head *temp; + + list_for_each_safe(pos, temp, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, transports); + if (!sctp_cmp_addr_exact(&t->ipaddr, + &asoc->peer.primary_addr)) { + sctp_assoc_del_peer(asoc, &t->ipaddr); + } + } + + return; +} + +/* These three macros allow us to pull the debugging code out of the + * main flow of sctp_do_sm() to keep attention focused on the real + * functionality there. + */ +#define DEBUG_PRE \ + SCTP_DEBUG_PRINTK("sctp_do_sm prefn: " \ + "ep %p, %s, %s, asoc %p[%s], %s\n", \ + ep, sctp_evttype_tbl[event_type], \ + (*debug_fn)(subtype), asoc, \ + sctp_state_tbl[state], state_fn->name) + +#define DEBUG_POST \ + SCTP_DEBUG_PRINTK("sctp_do_sm postfn: " \ + "asoc %p, status: %s\n", \ + asoc, sctp_status_tbl[status]) + +#define DEBUG_POST_SFX \ + SCTP_DEBUG_PRINTK("sctp_do_sm post sfx: error %d, asoc %p[%s]\n", \ + error, asoc, \ + sctp_state_tbl[(asoc && sctp_id2assoc(ep->base.sk, \ + sctp_assoc2id(asoc)))?asoc->state:SCTP_STATE_CLOSED]) + +/* + * This is the master state machine processing function. + * + * If you want to understand all of lksctp, this is a + * good place to start. + */ +int sctp_do_sm(sctp_event_t event_type, sctp_subtype_t subtype, + sctp_state_t state, + struct sctp_endpoint *ep, + struct sctp_association *asoc, + void *event_arg, + int gfp) +{ + sctp_cmd_seq_t commands; + const sctp_sm_table_entry_t *state_fn; + sctp_disposition_t status; + int error = 0; + typedef const char *(printfn_t)(sctp_subtype_t); + + static printfn_t *table[] = { + NULL, sctp_cname, sctp_tname, sctp_oname, sctp_pname, + }; + printfn_t *debug_fn __attribute__ ((unused)) = table[event_type]; + + /* Look up the state function, run it, and then process the + * side effects. These three steps are the heart of lksctp. + */ + state_fn = sctp_sm_lookup_event(event_type, state, subtype); + + sctp_init_cmd_seq(&commands); + + DEBUG_PRE; + status = (*state_fn->fn)(ep, asoc, subtype, event_arg, &commands); + DEBUG_POST; + + error = sctp_side_effects(event_type, subtype, state, + ep, asoc, event_arg, status, + &commands, gfp); + DEBUG_POST_SFX; + + return error; +} + +#undef DEBUG_PRE +#undef DEBUG_POST + +/***************************************************************** + * This the master state function side effect processing function. + *****************************************************************/ +static int sctp_side_effects(sctp_event_t event_type, sctp_subtype_t subtype, + sctp_state_t state, + struct sctp_endpoint *ep, + struct sctp_association *asoc, + void *event_arg, + sctp_disposition_t status, + sctp_cmd_seq_t *commands, + int gfp) +{ + int error; + + /* FIXME - Most of the dispositions left today would be categorized + * as "exceptional" dispositions. For those dispositions, it + * may not be proper to run through any of the commands at all. + * For example, the command interpreter might be run only with + * disposition SCTP_DISPOSITION_CONSUME. + */ + if (0 != (error = sctp_cmd_interpreter(event_type, subtype, state, + ep, asoc, + event_arg, status, + commands, gfp))) + goto bail; + + switch (status) { + case SCTP_DISPOSITION_DISCARD: + SCTP_DEBUG_PRINTK("Ignored sctp protocol event - state %d, " + "event_type %d, event_id %d\n", + state, event_type, subtype.chunk); + break; + + case SCTP_DISPOSITION_NOMEM: + /* We ran out of memory, so we need to discard this + * packet. + */ + /* BUG--we should now recover some memory, probably by + * reneging... + */ + error = -ENOMEM; + break; + + case SCTP_DISPOSITION_DELETE_TCB: + /* This should now be a command. */ + break; + + case SCTP_DISPOSITION_CONSUME: + case SCTP_DISPOSITION_ABORT: + /* + * We should no longer have much work to do here as the + * real work has been done as explicit commands above. + */ + break; + + case SCTP_DISPOSITION_VIOLATION: + printk(KERN_ERR "sctp protocol violation state %d " + "chunkid %d\n", state, subtype.chunk); + break; + + case SCTP_DISPOSITION_NOT_IMPL: + printk(KERN_WARNING "sctp unimplemented feature in state %d, " + "event_type %d, event_id %d\n", + state, event_type, subtype.chunk); + break; + + case SCTP_DISPOSITION_BUG: + printk(KERN_ERR "sctp bug in state %d, " + "event_type %d, event_id %d\n", + state, event_type, subtype.chunk); + BUG(); + break; + + default: + printk(KERN_ERR "sctp impossible disposition %d " + "in state %d, event_type %d, event_id %d\n", + status, state, event_type, subtype.chunk); + BUG(); + break; + }; + +bail: + return error; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* This is the side-effect interpreter. */ +static int sctp_cmd_interpreter(sctp_event_t event_type, + sctp_subtype_t subtype, + sctp_state_t state, + struct sctp_endpoint *ep, + struct sctp_association *asoc, + void *event_arg, + sctp_disposition_t status, + sctp_cmd_seq_t *commands, + int gfp) +{ + int error = 0; + int force; + sctp_cmd_t *cmd; + struct sctp_chunk *new_obj; + struct sctp_chunk *chunk = NULL; + struct sctp_packet *packet; + struct list_head *pos; + struct timer_list *timer; + unsigned long timeout; + struct sctp_transport *t; + struct sctp_sackhdr sackh; + int local_cork = 0; + + if (SCTP_EVENT_T_TIMEOUT != event_type) + chunk = (struct sctp_chunk *) event_arg; + + /* Note: This whole file is a huge candidate for rework. + * For example, each command could either have its own handler, so + * the loop would look like: + * while (cmds) + * cmd->handle(x, y, z) + * --jgrimm + */ + while (NULL != (cmd = sctp_next_cmd(commands))) { + switch (cmd->verb) { + case SCTP_CMD_NOP: + /* Do nothing. */ + break; + + case SCTP_CMD_NEW_ASOC: + /* Register a new association. */ + if (local_cork) { + sctp_outq_uncork(&asoc->outqueue); + local_cork = 0; + } + asoc = cmd->obj.ptr; + /* Register with the endpoint. */ + sctp_endpoint_add_asoc(ep, asoc); + sctp_hash_established(asoc); + break; + + case SCTP_CMD_UPDATE_ASSOC: + sctp_assoc_update(asoc, cmd->obj.ptr); + break; + + case SCTP_CMD_PURGE_OUTQUEUE: + sctp_outq_teardown(&asoc->outqueue); + break; + + case SCTP_CMD_DELETE_TCB: + if (local_cork) { + sctp_outq_uncork(&asoc->outqueue); + local_cork = 0; + } + /* Delete the current association. */ + sctp_cmd_delete_tcb(commands, asoc); + asoc = NULL; + break; + + case SCTP_CMD_NEW_STATE: + /* Enter a new state. */ + sctp_cmd_new_state(commands, asoc, cmd->obj.state); + break; + + case SCTP_CMD_REPORT_TSN: + /* Record the arrival of a TSN. */ + sctp_tsnmap_mark(&asoc->peer.tsn_map, cmd->obj.u32); + break; + + case SCTP_CMD_REPORT_FWDTSN: + /* Move the Cumulattive TSN Ack ahead. */ + sctp_tsnmap_skip(&asoc->peer.tsn_map, cmd->obj.u32); + + /* Abort any in progress partial delivery. */ + sctp_ulpq_abort_pd(&asoc->ulpq, GFP_ATOMIC); + break; + + case SCTP_CMD_PROCESS_FWDTSN: + sctp_cmd_process_fwdtsn(&asoc->ulpq, cmd->obj.ptr); + break; + + case SCTP_CMD_GEN_SACK: + /* Generate a Selective ACK. + * The argument tells us whether to just count + * the packet and MAYBE generate a SACK, or + * force a SACK out. + */ + force = cmd->obj.i32; + error = sctp_gen_sack(asoc, force, commands); + break; + + case SCTP_CMD_PROCESS_SACK: + /* Process an inbound SACK. */ + error = sctp_cmd_process_sack(commands, asoc, + cmd->obj.ptr); + break; + + case SCTP_CMD_GEN_INIT_ACK: + /* Generate an INIT ACK chunk. */ + new_obj = sctp_make_init_ack(asoc, chunk, GFP_ATOMIC, + 0); + if (!new_obj) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(new_obj)); + break; + + case SCTP_CMD_PEER_INIT: + /* Process a unified INIT from the peer. + * Note: Only used during INIT-ACK processing. If + * there is an error just return to the outter + * layer which will bail. + */ + error = sctp_cmd_process_init(commands, asoc, chunk, + cmd->obj.ptr, gfp); + break; + + case SCTP_CMD_GEN_COOKIE_ECHO: + /* Generate a COOKIE ECHO chunk. */ + new_obj = sctp_make_cookie_echo(asoc, chunk); + if (!new_obj) { + if (cmd->obj.ptr) + sctp_chunk_free(cmd->obj.ptr); + goto nomem; + } + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(new_obj)); + + /* If there is an ERROR chunk to be sent along with + * the COOKIE_ECHO, send it, too. + */ + if (cmd->obj.ptr) + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(cmd->obj.ptr)); + + /* FIXME - Eventually come up with a cleaner way to + * enabling COOKIE-ECHO + DATA bundling during + * multihoming stale cookie scenarios, the following + * command plays with asoc->peer.retran_path to + * avoid the problem of sending the COOKIE-ECHO and + * DATA in different paths, which could result + * in the association being ABORTed if the DATA chunk + * is processed first by the server. Checking the + * init error counter simply causes this command + * to be executed only during failed attempts of + * association establishment. + */ + if ((asoc->peer.retran_path != + asoc->peer.primary_path) && + (asoc->counters[SCTP_COUNTER_INIT_ERROR] > 0)) { + sctp_add_cmd_sf(commands, + SCTP_CMD_FORCE_PRIM_RETRAN, + SCTP_NULL()); + } + + break; + + case SCTP_CMD_GEN_SHUTDOWN: + /* Generate SHUTDOWN when in SHUTDOWN_SENT state. + * Reset error counts. + */ + asoc->overall_error_count = 0; + + /* Generate a SHUTDOWN chunk. */ + new_obj = sctp_make_shutdown(asoc, chunk); + if (!new_obj) + goto nomem; + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(new_obj)); + break; + + case SCTP_CMD_CHUNK_ULP: + /* Send a chunk to the sockets layer. */ + SCTP_DEBUG_PRINTK("sm_sideff: %s %p, %s %p.\n", + "chunk_up:", cmd->obj.ptr, + "ulpq:", &asoc->ulpq); + sctp_ulpq_tail_data(&asoc->ulpq, cmd->obj.ptr, + GFP_ATOMIC); + break; + + case SCTP_CMD_EVENT_ULP: + /* Send a notification to the sockets layer. */ + SCTP_DEBUG_PRINTK("sm_sideff: %s %p, %s %p.\n", + "event_up:",cmd->obj.ptr, + "ulpq:",&asoc->ulpq); + sctp_ulpq_tail_event(&asoc->ulpq, cmd->obj.ptr); + break; + + case SCTP_CMD_REPLY: + /* If an caller has not already corked, do cork. */ + if (!asoc->outqueue.cork) { + sctp_outq_cork(&asoc->outqueue); + local_cork = 1; + } + /* Send a chunk to our peer. */ + error = sctp_outq_tail(&asoc->outqueue, cmd->obj.ptr); + break; + + case SCTP_CMD_SEND_PKT: + /* Send a full packet to our peer. */ + packet = cmd->obj.ptr; + sctp_packet_transmit(packet); + sctp_ootb_pkt_free(packet); + break; + + case SCTP_CMD_RETRAN: + /* Mark a transport for retransmission. */ + sctp_retransmit(&asoc->outqueue, cmd->obj.transport, + SCTP_RTXR_T3_RTX); + break; + + case SCTP_CMD_TRANSMIT: + /* Kick start transmission. */ + error = sctp_outq_uncork(&asoc->outqueue); + local_cork = 0; + break; + + case SCTP_CMD_ECN_CE: + /* Do delayed CE processing. */ + sctp_do_ecn_ce_work(asoc, cmd->obj.u32); + break; + + case SCTP_CMD_ECN_ECNE: + /* Do delayed ECNE processing. */ + new_obj = sctp_do_ecn_ecne_work(asoc, cmd->obj.u32, + chunk); + if (new_obj) + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(new_obj)); + break; + + case SCTP_CMD_ECN_CWR: + /* Do delayed CWR processing. */ + sctp_do_ecn_cwr_work(asoc, cmd->obj.u32); + break; + + case SCTP_CMD_SETUP_T2: + sctp_cmd_setup_t2(commands, asoc, cmd->obj.ptr); + break; + + case SCTP_CMD_TIMER_START: + timer = &asoc->timers[cmd->obj.to]; + timeout = asoc->timeouts[cmd->obj.to]; + if (!timeout) + BUG(); + + timer->expires = jiffies + timeout; + sctp_association_hold(asoc); + add_timer(timer); + break; + + case SCTP_CMD_TIMER_RESTART: + timer = &asoc->timers[cmd->obj.to]; + timeout = asoc->timeouts[cmd->obj.to]; + if (!mod_timer(timer, jiffies + timeout)) + sctp_association_hold(asoc); + break; + + case SCTP_CMD_TIMER_STOP: + timer = &asoc->timers[cmd->obj.to]; + if (timer_pending(timer) && del_timer(timer)) + sctp_association_put(asoc); + break; + + case SCTP_CMD_INIT_RESTART: + /* Do the needed accounting and updates + * associated with restarting an initialization + * timer. + */ + asoc->counters[SCTP_COUNTER_INIT_ERROR]++; + asoc->timeouts[cmd->obj.to] *= 2; + if (asoc->timeouts[cmd->obj.to] > + asoc->max_init_timeo) { + asoc->timeouts[cmd->obj.to] = + asoc->max_init_timeo; + } + + /* If we've sent any data bundled with + * COOKIE-ECHO we need to resend. + */ + list_for_each(pos, &asoc->peer.transport_addr_list) { + t = list_entry(pos, struct sctp_transport, + transports); + sctp_retransmit_mark(&asoc->outqueue, t, 0); + } + + sctp_add_cmd_sf(commands, + SCTP_CMD_TIMER_RESTART, + SCTP_TO(cmd->obj.to)); + break; + + case SCTP_CMD_INIT_FAILED: + sctp_cmd_init_failed(commands, asoc, cmd->obj.u32); + break; + + case SCTP_CMD_ASSOC_FAILED: + sctp_cmd_assoc_failed(commands, asoc, event_type, + subtype, chunk, cmd->obj.u32); + break; + + case SCTP_CMD_COUNTER_INC: + asoc->counters[cmd->obj.counter]++; + break; + + case SCTP_CMD_COUNTER_RESET: + asoc->counters[cmd->obj.counter] = 0; + break; + + case SCTP_CMD_REPORT_DUP: + sctp_tsnmap_mark_dup(&asoc->peer.tsn_map, + cmd->obj.u32); + break; + + case SCTP_CMD_REPORT_BAD_TAG: + SCTP_DEBUG_PRINTK("vtag mismatch!\n"); + break; + + case SCTP_CMD_STRIKE: + /* Mark one strike against a transport. */ + sctp_do_8_2_transport_strike(asoc, cmd->obj.transport); + break; + + case SCTP_CMD_TRANSPORT_RESET: + t = cmd->obj.transport; + sctp_cmd_transport_reset(commands, asoc, t); + break; + + case SCTP_CMD_TRANSPORT_ON: + t = cmd->obj.transport; + sctp_cmd_transport_on(commands, asoc, t, chunk); + break; + + case SCTP_CMD_HB_TIMERS_START: + sctp_cmd_hb_timers_start(commands, asoc); + break; + + case SCTP_CMD_HB_TIMER_UPDATE: + t = cmd->obj.transport; + sctp_cmd_hb_timer_update(commands, asoc, t); + break; + + case SCTP_CMD_HB_TIMERS_STOP: + sctp_cmd_hb_timers_stop(commands, asoc); + break; + + case SCTP_CMD_REPORT_ERROR: + error = cmd->obj.error; + break; + + case SCTP_CMD_PROCESS_CTSN: + /* Dummy up a SACK for processing. */ + sackh.cum_tsn_ack = cmd->obj.u32; + sackh.a_rwnd = 0; + sackh.num_gap_ack_blocks = 0; + sackh.num_dup_tsns = 0; + sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_SACK, + SCTP_SACKH(&sackh)); + break; + + case SCTP_CMD_DISCARD_PACKET: + /* We need to discard the whole packet. */ + chunk->pdiscard = 1; + break; + + case SCTP_CMD_RTO_PENDING: + t = cmd->obj.transport; + t->rto_pending = 1; + break; + + case SCTP_CMD_PART_DELIVER: + sctp_ulpq_partial_delivery(&asoc->ulpq, cmd->obj.ptr, + GFP_ATOMIC); + break; + + case SCTP_CMD_RENEGE: + sctp_ulpq_renege(&asoc->ulpq, cmd->obj.ptr, + GFP_ATOMIC); + break; + + case SCTP_CMD_SETUP_T4: + sctp_cmd_setup_t4(commands, asoc, cmd->obj.ptr); + break; + + case SCTP_CMD_PROCESS_OPERR: + sctp_cmd_process_operr(commands, asoc, chunk); + break; + case SCTP_CMD_CLEAR_INIT_TAG: + asoc->peer.i.init_tag = 0; + break; + case SCTP_CMD_DEL_NON_PRIMARY: + sctp_cmd_del_non_primary(asoc); + break; + case SCTP_CMD_T3_RTX_TIMERS_STOP: + sctp_cmd_t3_rtx_timers_stop(commands, asoc); + break; + case SCTP_CMD_FORCE_PRIM_RETRAN: + t = asoc->peer.retran_path; + asoc->peer.retran_path = asoc->peer.primary_path; + error = sctp_outq_uncork(&asoc->outqueue); + local_cork = 0; + asoc->peer.retran_path = t; + break; + default: + printk(KERN_WARNING "Impossible command: %u, %p\n", + cmd->verb, cmd->obj.ptr); + break; + }; + if (error) + break; + } + +out: + if (local_cork) + sctp_outq_uncork(&asoc->outqueue); + return error; +nomem: + error = -ENOMEM; + goto out; +} + diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c new file mode 100644 index 000000000000..278c56a2d076 --- /dev/null +++ b/net/sctp/sm_statefuns.c @@ -0,0 +1,5238 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001-2002 Intel Corp. + * Copyright (c) 2002 Nokia Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * This is part of the SCTP Linux Kernel Reference Implementation. + * + * These are the state functions for the state machine. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Mathew Kotowsky <kotowsky@sctp.org> + * Sridhar Samudrala <samudrala@us.ibm.com> + * Jon Grimm <jgrimm@us.ibm.com> + * Hui Huang <hui.huang@nokia.com> + * Dajiang Zhang <dajiang.zhang@nokia.com> + * Daisy Chang <daisyc@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * Ryan Layer <rmlayer@us.ibm.com> + * Kevin Gao <kevin.gao@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <linux/net.h> +#include <linux/inet.h> +#include <net/sock.h> +#include <net/inet_ecn.h> +#include <linux/skbuff.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> +#include <net/sctp/structs.h> + +static struct sctp_packet *sctp_abort_pkt_new(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + const void *payload, + size_t paylen); +static int sctp_eat_data(const struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands); +static struct sctp_packet *sctp_ootb_pkt_new(const struct sctp_association *asoc, + const struct sctp_chunk *chunk); +static void sctp_send_stale_cookie_err(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands, + struct sctp_chunk *err_chunk); +static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands); +static sctp_disposition_t sctp_sf_shut_8_4_5(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands); +static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk); + + +/* Small helper function that checks if the chunk length + * is of the appropriate length. The 'required_length' argument + * is set to be the size of a specific chunk we are testing. + * Return Values: 1 = Valid length + * 0 = Invalid length + * + */ +static inline int +sctp_chunk_length_valid(struct sctp_chunk *chunk, + __u16 required_length) +{ + __u16 chunk_length = ntohs(chunk->chunk_hdr->length); + + if (unlikely(chunk_length < required_length)) + return 0; + + return 1; +} + +/********************************************************** + * These are the state functions for handling chunk events. + **********************************************************/ + +/* + * Process the final SHUTDOWN COMPLETE. + * + * Section: 4 (C) (diagram), 9.2 + * Upon reception of the SHUTDOWN COMPLETE chunk the endpoint will verify + * that it is in SHUTDOWN-ACK-SENT state, if it is not the chunk should be + * discarded. If the endpoint is in the SHUTDOWN-ACK-SENT state the endpoint + * should stop the T2-shutdown timer and remove all knowledge of the + * association (and thus the association enters the CLOSED state). + * + * Verification Tag: 8.5.1(C) + * C) Rules for packet carrying SHUTDOWN COMPLETE: + * ... + * - The receiver of a SHUTDOWN COMPLETE shall accept the packet if the + * Verification Tag field of the packet matches its own tag OR it is + * set to its peer's tag and the T bit is set in the Chunk Flags. + * Otherwise, the receiver MUST silently discard the packet and take + * no further action. An endpoint MUST ignore the SHUTDOWN COMPLETE if + * it is not in the SHUTDOWN-ACK-SENT state. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_4_C(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_ulpevent *ev; + + /* RFC 2960 6.10 Bundling + * + * An endpoint MUST NOT bundle INIT, INIT ACK or + * SHUTDOWN COMPLETE with any other chunks. + */ + if (!chunk->singleton) + return SCTP_DISPOSITION_VIOLATION; + + if (!sctp_vtag_verify_either(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* RFC 2960 10.2 SCTP-to-ULP + * + * H) SHUTDOWN COMPLETE notification + * + * When SCTP completes the shutdown procedures (section 9.2) this + * notification is passed to the upper layer. + */ + ev = sctp_ulpevent_make_assoc_change(asoc, 0, SCTP_SHUTDOWN_COMP, + 0, 0, 0, GFP_ATOMIC); + if (!ev) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + + /* Upon reception of the SHUTDOWN COMPLETE chunk the endpoint + * will verify that it is in SHUTDOWN-ACK-SENT state, if it is + * not the chunk should be discarded. If the endpoint is in + * the SHUTDOWN-ACK-SENT state the endpoint should stop the + * T2-shutdown timer and remove all knowledge of the + * association (and thus the association enters the CLOSED + * state). + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + + SCTP_INC_STATS(SCTP_MIB_SHUTDOWNS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + + return SCTP_DISPOSITION_DELETE_TCB; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Respond to a normal INIT chunk. + * We are the side that is being asked for an association. + * + * Section: 5.1 Normal Establishment of an Association, B + * B) "Z" shall respond immediately with an INIT ACK chunk. The + * destination IP address of the INIT ACK MUST be set to the source + * IP address of the INIT to which this INIT ACK is responding. In + * the response, besides filling in other parameters, "Z" must set the + * Verification Tag field to Tag_A, and also provide its own + * Verification Tag (Tag_Z) in the Initiate Tag field. + * + * Verification Tag: Must be 0. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_1B_init(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_chunk *repl; + struct sctp_association *new_asoc; + struct sctp_chunk *err_chunk; + struct sctp_packet *packet; + sctp_unrecognized_param_t *unk_param; + struct sock *sk; + int len; + + /* 6.10 Bundling + * An endpoint MUST NOT bundle INIT, INIT ACK or + * SHUTDOWN COMPLETE with any other chunks. + * + * IG Section 2.11.2 + * Furthermore, we require that the receiver of an INIT chunk MUST + * enforce these rules by silently discarding an arriving packet + * with an INIT chunk that is bundled with other chunks. + */ + if (!chunk->singleton) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* If the packet is an OOTB packet which is temporarily on the + * control endpoint, respond with an ABORT. + */ + if (ep == sctp_sk((sctp_get_ctl_sock()))->ep) + return sctp_sf_tabort_8_4_8(ep, asoc, type, arg, commands); + + sk = ep->base.sk; + /* If the endpoint is not listening or if the number of associations + * on the TCP-style socket exceed the max backlog, respond with an + * ABORT. + */ + if (!sctp_sstate(sk, LISTENING) || + (sctp_style(sk, TCP) && + sk_acceptq_is_full(sk))) + return sctp_sf_tabort_8_4_8(ep, asoc, type, arg, commands); + + /* 3.1 A packet containing an INIT chunk MUST have a zero Verification + * Tag. + */ + if (chunk->sctp_hdr->vtag != 0) + return sctp_sf_tabort_8_4_8(ep, asoc, type, arg, commands); + + /* Make sure that the INIT chunk has a valid length. + * Normally, this would cause an ABORT with a Protocol Violation + * error, but since we don't have an association, we'll + * just discard the packet. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_init_chunk_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Verify the INIT chunk before processing it. */ + err_chunk = NULL; + if (!sctp_verify_init(asoc, chunk->chunk_hdr->type, + (sctp_init_chunk_t *)chunk->chunk_hdr, chunk, + &err_chunk)) { + /* This chunk contains fatal error. It is to be discarded. + * Send an ABORT, with causes if there is any. + */ + if (err_chunk) { + packet = sctp_abort_pkt_new(ep, asoc, arg, + (__u8 *)(err_chunk->chunk_hdr) + + sizeof(sctp_chunkhdr_t), + ntohs(err_chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t)); + + sctp_chunk_free(err_chunk); + + if (packet) { + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, + SCTP_PACKET(packet)); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + return SCTP_DISPOSITION_CONSUME; + } else { + return SCTP_DISPOSITION_NOMEM; + } + } else { + return sctp_sf_tabort_8_4_8(ep, asoc, type, arg, + commands); + } + } + + /* Grab the INIT header. */ + chunk->subh.init_hdr = (sctp_inithdr_t *)chunk->skb->data; + + /* Tag the variable length parameters. */ + chunk->param_hdr.v = skb_pull(chunk->skb, sizeof(sctp_inithdr_t)); + + new_asoc = sctp_make_temp_asoc(ep, chunk, GFP_ATOMIC); + if (!new_asoc) + goto nomem; + + /* The call, sctp_process_init(), can fail on memory allocation. */ + if (!sctp_process_init(new_asoc, chunk->chunk_hdr->type, + sctp_source(chunk), + (sctp_init_chunk_t *)chunk->chunk_hdr, + GFP_ATOMIC)) + goto nomem_init; + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(new_asoc)); + + /* B) "Z" shall respond immediately with an INIT ACK chunk. */ + + /* If there are errors need to be reported for unknown parameters, + * make sure to reserve enough room in the INIT ACK for them. + */ + len = 0; + if (err_chunk) + len = ntohs(err_chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t); + + if (sctp_assoc_set_bind_addr_from_ep(new_asoc, GFP_ATOMIC) < 0) + goto nomem_ack; + + repl = sctp_make_init_ack(new_asoc, chunk, GFP_ATOMIC, len); + if (!repl) + goto nomem_ack; + + /* If there are errors need to be reported for unknown parameters, + * include them in the outgoing INIT ACK as "Unrecognized parameter" + * parameter. + */ + if (err_chunk) { + /* Get the "Unrecognized parameter" parameter(s) out of the + * ERROR chunk generated by sctp_verify_init(). Since the + * error cause code for "unknown parameter" and the + * "Unrecognized parameter" type is the same, we can + * construct the parameters in INIT ACK by copying the + * ERROR causes over. + */ + unk_param = (sctp_unrecognized_param_t *) + ((__u8 *)(err_chunk->chunk_hdr) + + sizeof(sctp_chunkhdr_t)); + /* Replace the cause code with the "Unrecognized parameter" + * parameter type. + */ + sctp_addto_chunk(repl, len, unk_param); + sctp_chunk_free(err_chunk); + } + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + + /* + * Note: After sending out INIT ACK with the State Cookie parameter, + * "Z" MUST NOT allocate any resources, nor keep any states for the + * new association. Otherwise, "Z" will be vulnerable to resource + * attacks. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + + return SCTP_DISPOSITION_DELETE_TCB; + +nomem_ack: + if (err_chunk) + sctp_chunk_free(err_chunk); +nomem_init: + sctp_association_free(new_asoc); +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Respond to a normal INIT ACK chunk. + * We are the side that is initiating the association. + * + * Section: 5.1 Normal Establishment of an Association, C + * C) Upon reception of the INIT ACK from "Z", "A" shall stop the T1-init + * timer and leave COOKIE-WAIT state. "A" shall then send the State + * Cookie received in the INIT ACK chunk in a COOKIE ECHO chunk, start + * the T1-cookie timer, and enter the COOKIE-ECHOED state. + * + * Note: The COOKIE ECHO chunk can be bundled with any pending outbound + * DATA chunks, but it MUST be the first chunk in the packet and + * until the COOKIE ACK is returned the sender MUST NOT send any + * other packets to the peer. + * + * Verification Tag: 3.3.3 + * If the value of the Initiate Tag in a received INIT ACK chunk is + * found to be 0, the receiver MUST treat it as an error and close the + * association by transmitting an ABORT. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_1C_ack(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + sctp_init_chunk_t *initchunk; + __u32 init_tag; + struct sctp_chunk *err_chunk; + struct sctp_packet *packet; + sctp_disposition_t ret; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the INIT-ACK chunk has a valid length */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_initack_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + /* 6.10 Bundling + * An endpoint MUST NOT bundle INIT, INIT ACK or + * SHUTDOWN COMPLETE with any other chunks. + */ + if (!chunk->singleton) + return SCTP_DISPOSITION_VIOLATION; + + /* Grab the INIT header. */ + chunk->subh.init_hdr = (sctp_inithdr_t *) chunk->skb->data; + + init_tag = ntohl(chunk->subh.init_hdr->init_tag); + + /* Verification Tag: 3.3.3 + * If the value of the Initiate Tag in a received INIT ACK + * chunk is found to be 0, the receiver MUST treat it as an + * error and close the association by transmitting an ABORT. + */ + if (!init_tag) { + struct sctp_chunk *reply = sctp_make_abort(asoc, chunk, 0); + if (!reply) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + return SCTP_DISPOSITION_DELETE_TCB; + } + + /* Verify the INIT chunk before processing it. */ + err_chunk = NULL; + if (!sctp_verify_init(asoc, chunk->chunk_hdr->type, + (sctp_init_chunk_t *)chunk->chunk_hdr, chunk, + &err_chunk)) { + + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + + /* This chunk contains fatal error. It is to be discarded. + * Send an ABORT, with causes if there is any. + */ + if (err_chunk) { + packet = sctp_abort_pkt_new(ep, asoc, arg, + (__u8 *)(err_chunk->chunk_hdr) + + sizeof(sctp_chunkhdr_t), + ntohs(err_chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t)); + + sctp_chunk_free(err_chunk); + + if (packet) { + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, + SCTP_PACKET(packet)); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, + SCTP_NULL()); + return SCTP_DISPOSITION_CONSUME; + } else { + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, + SCTP_NULL()); + return SCTP_DISPOSITION_NOMEM; + } + } else { + ret = sctp_sf_tabort_8_4_8(ep, asoc, type, arg, + commands); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, + SCTP_NULL()); + return ret; + } + } + + /* Tag the variable length parameters. Note that we never + * convert the parameters in an INIT chunk. + */ + chunk->param_hdr.v = skb_pull(chunk->skb, sizeof(sctp_inithdr_t)); + + initchunk = (sctp_init_chunk_t *) chunk->chunk_hdr; + + sctp_add_cmd_sf(commands, SCTP_CMD_PEER_INIT, + SCTP_PEER_INIT(initchunk)); + + /* 5.1 C) "A" shall stop the T1-init timer and leave + * COOKIE-WAIT state. "A" shall then ... start the T1-cookie + * timer, and enter the COOKIE-ECHOED state. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_COOKIE_ECHOED)); + + /* 5.1 C) "A" shall then send the State Cookie received in the + * INIT ACK chunk in a COOKIE ECHO chunk, ... + */ + /* If there is any errors to report, send the ERROR chunk generated + * for unknown parameters as well. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_COOKIE_ECHO, + SCTP_CHUNK(err_chunk)); + + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Respond to a normal COOKIE ECHO chunk. + * We are the side that is being asked for an association. + * + * Section: 5.1 Normal Establishment of an Association, D + * D) Upon reception of the COOKIE ECHO chunk, Endpoint "Z" will reply + * with a COOKIE ACK chunk after building a TCB and moving to + * the ESTABLISHED state. A COOKIE ACK chunk may be bundled with + * any pending DATA chunks (and/or SACK chunks), but the COOKIE ACK + * chunk MUST be the first chunk in the packet. + * + * IMPLEMENTATION NOTE: An implementation may choose to send the + * Communication Up notification to the SCTP user upon reception + * of a valid COOKIE ECHO chunk. + * + * Verification Tag: 8.5.1 Exceptions in Verification Tag Rules + * D) Rules for packet carrying a COOKIE ECHO + * + * - When sending a COOKIE ECHO, the endpoint MUST use the value of the + * Initial Tag received in the INIT ACK. + * + * - The receiver of a COOKIE ECHO follows the procedures in Section 5. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_1D_ce(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_association *new_asoc; + sctp_init_chunk_t *peer_init; + struct sctp_chunk *repl; + struct sctp_ulpevent *ev; + int error = 0; + struct sctp_chunk *err_chk_p; + + /* If the packet is an OOTB packet which is temporarily on the + * control endpoint, respond with an ABORT. + */ + if (ep == sctp_sk((sctp_get_ctl_sock()))->ep) + return sctp_sf_ootb(ep, asoc, type, arg, commands); + + /* Make sure that the COOKIE_ECHO chunk has a valid length. + * In this case, we check that we have enough for at least a + * chunk header. More detailed verification is done + * in sctp_unpack_cookie(). + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_chunkhdr_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* "Decode" the chunk. We have no optional parameters so we + * are in good shape. + */ + chunk->subh.cookie_hdr = + (struct sctp_signed_cookie *)chunk->skb->data; + skb_pull(chunk->skb, + ntohs(chunk->chunk_hdr->length) - sizeof(sctp_chunkhdr_t)); + + /* 5.1 D) Upon reception of the COOKIE ECHO chunk, Endpoint + * "Z" will reply with a COOKIE ACK chunk after building a TCB + * and moving to the ESTABLISHED state. + */ + new_asoc = sctp_unpack_cookie(ep, asoc, chunk, GFP_ATOMIC, &error, + &err_chk_p); + + /* FIXME: + * If the re-build failed, what is the proper error path + * from here? + * + * [We should abort the association. --piggy] + */ + if (!new_asoc) { + /* FIXME: Several errors are possible. A bad cookie should + * be silently discarded, but think about logging it too. + */ + switch (error) { + case -SCTP_IERROR_NOMEM: + goto nomem; + + case -SCTP_IERROR_STALE_COOKIE: + sctp_send_stale_cookie_err(ep, asoc, chunk, commands, + err_chk_p); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + case -SCTP_IERROR_BAD_SIG: + default: + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + }; + } + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(new_asoc)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_ESTABLISHED)); + SCTP_INC_STATS(SCTP_MIB_CURRESTAB); + SCTP_INC_STATS(SCTP_MIB_PASSIVEESTABS); + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_START, SCTP_NULL()); + + if (new_asoc->autoclose) + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); + + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL()); + + /* Re-build the bind address for the association is done in + * the sctp_unpack_cookie() already. + */ + /* This is a brand-new association, so these are not yet side + * effects--it is safe to run them here. + */ + peer_init = &chunk->subh.cookie_hdr->c.peer_init[0]; + + if (!sctp_process_init(new_asoc, chunk->chunk_hdr->type, + &chunk->subh.cookie_hdr->c.peer_addr, + peer_init, GFP_ATOMIC)) + goto nomem_init; + + repl = sctp_make_cookie_ack(new_asoc, chunk); + if (!repl) + goto nomem_repl; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + + /* RFC 2960 5.1 Normal Establishment of an Association + * + * D) IMPLEMENTATION NOTE: An implementation may choose to + * send the Communication Up notification to the SCTP user + * upon reception of a valid COOKIE ECHO chunk. + */ + ev = sctp_ulpevent_make_assoc_change(new_asoc, 0, SCTP_COMM_UP, 0, + new_asoc->c.sinit_num_ostreams, + new_asoc->c.sinit_max_instreams, + GFP_ATOMIC); + if (!ev) + goto nomem_ev; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + + /* Sockets API Draft Section 5.3.1.6 + * When a peer sends a Adaption Layer Indication parameter , SCTP + * delivers this notification to inform the application that of the + * peers requested adaption layer. + */ + if (new_asoc->peer.adaption_ind) { + ev = sctp_ulpevent_make_adaption_indication(new_asoc, + GFP_ATOMIC); + if (!ev) + goto nomem_ev; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(ev)); + } + + return SCTP_DISPOSITION_CONSUME; + +nomem_ev: + sctp_chunk_free(repl); +nomem_repl: +nomem_init: + sctp_association_free(new_asoc); +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Respond to a normal COOKIE ACK chunk. + * We are the side that is being asked for an association. + * + * RFC 2960 5.1 Normal Establishment of an Association + * + * E) Upon reception of the COOKIE ACK, endpoint "A" will move from the + * COOKIE-ECHOED state to the ESTABLISHED state, stopping the T1-cookie + * timer. It may also notify its ULP about the successful + * establishment of the association with a Communication Up + * notification (see Section 10). + * + * Verification Tag: + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_1E_ca(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_ulpevent *ev; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Verify that the chunk length for the COOKIE-ACK is OK. + * If we don't do this, any bundled chunks may be junked. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_chunkhdr_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* Reset init error count upon receipt of COOKIE-ACK, + * to avoid problems with the managemement of this + * counter in stale cookie situations when a transition back + * from the COOKIE-ECHOED state to the COOKIE-WAIT + * state is performed. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_RESET, + SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR)); + + /* RFC 2960 5.1 Normal Establishment of an Association + * + * E) Upon reception of the COOKIE ACK, endpoint "A" will move + * from the COOKIE-ECHOED state to the ESTABLISHED state, + * stopping the T1-cookie timer. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_ESTABLISHED)); + SCTP_INC_STATS(SCTP_MIB_CURRESTAB); + SCTP_INC_STATS(SCTP_MIB_ACTIVEESTABS); + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_START, SCTP_NULL()); + if (asoc->autoclose) + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL()); + + /* It may also notify its ULP about the successful + * establishment of the association with a Communication Up + * notification (see Section 10). + */ + ev = sctp_ulpevent_make_assoc_change(asoc, 0, SCTP_COMM_UP, + 0, asoc->c.sinit_num_ostreams, + asoc->c.sinit_max_instreams, + GFP_ATOMIC); + + if (!ev) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + + /* Sockets API Draft Section 5.3.1.6 + * When a peer sends a Adaption Layer Indication parameter , SCTP + * delivers this notification to inform the application that of the + * peers requested adaption layer. + */ + if (asoc->peer.adaption_ind) { + ev = sctp_ulpevent_make_adaption_indication(asoc, GFP_ATOMIC); + if (!ev) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(ev)); + } + + return SCTP_DISPOSITION_CONSUME; +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* Generate and sendout a heartbeat packet. */ +static sctp_disposition_t sctp_sf_heartbeat(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_transport *transport = (struct sctp_transport *) arg; + struct sctp_chunk *reply; + sctp_sender_hb_info_t hbinfo; + size_t paylen = 0; + + hbinfo.param_hdr.type = SCTP_PARAM_HEARTBEAT_INFO; + hbinfo.param_hdr.length = htons(sizeof(sctp_sender_hb_info_t)); + hbinfo.daddr = transport->ipaddr; + hbinfo.sent_at = jiffies; + + /* Send a heartbeat to our peer. */ + paylen = sizeof(sctp_sender_hb_info_t); + reply = sctp_make_heartbeat(asoc, transport, &hbinfo, paylen); + if (!reply) + return SCTP_DISPOSITION_NOMEM; + + /* Set rto_pending indicating that an RTT measurement + * is started with this heartbeat chunk. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_RTO_PENDING, + SCTP_TRANSPORT(transport)); + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + return SCTP_DISPOSITION_CONSUME; +} + +/* Generate a HEARTBEAT packet on the given transport. */ +sctp_disposition_t sctp_sf_sendbeat_8_3(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_transport *transport = (struct sctp_transport *) arg; + + if (asoc->overall_error_count > asoc->max_retrans) { + /* CMD_ASSOC_FAILED calls CMD_DELETE_TCB. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_DELETE_TCB; + } + + /* Section 3.3.5. + * The Sender-specific Heartbeat Info field should normally include + * information about the sender's current time when this HEARTBEAT + * chunk is sent and the destination transport address to which this + * HEARTBEAT is sent (see Section 8.3). + */ + + if (transport->hb_allowed) { + if (SCTP_DISPOSITION_NOMEM == + sctp_sf_heartbeat(ep, asoc, type, arg, + commands)) + return SCTP_DISPOSITION_NOMEM; + /* Set transport error counter and association error counter + * when sending heartbeat. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSPORT_RESET, + SCTP_TRANSPORT(transport)); + } + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMER_UPDATE, + SCTP_TRANSPORT(transport)); + + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Process an heartbeat request. + * + * Section: 8.3 Path Heartbeat + * The receiver of the HEARTBEAT should immediately respond with a + * HEARTBEAT ACK that contains the Heartbeat Information field copied + * from the received HEARTBEAT chunk. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * When receiving an SCTP packet, the endpoint MUST ensure that the + * value in the Verification Tag field of the received SCTP packet + * matches its own Tag. If the received Verification Tag value does not + * match the receiver's own tag value, the receiver shall silently + * discard the packet and shall not process it any further except for + * those cases listed in Section 8.5.1 below. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_beat_8_3(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_chunk *reply; + size_t paylen = 0; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the HEARTBEAT chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_heartbeat_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* 8.3 The receiver of the HEARTBEAT should immediately + * respond with a HEARTBEAT ACK that contains the Heartbeat + * Information field copied from the received HEARTBEAT chunk. + */ + chunk->subh.hb_hdr = (sctp_heartbeathdr_t *) chunk->skb->data; + paylen = ntohs(chunk->chunk_hdr->length) - sizeof(sctp_chunkhdr_t); + skb_pull(chunk->skb, paylen); + + reply = sctp_make_heartbeat_ack(asoc, chunk, + chunk->subh.hb_hdr, paylen); + if (!reply) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Process the returning HEARTBEAT ACK. + * + * Section: 8.3 Path Heartbeat + * Upon the receipt of the HEARTBEAT ACK, the sender of the HEARTBEAT + * should clear the error counter of the destination transport + * address to which the HEARTBEAT was sent, and mark the destination + * transport address as active if it is not so marked. The endpoint may + * optionally report to the upper layer when an inactive destination + * address is marked as active due to the reception of the latest + * HEARTBEAT ACK. The receiver of the HEARTBEAT ACK must also + * clear the association overall error count as well (as defined + * in section 8.1). + * + * The receiver of the HEARTBEAT ACK should also perform an RTT + * measurement for that destination transport address using the time + * value carried in the HEARTBEAT ACK chunk. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_backbeat_8_3(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + union sctp_addr from_addr; + struct sctp_transport *link; + sctp_sender_hb_info_t *hbinfo; + unsigned long max_interval; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the HEARTBEAT-ACK chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_heartbeat_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + hbinfo = (sctp_sender_hb_info_t *) chunk->skb->data; + from_addr = hbinfo->daddr; + link = sctp_assoc_lookup_paddr(asoc, &from_addr); + + /* This should never happen, but lets log it if so. */ + if (!link) { + printk(KERN_WARNING + "%s: Could not find address %d.%d.%d.%d\n", + __FUNCTION__, NIPQUAD(from_addr.v4.sin_addr)); + return SCTP_DISPOSITION_DISCARD; + } + + max_interval = link->hb_interval + link->rto; + + /* Check if the timestamp looks valid. */ + if (time_after(hbinfo->sent_at, jiffies) || + time_after(jiffies, hbinfo->sent_at + max_interval)) { + SCTP_DEBUG_PRINTK("%s: HEARTBEAT ACK with invalid timestamp" + "received for transport: %p\n", + __FUNCTION__, link); + return SCTP_DISPOSITION_DISCARD; + } + + /* 8.3 Upon the receipt of the HEARTBEAT ACK, the sender of + * the HEARTBEAT should clear the error counter of the + * destination transport address to which the HEARTBEAT was + * sent and mark the destination transport address as active if + * it is not so marked. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSPORT_ON, SCTP_TRANSPORT(link)); + + return SCTP_DISPOSITION_CONSUME; +} + +/* Helper function to send out an abort for the restart + * condition. + */ +static int sctp_sf_send_restart_abort(union sctp_addr *ssa, + struct sctp_chunk *init, + sctp_cmd_seq_t *commands) +{ + int len; + struct sctp_packet *pkt; + union sctp_addr_param *addrparm; + struct sctp_errhdr *errhdr; + struct sctp_endpoint *ep; + char buffer[sizeof(struct sctp_errhdr)+sizeof(union sctp_addr_param)]; + struct sctp_af *af = sctp_get_af_specific(ssa->v4.sin_family); + + /* Build the error on the stack. We are way to malloc crazy + * throughout the code today. + */ + errhdr = (struct sctp_errhdr *)buffer; + addrparm = (union sctp_addr_param *)errhdr->variable; + + /* Copy into a parm format. */ + len = af->to_addr_param(ssa, addrparm); + len += sizeof(sctp_errhdr_t); + + errhdr->cause = SCTP_ERROR_RESTART; + errhdr->length = htons(len); + + /* Assign to the control socket. */ + ep = sctp_sk((sctp_get_ctl_sock()))->ep; + + /* Association is NULL since this may be a restart attack and we + * want to send back the attacker's vtag. + */ + pkt = sctp_abort_pkt_new(ep, NULL, init, errhdr, len); + + if (!pkt) + goto out; + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, SCTP_PACKET(pkt)); + + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + + /* Discard the rest of the inbound packet. */ + sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET, SCTP_NULL()); + +out: + /* Even if there is no memory, treat as a failure so + * the packet will get dropped. + */ + return 0; +} + +/* A restart is occurring, check to make sure no new addresses + * are being added as we may be under a takeover attack. + */ +static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, + const struct sctp_association *asoc, + struct sctp_chunk *init, + sctp_cmd_seq_t *commands) +{ + struct sctp_transport *new_addr, *addr; + struct list_head *pos, *pos2; + int found; + + /* Implementor's Guide - Sectin 5.2.2 + * ... + * Before responding the endpoint MUST check to see if the + * unexpected INIT adds new addresses to the association. If new + * addresses are added to the association, the endpoint MUST respond + * with an ABORT.. + */ + + /* Search through all current addresses and make sure + * we aren't adding any new ones. + */ + new_addr = NULL; + found = 0; + + list_for_each(pos, &new_asoc->peer.transport_addr_list) { + new_addr = list_entry(pos, struct sctp_transport, transports); + found = 0; + list_for_each(pos2, &asoc->peer.transport_addr_list) { + addr = list_entry(pos2, struct sctp_transport, + transports); + if (sctp_cmp_addr_exact(&new_addr->ipaddr, + &addr->ipaddr)) { + found = 1; + break; + } + } + if (!found) + break; + } + + /* If a new address was added, ABORT the sender. */ + if (!found && new_addr) { + sctp_sf_send_restart_abort(&new_addr->ipaddr, init, commands); + } + + /* Return success if all addresses were found. */ + return found; +} + +/* Populate the verification/tie tags based on overlapping INIT + * scenario. + * + * Note: Do not use in CLOSED or SHUTDOWN-ACK-SENT state. + */ +static void sctp_tietags_populate(struct sctp_association *new_asoc, + const struct sctp_association *asoc) +{ + switch (asoc->state) { + + /* 5.2.1 INIT received in COOKIE-WAIT or COOKIE-ECHOED State */ + + case SCTP_STATE_COOKIE_WAIT: + new_asoc->c.my_vtag = asoc->c.my_vtag; + new_asoc->c.my_ttag = asoc->c.my_vtag; + new_asoc->c.peer_ttag = 0; + break; + + case SCTP_STATE_COOKIE_ECHOED: + new_asoc->c.my_vtag = asoc->c.my_vtag; + new_asoc->c.my_ttag = asoc->c.my_vtag; + new_asoc->c.peer_ttag = asoc->c.peer_vtag; + break; + + /* 5.2.2 Unexpected INIT in States Other than CLOSED, COOKIE-ECHOED, + * COOKIE-WAIT and SHUTDOWN-ACK-SENT + */ + default: + new_asoc->c.my_ttag = asoc->c.my_vtag; + new_asoc->c.peer_ttag = asoc->c.peer_vtag; + break; + }; + + /* Other parameters for the endpoint SHOULD be copied from the + * existing parameters of the association (e.g. number of + * outbound streams) into the INIT ACK and cookie. + */ + new_asoc->rwnd = asoc->rwnd; + new_asoc->c.sinit_num_ostreams = asoc->c.sinit_num_ostreams; + new_asoc->c.sinit_max_instreams = asoc->c.sinit_max_instreams; + new_asoc->c.initial_tsn = asoc->c.initial_tsn; +} + +/* + * Compare vtag/tietag values to determine unexpected COOKIE-ECHO + * handling action. + * + * RFC 2960 5.2.4 Handle a COOKIE ECHO when a TCB exists. + * + * Returns value representing action to be taken. These action values + * correspond to Action/Description values in RFC 2960, Table 2. + */ +static char sctp_tietags_compare(struct sctp_association *new_asoc, + const struct sctp_association *asoc) +{ + /* In this case, the peer may have restarted. */ + if ((asoc->c.my_vtag != new_asoc->c.my_vtag) && + (asoc->c.peer_vtag != new_asoc->c.peer_vtag) && + (asoc->c.my_vtag == new_asoc->c.my_ttag) && + (asoc->c.peer_vtag == new_asoc->c.peer_ttag)) + return 'A'; + + /* Collision case B. */ + if ((asoc->c.my_vtag == new_asoc->c.my_vtag) && + ((asoc->c.peer_vtag != new_asoc->c.peer_vtag) || + (0 == asoc->c.peer_vtag))) { + return 'B'; + } + + /* Collision case D. */ + if ((asoc->c.my_vtag == new_asoc->c.my_vtag) && + (asoc->c.peer_vtag == new_asoc->c.peer_vtag)) + return 'D'; + + /* Collision case C. */ + if ((asoc->c.my_vtag != new_asoc->c.my_vtag) && + (asoc->c.peer_vtag == new_asoc->c.peer_vtag) && + (0 == new_asoc->c.my_ttag) && + (0 == new_asoc->c.peer_ttag)) + return 'C'; + + /* No match to any of the special cases; discard this packet. */ + return 'E'; +} + +/* Common helper routine for both duplicate and simulataneous INIT + * chunk handling. + */ +static sctp_disposition_t sctp_sf_do_unexpected_init( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, sctp_cmd_seq_t *commands) +{ + sctp_disposition_t retval; + struct sctp_chunk *chunk = arg; + struct sctp_chunk *repl; + struct sctp_association *new_asoc; + struct sctp_chunk *err_chunk; + struct sctp_packet *packet; + sctp_unrecognized_param_t *unk_param; + int len; + + /* 6.10 Bundling + * An endpoint MUST NOT bundle INIT, INIT ACK or + * SHUTDOWN COMPLETE with any other chunks. + * + * IG Section 2.11.2 + * Furthermore, we require that the receiver of an INIT chunk MUST + * enforce these rules by silently discarding an arriving packet + * with an INIT chunk that is bundled with other chunks. + */ + if (!chunk->singleton) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* 3.1 A packet containing an INIT chunk MUST have a zero Verification + * Tag. + */ + if (chunk->sctp_hdr->vtag != 0) + return sctp_sf_tabort_8_4_8(ep, asoc, type, arg, commands); + + /* Make sure that the INIT chunk has a valid length. + * In this case, we generate a protocol violation since we have + * an association established. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_init_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + /* Grab the INIT header. */ + chunk->subh.init_hdr = (sctp_inithdr_t *) chunk->skb->data; + + /* Tag the variable length parameters. */ + chunk->param_hdr.v = skb_pull(chunk->skb, sizeof(sctp_inithdr_t)); + + /* Verify the INIT chunk before processing it. */ + err_chunk = NULL; + if (!sctp_verify_init(asoc, chunk->chunk_hdr->type, + (sctp_init_chunk_t *)chunk->chunk_hdr, chunk, + &err_chunk)) { + /* This chunk contains fatal error. It is to be discarded. + * Send an ABORT, with causes if there is any. + */ + if (err_chunk) { + packet = sctp_abort_pkt_new(ep, asoc, arg, + (__u8 *)(err_chunk->chunk_hdr) + + sizeof(sctp_chunkhdr_t), + ntohs(err_chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t)); + + if (packet) { + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, + SCTP_PACKET(packet)); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + retval = SCTP_DISPOSITION_CONSUME; + } else { + retval = SCTP_DISPOSITION_NOMEM; + } + goto cleanup; + } else { + return sctp_sf_tabort_8_4_8(ep, asoc, type, arg, + commands); + } + } + + /* + * Other parameters for the endpoint SHOULD be copied from the + * existing parameters of the association (e.g. number of + * outbound streams) into the INIT ACK and cookie. + * FIXME: We are copying parameters from the endpoint not the + * association. + */ + new_asoc = sctp_make_temp_asoc(ep, chunk, GFP_ATOMIC); + if (!new_asoc) + goto nomem; + + /* In the outbound INIT ACK the endpoint MUST copy its current + * Verification Tag and Peers Verification tag into a reserved + * place (local tie-tag and per tie-tag) within the state cookie. + */ + if (!sctp_process_init(new_asoc, chunk->chunk_hdr->type, + sctp_source(chunk), + (sctp_init_chunk_t *)chunk->chunk_hdr, + GFP_ATOMIC)) { + retval = SCTP_DISPOSITION_NOMEM; + goto nomem_init; + } + + /* Make sure no new addresses are being added during the + * restart. Do not do this check for COOKIE-WAIT state, + * since there are no peer addresses to check against. + * Upon return an ABORT will have been sent if needed. + */ + if (!sctp_state(asoc, COOKIE_WAIT)) { + if (!sctp_sf_check_restart_addrs(new_asoc, asoc, chunk, + commands)) { + retval = SCTP_DISPOSITION_CONSUME; + goto cleanup_asoc; + } + } + + sctp_tietags_populate(new_asoc, asoc); + + /* B) "Z" shall respond immediately with an INIT ACK chunk. */ + + /* If there are errors need to be reported for unknown parameters, + * make sure to reserve enough room in the INIT ACK for them. + */ + len = 0; + if (err_chunk) { + len = ntohs(err_chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t); + } + + if (sctp_assoc_set_bind_addr_from_ep(new_asoc, GFP_ATOMIC) < 0) + goto nomem; + + repl = sctp_make_init_ack(new_asoc, chunk, GFP_ATOMIC, len); + if (!repl) + goto nomem; + + /* If there are errors need to be reported for unknown parameters, + * include them in the outgoing INIT ACK as "Unrecognized parameter" + * parameter. + */ + if (err_chunk) { + /* Get the "Unrecognized parameter" parameter(s) out of the + * ERROR chunk generated by sctp_verify_init(). Since the + * error cause code for "unknown parameter" and the + * "Unrecognized parameter" type is the same, we can + * construct the parameters in INIT ACK by copying the + * ERROR causes over. + */ + unk_param = (sctp_unrecognized_param_t *) + ((__u8 *)(err_chunk->chunk_hdr) + + sizeof(sctp_chunkhdr_t)); + /* Replace the cause code with the "Unrecognized parameter" + * parameter type. + */ + sctp_addto_chunk(repl, len, unk_param); + } + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(new_asoc)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + + /* + * Note: After sending out INIT ACK with the State Cookie parameter, + * "Z" MUST NOT allocate any resources for this new association. + * Otherwise, "Z" will be vulnerable to resource attacks. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + retval = SCTP_DISPOSITION_CONSUME; + +cleanup: + if (err_chunk) + sctp_chunk_free(err_chunk); + return retval; +nomem: + retval = SCTP_DISPOSITION_NOMEM; + goto cleanup; +nomem_init: +cleanup_asoc: + sctp_association_free(new_asoc); + goto cleanup; +} + +/* + * Handle simultanous INIT. + * This means we started an INIT and then we got an INIT request from + * our peer. + * + * Section: 5.2.1 INIT received in COOKIE-WAIT or COOKIE-ECHOED State (Item B) + * This usually indicates an initialization collision, i.e., each + * endpoint is attempting, at about the same time, to establish an + * association with the other endpoint. + * + * Upon receipt of an INIT in the COOKIE-WAIT or COOKIE-ECHOED state, an + * endpoint MUST respond with an INIT ACK using the same parameters it + * sent in its original INIT chunk (including its Verification Tag, + * unchanged). These original parameters are combined with those from the + * newly received INIT chunk. The endpoint shall also generate a State + * Cookie with the INIT ACK. The endpoint uses the parameters sent in its + * INIT to calculate the State Cookie. + * + * After that, the endpoint MUST NOT change its state, the T1-init + * timer shall be left running and the corresponding TCB MUST NOT be + * destroyed. The normal procedures for handling State Cookies when + * a TCB exists will resolve the duplicate INITs to a single association. + * + * For an endpoint that is in the COOKIE-ECHOED state it MUST populate + * its Tie-Tags with the Tag information of itself and its peer (see + * section 5.2.2 for a description of the Tie-Tags). + * + * Verification Tag: Not explicit, but an INIT can not have a valid + * verification tag, so we skip the check. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_2_1_siminit(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* Call helper to do the real work for both simulataneous and + * duplicate INIT chunk handling. + */ + return sctp_sf_do_unexpected_init(ep, asoc, type, arg, commands); +} + +/* + * Handle duplicated INIT messages. These are usually delayed + * restransmissions. + * + * Section: 5.2.2 Unexpected INIT in States Other than CLOSED, + * COOKIE-ECHOED and COOKIE-WAIT + * + * Unless otherwise stated, upon reception of an unexpected INIT for + * this association, the endpoint shall generate an INIT ACK with a + * State Cookie. In the outbound INIT ACK the endpoint MUST copy its + * current Verification Tag and peer's Verification Tag into a reserved + * place within the state cookie. We shall refer to these locations as + * the Peer's-Tie-Tag and the Local-Tie-Tag. The outbound SCTP packet + * containing this INIT ACK MUST carry a Verification Tag value equal to + * the Initiation Tag found in the unexpected INIT. And the INIT ACK + * MUST contain a new Initiation Tag (randomly generated see Section + * 5.3.1). Other parameters for the endpoint SHOULD be copied from the + * existing parameters of the association (e.g. number of outbound + * streams) into the INIT ACK and cookie. + * + * After sending out the INIT ACK, the endpoint shall take no further + * actions, i.e., the existing association, including its current state, + * and the corresponding TCB MUST NOT be changed. + * + * Note: Only when a TCB exists and the association is not in a COOKIE- + * WAIT state are the Tie-Tags populated. For a normal association INIT + * (i.e. the endpoint is in a COOKIE-WAIT state), the Tie-Tags MUST be + * set to 0 (indicating that no previous TCB existed). The INIT ACK and + * State Cookie are populated as specified in section 5.2.1. + * + * Verification Tag: Not specified, but an INIT has no way of knowing + * what the verification tag could be, so we ignore it. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_2_2_dupinit(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* Call helper to do the real work for both simulataneous and + * duplicate INIT chunk handling. + */ + return sctp_sf_do_unexpected_init(ep, asoc, type, arg, commands); +} + + + +/* Unexpected COOKIE-ECHO handler for peer restart (Table 2, action 'A') + * + * Section 5.2.4 + * A) In this case, the peer may have restarted. + */ +static sctp_disposition_t sctp_sf_do_dupcook_a(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands, + struct sctp_association *new_asoc) +{ + sctp_init_chunk_t *peer_init; + struct sctp_ulpevent *ev; + struct sctp_chunk *repl; + struct sctp_chunk *err; + sctp_disposition_t disposition; + + /* new_asoc is a brand-new association, so these are not yet + * side effects--it is safe to run them here. + */ + peer_init = &chunk->subh.cookie_hdr->c.peer_init[0]; + + if (!sctp_process_init(new_asoc, chunk->chunk_hdr->type, + sctp_source(chunk), peer_init, + GFP_ATOMIC)) + goto nomem; + + /* Make sure no new addresses are being added during the + * restart. Though this is a pretty complicated attack + * since you'd have to get inside the cookie. + */ + if (!sctp_sf_check_restart_addrs(new_asoc, asoc, chunk, commands)) { + return SCTP_DISPOSITION_CONSUME; + } + + /* If the endpoint is in the SHUTDOWN-ACK-SENT state and recognizes + * the peer has restarted (Action A), it MUST NOT setup a new + * association but instead resend the SHUTDOWN ACK and send an ERROR + * chunk with a "Cookie Received while Shutting Down" error cause to + * its peer. + */ + if (sctp_state(asoc, SHUTDOWN_ACK_SENT)) { + disposition = sctp_sf_do_9_2_reshutack(ep, asoc, + SCTP_ST_CHUNK(chunk->chunk_hdr->type), + chunk, commands); + if (SCTP_DISPOSITION_NOMEM == disposition) + goto nomem; + + err = sctp_make_op_error(asoc, chunk, + SCTP_ERROR_COOKIE_IN_SHUTDOWN, + NULL, 0); + if (err) + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(err)); + + return SCTP_DISPOSITION_CONSUME; + } + + /* For now, fail any unsent/unacked data. Consider the optional + * choice of resending of this data. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_PURGE_OUTQUEUE, SCTP_NULL()); + + /* Update the content of current association. */ + sctp_add_cmd_sf(commands, SCTP_CMD_UPDATE_ASSOC, SCTP_ASOC(new_asoc)); + + repl = sctp_make_cookie_ack(new_asoc, chunk); + if (!repl) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + + /* Report association restart to upper layer. */ + ev = sctp_ulpevent_make_assoc_change(asoc, 0, SCTP_RESTART, 0, + new_asoc->c.sinit_num_ostreams, + new_asoc->c.sinit_max_instreams, + GFP_ATOMIC); + if (!ev) + goto nomem_ev; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + return SCTP_DISPOSITION_CONSUME; + +nomem_ev: + sctp_chunk_free(repl); +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* Unexpected COOKIE-ECHO handler for setup collision (Table 2, action 'B') + * + * Section 5.2.4 + * B) In this case, both sides may be attempting to start an association + * at about the same time but the peer endpoint started its INIT + * after responding to the local endpoint's INIT + */ +/* This case represents an initialization collision. */ +static sctp_disposition_t sctp_sf_do_dupcook_b(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands, + struct sctp_association *new_asoc) +{ + sctp_init_chunk_t *peer_init; + struct sctp_ulpevent *ev; + struct sctp_chunk *repl; + + /* new_asoc is a brand-new association, so these are not yet + * side effects--it is safe to run them here. + */ + peer_init = &chunk->subh.cookie_hdr->c.peer_init[0]; + if (!sctp_process_init(new_asoc, chunk->chunk_hdr->type, + sctp_source(chunk), peer_init, + GFP_ATOMIC)) + goto nomem; + + /* Update the content of current association. */ + sctp_add_cmd_sf(commands, SCTP_CMD_UPDATE_ASSOC, SCTP_ASOC(new_asoc)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_ESTABLISHED)); + SCTP_INC_STATS(SCTP_MIB_CURRESTAB); + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_START, SCTP_NULL()); + + repl = sctp_make_cookie_ack(new_asoc, chunk); + if (!repl) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL()); + + /* RFC 2960 5.1 Normal Establishment of an Association + * + * D) IMPLEMENTATION NOTE: An implementation may choose to + * send the Communication Up notification to the SCTP user + * upon reception of a valid COOKIE ECHO chunk. + */ + ev = sctp_ulpevent_make_assoc_change(asoc, 0, SCTP_COMM_UP, 0, + new_asoc->c.sinit_num_ostreams, + new_asoc->c.sinit_max_instreams, + GFP_ATOMIC); + if (!ev) + goto nomem_ev; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + + /* Sockets API Draft Section 5.3.1.6 + * When a peer sends a Adaption Layer Indication parameter , SCTP + * delivers this notification to inform the application that of the + * peers requested adaption layer. + */ + if (asoc->peer.adaption_ind) { + ev = sctp_ulpevent_make_adaption_indication(asoc, GFP_ATOMIC); + if (!ev) + goto nomem_ev; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(ev)); + } + + return SCTP_DISPOSITION_CONSUME; + +nomem_ev: + sctp_chunk_free(repl); +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* Unexpected COOKIE-ECHO handler for setup collision (Table 2, action 'C') + * + * Section 5.2.4 + * C) In this case, the local endpoint's cookie has arrived late. + * Before it arrived, the local endpoint sent an INIT and received an + * INIT-ACK and finally sent a COOKIE ECHO with the peer's same tag + * but a new tag of its own. + */ +/* This case represents an initialization collision. */ +static sctp_disposition_t sctp_sf_do_dupcook_c(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands, + struct sctp_association *new_asoc) +{ + /* The cookie should be silently discarded. + * The endpoint SHOULD NOT change states and should leave + * any timers running. + */ + return SCTP_DISPOSITION_DISCARD; +} + +/* Unexpected COOKIE-ECHO handler lost chunk (Table 2, action 'D') + * + * Section 5.2.4 + * + * D) When both local and remote tags match the endpoint should always + * enter the ESTABLISHED state, if it has not already done so. + */ +/* This case represents an initialization collision. */ +static sctp_disposition_t sctp_sf_do_dupcook_d(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands, + struct sctp_association *new_asoc) +{ + struct sctp_ulpevent *ev = NULL; + struct sctp_chunk *repl; + + /* Clarification from Implementor's Guide: + * D) When both local and remote tags match the endpoint should + * enter the ESTABLISHED state, if it is in the COOKIE-ECHOED state. + * It should stop any cookie timer that may be running and send + * a COOKIE ACK. + */ + + /* Don't accidentally move back into established state. */ + if (asoc->state < SCTP_STATE_ESTABLISHED) { + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_ESTABLISHED)); + SCTP_INC_STATS(SCTP_MIB_CURRESTAB); + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_START, + SCTP_NULL()); + + /* RFC 2960 5.1 Normal Establishment of an Association + * + * D) IMPLEMENTATION NOTE: An implementation may choose + * to send the Communication Up notification to the + * SCTP user upon reception of a valid COOKIE + * ECHO chunk. + */ + ev = sctp_ulpevent_make_assoc_change(new_asoc, 0, + SCTP_COMM_UP, 0, + new_asoc->c.sinit_num_ostreams, + new_asoc->c.sinit_max_instreams, + GFP_ATOMIC); + if (!ev) + goto nomem; + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(ev)); + + /* Sockets API Draft Section 5.3.1.6 + * When a peer sends a Adaption Layer Indication parameter, + * SCTP delivers this notification to inform the application + * that of the peers requested adaption layer. + */ + if (new_asoc->peer.adaption_ind) { + ev = sctp_ulpevent_make_adaption_indication(new_asoc, + GFP_ATOMIC); + if (!ev) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(ev)); + } + } + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL()); + + repl = sctp_make_cookie_ack(new_asoc, chunk); + if (!repl) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + sctp_add_cmd_sf(commands, SCTP_CMD_TRANSMIT, SCTP_NULL()); + + return SCTP_DISPOSITION_CONSUME; + +nomem: + if (ev) + sctp_ulpevent_free(ev); + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Handle a duplicate COOKIE-ECHO. This usually means a cookie-carrying + * chunk was retransmitted and then delayed in the network. + * + * Section: 5.2.4 Handle a COOKIE ECHO when a TCB exists + * + * Verification Tag: None. Do cookie validation. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_5_2_4_dupcook(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_disposition_t retval; + struct sctp_chunk *chunk = arg; + struct sctp_association *new_asoc; + int error = 0; + char action; + struct sctp_chunk *err_chk_p; + + /* Make sure that the chunk has a valid length from the protocol + * perspective. In this case check to make sure we have at least + * enough for the chunk header. Cookie length verification is + * done later. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_chunkhdr_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* "Decode" the chunk. We have no optional parameters so we + * are in good shape. + */ + chunk->subh.cookie_hdr = (struct sctp_signed_cookie *)chunk->skb->data; + skb_pull(chunk->skb, ntohs(chunk->chunk_hdr->length) - + sizeof(sctp_chunkhdr_t)); + + /* In RFC 2960 5.2.4 3, if both Verification Tags in the State Cookie + * of a duplicate COOKIE ECHO match the Verification Tags of the + * current association, consider the State Cookie valid even if + * the lifespan is exceeded. + */ + new_asoc = sctp_unpack_cookie(ep, asoc, chunk, GFP_ATOMIC, &error, + &err_chk_p); + + /* FIXME: + * If the re-build failed, what is the proper error path + * from here? + * + * [We should abort the association. --piggy] + */ + if (!new_asoc) { + /* FIXME: Several errors are possible. A bad cookie should + * be silently discarded, but think about logging it too. + */ + switch (error) { + case -SCTP_IERROR_NOMEM: + goto nomem; + + case -SCTP_IERROR_STALE_COOKIE: + sctp_send_stale_cookie_err(ep, asoc, chunk, commands, + err_chk_p); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + case -SCTP_IERROR_BAD_SIG: + default: + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + }; + } + + /* Compare the tie_tag in cookie with the verification tag of + * current association. + */ + action = sctp_tietags_compare(new_asoc, asoc); + + switch (action) { + case 'A': /* Association restart. */ + retval = sctp_sf_do_dupcook_a(ep, asoc, chunk, commands, + new_asoc); + break; + + case 'B': /* Collision case B. */ + retval = sctp_sf_do_dupcook_b(ep, asoc, chunk, commands, + new_asoc); + break; + + case 'C': /* Collision case C. */ + retval = sctp_sf_do_dupcook_c(ep, asoc, chunk, commands, + new_asoc); + break; + + case 'D': /* Collision case D. */ + retval = sctp_sf_do_dupcook_d(ep, asoc, chunk, commands, + new_asoc); + break; + + default: /* Discard packet for all others. */ + retval = sctp_sf_pdiscard(ep, asoc, type, arg, commands); + break; + }; + + /* Delete the tempory new association. */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, SCTP_ASOC(new_asoc)); + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + + return retval; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Process an ABORT. (SHUTDOWN-PENDING state) + * + * See sctp_sf_do_9_1_abort(). + */ +sctp_disposition_t sctp_sf_shutdown_pending_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + + if (!sctp_vtag_verify_either(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the ABORT chunk has a valid length. + * Since this is an ABORT chunk, we have to discard it + * because of the following text: + * RFC 2960, Section 3.3.7 + * If an endpoint receives an ABORT with a format error or for an + * association that doesn't exist, it MUST silently discard it. + * Becasue the length is "invalid", we can't really discard just + * as we do not know its true length. So, to be safe, discard the + * packet. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Stop the T5-shutdown guard timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + return sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands); +} + +/* + * Process an ABORT. (SHUTDOWN-SENT state) + * + * See sctp_sf_do_9_1_abort(). + */ +sctp_disposition_t sctp_sf_shutdown_sent_abort(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + + if (!sctp_vtag_verify_either(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the ABORT chunk has a valid length. + * Since this is an ABORT chunk, we have to discard it + * because of the following text: + * RFC 2960, Section 3.3.7 + * If an endpoint receives an ABORT with a format error or for an + * association that doesn't exist, it MUST silently discard it. + * Becasue the length is "invalid", we can't really discard just + * as we do not know its true length. So, to be safe, discard the + * packet. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Stop the T2-shutdown timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + /* Stop the T5-shutdown guard timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + return sctp_sf_do_9_1_abort(ep, asoc, type, arg, commands); +} + +/* + * Process an ABORT. (SHUTDOWN-ACK-SENT state) + * + * See sctp_sf_do_9_1_abort(). + */ +sctp_disposition_t sctp_sf_shutdown_ack_sent_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* The same T2 timer, so we should be able to use + * common function with the SHUTDOWN-SENT state. + */ + return sctp_sf_shutdown_sent_abort(ep, asoc, type, arg, commands); +} + +/* + * Handle an Error received in COOKIE_ECHOED state. + * + * Only handle the error type of stale COOKIE Error, the other errors will + * be ignored. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_cookie_echoed_err(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + sctp_errhdr_t *err; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the ERROR chunk has a valid length. + * The parameter walking depends on this as well. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_operr_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* Process the error here */ + /* FUTURE FIXME: When PR-SCTP related and other optional + * parms are emitted, this will have to change to handle multiple + * errors. + */ + sctp_walk_errors(err, chunk->chunk_hdr) { + if (SCTP_ERROR_STALE_COOKIE == err->cause) + return sctp_sf_do_5_2_6_stale(ep, asoc, type, + arg, commands); + } + + /* It is possible to have malformed error causes, and that + * will cause us to end the walk early. However, since + * we are discarding the packet, there should be no adverse + * affects. + */ + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); +} + +/* + * Handle a Stale COOKIE Error + * + * Section: 5.2.6 Handle Stale COOKIE Error + * If the association is in the COOKIE-ECHOED state, the endpoint may elect + * one of the following three alternatives. + * ... + * 3) Send a new INIT chunk to the endpoint, adding a Cookie + * Preservative parameter requesting an extension to the lifetime of + * the State Cookie. When calculating the time extension, an + * implementation SHOULD use the RTT information measured based on the + * previous COOKIE ECHO / ERROR exchange, and should add no more + * than 1 second beyond the measured RTT, due to long State Cookie + * lifetimes making the endpoint more subject to a replay attack. + * + * Verification Tag: Not explicit, but safe to ignore. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +static sctp_disposition_t sctp_sf_do_5_2_6_stale(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + time_t stale; + sctp_cookie_preserve_param_t bht; + sctp_errhdr_t *err; + struct sctp_chunk *reply; + struct sctp_bind_addr *bp; + int attempts; + + attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1; + + if (attempts >= asoc->max_init_attempts) { + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, + SCTP_U32(SCTP_ERROR_STALE_COOKIE)); + return SCTP_DISPOSITION_DELETE_TCB; + } + + err = (sctp_errhdr_t *)(chunk->skb->data); + + /* When calculating the time extension, an implementation + * SHOULD use the RTT information measured based on the + * previous COOKIE ECHO / ERROR exchange, and should add no + * more than 1 second beyond the measured RTT, due to long + * State Cookie lifetimes making the endpoint more subject to + * a replay attack. + * Measure of Staleness's unit is usec. (1/1000000 sec) + * Suggested Cookie Life-span Increment's unit is msec. + * (1/1000 sec) + * In general, if you use the suggested cookie life, the value + * found in the field of measure of staleness should be doubled + * to give ample time to retransmit the new cookie and thus + * yield a higher probability of success on the reattempt. + */ + stale = ntohl(*(suseconds_t *)((u8 *)err + sizeof(sctp_errhdr_t))); + stale = (stale * 2) / 1000; + + bht.param_hdr.type = SCTP_PARAM_COOKIE_PRESERVATIVE; + bht.param_hdr.length = htons(sizeof(bht)); + bht.lifespan_increment = htonl(stale); + + /* Build that new INIT chunk. */ + bp = (struct sctp_bind_addr *) &asoc->base.bind_addr; + reply = sctp_make_init(asoc, bp, GFP_ATOMIC, sizeof(bht)); + if (!reply) + goto nomem; + + sctp_addto_chunk(reply, sizeof(bht), &bht); + + /* Clear peer's init_tag cached in assoc as we are sending a new INIT */ + sctp_add_cmd_sf(commands, SCTP_CMD_CLEAR_INIT_TAG, SCTP_NULL()); + + /* Stop pending T3-rtx and heartbeat timers */ + sctp_add_cmd_sf(commands, SCTP_CMD_T3_RTX_TIMERS_STOP, SCTP_NULL()); + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_STOP, SCTP_NULL()); + + /* Delete non-primary peer ip addresses since we are transitioning + * back to the COOKIE-WAIT state + */ + sctp_add_cmd_sf(commands, SCTP_CMD_DEL_NON_PRIMARY, SCTP_NULL()); + + /* If we've sent any data bundled with COOKIE-ECHO we will need to + * resend + */ + sctp_add_cmd_sf(commands, SCTP_CMD_RETRAN, + SCTP_TRANSPORT(asoc->peer.primary_path)); + + /* Cast away the const modifier, as we want to just + * rerun it through as a sideffect. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_COUNTER_INC, + SCTP_COUNTER(SCTP_COUNTER_INIT_ERROR)); + + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_COOKIE)); + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_COOKIE_WAIT)); + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Process an ABORT. + * + * Section: 9.1 + * After checking the Verification Tag, the receiving endpoint shall + * remove the association from its record, and shall report the + * termination to its upper layer. + * + * Verification Tag: 8.5.1 Exceptions in Verification Tag Rules + * B) Rules for packet carrying ABORT: + * + * - The endpoint shall always fill in the Verification Tag field of the + * outbound packet with the destination endpoint's tag value if it + * is known. + * + * - If the ABORT is sent in response to an OOTB packet, the endpoint + * MUST follow the procedure described in Section 8.4. + * + * - The receiver MUST accept the packet if the Verification Tag + * matches either its own tag, OR the tag of its peer. Otherwise, the + * receiver MUST silently discard the packet and take no further + * action. + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_9_1_abort(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + unsigned len; + __u16 error = SCTP_ERROR_NO_ERROR; + + if (!sctp_vtag_verify_either(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the ABORT chunk has a valid length. + * Since this is an ABORT chunk, we have to discard it + * because of the following text: + * RFC 2960, Section 3.3.7 + * If an endpoint receives an ABORT with a format error or for an + * association that doesn't exist, it MUST silently discard it. + * Becasue the length is "invalid", we can't really discard just + * as we do not know its true length. So, to be safe, discard the + * packet. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* See if we have an error cause code in the chunk. */ + len = ntohs(chunk->chunk_hdr->length); + if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr)) + error = ((sctp_errhdr_t *)chunk->skb->data)->cause; + + /* ASSOC_FAILED will DELETE_TCB. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, SCTP_U32(error)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + + return SCTP_DISPOSITION_ABORT; +} + +/* + * Process an ABORT. (COOKIE-WAIT state) + * + * See sctp_sf_do_9_1_abort() above. + */ +sctp_disposition_t sctp_sf_cookie_wait_abort(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + unsigned len; + __u16 error = SCTP_ERROR_NO_ERROR; + + if (!sctp_vtag_verify_either(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the ABORT chunk has a valid length. + * Since this is an ABORT chunk, we have to discard it + * because of the following text: + * RFC 2960, Section 3.3.7 + * If an endpoint receives an ABORT with a format error or for an + * association that doesn't exist, it MUST silently discard it. + * Becasue the length is "invalid", we can't really discard just + * as we do not know its true length. So, to be safe, discard the + * packet. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_abort_chunk_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* See if we have an error cause code in the chunk. */ + len = ntohs(chunk->chunk_hdr->length); + if (len >= sizeof(struct sctp_chunkhdr) + sizeof(struct sctp_errhdr)) + error = ((sctp_errhdr_t *)chunk->skb->data)->cause; + + sctp_stop_t1_and_abort(commands, error); + return SCTP_DISPOSITION_ABORT; +} + +/* + * Process an incoming ICMP as an ABORT. (COOKIE-WAIT state) + */ +sctp_disposition_t sctp_sf_cookie_wait_icmp_abort(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_stop_t1_and_abort(commands, SCTP_ERROR_NO_ERROR); + return SCTP_DISPOSITION_ABORT; +} + +/* + * Process an ABORT. (COOKIE-ECHOED state) + */ +sctp_disposition_t sctp_sf_cookie_echoed_abort(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* There is a single T1 timer, so we should be able to use + * common function with the COOKIE-WAIT state. + */ + return sctp_sf_cookie_wait_abort(ep, asoc, type, arg, commands); +} + +/* + * Stop T1 timer and abort association with "INIT failed". + * + * This is common code called by several sctp_sf_*_abort() functions above. + */ +void sctp_stop_t1_and_abort(sctp_cmd_seq_t *commands, __u16 error) +{ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + /* CMD_INIT_FAILED will DELETE_TCB. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, + SCTP_U32(error)); +} + +/* + * sctp_sf_do_9_2_shut + * + * Section: 9.2 + * Upon the reception of the SHUTDOWN, the peer endpoint shall + * - enter the SHUTDOWN-RECEIVED state, + * + * - stop accepting new data from its SCTP user + * + * - verify, by checking the Cumulative TSN Ack field of the chunk, + * that all its outstanding DATA chunks have been received by the + * SHUTDOWN sender. + * + * Once an endpoint as reached the SHUTDOWN-RECEIVED state it MUST NOT + * send a SHUTDOWN in response to a ULP request. And should discard + * subsequent SHUTDOWN chunks. + * + * If there are still outstanding DATA chunks left, the SHUTDOWN + * receiver shall continue to follow normal data transmission + * procedures defined in Section 6 until all outstanding DATA chunks + * are acknowledged; however, the SHUTDOWN receiver MUST NOT accept + * new data from its SCTP user. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_9_2_shutdown(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + sctp_shutdownhdr_t *sdh; + sctp_disposition_t disposition; + struct sctp_ulpevent *ev; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the SHUTDOWN chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, + sizeof(struct sctp_shutdown_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* Convert the elaborate header. */ + sdh = (sctp_shutdownhdr_t *)chunk->skb->data; + skb_pull(chunk->skb, sizeof(sctp_shutdownhdr_t)); + chunk->subh.shutdown_hdr = sdh; + + /* Upon the reception of the SHUTDOWN, the peer endpoint shall + * - enter the SHUTDOWN-RECEIVED state, + * - stop accepting new data from its SCTP user + * + * [This is implicit in the new state.] + */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_SHUTDOWN_RECEIVED)); + disposition = SCTP_DISPOSITION_CONSUME; + + if (sctp_outq_is_empty(&asoc->outqueue)) { + disposition = sctp_sf_do_9_2_shutdown_ack(ep, asoc, type, + arg, commands); + } + + if (SCTP_DISPOSITION_NOMEM == disposition) + goto out; + + /* - verify, by checking the Cumulative TSN Ack field of the + * chunk, that all its outstanding DATA chunks have been + * received by the SHUTDOWN sender. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_CTSN, + SCTP_U32(chunk->subh.shutdown_hdr->cum_tsn_ack)); + + /* API 5.3.1.5 SCTP_SHUTDOWN_EVENT + * When a peer sends a SHUTDOWN, SCTP delivers this notification to + * inform the application that it should cease sending data. + */ + ev = sctp_ulpevent_make_shutdown_event(asoc, 0, GFP_ATOMIC); + if (!ev) { + disposition = SCTP_DISPOSITION_NOMEM; + goto out; + } + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + +out: + return disposition; +} + +/* RFC 2960 9.2 + * If an endpoint is in SHUTDOWN-ACK-SENT state and receives an INIT chunk + * (e.g., if the SHUTDOWN COMPLETE was lost) with source and destination + * transport addresses (either in the IP addresses or in the INIT chunk) + * that belong to this association, it should discard the INIT chunk and + * retransmit the SHUTDOWN ACK chunk. + */ +sctp_disposition_t sctp_sf_do_9_2_reshutack(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = (struct sctp_chunk *) arg; + struct sctp_chunk *reply; + + /* Since we are not going to really process this INIT, there + * is no point in verifying chunk boundries. Just generate + * the SHUTDOWN ACK. + */ + reply = sctp_make_shutdown_ack(asoc, chunk); + if (NULL == reply) + goto nomem; + + /* Set the transport for the SHUTDOWN ACK chunk and the timeout for + * the T2-SHUTDOWN timer. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_SETUP_T2, SCTP_CHUNK(reply)); + + /* and restart the T2-shutdown timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + + return SCTP_DISPOSITION_CONSUME; +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * sctp_sf_do_ecn_cwr + * + * Section: Appendix A: Explicit Congestion Notification + * + * CWR: + * + * RFC 2481 details a specific bit for a sender to send in the header of + * its next outbound TCP segment to indicate to its peer that it has + * reduced its congestion window. This is termed the CWR bit. For + * SCTP the same indication is made by including the CWR chunk. + * This chunk contains one data element, i.e. the TSN number that + * was sent in the ECNE chunk. This element represents the lowest + * TSN number in the datagram that was originally marked with the + * CE bit. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_ecn_cwr(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_cwrhdr_t *cwr; + struct sctp_chunk *chunk = arg; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_ecne_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + cwr = (sctp_cwrhdr_t *) chunk->skb->data; + skb_pull(chunk->skb, sizeof(sctp_cwrhdr_t)); + + cwr->lowest_tsn = ntohl(cwr->lowest_tsn); + + /* Does this CWR ack the last sent congestion notification? */ + if (TSN_lte(asoc->last_ecne_tsn, cwr->lowest_tsn)) { + /* Stop sending ECNE. */ + sctp_add_cmd_sf(commands, + SCTP_CMD_ECN_CWR, + SCTP_U32(cwr->lowest_tsn)); + } + return SCTP_DISPOSITION_CONSUME; +} + +/* + * sctp_sf_do_ecne + * + * Section: Appendix A: Explicit Congestion Notification + * + * ECN-Echo + * + * RFC 2481 details a specific bit for a receiver to send back in its + * TCP acknowledgements to notify the sender of the Congestion + * Experienced (CE) bit having arrived from the network. For SCTP this + * same indication is made by including the ECNE chunk. This chunk + * contains one data element, i.e. the lowest TSN associated with the IP + * datagram marked with the CE bit..... + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_ecne(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_ecnehdr_t *ecne; + struct sctp_chunk *chunk = arg; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_ecne_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + ecne = (sctp_ecnehdr_t *) chunk->skb->data; + skb_pull(chunk->skb, sizeof(sctp_ecnehdr_t)); + + /* If this is a newer ECNE than the last CWR packet we sent out */ + sctp_add_cmd_sf(commands, SCTP_CMD_ECN_ECNE, + SCTP_U32(ntohl(ecne->lowest_tsn))); + + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Section: 6.2 Acknowledgement on Reception of DATA Chunks + * + * The SCTP endpoint MUST always acknowledge the reception of each valid + * DATA chunk. + * + * The guidelines on delayed acknowledgement algorithm specified in + * Section 4.2 of [RFC2581] SHOULD be followed. Specifically, an + * acknowledgement SHOULD be generated for at least every second packet + * (not every second DATA chunk) received, and SHOULD be generated within + * 200 ms of the arrival of any unacknowledged DATA chunk. In some + * situations it may be beneficial for an SCTP transmitter to be more + * conservative than the algorithms detailed in this document allow. + * However, an SCTP transmitter MUST NOT be more aggressive than the + * following algorithms allow. + * + * A SCTP receiver MUST NOT generate more than one SACK for every + * incoming packet, other than to update the offered window as the + * receiving application consumes new data. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_eat_data_6_2(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + int error; + + if (!sctp_vtag_verify(chunk, asoc)) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, + SCTP_NULL()); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + } + + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_data_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + error = sctp_eat_data(asoc, chunk, commands ); + switch (error) { + case SCTP_IERROR_NO_ERROR: + break; + case SCTP_IERROR_HIGH_TSN: + case SCTP_IERROR_BAD_STREAM: + goto discard_noforce; + case SCTP_IERROR_DUP_TSN: + case SCTP_IERROR_IGNORE_TSN: + goto discard_force; + case SCTP_IERROR_NO_DATA: + goto consume; + default: + BUG(); + } + + if (asoc->autoclose) { + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); + } + + /* If this is the last chunk in a packet, we need to count it + * toward sack generation. Note that we need to SACK every + * OTHER packet containing data chunks, EVEN IF WE DISCARD + * THEM. We elect to NOT generate SACK's if the chunk fails + * the verification tag test. + * + * RFC 2960 6.2 Acknowledgement on Reception of DATA Chunks + * + * The SCTP endpoint MUST always acknowledge the reception of + * each valid DATA chunk. + * + * The guidelines on delayed acknowledgement algorithm + * specified in Section 4.2 of [RFC2581] SHOULD be followed. + * Specifically, an acknowledgement SHOULD be generated for at + * least every second packet (not every second DATA chunk) + * received, and SHOULD be generated within 200 ms of the + * arrival of any unacknowledged DATA chunk. In some + * situations it may be beneficial for an SCTP transmitter to + * be more conservative than the algorithms detailed in this + * document allow. However, an SCTP transmitter MUST NOT be + * more aggressive than the following algorithms allow. + */ + if (chunk->end_of_packet) { + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_NOFORCE()); + + /* Start the SACK timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_SACK)); + } + + return SCTP_DISPOSITION_CONSUME; + +discard_force: + /* RFC 2960 6.2 Acknowledgement on Reception of DATA Chunks + * + * When a packet arrives with duplicate DATA chunk(s) and with + * no new DATA chunk(s), the endpoint MUST immediately send a + * SACK with no delay. If a packet arrives with duplicate + * DATA chunk(s) bundled with new DATA chunks, the endpoint + * MAY immediately send a SACK. Normally receipt of duplicate + * DATA chunks will occur when the original SACK chunk was lost + * and the peer's RTO has expired. The duplicate TSN number(s) + * SHOULD be reported in the SACK as duplicate. + */ + /* In our case, we split the MAY SACK advice up whether or not + * the last chunk is a duplicate.' + */ + if (chunk->end_of_packet) + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_FORCE()); + return SCTP_DISPOSITION_DISCARD; + +discard_noforce: + if (chunk->end_of_packet) { + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_NOFORCE()); + + /* Start the SACK timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_SACK)); + } + return SCTP_DISPOSITION_DISCARD; +consume: + return SCTP_DISPOSITION_CONSUME; + +} + +/* + * sctp_sf_eat_data_fast_4_4 + * + * Section: 4 (4) + * (4) In SHUTDOWN-SENT state the endpoint MUST acknowledge any received + * DATA chunks without delay. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_eat_data_fast_4_4(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + int error; + + if (!sctp_vtag_verify(chunk, asoc)) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, + SCTP_NULL()); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + } + + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_data_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + error = sctp_eat_data(asoc, chunk, commands ); + switch (error) { + case SCTP_IERROR_NO_ERROR: + case SCTP_IERROR_HIGH_TSN: + case SCTP_IERROR_DUP_TSN: + case SCTP_IERROR_IGNORE_TSN: + case SCTP_IERROR_BAD_STREAM: + break; + case SCTP_IERROR_NO_DATA: + goto consume; + default: + BUG(); + } + + /* Go a head and force a SACK, since we are shutting down. */ + + /* Implementor's Guide. + * + * While in SHUTDOWN-SENT state, the SHUTDOWN sender MUST immediately + * respond to each received packet containing one or more DATA chunk(s) + * with a SACK, a SHUTDOWN chunk, and restart the T2-shutdown timer + */ + if (chunk->end_of_packet) { + /* We must delay the chunk creation since the cumulative + * TSN has not been updated yet. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SHUTDOWN, SCTP_NULL()); + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_FORCE()); + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + } + +consume: + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Section: 6.2 Processing a Received SACK + * D) Any time a SACK arrives, the endpoint performs the following: + * + * i) If Cumulative TSN Ack is less than the Cumulative TSN Ack Point, + * then drop the SACK. Since Cumulative TSN Ack is monotonically + * increasing, a SACK whose Cumulative TSN Ack is less than the + * Cumulative TSN Ack Point indicates an out-of-order SACK. + * + * ii) Set rwnd equal to the newly received a_rwnd minus the number + * of bytes still outstanding after processing the Cumulative TSN Ack + * and the Gap Ack Blocks. + * + * iii) If the SACK is missing a TSN that was previously + * acknowledged via a Gap Ack Block (e.g., the data receiver + * reneged on the data), then mark the corresponding DATA chunk + * as available for retransmit: Mark it as missing for fast + * retransmit as described in Section 7.2.4 and if no retransmit + * timer is running for the destination address to which the DATA + * chunk was originally transmitted, then T3-rtx is started for + * that destination address. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_eat_sack_6_2(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + sctp_sackhdr_t *sackh; + __u32 ctsn; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the SACK chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_sack_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* Pull the SACK chunk from the data buffer */ + sackh = sctp_sm_pull_sack(chunk); + /* Was this a bogus SACK? */ + if (!sackh) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + chunk->subh.sack_hdr = sackh; + ctsn = ntohl(sackh->cum_tsn_ack); + + /* i) If Cumulative TSN Ack is less than the Cumulative TSN + * Ack Point, then drop the SACK. Since Cumulative TSN + * Ack is monotonically increasing, a SACK whose + * Cumulative TSN Ack is less than the Cumulative TSN Ack + * Point indicates an out-of-order SACK. + */ + if (TSN_lt(ctsn, asoc->ctsn_ack_point)) { + SCTP_DEBUG_PRINTK("ctsn %x\n", ctsn); + SCTP_DEBUG_PRINTK("ctsn_ack_point %x\n", asoc->ctsn_ack_point); + return SCTP_DISPOSITION_DISCARD; + } + + /* Return this SACK for further processing. */ + sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_SACK, SCTP_SACKH(sackh)); + + /* Note: We do the rest of the work on the PROCESS_SACK + * sideeffect. + */ + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Generate an ABORT in response to a packet. + * + * Section: 8.4 Handle "Out of the blue" Packets + * + * 8) The receiver should respond to the sender of the OOTB packet + * with an ABORT. When sending the ABORT, the receiver of the + * OOTB packet MUST fill in the Verification Tag field of the + * outbound packet with the value found in the Verification Tag + * field of the OOTB packet and set the T-bit in the Chunk Flags + * to indicate that no TCB was found. After sending this ABORT, + * the receiver of the OOTB packet shall discard the OOTB packet + * and take no further action. + * + * Verification Tag: + * + * The return value is the disposition of the chunk. +*/ +sctp_disposition_t sctp_sf_tabort_8_4_8(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_packet *packet = NULL; + struct sctp_chunk *chunk = arg; + struct sctp_chunk *abort; + + packet = sctp_ootb_pkt_new(asoc, chunk); + + if (packet) { + /* Make an ABORT. The T bit will be set if the asoc + * is NULL. + */ + abort = sctp_make_abort(asoc, chunk, 0); + if (!abort) { + sctp_ootb_pkt_free(packet); + return SCTP_DISPOSITION_NOMEM; + } + + /* Set the skb to the belonging sock for accounting. */ + abort->skb->sk = ep->base.sk; + + sctp_packet_append_chunk(packet, abort); + + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, + SCTP_PACKET(packet)); + + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + + return SCTP_DISPOSITION_CONSUME; + } + + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Received an ERROR chunk from peer. Generate SCTP_REMOTE_ERROR + * event as ULP notification for each cause included in the chunk. + * + * API 5.3.1.3 - SCTP_REMOTE_ERROR + * + * The return value is the disposition of the chunk. +*/ +sctp_disposition_t sctp_sf_operr_notify(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_ulpevent *ev; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the ERROR chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_operr_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + while (chunk->chunk_end > chunk->skb->data) { + ev = sctp_ulpevent_make_remote_error(asoc, chunk, 0, + GFP_ATOMIC); + if (!ev) + goto nomem; + + if (!sctp_add_cmd(commands, SCTP_CMD_EVENT_ULP, + SCTP_ULPEVENT(ev))) { + sctp_ulpevent_free(ev); + goto nomem; + } + + sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_OPERR, + SCTP_CHUNK(chunk)); + } + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Process an inbound SHUTDOWN ACK. + * + * From Section 9.2: + * Upon the receipt of the SHUTDOWN ACK, the SHUTDOWN sender shall + * stop the T2-shutdown timer, send a SHUTDOWN COMPLETE chunk to its + * peer, and remove all record of the association. + * + * The return value is the disposition. + */ +sctp_disposition_t sctp_sf_do_9_2_final(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_chunk *reply; + struct sctp_ulpevent *ev; + + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the SHUTDOWN_ACK chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_chunkhdr_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + /* 10.2 H) SHUTDOWN COMPLETE notification + * + * When SCTP completes the shutdown procedures (section 9.2) this + * notification is passed to the upper layer. + */ + ev = sctp_ulpevent_make_assoc_change(asoc, 0, SCTP_SHUTDOWN_COMP, + 0, 0, 0, GFP_ATOMIC); + if (!ev) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_EVENT_ULP, SCTP_ULPEVENT(ev)); + + /* Upon the receipt of the SHUTDOWN ACK, the SHUTDOWN sender shall + * stop the T2-shutdown timer, + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + /* ...send a SHUTDOWN COMPLETE chunk to its peer, */ + reply = sctp_make_shutdown_complete(asoc, chunk); + if (!reply) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + SCTP_INC_STATS(SCTP_MIB_SHUTDOWNS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + + /* ...and remove all record of the association. */ + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + return SCTP_DISPOSITION_DELETE_TCB; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * RFC 2960, 8.4 - Handle "Out of the blue" Packets + * 5) If the packet contains a SHUTDOWN ACK chunk, the receiver should + * respond to the sender of the OOTB packet with a SHUTDOWN COMPLETE. + * When sending the SHUTDOWN COMPLETE, the receiver of the OOTB + * packet must fill in the Verification Tag field of the outbound + * packet with the Verification Tag received in the SHUTDOWN ACK and + * set the T-bit in the Chunk Flags to indicate that no TCB was + * found. Otherwise, + * + * 8) The receiver should respond to the sender of the OOTB packet with + * an ABORT. When sending the ABORT, the receiver of the OOTB packet + * MUST fill in the Verification Tag field of the outbound packet + * with the value found in the Verification Tag field of the OOTB + * packet and set the T-bit in the Chunk Flags to indicate that no + * TCB was found. After sending this ABORT, the receiver of the OOTB + * packet shall discard the OOTB packet and take no further action. + */ +sctp_disposition_t sctp_sf_ootb(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sk_buff *skb = chunk->skb; + sctp_chunkhdr_t *ch; + __u8 *ch_end; + int ootb_shut_ack = 0; + + SCTP_INC_STATS(SCTP_MIB_OUTOFBLUES); + + ch = (sctp_chunkhdr_t *) chunk->chunk_hdr; + do { + /* Break out if chunk length is less then minimal. */ + if (ntohs(ch->length) < sizeof(sctp_chunkhdr_t)) + break; + + ch_end = ((__u8 *)ch) + WORD_ROUND(ntohs(ch->length)); + + if (SCTP_CID_SHUTDOWN_ACK == ch->type) + ootb_shut_ack = 1; + + /* RFC 2960, Section 3.3.7 + * Moreover, under any circumstances, an endpoint that + * receives an ABORT MUST NOT respond to that ABORT by + * sending an ABORT of its own. + */ + if (SCTP_CID_ABORT == ch->type) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + ch = (sctp_chunkhdr_t *) ch_end; + } while (ch_end < skb->tail); + + if (ootb_shut_ack) + sctp_sf_shut_8_4_5(ep, asoc, type, arg, commands); + else + sctp_sf_tabort_8_4_8(ep, asoc, type, arg, commands); + + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); +} + +/* + * Handle an "Out of the blue" SHUTDOWN ACK. + * + * Section: 8.4 5) + * 5) If the packet contains a SHUTDOWN ACK chunk, the receiver should + * respond to the sender of the OOTB packet with a SHUTDOWN COMPLETE. + * When sending the SHUTDOWN COMPLETE, the receiver of the OOTB packet + * must fill in the Verification Tag field of the outbound packet with + * the Verification Tag received in the SHUTDOWN ACK and set the + * T-bit in the Chunk Flags to indicate that no TCB was found. + * + * Inputs + * (endpoint, asoc, type, arg, commands) + * + * Outputs + * (sctp_disposition_t) + * + * The return value is the disposition of the chunk. + */ +static sctp_disposition_t sctp_sf_shut_8_4_5(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_packet *packet = NULL; + struct sctp_chunk *chunk = arg; + struct sctp_chunk *shut; + + packet = sctp_ootb_pkt_new(asoc, chunk); + + if (packet) { + /* Make an SHUTDOWN_COMPLETE. + * The T bit will be set if the asoc is NULL. + */ + shut = sctp_make_shutdown_complete(asoc, chunk); + if (!shut) { + sctp_ootb_pkt_free(packet); + return SCTP_DISPOSITION_NOMEM; + } + + /* Set the skb to the belonging sock for accounting. */ + shut->skb->sk = ep->base.sk; + + sctp_packet_append_chunk(packet, shut); + + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, + SCTP_PACKET(packet)); + + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + + /* If the chunk length is invalid, we don't want to process + * the reset of the packet. + */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_chunkhdr_t))) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + return SCTP_DISPOSITION_CONSUME; + } + + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Handle SHUTDOWN ACK in COOKIE_ECHOED or COOKIE_WAIT state. + * + * Verification Tag: 8.5.1 E) Rules for packet carrying a SHUTDOWN ACK + * If the receiver is in COOKIE-ECHOED or COOKIE-WAIT state the + * procedures in section 8.4 SHOULD be followed, in other words it + * should be treated as an Out Of The Blue packet. + * [This means that we do NOT check the Verification Tag on these + * chunks. --piggy ] + * + */ +sctp_disposition_t sctp_sf_do_8_5_1_E_sa(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* Although we do have an association in this case, it corresponds + * to a restarted association. So the packet is treated as an OOTB + * packet and the state function that handles OOTB SHUTDOWN_ACK is + * called with a NULL association. + */ + return sctp_sf_shut_8_4_5(ep, NULL, type, arg, commands); +} + +/* ADDIP Section 4.2 Upon reception of an ASCONF Chunk. */ +sctp_disposition_t sctp_sf_do_asconf(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_chunk *asconf_ack = NULL; + sctp_addiphdr_t *hdr; + __u32 serial; + + if (!sctp_vtag_verify(chunk, asoc)) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, + SCTP_NULL()); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + } + + /* Make sure that the ASCONF ADDIP chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(sctp_addip_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + hdr = (sctp_addiphdr_t *)chunk->skb->data; + serial = ntohl(hdr->serial); + + /* ADDIP 4.2 C1) Compare the value of the serial number to the value + * the endpoint stored in a new association variable + * 'Peer-Serial-Number'. + */ + if (serial == asoc->peer.addip_serial + 1) { + /* ADDIP 4.2 C2) If the value found in the serial number is + * equal to the ('Peer-Serial-Number' + 1), the endpoint MUST + * do V1-V5. + */ + asconf_ack = sctp_process_asconf((struct sctp_association *) + asoc, chunk); + if (!asconf_ack) + return SCTP_DISPOSITION_NOMEM; + } else if (serial == asoc->peer.addip_serial) { + /* ADDIP 4.2 C3) If the value found in the serial number is + * equal to the value stored in the 'Peer-Serial-Number' + * IMPLEMENTATION NOTE: As an optimization a receiver may wish + * to save the last ASCONF-ACK for some predetermined period of + * time and instead of re-processing the ASCONF (with the same + * serial number) it may just re-transmit the ASCONF-ACK. + */ + if (asoc->addip_last_asconf_ack) + asconf_ack = asoc->addip_last_asconf_ack; + else + return SCTP_DISPOSITION_DISCARD; + } else { + /* ADDIP 4.2 C4) Otherwise, the ASCONF Chunk is discarded since + * it must be either a stale packet or from an attacker. + */ + return SCTP_DISPOSITION_DISCARD; + } + + /* ADDIP 4.2 C5) In both cases C2 and C3 the ASCONF-ACK MUST be sent + * back to the source address contained in the IP header of the ASCONF + * being responded to. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(asconf_ack)); + + return SCTP_DISPOSITION_CONSUME; +} + +/* + * ADDIP Section 4.3 General rules for address manipulation + * When building TLV parameters for the ASCONF Chunk that will add or + * delete IP addresses the D0 to D13 rules should be applied: + */ +sctp_disposition_t sctp_sf_do_asconf_ack(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *asconf_ack = arg; + struct sctp_chunk *last_asconf = asoc->addip_last_asconf; + struct sctp_chunk *abort; + sctp_addiphdr_t *addip_hdr; + __u32 sent_serial, rcvd_serial; + + if (!sctp_vtag_verify(asconf_ack, asoc)) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, + SCTP_NULL()); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + } + + /* Make sure that the ADDIP chunk has a valid length. */ + if (!sctp_chunk_length_valid(asconf_ack, sizeof(sctp_addip_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + addip_hdr = (sctp_addiphdr_t *)asconf_ack->skb->data; + rcvd_serial = ntohl(addip_hdr->serial); + + if (last_asconf) { + addip_hdr = (sctp_addiphdr_t *)last_asconf->subh.addip_hdr; + sent_serial = ntohl(addip_hdr->serial); + } else { + sent_serial = asoc->addip_serial - 1; + } + + /* D0) If an endpoint receives an ASCONF-ACK that is greater than or + * equal to the next serial number to be used but no ASCONF chunk is + * outstanding the endpoint MUST ABORT the association. Note that a + * sequence number is greater than if it is no more than 2^^31-1 + * larger than the current sequence number (using serial arithmetic). + */ + if (ADDIP_SERIAL_gte(rcvd_serial, sent_serial + 1) && + !(asoc->addip_last_asconf)) { + abort = sctp_make_abort(asoc, asconf_ack, + sizeof(sctp_errhdr_t)); + if (abort) { + sctp_init_cause(abort, SCTP_ERROR_ASCONF_ACK, NULL, 0); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(abort)); + } + /* We are going to ABORT, so we might as well stop + * processing the rest of the chunks in the packet. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T4_RTO)); + sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET,SCTP_NULL()); + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_ASCONF_ACK)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_ABORT; + } + + if ((rcvd_serial == sent_serial) && asoc->addip_last_asconf) { + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T4_RTO)); + + if (!sctp_process_asconf_ack((struct sctp_association *)asoc, + asconf_ack)) + return SCTP_DISPOSITION_CONSUME; + + abort = sctp_make_abort(asoc, asconf_ack, + sizeof(sctp_errhdr_t)); + if (abort) { + sctp_init_cause(abort, SCTP_ERROR_RSRC_LOW, NULL, 0); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(abort)); + } + /* We are going to ABORT, so we might as well stop + * processing the rest of the chunks in the packet. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET,SCTP_NULL()); + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_ASCONF_ACK)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_ABORT; + } + + return SCTP_DISPOSITION_DISCARD; +} + +/* + * PR-SCTP Section 3.6 Receiver Side Implementation of PR-SCTP + * + * When a FORWARD TSN chunk arrives, the data receiver MUST first update + * its cumulative TSN point to the value carried in the FORWARD TSN + * chunk, and then MUST further advance its cumulative TSN point locally + * if possible. + * After the above processing, the data receiver MUST stop reporting any + * missing TSNs earlier than or equal to the new cumulative TSN point. + * + * Verification Tag: 8.5 Verification Tag [Normal verification] + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_eat_fwd_tsn(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_fwdtsn_hdr *fwdtsn_hdr; + __u16 len; + __u32 tsn; + + if (!sctp_vtag_verify(chunk, asoc)) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, + SCTP_NULL()); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + } + + /* Make sure that the FORWARD_TSN chunk has valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_fwdtsn_chunk))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + fwdtsn_hdr = (struct sctp_fwdtsn_hdr *)chunk->skb->data; + chunk->subh.fwdtsn_hdr = fwdtsn_hdr; + len = ntohs(chunk->chunk_hdr->length); + len -= sizeof(struct sctp_chunkhdr); + skb_pull(chunk->skb, len); + + tsn = ntohl(fwdtsn_hdr->new_cum_tsn); + SCTP_DEBUG_PRINTK("%s: TSN 0x%x.\n", __FUNCTION__, tsn); + + /* The TSN is too high--silently discard the chunk and count on it + * getting retransmitted later. + */ + if (sctp_tsnmap_check(&asoc->peer.tsn_map, tsn) < 0) + goto discard_noforce; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_FWDTSN, SCTP_U32(tsn)); + if (len > sizeof(struct sctp_fwdtsn_hdr)) + sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_FWDTSN, + SCTP_CHUNK(chunk)); + + /* Count this as receiving DATA. */ + if (asoc->autoclose) { + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); + } + + /* FIXME: For now send a SACK, but DATA processing may + * send another. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_NOFORCE()); + /* Start the SACK timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_SACK)); + + return SCTP_DISPOSITION_CONSUME; + +discard_noforce: + return SCTP_DISPOSITION_DISCARD; +} + +sctp_disposition_t sctp_sf_eat_fwd_tsn_fast( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_fwdtsn_hdr *fwdtsn_hdr; + __u16 len; + __u32 tsn; + + if (!sctp_vtag_verify(chunk, asoc)) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_BAD_TAG, + SCTP_NULL()); + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + } + + /* Make sure that the FORWARD_TSN chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_fwdtsn_chunk))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + fwdtsn_hdr = (struct sctp_fwdtsn_hdr *)chunk->skb->data; + chunk->subh.fwdtsn_hdr = fwdtsn_hdr; + len = ntohs(chunk->chunk_hdr->length); + len -= sizeof(struct sctp_chunkhdr); + skb_pull(chunk->skb, len); + + tsn = ntohl(fwdtsn_hdr->new_cum_tsn); + SCTP_DEBUG_PRINTK("%s: TSN 0x%x.\n", __FUNCTION__, tsn); + + /* The TSN is too high--silently discard the chunk and count on it + * getting retransmitted later. + */ + if (sctp_tsnmap_check(&asoc->peer.tsn_map, tsn) < 0) + goto gen_shutdown; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_FWDTSN, SCTP_U32(tsn)); + if (len > sizeof(struct sctp_fwdtsn_hdr)) + sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_FWDTSN, + SCTP_CHUNK(chunk)); + + /* Go a head and force a SACK, since we are shutting down. */ +gen_shutdown: + /* Implementor's Guide. + * + * While in SHUTDOWN-SENT state, the SHUTDOWN sender MUST immediately + * respond to each received packet containing one or more DATA chunk(s) + * with a SACK, a SHUTDOWN chunk, and restart the T2-shutdown timer + */ + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SHUTDOWN, SCTP_NULL()); + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_FORCE()); + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Process an unknown chunk. + * + * Section: 3.2. Also, 2.1 in the implementor's guide. + * + * Chunk Types are encoded such that the highest-order two bits specify + * the action that must be taken if the processing endpoint does not + * recognize the Chunk Type. + * + * 00 - Stop processing this SCTP packet and discard it, do not process + * any further chunks within it. + * + * 01 - Stop processing this SCTP packet and discard it, do not process + * any further chunks within it, and report the unrecognized + * chunk in an 'Unrecognized Chunk Type'. + * + * 10 - Skip this chunk and continue processing. + * + * 11 - Skip this chunk and continue processing, but report in an ERROR + * Chunk using the 'Unrecognized Chunk Type' cause of error. + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_unk_chunk(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *unk_chunk = arg; + struct sctp_chunk *err_chunk; + sctp_chunkhdr_t *hdr; + + SCTP_DEBUG_PRINTK("Processing the unknown chunk id %d.\n", type.chunk); + + if (!sctp_vtag_verify(unk_chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the chunk has a valid length. + * Since we don't know the chunk type, we use a general + * chunkhdr structure to make a comparison. + */ + if (!sctp_chunk_length_valid(unk_chunk, sizeof(sctp_chunkhdr_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + + switch (type.chunk & SCTP_CID_ACTION_MASK) { + case SCTP_CID_ACTION_DISCARD: + /* Discard the packet. */ + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + break; + case SCTP_CID_ACTION_DISCARD_ERR: + /* Discard the packet. */ + sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Generate an ERROR chunk as response. */ + hdr = unk_chunk->chunk_hdr; + err_chunk = sctp_make_op_error(asoc, unk_chunk, + SCTP_ERROR_UNKNOWN_CHUNK, hdr, + WORD_ROUND(ntohs(hdr->length))); + if (err_chunk) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(err_chunk)); + } + return SCTP_DISPOSITION_CONSUME; + break; + case SCTP_CID_ACTION_SKIP: + /* Skip the chunk. */ + return SCTP_DISPOSITION_DISCARD; + break; + case SCTP_CID_ACTION_SKIP_ERR: + /* Generate an ERROR chunk as response. */ + hdr = unk_chunk->chunk_hdr; + err_chunk = sctp_make_op_error(asoc, unk_chunk, + SCTP_ERROR_UNKNOWN_CHUNK, hdr, + WORD_ROUND(ntohs(hdr->length))); + if (err_chunk) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(err_chunk)); + } + /* Skip the chunk. */ + return SCTP_DISPOSITION_CONSUME; + break; + default: + break; + } + + return SCTP_DISPOSITION_DISCARD; +} + +/* + * Discard the chunk. + * + * Section: 0.2, 5.2.3, 5.2.5, 5.2.6, 6.0, 8.4.6, 8.5.1c, 9.2 + * [Too numerous to mention...] + * Verification Tag: No verification needed. + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_discard_chunk(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + SCTP_DEBUG_PRINTK("Chunk %d is discarded\n", type.chunk); + return SCTP_DISPOSITION_DISCARD; +} + +/* + * Discard the whole packet. + * + * Section: 8.4 2) + * + * 2) If the OOTB packet contains an ABORT chunk, the receiver MUST + * silently discard the OOTB packet and take no further action. + * Otherwise, + * + * Verification Tag: No verification necessary + * + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_pdiscard(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET, SCTP_NULL()); + + return SCTP_DISPOSITION_CONSUME; +} + + +/* + * The other end is violating protocol. + * + * Section: Not specified + * Verification Tag: Not specified + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (asoc, reply_msg, msg_up, timers, counters) + * + * We simply tag the chunk as a violation. The state machine will log + * the violation and continue. + */ +sctp_disposition_t sctp_sf_violation(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + return SCTP_DISPOSITION_VIOLATION; +} + + +/* + * Handle a protocol violation when the chunk length is invalid. + * "Invalid" length is identified as smaller then the minimal length a + * given chunk can be. For example, a SACK chunk has invalid length + * if it's length is set to be smaller then the size of sctp_sack_chunk_t. + * + * We inform the other end by sending an ABORT with a Protocol Violation + * error code. + * + * Section: Not specified + * Verification Tag: Nothing to do + * Inputs + * (endpoint, asoc, chunk) + * + * Outputs + * (reply_msg, msg_up, counters) + * + * Generate an ABORT chunk and terminate the association. + */ +sctp_disposition_t sctp_sf_violation_chunklen(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + struct sctp_chunk *abort = NULL; + char err_str[]="The following chunk had invalid length:"; + + /* Make the abort chunk. */ + abort = sctp_make_abort_violation(asoc, chunk, err_str, + sizeof(err_str)); + if (!abort) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + + if (asoc->state <= SCTP_STATE_COOKIE_ECHOED) { + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, + SCTP_U32(SCTP_ERROR_PROTO_VIOLATION)); + } else { + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_PROTO_VIOLATION)); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + } + + sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET, SCTP_NULL()); + + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + + return SCTP_DISPOSITION_ABORT; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/*************************************************************************** + * These are the state functions for handling primitive (Section 10) events. + ***************************************************************************/ +/* + * sctp_sf_do_prm_asoc + * + * Section: 10.1 ULP-to-SCTP + * B) Associate + * + * Format: ASSOCIATE(local SCTP instance name, destination transport addr, + * outbound stream count) + * -> association id [,destination transport addr list] [,outbound stream + * count] + * + * This primitive allows the upper layer to initiate an association to a + * specific peer endpoint. + * + * The peer endpoint shall be specified by one of the transport addresses + * which defines the endpoint (see Section 1.4). If the local SCTP + * instance has not been initialized, the ASSOCIATE is considered an + * error. + * [This is not relevant for the kernel implementation since we do all + * initialization at boot time. It we hadn't initialized we wouldn't + * get anywhere near this code.] + * + * An association id, which is a local handle to the SCTP association, + * will be returned on successful establishment of the association. If + * SCTP is not able to open an SCTP association with the peer endpoint, + * an error is returned. + * [In the kernel implementation, the struct sctp_association needs to + * be created BEFORE causing this primitive to run.] + * + * Other association parameters may be returned, including the + * complete destination transport addresses of the peer as well as the + * outbound stream count of the local endpoint. One of the transport + * address from the returned destination addresses will be selected by + * the local endpoint as default primary path for sending SCTP packets + * to this peer. The returned "destination transport addr list" can + * be used by the ULP to change the default primary path or to force + * sending a packet to a specific transport address. [All of this + * stuff happens when the INIT ACK arrives. This is a NON-BLOCKING + * function.] + * + * Mandatory attributes: + * + * o local SCTP instance name - obtained from the INITIALIZE operation. + * [This is the argument asoc.] + * o destination transport addr - specified as one of the transport + * addresses of the peer endpoint with which the association is to be + * established. + * [This is asoc->peer.active_path.] + * o outbound stream count - the number of outbound streams the ULP + * would like to open towards this peer endpoint. + * [BUG: This is not currently implemented.] + * Optional attributes: + * + * None. + * + * The return value is a disposition. + */ +sctp_disposition_t sctp_sf_do_prm_asoc(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *repl; + + /* The comment below says that we enter COOKIE-WAIT AFTER + * sending the INIT, but that doesn't actually work in our + * implementation... + */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_COOKIE_WAIT)); + + /* RFC 2960 5.1 Normal Establishment of an Association + * + * A) "A" first sends an INIT chunk to "Z". In the INIT, "A" + * must provide its Verification Tag (Tag_A) in the Initiate + * Tag field. Tag_A SHOULD be a random number in the range of + * 1 to 4294967295 (see 5.3.1 for Tag value selection). ... + */ + + repl = sctp_make_init(asoc, &asoc->base.bind_addr, GFP_ATOMIC, 0); + if (!repl) + goto nomem; + + /* Cast away the const modifier, as we want to just + * rerun it through as a sideffect. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_ASOC, + SCTP_ASOC((struct sctp_association *) asoc)); + + /* After sending the INIT, "A" starts the T1-init timer and + * enters the COOKIE-WAIT state. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Process the SEND primitive. + * + * Section: 10.1 ULP-to-SCTP + * E) Send + * + * Format: SEND(association id, buffer address, byte count [,context] + * [,stream id] [,life time] [,destination transport address] + * [,unorder flag] [,no-bundle flag] [,payload protocol-id] ) + * -> result + * + * This is the main method to send user data via SCTP. + * + * Mandatory attributes: + * + * o association id - local handle to the SCTP association + * + * o buffer address - the location where the user message to be + * transmitted is stored; + * + * o byte count - The size of the user data in number of bytes; + * + * Optional attributes: + * + * o context - an optional 32 bit integer that will be carried in the + * sending failure notification to the ULP if the transportation of + * this User Message fails. + * + * o stream id - to indicate which stream to send the data on. If not + * specified, stream 0 will be used. + * + * o life time - specifies the life time of the user data. The user data + * will not be sent by SCTP after the life time expires. This + * parameter can be used to avoid efforts to transmit stale + * user messages. SCTP notifies the ULP if the data cannot be + * initiated to transport (i.e. sent to the destination via SCTP's + * send primitive) within the life time variable. However, the + * user data will be transmitted if SCTP has attempted to transmit a + * chunk before the life time expired. + * + * o destination transport address - specified as one of the destination + * transport addresses of the peer endpoint to which this packet + * should be sent. Whenever possible, SCTP should use this destination + * transport address for sending the packets, instead of the current + * primary path. + * + * o unorder flag - this flag, if present, indicates that the user + * would like the data delivered in an unordered fashion to the peer + * (i.e., the U flag is set to 1 on all DATA chunks carrying this + * message). + * + * o no-bundle flag - instructs SCTP not to bundle this user data with + * other outbound DATA chunks. SCTP MAY still bundle even when + * this flag is present, when faced with network congestion. + * + * o payload protocol-id - A 32 bit unsigned integer that is to be + * passed to the peer indicating the type of payload protocol data + * being transmitted. This value is passed as opaque data by SCTP. + * + * The return value is the disposition. + */ +sctp_disposition_t sctp_sf_do_prm_send(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(chunk)); + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Process the SHUTDOWN primitive. + * + * Section: 10.1: + * C) Shutdown + * + * Format: SHUTDOWN(association id) + * -> result + * + * Gracefully closes an association. Any locally queued user data + * will be delivered to the peer. The association will be terminated only + * after the peer acknowledges all the SCTP packets sent. A success code + * will be returned on successful termination of the association. If + * attempting to terminate the association results in a failure, an error + * code shall be returned. + * + * Mandatory attributes: + * + * o association id - local handle to the SCTP association + * + * Optional attributes: + * + * None. + * + * The return value is the disposition. + */ +sctp_disposition_t sctp_sf_do_9_2_prm_shutdown( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + int disposition; + + /* From 9.2 Shutdown of an Association + * Upon receipt of the SHUTDOWN primitive from its upper + * layer, the endpoint enters SHUTDOWN-PENDING state and + * remains there until all outstanding data has been + * acknowledged by its peer. The endpoint accepts no new data + * from its upper layer, but retransmits data to the far end + * if necessary to fill gaps. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_SHUTDOWN_PENDING)); + + /* sctpimpguide-05 Section 2.12.2 + * The sender of the SHUTDOWN MAY also start an overall guard timer + * 'T5-shutdown-guard' to bound the overall time for shutdown sequence. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + disposition = SCTP_DISPOSITION_CONSUME; + if (sctp_outq_is_empty(&asoc->outqueue)) { + disposition = sctp_sf_do_9_2_start_shutdown(ep, asoc, type, + arg, commands); + } + return disposition; +} + +/* + * Process the ABORT primitive. + * + * Section: 10.1: + * C) Abort + * + * Format: Abort(association id [, cause code]) + * -> result + * + * Ungracefully closes an association. Any locally queued user data + * will be discarded and an ABORT chunk is sent to the peer. A success code + * will be returned on successful abortion of the association. If + * attempting to abort the association results in a failure, an error + * code shall be returned. + * + * Mandatory attributes: + * + * o association id - local handle to the SCTP association + * + * Optional attributes: + * + * o cause code - reason of the abort to be passed to the peer + * + * None. + * + * The return value is the disposition. + */ +sctp_disposition_t sctp_sf_do_9_1_prm_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* From 9.1 Abort of an Association + * Upon receipt of the ABORT primitive from its upper + * layer, the endpoint enters CLOSED state and + * discard all outstanding data has been + * acknowledged by its peer. The endpoint accepts no new data + * from its upper layer, but retransmits data to the far end + * if necessary to fill gaps. + */ + struct msghdr *msg = arg; + struct sctp_chunk *abort; + sctp_disposition_t retval; + + retval = SCTP_DISPOSITION_CONSUME; + + /* Generate ABORT chunk to send the peer. */ + abort = sctp_make_abort_user(asoc, NULL, msg); + if (!abort) + retval = SCTP_DISPOSITION_NOMEM; + else + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); + + /* Even if we can't send the ABORT due to low memory delete the + * TCB. This is a departure from our typical NOMEM handling. + */ + + /* Delete the established association. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_USER_ABORT)); + + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + + return retval; +} + +/* We tried an illegal operation on an association which is closed. */ +sctp_disposition_t sctp_sf_error_closed(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_ERROR, SCTP_ERROR(-EINVAL)); + return SCTP_DISPOSITION_CONSUME; +} + +/* We tried an illegal operation on an association which is shutting + * down. + */ +sctp_disposition_t sctp_sf_error_shutdown(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_ERROR, + SCTP_ERROR(-ESHUTDOWN)); + return SCTP_DISPOSITION_CONSUME; +} + +/* + * sctp_cookie_wait_prm_shutdown + * + * Section: 4 Note: 2 + * Verification Tag: + * Inputs + * (endpoint, asoc) + * + * The RFC does not explicitly address this issue, but is the route through the + * state table when someone issues a shutdown while in COOKIE_WAIT state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_cookie_wait_prm_shutdown( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + + SCTP_INC_STATS(SCTP_MIB_SHUTDOWNS); + + sctp_add_cmd_sf(commands, SCTP_CMD_DELETE_TCB, SCTP_NULL()); + + return SCTP_DISPOSITION_DELETE_TCB; +} + +/* + * sctp_cookie_echoed_prm_shutdown + * + * Section: 4 Note: 2 + * Verification Tag: + * Inputs + * (endpoint, asoc) + * + * The RFC does not explcitly address this issue, but is the route through the + * state table when someone issues a shutdown while in COOKIE_ECHOED state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_cookie_echoed_prm_shutdown( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, sctp_cmd_seq_t *commands) +{ + /* There is a single T1 timer, so we should be able to use + * common function with the COOKIE-WAIT state. + */ + return sctp_sf_cookie_wait_prm_shutdown(ep, asoc, type, arg, commands); +} + +/* + * sctp_sf_cookie_wait_prm_abort + * + * Section: 4 Note: 2 + * Verification Tag: + * Inputs + * (endpoint, asoc) + * + * The RFC does not explicitly address this issue, but is the route through the + * state table when someone issues an abort while in COOKIE_WAIT state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_cookie_wait_prm_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct msghdr *msg = arg; + struct sctp_chunk *abort; + sctp_disposition_t retval; + + /* Stop T1-init timer */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); + retval = SCTP_DISPOSITION_CONSUME; + + /* Generate ABORT chunk to send the peer */ + abort = sctp_make_abort_user(asoc, NULL, msg); + if (!abort) + retval = SCTP_DISPOSITION_NOMEM; + else + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); + + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_CLOSED)); + + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + + /* Even if we can't send the ABORT due to low memory delete the + * TCB. This is a departure from our typical NOMEM handling. + */ + + /* Delete the established association. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, + SCTP_U32(SCTP_ERROR_USER_ABORT)); + + return retval; +} + +/* + * sctp_sf_cookie_echoed_prm_abort + * + * Section: 4 Note: 3 + * Verification Tag: + * Inputs + * (endpoint, asoc) + * + * The RFC does not explcitly address this issue, but is the route through the + * state table when someone issues an abort while in COOKIE_ECHOED state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_cookie_echoed_prm_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* There is a single T1 timer, so we should be able to use + * common function with the COOKIE-WAIT state. + */ + return sctp_sf_cookie_wait_prm_abort(ep, asoc, type, arg, commands); +} + +/* + * sctp_sf_shutdown_pending_prm_abort + * + * Inputs + * (endpoint, asoc) + * + * The RFC does not explicitly address this issue, but is the route through the + * state table when someone issues an abort while in SHUTDOWN-PENDING state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_shutdown_pending_prm_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* Stop the T5-shutdown guard timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + return sctp_sf_do_9_1_prm_abort(ep, asoc, type, arg, commands); +} + +/* + * sctp_sf_shutdown_sent_prm_abort + * + * Inputs + * (endpoint, asoc) + * + * The RFC does not explicitly address this issue, but is the route through the + * state table when someone issues an abort while in SHUTDOWN-SENT state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_shutdown_sent_prm_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* Stop the T2-shutdown timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + /* Stop the T5-shutdown guard timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + + return sctp_sf_do_9_1_prm_abort(ep, asoc, type, arg, commands); +} + +/* + * sctp_sf_cookie_echoed_prm_abort + * + * Inputs + * (endpoint, asoc) + * + * The RFC does not explcitly address this issue, but is the route through the + * state table when someone issues an abort while in COOKIE_ECHOED state. + * + * Outputs + * (timers) + */ +sctp_disposition_t sctp_sf_shutdown_ack_sent_prm_abort( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + /* The same T2 timer, so we should be able to use + * common function with the SHUTDOWN-SENT state. + */ + return sctp_sf_shutdown_sent_prm_abort(ep, asoc, type, arg, commands); +} + +/* + * Process the REQUESTHEARTBEAT primitive + * + * 10.1 ULP-to-SCTP + * J) Request Heartbeat + * + * Format: REQUESTHEARTBEAT(association id, destination transport address) + * + * -> result + * + * Instructs the local endpoint to perform a HeartBeat on the specified + * destination transport address of the given association. The returned + * result should indicate whether the transmission of the HEARTBEAT + * chunk to the destination address is successful. + * + * Mandatory attributes: + * + * o association id - local handle to the SCTP association + * + * o destination transport address - the transport address of the + * association on which a heartbeat should be issued. + */ +sctp_disposition_t sctp_sf_do_prm_requestheartbeat( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + return sctp_sf_heartbeat(ep, asoc, type, (struct sctp_transport *)arg, + commands); +} + +/* + * ADDIP Section 4.1 ASCONF Chunk Procedures + * When an endpoint has an ASCONF signaled change to be sent to the + * remote endpoint it should do A1 to A9 + */ +sctp_disposition_t sctp_sf_do_prm_asconf(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = arg; + + sctp_add_cmd_sf(commands, SCTP_CMD_SETUP_T4, SCTP_CHUNK(chunk)); + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T4_RTO)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(chunk)); + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Ignore the primitive event + * + * The return value is the disposition of the primitive. + */ +sctp_disposition_t sctp_sf_ignore_primitive( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + SCTP_DEBUG_PRINTK("Primitive type %d is ignored.\n", type.primitive); + return SCTP_DISPOSITION_DISCARD; +} + +/*************************************************************************** + * These are the state functions for the OTHER events. + ***************************************************************************/ + +/* + * Start the shutdown negotiation. + * + * From Section 9.2: + * Once all its outstanding data has been acknowledged, the endpoint + * shall send a SHUTDOWN chunk to its peer including in the Cumulative + * TSN Ack field the last sequential TSN it has received from the peer. + * It shall then start the T2-shutdown timer and enter the SHUTDOWN-SENT + * state. If the timer expires, the endpoint must re-send the SHUTDOWN + * with the updated last sequential TSN received from its peer. + * + * The return value is the disposition. + */ +sctp_disposition_t sctp_sf_do_9_2_start_shutdown( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *reply; + + /* Once all its outstanding data has been acknowledged, the + * endpoint shall send a SHUTDOWN chunk to its peer including + * in the Cumulative TSN Ack field the last sequential TSN it + * has received from the peer. + */ + reply = sctp_make_shutdown(asoc, NULL); + if (!reply) + goto nomem; + + /* Set the transport for the SHUTDOWN chunk and the timeout for the + * T2-shutdown timer. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_SETUP_T2, SCTP_CHUNK(reply)); + + /* It shall then start the T2-shutdown timer */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + if (asoc->autoclose) + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); + + /* and enter the SHUTDOWN-SENT state. */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_SHUTDOWN_SENT)); + + /* sctp-implguide 2.10 Issues with Heartbeating and failover + * + * HEARTBEAT ... is discontinued after sending either SHUTDOWN + * or SHUTDOWN-ACK. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_STOP, SCTP_NULL()); + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Generate a SHUTDOWN ACK now that everything is SACK'd. + * + * From Section 9.2: + * + * If it has no more outstanding DATA chunks, the SHUTDOWN receiver + * shall send a SHUTDOWN ACK and start a T2-shutdown timer of its own, + * entering the SHUTDOWN-ACK-SENT state. If the timer expires, the + * endpoint must re-send the SHUTDOWN ACK. + * + * The return value is the disposition. + */ +sctp_disposition_t sctp_sf_do_9_2_shutdown_ack( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = (struct sctp_chunk *) arg; + struct sctp_chunk *reply; + + /* There are 2 ways of getting here: + * 1) called in response to a SHUTDOWN chunk + * 2) called when SCTP_EVENT_NO_PENDING_TSN event is issued. + * + * For the case (2), the arg parameter is set to NULL. We need + * to check that we have a chunk before accessing it's fields. + */ + if (chunk) { + if (!sctp_vtag_verify(chunk, asoc)) + return sctp_sf_pdiscard(ep, asoc, type, arg, commands); + + /* Make sure that the SHUTDOWN chunk has a valid length. */ + if (!sctp_chunk_length_valid(chunk, sizeof(struct sctp_shutdown_chunk_t))) + return sctp_sf_violation_chunklen(ep, asoc, type, arg, + commands); + } + + /* If it has no more outstanding DATA chunks, the SHUTDOWN receiver + * shall send a SHUTDOWN ACK ... + */ + reply = sctp_make_shutdown_ack(asoc, chunk); + if (!reply) + goto nomem; + + /* Set the transport for the SHUTDOWN ACK chunk and the timeout for + * the T2-shutdown timer. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_SETUP_T2, SCTP_CHUNK(reply)); + + /* and start/restart a T2-shutdown timer of its own, */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + + if (asoc->autoclose) + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_AUTOCLOSE)); + + /* Enter the SHUTDOWN-ACK-SENT state. */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_SHUTDOWN_ACK_SENT)); + + /* sctp-implguide 2.10 Issues with Heartbeating and failover + * + * HEARTBEAT ... is discontinued after sending either SHUTDOWN + * or SHUTDOWN-ACK. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_HB_TIMERS_STOP, SCTP_NULL()); + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * Ignore the event defined as other + * + * The return value is the disposition of the event. + */ +sctp_disposition_t sctp_sf_ignore_other(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + SCTP_DEBUG_PRINTK("The event other type %d is ignored\n", type.other); + return SCTP_DISPOSITION_DISCARD; +} + +/************************************************************ + * These are the state functions for handling timeout events. + ************************************************************/ + +/* + * RTX Timeout + * + * Section: 6.3.3 Handle T3-rtx Expiration + * + * Whenever the retransmission timer T3-rtx expires for a destination + * address, do the following: + * [See below] + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_do_6_3_3_rtx(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_transport *transport = arg; + + if (asoc->overall_error_count >= asoc->max_retrans) { + /* CMD_ASSOC_FAILED calls CMD_DELETE_TCB. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_DELETE_TCB; + } + + /* E1) For the destination address for which the timer + * expires, adjust its ssthresh with rules defined in Section + * 7.2.3 and set the cwnd <- MTU. + */ + + /* E2) For the destination address for which the timer + * expires, set RTO <- RTO * 2 ("back off the timer"). The + * maximum value discussed in rule C7 above (RTO.max) may be + * used to provide an upper bound to this doubling operation. + */ + + /* E3) Determine how many of the earliest (i.e., lowest TSN) + * outstanding DATA chunks for the address for which the + * T3-rtx has expired will fit into a single packet, subject + * to the MTU constraint for the path corresponding to the + * destination transport address to which the retransmission + * is being sent (this may be different from the address for + * which the timer expires [see Section 6.4]). Call this + * value K. Bundle and retransmit those K DATA chunks in a + * single packet to the destination endpoint. + * + * Note: Any DATA chunks that were sent to the address for + * which the T3-rtx timer expired but did not fit in one MTU + * (rule E3 above), should be marked for retransmission and + * sent as soon as cwnd allows (normally when a SACK arrives). + */ + + /* NB: Rules E4 and F1 are implicit in R1. */ + sctp_add_cmd_sf(commands, SCTP_CMD_RETRAN, SCTP_TRANSPORT(transport)); + + /* Do some failure management (Section 8.2). */ + sctp_add_cmd_sf(commands, SCTP_CMD_STRIKE, SCTP_TRANSPORT(transport)); + + return SCTP_DISPOSITION_CONSUME; +} + +/* + * Generate delayed SACK on timeout + * + * Section: 6.2 Acknowledgement on Reception of DATA Chunks + * + * The guidelines on delayed acknowledgement algorithm specified in + * Section 4.2 of [RFC2581] SHOULD be followed. Specifically, an + * acknowledgement SHOULD be generated for at least every second packet + * (not every second DATA chunk) received, and SHOULD be generated + * within 200 ms of the arrival of any unacknowledged DATA chunk. In + * some situations it may be beneficial for an SCTP transmitter to be + * more conservative than the algorithms detailed in this document + * allow. However, an SCTP transmitter MUST NOT be more aggressive than + * the following algorithms allow. + */ +sctp_disposition_t sctp_sf_do_6_2_sack(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + sctp_add_cmd_sf(commands, SCTP_CMD_GEN_SACK, SCTP_FORCE()); + return SCTP_DISPOSITION_CONSUME; +} + +/* + * sctp_sf_t1_timer_expire + * + * Section: 4 Note: 2 + * Verification Tag: + * Inputs + * (endpoint, asoc) + * + * RFC 2960 Section 4 Notes + * 2) If the T1-init timer expires, the endpoint MUST retransmit INIT + * and re-start the T1-init timer without changing state. This MUST + * be repeated up to 'Max.Init.Retransmits' times. After that, the + * endpoint MUST abort the initialization process and report the + * error to SCTP user. + * + * 3) If the T1-cookie timer expires, the endpoint MUST retransmit + * COOKIE ECHO and re-start the T1-cookie timer without changing + * state. This MUST be repeated up to 'Max.Init.Retransmits' times. + * After that, the endpoint MUST abort the initialization process and + * report the error to SCTP user. + * + * Outputs + * (timers, events) + * + */ +sctp_disposition_t sctp_sf_t1_timer_expire(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *repl; + struct sctp_bind_addr *bp; + sctp_event_timeout_t timer = (sctp_event_timeout_t) arg; + int timeout; + int attempts; + + timeout = asoc->timeouts[timer]; + attempts = asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1; + repl = NULL; + + SCTP_DEBUG_PRINTK("Timer T1 expired.\n"); + + if (attempts < asoc->max_init_attempts) { + switch (timer) { + case SCTP_EVENT_TIMEOUT_T1_INIT: + bp = (struct sctp_bind_addr *) &asoc->base.bind_addr; + repl = sctp_make_init(asoc, bp, GFP_ATOMIC, 0); + break; + + case SCTP_EVENT_TIMEOUT_T1_COOKIE: + repl = sctp_make_cookie_echo(asoc, NULL); + break; + + default: + BUG(); + break; + }; + + if (!repl) + goto nomem; + + /* Issue a sideeffect to do the needed accounting. */ + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_RESTART, + SCTP_TO(timer)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(repl)); + } else { + sctp_add_cmd_sf(commands, SCTP_CMD_INIT_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + return SCTP_DISPOSITION_DELETE_TCB; + } + + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* RFC2960 9.2 If the timer expires, the endpoint must re-send the SHUTDOWN + * with the updated last sequential TSN received from its peer. + * + * An endpoint should limit the number of retransmissions of the + * SHUTDOWN chunk to the protocol parameter 'Association.Max.Retrans'. + * If this threshold is exceeded the endpoint should destroy the TCB and + * MUST report the peer endpoint unreachable to the upper layer (and + * thus the association enters the CLOSED state). The reception of any + * packet from its peer (i.e. as the peer sends all of its queued DATA + * chunks) should clear the endpoint's retransmission count and restart + * the T2-Shutdown timer, giving its peer ample opportunity to transmit + * all of its queued DATA chunks that have not yet been sent. + */ +sctp_disposition_t sctp_sf_t2_timer_expire(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *reply = NULL; + + SCTP_DEBUG_PRINTK("Timer T2 expired.\n"); + if (asoc->overall_error_count >= asoc->max_retrans) { + /* Note: CMD_ASSOC_FAILED calls CMD_DELETE_TCB. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_DELETE_TCB; + } + + switch (asoc->state) { + case SCTP_STATE_SHUTDOWN_SENT: + reply = sctp_make_shutdown(asoc, NULL); + break; + + case SCTP_STATE_SHUTDOWN_ACK_SENT: + reply = sctp_make_shutdown_ack(asoc, NULL); + break; + + default: + BUG(); + break; + }; + + if (!reply) + goto nomem; + + /* Do some failure management (Section 8.2). */ + sctp_add_cmd_sf(commands, SCTP_CMD_STRIKE, + SCTP_TRANSPORT(asoc->shutdown_last_sent_to)); + + /* Set the transport for the SHUTDOWN/ACK chunk and the timeout for + * the T2-shutdown timer. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_SETUP_T2, SCTP_CHUNK(reply)); + + /* Restart the T2-shutdown timer. */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T2_SHUTDOWN)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + return SCTP_DISPOSITION_CONSUME; + +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* + * ADDIP Section 4.1 ASCONF CHunk Procedures + * If the T4 RTO timer expires the endpoint should do B1 to B5 + */ +sctp_disposition_t sctp_sf_t4_timer_expire( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *chunk = asoc->addip_last_asconf; + struct sctp_transport *transport = chunk->transport; + + /* ADDIP 4.1 B1) Increment the error counters and perform path failure + * detection on the appropriate destination address as defined in + * RFC2960 [5] section 8.1 and 8.2. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_STRIKE, SCTP_TRANSPORT(transport)); + + /* Reconfig T4 timer and transport. */ + sctp_add_cmd_sf(commands, SCTP_CMD_SETUP_T4, SCTP_CHUNK(chunk)); + + /* ADDIP 4.1 B2) Increment the association error counters and perform + * endpoint failure detection on the association as defined in + * RFC2960 [5] section 8.1 and 8.2. + * association error counter is incremented in SCTP_CMD_STRIKE. + */ + if (asoc->overall_error_count >= asoc->max_retrans) { + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_STOP, + SCTP_TO(SCTP_EVENT_TIMEOUT_T4_RTO)); + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_INC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_ABORT; + } + + /* ADDIP 4.1 B3) Back-off the destination address RTO value to which + * the ASCONF chunk was sent by doubling the RTO timer value. + * This is done in SCTP_CMD_STRIKE. + */ + + /* ADDIP 4.1 B4) Re-transmit the ASCONF Chunk last sent and if possible + * choose an alternate destination address (please refer to RFC2960 + * [5] section 6.4.1). An endpoint MUST NOT add new parameters to this + * chunk, it MUST be the same (including its serial number) as the last + * ASCONF sent. + */ + sctp_chunk_hold(asoc->addip_last_asconf); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(asoc->addip_last_asconf)); + + /* ADDIP 4.1 B5) Restart the T-4 RTO timer. Note that if a different + * destination is selected, then the RTO used will be that of the new + * destination address. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, + SCTP_TO(SCTP_EVENT_TIMEOUT_T4_RTO)); + + return SCTP_DISPOSITION_CONSUME; +} + +/* sctpimpguide-05 Section 2.12.2 + * The sender of the SHUTDOWN MAY also start an overall guard timer + * 'T5-shutdown-guard' to bound the overall time for shutdown sequence. + * At the expiration of this timer the sender SHOULD abort the association + * by sending an ABORT chunk. + */ +sctp_disposition_t sctp_sf_t5_timer_expire(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + struct sctp_chunk *reply = NULL; + + SCTP_DEBUG_PRINTK("Timer T5 expired.\n"); + + reply = sctp_make_abort(asoc, NULL, 0); + if (!reply) + goto nomem; + + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(reply)); + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_NO_ERROR)); + + return SCTP_DISPOSITION_DELETE_TCB; +nomem: + return SCTP_DISPOSITION_NOMEM; +} + +/* Handle expiration of AUTOCLOSE timer. When the autoclose timer expires, + * the association is automatically closed by starting the shutdown process. + * The work that needs to be done is same as when SHUTDOWN is initiated by + * the user. So this routine looks same as sctp_sf_do_9_2_prm_shutdown(). + */ +sctp_disposition_t sctp_sf_autoclose_timer_expire( + const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + int disposition; + + /* From 9.2 Shutdown of an Association + * Upon receipt of the SHUTDOWN primitive from its upper + * layer, the endpoint enters SHUTDOWN-PENDING state and + * remains there until all outstanding data has been + * acknowledged by its peer. The endpoint accepts no new data + * from its upper layer, but retransmits data to the far end + * if necessary to fill gaps. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, + SCTP_STATE(SCTP_STATE_SHUTDOWN_PENDING)); + + /* sctpimpguide-05 Section 2.12.2 + * The sender of the SHUTDOWN MAY also start an overall guard timer + * 'T5-shutdown-guard' to bound the overall time for shutdown sequence. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + disposition = SCTP_DISPOSITION_CONSUME; + if (sctp_outq_is_empty(&asoc->outqueue)) { + disposition = sctp_sf_do_9_2_start_shutdown(ep, asoc, type, + arg, commands); + } + return disposition; +} + +/***************************************************************************** + * These are sa state functions which could apply to all types of events. + ****************************************************************************/ + +/* + * This table entry is not implemented. + * + * Inputs + * (endpoint, asoc, chunk) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_not_impl(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + return SCTP_DISPOSITION_NOT_IMPL; +} + +/* + * This table entry represents a bug. + * + * Inputs + * (endpoint, asoc, chunk) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_bug(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + return SCTP_DISPOSITION_BUG; +} + +/* + * This table entry represents the firing of a timer in the wrong state. + * Since timer deletion cannot be guaranteed a timer 'may' end up firing + * when the association is in the wrong state. This event should + * be ignored, so as to prevent any rearming of the timer. + * + * Inputs + * (endpoint, asoc, chunk) + * + * The return value is the disposition of the chunk. + */ +sctp_disposition_t sctp_sf_timer_ignore(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const sctp_subtype_t type, + void *arg, + sctp_cmd_seq_t *commands) +{ + SCTP_DEBUG_PRINTK("Timer %d ignored.\n", type.chunk); + return SCTP_DISPOSITION_CONSUME; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* Pull the SACK chunk based on the SACK header. */ +static struct sctp_sackhdr *sctp_sm_pull_sack(struct sctp_chunk *chunk) +{ + struct sctp_sackhdr *sack; + unsigned int len; + __u16 num_blocks; + __u16 num_dup_tsns; + + /* Protect ourselves from reading too far into + * the skb from a bogus sender. + */ + sack = (struct sctp_sackhdr *) chunk->skb->data; + + num_blocks = ntohs(sack->num_gap_ack_blocks); + num_dup_tsns = ntohs(sack->num_dup_tsns); + len = sizeof(struct sctp_sackhdr); + len += (num_blocks + num_dup_tsns) * sizeof(__u32); + if (len > chunk->skb->len) + return NULL; + + skb_pull(chunk->skb, len); + + return sack; +} + +/* Create an ABORT packet to be sent as a response, with the specified + * error causes. + */ +static struct sctp_packet *sctp_abort_pkt_new(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + struct sctp_chunk *chunk, + const void *payload, + size_t paylen) +{ + struct sctp_packet *packet; + struct sctp_chunk *abort; + + packet = sctp_ootb_pkt_new(asoc, chunk); + + if (packet) { + /* Make an ABORT. + * The T bit will be set if the asoc is NULL. + */ + abort = sctp_make_abort(asoc, chunk, paylen); + if (!abort) { + sctp_ootb_pkt_free(packet); + return NULL; + } + /* Add specified error causes, i.e., payload, to the + * end of the chunk. + */ + sctp_addto_chunk(abort, paylen, payload); + + /* Set the skb to the belonging sock for accounting. */ + abort->skb->sk = ep->base.sk; + + sctp_packet_append_chunk(packet, abort); + + } + + return packet; +} + +/* Allocate a packet for responding in the OOTB conditions. */ +static struct sctp_packet *sctp_ootb_pkt_new(const struct sctp_association *asoc, + const struct sctp_chunk *chunk) +{ + struct sctp_packet *packet; + struct sctp_transport *transport; + __u16 sport; + __u16 dport; + __u32 vtag; + + /* Get the source and destination port from the inbound packet. */ + sport = ntohs(chunk->sctp_hdr->dest); + dport = ntohs(chunk->sctp_hdr->source); + + /* The V-tag is going to be the same as the inbound packet if no + * association exists, otherwise, use the peer's vtag. + */ + if (asoc) { + vtag = asoc->peer.i.init_tag; + } else { + /* Special case the INIT and stale COOKIE_ECHO as there is no + * vtag yet. + */ + switch(chunk->chunk_hdr->type) { + case SCTP_CID_INIT: + { + sctp_init_chunk_t *init; + + init = (sctp_init_chunk_t *)chunk->chunk_hdr; + vtag = ntohl(init->init_hdr.init_tag); + break; + } + default: + vtag = ntohl(chunk->sctp_hdr->vtag); + break; + } + } + + /* Make a transport for the bucket, Eliza... */ + transport = sctp_transport_new(sctp_source(chunk), GFP_ATOMIC); + if (!transport) + goto nomem; + + /* Cache a route for the transport with the chunk's destination as + * the source address. + */ + sctp_transport_route(transport, (union sctp_addr *)&chunk->dest, + sctp_sk(sctp_get_ctl_sock())); + + packet = sctp_packet_init(&transport->packet, transport, sport, dport); + packet = sctp_packet_config(packet, vtag, 0); + + return packet; + +nomem: + return NULL; +} + +/* Free the packet allocated earlier for responding in the OOTB condition. */ +void sctp_ootb_pkt_free(struct sctp_packet *packet) +{ + sctp_transport_free(packet->transport); +} + +/* Send a stale cookie error when a invalid COOKIE ECHO chunk is found */ +static void sctp_send_stale_cookie_err(const struct sctp_endpoint *ep, + const struct sctp_association *asoc, + const struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands, + struct sctp_chunk *err_chunk) +{ + struct sctp_packet *packet; + + if (err_chunk) { + packet = sctp_ootb_pkt_new(asoc, chunk); + if (packet) { + struct sctp_signed_cookie *cookie; + + /* Override the OOTB vtag from the cookie. */ + cookie = chunk->subh.cookie_hdr; + packet->vtag = cookie->c.peer_vtag; + + /* Set the skb to the belonging sock for accounting. */ + err_chunk->skb->sk = ep->base.sk; + sctp_packet_append_chunk(packet, err_chunk); + sctp_add_cmd_sf(commands, SCTP_CMD_SEND_PKT, + SCTP_PACKET(packet)); + SCTP_INC_STATS(SCTP_MIB_OUTCTRLCHUNKS); + } else + sctp_chunk_free (err_chunk); + } +} + + +/* Process a data chunk */ +static int sctp_eat_data(const struct sctp_association *asoc, + struct sctp_chunk *chunk, + sctp_cmd_seq_t *commands) +{ + sctp_datahdr_t *data_hdr; + struct sctp_chunk *err; + size_t datalen; + sctp_verb_t deliver; + int tmp; + __u32 tsn; + + data_hdr = chunk->subh.data_hdr = (sctp_datahdr_t *)chunk->skb->data; + skb_pull(chunk->skb, sizeof(sctp_datahdr_t)); + + tsn = ntohl(data_hdr->tsn); + SCTP_DEBUG_PRINTK("eat_data: TSN 0x%x.\n", tsn); + + /* ASSERT: Now skb->data is really the user data. */ + + /* Process ECN based congestion. + * + * Since the chunk structure is reused for all chunks within + * a packet, we use ecn_ce_done to track if we've already + * done CE processing for this packet. + * + * We need to do ECN processing even if we plan to discard the + * chunk later. + */ + + if (!chunk->ecn_ce_done) { + struct sctp_af *af; + chunk->ecn_ce_done = 1; + + af = sctp_get_af_specific( + ipver2af(chunk->skb->nh.iph->version)); + + if (af && af->is_ce(chunk->skb) && asoc->peer.ecn_capable) { + /* Do real work as sideffect. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ECN_CE, + SCTP_U32(tsn)); + } + } + + tmp = sctp_tsnmap_check(&asoc->peer.tsn_map, tsn); + if (tmp < 0) { + /* The TSN is too high--silently discard the chunk and + * count on it getting retransmitted later. + */ + return SCTP_IERROR_HIGH_TSN; + } else if (tmp > 0) { + /* This is a duplicate. Record it. */ + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_DUP, SCTP_U32(tsn)); + return SCTP_IERROR_DUP_TSN; + } + + /* This is a new TSN. */ + + /* Discard if there is no room in the receive window. + * Actually, allow a little bit of overflow (up to a MTU). + */ + datalen = ntohs(chunk->chunk_hdr->length); + datalen -= sizeof(sctp_data_chunk_t); + + deliver = SCTP_CMD_CHUNK_ULP; + + /* Think about partial delivery. */ + if ((datalen >= asoc->rwnd) && (!asoc->ulpq.pd_mode)) { + + /* Even if we don't accept this chunk there is + * memory pressure. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_PART_DELIVER, SCTP_NULL()); + } + + /* Spill over rwnd a little bit. Note: While allowed, this spill over + * seems a bit troublesome in that frag_point varies based on + * PMTU. In cases, such as loopback, this might be a rather + * large spill over. + */ + if (!asoc->rwnd || asoc->rwnd_over || + (datalen > asoc->rwnd + asoc->frag_point)) { + + /* If this is the next TSN, consider reneging to make + * room. Note: Playing nice with a confused sender. A + * malicious sender can still eat up all our buffer + * space and in the future we may want to detect and + * do more drastic reneging. + */ + if (sctp_tsnmap_has_gap(&asoc->peer.tsn_map) && + (sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map) + 1) == tsn) { + SCTP_DEBUG_PRINTK("Reneging for tsn:%u\n", tsn); + deliver = SCTP_CMD_RENEGE; + } else { + SCTP_DEBUG_PRINTK("Discard tsn: %u len: %Zd, " + "rwnd: %d\n", tsn, datalen, + asoc->rwnd); + return SCTP_IERROR_IGNORE_TSN; + } + } + + /* + * Section 3.3.10.9 No User Data (9) + * + * Cause of error + * --------------- + * No User Data: This error cause is returned to the originator of a + * DATA chunk if a received DATA chunk has no user data. + */ + if (unlikely(0 == datalen)) { + err = sctp_make_abort_no_data(asoc, chunk, tsn); + if (err) { + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(err)); + } + /* We are going to ABORT, so we might as well stop + * processing the rest of the chunks in the packet. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_DISCARD_PACKET,SCTP_NULL()); + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_U32(SCTP_ERROR_NO_DATA)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_IERROR_NO_DATA; + } + + /* If definately accepting the DATA chunk, record its TSN, otherwise + * wait for renege processing. + */ + if (SCTP_CMD_CHUNK_ULP == deliver) + sctp_add_cmd_sf(commands, SCTP_CMD_REPORT_TSN, SCTP_U32(tsn)); + + /* Note: Some chunks may get overcounted (if we drop) or overcounted + * if we renege and the chunk arrives again. + */ + if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) + SCTP_INC_STATS(SCTP_MIB_INUNORDERCHUNKS); + else + SCTP_INC_STATS(SCTP_MIB_INORDERCHUNKS); + + /* RFC 2960 6.5 Stream Identifier and Stream Sequence Number + * + * If an endpoint receive a DATA chunk with an invalid stream + * identifier, it shall acknowledge the reception of the DATA chunk + * following the normal procedure, immediately send an ERROR chunk + * with cause set to "Invalid Stream Identifier" (See Section 3.3.10) + * and discard the DATA chunk. + */ + if (ntohs(data_hdr->stream) >= asoc->c.sinit_max_instreams) { + err = sctp_make_op_error(asoc, chunk, SCTP_ERROR_INV_STRM, + &data_hdr->stream, + sizeof(data_hdr->stream)); + if (err) + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, + SCTP_CHUNK(err)); + return SCTP_IERROR_BAD_STREAM; + } + + /* Send the data up to the user. Note: Schedule the + * SCTP_CMD_CHUNK_ULP cmd before the SCTP_CMD_GEN_SACK, as the SACK + * chunk needs the updated rwnd. + */ + sctp_add_cmd_sf(commands, deliver, SCTP_CHUNK(chunk)); + + return SCTP_IERROR_NO_ERROR; +} diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c new file mode 100644 index 000000000000..8967846f69e8 --- /dev/null +++ b/net/sctp/sm_statetable.c @@ -0,0 +1,1004 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * Copyright (c) 2001 Nokia, Inc. + * + * This file is part of the SCTP kernel reference Implementation + * + * These are the state tables for the SCTP state machine. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@us.ibm.com> + * Hui Huang <hui.huang@nokia.com> + * Daisy Chang <daisyc@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/skbuff.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +static const sctp_sm_table_entry_t +primitive_event_table[SCTP_NUM_PRIMITIVE_TYPES][SCTP_STATE_NUM_STATES]; +static const sctp_sm_table_entry_t +other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_STATE_NUM_STATES]; +static const sctp_sm_table_entry_t +timeout_event_table[SCTP_NUM_TIMEOUT_TYPES][SCTP_STATE_NUM_STATES]; + +static const sctp_sm_table_entry_t *sctp_chunk_event_lookup(sctp_cid_t cid, + sctp_state_t state); + + +static const sctp_sm_table_entry_t bug = { + .fn = sctp_sf_bug, + .name = "sctp_sf_bug" +}; + +#define DO_LOOKUP(_max, _type, _table) \ + if ((event_subtype._type > (_max))) { \ + printk(KERN_WARNING \ + "sctp table %p possible attack:" \ + " event %d exceeds max %d\n", \ + _table, event_subtype._type, _max); \ + return &bug; \ + } \ + return &_table[event_subtype._type][(int)state]; + +const sctp_sm_table_entry_t *sctp_sm_lookup_event(sctp_event_t event_type, + sctp_state_t state, + sctp_subtype_t event_subtype) +{ + switch (event_type) { + case SCTP_EVENT_T_CHUNK: + return sctp_chunk_event_lookup(event_subtype.chunk, state); + break; + case SCTP_EVENT_T_TIMEOUT: + DO_LOOKUP(SCTP_EVENT_TIMEOUT_MAX, timeout, + timeout_event_table); + break; + + case SCTP_EVENT_T_OTHER: + DO_LOOKUP(SCTP_EVENT_OTHER_MAX, other, other_event_table); + break; + + case SCTP_EVENT_T_PRIMITIVE: + DO_LOOKUP(SCTP_EVENT_PRIMITIVE_MAX, primitive, + primitive_event_table); + break; + + default: + /* Yikes! We got an illegal event type. */ + return &bug; + }; +} + +#define TYPE_SCTP_DATA { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_eat_data_6_2, .name = "sctp_sf_eat_data_6_2"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_eat_data_6_2, .name = "sctp_sf_eat_data_6_2"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_eat_data_fast_4_4, .name = "sctp_sf_eat_data_fast_4_4"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_DATA */ + +#define TYPE_SCTP_INIT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_do_5_1B_init, .name = "sctp_sf_do_5_1B_init"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_do_5_2_1_siminit, .name = "sctp_sf_do_5_2_1_siminit"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_5_2_1_siminit, .name = "sctp_sf_do_5_2_1_siminit"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_5_2_2_dupinit, .name = "sctp_sf_do_5_2_2_dupinit"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_5_2_2_dupinit, .name = "sctp_sf_do_5_2_2_dupinit"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_5_2_2_dupinit, .name = "sctp_sf_do_5_2_2_dupinit"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_5_2_2_dupinit, .name = "sctp_sf_do_5_2_2_dupinit"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_do_9_2_reshutack, .name = "sctp_sf_do_9_2_reshutack"}, \ +} /* TYPE_SCTP_INIT */ + +#define TYPE_SCTP_INIT_ACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_do_5_1C_ack, .name = "sctp_sf_do_5_1C_ack"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_INIT_ACK */ + +#define TYPE_SCTP_SACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_eat_sack_6_2, .name = "sctp_sf_eat_sack_6_2"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_eat_sack_6_2, .name = "sctp_sf_eat_sack_6_2"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_eat_sack_6_2, .name = "sctp_sf_eat_sack_6_2"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_eat_sack_6_2, .name = "sctp_sf_eat_sack_6_2"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_SACK */ + +#define TYPE_SCTP_HEARTBEAT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_beat_8_3, .name = "sctp_sf_beat_8_3"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_beat_8_3, .name = "sctp_sf_beat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_beat_8_3, .name = "sctp_sf_beat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_beat_8_3, .name = "sctp_sf_beat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_beat_8_3, .name = "sctp_sf_beat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + /* This should not happen, but we are nice. */ \ + {.fn = sctp_sf_beat_8_3, .name = "sctp_sf_beat_8_3"}, \ +} /* TYPE_SCTP_HEARTBEAT */ + +#define TYPE_SCTP_HEARTBEAT_ACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_violation, .name = "sctp_sf_violation"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_backbeat_8_3, .name = "sctp_sf_backbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_backbeat_8_3, .name = "sctp_sf_backbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_backbeat_8_3, .name = "sctp_sf_backbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_backbeat_8_3, .name = "sctp_sf_backbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_HEARTBEAT_ACK */ + +#define TYPE_SCTP_ABORT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_pdiscard, .name = "sctp_sf_pdiscard"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_cookie_wait_abort, .name = "sctp_sf_cookie_wait_abort"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_cookie_echoed_abort, \ + .name = "sctp_sf_cookie_echoed_abort"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_9_1_abort, .name = "sctp_sf_do_9_1_abort"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_shutdown_pending_abort, \ + .name = "sctp_sf_shutdown_pending_abort"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_shutdown_sent_abort, \ + .name = "sctp_sf_shutdown_sent_abort"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_9_1_abort, .name = "sctp_sf_do_9_1_abort"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_shutdown_ack_sent_abort, \ + .name = "sctp_sf_shutdown_ack_sent_abort"}, \ +} /* TYPE_SCTP_ABORT */ + +#define TYPE_SCTP_SHUTDOWN { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_9_2_shutdown, .name = "sctp_sf_do_9_2_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_9_2_shutdown_ack, \ + .name = "sctp_sf_do_9_2_shutdown_ack"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_SHUTDOWN */ + +#define TYPE_SCTP_SHUTDOWN_ACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_do_8_5_1_E_sa, .name = "sctp_sf_do_8_5_1_E_sa"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_8_5_1_E_sa, .name = "sctp_sf_do_8_5_1_E_sa"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_violation, .name = "sctp_sf_violation"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_violation, .name = "sctp_sf_violation"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_9_2_final, .name = "sctp_sf_do_9_2_final"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_violation, .name = "sctp_sf_violation"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_do_9_2_final, .name = "sctp_sf_do_9_2_final"}, \ +} /* TYPE_SCTP_SHUTDOWN_ACK */ + +#define TYPE_SCTP_ERROR { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_cookie_echoed_err, .name = "sctp_sf_cookie_echoed_err"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_operr_notify, .name = "sctp_sf_operr_notify"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_operr_notify, .name = "sctp_sf_operr_notify"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_operr_notify, .name = "sctp_sf_operr_notify"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_ERROR */ + +#define TYPE_SCTP_COOKIE_ECHO { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_do_5_1D_ce, .name = "sctp_sf_do_5_1D_ce"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_do_5_2_4_dupcook, .name = "sctp_sf_do_5_2_4_dupcook"}, \ +} /* TYPE_SCTP_COOKIE_ECHO */ + +#define TYPE_SCTP_COOKIE_ACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_5_1E_ca, .name = "sctp_sf_do_5_1E_ca"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_COOKIE_ACK */ + +#define TYPE_SCTP_ECN_ECNE { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_ecne, .name = "sctp_sf_do_ecne"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ +} /* TYPE_SCTP_ECN_ECNE */ + +#define TYPE_SCTP_ECN_CWR { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_ecn_cwr, .name = "sctp_sf_do_ecn_cwr"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_ecn_cwr, .name = "sctp_sf_do_ecn_cwr"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_ecn_cwr, .name = "sctp_sf_do_ecn_cwr"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ +} /* TYPE_SCTP_ECN_CWR */ + +#define TYPE_SCTP_SHUTDOWN_COMPLETE { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_do_4_C, .name = "sctp_sf_do_4_C"}, \ +} /* TYPE_SCTP_SHUTDOWN_COMPLETE */ + +/* The primary index for this table is the chunk type. + * The secondary index for this table is the state. + * + * For base protocol (RFC 2960). + */ +static const sctp_sm_table_entry_t chunk_event_table[SCTP_NUM_BASE_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { + TYPE_SCTP_DATA, + TYPE_SCTP_INIT, + TYPE_SCTP_INIT_ACK, + TYPE_SCTP_SACK, + TYPE_SCTP_HEARTBEAT, + TYPE_SCTP_HEARTBEAT_ACK, + TYPE_SCTP_ABORT, + TYPE_SCTP_SHUTDOWN, + TYPE_SCTP_SHUTDOWN_ACK, + TYPE_SCTP_ERROR, + TYPE_SCTP_COOKIE_ECHO, + TYPE_SCTP_COOKIE_ACK, + TYPE_SCTP_ECN_ECNE, + TYPE_SCTP_ECN_CWR, + TYPE_SCTP_SHUTDOWN_COMPLETE, +}; /* state_fn_t chunk_event_table[][] */ + +#define TYPE_SCTP_ASCONF { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_asconf, .name = "sctp_sf_do_asconf"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_ASCONF */ + +#define TYPE_SCTP_ASCONF_ACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_asconf_ack, .name = "sctp_sf_do_asconf_ack"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_ASCONF_ACK */ + +/* The primary index for this table is the chunk type. + * The secondary index for this table is the state. + */ +static const sctp_sm_table_entry_t addip_chunk_event_table[SCTP_NUM_ADDIP_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { + TYPE_SCTP_ASCONF, + TYPE_SCTP_ASCONF_ACK, +}; /*state_fn_t addip_chunk_event_table[][] */ + +#define TYPE_SCTP_FWD_TSN { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_eat_fwd_tsn, .name = "sctp_sf_eat_fwd_tsn"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_eat_fwd_tsn, .name = "sctp_sf_eat_fwd_tsn"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_eat_fwd_tsn_fast, .name = "sctp_sf_eat_fwd_tsn_fast"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_discard_chunk, .name = "sctp_sf_discard_chunk"}, \ +} /* TYPE_SCTP_FWD_TSN */ + +/* The primary index for this table is the chunk type. + * The secondary index for this table is the state. + */ +static const sctp_sm_table_entry_t prsctp_chunk_event_table[SCTP_NUM_PRSCTP_CHUNK_TYPES][SCTP_STATE_NUM_STATES] = { + TYPE_SCTP_FWD_TSN, +}; /*state_fn_t prsctp_chunk_event_table[][] */ + +static const sctp_sm_table_entry_t +chunk_event_table_unknown[SCTP_STATE_NUM_STATES] = { + /* SCTP_STATE_EMPTY */ + {.fn = sctp_sf_ootb, .name = "sctp_sf_ootb"}, + /* SCTP_STATE_CLOSED */ + {.fn = sctp_sf_tabort_8_4_8, .name = "sctp_sf_tabort_8_4_8"}, + /* SCTP_STATE_COOKIE_WAIT */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, + /* SCTP_STATE_COOKIE_ECHOED */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, + /* SCTP_STATE_ESTABLISHED */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, + /* SCTP_STATE_SHUTDOWN_PENDING */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, + /* SCTP_STATE_SHUTDOWN_SENT */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, + /* SCTP_STATE_SHUTDOWN_RECEIVED */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ + {.fn = sctp_sf_unk_chunk, .name = "sctp_sf_unk_chunk"}, +}; /* chunk unknown */ + + +#define TYPE_SCTP_PRIMITIVE_ASSOCIATE { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_do_prm_asoc, .name = "sctp_sf_do_prm_asoc"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_not_impl, .name = "sctp_sf_not_impl"}, \ +} /* TYPE_SCTP_PRIMITIVE_ASSOCIATE */ + +#define TYPE_SCTP_PRIMITIVE_SHUTDOWN { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_cookie_wait_prm_shutdown, \ + .name = "sctp_sf_cookie_wait_prm_shutdown"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_cookie_echoed_prm_shutdown, \ + .name = "sctp_sf_cookie_echoed_prm_shutdown"},\ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_9_2_prm_shutdown, \ + .name = "sctp_sf_do_9_2_prm_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_ignore_primitive, .name = "sctp_sf_ignore_primitive"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_ignore_primitive, .name = "sctp_sf_ignore_primitive"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_ignore_primitive, .name = "sctp_sf_ignore_primitive"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_ignore_primitive, .name = "sctp_sf_ignore_primitive"}, \ +} /* TYPE_SCTP_PRIMITIVE_SHUTDOWN */ + +#define TYPE_SCTP_PRIMITIVE_ABORT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_cookie_wait_prm_abort, \ + .name = "sctp_sf_cookie_wait_prm_abort"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_cookie_echoed_prm_abort, \ + .name = "sctp_sf_cookie_echoed_prm_abort"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_9_1_prm_abort, \ + .name = "sctp_sf_do_9_1_prm_abort"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_shutdown_pending_prm_abort, \ + .name = "sctp_sf_shutdown_pending_prm_abort"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_shutdown_sent_prm_abort, \ + .name = "sctp_sf_shutdown_sent_prm_abort"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_9_1_prm_abort, \ + .name = "sctp_sf_do_9_1_prm_abort"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_shutdown_ack_sent_prm_abort, \ + .name = "sctp_sf_shutdown_ack_sent_prm_abort"}, \ +} /* TYPE_SCTP_PRIMITIVE_ABORT */ + +#define TYPE_SCTP_PRIMITIVE_SEND { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_do_prm_send, .name = "sctp_sf_do_prm_send"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_prm_send, .name = "sctp_sf_do_prm_send"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_prm_send, .name = "sctp_sf_do_prm_send"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ +} /* TYPE_SCTP_PRIMITIVE_SEND */ + +#define TYPE_SCTP_PRIMITIVE_REQUESTHEARTBEAT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_do_prm_requestheartbeat, \ + .name = "sctp_sf_do_prm_requestheartbeat"}, \ +} /* TYPE_SCTP_PRIMITIVE_REQUESTHEARTBEAT */ + +#define TYPE_SCTP_PRIMITIVE_ASCONF { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_error_closed, .name = "sctp_sf_error_closed"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_prm_asconf, .name = "sctp_sf_do_prm_asconf"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_error_shutdown, .name = "sctp_sf_error_shutdown"}, \ +} /* TYPE_SCTP_PRIMITIVE_REQUESTHEARTBEAT */ + +/* The primary index for this table is the primitive type. + * The secondary index for this table is the state. + */ +static const sctp_sm_table_entry_t primitive_event_table[SCTP_NUM_PRIMITIVE_TYPES][SCTP_STATE_NUM_STATES] = { + TYPE_SCTP_PRIMITIVE_ASSOCIATE, + TYPE_SCTP_PRIMITIVE_SHUTDOWN, + TYPE_SCTP_PRIMITIVE_ABORT, + TYPE_SCTP_PRIMITIVE_SEND, + TYPE_SCTP_PRIMITIVE_REQUESTHEARTBEAT, + TYPE_SCTP_PRIMITIVE_ASCONF, +}; + +#define TYPE_SCTP_OTHER_NO_PENDING_TSN { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_9_2_start_shutdown, \ + .name = "sctp_do_9_2_start_shutdown"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_9_2_shutdown_ack, \ + .name = "sctp_sf_do_9_2_shutdown_ack"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ +} + +#define TYPE_SCTP_OTHER_ICMP_PROTO_UNREACH { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_cookie_wait_icmp_abort, \ + .name = "sctp_sf_cookie_wait_icmp_abort"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_ignore_other, .name = "sctp_sf_ignore_other"}, \ +} + +static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_STATE_NUM_STATES] = { + TYPE_SCTP_OTHER_NO_PENDING_TSN, + TYPE_SCTP_OTHER_ICMP_PROTO_UNREACH, +}; + +#define TYPE_SCTP_EVENT_TIMEOUT_NONE { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_T1_COOKIE { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_T1_INIT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_t1_timer_expire, .name = "sctp_sf_t1_timer_expire"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_T2_SHUTDOWN { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_t2_timer_expire, .name = "sctp_sf_t2_timer_expire"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_t2_timer_expire, .name = "sctp_sf_t2_timer_expire"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_T3_RTX { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_do_6_3_3_rtx, .name = "sctp_sf_do_6_3_3_rtx"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_6_3_3_rtx, .name = "sctp_sf_do_6_3_3_rtx"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_6_3_3_rtx, .name = "sctp_sf_do_6_3_3_rtx"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_do_6_3_3_rtx, .name = "sctp_sf_do_6_3_3_rtx"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_T4_RTO { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_t4_timer_expire, .name = "sctp_sf_t4_timer_expire"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_t5_timer_expire, .name = "sctp_sf_t5_timer_expire"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_t5_timer_expire, .name = "sctp_sf_t5_timer_expire"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_HEARTBEAT { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_sendbeat_8_3, .name = "sctp_sf_sendbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_sendbeat_8_3, .name = "sctp_sf_sendbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_sendbeat_8_3, .name = "sctp_sf_sendbeat_8_3"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_SACK { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_bug, .name = "sctp_sf_bug"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_do_6_2_sack, .name = "sctp_sf_do_6_2_sack"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_do_6_2_sack, .name = "sctp_sf_do_6_2_sack"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_do_6_2_sack, .name = "sctp_sf_do_6_2_sack"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +#define TYPE_SCTP_EVENT_TIMEOUT_AUTOCLOSE { \ + /* SCTP_STATE_EMPTY */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_CLOSED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_WAIT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_COOKIE_ECHOED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_ESTABLISHED */ \ + {.fn = sctp_sf_autoclose_timer_expire, \ + .name = "sctp_sf_autoclose_timer_expire"}, \ + /* SCTP_STATE_SHUTDOWN_PENDING */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ + /* SCTP_STATE_SHUTDOWN_ACK_SENT */ \ + {.fn = sctp_sf_timer_ignore, .name = "sctp_sf_timer_ignore"}, \ +} + +static const sctp_sm_table_entry_t timeout_event_table[SCTP_NUM_TIMEOUT_TYPES][SCTP_STATE_NUM_STATES] = { + TYPE_SCTP_EVENT_TIMEOUT_NONE, + TYPE_SCTP_EVENT_TIMEOUT_T1_COOKIE, + TYPE_SCTP_EVENT_TIMEOUT_T1_INIT, + TYPE_SCTP_EVENT_TIMEOUT_T2_SHUTDOWN, + TYPE_SCTP_EVENT_TIMEOUT_T3_RTX, + TYPE_SCTP_EVENT_TIMEOUT_T4_RTO, + TYPE_SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD, + TYPE_SCTP_EVENT_TIMEOUT_HEARTBEAT, + TYPE_SCTP_EVENT_TIMEOUT_SACK, + TYPE_SCTP_EVENT_TIMEOUT_AUTOCLOSE, +}; + +static const sctp_sm_table_entry_t *sctp_chunk_event_lookup(sctp_cid_t cid, + sctp_state_t state) +{ + if (state > SCTP_STATE_MAX) + return &bug; + + if (cid >= 0 && cid <= SCTP_CID_BASE_MAX) + return &chunk_event_table[cid][state]; + + if (sctp_prsctp_enable) { + if (cid == SCTP_CID_FWD_TSN) + return &prsctp_chunk_event_table[0][state]; + } + + if (sctp_addip_enable) { + if (cid == SCTP_CID_ASCONF) + return &addip_chunk_event_table[0][state]; + + if (cid == SCTP_CID_ASCONF_ACK) + return &addip_chunk_event_table[1][state]; + } + + return &chunk_event_table_unknown[state]; +} diff --git a/net/sctp/socket.c b/net/sctp/socket.c new file mode 100644 index 000000000000..e8c210182571 --- /dev/null +++ b/net/sctp/socket.c @@ -0,0 +1,4797 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001-2003 Intel Corp. + * Copyright (c) 2001-2002 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions interface with the sockets layer to implement the + * SCTP Extensions for the Sockets API. + * + * Note that the descriptions from the specification are USER level + * functions--this file is the functions which populate the struct proto + * for SCTP which is the BOTTOM of the sockets interface. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Narasimha Budihal <narsi@refcode.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@us.ibm.com> + * Xingang Guo <xingang.guo@intel.com> + * Daisy Chang <daisyc@us.ibm.com> + * Sridhar Samudrala <samudrala@us.ibm.com> + * Inaky Perez-Gonzalez <inaky.gonzalez@intel.com> + * Ardelle Fan <ardelle.fan@intel.com> + * Ryan Layer <rmlayer@us.ibm.com> + * Anup Pemmaiah <pemmaiah@cc.usu.edu> + * Kevin Gao <kevin.gao@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * 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/fcntl.h> +#include <linux/poll.h> +#include <linux/init.h> +#include <linux/crypto.h> + +#include <net/ip.h> +#include <net/icmp.h> +#include <net/route.h> +#include <net/ipv6.h> +#include <net/inet_common.h> + +#include <linux/socket.h> /* for sa_family_t */ +#include <net/sock.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* WARNING: Please do not remove the SCTP_STATIC attribute to + * any of the functions below as they are used to export functions + * used by a project regression testsuite. + */ + +/* Forward declarations for internal helper functions. */ +static int sctp_writeable(struct sock *sk); +static void sctp_wfree(struct sk_buff *skb); +static int sctp_wait_for_sndbuf(struct sctp_association *, long *timeo_p, + size_t msg_len); +static int sctp_wait_for_packet(struct sock * sk, int *err, long *timeo_p); +static int sctp_wait_for_connect(struct sctp_association *, long *timeo_p); +static int sctp_wait_for_accept(struct sock *sk, long timeo); +static void sctp_wait_for_close(struct sock *sk, long timeo); +static struct sctp_af *sctp_sockaddr_af(struct sctp_sock *opt, + union sctp_addr *addr, int len); +static int sctp_bindx_add(struct sock *, struct sockaddr *, int); +static int sctp_bindx_rem(struct sock *, struct sockaddr *, int); +static int sctp_send_asconf_add_ip(struct sock *, struct sockaddr *, int); +static int sctp_send_asconf_del_ip(struct sock *, struct sockaddr *, int); +static int sctp_send_asconf(struct sctp_association *asoc, + struct sctp_chunk *chunk); +static int sctp_do_bind(struct sock *, union sctp_addr *, int); +static int sctp_autobind(struct sock *sk); +static void sctp_sock_migrate(struct sock *, struct sock *, + struct sctp_association *, sctp_socket_type_t); +static char *sctp_hmac_alg = SCTP_COOKIE_HMAC_ALG; + +extern kmem_cache_t *sctp_bucket_cachep; + +/* Get the sndbuf space available at the time on the association. */ +static inline int sctp_wspace(struct sctp_association *asoc) +{ + struct sock *sk = asoc->base.sk; + int amt = 0; + + amt = sk->sk_sndbuf - asoc->sndbuf_used; + if (amt < 0) + amt = 0; + return amt; +} + +/* Increment the used sndbuf space count of the corresponding association by + * the size of the outgoing data chunk. + * Also, set the skb destructor for sndbuf accounting later. + * + * Since it is always 1-1 between chunk and skb, and also a new skb is always + * allocated for chunk bundling in sctp_packet_transmit(), we can use the + * destructor in the data chunk skb for the purpose of the sndbuf space + * tracking. + */ +static inline void sctp_set_owner_w(struct sctp_chunk *chunk) +{ + struct sctp_association *asoc = chunk->asoc; + struct sock *sk = asoc->base.sk; + + /* The sndbuf space is tracked per association. */ + sctp_association_hold(asoc); + + chunk->skb->destructor = sctp_wfree; + /* Save the chunk pointer in skb for sctp_wfree to use later. */ + *((struct sctp_chunk **)(chunk->skb->cb)) = chunk; + + asoc->sndbuf_used += SCTP_DATA_SNDSIZE(chunk); + sk->sk_wmem_queued += SCTP_DATA_SNDSIZE(chunk); +} + +/* Verify that this is a valid address. */ +static inline int sctp_verify_addr(struct sock *sk, union sctp_addr *addr, + int len) +{ + struct sctp_af *af; + + /* Verify basic sockaddr. */ + af = sctp_sockaddr_af(sctp_sk(sk), addr, len); + if (!af) + return -EINVAL; + + /* Is this a valid SCTP address? */ + if (!af->addr_valid(addr, sctp_sk(sk))) + return -EINVAL; + + if (!sctp_sk(sk)->pf->send_verify(sctp_sk(sk), (addr))) + return -EINVAL; + + return 0; +} + +/* Look up the association by its id. If this is not a UDP-style + * socket, the ID field is always ignored. + */ +struct sctp_association *sctp_id2assoc(struct sock *sk, sctp_assoc_t id) +{ + struct sctp_association *asoc = NULL; + + /* If this is not a UDP-style socket, assoc id should be ignored. */ + if (!sctp_style(sk, UDP)) { + /* Return NULL if the socket state is not ESTABLISHED. It + * could be a TCP-style listening socket or a socket which + * hasn't yet called connect() to establish an association. + */ + if (!sctp_sstate(sk, ESTABLISHED)) + return NULL; + + /* Get the first and the only association from the list. */ + if (!list_empty(&sctp_sk(sk)->ep->asocs)) + asoc = list_entry(sctp_sk(sk)->ep->asocs.next, + struct sctp_association, asocs); + return asoc; + } + + /* Otherwise this is a UDP-style socket. */ + if (!id || (id == (sctp_assoc_t)-1)) + return NULL; + + spin_lock_bh(&sctp_assocs_id_lock); + asoc = (struct sctp_association *)idr_find(&sctp_assocs_id, (int)id); + spin_unlock_bh(&sctp_assocs_id_lock); + + if (!asoc || (asoc->base.sk != sk) || asoc->base.dead) + return NULL; + + return asoc; +} + +/* Look up the transport from an address and an assoc id. If both address and + * id are specified, the associations matching the address and the id should be + * the same. + */ +static struct sctp_transport *sctp_addr_id2transport(struct sock *sk, + struct sockaddr_storage *addr, + sctp_assoc_t id) +{ + struct sctp_association *addr_asoc = NULL, *id_asoc = NULL; + 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, + &transport); + laddr->v4.sin_port = htons(laddr->v4.sin_port); + + if (!addr_asoc) + return NULL; + + id_asoc = sctp_id2assoc(sk, id); + if (id_asoc && (id_asoc != addr_asoc)) + return NULL; + + sctp_get_pf_specific(sk->sk_family)->addr_v4map(sctp_sk(sk), + (union sctp_addr *)addr); + + return transport; +} + +/* API 3.1.2 bind() - UDP Style Syntax + * The syntax of bind() is, + * + * ret = bind(int sd, struct sockaddr *addr, int addrlen); + * + * sd - the socket descriptor returned by socket(). + * addr - the address structure (struct sockaddr_in or struct + * sockaddr_in6 [RFC 2553]), + * addr_len - the size of the address structure. + */ +SCTP_STATIC int sctp_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +{ + int retval = 0; + + sctp_lock_sock(sk); + + SCTP_DEBUG_PRINTK("sctp_bind(sk: %p, uaddr: %p, addr_len: %d)\n", + sk, uaddr, addr_len); + + /* Disallow binding twice. */ + if (!sctp_sk(sk)->ep->base.bind_addr.port) + retval = sctp_do_bind(sk, (union sctp_addr *)uaddr, + addr_len); + else + retval = -EINVAL; + + sctp_release_sock(sk); + + return retval; +} + +static long sctp_get_port_local(struct sock *, union sctp_addr *); + +/* Verify this is a valid sockaddr. */ +static struct sctp_af *sctp_sockaddr_af(struct sctp_sock *opt, + union sctp_addr *addr, int len) +{ + struct sctp_af *af; + + /* Check minimum size. */ + if (len < sizeof (struct sockaddr)) + return NULL; + + /* Does this PF support this AF? */ + if (!opt->pf->af_supported(addr->sa.sa_family, opt)) + return NULL; + + /* If we get this far, af is valid. */ + af = sctp_get_af_specific(addr->sa.sa_family); + + if (len < af->sockaddr_len) + return NULL; + + return af; +} + +/* Bind a local address either to an endpoint or to an association. */ +SCTP_STATIC int sctp_do_bind(struct sock *sk, union sctp_addr *addr, int len) +{ + struct sctp_sock *sp = sctp_sk(sk); + struct sctp_endpoint *ep = sp->ep; + struct sctp_bind_addr *bp = &ep->base.bind_addr; + struct sctp_af *af; + unsigned short snum; + int ret = 0; + + SCTP_DEBUG_PRINTK("sctp_do_bind(sk: %p, newaddr: %p, len: %d)\n", + sk, addr, len); + + /* Common sockaddr verification. */ + af = sctp_sockaddr_af(sp, addr, len); + if (!af) + return -EINVAL; + + /* PF specific bind() address verification. */ + if (!sp->pf->bind_verify(sp, addr)) + return -EADDRNOTAVAIL; + + snum= ntohs(addr->v4.sin_port); + + SCTP_DEBUG_PRINTK("sctp_do_bind: port: %d, new port: %d\n", + bp->port, snum); + + /* We must either be unbound, or bind to the same port. */ + if (bp->port && (snum != bp->port)) { + SCTP_DEBUG_PRINTK("sctp_do_bind:" + " New port %d does not match existing port " + "%d.\n", snum, bp->port); + return -EINVAL; + } + + if (snum && snum < PROT_SOCK && !capable(CAP_NET_BIND_SERVICE)) + return -EACCES; + + /* Make sure we are allowed to bind here. + * The function sctp_get_port_local() does duplicate address + * detection. + */ + if ((ret = sctp_get_port_local(sk, addr))) { + if (ret == (long) sk) { + /* This endpoint has a conflicting address. */ + return -EINVAL; + } else { + return -EADDRINUSE; + } + } + + /* Refresh ephemeral port. */ + if (!bp->port) + bp->port = inet_sk(sk)->num; + + /* Add the address to the bind address list. */ + sctp_local_bh_disable(); + 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); + sctp_write_unlock(&ep->base.addr_lock); + sctp_local_bh_enable(); + + /* Copy back into socket for getsockname() use. */ + if (!ret) { + inet_sk(sk)->sport = htons(inet_sk(sk)->num); + af->to_sk_saddr(addr, sk); + } + + return ret; +} + + /* ADDIP Section 4.1.1 Congestion Control of ASCONF Chunks + * + * 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 + * 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 + * from each endpoint). + */ +static int sctp_send_asconf(struct sctp_association *asoc, + struct sctp_chunk *chunk) +{ + int retval = 0; + + /* If there is an outstanding ASCONF chunk, queue it for later + * transmission. + */ + if (asoc->addip_last_asconf) { + __skb_queue_tail(&asoc->addip_chunks, (struct sk_buff *)chunk); + goto out; + } + + /* Hold the chunk until an ASCONF_ACK is received. */ + sctp_chunk_hold(chunk); + retval = sctp_primitive_ASCONF(asoc, chunk); + if (retval) + sctp_chunk_free(chunk); + else + asoc->addip_last_asconf = chunk; + +out: + return retval; +} + +/* Add a list of addresses as bind addresses to local endpoint or + * association. + * + * Basically run through each address specified in the addrs/addrcnt + * array/length pair, determine if it is IPv6 or IPv4 and call + * sctp_do_bind() on it. + * + * If any of them fails, then the operation will be reversed and the + * ones that were added will be removed. + * + * Only sctp_setsockopt_bindx() is supposed to call this function. + */ +int sctp_bindx_add(struct sock *sk, struct sockaddr *addrs, int addrcnt) +{ + int cnt; + int retval = 0; + void *addr_buf; + struct sockaddr *sa_addr; + struct sctp_af *af; + + SCTP_DEBUG_PRINTK("sctp_bindx_add (sk: %p, addrs: %p, addrcnt: %d)\n", + sk, addrs, addrcnt); + + addr_buf = addrs; + for (cnt = 0; cnt < addrcnt; cnt++) { + /* The list may contain either IPv4 or IPv6 address; + * determine the address length for walking thru the list. + */ + sa_addr = (struct sockaddr *)addr_buf; + af = sctp_get_af_specific(sa_addr->sa_family); + if (!af) { + retval = -EINVAL; + goto err_bindx_add; + } + + retval = sctp_do_bind(sk, (union sctp_addr *)sa_addr, + af->sockaddr_len); + + addr_buf += af->sockaddr_len; + +err_bindx_add: + if (retval < 0) { + /* Failed. Cleanup the ones that have been added */ + if (cnt > 0) + sctp_bindx_rem(sk, addrs, cnt); + return retval; + } + } + + return retval; +} + +/* Send an ASCONF chunk with Add IP address parameters to all the peers of the + * 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 + * 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, + struct sockaddr *addrs, + int addrcnt) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sctp_association *asoc; + struct sctp_bind_addr *bp; + struct sctp_chunk *chunk; + struct sctp_sockaddr_entry *laddr; + union sctp_addr *addr; + void *addr_buf; + struct sctp_af *af; + struct list_head *pos; + struct list_head *p; + int i; + int retval = 0; + + if (!sctp_addip_enable) + return retval; + + sp = sctp_sk(sk); + ep = sp->ep; + + SCTP_DEBUG_PRINTK("%s: (sk: %p, addrs: %p, addrcnt: %d)\n", + __FUNCTION__, sk, addrs, addrcnt); + + list_for_each(pos, &ep->asocs) { + asoc = list_entry(pos, struct sctp_association, asocs); + + if (!asoc->peer.asconf_capable) + continue; + + if (asoc->peer.addip_disabled_mask & SCTP_PARAM_ADD_IP) + continue; + + if (!sctp_state(asoc, ESTABLISHED)) + 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 + * other associations. + */ + addr_buf = addrs; + for (i = 0; i < addrcnt; i++) { + addr = (union sctp_addr *)addr_buf; + af = sctp_get_af_specific(addr->v4.sin_family); + if (!af) { + retval = -EINVAL; + goto out; + } + + if (sctp_assoc_lookup_laddr(asoc, addr)) + break; + + addr_buf += af->sockaddr_len; + } + if (i < addrcnt) + continue; + + /* Use the first address in bind addr list of association as + * Address Parameter of ASCONF CHUNK. + */ + sctp_read_lock(&asoc->base.addr_lock); + bp = &asoc->base.bind_addr; + p = bp->address_list.next; + laddr = list_entry(p, struct sctp_sockaddr_entry, list); + sctp_read_unlock(&asoc->base.addr_lock); + + chunk = sctp_make_asconf_update_ip(asoc, &laddr->a, addrs, + addrcnt, SCTP_PARAM_ADD_IP); + if (!chunk) { + retval = -ENOMEM; + goto out; + } + + retval = sctp_send_asconf(asoc, chunk); + + /* 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. + */ + } + +out: + return retval; +} + +/* Remove a list of addresses from bind addresses list. Do not remove the + * last address. + * + * Basically run through each address specified in the addrs/addrcnt + * array/length pair, determine if it is IPv6 or IPv4 and call + * sctp_del_bind() on it. + * + * If any of them fails, then the operation will be reversed and the + * ones that were removed will be added back. + * + * At least one address has to be left; if only one address is + * available, the operation will return -EBUSY. + * + * Only sctp_setsockopt_bindx() is supposed to call this function. + */ +int sctp_bindx_rem(struct sock *sk, struct sockaddr *addrs, int addrcnt) +{ + struct sctp_sock *sp = sctp_sk(sk); + struct sctp_endpoint *ep = sp->ep; + 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; + struct sctp_af *af; + + SCTP_DEBUG_PRINTK("sctp_bindx_rem (sk: %p, addrs: %p, addrcnt: %d)\n", + sk, addrs, addrcnt); + + addr_buf = addrs; + for (cnt = 0; cnt < addrcnt; cnt++) { + /* If the bind address list is empty or if there is only one + * bind address, there is nothing more to be removed (we need + * at least one address here). + */ + if (list_empty(&bp->address_list) || + (sctp_list_single_entry(&bp->address_list))) { + retval = -EBUSY; + 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); + 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) { + retval = -EINVAL; + goto err_bindx_rem; + } + + /* FIXME - There is probably a need to check if sk->sk_saddr and + * sk->sk_rcv_addr are currently set to one of the addresses to + * be removed. This is something which needs to be looked into + * when we are fixing the outstanding issues with multi-homing + * socket routing and failover schemes. Refer to comments in + * sctp_do_bind(). -daisy + */ + sctp_local_bh_disable(); + sctp_write_lock(&ep->base.addr_lock); + + retval = sctp_del_bind_addr(bp, &saveaddr); + + sctp_write_unlock(&ep->base.addr_lock); + sctp_local_bh_enable(); + + addr_buf += af->sockaddr_len; +err_bindx_rem: + if (retval < 0) { + /* Failed. Add the ones that has been removed back */ + if (cnt > 0) + sctp_bindx_add(sk, addrs, cnt); + return retval; + } + } + + return retval; +} + +/* Send an ASCONF chunk with Delete IP address parameters to all the peers of + * 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 + * 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_del_ip(struct sock *sk, + struct sockaddr *addrs, + int addrcnt) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sctp_association *asoc; + struct sctp_bind_addr *bp; + struct sctp_chunk *chunk; + union sctp_addr *laddr; + void *addr_buf; + struct sctp_af *af; + struct list_head *pos; + int i; + int retval = 0; + + if (!sctp_addip_enable) + return retval; + + sp = sctp_sk(sk); + ep = sp->ep; + + SCTP_DEBUG_PRINTK("%s: (sk: %p, addrs: %p, addrcnt: %d)\n", + __FUNCTION__, sk, addrs, addrcnt); + + list_for_each(pos, &ep->asocs) { + asoc = list_entry(pos, struct sctp_association, asocs); + + if (!asoc->peer.asconf_capable) + continue; + + if (asoc->peer.addip_disabled_mask & SCTP_PARAM_DEL_IP) + continue; + + if (!sctp_state(asoc, ESTABLISHED)) + continue; + + /* Check if any address in the packed array of addresses is + * 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. + */ + addr_buf = addrs; + for (i = 0; i < addrcnt; i++) { + laddr = (union sctp_addr *)addr_buf; + af = sctp_get_af_specific(laddr->v4.sin_family); + if (!af) { + retval = -EINVAL; + goto out; + } + + if (!sctp_assoc_lookup_laddr(asoc, laddr)) + break; + + addr_buf += af->sockaddr_len; + } + if (i < addrcnt) + continue; + + /* Find one address in the association's bind address list + * that is not in the packed array of addresses. This is to + * make sure that we do not delete all the addresses in the + * association. + */ + sctp_read_lock(&asoc->base.addr_lock); + bp = &asoc->base.bind_addr; + laddr = sctp_find_unmatch_addr(bp, (union sctp_addr *)addrs, + addrcnt, sp); + sctp_read_unlock(&asoc->base.addr_lock); + if (!laddr) + continue; + + chunk = sctp_make_asconf_update_ip(asoc, laddr, addrs, addrcnt, + SCTP_PARAM_DEL_IP); + if (!chunk) { + retval = -ENOMEM; + goto out; + } + + retval = sctp_send_asconf(asoc, chunk); + + /* 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. + */ + } +out: + return retval; +} + +/* Helper for tunneling sctp_bindx() requests through sctp_setsockopt() + * + * API 8.1 + * int sctp_bindx(int sd, struct sockaddr *addrs, int addrcnt, + * int flags); + * + * If sd is an IPv4 socket, the addresses passed must be IPv4 addresses. + * If the sd is an IPv6 socket, the addresses passed can either be IPv4 + * or IPv6 addresses. + * + * A single address may be specified as INADDR_ANY or IN6ADDR_ANY, see + * Section 3.1.2 for this usage. + * + * 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 + * representation is termed a "packed array" of addresses). The caller + * specifies the number of addresses in the array with addrcnt. + * + * On success, sctp_bindx() returns 0. On failure, sctp_bindx() returns + * -1, and sets errno to the appropriate error code. + * + * For SCTP, the port given in each socket address must be the same, or + * sctp_bindx() will fail, setting errno to EINVAL. + * + * The flags parameter is formed from the bitwise OR of zero or more of + * the following currently defined flags: + * + * SCTP_BINDX_ADD_ADDR + * + * SCTP_BINDX_REM_ADDR + * + * SCTP_BINDX_ADD_ADDR directs SCTP to add the given addresses to the + * association, and SCTP_BINDX_REM_ADDR directs SCTP to remove the given + * addresses from the association. The two flags are mutually exclusive; + * if both are given, sctp_bindx() will fail with EINVAL. A caller may + * not remove all addresses from an association; sctp_bindx() will + * reject such an attempt with EINVAL. + * + * An application can use sctp_bindx(SCTP_BINDX_ADD_ADDR) to associate + * additional addresses with an endpoint after calling bind(). Or use + * sctp_bindx(SCTP_BINDX_REM_ADDR) to remove some addresses a listening + * socket is associated with so that no new association accepted will be + * associated with those addresses. If the endpoint supports dynamic + * address a SCTP_BINDX_REM_ADDR or SCTP_BINDX_ADD_ADDR may cause a + * endpoint to send the appropriate message to the peer to change the + * peers address lists. + * + * Adding and removing addresses from a connected association is + * optional functionality. Implementations that do not support this + * functionality should return EOPNOTSUPP. + * + * Basically do nothing but copying the addresses from user to kernel + * land and invoking either sctp_bindx_add() or sctp_bindx_rem() on the sk. + * This is used for tunneling the sctp_bindx() request through sctp_setsockopt() * from userspace. + * + * We don't use copy_from_user() for optimization: we first do the + * sanity checks (buffer size -fast- and access check-healthy + * pointer); if all of those succeed, then we can alloc the memory + * (expensive operation) needed to copy the data to kernel. Then we do + * the copying without checking the user space area + * (__copy_from_user()). + * + * On exit there is no need to do sockfd_put(), sys_setsockopt() does + * it. + * + * sk The sk of the socket + * addrs The pointer to the addresses in user land + * addrssize Size of the addrs buffer + * op Operation to perform (add or remove, see the flags of + * sctp_bindx) + * + * Returns 0 if ok, <0 errno code on error. + */ +SCTP_STATIC int sctp_setsockopt_bindx(struct sock* sk, + struct sockaddr __user *addrs, + int addrs_size, int op) +{ + struct sockaddr *kaddrs; + int err; + int addrcnt = 0; + int walk_size = 0; + struct sockaddr *sa_addr; + void *addr_buf; + struct sctp_af *af; + + SCTP_DEBUG_PRINTK("sctp_setsocktopt_bindx: sk %p addrs %p" + " addrs_size %d opt %d\n", sk, addrs, addrs_size, op); + + if (unlikely(addrs_size <= 0)) + return -EINVAL; + + /* Check the user passed a healthy pointer. */ + if (unlikely(!access_ok(VERIFY_READ, addrs, addrs_size))) + return -EFAULT; + + /* Alloc space for the address array in kernel memory. */ + kaddrs = (struct sockaddr *)kmalloc(addrs_size, GFP_KERNEL); + if (unlikely(!kaddrs)) + return -ENOMEM; + + if (__copy_from_user(kaddrs, addrs, addrs_size)) { + kfree(kaddrs); + return -EFAULT; + } + + /* 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); + + /* 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; + } + addrcnt++; + addr_buf += af->sockaddr_len; + walk_size += af->sockaddr_len; + } + + /* Do the work. */ + switch (op) { + case SCTP_BINDX_ADD_ADDR: + err = sctp_bindx_add(sk, kaddrs, addrcnt); + if (err) + goto out; + err = sctp_send_asconf_add_ip(sk, kaddrs, addrcnt); + break; + + case SCTP_BINDX_REM_ADDR: + err = sctp_bindx_rem(sk, kaddrs, addrcnt); + if (err) + goto out; + err = sctp_send_asconf_del_ip(sk, kaddrs, addrcnt); + break; + + default: + err = -EINVAL; + break; + }; + +out: + kfree(kaddrs); + + return err; +} + +/* API 3.1.4 close() - UDP Style Syntax + * Applications use close() to perform graceful shutdown (as described in + * Section 10.1 of [SCTP]) on ALL the associations currently represented + * by a UDP-style socket. + * + * The syntax is + * + * ret = close(int sd); + * + * sd - the socket descriptor of the associations to be closed. + * + * To gracefully shutdown a specific association represented by the + * UDP-style socket, an application should use the sendmsg() call, + * passing no user data, but including the appropriate flag in the + * ancillary data (see Section xxxx). + * + * If sd in the close() call is a branched-off socket representing only + * one association, the shutdown is performed on that association only. + * + * 4.1.6 close() - TCP Style Syntax + * + * Applications use close() to gracefully close down an association. + * + * The syntax is: + * + * int close(int sd); + * + * sd - the socket descriptor of the association to be closed. + * + * After an application calls close() on a socket descriptor, no further + * socket operations will succeed on that descriptor. + * + * API 7.1.4 SO_LINGER + * + * An application using the TCP-style socket can use this option to + * perform the SCTP ABORT primitive. The linger option structure is: + * + * struct linger { + * int l_onoff; // option on/off + * int l_linger; // linger time + * }; + * + * To enable the option, set l_onoff to 1. If the l_linger value is set + * to 0, calling close() is the same as the ABORT primitive. If the + * value is set to a negative value, the setsockopt() call will return + * an error. If the value is set to a positive value linger_time, the + * close() can be blocked for at most linger_time ms. If the graceful + * shutdown phase does not finish during this period, close() will + * return but the graceful shutdown phase continues in the system. + */ +SCTP_STATIC void sctp_close(struct sock *sk, long timeout) +{ + struct sctp_endpoint *ep; + struct sctp_association *asoc; + struct list_head *pos, *temp; + + SCTP_DEBUG_PRINTK("sctp_close(sk: 0x%p, timeout:%ld)\n", sk, timeout); + + sctp_lock_sock(sk); + sk->sk_shutdown = SHUTDOWN_MASK; + + ep = sctp_sk(sk)->ep; + + /* Walk all associations on a socket, not on an endpoint. */ + list_for_each_safe(pos, temp, &ep->asocs) { + asoc = list_entry(pos, struct sctp_association, asocs); + + if (sctp_style(sk, TCP)) { + /* A closed association can still be in the list if + * it belongs to a TCP-style listening socket that is + * not yet accepted. If so, free it. If not, send an + * ABORT or SHUTDOWN based on the linger options. + */ + if (sctp_state(asoc, CLOSED)) { + sctp_unhash_established(asoc); + sctp_association_free(asoc); + + } else if (sock_flag(sk, SOCK_LINGER) && + !sk->sk_lingertime) + sctp_primitive_ABORT(asoc, NULL); + else + sctp_primitive_SHUTDOWN(asoc, NULL); + } else + sctp_primitive_SHUTDOWN(asoc, NULL); + } + + /* Clean up any skbs sitting on the receive queue. */ + sctp_queue_purge_ulpevents(&sk->sk_receive_queue); + sctp_queue_purge_ulpevents(&sctp_sk(sk)->pd_lobby); + + /* On a TCP-style socket, block for at most linger_time if set. */ + if (sctp_style(sk, TCP) && timeout) + sctp_wait_for_close(sk, timeout); + + /* This will run the backlog queue. */ + sctp_release_sock(sk); + + /* Supposedly, no process has access to the socket, but + * the net layers still may. + */ + sctp_local_bh_disable(); + sctp_bh_lock_sock(sk); + + /* Hold the sock, since sk_common_release() will put sock_put() + * and we have just a little more cleanup. + */ + sock_hold(sk); + sk_common_release(sk); + + sctp_bh_unlock_sock(sk); + sctp_local_bh_enable(); + + sock_put(sk); + + SCTP_DBG_OBJCNT_DEC(sock); +} + +/* Handle EPIPE error. */ +static int sctp_error(struct sock *sk, int flags, int err) +{ + if (err == -EPIPE) + err = sock_error(sk) ? : -EPIPE; + if (err == -EPIPE && !(flags & MSG_NOSIGNAL)) + send_sig(SIGPIPE, current, 0); + return err; +} + +/* API 3.1.3 sendmsg() - UDP Style Syntax + * + * An application uses sendmsg() and recvmsg() calls to transmit data to + * and receive data from its peer. + * + * ssize_t sendmsg(int socket, const struct msghdr *message, + * int flags); + * + * socket - the socket descriptor of the endpoint. + * message - pointer to the msghdr structure which contains a single + * user message and possibly some ancillary data. + * + * See Section 5 for complete description of the data + * structures. + * + * flags - flags sent or received with the user message, see Section + * 5 for complete description of the flags. + * + * Note: This function could use a rewrite especially when explicit + * connect support comes in. + */ +/* BUG: We do not implement the equivalent of sk_stream_wait_memory(). */ + +SCTP_STATIC int sctp_msghdr_parse(const struct msghdr *, sctp_cmsgs_t *); + +SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t msg_len) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sctp_association *new_asoc=NULL, *asoc=NULL; + struct sctp_transport *transport, *chunk_tp; + struct sctp_chunk *chunk; + union sctp_addr to; + struct sockaddr *msg_name = NULL; + struct sctp_sndrcvinfo default_sinfo = { 0 }; + struct sctp_sndrcvinfo *sinfo; + struct sctp_initmsg *sinit; + sctp_assoc_t associd = 0; + sctp_cmsgs_t cmsgs = { NULL }; + int err; + sctp_scope_t scope; + long timeo; + __u16 sinfo_flags = 0; + struct sctp_datamsg *datamsg; + struct list_head *pos; + int msg_flags = msg->msg_flags; + + SCTP_DEBUG_PRINTK("sctp_sendmsg(sk: %p, msg: %p, msg_len: %zu)\n", + sk, msg, msg_len); + + err = 0; + sp = sctp_sk(sk); + ep = sp->ep; + + SCTP_DEBUG_PRINTK("Using endpoint: %s.\n", ep->debug_name); + + /* We cannot send a message over a TCP-style listening socket. */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) { + err = -EPIPE; + goto out_nounlock; + } + + /* Parse out the SCTP CMSGs. */ + err = sctp_msghdr_parse(msg, &cmsgs); + + if (err) { + SCTP_DEBUG_PRINTK("msghdr parse err = %x\n", err); + goto out_nounlock; + } + + /* Fetch the destination address for this packet. This + * address only selects the association--it is not necessarily + * the address we will send to. + * For a peeled-off socket, msg_name is ignored. + */ + if (!sctp_style(sk, UDP_HIGH_BANDWIDTH) && msg->msg_name) { + int msg_namelen = msg->msg_namelen; + + err = sctp_verify_addr(sk, (union sctp_addr *)msg->msg_name, + msg_namelen); + if (err) + return err; + + 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; + } + + sinfo = cmsgs.info; + sinit = cmsgs.init; + + /* Did the user specify SNDRCVINFO? */ + if (sinfo) { + sinfo_flags = sinfo->sinfo_flags; + associd = sinfo->sinfo_assoc_id; + } + + SCTP_DEBUG_PRINTK("msg_len: %zu, sinfo_flags: 0x%x\n", + msg_len, sinfo_flags); + + /* MSG_EOF or MSG_ABORT cannot be set on a TCP-style socket. */ + if (sctp_style(sk, TCP) && (sinfo_flags & (MSG_EOF | MSG_ABORT))) { + err = -EINVAL; + goto out_nounlock; + } + + /* If MSG_EOF is set, no data can be sent. Disallow sending zero + * length messages when MSG_EOF|MSG_ABORT is not set. + * If MSG_ABORT is set, the message length could be non zero with + * the msg_iov set to the user abort reason. + */ + if (((sinfo_flags & MSG_EOF) && (msg_len > 0)) || + (!(sinfo_flags & (MSG_EOF|MSG_ABORT)) && (msg_len == 0))) { + err = -EINVAL; + goto out_nounlock; + } + + /* If MSG_ADDR_OVER is set, there must be an address + * specified in msg_name. + */ + if ((sinfo_flags & MSG_ADDR_OVER) && (!msg->msg_name)) { + err = -EINVAL; + goto out_nounlock; + } + + transport = NULL; + + SCTP_DEBUG_PRINTK("About to look up association.\n"); + + sctp_lock_sock(sk); + + /* If a msg_name has been specified, assume this is to be used. */ + if (msg_name) { + /* Look for a matching association on the endpoint. */ + asoc = sctp_endpoint_lookup_assoc(ep, &to, &transport); + if (!asoc) { + /* If we could not find a matching association on the + * endpoint, make sure that it is not a TCP-style + * socket that already has an association or there is + * no peeled-off association on another socket. + */ + if ((sctp_style(sk, TCP) && + sctp_sstate(sk, ESTABLISHED)) || + sctp_endpoint_is_peeled_off(ep, &to)) { + err = -EADDRNOTAVAIL; + goto out_unlock; + } + } + } else { + asoc = sctp_id2assoc(sk, associd); + if (!asoc) { + err = -EPIPE; + goto out_unlock; + } + } + + if (asoc) { + SCTP_DEBUG_PRINTK("Just looked up association: %p.\n", asoc); + + /* We cannot send a message on a TCP-style SCTP_SS_ESTABLISHED + * socket that has an association in CLOSED state. This can + * happen when an accepted socket has an association that is + * already CLOSED. + */ + if (sctp_state(asoc, CLOSED) && sctp_style(sk, TCP)) { + err = -EPIPE; + goto out_unlock; + } + + if (sinfo_flags & MSG_EOF) { + SCTP_DEBUG_PRINTK("Shutting down association: %p\n", + asoc); + sctp_primitive_SHUTDOWN(asoc, NULL); + err = 0; + goto out_unlock; + } + if (sinfo_flags & MSG_ABORT) { + SCTP_DEBUG_PRINTK("Aborting association: %p\n", asoc); + sctp_primitive_ABORT(asoc, msg); + err = 0; + goto out_unlock; + } + } + + /* Do we need to create the association? */ + if (!asoc) { + SCTP_DEBUG_PRINTK("There is no association yet.\n"); + + if (sinfo_flags & (MSG_EOF | MSG_ABORT)) { + err = -EINVAL; + goto out_unlock; + } + + /* Check for invalid stream against the stream counts, + * either the default or the user specified stream counts. + */ + if (sinfo) { + if (!sinit || (sinit && !sinit->sinit_num_ostreams)) { + /* Check against the defaults. */ + if (sinfo->sinfo_stream >= + sp->initmsg.sinit_num_ostreams) { + err = -EINVAL; + goto out_unlock; + } + } else { + /* Check against the requested. */ + if (sinfo->sinfo_stream >= + sinit->sinit_num_ostreams) { + err = -EINVAL; + goto out_unlock; + } + } + } + + /* + * API 3.1.2 bind() - UDP Style Syntax + * If a bind() or sctp_bindx() is not called prior to a + * sendmsg() call that initiates a new association, the + * system picks an ephemeral port and will choose an address + * set equivalent to binding with a wildcard address. + */ + if (!ep->base.bind_addr.port) { + if (sctp_autobind(sk)) { + err = -EAGAIN; + goto out_unlock; + } + } + + scope = sctp_scope(&to); + new_asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL); + if (!new_asoc) { + err = -ENOMEM; + goto out_unlock; + } + asoc = new_asoc; + + /* If the SCTP_INIT ancillary data is specified, set all + * the association init values accordingly. + */ + if (sinit) { + if (sinit->sinit_num_ostreams) { + asoc->c.sinit_num_ostreams = + sinit->sinit_num_ostreams; + } + if (sinit->sinit_max_instreams) { + asoc->c.sinit_max_instreams = + sinit->sinit_max_instreams; + } + if (sinit->sinit_max_attempts) { + asoc->max_init_attempts + = sinit->sinit_max_attempts; + } + if (sinit->sinit_max_init_timeo) { + asoc->max_init_timeo = + msecs_to_jiffies(sinit->sinit_max_init_timeo); + } + } + + /* Prime the peer's transport structures. */ + transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL); + if (!transport) { + err = -ENOMEM; + goto out_free; + } + err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL); + if (err < 0) { + err = -ENOMEM; + goto out_free; + } + } + + /* ASSERT: we have a valid association at this point. */ + SCTP_DEBUG_PRINTK("We have a valid association.\n"); + + if (!sinfo) { + /* If the user didn't specify SNDRCVINFO, make up one with + * some defaults. + */ + default_sinfo.sinfo_stream = asoc->default_stream; + default_sinfo.sinfo_flags = asoc->default_flags; + default_sinfo.sinfo_ppid = asoc->default_ppid; + default_sinfo.sinfo_context = asoc->default_context; + default_sinfo.sinfo_timetolive = asoc->default_timetolive; + default_sinfo.sinfo_assoc_id = sctp_assoc2id(asoc); + sinfo = &default_sinfo; + } + + /* API 7.1.7, the sndbuf size per association bounds the + * maximum size of data that can be sent in a single send call. + */ + if (msg_len > sk->sk_sndbuf) { + err = -EMSGSIZE; + goto out_free; + } + + /* If fragmentation is disabled and the message length exceeds the + * association fragmentation point, return EMSGSIZE. The I-D + * does not specify what this error is, but this looks like + * a great fit. + */ + if (sctp_sk(sk)->disable_fragments && (msg_len > asoc->frag_point)) { + err = -EMSGSIZE; + goto out_free; + } + + if (sinfo) { + /* Check for invalid stream. */ + if (sinfo->sinfo_stream >= asoc->c.sinit_num_ostreams) { + err = -EINVAL; + goto out_free; + } + } + + timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT); + if (!sctp_wspace(asoc)) { + err = sctp_wait_for_sndbuf(asoc, &timeo, msg_len); + if (err) + goto out_free; + } + + /* If an address is passed with the sendto/sendmsg call, it is used + * to override the primary destination address in the TCP model, or + * when MSG_ADDR_OVER flag is set in the UDP model. + */ + if ((sctp_style(sk, TCP) && msg_name) || + (sinfo_flags & MSG_ADDR_OVER)) { + chunk_tp = sctp_assoc_lookup_paddr(asoc, &to); + if (!chunk_tp) { + err = -EINVAL; + goto out_free; + } + } else + chunk_tp = NULL; + + /* Auto-connect, if we aren't connected already. */ + if (sctp_state(asoc, CLOSED)) { + err = sctp_primitive_ASSOCIATE(asoc, NULL); + if (err < 0) + goto out_free; + SCTP_DEBUG_PRINTK("We associated primitively.\n"); + } + + /* Break the message into multiple chunks of maximum size. */ + datamsg = sctp_datamsg_from_user(asoc, sinfo, msg, msg_len); + if (!datamsg) { + err = -ENOMEM; + goto out_free; + } + + /* Now send the (possibly) fragmented message. */ + list_for_each(pos, &datamsg->chunks) { + chunk = list_entry(pos, struct sctp_chunk, frag_list); + sctp_datamsg_track(chunk); + + /* Do accounting for the write space. */ + sctp_set_owner_w(chunk); + + chunk->transport = chunk_tp; + + /* Send it to the lower layers. Note: all chunks + * must either fail or succeed. The lower layer + * works that way today. Keep it that way or this + * breaks. + */ + err = sctp_primitive_SEND(asoc, chunk); + /* Did the lower layer accept the chunk? */ + if (err) + sctp_chunk_free(chunk); + SCTP_DEBUG_PRINTK("We sent primitively.\n"); + } + + sctp_datamsg_free(datamsg); + if (err) + goto out_free; + else + err = msg_len; + + /* If we are already past ASSOCIATE, the lower + * layers are responsible for association cleanup. + */ + goto out_unlock; + +out_free: + if (new_asoc) + sctp_association_free(asoc); +out_unlock: + sctp_release_sock(sk); + +out_nounlock: + return sctp_error(sk, msg_flags, err); + +#if 0 +do_sock_err: + if (msg_len) + err = msg_len; + else + err = sock_error(sk); + goto out; + +do_interrupted: + if (msg_len) + err = msg_len; + goto out; +#endif /* 0 */ +} + +/* This is an extended version of skb_pull() that removes the data from the + * start of a skb even when data is spread across the list of skb's in the + * frag_list. len specifies the total amount of data that needs to be removed. + * when 'len' bytes could be removed from the skb, it returns 0. + * If 'len' exceeds the total skb length, it returns the no. of bytes that + * could not be removed. + */ +static int sctp_skb_pull(struct sk_buff *skb, int len) +{ + struct sk_buff *list; + int skb_len = skb_headlen(skb); + int rlen; + + if (len <= skb_len) { + __skb_pull(skb, len); + return 0; + } + len -= skb_len; + __skb_pull(skb, skb_len); + + for (list = skb_shinfo(skb)->frag_list; list; list = list->next) { + rlen = sctp_skb_pull(list, len); + skb->len -= (len-rlen); + skb->data_len -= (len-rlen); + + if (!rlen) + return 0; + + len = rlen; + } + + return len; +} + +/* API 3.1.3 recvmsg() - UDP Style Syntax + * + * ssize_t recvmsg(int socket, struct msghdr *message, + * int flags); + * + * socket - the socket descriptor of the endpoint. + * message - pointer to the msghdr structure which contains a single + * user message and possibly some ancillary data. + * + * See Section 5 for complete description of the data + * structures. + * + * flags - flags sent or received with the user message, see Section + * 5 for complete description of the flags. + */ +static struct sk_buff *sctp_skb_recv_datagram(struct sock *, int, int, int *); + +SCTP_STATIC int sctp_recvmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len, int noblock, + int flags, int *addr_len) +{ + struct sctp_ulpevent *event = NULL; + struct sctp_sock *sp = sctp_sk(sk); + struct sk_buff *skb; + int copied; + int err = 0; + int skb_len; + + SCTP_DEBUG_PRINTK("sctp_recvmsg(%s: %p, %s: %p, %s: %zd, %s: %d, %s: " + "0x%x, %s: %p)\n", "sk", sk, "msghdr", msg, + "len", len, "knoblauch", noblock, + "flags", flags, "addr_len", addr_len); + + sctp_lock_sock(sk); + + if (sctp_style(sk, TCP) && !sctp_sstate(sk, ESTABLISHED)) { + err = -ENOTCONN; + goto out; + } + + skb = sctp_skb_recv_datagram(sk, flags, noblock, &err); + if (!skb) + goto out; + + /* Get the total length of the skb including any skb's in the + * frag_list. + */ + skb_len = skb->len; + + copied = skb_len; + if (copied > len) + copied = len; + + err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied); + + event = sctp_skb2event(skb); + + if (err) + goto out_free; + + sock_recv_timestamp(msg, sk, skb); + if (sctp_ulpevent_is_notification(event)) { + msg->msg_flags |= MSG_NOTIFICATION; + sp->pf->event_msgname(event, msg->msg_name, addr_len); + } else { + sp->pf->skb_msgname(skb, msg->msg_name, addr_len); + } + + /* Check if we allow SCTP_SNDRCVINFO. */ + if (sp->subscribe.sctp_data_io_event) + sctp_ulpevent_read_sndrcvinfo(event, msg); +#if 0 + /* FIXME: we should be calling IP/IPv6 layers. */ + if (sk->sk_protinfo.af_inet.cmsg_flags) + ip_cmsg_recv(msg, skb); +#endif + + err = copied; + + /* If skb's length exceeds the user's buffer, update the skb and + * push it back to the receive_queue so that the next call to + * recvmsg() will return the remaining data. Don't set MSG_EOR. + */ + if (skb_len > copied) { + msg->msg_flags &= ~MSG_EOR; + if (flags & MSG_PEEK) + goto out_free; + sctp_skb_pull(skb, copied); + skb_queue_head(&sk->sk_receive_queue, skb); + + /* When only partial message is copied to the user, increase + * rwnd by that amount. If all the data in the skb is read, + * rwnd is updated when the event is freed. + */ + sctp_assoc_rwnd_increase(event->asoc, copied); + goto out; + } else if ((event->msg_flags & MSG_NOTIFICATION) || + (event->msg_flags & MSG_EOR)) + msg->msg_flags |= MSG_EOR; + else + msg->msg_flags &= ~MSG_EOR; + +out_free: + if (flags & MSG_PEEK) { + /* Release the skb reference acquired after peeking the skb in + * sctp_skb_recv_datagram(). + */ + kfree_skb(skb); + } else { + /* Free the event which includes releasing the reference to + * the owner of the skb, freeing the skb and updating the + * rwnd. + */ + sctp_ulpevent_free(event); + } +out: + sctp_release_sock(sk); + return err; +} + +/* 7.1.12 Enable/Disable message fragmentation (SCTP_DISABLE_FRAGMENTS) + * + * This option is a on/off flag. If enabled no SCTP message + * fragmentation will be performed. Instead if a message being sent + * exceeds the current PMTU size, the message will NOT be sent and + * instead a error will be indicated to the user. + */ +static int sctp_setsockopt_disable_fragments(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; + + sctp_sk(sk)->disable_fragments = (val == 0) ? 0 : 1; + + return 0; +} + +static int sctp_setsockopt_events(struct sock *sk, char __user *optval, + int optlen) +{ + if (optlen != sizeof(struct sctp_event_subscribe)) + return -EINVAL; + if (copy_from_user(&sctp_sk(sk)->subscribe, optval, optlen)) + return -EFAULT; + return 0; +} + +/* 7.1.8 Automatic Close of associations (SCTP_AUTOCLOSE) + * + * This socket option is applicable to the UDP-style socket only. When + * set it will cause associations that are idle for more than the + * specified number of seconds to automatically close. An association + * being idle is defined an association that has NOT sent or received + * user data. The special value of '0' indicates that no automatic + * close of any associations should be performed. The option expects an + * integer defining the number of seconds of idle time before an + * association is closed. + */ +static int sctp_setsockopt_autoclose(struct sock *sk, char __user *optval, + int optlen) +{ + struct sctp_sock *sp = sctp_sk(sk); + + /* Applicable to UDP-style socket only */ + if (sctp_style(sk, TCP)) + return -EOPNOTSUPP; + if (optlen != sizeof(int)) + return -EINVAL; + if (copy_from_user(&sp->autoclose, optval, optlen)) + return -EFAULT; + + sp->ep->timeouts[SCTP_EVENT_TIMEOUT_AUTOCLOSE] = sp->autoclose * HZ; + return 0; +} + +/* 7.1.13 Peer Address Parameters (SCTP_PEER_ADDR_PARAMS) + * + * 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 following structure is used to access and modify an + * 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. + * 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. + * spp_pathmaxrxt - This contains the maximum number of + * retransmissions before this address shall be + * considered unreachable. + */ +static int sctp_setsockopt_peer_addr_params(struct sock *sk, + char __user *optval, int optlen) +{ + struct sctp_paddrparams params; + struct sctp_transport *trans; + int error; + + if (optlen != sizeof(struct sctp_paddrparams)) + 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). + */ + + /* 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) + 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) + 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. + */ + 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_hbinterval) { + trans->hb_allowed = 1; + trans->hb_interval = + msecs_to_jiffies(params.spp_hbinterval); + } else + trans->hb_allowed = 0; + } + + /* spp_pathmaxrxt contains the maximum number of retransmissions + * before this address shall be considered unreachable. + */ + if (params.spp_pathmaxrxt) + trans->max_retrans = params.spp_pathmaxrxt; + + return 0; +} + +/* 7.1.3 Initialization Parameters (SCTP_INITMSG) + * + * Applications can specify protocol parameters for the default association + * initialization. The option name argument to setsockopt() and getsockopt() + * is SCTP_INITMSG. + * + * Setting initialization parameters is effective only on an unconnected + * socket (for UDP-style sockets only future associations are effected + * by the change). With TCP-style sockets, this option is inherited by + * sockets derived from a listener socket. + */ +static int sctp_setsockopt_initmsg(struct sock *sk, char __user *optval, int optlen) +{ + struct sctp_initmsg sinit; + struct sctp_sock *sp = sctp_sk(sk); + + if (optlen != sizeof(struct sctp_initmsg)) + return -EINVAL; + if (copy_from_user(&sinit, optval, optlen)) + return -EFAULT; + + if (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; + if (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; + + return 0; +} + +/* + * 7.1.14 Set default send parameters (SCTP_DEFAULT_SEND_PARAM) + * + * Applications that wish to use the sendto() system call may wish to + * specify a default set of parameters that would normally be supplied + * through the inclusion of ancillary data. This socket option allows + * such an application to set the default sctp_sndrcvinfo structure. + * The application that wishes to use this socket option simply passes + * in to this call the sctp_sndrcvinfo structure defined in Section + * 5.2.2) The input parameters accepted by this call include + * sinfo_stream, sinfo_flags, sinfo_ppid, sinfo_context, + * sinfo_timetolive. The user must provide the sinfo_assoc_id field in + * to this call if the caller is using the UDP model. + */ +static int sctp_setsockopt_default_send_param(struct sock *sk, + char __user *optval, int optlen) +{ + struct sctp_sndrcvinfo info; + struct sctp_association *asoc; + struct sctp_sock *sp = sctp_sk(sk); + + if (optlen != sizeof(struct sctp_sndrcvinfo)) + return -EINVAL; + if (copy_from_user(&info, optval, optlen)) + return -EFAULT; + + asoc = sctp_id2assoc(sk, info.sinfo_assoc_id); + if (!asoc && info.sinfo_assoc_id && sctp_style(sk, UDP)) + return -EINVAL; + + if (asoc) { + asoc->default_stream = info.sinfo_stream; + asoc->default_flags = info.sinfo_flags; + asoc->default_ppid = info.sinfo_ppid; + asoc->default_context = info.sinfo_context; + asoc->default_timetolive = info.sinfo_timetolive; + } else { + sp->default_stream = info.sinfo_stream; + sp->default_flags = info.sinfo_flags; + sp->default_ppid = info.sinfo_ppid; + sp->default_context = info.sinfo_context; + sp->default_timetolive = info.sinfo_timetolive; + } + + return 0; +} + +/* 7.1.10 Set Primary Address (SCTP_PRIMARY_ADDR) + * + * Requests that the local SCTP stack use the enclosed peer address as + * the association primary. The enclosed address must be one of the + * association peer's addresses. + */ +static int sctp_setsockopt_primary_addr(struct sock *sk, char __user *optval, + int optlen) +{ + struct sctp_prim prim; + struct sctp_transport *trans; + + if (optlen != sizeof(struct sctp_prim)) + return -EINVAL; + + if (copy_from_user(&prim, optval, sizeof(struct sctp_prim))) + return -EFAULT; + + trans = sctp_addr_id2transport(sk, &prim.ssp_addr, prim.ssp_assoc_id); + if (!trans) + return -EINVAL; + + sctp_assoc_set_primary(trans->asoc, trans); + + return 0; +} + +/* + * 7.1.5 SCTP_NODELAY + * + * Turn on/off any Nagle-like algorithm. This means that packets are + * generally sent as soon as possible and no unnecessary delays are + * introduced, at the cost of more packets in the network. Expects an + * integer boolean flag. + */ +static int sctp_setsockopt_nodelay(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; + + sctp_sk(sk)->nodelay = (val == 0) ? 0 : 1; + return 0; +} + +/* + * + * 7.1.1 SCTP_RTOINFO + * + * The protocol parameters used to initialize and bound retransmission + * timeout (RTO) are tunable. sctp_rtoinfo structure is used to access + * and modify these parameters. + * All parameters are time values, in milliseconds. A value of 0, when + * modifying the parameters, indicates that the current value should not + * be changed. + * + */ +static int sctp_setsockopt_rtoinfo(struct sock *sk, char __user *optval, int optlen) { + struct sctp_rtoinfo rtoinfo; + struct sctp_association *asoc; + + if (optlen != sizeof (struct sctp_rtoinfo)) + return -EINVAL; + + if (copy_from_user(&rtoinfo, optval, optlen)) + return -EFAULT; + + asoc = sctp_id2assoc(sk, rtoinfo.srto_assoc_id); + + /* Set the values to the specific association */ + if (!asoc && rtoinfo.srto_assoc_id && sctp_style(sk, UDP)) + return -EINVAL; + + if (asoc) { + if (rtoinfo.srto_initial != 0) + asoc->rto_initial = + msecs_to_jiffies(rtoinfo.srto_initial); + if (rtoinfo.srto_max != 0) + asoc->rto_max = msecs_to_jiffies(rtoinfo.srto_max); + if (rtoinfo.srto_min != 0) + asoc->rto_min = msecs_to_jiffies(rtoinfo.srto_min); + } else { + /* If there is no association or the association-id = 0 + * set the values to the endpoint. + */ + struct sctp_sock *sp = sctp_sk(sk); + + if (rtoinfo.srto_initial != 0) + sp->rtoinfo.srto_initial = rtoinfo.srto_initial; + if (rtoinfo.srto_max != 0) + sp->rtoinfo.srto_max = rtoinfo.srto_max; + if (rtoinfo.srto_min != 0) + sp->rtoinfo.srto_min = rtoinfo.srto_min; + } + + return 0; +} + +/* + * + * 7.1.2 SCTP_ASSOCINFO + * + * This option is used to tune the 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. + * See [SCTP] for more information. + * + */ +static int sctp_setsockopt_associnfo(struct sock *sk, char __user *optval, int optlen) +{ + + struct sctp_assocparams assocparams; + struct sctp_association *asoc; + + if (optlen != sizeof(struct sctp_assocparams)) + return -EINVAL; + if (copy_from_user(&assocparams, optval, optlen)) + return -EFAULT; + + asoc = sctp_id2assoc(sk, assocparams.sasoc_assoc_id); + + if (!asoc && assocparams.sasoc_assoc_id && sctp_style(sk, UDP)) + return -EINVAL; + + /* Set the values to the specific association */ + if (asoc) { + if (assocparams.sasoc_asocmaxrxt != 0) + asoc->max_retrans = assocparams.sasoc_asocmaxrxt; + if (assocparams.sasoc_cookie_life != 0) { + asoc->cookie_life.tv_sec = + assocparams.sasoc_cookie_life / 1000; + asoc->cookie_life.tv_usec = + (assocparams.sasoc_cookie_life % 1000) + * 1000; + } + } else { + /* Set the values to the endpoint */ + struct sctp_sock *sp = sctp_sk(sk); + + if (assocparams.sasoc_asocmaxrxt != 0) + sp->assocparams.sasoc_asocmaxrxt = + assocparams.sasoc_asocmaxrxt; + if (assocparams.sasoc_cookie_life != 0) + sp->assocparams.sasoc_cookie_life = + assocparams.sasoc_cookie_life; + } + return 0; +} + +/* + * 7.1.16 Set/clear IPv4 mapped addresses (SCTP_I_WANT_MAPPED_V4_ADDR) + * + * This socket option is a boolean flag which turns on or off mapped V4 + * addresses. If this option is turned on and the socket is type + * PF_INET6, then IPv4 addresses will be mapped to V6 representation. + * If this option is turned off, then no mapping will be done of V4 + * addresses and a user will receive both PF_INET6 and PF_INET type + * addresses on the socket. + */ +static int sctp_setsockopt_mappedv4(struct sock *sk, char __user *optval, int optlen) +{ + int val; + struct sctp_sock *sp = sctp_sk(sk); + + if (optlen < sizeof(int)) + return -EINVAL; + if (get_user(val, (int __user *)optval)) + return -EFAULT; + if (val) + sp->v4mapped = 1; + else + sp->v4mapped = 0; + + return 0; +} + +/* + * 7.1.17 Set the maximum fragrmentation size (SCTP_MAXSEG) + * + * This socket option specifies the maximum size to put in any outgoing + * SCTP chunk. If a message is larger than this size it will be + * fragmented by SCTP into the specified size. Note that the underlying + * SCTP implementation may fragment into smaller sized chunks when the + * PMTU of the underlying association is smaller than the value set by + * the user. + */ +static int sctp_setsockopt_maxseg(struct sock *sk, char __user *optval, int optlen) +{ + struct sctp_association *asoc; + struct list_head *pos; + struct sctp_sock *sp = sctp_sk(sk); + int val; + + if (optlen < sizeof(int)) + return -EINVAL; + if (get_user(val, (int __user *)optval)) + return -EFAULT; + if ((val < 8) || (val > SCTP_MAX_CHUNK_LEN)) + return -EINVAL; + sp->user_frag = val; + + if (val) { + /* 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); + } + } + + return 0; +} + + +/* + * 7.1.9 Set Peer Primary Address (SCTP_SET_PEER_PRIMARY_ADDR) + * + * Requests that the peer mark the enclosed address as the association + * primary. The enclosed address must be one of the association's + * locally bound addresses. The following structure is used to make a + * set primary request: + */ +static int sctp_setsockopt_peer_primary_addr(struct sock *sk, char __user *optval, + int optlen) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sctp_association *asoc = NULL; + struct sctp_setpeerprim prim; + struct sctp_chunk *chunk; + int err; + + sp = sctp_sk(sk); + ep = sp->ep; + + if (!sctp_addip_enable) + return -EPERM; + + if (optlen != sizeof(struct sctp_setpeerprim)) + return -EINVAL; + + if (copy_from_user(&prim, optval, optlen)) + return -EFAULT; + + asoc = sctp_id2assoc(sk, prim.sspp_assoc_id); + if (!asoc) + return -EINVAL; + + if (!asoc->peer.asconf_capable) + return -EPERM; + + if (asoc->peer.addip_disabled_mask & SCTP_PARAM_SET_PRIMARY) + return -EPERM; + + if (!sctp_state(asoc, ESTABLISHED)) + return -ENOTCONN; + + if (!sctp_assoc_lookup_laddr(asoc, (union sctp_addr *)&prim.sspp_addr)) + return -EADDRNOTAVAIL; + + /* Create an ASCONF chunk with SET_PRIMARY parameter */ + chunk = sctp_make_asconf_set_prim(asoc, + (union sctp_addr *)&prim.sspp_addr); + if (!chunk) + return -ENOMEM; + + err = sctp_send_asconf(asoc, chunk); + + SCTP_DEBUG_PRINTK("We set peer primary addr primitively.\n"); + + return err; +} + +static int sctp_setsockopt_adaption_layer(struct sock *sk, char __user *optval, + int optlen) +{ + __u32 val; + + if (optlen < sizeof(__u32)) + return -EINVAL; + if (copy_from_user(&val, optval, sizeof(__u32))) + return -EFAULT; + + sctp_sk(sk)->adaption_ind = val; + + return 0; +} + +/* API 6.2 setsockopt(), getsockopt() + * + * Applications use setsockopt() and getsockopt() to set or retrieve + * socket options. Socket options are used to change the default + * behavior of sockets calls. They are described in Section 7. + * + * The syntax is: + * + * ret = getsockopt(int sd, int level, int optname, void __user *optval, + * int __user *optlen); + * ret = setsockopt(int sd, int level, int optname, const void __user *optval, + * int optlen); + * + * sd - the socket descript. + * level - set to IPPROTO_SCTP for all SCTP options. + * optname - the option name. + * optval - the buffer to store the value of the option. + * optlen - the size of the buffer. + */ +SCTP_STATIC int sctp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) +{ + int retval = 0; + + SCTP_DEBUG_PRINTK("sctp_setsockopt(sk: %p... optname: %d)\n", + sk, optname); + + /* I can hardly begin to describe how wrong this is. This is + * so broken as to be worse than useless. The API draft + * REALLY is NOT helpful here... I am not convinced that the + * semantics of setsockopt() with a level OTHER THAN SOL_SCTP + * are at all well-founded. + */ + if (level != SOL_SCTP) { + struct sctp_af *af = sctp_sk(sk)->pf->af; + retval = af->setsockopt(sk, level, optname, optval, optlen); + goto out_nounlock; + } + + sctp_lock_sock(sk); + + switch (optname) { + case SCTP_SOCKOPT_BINDX_ADD: + /* 'optlen' is the size of the addresses buffer. */ + retval = sctp_setsockopt_bindx(sk, (struct sockaddr __user *)optval, + optlen, SCTP_BINDX_ADD_ADDR); + break; + + case SCTP_SOCKOPT_BINDX_REM: + /* 'optlen' is the size of the addresses buffer. */ + retval = sctp_setsockopt_bindx(sk, (struct sockaddr __user *)optval, + optlen, SCTP_BINDX_REM_ADDR); + break; + + case SCTP_DISABLE_FRAGMENTS: + retval = sctp_setsockopt_disable_fragments(sk, optval, optlen); + break; + + case SCTP_EVENTS: + retval = sctp_setsockopt_events(sk, optval, optlen); + break; + + case SCTP_AUTOCLOSE: + retval = sctp_setsockopt_autoclose(sk, optval, optlen); + break; + + case SCTP_PEER_ADDR_PARAMS: + retval = sctp_setsockopt_peer_addr_params(sk, optval, optlen); + break; + + case SCTP_INITMSG: + retval = sctp_setsockopt_initmsg(sk, optval, optlen); + break; + case SCTP_DEFAULT_SEND_PARAM: + retval = sctp_setsockopt_default_send_param(sk, optval, + optlen); + break; + case SCTP_PRIMARY_ADDR: + retval = sctp_setsockopt_primary_addr(sk, optval, optlen); + break; + case SCTP_SET_PEER_PRIMARY_ADDR: + retval = sctp_setsockopt_peer_primary_addr(sk, optval, optlen); + break; + case SCTP_NODELAY: + retval = sctp_setsockopt_nodelay(sk, optval, optlen); + break; + case SCTP_RTOINFO: + retval = sctp_setsockopt_rtoinfo(sk, optval, optlen); + break; + case SCTP_ASSOCINFO: + retval = sctp_setsockopt_associnfo(sk, optval, optlen); + break; + case SCTP_I_WANT_MAPPED_V4_ADDR: + retval = sctp_setsockopt_mappedv4(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); + break; + + default: + retval = -ENOPROTOOPT; + break; + }; + + sctp_release_sock(sk); + +out_nounlock: + return retval; +} + +/* API 3.1.6 connect() - UDP Style Syntax + * + * An application may use the connect() call in the UDP model to initiate an + * association without sending data. + * + * The syntax is: + * + * ret = connect(int sd, const struct sockaddr *nam, socklen_t len); + * + * sd: the socket descriptor to have a new association added to. + * + * nam: the address structure (either struct sockaddr_in or struct + * sockaddr_in6 defined in RFC2553 [7]). + * + * len: the size of the address. + */ +SCTP_STATIC int sctp_connect(struct sock *sk, struct sockaddr *uaddr, + int addr_len) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sctp_association *asoc; + struct sctp_transport *transport; + union sctp_addr to; + struct sctp_af *af; + sctp_scope_t scope; + long timeo; + int err = 0; + + sctp_lock_sock(sk); + + SCTP_DEBUG_PRINTK("%s - sk: %p, sockaddr: %p, addr_len: %d)\n", + __FUNCTION__, sk, uaddr, addr_len); + + sp = sctp_sk(sk); + ep = sp->ep; + + /* connect() cannot be done on a socket that is already in ESTABLISHED + * state - UDP-style peeled off socket or a TCP-style socket that + * is already connected. + * It cannot be done even on a TCP-style listening socket. + */ + if (sctp_sstate(sk, ESTABLISHED) || + (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING))) { + err = -EISCONN; + goto out_unlock; + } + + err = sctp_verify_addr(sk, (union sctp_addr *)uaddr, addr_len); + if (err) + goto out_unlock; + + if (addr_len > sizeof(to)) + addr_len = sizeof(to); + memcpy(&to, uaddr, addr_len); + to.v4.sin_port = ntohs(to.v4.sin_port); + + asoc = sctp_endpoint_lookup_assoc(ep, &to, &transport); + if (asoc) { + if (asoc->state >= SCTP_STATE_ESTABLISHED) + err = -EISCONN; + else + err = -EALREADY; + goto out_unlock; + } + + /* If we could not find a matching association on the endpoint, + * 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)) { + err = -EADDRNOTAVAIL; + goto out_unlock; + } + + /* If a bind() or sctp_bindx() is not called prior to a connect() + * call, the system picks an ephemeral port and will choose an address + * set equivalent to binding with a wildcard address. + */ + if (!ep->base.bind_addr.port) { + if (sctp_autobind(sk)) { + err = -EAGAIN; + goto out_unlock; + } + } + + scope = sctp_scope(&to); + asoc = sctp_association_new(ep, sk, scope, GFP_KERNEL); + if (!asoc) { + err = -ENOMEM; + goto out_unlock; + } + + /* Prime the peer's transport structures. */ + transport = sctp_assoc_add_peer(asoc, &to, GFP_KERNEL); + if (!transport) { + sctp_association_free(asoc); + goto out_unlock; + } + err = sctp_assoc_set_bind_addr_from_ep(asoc, GFP_KERNEL); + if (err < 0) { + sctp_association_free(asoc); + goto out_unlock; + } + + err = sctp_primitive_ASSOCIATE(asoc, NULL); + if (err < 0) { + sctp_association_free(asoc); + goto out_unlock; + } + + /* Initialize sk's dport and daddr for getpeername() */ + inet_sk(sk)->dport = htons(asoc->peer.port); + af = sctp_get_af_specific(to.sa.sa_family); + af->to_sk_daddr(&to, sk); + + timeo = sock_sndtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK); + err = sctp_wait_for_connect(asoc, &timeo); + +out_unlock: + sctp_release_sock(sk); + + return err; +} + +/* FIXME: Write comments. */ +SCTP_STATIC int sctp_disconnect(struct sock *sk, int flags) +{ + return -EOPNOTSUPP; /* STUB */ +} + +/* 4.1.4 accept() - TCP Style Syntax + * + * Applications use accept() call to remove an established SCTP + * association from the accept queue of the endpoint. A new socket + * descriptor will be returned from accept() to represent the newly + * formed association. + */ +SCTP_STATIC struct sock *sctp_accept(struct sock *sk, int flags, int *err) +{ + struct sctp_sock *sp; + struct sctp_endpoint *ep; + struct sock *newsk = NULL; + struct sctp_association *asoc; + long timeo; + int error = 0; + + sctp_lock_sock(sk); + + sp = sctp_sk(sk); + ep = sp->ep; + + if (!sctp_style(sk, TCP)) { + error = -EOPNOTSUPP; + goto out; + } + + if (!sctp_sstate(sk, LISTENING)) { + error = -EINVAL; + goto out; + } + + timeo = sock_rcvtimeo(sk, sk->sk_socket->file->f_flags & O_NONBLOCK); + + error = sctp_wait_for_accept(sk, timeo); + if (error) + goto out; + + /* We treat the list of associations on the endpoint as the accept + * queue and pick the first association on the list. + */ + asoc = list_entry(ep->asocs.next, struct sctp_association, asocs); + + newsk = sp->pf->create_accept_sk(sk, asoc); + if (!newsk) { + error = -ENOMEM; + goto out; + } + + /* Populate the fields of the newsk from the oldsk and migrate the + * asoc to the newsk. + */ + sctp_sock_migrate(sk, newsk, asoc, SCTP_SOCKET_TCP); + +out: + sctp_release_sock(sk); + *err = error; + return newsk; +} + +/* The SCTP ioctl handler. */ +SCTP_STATIC int sctp_ioctl(struct sock *sk, int cmd, unsigned long arg) +{ + return -ENOIOCTLCMD; +} + +/* This is the function which gets called during socket creation to + * initialized the SCTP-specific portion of the sock. + * The sock structure should already be zero-filled memory. + */ +SCTP_STATIC int sctp_init_sock(struct sock *sk) +{ + struct sctp_endpoint *ep; + struct sctp_sock *sp; + + SCTP_DEBUG_PRINTK("sctp_init_sock(sk: %p)\n", sk); + + sp = sctp_sk(sk); + + /* Initialize the SCTP per socket area. */ + switch (sk->sk_type) { + case SOCK_SEQPACKET: + sp->type = SCTP_SOCKET_UDP; + break; + case SOCK_STREAM: + sp->type = SCTP_SOCKET_TCP; + break; + default: + return -ESOCKTNOSUPPORT; + } + + /* Initialize default send parameters. These parameters can be + * modified with the SCTP_DEFAULT_SEND_PARAM socket option. + */ + sp->default_stream = 0; + sp->default_ppid = 0; + sp->default_flags = 0; + sp->default_context = 0; + sp->default_timetolive = 0; + + /* 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); + + /* 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); + + /* Initialize default association related parameters. These parameters + * can be modified with the SCTP_ASSOCINFO socket option. + */ + sp->assocparams.sasoc_asocmaxrxt = sctp_max_retrans_association; + 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); + + /* Initialize default event subscriptions. By default, all the + * 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; + + /* 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 by default. */ + sp->v4mapped = 1; + + /* Auto-close idle associations after the configured + * number of seconds. A value of 0 disables this + * feature. Configure through the SCTP_AUTOCLOSE socket option, + * for UDP-style sockets only. + */ + sp->autoclose = 0; + + /* User specified fragmentation limit. */ + sp->user_frag = 0; + + sp->adaption_ind = 0; + + sp->pf = sctp_get_pf_specific(sk->sk_family); + + /* Control variables for partial data delivery. */ + sp->pd_mode = 0; + skb_queue_head_init(&sp->pd_lobby); + + /* Create a per socket endpoint structure. Even if we + * change the data structure relationships, this may still + * be useful for storing pre-connect address information. + */ + ep = sctp_endpoint_new(sk, GFP_KERNEL); + if (!ep) + return -ENOMEM; + + sp->ep = ep; + sp->hmac = NULL; + + SCTP_DBG_OBJCNT_INC(sock); + return 0; +} + +/* Cleanup any SCTP per socket resources. */ +SCTP_STATIC int sctp_destroy_sock(struct sock *sk) +{ + struct sctp_endpoint *ep; + + SCTP_DEBUG_PRINTK("sctp_destroy_sock(sk: %p)\n", sk); + + /* Release our hold on the endpoint. */ + ep = sctp_sk(sk)->ep; + sctp_endpoint_free(ep); + + return 0; +} + +/* API 4.1.7 shutdown() - TCP Style Syntax + * int shutdown(int socket, int how); + * + * sd - the socket descriptor of the association to be closed. + * how - Specifies the type of shutdown. The values are + * as follows: + * SHUT_RD + * Disables further receive operations. No SCTP + * protocol action is taken. + * SHUT_WR + * Disables further send operations, and initiates + * the SCTP shutdown sequence. + * SHUT_RDWR + * Disables further send and receive operations + * and initiates the SCTP shutdown sequence. + */ +SCTP_STATIC void sctp_shutdown(struct sock *sk, int how) +{ + struct sctp_endpoint *ep; + struct sctp_association *asoc; + + if (!sctp_style(sk, TCP)) + return; + + if (how & SEND_SHUTDOWN) { + ep = sctp_sk(sk)->ep; + if (!list_empty(&ep->asocs)) { + asoc = list_entry(ep->asocs.next, + struct sctp_association, asocs); + sctp_primitive_SHUTDOWN(asoc, NULL); + } + } +} + +/* 7.2.1 Association Status (SCTP_STATUS) + + * Applications can retrieve current status information about an + * association, including association state, peer receiver window size, + * number of unacked data chunks, and number of data chunks pending + * receipt. This information is read-only. + */ +static int sctp_getsockopt_sctp_status(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_status status; + struct sctp_association *asoc = NULL; + struct sctp_transport *transport; + sctp_assoc_t associd; + int retval = 0; + + if (len != sizeof(status)) { + retval = -EINVAL; + goto out; + } + + if (copy_from_user(&status, optval, sizeof(status))) { + retval = -EFAULT; + goto out; + } + + associd = status.sstat_assoc_id; + asoc = sctp_id2assoc(sk, associd); + if (!asoc) { + retval = -EINVAL; + goto out; + } + + transport = asoc->peer.primary_path; + + status.sstat_assoc_id = sctp_assoc2id(asoc); + status.sstat_state = asoc->state; + status.sstat_rwnd = asoc->peer.rwnd; + status.sstat_unackdata = asoc->unack_data; + + status.sstat_penddata = sctp_tsnmap_pending(&asoc->peer.tsn_map); + status.sstat_instrms = asoc->c.sinit_max_instreams; + 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)); + /* 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_state = transport->active; + 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; + + if (put_user(len, optlen)) { + retval = -EFAULT; + goto out; + } + + SCTP_DEBUG_PRINTK("sctp_getsockopt_sctp_status(%d): %d %d %d\n", + len, status.sstat_state, status.sstat_rwnd, + status.sstat_assoc_id); + + if (copy_to_user(optval, &status, len)) { + retval = -EFAULT; + goto out; + } + +out: + return (retval); +} + + +/* 7.2.2 Peer Address Information (SCTP_GET_PEER_ADDR_INFO) + * + * Applications can retrieve information about a specific peer address + * of an association, including its reachability state, congestion + * window, and retransmission timer values. This information is + * read-only. + */ +static int sctp_getsockopt_peer_addr_info(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + struct sctp_paddrinfo pinfo; + struct sctp_transport *transport; + int retval = 0; + + if (len != sizeof(pinfo)) { + retval = -EINVAL; + goto out; + } + + if (copy_from_user(&pinfo, optval, sizeof(pinfo))) { + retval = -EFAULT; + goto out; + } + + transport = sctp_addr_id2transport(sk, &pinfo.spinfo_address, + pinfo.spinfo_assoc_id); + if (!transport) + return -EINVAL; + + pinfo.spinfo_assoc_id = sctp_assoc2id(transport->asoc); + pinfo.spinfo_state = transport->active; + pinfo.spinfo_cwnd = transport->cwnd; + pinfo.spinfo_srtt = transport->srtt; + pinfo.spinfo_rto = jiffies_to_msecs(transport->rto); + pinfo.spinfo_mtu = transport->pmtu; + + if (put_user(len, optlen)) { + retval = -EFAULT; + goto out; + } + + if (copy_to_user(optval, &pinfo, len)) { + retval = -EFAULT; + goto out; + } + +out: + return (retval); +} + +/* 7.1.12 Enable/Disable message fragmentation (SCTP_DISABLE_FRAGMENTS) + * + * This option is a on/off flag. If enabled no SCTP message + * fragmentation will be performed. Instead if a message being sent + * exceeds the current PMTU size, the message will NOT be sent and + * instead a error will be indicated to the user. + */ +static int sctp_getsockopt_disable_fragments(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)->disable_fragments == 1); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, len)) + return -EFAULT; + return 0; +} + +/* 7.1.15 Set notification and ancillary events (SCTP_EVENTS) + * + * This socket option is used to specify various notifications and + * ancillary data the user wishes to receive. + */ +static int sctp_getsockopt_events(struct sock *sk, int len, char __user *optval, + int __user *optlen) +{ + if (len != sizeof(struct sctp_event_subscribe)) + return -EINVAL; + if (copy_to_user(optval, &sctp_sk(sk)->subscribe, len)) + return -EFAULT; + return 0; +} + +/* 7.1.8 Automatic Close of associations (SCTP_AUTOCLOSE) + * + * This socket option is applicable to the UDP-style socket only. When + * set it will cause associations that are idle for more than the + * specified number of seconds to automatically close. An association + * being idle is defined an association that has NOT sent or received + * user data. The special value of '0' indicates that no automatic + * close of any associations should be performed. The option expects an + * integer defining the number of seconds of idle time before an + * association is closed. + */ +static int sctp_getsockopt_autoclose(struct sock *sk, int len, char __user *optval, int __user *optlen) +{ + /* Applicable to UDP-style socket only */ + if (sctp_style(sk, TCP)) + return -EOPNOTSUPP; + if (len != sizeof(int)) + return -EINVAL; + if (copy_to_user(optval, &sctp_sk(sk)->autoclose, len)) + return -EFAULT; + return 0; +} + +/* Helper routine to branch off an association to a new socket. */ +SCTP_STATIC int sctp_do_peeloff(struct sctp_association *asoc, + struct socket **sockp) +{ + struct sock *sk = asoc->base.sk; + struct socket *sock; + int err = 0; + + /* An association cannot be branched off from an already peeled-off + * socket, nor is this supported for tcp style sockets. + */ + if (!sctp_style(sk, UDP)) + return -EINVAL; + + /* Create a new socket. */ + err = sock_create(sk->sk_family, SOCK_SEQPACKET, IPPROTO_SCTP, &sock); + if (err < 0) + return err; + + /* Populate the fields of the newsk from the oldsk and migrate the + * asoc to the newsk. + */ + sctp_sock_migrate(sk, sock->sk, asoc, SCTP_SOCKET_UDP_HIGH_BANDWIDTH); + *sockp = sock; + + return err; +} + +static int sctp_getsockopt_peeloff(struct sock *sk, int len, char __user *optval, int __user *optlen) +{ + sctp_peeloff_arg_t peeloff; + struct socket *newsock; + int retval = 0; + struct sctp_association *asoc; + + if (len != sizeof(sctp_peeloff_arg_t)) + return -EINVAL; + if (copy_from_user(&peeloff, optval, len)) + return -EFAULT; + + asoc = sctp_id2assoc(sk, peeloff.associd); + if (!asoc) { + retval = -EINVAL; + goto out; + } + + SCTP_DEBUG_PRINTK("%s: sk: %p asoc: %p\n", __FUNCTION__, sk, asoc); + + retval = sctp_do_peeloff(asoc, &newsock); + if (retval < 0) + goto out; + + /* Map the socket to an unused fd that can be returned to the user. */ + retval = sock_map_fd(newsock); + if (retval < 0) { + sock_release(newsock); + goto out; + } + + SCTP_DEBUG_PRINTK("%s: sk: %p asoc: %p newsk: %p sd: %d\n", + __FUNCTION__, sk, asoc, newsock->sk, retval); + + /* Return the fd mapped to the new socket. */ + peeloff.sd = retval; + if (copy_to_user(optval, &peeloff, len)) + retval = -EFAULT; + +out: + return retval; +} + +/* 7.1.13 Peer Address Parameters (SCTP_PEER_ADDR_PARAMS) + * + * 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 following structure is used to access and modify an + * 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. + * 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. + * spp_pathmaxrxt - This contains the maximum number of + * retransmissions before this address shall be + * considered unreachable. + */ +static int sctp_getsockopt_peer_addr_params(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + struct sctp_paddrparams params; + struct sctp_transport *trans; + + 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 (!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; + } + + trans = sctp_addr_id2transport(sk, ¶ms.spp_address, + params.spp_assoc_id); + if (!trans) + 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); + + /* spp_pathmaxrxt contains the maximum number of retransmissions + * before this address shall be considered unreachable. + */ + params.spp_pathmaxrxt = trans->max_retrans; + +done: + if (copy_to_user(optval, ¶ms, len)) + return -EFAULT; + + if (put_user(len, optlen)) + return -EFAULT; + + return 0; +} + +/* 7.1.3 Initialization Parameters (SCTP_INITMSG) + * + * Applications can specify protocol parameters for the default association + * initialization. The option name argument to setsockopt() and getsockopt() + * is SCTP_INITMSG. + * + * Setting initialization parameters is effective only on an unconnected + * socket (for UDP-style sockets only future associations are effected + * by the change). With TCP-style sockets, this option is inherited by + * sockets derived from a listener socket. + */ +static int sctp_getsockopt_initmsg(struct sock *sk, int len, char __user *optval, int __user *optlen) +{ + if (len != sizeof(struct sctp_initmsg)) + return -EINVAL; + if (copy_to_user(optval, &sctp_sk(sk)->initmsg, len)) + return -EFAULT; + return 0; +} + +static int sctp_getsockopt_peer_addrs_num(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + sctp_assoc_t id; + struct sctp_association *asoc; + struct list_head *pos; + int cnt = 0; + + if (len != sizeof(sctp_assoc_t)) + return -EINVAL; + + if (copy_from_user(&id, optval, sizeof(sctp_assoc_t))) + return -EFAULT; + + /* For UDP-style sockets, id specifies the association to query. */ + asoc = sctp_id2assoc(sk, id); + if (!asoc) + return -EINVAL; + + list_for_each(pos, &asoc->peer.transport_addr_list) { + cnt ++; + } + + return cnt; +} + +static int sctp_getsockopt_peer_addrs(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + struct sctp_association *asoc; + struct list_head *pos; + int cnt = 0; + struct sctp_getaddrs getaddrs; + struct sctp_transport *from; + void __user *to; + union sctp_addr temp; + struct sctp_sock *sp = sctp_sk(sk); + int addrlen; + + if (len != sizeof(struct sctp_getaddrs)) + return -EINVAL; + + if (copy_from_user(&getaddrs, optval, sizeof(struct sctp_getaddrs))) + return -EFAULT; + + if (getaddrs.addr_num <= 0) return -EINVAL; + + /* For UDP-style sockets, id specifies the association to query. */ + asoc = sctp_id2assoc(sk, getaddrs.assoc_id); + if (!asoc) + return -EINVAL; + + to = (void __user *)getaddrs.addrs; + list_for_each(pos, &asoc->peer.transport_addr_list) { + from = list_entry(pos, struct sctp_transport, transports); + 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 ; + cnt ++; + if (cnt >= getaddrs.addr_num) break; + } + getaddrs.addr_num = cnt; + if (copy_to_user(optval, &getaddrs, sizeof(struct sctp_getaddrs))) + return -EFAULT; + + return 0; +} + +static int sctp_getsockopt_local_addrs_num(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + sctp_assoc_t id; + struct sctp_bind_addr *bp; + struct sctp_association *asoc; + struct list_head *pos; + struct sctp_sockaddr_entry *addr; + rwlock_t *addr_lock; + unsigned long flags; + int cnt = 0; + + if (len != sizeof(sctp_assoc_t)) + return -EINVAL; + + if (copy_from_user(&id, optval, sizeof(sctp_assoc_t))) + return -EFAULT; + + /* + * For UDP-style sockets, id specifies the association to query. + * If the id field is set to the value '0' then the locally bound + * addresses are returned without regard to any particular + * association. + */ + if (0 == id) { + bp = &sctp_sk(sk)->ep->base.bind_addr; + addr_lock = &sctp_sk(sk)->ep->base.addr_lock; + } else { + asoc = sctp_id2assoc(sk, id); + if (!asoc) + return -EINVAL; + bp = &asoc->base.bind_addr; + addr_lock = &asoc->base.addr_lock; + } + + sctp_read_lock(addr_lock); + + /* If the endpoint is bound to 0.0.0.0 or ::0, count the valid + * addresses from the global local address list. + */ + if (sctp_list_single_entry(&bp->address_list)) { + 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) { + addr = list_entry(pos, + struct sctp_sockaddr_entry, + list); + 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; + } + goto done; + } + + list_for_each(pos, &bp->address_list) { + cnt ++; + } + +done: + sctp_read_unlock(addr_lock); + return cnt; +} + +/* Helper function that copies local addresses to user and returns the number + * of addresses copied. + */ +static int sctp_copy_laddrs_to_user(struct sock *sk, __u16 port, int max_addrs, + void __user *to) +{ + struct list_head *pos; + 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) { + addr = list_entry(pos, struct sctp_sockaddr_entry, list); + 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; + } + to += addrlen; + cnt ++; + if (cnt >= max_addrs) break; + } + sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags); + + return cnt; +} + +static int sctp_getsockopt_local_addrs(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + struct sctp_bind_addr *bp; + struct sctp_association *asoc; + struct list_head *pos; + int cnt = 0; + struct sctp_getaddrs getaddrs; + struct sctp_sockaddr_entry *addr; + void __user *to; + union sctp_addr temp; + struct sctp_sock *sp = sctp_sk(sk); + int addrlen; + rwlock_t *addr_lock; + int err = 0; + + if (len != sizeof(struct sctp_getaddrs)) + return -EINVAL; + + if (copy_from_user(&getaddrs, optval, sizeof(struct sctp_getaddrs))) + return -EFAULT; + + if (getaddrs.addr_num <= 0) return -EINVAL; + /* + * For UDP-style sockets, id specifies the association to query. + * If the id field is set to the value '0' then the locally bound + * addresses are returned without regard to any particular + * association. + */ + if (0 == getaddrs.assoc_id) { + bp = &sctp_sk(sk)->ep->base.bind_addr; + addr_lock = &sctp_sk(sk)->ep->base.addr_lock; + } else { + asoc = sctp_id2assoc(sk, getaddrs.assoc_id); + if (!asoc) + return -EINVAL; + bp = &asoc->base.bind_addr; + addr_lock = &asoc->base.addr_lock; + } + + to = getaddrs.addrs; + + sctp_read_lock(addr_lock); + + /* If the endpoint is bound to 0.0.0.0 or ::0, get the valid + * addresses from the global local address list. + */ + if (sctp_list_single_entry(&bp->address_list)) { + 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, + getaddrs.addr_num, to); + if (cnt < 0) { + err = cnt; + goto unlock; + } + goto copy_getaddrs; + } + } + + 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; + cnt ++; + if (cnt >= getaddrs.addr_num) break; + } + +copy_getaddrs: + getaddrs.addr_num = cnt; + if (copy_to_user(optval, &getaddrs, sizeof(struct sctp_getaddrs))) + err = -EFAULT; + +unlock: + sctp_read_unlock(addr_lock); + return err; +} + +/* 7.1.10 Set Primary Address (SCTP_PRIMARY_ADDR) + * + * Requests that the local SCTP stack use the enclosed peer address as + * the association primary. The enclosed address must be one of the + * association peer's addresses. + */ +static int sctp_getsockopt_primary_addr(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + struct sctp_prim prim; + struct sctp_association *asoc; + struct sctp_sock *sp = sctp_sk(sk); + + if (len != sizeof(struct sctp_prim)) + return -EINVAL; + + if (copy_from_user(&prim, optval, sizeof(struct sctp_prim))) + return -EFAULT; + + asoc = sctp_id2assoc(sk, prim.ssp_assoc_id); + if (!asoc) + return -EINVAL; + + 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); + + sctp_get_pf_specific(sk->sk_family)->addr_v4map(sp, + (union sctp_addr *)&prim.ssp_addr); + + if (copy_to_user(optval, &prim, sizeof(struct sctp_prim))) + return -EFAULT; + + return 0; +} + +/* + * 7.1.11 Set Adaption Layer Indicator (SCTP_ADAPTION_LAYER) + * + * Requests that the local endpoint set the specified Adaption Layer + * Indication parameter for all future INIT and INIT-ACK exchanges. + */ +static int sctp_getsockopt_adaption_layer(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)->adaption_ind; + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, len)) + return -EFAULT; + return 0; +} + +/* + * + * 7.1.14 Set default send parameters (SCTP_DEFAULT_SEND_PARAM) + * + * Applications that wish to use the sendto() system call may wish to + * specify a default set of parameters that would normally be supplied + * through the inclusion of ancillary data. This socket option allows + * such an application to set the default sctp_sndrcvinfo structure. + + + * The application that wishes to use this socket option simply passes + * in to this call the sctp_sndrcvinfo structure defined in Section + * 5.2.2) The input parameters accepted by this call include + * sinfo_stream, sinfo_flags, sinfo_ppid, sinfo_context, + * sinfo_timetolive. The user must provide the sinfo_assoc_id field in + * to this call if the caller is using the UDP model. + * + * For getsockopt, it get the default sctp_sndrcvinfo structure. + */ +static int sctp_getsockopt_default_send_param(struct sock *sk, + int len, char __user *optval, + int __user *optlen) +{ + struct sctp_sndrcvinfo info; + struct sctp_association *asoc; + struct sctp_sock *sp = sctp_sk(sk); + + if (len != sizeof(struct sctp_sndrcvinfo)) + return -EINVAL; + if (copy_from_user(&info, optval, sizeof(struct sctp_sndrcvinfo))) + return -EFAULT; + + asoc = sctp_id2assoc(sk, info.sinfo_assoc_id); + if (!asoc && info.sinfo_assoc_id && sctp_style(sk, UDP)) + return -EINVAL; + + if (asoc) { + info.sinfo_stream = asoc->default_stream; + info.sinfo_flags = asoc->default_flags; + info.sinfo_ppid = asoc->default_ppid; + info.sinfo_context = asoc->default_context; + info.sinfo_timetolive = asoc->default_timetolive; + } else { + info.sinfo_stream = sp->default_stream; + info.sinfo_flags = sp->default_flags; + info.sinfo_ppid = sp->default_ppid; + info.sinfo_context = sp->default_context; + info.sinfo_timetolive = sp->default_timetolive; + } + + if (copy_to_user(optval, &info, sizeof(struct sctp_sndrcvinfo))) + return -EFAULT; + + return 0; +} + +/* + * + * 7.1.5 SCTP_NODELAY + * + * Turn on/off any Nagle-like algorithm. This means that packets are + * generally sent as soon as possible and no unnecessary delays are + * introduced, at the cost of more packets in the network. Expects an + * integer boolean flag. + */ + +static int sctp_getsockopt_nodelay(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)->nodelay == 1); + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, len)) + return -EFAULT; + return 0; +} + +/* + * + * 7.1.1 SCTP_RTOINFO + * + * The protocol parameters used to initialize and bound retransmission + * timeout (RTO) are tunable. sctp_rtoinfo structure is used to access + * and modify these parameters. + * All parameters are time values, in milliseconds. A value of 0, when + * modifying the parameters, indicates that the current value should not + * be changed. + * + */ +static int sctp_getsockopt_rtoinfo(struct sock *sk, int len, + char __user *optval, + int __user *optlen) { + struct sctp_rtoinfo rtoinfo; + struct sctp_association *asoc; + + if (len != sizeof (struct sctp_rtoinfo)) + return -EINVAL; + + if (copy_from_user(&rtoinfo, optval, sizeof (struct sctp_rtoinfo))) + return -EFAULT; + + asoc = sctp_id2assoc(sk, rtoinfo.srto_assoc_id); + + if (!asoc && rtoinfo.srto_assoc_id && sctp_style(sk, UDP)) + return -EINVAL; + + /* Values corresponding to the specific association. */ + if (asoc) { + rtoinfo.srto_initial = jiffies_to_msecs(asoc->rto_initial); + rtoinfo.srto_max = jiffies_to_msecs(asoc->rto_max); + rtoinfo.srto_min = jiffies_to_msecs(asoc->rto_min); + } else { + /* Values corresponding to the endpoint. */ + struct sctp_sock *sp = sctp_sk(sk); + + rtoinfo.srto_initial = sp->rtoinfo.srto_initial; + rtoinfo.srto_max = sp->rtoinfo.srto_max; + rtoinfo.srto_min = sp->rtoinfo.srto_min; + } + + if (put_user(len, optlen)) + return -EFAULT; + + if (copy_to_user(optval, &rtoinfo, len)) + return -EFAULT; + + return 0; +} + +/* + * + * 7.1.2 SCTP_ASSOCINFO + * + * This option is used to tune the 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. + * See [SCTP] for more information. + * + */ +static int sctp_getsockopt_associnfo(struct sock *sk, int len, + char __user *optval, + int __user *optlen) +{ + + struct sctp_assocparams assocparams; + struct sctp_association *asoc; + struct list_head *pos; + int cnt = 0; + + if (len != sizeof (struct sctp_assocparams)) + return -EINVAL; + + if (copy_from_user(&assocparams, optval, + sizeof (struct sctp_assocparams))) + return -EFAULT; + + asoc = sctp_id2assoc(sk, assocparams.sasoc_assoc_id); + + if (!asoc && assocparams.sasoc_assoc_id && sctp_style(sk, UDP)) + return -EINVAL; + + /* Values correspoinding to the specific association */ + if (assocparams.sasoc_assoc_id != 0) { + assocparams.sasoc_asocmaxrxt = asoc->max_retrans; + assocparams.sasoc_peer_rwnd = asoc->peer.rwnd; + assocparams.sasoc_local_rwnd = asoc->a_rwnd; + assocparams.sasoc_cookie_life = (asoc->cookie_life.tv_sec + * 1000) + + (asoc->cookie_life.tv_usec + / 1000); + + list_for_each(pos, &asoc->peer.transport_addr_list) { + cnt ++; + } + + assocparams.sasoc_number_peer_destinations = cnt; + } else { + /* Values corresponding to the endpoint */ + struct sctp_sock *sp = sctp_sk(sk); + + assocparams.sasoc_asocmaxrxt = sp->assocparams.sasoc_asocmaxrxt; + assocparams.sasoc_peer_rwnd = sp->assocparams.sasoc_peer_rwnd; + assocparams.sasoc_local_rwnd = sp->assocparams.sasoc_local_rwnd; + assocparams.sasoc_cookie_life = + sp->assocparams.sasoc_cookie_life; + assocparams.sasoc_number_peer_destinations = + sp->assocparams. + sasoc_number_peer_destinations; + } + + if (put_user(len, optlen)) + return -EFAULT; + + if (copy_to_user(optval, &assocparams, len)) + return -EFAULT; + + return 0; +} + +/* + * 7.1.16 Set/clear IPv4 mapped addresses (SCTP_I_WANT_MAPPED_V4_ADDR) + * + * This socket option is a boolean flag which turns on or off mapped V4 + * addresses. If this option is turned on and the socket is type + * PF_INET6, then IPv4 addresses will be mapped to V6 representation. + * If this option is turned off, then no mapping will be done of V4 + * addresses and a user will receive both PF_INET6 and PF_INET type + * addresses on the socket. + */ +static int sctp_getsockopt_mappedv4(struct sock *sk, int len, + char __user *optval, int __user *optlen) +{ + int val; + struct sctp_sock *sp = sctp_sk(sk); + + if (len < sizeof(int)) + return -EINVAL; + + len = sizeof(int); + val = sp->v4mapped; + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, 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 + * SCTP chunk. If a message is larger than this size it will be + * fragmented by SCTP into the specified size. Note that the underlying + * SCTP implementation may fragment into smaller sized chunks when the + * PMTU of the underlying association is smaller than the value set by + * the user. + */ +static int sctp_getsockopt_maxseg(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)->user_frag; + if (put_user(len, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, len)) + return -EFAULT; + + return 0; +} + +SCTP_STATIC int sctp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) +{ + int retval = 0; + int len; + + SCTP_DEBUG_PRINTK("sctp_getsockopt(sk: %p, ...)\n", sk); + + /* I can hardly begin to describe how wrong this is. This is + * so broken as to be worse than useless. The API draft + * REALLY is NOT helpful here... I am not convinced that the + * semantics of getsockopt() with a level OTHER THAN SOL_SCTP + * are at all well-founded. + */ + if (level != SOL_SCTP) { + struct sctp_af *af = sctp_sk(sk)->pf->af; + + retval = af->getsockopt(sk, level, optname, optval, optlen); + return retval; + } + + if (get_user(len, optlen)) + return -EFAULT; + + sctp_lock_sock(sk); + + switch (optname) { + case SCTP_STATUS: + retval = sctp_getsockopt_sctp_status(sk, len, optval, optlen); + break; + case SCTP_DISABLE_FRAGMENTS: + retval = sctp_getsockopt_disable_fragments(sk, len, optval, + optlen); + break; + case SCTP_EVENTS: + retval = sctp_getsockopt_events(sk, len, optval, optlen); + break; + case SCTP_AUTOCLOSE: + retval = sctp_getsockopt_autoclose(sk, len, optval, optlen); + break; + case SCTP_SOCKOPT_PEELOFF: + retval = sctp_getsockopt_peeloff(sk, len, optval, optlen); + break; + case SCTP_PEER_ADDR_PARAMS: + retval = sctp_getsockopt_peer_addr_params(sk, len, optval, + optlen); + break; + case SCTP_INITMSG: + retval = sctp_getsockopt_initmsg(sk, len, optval, optlen); + break; + case SCTP_GET_PEER_ADDRS_NUM: + retval = sctp_getsockopt_peer_addrs_num(sk, len, optval, + optlen); + break; + case SCTP_GET_LOCAL_ADDRS_NUM: + retval = sctp_getsockopt_local_addrs_num(sk, len, optval, + optlen); + break; + case SCTP_GET_PEER_ADDRS: + retval = sctp_getsockopt_peer_addrs(sk, len, optval, + optlen); + break; + case SCTP_GET_LOCAL_ADDRS: + retval = sctp_getsockopt_local_addrs(sk, len, optval, + optlen); + break; + case SCTP_DEFAULT_SEND_PARAM: + retval = sctp_getsockopt_default_send_param(sk, len, + optval, optlen); + break; + case SCTP_PRIMARY_ADDR: + retval = sctp_getsockopt_primary_addr(sk, len, optval, optlen); + break; + case SCTP_NODELAY: + retval = sctp_getsockopt_nodelay(sk, len, optval, optlen); + break; + case SCTP_RTOINFO: + retval = sctp_getsockopt_rtoinfo(sk, len, optval, optlen); + break; + case SCTP_ASSOCINFO: + retval = sctp_getsockopt_associnfo(sk, len, optval, optlen); + break; + case SCTP_I_WANT_MAPPED_V4_ADDR: + retval = sctp_getsockopt_mappedv4(sk, len, optval, optlen); + break; + case SCTP_MAXSEG: + retval = sctp_getsockopt_maxseg(sk, len, optval, optlen); + break; + case SCTP_GET_PEER_ADDR_INFO: + retval = sctp_getsockopt_peer_addr_info(sk, len, optval, + optlen); + break; + case SCTP_ADAPTION_LAYER: + retval = sctp_getsockopt_adaption_layer(sk, len, optval, + optlen); + break; + default: + retval = -ENOPROTOOPT; + break; + }; + + sctp_release_sock(sk); + return retval; +} + +static void sctp_hash(struct sock *sk) +{ + /* STUB */ +} + +static void sctp_unhash(struct sock *sk) +{ + /* STUB */ +} + +/* Check if port is acceptable. Possibly find first available port. + * + * The port hash table (contained in the 'global' SCTP protocol storage + * returned by struct sctp_protocol *sctp_get_protocol()). The hash + * table is an array of 4096 lists (sctp_bind_hashbucket). Each + * list (the list number is the port number hashed out, so as you + * would expect from a hash function, all the ports in a given list have + * such a number that hashes out to the same list number; you were + * expecting that, right?); so each list has a set of ports, with a + * link to the socket (struct sock) that uses it, the port number and + * a fastreuse flag (FIXME: NPI ipg). + */ +static struct sctp_bind_bucket *sctp_bucket_create( + struct sctp_bind_hashbucket *head, unsigned short snum); + +static long sctp_get_port_local(struct sock *sk, union sctp_addr *addr) +{ + struct sctp_bind_hashbucket *head; /* hash list */ + struct sctp_bind_bucket *pp; /* hash list port iterator */ + 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; + + SCTP_DEBUG_PRINTK("sctp_get_port() begins, snum=%d\n", snum); + sctp_local_bh_disable(); + + if (snum == 0) { + /* Search for an available port. + * + * 'sctp_port_rover' was the last port assigned, so + * we start to search from 'sctp_port_rover + + * 1'. What we do is first check if port 'rover' is + * already in the hash table; if not, we use that; if + * it is, we try next. + */ + int low = sysctl_local_port_range[0]; + int high = sysctl_local_port_range[1]; + int remaining = (high - low) + 1; + int rover; + int index; + + sctp_spin_lock(&sctp_port_alloc_lock); + rover = sctp_port_rover; + do { + rover++; + if ((rover < low) || (rover > high)) + rover = low; + index = sctp_phashfn(rover); + head = &sctp_port_hashtable[index]; + sctp_spin_lock(&head->lock); + for (pp = head->chain; pp; pp = pp->next) + if (pp->port == rover) + goto next; + break; + next: + sctp_spin_unlock(&head->lock); + } while (--remaining > 0); + sctp_port_rover = rover; + sctp_spin_unlock(&sctp_port_alloc_lock); + + /* Exhausted local port range during search? */ + ret = 1; + if (remaining <= 0) + goto fail; + + /* OK, here is the one we will use. HEAD (the port + * hash table list entry) is non-NULL and we hold it's + * mutex. + */ + snum = rover; + } else { + /* We are given an specific port number; we verify + * that it is not being used. If it is used, we will + * exahust the search in the hash list corresponding + * to the port number (snum) - we detect that with the + * port iterator, pp being NULL. + */ + head = &sctp_port_hashtable[sctp_phashfn(snum)]; + sctp_spin_lock(&head->lock); + for (pp = head->chain; pp; pp = pp->next) { + if (pp->port == snum) + goto pp_found; + } + } + pp = NULL; + goto pp_not_found; +pp_found: + if (!hlist_empty(&pp->owner)) { + /* We had a port hash table hit - there is an + * available port (pp != NULL) and it is being + * used by other socket (pp->owner not empty); that other + * socket is going to be sk2. + */ + int reuse = sk->sk_reuse; + struct sock *sk2; + struct hlist_node *node; + + SCTP_DEBUG_PRINTK("sctp_get_port() found a possible match\n"); + if (pp->fastreuse && sk->sk_reuse) + goto success; + + /* Run through the list of sockets bound to the port + * (pp->port) [via the pointers bind_next and + * bind_pprev in the struct sock *sk2 (pp->sk)]. On each one, + * we get the endpoint they describe and run through + * the endpoint's list of IP (v4 or v6) addresses, + * comparing each of the addresses with the address of + * the socket sk. If we find a match, then that means + * that this port/socket (sk) combination are already + * in an endpoint. + */ + sk_for_each_bound(sk2, node, &pp->owner) { + struct sctp_endpoint *ep2; + ep2 = sctp_sk(sk2)->ep; + + if (reuse && sk2->sk_reuse) + continue; + + if (sctp_bind_addr_match(&ep2->base.bind_addr, addr, + sctp_sk(sk))) { + ret = (long)sk2; + goto fail_unlock; + } + } + SCTP_DEBUG_PRINTK("sctp_get_port(): Found a match\n"); + } +pp_not_found: + /* If there was a hash table miss, create a new port. */ + ret = 1; + if (!pp && !(pp = sctp_bucket_create(head, snum))) + goto fail_unlock; + + /* In either case (hit or miss), make sure fastreuse is 1 only + * 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) + pp->fastreuse = 0; + + /* We are set, so fill up all the data in the hash table + * entry, tie the socket list information with the rest of the + * sockets FIXME: Blurry, NPI (ipg). + */ +success: + inet_sk(sk)->num = snum; + if (!sctp_sk(sk)->bind_hash) { + sk_add_bind_node(sk, &pp->owner); + sctp_sk(sk)->bind_hash = pp; + } + ret = 0; + +fail_unlock: + sctp_spin_unlock(&head->lock); + +fail: + sctp_local_bh_enable(); + addr->v4.sin_port = htons(addr->v4.sin_port); + return ret; +} + +/* Assign a 'snum' port to the socket. If snum == 0, an ephemeral + * port is requested. + */ +static int sctp_get_port(struct sock *sk, unsigned short snum) +{ + long ret; + union sctp_addr addr; + struct sctp_af *af = sctp_sk(sk)->pf->af; + + /* Set up a dummy address struct from the sk. */ + af->from_sk(&addr, sk); + addr.v4.sin_port = htons(snum); + + /* Note: sk->sk_num gets filled in if ephemeral port request. */ + ret = sctp_get_port_local(sk, &addr); + + return (ret ? 1 : 0); +} + +/* + * 3.1.3 listen() - UDP Style Syntax + * + * By default, new associations are not accepted for UDP style sockets. + * An application uses listen() to mark a socket as being able to + * accept new associations. + */ +SCTP_STATIC int sctp_seqpacket_listen(struct sock *sk, int backlog) +{ + struct sctp_sock *sp = sctp_sk(sk); + struct sctp_endpoint *ep = sp->ep; + + /* Only UDP style sockets that are not peeled off are allowed to + * listen(). + */ + if (!sctp_style(sk, UDP)) + return -EINVAL; + + /* If backlog is zero, disable listening. */ + 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 + * picks an ephemeral port and will choose an address set equivalent + * to binding with a wildcard address. + * + * This is not currently spelled out in the SCTP sockets + * extensions draft, but follows the practice as seen in TCP + * sockets. + */ + if (!ep->base.bind_addr.port) { + if (sctp_autobind(sk)) + return -EAGAIN; + } + sk->sk_state = SCTP_SS_LISTENING; + sctp_hash_endpoint(ep); + return 0; +} + +/* + * 4.1.3 listen() - TCP Style Syntax + * + * Applications uses listen() to ready the SCTP endpoint for accepting + * inbound associations. + */ +SCTP_STATIC int sctp_stream_listen(struct sock *sk, int backlog) +{ + struct sctp_sock *sp = sctp_sk(sk); + struct sctp_endpoint *ep = sp->ep; + + /* If backlog is zero, disable listening. */ + if (!backlog) { + if (sctp_sstate(sk, CLOSED)) + return 0; + + sctp_unhash_endpoint(ep); + sk->sk_state = SCTP_SS_CLOSED; + } + + 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 + * picks an ephemeral port and will choose an address set equivalent + * to binding with a wildcard address. + * + * This is not currently spelled out in the SCTP sockets + * extensions draft, but follows the practice as seen in TCP + * sockets. + */ + if (!ep->base.bind_addr.port) { + if (sctp_autobind(sk)) + return -EAGAIN; + } + sk->sk_state = SCTP_SS_LISTENING; + sk->sk_max_ack_backlog = backlog; + sctp_hash_endpoint(ep); + return 0; +} + +/* + * Move a socket to LISTENING state. + */ +int sctp_inet_listen(struct socket *sock, int backlog) +{ + struct sock *sk = sock->sk; + struct crypto_tfm *tfm=NULL; + int err = -EINVAL; + + if (unlikely(backlog < 0)) + goto out; + + sctp_lock_sock(sk); + + if (sock->state != SS_UNCONNECTED) + goto out; + + /* Allocate HMAC for generating cookie. */ + if (sctp_hmac_alg) { + tfm = sctp_crypto_alloc_tfm(sctp_hmac_alg, 0); + if (!tfm) { + err = -ENOSYS; + goto out; + } + } + + switch (sock->type) { + case SOCK_SEQPACKET: + err = sctp_seqpacket_listen(sk, backlog); + break; + case SOCK_STREAM: + err = sctp_stream_listen(sk, backlog); + break; + default: + break; + }; + if (err) + goto cleanup; + + /* Store away the transform reference. */ + sctp_sk(sk)->hmac = tfm; +out: + sctp_release_sock(sk); + return err; +cleanup: + if (tfm) + sctp_crypto_free_tfm(tfm); + goto out; +} + +/* + * This function is done by modeling the current datagram_poll() and the + * 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 + * in this place. We assume that we don't need locks either until proven + * otherwise. + * + * Another thing to note is that we include the Async I/O support + * here, again, by modeling the current TCP/UDP code. We don't have + * a good way to test with it yet. + */ +unsigned int sctp_poll(struct file *file, struct socket *sock, poll_table *wait) +{ + struct sock *sk = sock->sk; + struct sctp_sock *sp = sctp_sk(sk); + unsigned int mask; + + poll_wait(file, sk->sk_sleep, wait); + + /* A TCP-style listening socket becomes readable when the accept queue + * is not empty. + */ + if (sctp_style(sk, TCP) && sctp_sstate(sk, LISTENING)) + return (!list_empty(&sp->ep->asocs)) ? + (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 == SHUTDOWN_MASK) + mask |= POLLHUP; + + /* Is it readable? Reconsider this code with TCP-style support. */ + if (!skb_queue_empty(&sk->sk_receive_queue) || + (sk->sk_shutdown & RCV_SHUTDOWN)) + mask |= POLLIN | POLLRDNORM; + + /* The association is either gone or not ready. */ + if (!sctp_style(sk, UDP) && sctp_sstate(sk, CLOSED)) + return mask; + + /* Is it writable? */ + if (sctp_writeable(sk)) { + mask |= POLLOUT | POLLWRNORM; + } else { + set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags); + /* + * Since the socket is not locked, the buffer + * might be made available after the writeable check and + * before the bit is set. This could cause a lost I/O + * signal. tcp_poll() has a race breaker for this race + * condition. Based on their implementation, we put + * in the following code to cover it as well. + */ + if (sctp_writeable(sk)) + mask |= POLLOUT | POLLWRNORM; + } + return mask; +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +static struct sctp_bind_bucket *sctp_bucket_create( + struct sctp_bind_hashbucket *head, unsigned short snum) +{ + struct sctp_bind_bucket *pp; + + pp = kmem_cache_alloc(sctp_bucket_cachep, SLAB_ATOMIC); + SCTP_DBG_OBJCNT_INC(bind_bucket); + if (pp) { + pp->port = snum; + pp->fastreuse = 0; + INIT_HLIST_HEAD(&pp->owner); + if ((pp->next = head->chain) != NULL) + pp->next->pprev = &pp->next; + head->chain = pp; + pp->pprev = &head->chain; + } + return pp; +} + +/* 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->next) + pp->next->pprev = pp->pprev; + *(pp->pprev) = pp->next; + kmem_cache_free(sctp_bucket_cachep, pp); + SCTP_DBG_OBJCNT_DEC(bind_bucket); + } +} + +/* Release this socket's reference to a local port. */ +static inline void __sctp_put_port(struct sock *sk) +{ + struct sctp_bind_hashbucket *head = + &sctp_port_hashtable[sctp_phashfn(inet_sk(sk)->num)]; + struct sctp_bind_bucket *pp; + + sctp_spin_lock(&head->lock); + pp = sctp_sk(sk)->bind_hash; + __sk_del_bind_node(sk); + sctp_sk(sk)->bind_hash = NULL; + inet_sk(sk)->num = 0; + sctp_bucket_destroy(pp); + sctp_spin_unlock(&head->lock); +} + +void sctp_put_port(struct sock *sk) +{ + sctp_local_bh_disable(); + __sctp_put_port(sk); + sctp_local_bh_enable(); +} + +/* + * The system picks an ephemeral port and choose an address set equivalent + * to binding with a wildcard address. + * One of those addresses will be the primary address for the association. + * This automatically enables the multihoming capability of SCTP. + */ +static int sctp_autobind(struct sock *sk) +{ + union sctp_addr autoaddr; + struct sctp_af *af; + unsigned short port; + + /* Initialize a local sockaddr structure to INADDR_ANY. */ + af = sctp_sk(sk)->pf->af; + + port = htons(inet_sk(sk)->num); + af->inaddr_any(&autoaddr, port); + + return sctp_do_bind(sk, &autoaddr, af->sockaddr_len); +} + +/* Parse out IPPROTO_SCTP CMSG headers. Perform only minimal validation. + * + * From RFC 2292 + * 4.2 The cmsghdr Structure * + * + * When ancillary data is sent or received, any number of ancillary data + * objects can be specified by the msg_control and msg_controllen members of + * the msghdr structure, because each object is preceded by + * a cmsghdr structure defining the object's length (the cmsg_len member). + * Historically Berkeley-derived implementations have passed only one object + * at a time, but this API allows multiple objects to be + * passed in a single call to sendmsg() or recvmsg(). The following example + * shows two ancillary data objects in a control buffer. + * + * |<--------------------------- msg_controllen -------------------------->| + * | | + * + * |<----- ancillary data object ----->|<----- ancillary data object ----->| + * + * |<---------- CMSG_SPACE() --------->|<---------- CMSG_SPACE() --------->| + * | | | + * + * |<---------- cmsg_len ---------->| |<--------- cmsg_len ----------->| | + * + * |<--------- CMSG_LEN() --------->| |<-------- CMSG_LEN() ---------->| | + * | | | | | + * + * +-----+-----+-----+--+-----------+--+-----+-----+-----+--+-----------+--+ + * |cmsg_|cmsg_|cmsg_|XX| |XX|cmsg_|cmsg_|cmsg_|XX| |XX| + * + * |len |level|type |XX|cmsg_data[]|XX|len |level|type |XX|cmsg_data[]|XX| + * + * +-----+-----+-----+--+-----------+--+-----+-----+-----+--+-----------+--+ + * ^ + * | + * + * msg_control + * points here + */ +SCTP_STATIC int sctp_msghdr_parse(const struct msghdr *msg, + sctp_cmsgs_t *cmsgs) +{ + struct cmsghdr *cmsg; + + for (cmsg = CMSG_FIRSTHDR(msg); + cmsg != NULL; + cmsg = CMSG_NXTHDR((struct msghdr*)msg, cmsg)) { + if (!CMSG_OK(msg, cmsg)) + return -EINVAL; + + /* Should we parse this header or ignore? */ + if (cmsg->cmsg_level != IPPROTO_SCTP) + continue; + + /* Strictly check lengths following example in SCM code. */ + switch (cmsg->cmsg_type) { + case SCTP_INIT: + /* SCTP Socket API Extension + * 5.2.1 SCTP Initiation Structure (SCTP_INIT) + * + * This cmsghdr structure provides information for + * initializing new SCTP associations with sendmsg(). + * The SCTP_INITMSG socket option uses this same data + * structure. This structure is not used for + * recvmsg(). + * + * cmsg_level cmsg_type cmsg_data[] + * ------------ ------------ ---------------------- + * IPPROTO_SCTP SCTP_INIT struct sctp_initmsg + */ + if (cmsg->cmsg_len != + CMSG_LEN(sizeof(struct sctp_initmsg))) + return -EINVAL; + cmsgs->init = (struct sctp_initmsg *)CMSG_DATA(cmsg); + break; + + case SCTP_SNDRCV: + /* SCTP Socket API Extension + * 5.2.2 SCTP Header Information Structure(SCTP_SNDRCV) + * + * This cmsghdr structure specifies SCTP options for + * sendmsg() and describes SCTP header information + * about a received message through recvmsg(). + * + * cmsg_level cmsg_type cmsg_data[] + * ------------ ------------ ---------------------- + * IPPROTO_SCTP SCTP_SNDRCV struct sctp_sndrcvinfo + */ + if (cmsg->cmsg_len != + CMSG_LEN(sizeof(struct sctp_sndrcvinfo))) + return -EINVAL; + + cmsgs->info = + (struct sctp_sndrcvinfo *)CMSG_DATA(cmsg); + + /* Minimally, validate the sinfo_flags. */ + if (cmsgs->info->sinfo_flags & + ~(MSG_UNORDERED | MSG_ADDR_OVER | + MSG_ABORT | MSG_EOF)) + return -EINVAL; + break; + + default: + return -EINVAL; + }; + } + return 0; +} + +/* + * Wait for a packet.. + * Note: This function is the same function as in core/datagram.c + * with a few modifications to make lksctp work. + */ +static int sctp_wait_for_packet(struct sock * sk, int *err, long *timeo_p) +{ + int error; + DEFINE_WAIT(wait); + + prepare_to_wait_exclusive(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); + + /* Socket errors? */ + error = sock_error(sk); + if (error) + goto out; + + if (!skb_queue_empty(&sk->sk_receive_queue)) + goto ready; + + /* Socket shut down? */ + if (sk->sk_shutdown & RCV_SHUTDOWN) + goto out; + + /* Sequenced packets can come disconnected. If so we report the + * problem. + */ + error = -ENOTCONN; + + /* Is there a good reason to think that we may receive some data? */ + if (list_empty(&sctp_sk(sk)->ep->asocs) && !sctp_sstate(sk, LISTENING)) + goto out; + + /* Handle signals. */ + if (signal_pending(current)) + goto interrupted; + + /* Let another process have a go. Since we are going to sleep + * anyway. Note: This may cause odd behaviors if the message + * does not fit in the user's buffer, but this seems to be the + * only way to honor MSG_DONTWAIT realistically. + */ + sctp_release_sock(sk); + *timeo_p = schedule_timeout(*timeo_p); + sctp_lock_sock(sk); + +ready: + finish_wait(sk->sk_sleep, &wait); + return 0; + +interrupted: + error = sock_intr_errno(*timeo_p); + +out: + finish_wait(sk->sk_sleep, &wait); + *err = error; + return error; +} + +/* Receive a datagram. + * Note: This is pretty much the same routine as in core/datagram.c + * with a few changes to make lksctp work. + */ +static struct sk_buff *sctp_skb_recv_datagram(struct sock *sk, int flags, + int noblock, int *err) +{ + int error; + 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", + timeo, MAX_SCHEDULE_TIMEOUT); + + do { + /* Again only user level code calls this function, + * so nothing interrupt level + * will suddenly eat the receive_queue. + * + * Look at current nfs client by the way... + * However, this function was corrent in any case. 8) + */ + if (flags & MSG_PEEK) { + unsigned long cpu_flags; + + sctp_spin_lock_irqsave(&sk->sk_receive_queue.lock, + cpu_flags); + skb = skb_peek(&sk->sk_receive_queue); + if (skb) + atomic_inc(&skb->users); + sctp_spin_unlock_irqrestore(&sk->sk_receive_queue.lock, + cpu_flags); + } else { + skb = skb_dequeue(&sk->sk_receive_queue); + } + + if (skb) + return skb; + + if (sk->sk_shutdown & RCV_SHUTDOWN) + break; + + /* User doesn't want to wait. */ + error = -EAGAIN; + if (!timeo) + goto no_packet; + } while (sctp_wait_for_packet(sk, err, &timeo) == 0); + + return NULL; + +no_packet: + *err = error; + return NULL; +} + +/* If sndbuf has changed, wake up per association sndbuf waiters. */ +static void __sctp_write_space(struct sctp_association *asoc) +{ + struct sock *sk = asoc->base.sk; + struct socket *sock = sk->sk_socket; + + if ((sctp_wspace(asoc) > 0) && sock) { + if (waitqueue_active(&asoc->wait)) + wake_up_interruptible(&asoc->wait); + + if (sctp_writeable(sk)) { + if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) + wake_up_interruptible(sk->sk_sleep); + + /* Note that we try to include the Async I/O support + * here by modeling from the current TCP/UDP code. + * We have not tested with it yet. + */ + if (sock->fasync_list && + !(sk->sk_shutdown & SEND_SHUTDOWN)) + sock_wake_async(sock, 2, POLL_OUT); + } + } +} + +/* Do accounting for the sndbuf space. + * Decrement the used sndbuf space of the corresponding association by the + * data size which was just transmitted(freed). + */ +static void sctp_wfree(struct sk_buff *skb) +{ + struct sctp_association *asoc; + struct sctp_chunk *chunk; + struct sock *sk; + + /* Get the saved chunk pointer. */ + chunk = *((struct sctp_chunk **)(skb->cb)); + asoc = chunk->asoc; + sk = asoc->base.sk; + asoc->sndbuf_used -= SCTP_DATA_SNDSIZE(chunk); + sk->sk_wmem_queued -= SCTP_DATA_SNDSIZE(chunk); + __sctp_write_space(asoc); + + sctp_association_put(asoc); +} + +/* 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) +{ + struct sock *sk = asoc->base.sk; + int err = 0; + long current_timeo = *timeo_p; + DEFINE_WAIT(wait); + + SCTP_DEBUG_PRINTK("wait_for_sndbuf: asoc=%p, timeo=%ld, msg_len=%zu\n", + asoc, (long)(*timeo_p), msg_len); + + /* Increment the association's refcnt. */ + sctp_association_hold(asoc); + + /* Wait on the association specific sndbuf space. */ + for (;;) { + prepare_to_wait_exclusive(&asoc->wait, &wait, + TASK_INTERRUPTIBLE); + if (!*timeo_p) + goto do_nonblock; + if (sk->sk_err || asoc->state >= SCTP_STATE_SHUTDOWN_PENDING || + asoc->base.dead) + goto do_error; + if (signal_pending(current)) + goto do_interrupted; + if (msg_len <= sctp_wspace(asoc)) + break; + + /* Let another process have a go. Since we are going + * to sleep anyway. + */ + sctp_release_sock(sk); + current_timeo = schedule_timeout(current_timeo); + sctp_lock_sock(sk); + + *timeo_p = current_timeo; + } + +out: + finish_wait(&asoc->wait, &wait); + + /* Release the association's refcnt. */ + sctp_association_put(asoc); + + return err; + +do_error: + err = -EPIPE; + goto out; + +do_interrupted: + err = sock_intr_errno(*timeo_p); + goto out; + +do_nonblock: + err = -EAGAIN; + goto out; +} + +/* If socket sndbuf has changed, wake up all per association waiters. */ +void sctp_write_space(struct sock *sk) +{ + struct sctp_association *asoc; + struct list_head *pos; + + /* Wake up the tasks in each wait queue. */ + list_for_each(pos, &((sctp_sk(sk))->ep->asocs)) { + asoc = list_entry(pos, struct sctp_association, asocs); + __sctp_write_space(asoc); + } +} + +/* Is there any sndbuf space available on the socket? + * + * Note that wmem_queued 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 + * a premature "unwriteable" is better than an accidental "writeable" which + * would cause an unwanted block under certain circumstances. For the 1-1 + * UDP-style sockets or TCP-style sockets, this code should work. + * - Daisy + */ +static int sctp_writeable(struct sock *sk) +{ + int amt = 0; + + amt = sk->sk_sndbuf - sk->sk_wmem_queued; + if (amt < 0) + amt = 0; + return amt; +} + +/* Wait for an association to go into ESTABLISHED state. If timeout is 0, + * returns immediately with EINPROGRESS. + */ +static int sctp_wait_for_connect(struct sctp_association *asoc, long *timeo_p) +{ + struct sock *sk = asoc->base.sk; + int err = 0; + long current_timeo = *timeo_p; + DEFINE_WAIT(wait); + + SCTP_DEBUG_PRINTK("%s: asoc=%p, timeo=%ld\n", __FUNCTION__, asoc, + (long)(*timeo_p)); + + /* Increment the association's refcnt. */ + sctp_association_hold(asoc); + + for (;;) { + prepare_to_wait_exclusive(&asoc->wait, &wait, + TASK_INTERRUPTIBLE); + if (!*timeo_p) + goto do_nonblock; + if (sk->sk_shutdown & RCV_SHUTDOWN) + break; + if (sk->sk_err || asoc->state >= SCTP_STATE_SHUTDOWN_PENDING || + asoc->base.dead) + goto do_error; + if (signal_pending(current)) + goto do_interrupted; + + if (sctp_state(asoc, ESTABLISHED)) + break; + + /* Let another process have a go. Since we are going + * to sleep anyway. + */ + sctp_release_sock(sk); + current_timeo = schedule_timeout(current_timeo); + sctp_lock_sock(sk); + + *timeo_p = current_timeo; + } + +out: + finish_wait(&asoc->wait, &wait); + + /* Release the association's refcnt. */ + sctp_association_put(asoc); + + return err; + +do_error: + if (asoc->counters[SCTP_COUNTER_INIT_ERROR] + 1 >= + asoc->max_init_attempts) + err = -ETIMEDOUT; + else + err = -ECONNREFUSED; + goto out; + +do_interrupted: + err = sock_intr_errno(*timeo_p); + goto out; + +do_nonblock: + err = -EINPROGRESS; + goto out; +} + +static int sctp_wait_for_accept(struct sock *sk, long timeo) +{ + struct sctp_endpoint *ep; + int err = 0; + DEFINE_WAIT(wait); + + ep = sctp_sk(sk)->ep; + + + for (;;) { + prepare_to_wait_exclusive(sk->sk_sleep, &wait, + TASK_INTERRUPTIBLE); + + if (list_empty(&ep->asocs)) { + sctp_release_sock(sk); + timeo = schedule_timeout(timeo); + sctp_lock_sock(sk); + } + + err = -EINVAL; + if (!sctp_sstate(sk, LISTENING)) + break; + + err = 0; + if (!list_empty(&ep->asocs)) + break; + + err = sock_intr_errno(timeo); + if (signal_pending(current)) + break; + + err = -EAGAIN; + if (!timeo) + break; + } + + finish_wait(sk->sk_sleep, &wait); + + return err; +} + +void sctp_wait_for_close(struct sock *sk, long timeout) +{ + DEFINE_WAIT(wait); + + do { + prepare_to_wait(sk->sk_sleep, &wait, TASK_INTERRUPTIBLE); + if (list_empty(&sctp_sk(sk)->ep->asocs)) + break; + sctp_release_sock(sk); + timeout = schedule_timeout(timeout); + sctp_lock_sock(sk); + } while (!signal_pending(current) && timeout); + + finish_wait(sk->sk_sleep, &wait); +} + +/* Populate the fields of the newsk from the oldsk and migrate the assoc + * and its messages to the newsk. + */ +static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, + struct sctp_association *assoc, + sctp_socket_type_t type) +{ + struct sctp_sock *oldsp = sctp_sk(oldsk); + struct sctp_sock *newsp = sctp_sk(newsk); + struct sctp_bind_bucket *pp; /* hash list port iterator */ + struct sctp_endpoint *newep = newsp->ep; + struct sk_buff *skb, *tmp; + struct sctp_ulpevent *event; + + /* Migrate socket buffer sizes and all the socket level options to the + * new socket. + */ + newsk->sk_sndbuf = oldsk->sk_sndbuf; + newsk->sk_rcvbuf = oldsk->sk_rcvbuf; + /* Brute force copy old sctp opt. */ + inet_sk_copy_descendant(newsk, oldsk); + + /* Restore the ep value that was overwritten with the above structure + * copy. + */ + newsp->ep = newep; + newsp->hmac = NULL; + + /* Hook this new socket in to the bind_hash list. */ + pp = sctp_sk(oldsk)->bind_hash; + sk_add_bind_node(newsk, &pp->owner); + sctp_sk(newsk)->bind_hash = pp; + inet_sk(newsk)->num = inet_sk(oldsk)->num; + + /* Move any messages in the old socket's receive queue that are for the + * peeled off association to the new socket's receive queue. + */ + sctp_skb_for_each(skb, &oldsk->sk_receive_queue, tmp) { + event = sctp_skb2event(skb); + if (event->asoc == assoc) { + __skb_unlink(skb, skb->list); + __skb_queue_tail(&newsk->sk_receive_queue, skb); + } + } + + /* Clean up any messages pending delivery due to partial + * delivery. Three cases: + * 1) No partial deliver; no work. + * 2) Peeling off partial delivery; keep pd_lobby in new pd_lobby. + * 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; + + if (sctp_sk(oldsk)->pd_mode) { + struct sk_buff_head *queue; + + /* Decide which queue to move pd_lobby skbs to. */ + if (assoc->ulpq.pd_mode) { + queue = &newsp->pd_lobby; + } else + queue = &newsk->sk_receive_queue; + + /* Walk through the pd_lobby, looking for skbs that + * need moved to the new socket. + */ + sctp_skb_for_each(skb, &oldsp->pd_lobby, tmp) { + event = sctp_skb2event(skb); + if (event->asoc == assoc) { + __skb_unlink(skb, skb->list); + __skb_queue_tail(queue, skb); + } + } + + /* Clear up any skbs waiting for the partial + * delivery to finish. + */ + if (assoc->ulpq.pd_mode) + sctp_clear_pd(oldsk); + + } + + /* Set the type of socket to indicate that it is peeled off from the + * original UDP-style socket or created with the accept() call on a + * TCP-style socket.. + */ + newsp->type = type; + + /* Migrate the association to the new socket. */ + sctp_assoc_migrate(assoc, newsk); + + /* If the association on the newsk is already closed before accept() + * is called, set RCV_SHUTDOWN flag. + */ + if (sctp_state(assoc, CLOSED) && sctp_style(newsk, TCP)) + newsk->sk_shutdown |= RCV_SHUTDOWN; + + newsk->sk_state = SCTP_SS_ESTABLISHED; +} + +/* This proto struct describes the ULP interface for SCTP. */ +struct proto sctp_prot = { + .name = "SCTP", + .owner = THIS_MODULE, + .close = sctp_close, + .connect = sctp_connect, + .disconnect = sctp_disconnect, + .accept = sctp_accept, + .ioctl = sctp_ioctl, + .init = sctp_init_sock, + .destroy = sctp_destroy_sock, + .shutdown = sctp_shutdown, + .setsockopt = sctp_setsockopt, + .getsockopt = sctp_getsockopt, + .sendmsg = sctp_sendmsg, + .recvmsg = sctp_recvmsg, + .bind = sctp_bind, + .backlog_rcv = sctp_backlog_rcv, + .hash = sctp_hash, + .unhash = sctp_unhash, + .get_port = sctp_get_port, + .obj_size = sizeof(struct sctp_sock), +}; + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +struct proto sctpv6_prot = { + .name = "SCTPv6", + .owner = THIS_MODULE, + .close = sctp_close, + .connect = sctp_connect, + .disconnect = sctp_disconnect, + .accept = sctp_accept, + .ioctl = sctp_ioctl, + .init = sctp_init_sock, + .destroy = sctp_destroy_sock, + .shutdown = sctp_shutdown, + .setsockopt = sctp_setsockopt, + .getsockopt = sctp_getsockopt, + .sendmsg = sctp_sendmsg, + .recvmsg = sctp_recvmsg, + .bind = sctp_bind, + .backlog_rcv = sctp_backlog_rcv, + .hash = sctp_hash, + .unhash = sctp_unhash, + .get_port = sctp_get_port, + .obj_size = sizeof(struct sctp6_sock), +}; +#endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ diff --git a/net/sctp/ssnmap.c b/net/sctp/ssnmap.c new file mode 100644 index 000000000000..e627d2b451b6 --- /dev/null +++ b/net/sctp/ssnmap.c @@ -0,0 +1,131 @@ +/* SCTP kernel reference Implementation + * Copyright (c) 2003 International Business Machines, Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions manipulate sctp SSN tracker. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Jon Grimm <jgrimm@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +#define MAX_KMALLOC_SIZE 131072 + +static struct sctp_ssnmap *sctp_ssnmap_init(struct sctp_ssnmap *map, __u16 in, + __u16 out); + +/* Storage size needed for map includes 2 headers and then the + * specific needs of in or out streams. + */ +static inline size_t sctp_ssnmap_size(__u16 in, __u16 out) +{ + return sizeof(struct sctp_ssnmap) + (in + out) * sizeof(__u16); +} + + +/* Create a new sctp_ssnmap. + * Allocate room to store at least 'len' contiguous TSNs. + */ +struct sctp_ssnmap *sctp_ssnmap_new(__u16 in, __u16 out, int gfp) +{ + struct sctp_ssnmap *retval; + int size; + + size = sctp_ssnmap_size(in, out); + if (size <= MAX_KMALLOC_SIZE) + retval = kmalloc(size, gfp); + else + retval = (struct sctp_ssnmap *) + __get_free_pages(gfp, get_order(size)); + if (!retval) + goto fail; + + if (!sctp_ssnmap_init(retval, in, out)) + goto fail_map; + + retval->malloced = 1; + SCTP_DBG_OBJCNT_INC(ssnmap); + + return retval; + +fail_map: + if (size <= MAX_KMALLOC_SIZE) + kfree(retval); + else + free_pages((unsigned long)retval, get_order(size)); +fail: + return NULL; +} + + +/* Initialize a block of memory as a ssnmap. */ +static struct sctp_ssnmap *sctp_ssnmap_init(struct sctp_ssnmap *map, __u16 in, + __u16 out) +{ + memset(map, 0x00, sctp_ssnmap_size(in, out)); + + /* Start 'in' stream just after the map header. */ + map->in.ssn = (__u16 *)&map[1]; + map->in.len = in; + + /* Start 'out' stream just after 'in'. */ + map->out.ssn = &map->in.ssn[in]; + map->out.len = out; + + return map; +} + +/* Clear out the ssnmap streams. */ +void sctp_ssnmap_clear(struct sctp_ssnmap *map) +{ + size_t size; + + size = (map->in.len + map->out.len) * sizeof(__u16); + memset(map->in.ssn, 0x00, size); +} + +/* Dispose of a ssnmap. */ +void sctp_ssnmap_free(struct sctp_ssnmap *map) +{ + if (map && map->malloced) { + int size; + + size = sctp_ssnmap_size(map->in.len, map->out.len); + if (size <= MAX_KMALLOC_SIZE) + kfree(map); + else + free_pages((unsigned long)map, get_order(size)); + SCTP_DBG_OBJCNT_DEC(ssnmap); + } +} diff --git a/net/sctp/sysctl.c b/net/sctp/sysctl.c new file mode 100644 index 000000000000..89fa20c73a5c --- /dev/null +++ b/net/sctp/sysctl.c @@ -0,0 +1,251 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2002, 2004 + * Copyright (c) 2002 Intel Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * Sysctl related interfaces for SCTP. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Mingqin Liu <liuming@us.ibm.com> + * Jon Grimm <jgrimm@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * Ryan Layer <rmlayer@us.ibm.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <net/sctp/structs.h> +#include <linux/sysctl.h> + +static ctl_handler sctp_sysctl_jiffies_ms; +static long rto_timer_min = 1; +static long rto_timer_max = 86400000; /* One day */ + +static ctl_table sctp_table[] = { + { + .ctl_name = NET_SCTP_RTO_INITIAL, + .procname = "rto_initial", + .data = &sctp_rto_initial, + .maxlen = sizeof(long), + .mode = 0644, + .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .strategy = &sctp_sysctl_jiffies_ms, + .extra1 = &rto_timer_min, + .extra2 = &rto_timer_max + }, + { + .ctl_name = NET_SCTP_RTO_MIN, + .procname = "rto_min", + .data = &sctp_rto_min, + .maxlen = sizeof(long), + .mode = 0644, + .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .strategy = &sctp_sysctl_jiffies_ms, + .extra1 = &rto_timer_min, + .extra2 = &rto_timer_max + }, + { + .ctl_name = NET_SCTP_RTO_MAX, + .procname = "rto_max", + .data = &sctp_rto_max, + .maxlen = sizeof(long), + .mode = 0644, + .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .strategy = &sctp_sysctl_jiffies_ms, + .extra1 = &rto_timer_min, + .extra2 = &rto_timer_max + }, + { + .ctl_name = NET_SCTP_VALID_COOKIE_LIFE, + .procname = "valid_cookie_life", + .data = &sctp_valid_cookie_life, + .maxlen = sizeof(long), + .mode = 0644, + .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .strategy = &sctp_sysctl_jiffies_ms, + .extra1 = &rto_timer_min, + .extra2 = &rto_timer_max + }, + { + .ctl_name = NET_SCTP_MAX_BURST, + .procname = "max_burst", + .data = &sctp_max_burst, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_ASSOCIATION_MAX_RETRANS, + .procname = "association_max_retrans", + .data = &sctp_max_retrans_association, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_PATH_MAX_RETRANS, + .procname = "path_max_retrans", + .data = &sctp_max_retrans_path, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_MAX_INIT_RETRANSMITS, + .procname = "max_init_retransmits", + .data = &sctp_max_retrans_init, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_HB_INTERVAL, + .procname = "hb_interval", + .data = &sctp_hb_interval, + .maxlen = sizeof(long), + .mode = 0644, + .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .strategy = &sctp_sysctl_jiffies_ms, + .extra1 = &rto_timer_min, + .extra2 = &rto_timer_max + }, + { + .ctl_name = NET_SCTP_PRESERVE_ENABLE, + .procname = "cookie_preserve_enable", + .data = &sctp_cookie_preserve_enable, + .maxlen = sizeof(long), + .mode = 0644, + .proc_handler = &proc_doulongvec_ms_jiffies_minmax, + .strategy = &sctp_sysctl_jiffies_ms, + .extra1 = &rto_timer_min, + .extra2 = &rto_timer_max + }, + { + .ctl_name = NET_SCTP_RTO_ALPHA, + .procname = "rto_alpha_exp_divisor", + .data = &sctp_rto_alpha, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_RTO_BETA, + .procname = "rto_beta_exp_divisor", + .data = &sctp_rto_beta, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_ADDIP_ENABLE, + .procname = "addip_enable", + .data = &sctp_addip_enable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_SCTP_PRSCTP_ENABLE, + .procname = "prsctp_enable", + .data = &sctp_prsctp_enable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { .ctl_name = 0 } +}; + +static ctl_table sctp_net_table[] = { + { + .ctl_name = NET_SCTP, + .procname = "sctp", + .mode = 0555, + .child = sctp_table + }, + { .ctl_name = 0 } +}; + +static ctl_table sctp_root_table[] = { + { + .ctl_name = CTL_NET, + .procname = "net", + .mode = 0555, + .child = sctp_net_table + }, + { .ctl_name = 0 } +}; + +static struct ctl_table_header * sctp_sysctl_header; + +/* Sysctl registration. */ +void sctp_sysctl_register(void) +{ + sctp_sysctl_header = register_sysctl_table(sctp_root_table, 0); +} + +/* Sysctl deregistration. */ +void sctp_sysctl_unregister(void) +{ + unregister_sysctl_table(sctp_sysctl_header); +} + +/* Strategy function to convert jiffies to milliseconds. */ +static int sctp_sysctl_jiffies_ms(ctl_table *table, int __user *name, int nlen, + void __user *oldval, size_t __user *oldlenp, + void __user *newval, size_t newlen, void **context) { + + if (oldval) { + size_t olen; + + if (oldlenp) { + if (get_user(olen, oldlenp)) + return -EFAULT; + + if (olen != sizeof (int)) + return -EINVAL; + } + if (put_user((*(int *)(table->data) * 1000) / HZ, + (int __user *)oldval) || + (oldlenp && put_user(sizeof (int), oldlenp))) + return -EFAULT; + } + if (newval && newlen) { + int new; + + if (newlen != sizeof (int)) + return -EINVAL; + + if (get_user(new, (int __user *)newval)) + return -EFAULT; + + *(int *)(table->data) = (new * HZ) / 1000; + } + return 1; +} diff --git a/net/sctp/transport.c b/net/sctp/transport.c new file mode 100644 index 000000000000..f30882e1e96a --- /dev/null +++ b/net/sctp/transport.c @@ -0,0 +1,514 @@ +/* SCTP kernel reference 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 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; + * 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Karl Knutson <karl@athena.chicago.il.us> + * Jon Grimm <jgrimm@us.ibm.com> + * Xingang Guo <xingang.guo@intel.com> + * Hui Huang <hui.huang@nokia.com> + * Sridhar Samudrala <sri@us.ibm.com> + * Ardelle Fan <ardelle.fan@intel.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* 1st Level Abstractions. */ + +/* 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) +{ + /* 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: + * + * C1) Until an RTT measurement has been made for a packet sent to the + * 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->last_time_heard = jiffies; + peer->last_time_used = jiffies; + peer->last_time_ecne_reduced = jiffies; + + peer->active = SCTP_ACTIVE; + peer->hb_allowed = 0; + + /* Initialize the default path max_retrans. */ + peer->max_retrans = sctp_max_retrans_path; + peer->error_count = 0; + + 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; + + /* Set up the heartbeat timer. */ + init_timer(&peer->hb_timer); + peer->hb_interval = SCTP_DEFAULT_TIMEOUT_HEARTBEAT; + peer->hb_timer.function = sctp_generate_heartbeat_event; + peer->hb_timer.data = (unsigned long)peer; + + 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 *transport; + + transport = t_new(struct sctp_transport, gfp); + if (!transport) + goto fail; + + if (!sctp_transport_init(transport, addr, gfp)) + goto fail_init; + + transport->malloced = 1; + 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 + * delay until it last reference count. + */ +void sctp_transport_free(struct sctp_transport *transport) +{ + transport->dead = 1; + + /* Try to delete the heartbeat timer. */ + if (del_timer(&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. + */ + if (timer_pending(&transport->T3_rtx_timer) && + del_timer(&transport->T3_rtx_timer)) + sctp_transport_put(transport); + + + sctp_transport_put(transport); +} + +/* Destroy the transport data structure. + * Assumes there are no more users of this structure. + */ +static void sctp_transport_destroy(struct sctp_transport *transport) +{ + SCTP_ASSERT(transport->dead, "Transport is not dead", return); + + if (transport->asoc) + sctp_association_put(transport->asoc); + + sctp_packet_free(&transport->packet); + + dst_release(transport->dst); + kfree(transport); + SCTP_DBG_OBJCNT_DEC(transport); +} + +/* Start T3_rtx timer if it is not already running and update the heartbeat + * timer. This routine is called every time a DATA chunk is sent. + */ +void sctp_transport_reset_timers(struct sctp_transport *transport) +{ + /* RFC 2960 6.3.2 Retransmission Timer Rules + * + * R1) Every time a DATA chunk is sent to any address(including a + * retransmission), if the T3-rtx timer of that address is not running + * start it running so that it will expire after the RTO of that + * address. + */ + + if (!timer_pending(&transport->T3_rtx_timer)) + if (!mod_timer(&transport->T3_rtx_timer, + jiffies + transport->rto)) + sctp_transport_hold(transport); + + /* When a data chunk is sent, reset the heartbeat interval. */ + if (!mod_timer(&transport->hb_timer, + sctp_transport_timeout(transport))) + 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. + */ +void sctp_transport_set_owner(struct sctp_transport *transport, + struct sctp_association *asoc) +{ + transport->asoc = asoc; + sctp_association_hold(asoc); +} + +/* Initialize the pmtu of a transport. */ +void sctp_transport_pmtu(struct sctp_transport *transport) +{ + struct dst_entry *dst; + + dst = transport->af_specific->get_dst(NULL, &transport->ipaddr, NULL); + + if (dst) { + transport->pmtu = dst_mtu(dst); + dst_release(dst); + } else + transport->pmtu = SCTP_DEFAULT_MAXSEGMENT; +} + +/* Caches the dst entry and source address for a transport's destination + * address. + */ +void sctp_transport_route(struct sctp_transport *transport, + union sctp_addr *saddr, struct sctp_sock *opt) +{ + struct sctp_association *asoc = transport->asoc; + struct sctp_af *af = transport->af_specific; + union sctp_addr *daddr = &transport->ipaddr; + struct dst_entry *dst; + + dst = af->get_dst(asoc, daddr, saddr); + + if (saddr) + memcpy(&transport->saddr, saddr, sizeof(union sctp_addr)); + else + af->get_saddr(asoc, dst, daddr, &transport->saddr); + + transport->dst = dst; + if (dst) { + transport->pmtu = 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); + } else + transport->pmtu = SCTP_DEFAULT_MAXSEGMENT; +} + +/* Hold a reference to a transport. */ +void sctp_transport_hold(struct sctp_transport *transport) +{ + atomic_inc(&transport->refcnt); +} + +/* Release a reference to a transport and clean up + * if there are no more references. + */ +void sctp_transport_put(struct sctp_transport *transport) +{ + if (atomic_dec_and_test(&transport->refcnt)) + sctp_transport_destroy(transport); +} + +/* Update transport's RTO based on the newly calculated RTT. */ +void sctp_transport_update_rto(struct sctp_transport *tp, __u32 rtt) +{ + /* Check for valid transport. */ + SCTP_ASSERT(tp, "NULL transport", return); + + /* We should not be doing any RTO updates unless rto_pending is set. */ + SCTP_ASSERT(tp->rto_pending, "rto_pending not set", return); + + if (tp->rttvar || tp->srtt) { + /* 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' + */ + + /* Note: The above algorithm has been rewritten to + * express rto_beta and rto_alpha as inverse powers + * of two. + * 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 >> sctp_rto_beta) + + ((abs(tp->srtt - rtt)) >> sctp_rto_beta); + tp->srtt = tp->srtt - (tp->srtt >> sctp_rto_alpha) + + (rtt >> sctp_rto_alpha); + } else { + /* 6.3.1 C2) When the first RTT measurement R is made, set + * SRTT <- R, RTTVAR <- R/2. + */ + tp->srtt = rtt; + tp->rttvar = rtt >> 1; + } + + /* 6.3.1 G1) Whenever RTTVAR is computed, if RTTVAR = 0, then + * adjust RTTVAR <- G, where G is the CLOCK GRANULARITY. + */ + if (tp->rttvar == 0) + tp->rttvar = SCTP_CLOCK_GRANULARITY; + + /* 6.3.1 C3) After the computation, update RTO <- SRTT + 4 * RTTVAR. */ + tp->rto = tp->srtt + (tp->rttvar << 2); + + /* 6.3.1 C6) Whenever RTO is computed, if it is less than RTO.Min + * seconds then it is rounded up to RTO.Min seconds. + */ + if (tp->rto < tp->asoc->rto_min) + tp->rto = tp->asoc->rto_min; + + /* 6.3.1 C7) A maximum value may be placed on RTO provided it is + * at least RTO.max seconds. + */ + if (tp->rto > tp->asoc->rto_max) + tp->rto = tp->asoc->rto_max; + + tp->rtt = rtt; + + /* Reset rto_pending so that a new RTT measurement is started when a + * new data chunk is sent. + */ + tp->rto_pending = 0; + + SCTP_DEBUG_PRINTK("%s: transport: %p, rtt: %d, srtt: %d " + "rttvar: %d, rto: %d\n", __FUNCTION__, + tp, rtt, tp->srtt, tp->rttvar, tp->rto); +} + +/* This routine updates the transport's cwnd and partial_bytes_acked + * parameters based on the bytes acked in the received SACK. + */ +void sctp_transport_raise_cwnd(struct sctp_transport *transport, + __u32 sack_ctsn, __u32 bytes_acked) +{ + __u32 cwnd, ssthresh, flight_size, pba, pmtu; + + cwnd = transport->cwnd; + flight_size = transport->flight_size; + + /* The appropriate cwnd increase algorithm is performed if, and only + * if the cumulative TSN has advanced and the congestion window is + * being fully utilized. + */ + if ((transport->asoc->ctsn_ack_point >= sack_ctsn) || + (flight_size < cwnd)) + return; + + ssthresh = transport->ssthresh; + pba = transport->partial_bytes_acked; + pmtu = transport->asoc->pmtu; + + 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. + */ + if (bytes_acked > pmtu) + cwnd += pmtu; + else + cwnd += bytes_acked; + SCTP_DEBUG_PRINTK("%s: SLOW START: transport: %p, " + "bytes_acked: %d, cwnd: %d, ssthresh: %d, " + "flight_size: %d, pba: %d\n", + __FUNCTION__, + transport, bytes_acked, cwnd, + ssthresh, flight_size, pba); + } else { + /* RFC 2960 7.2.2 Whenever cwnd is greater than ssthresh, + * upon each SACK arrival that advances the Cumulative TSN Ack + * Point, increase partial_bytes_acked by the total number of + * bytes of all new chunks acknowledged in that SACK including + * chunks acknowledged by the new Cumulative TSN Ack and by + * Gap Ack Blocks. + * + * When partial_bytes_acked is equal to or greater than cwnd + * and before the arrival of the SACK the sender had cwnd or + * more bytes of data outstanding (i.e., before arrival of the + * SACK, flightsize was greater than or equal to cwnd), + * increase cwnd by MTU, and reset partial_bytes_acked to + * (partial_bytes_acked - cwnd). + */ + pba += bytes_acked; + if (pba >= cwnd) { + cwnd += pmtu; + pba = ((cwnd < pba) ? (pba - cwnd) : 0); + } + SCTP_DEBUG_PRINTK("%s: CONGESTION AVOIDANCE: " + "transport: %p, bytes_acked: %d, cwnd: %d, " + "ssthresh: %d, flight_size: %d, pba: %d\n", + __FUNCTION__, + transport, bytes_acked, cwnd, + ssthresh, flight_size, pba); + } + + transport->cwnd = cwnd; + transport->partial_bytes_acked = pba; +} + +/* This routine is used to lower the transport's cwnd when congestion is + * detected. + */ +void sctp_transport_lower_cwnd(struct sctp_transport *transport, + sctp_lower_cwnd_t reason) +{ + switch (reason) { + case SCTP_LOWER_CWND_T3_RTX: + /* RFC 2960 Section 7.2.3, sctpimpguide + * When the T3-rtx timer expires on an address, SCTP should + * perform slow start by: + * ssthresh = max(cwnd/2, 4*MTU) + * cwnd = 1*MTU + * partial_bytes_acked = 0 + */ + transport->ssthresh = max(transport->cwnd/2, + 4*transport->asoc->pmtu); + transport->cwnd = transport->asoc->pmtu; + break; + + case SCTP_LOWER_CWND_FAST_RTX: + /* RFC 2960 7.2.4 Adjust the ssthresh and cwnd of the + * 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 + * 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 + */ + transport->ssthresh = max(transport->cwnd/2, + 4*transport->asoc->pmtu); + transport->cwnd = transport->ssthresh; + break; + + case SCTP_LOWER_CWND_ECNE: + /* RFC 2481 Section 6.1.2. + * If the sender receives an ECN-Echo ACK packet + * then the sender knows that congestion was encountered in the + * network on the path from the sender to the receiver. The + * indication of congestion should be treated just as a + * congestion loss in non-ECN Capable TCP. That is, the TCP + * source halves the congestion window "cwnd" and reduces the + * slow start threshold "ssthresh". + * A critical condition is that TCP does not react to + * 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) { + transport->ssthresh = max(transport->cwnd/2, + 4*transport->asoc->pmtu); + transport->cwnd = transport->ssthresh; + transport->last_time_ecne_reduced = jiffies; + } + break; + + case SCTP_LOWER_CWND_INACTIVE: + /* RFC 2960 Section 7.2.1, sctpimpguide + * When the endpoint does not transmit data on a given + * transport address, the cwnd of the transport address + * should be adjusted to max(cwnd/2, 4*MTU) per RTO. + * NOTE: Although the draft recommends that this check needs + * 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); + break; + }; + + transport->partial_bytes_acked = 0; + SCTP_DEBUG_PRINTK("%s: transport: %p reason: %d cwnd: " + "%d ssthresh: %d\n", __FUNCTION__, + transport, reason, + transport->cwnd, transport->ssthresh); +} + +/* 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 += jiffies; + return timeout; +} diff --git a/net/sctp/tsnmap.c b/net/sctp/tsnmap.c new file mode 100644 index 000000000000..ac4fae161bc7 --- /dev/null +++ b/net/sctp/tsnmap.c @@ -0,0 +1,417 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * + * This file is part of the SCTP kernel reference Implementation + * + * These functions manipulate sctp tsn mapping array. + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * La Monte H.P. Yarroll <piggy@acm.org> + * Jon Grimm <jgrimm@us.ibm.com> + * Karl Knutson <karl@athena.chicago.il.us> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +static void sctp_tsnmap_update(struct sctp_tsnmap *map); +static void sctp_tsnmap_find_gap_ack(__u8 *map, __u16 off, + __u16 len, __u16 base, + int *started, __u16 *start, + int *ended, __u16 *end); + +/* Initialize a block of memory as a tsnmap. */ +struct sctp_tsnmap *sctp_tsnmap_init(struct sctp_tsnmap *map, __u16 len, + __u32 initial_tsn) +{ + map->tsn_map = map->raw_map; + map->overflow_map = map->tsn_map + len; + map->len = len; + + /* Clear out a TSN ack status. */ + memset(map->tsn_map, 0x00, map->len + map->len); + + /* Keep track of TSNs represented by tsn_map. */ + map->base_tsn = initial_tsn; + map->overflow_tsn = initial_tsn + map->len; + map->cumulative_tsn_ack_point = initial_tsn - 1; + map->max_tsn_seen = map->cumulative_tsn_ack_point; + map->malloced = 0; + map->num_dup_tsns = 0; + + return map; +} + +/* Test the tracking state of this TSN. + * Returns: + * 0 if the TSN has not yet been seen + * >0 if the TSN has been seen (duplicate) + * <0 if the TSN is invalid (too large to track) + */ +int sctp_tsnmap_check(const struct sctp_tsnmap *map, __u32 tsn) +{ + __s32 gap; + int dup; + + /* Calculate the index into the mapping arrays. */ + gap = tsn - map->base_tsn; + + /* Verify that we can hold this TSN. */ + if (gap >= (/* base */ map->len + /* overflow */ map->len)) { + dup = -1; + goto out; + } + + /* Honk if we've already seen this TSN. + * We have three cases: + * 1. The TSN is ancient or belongs to a previous tsn_map. + * 2. The TSN is already marked in the tsn_map. + * 3. The TSN is already marked in the tsn_map_overflow. + */ + if (gap < 0 || + (gap < map->len && map->tsn_map[gap]) || + (gap >= map->len && map->overflow_map[gap - map->len])) + dup = 1; + else + dup = 0; + +out: + return dup; +} + + +/* Mark this TSN as seen. */ +void sctp_tsnmap_mark(struct sctp_tsnmap *map, __u32 tsn) +{ + __s32 gap; + + /* Vacuously mark any TSN which precedes the map base or + * exceeds the end of the map. + */ + if (TSN_lt(tsn, map->base_tsn)) + return; + if (!TSN_lt(tsn, map->base_tsn + map->len + map->len)) + return; + + /* Bump the max. */ + if (TSN_lt(map->max_tsn_seen, tsn)) + map->max_tsn_seen = tsn; + + /* Assert: TSN is in range. */ + gap = tsn - map->base_tsn; + + /* Mark the TSN as received. */ + if (gap < map->len) + map->tsn_map[gap]++; + else + map->overflow_map[gap - map->len]++; + + /* Go fixup any internal TSN mapping variables including + * cumulative_tsn_ack_point. + */ + sctp_tsnmap_update(map); +} + + +/* Initialize a Gap Ack Block iterator from memory being provided. */ +SCTP_STATIC void sctp_tsnmap_iter_init(const struct sctp_tsnmap *map, + struct sctp_tsnmap_iter *iter) +{ + /* Only start looking one past the Cumulative TSN Ack Point. */ + iter->start = map->cumulative_tsn_ack_point + 1; +} + +/* Get the next Gap Ack Blocks. Returns 0 if there was not another block + * to get. + */ +SCTP_STATIC int sctp_tsnmap_next_gap_ack(const struct sctp_tsnmap *map, + struct sctp_tsnmap_iter *iter, + __u16 *start, __u16 *end) +{ + int started, ended; + __u16 _start, _end, offset; + + /* We haven't found a gap yet. */ + started = ended = 0; + + /* If there are no more gap acks possible, get out fast. */ + if (TSN_lte(map->max_tsn_seen, iter->start)) + return 0; + + /* Search the first mapping array. */ + if (iter->start - map->base_tsn < map->len) { + + offset = iter->start - map->base_tsn; + sctp_tsnmap_find_gap_ack(map->tsn_map, offset, map->len, 0, + &started, &_start, &ended, &_end); + } + + /* Do we need to check the overflow map? */ + if (!ended) { + /* Fix up where we'd like to start searching in the + * overflow map. + */ + if (iter->start - map->base_tsn < map->len) + offset = 0; + else + offset = iter->start - map->base_tsn - map->len; + + /* Search the overflow map. */ + sctp_tsnmap_find_gap_ack(map->overflow_map, + offset, + map->len, + map->len, + &started, &_start, + &ended, &_end); + } + + /* The Gap Ack Block happens to end at the end of the + * overflow map. + */ + if (started && !ended) { + ended++; + _end = map->len + map->len - 1; + } + + /* If we found a Gap Ack Block, return the start and end and + * bump the iterator forward. + */ + if (ended) { + /* Fix up the start and end based on the + * Cumulative TSN Ack offset into the map. + */ + int gap = map->cumulative_tsn_ack_point - + map->base_tsn; + + *start = _start - gap; + *end = _end - gap; + + /* Move the iterator forward. */ + iter->start = map->cumulative_tsn_ack_point + *end + 1; + } + + return ended; +} + +/* Mark this and any lower TSN as seen. */ +void sctp_tsnmap_skip(struct sctp_tsnmap *map, __u32 tsn) +{ + __s32 gap; + + /* Vacuously mark any TSN which precedes the map base or + * exceeds the end of the map. + */ + if (TSN_lt(tsn, map->base_tsn)) + return; + if (!TSN_lt(tsn, map->base_tsn + map->len + map->len)) + return; + + /* Bump the max. */ + if (TSN_lt(map->max_tsn_seen, tsn)) + map->max_tsn_seen = tsn; + + /* Assert: TSN is in range. */ + gap = tsn - map->base_tsn + 1; + + /* Mark the TSNs as received. */ + if (gap <= map->len) + memset(map->tsn_map, 0x01, gap); + else { + memset(map->tsn_map, 0x01, map->len); + memset(map->overflow_map, 0x01, (gap - map->len)); + } + + /* Go fixup any internal TSN mapping variables including + * cumulative_tsn_ack_point. + */ + sctp_tsnmap_update(map); +} + +/******************************************************************** + * 2nd Level Abstractions + ********************************************************************/ + +/* This private helper function updates the tsnmap buffers and + * the Cumulative TSN Ack Point. + */ +static void sctp_tsnmap_update(struct sctp_tsnmap *map) +{ + __u32 ctsn; + + ctsn = map->cumulative_tsn_ack_point; + do { + ctsn++; + if (ctsn == map->overflow_tsn) { + /* Now tsn_map must have been all '1's, + * so we swap the map and check the overflow table + */ + __u8 *tmp = map->tsn_map; + memset(tmp, 0, map->len); + map->tsn_map = map->overflow_map; + map->overflow_map = tmp; + + /* Update the tsn_map boundaries. */ + map->base_tsn += map->len; + map->overflow_tsn += map->len; + } + } while (map->tsn_map[ctsn - map->base_tsn]); + + map->cumulative_tsn_ack_point = ctsn - 1; /* Back up one. */ +} + +/* How many data chunks are we missing from our peer? + */ +__u16 sctp_tsnmap_pending(struct sctp_tsnmap *map) +{ + __u32 cum_tsn = map->cumulative_tsn_ack_point; + __u32 max_tsn = map->max_tsn_seen; + __u32 base_tsn = map->base_tsn; + __u16 pending_data; + __s32 gap, start, end, i; + + pending_data = max_tsn - cum_tsn; + gap = max_tsn - base_tsn; + + if (gap <= 0 || gap >= (map->len + map->len)) + goto out; + + start = ((cum_tsn >= base_tsn) ? (cum_tsn - base_tsn + 1) : 0); + end = ((gap > map->len ) ? map->len : gap + 1); + + for (i = start; i < end; i++) { + if (map->tsn_map[i]) + pending_data--; + } + + if (gap >= map->len) { + start = 0; + end = gap - map->len + 1; + for (i = start; i < end; i++) { + if (map->overflow_map[i]) + pending_data--; + } + } + +out: + return pending_data; +} + +/* This is a private helper for finding Gap Ack Blocks. It searches a + * single array for the start and end of a Gap Ack Block. + * + * The flags "started" and "ended" tell is if we found the beginning + * or (respectively) the end of a Gap Ack Block. + */ +static void sctp_tsnmap_find_gap_ack(__u8 *map, __u16 off, + __u16 len, __u16 base, + int *started, __u16 *start, + int *ended, __u16 *end) +{ + int i = off; + + /* Look through the entire array, but break out + * early if we have found the end of the Gap Ack Block. + */ + + /* Also, stop looking past the maximum TSN seen. */ + + /* Look for the start. */ + if (!(*started)) { + for (; i < len; i++) { + if (map[i]) { + (*started)++; + *start = base + i; + break; + } + } + } + + /* Look for the end. */ + if (*started) { + /* We have found the start, let's find the + * end. If we find the end, break out. + */ + for (; i < len; i++) { + if (!map[i]) { + (*ended)++; + *end = base + i - 1; + break; + } + } + } +} + +/* Renege that we have seen a TSN. */ +void sctp_tsnmap_renege(struct sctp_tsnmap *map, __u32 tsn) +{ + __s32 gap; + + if (TSN_lt(tsn, map->base_tsn)) + return; + if (!TSN_lt(tsn, map->base_tsn + map->len + map->len)) + return; + + /* Assert: TSN is in range. */ + gap = tsn - map->base_tsn; + + /* Pretend we never saw the TSN. */ + if (gap < map->len) + map->tsn_map[gap] = 0; + else + map->overflow_map[gap - map->len] = 0; +} + +/* How many gap ack blocks do we have recorded? */ +__u16 sctp_tsnmap_num_gabs(struct sctp_tsnmap *map) +{ + struct sctp_tsnmap_iter iter; + int gabs = 0; + + /* Refresh the gap ack information. */ + if (sctp_tsnmap_has_gap(map)) { + sctp_tsnmap_iter_init(map, &iter); + while (sctp_tsnmap_next_gap_ack(map, &iter, + &map->gabs[gabs].start, + &map->gabs[gabs].end)) { + + map->gabs[gabs].start = htons(map->gabs[gabs].start); + map->gabs[gabs].end = htons(map->gabs[gabs].end); + gabs++; + if (gabs >= SCTP_MAX_GABS) + break; + } + } + return gabs; +} diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c new file mode 100644 index 000000000000..17d0ff534735 --- /dev/null +++ b/net/sctp/ulpevent.c @@ -0,0 +1,942 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * Copyright (c) 2001 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * These functions manipulate an sctp event. The struct ulpevent is used + * to carry notifications and data to the ULP (sockets). + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Jon Grimm <jgrimm@us.ibm.com> + * La Monte H.P. Yarroll <piggy@acm.org> + * Ardelle Fan <ardelle.fan@intel.com> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/skbuff.h> +#include <net/sctp/structs.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +static void sctp_ulpevent_receive_data(struct sctp_ulpevent *event, + struct sctp_association *asoc); +static void sctp_ulpevent_release_data(struct sctp_ulpevent *event); + +/* Stub skb destructor. */ +static void sctp_stub_rfree(struct sk_buff *skb) +{ +/* WARNING: This function is just a warning not to use the + * skb destructor. If the skb is shared, we may get the destructor + * callback on some processor that does not own the sock_lock. This + * was occuring with PACKET socket applications that were monitoring + * our skbs. We can't take the sock_lock, because we can't risk + * recursing if we do really own the sock lock. Instead, do all + * of our rwnd manipulation while we own the sock_lock outright. + */ +} + +/* Initialize an ULP event from an given skb. */ +SCTP_STATIC void sctp_ulpevent_init(struct sctp_ulpevent *event, int msg_flags) +{ + memset(event, 0, sizeof(struct sctp_ulpevent)); + event->msg_flags = msg_flags; +} + +/* Create a new sctp_ulpevent. */ +SCTP_STATIC struct sctp_ulpevent *sctp_ulpevent_new(int size, int msg_flags, + int gfp) +{ + struct sctp_ulpevent *event; + struct sk_buff *skb; + + skb = alloc_skb(size, gfp); + if (!skb) + goto fail; + + event = sctp_skb2event(skb); + sctp_ulpevent_init(event, msg_flags); + + return event; + +fail: + return NULL; +} + +/* Is this a MSG_NOTIFICATION? */ +int sctp_ulpevent_is_notification(const struct sctp_ulpevent *event) +{ + return MSG_NOTIFICATION == (event->msg_flags & MSG_NOTIFICATION); +} + +/* Hold the association in case the msg_name needs read out of + * the association. + */ +static inline void sctp_ulpevent_set_owner(struct sctp_ulpevent *event, + const struct sctp_association *asoc) +{ + struct sk_buff *skb; + + /* Cast away the const, as we are just wanting to + * bump the reference count. + */ + sctp_association_hold((struct sctp_association *)asoc); + skb = sctp_event2skb(event); + skb->sk = asoc->base.sk; + event->asoc = (struct sctp_association *)asoc; + skb->destructor = sctp_stub_rfree; +} + +/* A simple destructor to give up the reference to the association. */ +static inline void sctp_ulpevent_release_owner(struct sctp_ulpevent *event) +{ + sctp_association_put(event->asoc); +} + +/* Create and initialize an SCTP_ASSOC_CHANGE event. + * + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * Communication notifications inform the ULP that an SCTP association + * has either begun or ended. The identifier for a new association is + * provided by this notification. + * + * Note: There is no field checking here. If a field is unused it will be + * zero'd out. + */ +struct sctp_ulpevent *sctp_ulpevent_make_assoc_change( + const struct sctp_association *asoc, + __u16 flags, __u16 state, __u16 error, __u16 outbound, + __u16 inbound, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_assoc_change *sac; + struct sk_buff *skb; + + event = sctp_ulpevent_new(sizeof(struct sctp_assoc_change), + MSG_NOTIFICATION, gfp); + if (!event) + goto fail; + skb = sctp_event2skb(event); + sac = (struct sctp_assoc_change *) + skb_put(skb, sizeof(struct sctp_assoc_change)); + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_type: + * It should be SCTP_ASSOC_CHANGE. + */ + sac->sac_type = SCTP_ASSOC_CHANGE; + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_state: 32 bits (signed integer) + * This field holds one of a number of values that communicate the + * event that happened to the association. + */ + sac->sac_state = state; + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_flags: 16 bits (unsigned integer) + * Currently unused. + */ + sac->sac_flags = 0; + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_length: sizeof (__u32) + * This field is the total length of the notification data, including + * the notification header. + */ + sac->sac_length = sizeof(struct sctp_assoc_change); + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_error: 32 bits (signed integer) + * + * If the state was reached due to a error condition (e.g. + * COMMUNICATION_LOST) any relevant error information is available in + * this field. This corresponds to the protocol error codes defined in + * [SCTP]. + */ + sac->sac_error = error; + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_outbound_streams: 16 bits (unsigned integer) + * sac_inbound_streams: 16 bits (unsigned integer) + * + * The maximum number of streams allowed in each direction are + * available in sac_outbound_streams and sac_inbound streams. + */ + sac->sac_outbound_streams = outbound; + sac->sac_inbound_streams = inbound; + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * sac_assoc_id: sizeof (sctp_assoc_t) + * + * The association id field, holds the identifier for the association. + * All notifications for a given association have the same association + * identifier. For TCP style socket, this field is ignored. + */ + sctp_ulpevent_set_owner(event, asoc); + sac->sac_assoc_id = sctp_assoc2id(asoc); + + return event; + +fail: + return NULL; +} + +/* Create and initialize an SCTP_PEER_ADDR_CHANGE event. + * + * Socket Extensions for SCTP - draft-01 + * 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * When a destination address on a multi-homed peer encounters a change + * an interface details event is sent. + */ +struct sctp_ulpevent *sctp_ulpevent_make_peer_addr_change( + const struct sctp_association *asoc, + const struct sockaddr_storage *aaddr, + int flags, int state, int error, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_paddr_change *spc; + struct sk_buff *skb; + + event = sctp_ulpevent_new(sizeof(struct sctp_paddr_change), + MSG_NOTIFICATION, gfp); + if (!event) + goto fail; + + skb = sctp_event2skb(event); + spc = (struct sctp_paddr_change *) + skb_put(skb, sizeof(struct sctp_paddr_change)); + + /* Sockets API Extensions for SCTP + * Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * spc_type: + * + * It should be SCTP_PEER_ADDR_CHANGE. + */ + spc->spc_type = SCTP_PEER_ADDR_CHANGE; + + /* Sockets API Extensions for SCTP + * Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * spc_length: sizeof (__u32) + * + * This field is the total length of the notification data, including + * the notification header. + */ + spc->spc_length = sizeof(struct sctp_paddr_change); + + /* Sockets API Extensions for SCTP + * Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * spc_flags: 16 bits (unsigned integer) + * Currently unused. + */ + spc->spc_flags = 0; + + /* Sockets API Extensions for SCTP + * Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * spc_state: 32 bits (signed integer) + * + * This field holds one of a number of values that communicate the + * event that happened to the address. + */ + spc->spc_state = state; + + /* Sockets API Extensions for SCTP + * Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * spc_error: 32 bits (signed integer) + * + * If the state was reached due to any error condition (e.g. + * ADDRESS_UNREACHABLE) any relevant error information is available in + * this field. + */ + spc->spc_error = error; + + /* Socket Extensions for SCTP + * 5.3.1.1 SCTP_ASSOC_CHANGE + * + * spc_assoc_id: sizeof (sctp_assoc_t) + * + * The association id field, holds the identifier for the association. + * All notifications for a given association have the same association + * identifier. For TCP style socket, this field is ignored. + */ + sctp_ulpevent_set_owner(event, asoc); + spc->spc_assoc_id = sctp_assoc2id(asoc); + + /* Sockets API Extensions for SCTP + * Section 5.3.1.2 SCTP_PEER_ADDR_CHANGE + * + * spc_aaddr: sizeof (struct sockaddr_storage) + * + * The affected address field, holds the remote peer's address that is + * encountering the change of state. + */ + memcpy(&spc->spc_aaddr, aaddr, sizeof(struct sockaddr_storage)); + + /* Map ipv4 address into v4-mapped-on-v6 address. */ + sctp_get_pf_specific(asoc->base.sk->sk_family)->addr_v4map( + sctp_sk(asoc->base.sk), + (union sctp_addr *)&spc->spc_aaddr); + + return event; + +fail: + return NULL; +} + +/* Create and initialize an SCTP_REMOTE_ERROR notification. + * + * Note: This assumes that the chunk->skb->data already points to the + * operation error payload. + * + * Socket Extensions for SCTP - draft-01 + * 5.3.1.3 SCTP_REMOTE_ERROR + * + * A remote peer may send an Operational Error message to its peer. + * This message indicates a variety of error conditions on an + * association. The entire error TLV as it appears on the wire is + * included in a SCTP_REMOTE_ERROR event. Please refer to the SCTP + * specification [SCTP] and any extensions for a list of possible + * error formats. + */ +struct sctp_ulpevent *sctp_ulpevent_make_remote_error( + const struct sctp_association *asoc, struct sctp_chunk *chunk, + __u16 flags, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_remote_error *sre; + struct sk_buff *skb; + sctp_errhdr_t *ch; + __u16 cause; + int elen; + + ch = (sctp_errhdr_t *)(chunk->skb->data); + cause = ch->cause; + elen = WORD_ROUND(ntohs(ch->length)) - sizeof(sctp_errhdr_t); + + /* Pull off the ERROR header. */ + skb_pull(chunk->skb, sizeof(sctp_errhdr_t)); + + /* Copy the skb to a new skb with room for us to prepend + * notification with. + */ + skb = skb_copy_expand(chunk->skb, sizeof(struct sctp_remote_error), + 0, gfp); + + /* Pull off the rest of the cause TLV from the chunk. */ + skb_pull(chunk->skb, elen); + if (!skb) + goto fail; + + /* Embed the event fields inside the cloned skb. */ + event = sctp_skb2event(skb); + sctp_ulpevent_init(event, MSG_NOTIFICATION); + + sre = (struct sctp_remote_error *) + skb_push(skb, sizeof(struct sctp_remote_error)); + + /* Trim the buffer to the right length. */ + skb_trim(skb, sizeof(struct sctp_remote_error) + elen); + + /* Socket Extensions for SCTP + * 5.3.1.3 SCTP_REMOTE_ERROR + * + * sre_type: + * It should be SCTP_REMOTE_ERROR. + */ + sre->sre_type = SCTP_REMOTE_ERROR; + + /* + * Socket Extensions for SCTP + * 5.3.1.3 SCTP_REMOTE_ERROR + * + * sre_flags: 16 bits (unsigned integer) + * Currently unused. + */ + sre->sre_flags = 0; + + /* Socket Extensions for SCTP + * 5.3.1.3 SCTP_REMOTE_ERROR + * + * sre_length: sizeof (__u32) + * + * This field is the total length of the notification data, + * including the notification header. + */ + sre->sre_length = skb->len; + + /* Socket Extensions for SCTP + * 5.3.1.3 SCTP_REMOTE_ERROR + * + * sre_error: 16 bits (unsigned integer) + * This value represents one of the Operational Error causes defined in + * the SCTP specification, in network byte order. + */ + sre->sre_error = cause; + + /* Socket Extensions for SCTP + * 5.3.1.3 SCTP_REMOTE_ERROR + * + * sre_assoc_id: sizeof (sctp_assoc_t) + * + * The association id field, holds the identifier for the association. + * All notifications for a given association have the same association + * identifier. For TCP style socket, this field is ignored. + */ + sctp_ulpevent_set_owner(event, asoc); + sre->sre_assoc_id = sctp_assoc2id(asoc); + + return event; + +fail: + return NULL; +} + +/* Create and initialize a SCTP_SEND_FAILED notification. + * + * Socket Extensions for SCTP - draft-01 + * 5.3.1.4 SCTP_SEND_FAILED + */ +struct sctp_ulpevent *sctp_ulpevent_make_send_failed( + const struct sctp_association *asoc, struct sctp_chunk *chunk, + __u16 flags, __u32 error, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_send_failed *ssf; + struct sk_buff *skb; + + /* Pull off any padding. */ + int len = ntohs(chunk->chunk_hdr->length); + + /* Make skb with more room so we can prepend notification. */ + skb = skb_copy_expand(chunk->skb, + sizeof(struct sctp_send_failed), /* headroom */ + 0, /* tailroom */ + gfp); + if (!skb) + goto fail; + + /* Pull off the common chunk header and DATA header. */ + skb_pull(skb, sizeof(struct sctp_data_chunk)); + len -= sizeof(struct sctp_data_chunk); + + /* Embed the event fields inside the cloned skb. */ + event = sctp_skb2event(skb); + sctp_ulpevent_init(event, MSG_NOTIFICATION); + + ssf = (struct sctp_send_failed *) + skb_push(skb, sizeof(struct sctp_send_failed)); + + /* Socket Extensions for SCTP + * 5.3.1.4 SCTP_SEND_FAILED + * + * ssf_type: + * It should be SCTP_SEND_FAILED. + */ + ssf->ssf_type = SCTP_SEND_FAILED; + + /* Socket Extensions for SCTP + * 5.3.1.4 SCTP_SEND_FAILED + * + * ssf_flags: 16 bits (unsigned integer) + * The flag value will take one of the following values + * + * SCTP_DATA_UNSENT - Indicates that the data was never put on + * the wire. + * + * SCTP_DATA_SENT - Indicates that the data was put on the wire. + * Note that this does not necessarily mean that the + * data was (or was not) successfully delivered. + */ + ssf->ssf_flags = flags; + + /* Socket Extensions for SCTP + * 5.3.1.4 SCTP_SEND_FAILED + * + * ssf_length: sizeof (__u32) + * This field is the total length of the notification data, including + * the notification header. + */ + ssf->ssf_length = sizeof(struct sctp_send_failed) + len; + skb_trim(skb, ssf->ssf_length); + + /* Socket Extensions for SCTP + * 5.3.1.4 SCTP_SEND_FAILED + * + * ssf_error: 16 bits (unsigned integer) + * This value represents the reason why the send failed, and if set, + * will be a SCTP protocol error code as defined in [SCTP] section + * 3.3.10. + */ + ssf->ssf_error = error; + + /* Socket Extensions for SCTP + * 5.3.1.4 SCTP_SEND_FAILED + * + * ssf_info: sizeof (struct sctp_sndrcvinfo) + * The original send information associated with the undelivered + * message. + */ + memcpy(&ssf->ssf_info, &chunk->sinfo, sizeof(struct sctp_sndrcvinfo)); + + /* Per TSVWG discussion with Randy. Allow the application to + * ressemble a fragmented message. + */ + ssf->ssf_info.sinfo_flags = chunk->chunk_hdr->flags; + + /* Socket Extensions for SCTP + * 5.3.1.4 SCTP_SEND_FAILED + * + * ssf_assoc_id: sizeof (sctp_assoc_t) + * The association id field, sf_assoc_id, holds the identifier for the + * association. All notifications for a given association have the + * same association identifier. For TCP style socket, this field is + * ignored. + */ + sctp_ulpevent_set_owner(event, asoc); + ssf->ssf_assoc_id = sctp_assoc2id(asoc); + return event; + +fail: + return NULL; +} + +/* Create and initialize a SCTP_SHUTDOWN_EVENT notification. + * + * Socket Extensions for SCTP - draft-01 + * 5.3.1.5 SCTP_SHUTDOWN_EVENT + */ +struct sctp_ulpevent *sctp_ulpevent_make_shutdown_event( + const struct sctp_association *asoc, + __u16 flags, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_shutdown_event *sse; + struct sk_buff *skb; + + event = sctp_ulpevent_new(sizeof(struct sctp_shutdown_event), + MSG_NOTIFICATION, gfp); + if (!event) + goto fail; + + skb = sctp_event2skb(event); + sse = (struct sctp_shutdown_event *) + skb_put(skb, sizeof(struct sctp_shutdown_event)); + + /* Socket Extensions for SCTP + * 5.3.1.5 SCTP_SHUTDOWN_EVENT + * + * sse_type + * It should be SCTP_SHUTDOWN_EVENT + */ + sse->sse_type = SCTP_SHUTDOWN_EVENT; + + /* Socket Extensions for SCTP + * 5.3.1.5 SCTP_SHUTDOWN_EVENT + * + * sse_flags: 16 bits (unsigned integer) + * Currently unused. + */ + sse->sse_flags = 0; + + /* Socket Extensions for SCTP + * 5.3.1.5 SCTP_SHUTDOWN_EVENT + * + * sse_length: sizeof (__u32) + * This field is the total length of the notification data, including + * the notification header. + */ + sse->sse_length = sizeof(struct sctp_shutdown_event); + + /* Socket Extensions for SCTP + * 5.3.1.5 SCTP_SHUTDOWN_EVENT + * + * sse_assoc_id: sizeof (sctp_assoc_t) + * The association id field, holds the identifier for the association. + * All notifications for a given association have the same association + * identifier. For TCP style socket, this field is ignored. + */ + sctp_ulpevent_set_owner(event, asoc); + sse->sse_assoc_id = sctp_assoc2id(asoc); + + return event; + +fail: + return NULL; +} + +/* Create and initialize a SCTP_ADAPTION_INDICATION notification. + * + * Socket Extensions for SCTP + * 5.3.1.6 SCTP_ADAPTION_INDICATION + */ +struct sctp_ulpevent *sctp_ulpevent_make_adaption_indication( + const struct sctp_association *asoc, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_adaption_event *sai; + struct sk_buff *skb; + + event = sctp_ulpevent_new(sizeof(struct sctp_adaption_event), + MSG_NOTIFICATION, gfp); + if (!event) + goto fail; + + skb = sctp_event2skb(event); + sai = (struct sctp_adaption_event *) + skb_put(skb, sizeof(struct sctp_adaption_event)); + + sai->sai_type = SCTP_ADAPTION_INDICATION; + sai->sai_flags = 0; + sai->sai_length = sizeof(struct sctp_adaption_event); + sai->sai_adaption_ind = asoc->peer.adaption_ind; + sctp_ulpevent_set_owner(event, asoc); + sai->sai_assoc_id = sctp_assoc2id(asoc); + + return event; + +fail: + return NULL; +} + +/* A message has been received. Package this message as a notification + * to pass it to the upper layers. Go ahead and calculate the sndrcvinfo + * even if filtered out later. + * + * Socket Extensions for SCTP + * 5.2.2 SCTP Header Information Structure (SCTP_SNDRCV) + */ +struct sctp_ulpevent *sctp_ulpevent_make_rcvmsg(struct sctp_association *asoc, + struct sctp_chunk *chunk, + int gfp) +{ + struct sctp_ulpevent *event = NULL; + struct sk_buff *skb; + size_t padding, len; + + /* Clone the original skb, sharing the data. */ + skb = skb_clone(chunk->skb, gfp); + if (!skb) + goto fail; + + /* First calculate the padding, so we don't inadvertently + * pass up the wrong length to the user. + * + * RFC 2960 - Section 3.2 Chunk Field Descriptions + * + * The total length of a chunk(including Type, Length and Value fields) + * MUST be a multiple of 4 bytes. If the length of the chunk is not a + * multiple of 4 bytes, the sender MUST pad the chunk with all zero + * bytes and this padding is not included in the chunk length field. + * The sender should never pad with more than 3 bytes. The receiver + * MUST ignore the padding bytes. + */ + len = ntohs(chunk->chunk_hdr->length); + padding = WORD_ROUND(len) - len; + + /* Fixup cloned skb with just this chunks data. */ + skb_trim(skb, chunk->chunk_end - padding - skb->data); + + /* Embed the event fields inside the cloned skb. */ + event = sctp_skb2event(skb); + + /* Initialize event with flags 0. */ + sctp_ulpevent_init(event, 0); + + sctp_ulpevent_receive_data(event, asoc); + + event->stream = ntohs(chunk->subh.data_hdr->stream); + event->ssn = ntohs(chunk->subh.data_hdr->ssn); + event->ppid = chunk->subh.data_hdr->ppid; + if (chunk->chunk_hdr->flags & SCTP_DATA_UNORDERED) { + event->flags |= MSG_UNORDERED; + event->cumtsn = sctp_tsnmap_get_ctsn(&asoc->peer.tsn_map); + } + event->tsn = ntohl(chunk->subh.data_hdr->tsn); + event->msg_flags |= chunk->chunk_hdr->flags; + event->iif = sctp_chunk_iif(chunk); + +fail: + return event; +} + +/* Create a partial delivery related event. + * + * 5.3.1.7 SCTP_PARTIAL_DELIVERY_EVENT + * + * When a receiver is engaged in a partial delivery of a + * message this notification will be used to indicate + * various events. + */ +struct sctp_ulpevent *sctp_ulpevent_make_pdapi( + const struct sctp_association *asoc, __u32 indication, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_pdapi_event *pd; + struct sk_buff *skb; + + event = sctp_ulpevent_new(sizeof(struct sctp_pdapi_event), + MSG_NOTIFICATION, gfp); + if (!event) + goto fail; + + skb = sctp_event2skb(event); + pd = (struct sctp_pdapi_event *) + skb_put(skb, sizeof(struct sctp_pdapi_event)); + + /* pdapi_type + * It should be SCTP_PARTIAL_DELIVERY_EVENT + * + * pdapi_flags: 16 bits (unsigned integer) + * Currently unused. + */ + pd->pdapi_type = SCTP_PARTIAL_DELIVERY_EVENT; + pd->pdapi_flags = 0; + + /* pdapi_length: 32 bits (unsigned integer) + * + * This field is the total length of the notification data, including + * the notification header. It will generally be sizeof (struct + * sctp_pdapi_event). + */ + pd->pdapi_length = sizeof(struct sctp_pdapi_event); + + /* pdapi_indication: 32 bits (unsigned integer) + * + * This field holds the indication being sent to the application. + */ + pd->pdapi_indication = indication; + + /* pdapi_assoc_id: sizeof (sctp_assoc_t) + * + * The association id field, holds the identifier for the association. + */ + sctp_ulpevent_set_owner(event, asoc); + pd->pdapi_assoc_id = sctp_assoc2id(asoc); + + return event; +fail: + return NULL; +} + +/* Return the notification type, assuming this is a notification + * event. + */ +__u16 sctp_ulpevent_get_notification_type(const struct sctp_ulpevent *event) +{ + union sctp_notification *notification; + struct sk_buff *skb; + + skb = sctp_event2skb((struct sctp_ulpevent *)event); + notification = (union sctp_notification *) skb->data; + return notification->sn_header.sn_type; +} + +/* Copy out the sndrcvinfo into a msghdr. */ +void sctp_ulpevent_read_sndrcvinfo(const struct sctp_ulpevent *event, + struct msghdr *msghdr) +{ + struct sctp_sndrcvinfo sinfo; + + if (sctp_ulpevent_is_notification(event)) + return; + + /* Sockets API Extensions for SCTP + * Section 5.2.2 SCTP Header Information Structure (SCTP_SNDRCV) + * + * sinfo_stream: 16 bits (unsigned integer) + * + * For recvmsg() the SCTP stack places the message's stream number in + * this value. + */ + sinfo.sinfo_stream = event->stream; + /* sinfo_ssn: 16 bits (unsigned integer) + * + * For recvmsg() this value contains the stream sequence number that + * the remote endpoint placed in the DATA chunk. For fragmented + * messages this is the same number for all deliveries of the message + * (if more than one recvmsg() is needed to read the message). + */ + sinfo.sinfo_ssn = event->ssn; + /* sinfo_ppid: 32 bits (unsigned integer) + * + * In recvmsg() this value is + * the same information that was passed by the upper layer in the peer + * application. Please note that byte order issues are NOT accounted + * for and this information is passed opaquely by the SCTP stack from + * one end to the other. + */ + sinfo.sinfo_ppid = event->ppid; + /* sinfo_flags: 16 bits (unsigned integer) + * + * This field may contain any of the following flags and is composed of + * a bitwise OR of these values. + * + * recvmsg() flags: + * + * MSG_UNORDERED - This flag is present when the message was sent + * non-ordered. + */ + sinfo.sinfo_flags = event->flags; + /* sinfo_tsn: 32 bit (unsigned integer) + * + * For the receiving side, this field holds a TSN that was + * assigned to one of the SCTP Data Chunks. + */ + sinfo.sinfo_tsn = event->tsn; + /* sinfo_cumtsn: 32 bit (unsigned integer) + * + * This field will hold the current cumulative TSN as + * known by the underlying SCTP layer. Note this field is + * ignored when sending and only valid for a receive + * operation when sinfo_flags are set to MSG_UNORDERED. + */ + sinfo.sinfo_cumtsn = event->cumtsn; + /* sinfo_assoc_id: sizeof (sctp_assoc_t) + * + * The association handle field, sinfo_assoc_id, holds the identifier + * for the association announced in the COMMUNICATION_UP notification. + * All notifications for a given association have the same identifier. + * Ignored for one-to-one style sockets. + */ + sinfo.sinfo_assoc_id = sctp_assoc2id(event->asoc); + + /* These fields are not used while receiving. */ + sinfo.sinfo_context = 0; + sinfo.sinfo_timetolive = 0; + + put_cmsg(msghdr, IPPROTO_SCTP, SCTP_SNDRCV, + sizeof(struct sctp_sndrcvinfo), (void *)&sinfo); +} + +/* Do accounting for bytes received and hold a reference to the association + * for each skb. + */ +static void sctp_ulpevent_receive_data(struct sctp_ulpevent *event, + struct sctp_association *asoc) +{ + struct sk_buff *skb, *frag; + + skb = sctp_event2skb(event); + /* Set the owner and charge rwnd for bytes received. */ + sctp_ulpevent_set_owner(event, asoc); + sctp_assoc_rwnd_decrease(asoc, skb_headlen(skb)); + + if (!skb->data_len) + return; + + /* Note: Not clearing the entire event struct as this is just a + * fragment of the real event. However, we still need to do rwnd + * accounting. + * In general, the skb passed from IP can have only 1 level of + * fragments. But we allow multiple levels of fragments. + */ + for (frag = skb_shinfo(skb)->frag_list; frag; frag = frag->next) { + sctp_ulpevent_receive_data(sctp_skb2event(frag), asoc); + } +} + +/* Do accounting for bytes just read by user and release the references to + * the association. + */ +static void sctp_ulpevent_release_data(struct sctp_ulpevent *event) +{ + struct sk_buff *skb, *frag; + + /* Current stack structures assume that the rcv buffer is + * per socket. For UDP style sockets this is not true as + * multiple associations may be on a single UDP-style socket. + * Use the local private area of the skb to track the owning + * association. + */ + + skb = sctp_event2skb(event); + sctp_assoc_rwnd_increase(event->asoc, skb_headlen(skb)); + + if (!skb->data_len) + goto done; + + /* Don't forget the fragments. */ + for (frag = skb_shinfo(skb)->frag_list; frag; frag = frag->next) { + /* NOTE: skb_shinfos are recursive. Although IP returns + * skb's with only 1 level of fragments, SCTP reassembly can + * increase the levels. + */ + sctp_ulpevent_release_data(sctp_skb2event(frag)); + } + +done: + sctp_ulpevent_release_owner(event); +} + +/* Free a ulpevent that has an owner. It includes releasing the reference + * to the owner, updating the rwnd in case of a DATA event and freeing the + * skb. + * See comments in sctp_stub_rfree(). + */ +void sctp_ulpevent_free(struct sctp_ulpevent *event) +{ + if (sctp_ulpevent_is_notification(event)) + sctp_ulpevent_release_owner(event); + else + sctp_ulpevent_release_data(event); + + kfree_skb(sctp_event2skb(event)); +} + +/* Purge the skb lists holding ulpevents. */ +void sctp_queue_purge_ulpevents(struct sk_buff_head *list) +{ + struct sk_buff *skb; + while ((skb = skb_dequeue(list)) != NULL) + sctp_ulpevent_free(sctp_skb2event(skb)); +} diff --git a/net/sctp/ulpqueue.c b/net/sctp/ulpqueue.c new file mode 100644 index 000000000000..d5dd2cf7ac4a --- /dev/null +++ b/net/sctp/ulpqueue.c @@ -0,0 +1,864 @@ +/* SCTP kernel reference Implementation + * (C) Copyright IBM Corp. 2001, 2004 + * Copyright (c) 1999-2000 Cisco, Inc. + * Copyright (c) 1999-2001 Motorola, Inc. + * Copyright (c) 2001 Intel Corp. + * Copyright (c) 2001 Nokia, Inc. + * Copyright (c) 2001 La Monte H.P. Yarroll + * + * This abstraction carries sctp events to the ULP (sockets). + * + * The SCTP reference 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 + * will be useful, but WITHOUT ANY WARRANTY; without even the implied + * ************************ + * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * See the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with GNU CC; see the file COPYING. If not, write to + * the Free Software Foundation, 59 Temple Place - Suite 330, + * Boston, MA 02111-1307, USA. + * + * Please send any bug reports or fixes you make to the + * email address(es): + * lksctp developers <lksctp-developers@lists.sourceforge.net> + * + * Or submit a bug report through the following website: + * http://www.sf.net/projects/lksctp + * + * Written or modified by: + * Jon Grimm <jgrimm@us.ibm.com> + * La Monte H.P. Yarroll <piggy@acm.org> + * Sridhar Samudrala <sri@us.ibm.com> + * + * Any bugs reported given to us we will try to fix... any fixes shared will + * be incorporated into the next SCTP release. + */ + +#include <linux/types.h> +#include <linux/skbuff.h> +#include <net/sock.h> +#include <net/sctp/structs.h> +#include <net/sctp/sctp.h> +#include <net/sctp/sm.h> + +/* Forward declarations for internal helpers. */ +static struct sctp_ulpevent * sctp_ulpq_reasm(struct sctp_ulpq *ulpq, + struct sctp_ulpevent *); +static struct sctp_ulpevent * sctp_ulpq_order(struct sctp_ulpq *, + struct sctp_ulpevent *); + +/* 1st Level Abstractions */ + +/* Initialize a ULP queue from a block of memory. */ +struct sctp_ulpq *sctp_ulpq_init(struct sctp_ulpq *ulpq, + struct sctp_association *asoc) +{ + memset(ulpq, 0, sizeof(struct sctp_ulpq)); + + ulpq->asoc = asoc; + skb_queue_head_init(&ulpq->reasm); + skb_queue_head_init(&ulpq->lobby); + ulpq->pd_mode = 0; + ulpq->malloced = 0; + + return ulpq; +} + + +/* Flush the reassembly and ordering queues. */ +static void sctp_ulpq_flush(struct sctp_ulpq *ulpq) +{ + struct sk_buff *skb; + struct sctp_ulpevent *event; + + while ((skb = __skb_dequeue(&ulpq->lobby)) != NULL) { + event = sctp_skb2event(skb); + sctp_ulpevent_free(event); + } + + while ((skb = __skb_dequeue(&ulpq->reasm)) != NULL) { + event = sctp_skb2event(skb); + sctp_ulpevent_free(event); + } + +} + +/* Dispose of a ulpqueue. */ +void sctp_ulpq_free(struct sctp_ulpq *ulpq) +{ + sctp_ulpq_flush(ulpq); + if (ulpq->malloced) + kfree(ulpq); +} + +/* Process an incoming DATA chunk. */ +int sctp_ulpq_tail_data(struct sctp_ulpq *ulpq, struct sctp_chunk *chunk, + int gfp) +{ + struct sk_buff_head temp; + sctp_data_chunk_t *hdr; + struct sctp_ulpevent *event; + + hdr = (sctp_data_chunk_t *) chunk->chunk_hdr; + + /* Create an event from the incoming chunk. */ + event = sctp_ulpevent_make_rcvmsg(chunk->asoc, chunk, gfp); + if (!event) + return -ENOMEM; + + /* Do reassembly if needed. */ + event = sctp_ulpq_reasm(ulpq, event); + + /* Do ordering if needed. */ + if ((event) && (event->msg_flags & MSG_EOR)){ + /* Create a temporary list to collect chunks on. */ + skb_queue_head_init(&temp); + __skb_queue_tail(&temp, sctp_event2skb(event)); + + event = sctp_ulpq_order(ulpq, event); + } + + /* Send event to the ULP. */ + if (event) + sctp_ulpq_tail_event(ulpq, event); + + return 0; +} + +/* Add a new event for propagation to the ULP. */ +/* Clear the partial delivery mode for this socket. Note: This + * assumes that no association is currently in partial delivery mode. + */ +int sctp_clear_pd(struct sock *sk) +{ + struct sctp_sock *sp = sctp_sk(sk); + + sp->pd_mode = 0; + if (!skb_queue_empty(&sp->pd_lobby)) { + struct list_head *list; + sctp_skb_list_tail(&sp->pd_lobby, &sk->sk_receive_queue); + list = (struct list_head *)&sctp_sk(sk)->pd_lobby; + INIT_LIST_HEAD(list); + return 1; + } + return 0; +} + +/* Clear the pd_mode and restart any pending messages waiting for delivery. */ +static int sctp_ulpq_clear_pd(struct sctp_ulpq *ulpq) +{ + ulpq->pd_mode = 0; + return sctp_clear_pd(ulpq->asoc->base.sk); +} + + + +int sctp_ulpq_tail_event(struct sctp_ulpq *ulpq, struct sctp_ulpevent *event) +{ + struct sock *sk = ulpq->asoc->base.sk; + struct sk_buff_head *queue; + int clear_pd = 0; + + /* If the socket is just going to throw this away, do not + * even try to deliver it. + */ + if (sock_flag(sk, SOCK_DEAD) || (sk->sk_shutdown & RCV_SHUTDOWN)) + goto out_free; + + /* Check if the user wishes to receive this event. */ + if (!sctp_ulpevent_is_enabled(event, &sctp_sk(sk)->subscribe)) + goto out_free; + + /* If we are in partial delivery mode, post to the lobby until + * partial delivery is cleared, unless, of course _this_ is + * the association the cause of the partial delivery. + */ + + if (!sctp_sk(sk)->pd_mode) { + queue = &sk->sk_receive_queue; + } else if (ulpq->pd_mode) { + if (event->msg_flags & MSG_NOTIFICATION) + queue = &sctp_sk(sk)->pd_lobby; + else { + clear_pd = event->msg_flags & MSG_EOR; + queue = &sk->sk_receive_queue; + } + } else + queue = &sctp_sk(sk)->pd_lobby; + + + /* If we are harvesting multiple skbs they will be + * collected on a list. + */ + if (sctp_event2skb(event)->list) + sctp_skb_list_tail(sctp_event2skb(event)->list, queue); + else + __skb_queue_tail(queue, sctp_event2skb(event)); + + /* Did we just complete partial delivery and need to get + * rolling again? Move pending data to the receive + * queue. + */ + if (clear_pd) + sctp_ulpq_clear_pd(ulpq); + + if (queue == &sk->sk_receive_queue) + sk->sk_data_ready(sk, 0); + return 1; + +out_free: + if (sctp_event2skb(event)->list) + sctp_queue_purge_ulpevents(sctp_event2skb(event)->list); + else + sctp_ulpevent_free(event); + return 0; +} + +/* 2nd Level Abstractions */ + +/* Helper function to store chunks that need to be reassembled. */ +static inline void sctp_ulpq_store_reasm(struct sctp_ulpq *ulpq, + struct sctp_ulpevent *event) +{ + struct sk_buff *pos; + struct sctp_ulpevent *cevent; + __u32 tsn, ctsn; + + tsn = event->tsn; + + /* See if it belongs at the end. */ + pos = skb_peek_tail(&ulpq->reasm); + if (!pos) { + __skb_queue_tail(&ulpq->reasm, sctp_event2skb(event)); + return; + } + + /* Short circuit just dropping it at the end. */ + cevent = sctp_skb2event(pos); + ctsn = cevent->tsn; + if (TSN_lt(ctsn, tsn)) { + __skb_queue_tail(&ulpq->reasm, sctp_event2skb(event)); + return; + } + + /* Find the right place in this list. We store them by TSN. */ + skb_queue_walk(&ulpq->reasm, pos) { + cevent = sctp_skb2event(pos); + ctsn = cevent->tsn; + + if (TSN_lt(tsn, ctsn)) + break; + } + + /* Insert before pos. */ + __skb_insert(sctp_event2skb(event), pos->prev, pos, &ulpq->reasm); + +} + +/* Helper function to return an event corresponding to the reassembled + * datagram. + * This routine creates a re-assembled skb given the first and last skb's + * as stored in the reassembly queue. The skb's may be non-linear if the sctp + * payload was fragmented on the way and ip had to reassemble them. + * We add the rest of skb's to the first skb's fraglist. + */ +static struct sctp_ulpevent *sctp_make_reassembled_event(struct sk_buff *f_frag, struct sk_buff *l_frag) +{ + struct sk_buff *pos; + struct sctp_ulpevent *event; + struct sk_buff *pnext, *last; + struct sk_buff *list = skb_shinfo(f_frag)->frag_list; + + /* Store the pointer to the 2nd skb */ + if (f_frag == l_frag) + pos = NULL; + else + pos = f_frag->next; + + /* Get the last skb in the f_frag's frag_list if present. */ + for (last = list; list; last = list, list = list->next); + + /* Add the list of remaining fragments to the first fragments + * frag_list. + */ + if (last) + last->next = pos; + else + skb_shinfo(f_frag)->frag_list = pos; + + /* Remove the first fragment from the reassembly queue. */ + __skb_unlink(f_frag, f_frag->list); + while (pos) { + + pnext = pos->next; + + /* Update the len and data_len fields of the first fragment. */ + f_frag->len += pos->len; + f_frag->data_len += pos->len; + + /* Remove the fragment from the reassembly queue. */ + __skb_unlink(pos, pos->list); + + /* Break if we have reached the last fragment. */ + if (pos == l_frag) + break; + pos->next = pnext; + pos = pnext; + }; + + event = sctp_skb2event(f_frag); + SCTP_INC_STATS(SCTP_MIB_REASMUSRMSGS); + + return event; +} + + +/* Helper function to check if an incoming chunk has filled up the last + * missing fragment in a SCTP datagram and return the corresponding event. + */ +static inline struct sctp_ulpevent *sctp_ulpq_retrieve_reassembled(struct sctp_ulpq *ulpq) +{ + struct sk_buff *pos; + struct sctp_ulpevent *cevent; + struct sk_buff *first_frag = NULL; + __u32 ctsn, next_tsn; + struct sctp_ulpevent *retval = NULL; + + /* Initialized to 0 just to avoid compiler warning message. Will + * never be used with this value. It is referenced only after it + * is set when we find the first fragment of a message. + */ + next_tsn = 0; + + /* The chunks are held in the reasm queue sorted by TSN. + * Walk through the queue sequentially and look for a sequence of + * fragmented chunks that complete a datagram. + * 'first_frag' and next_tsn are reset when we find a chunk which + * is the first fragment of a datagram. Once these 2 fields are set + * we expect to find the remaining middle fragments and the last + * fragment in order. If not, first_frag is reset to NULL and we + * start the next pass when we find another first fragment. + */ + skb_queue_walk(&ulpq->reasm, pos) { + cevent = sctp_skb2event(pos); + ctsn = cevent->tsn; + + switch (cevent->msg_flags & SCTP_DATA_FRAG_MASK) { + case SCTP_DATA_FIRST_FRAG: + first_frag = pos; + next_tsn = ctsn + 1; + break; + + case SCTP_DATA_MIDDLE_FRAG: + if ((first_frag) && (ctsn == next_tsn)) + next_tsn++; + else + first_frag = NULL; + break; + + case SCTP_DATA_LAST_FRAG: + if (first_frag && (ctsn == next_tsn)) + goto found; + else + first_frag = NULL; + break; + }; + + } +done: + return retval; +found: + retval = sctp_make_reassembled_event(first_frag, pos); + if (retval) + retval->msg_flags |= MSG_EOR; + goto done; +} + +/* Retrieve the next set of fragments of a partial message. */ +static inline struct sctp_ulpevent *sctp_ulpq_retrieve_partial(struct sctp_ulpq *ulpq) +{ + struct sk_buff *pos, *last_frag, *first_frag; + struct sctp_ulpevent *cevent; + __u32 ctsn, next_tsn; + int is_last; + struct sctp_ulpevent *retval; + + /* The chunks are held in the reasm queue sorted by TSN. + * Walk through the queue sequentially and look for the first + * sequence of fragmented chunks. + */ + + if (skb_queue_empty(&ulpq->reasm)) + return NULL; + + last_frag = first_frag = NULL; + retval = NULL; + next_tsn = 0; + is_last = 0; + + skb_queue_walk(&ulpq->reasm, pos) { + cevent = sctp_skb2event(pos); + ctsn = cevent->tsn; + + switch (cevent->msg_flags & SCTP_DATA_FRAG_MASK) { + case SCTP_DATA_MIDDLE_FRAG: + if (!first_frag) { + first_frag = pos; + next_tsn = ctsn + 1; + last_frag = pos; + } else if (next_tsn == ctsn) + next_tsn++; + else + goto done; + break; + case SCTP_DATA_LAST_FRAG: + if (!first_frag) + first_frag = pos; + else if (ctsn != next_tsn) + goto done; + last_frag = pos; + is_last = 1; + goto done; + default: + return NULL; + }; + } + + /* We have the reassembled event. There is no need to look + * further. + */ +done: + retval = sctp_make_reassembled_event(first_frag, last_frag); + if (retval && is_last) + retval->msg_flags |= MSG_EOR; + + return retval; +} + + +/* Helper function to reassemble chunks. Hold chunks on the reasm queue that + * need reassembling. + */ +static struct sctp_ulpevent *sctp_ulpq_reasm(struct sctp_ulpq *ulpq, + struct sctp_ulpevent *event) +{ + struct sctp_ulpevent *retval = NULL; + + /* Check if this is part of a fragmented message. */ + if (SCTP_DATA_NOT_FRAG == (event->msg_flags & SCTP_DATA_FRAG_MASK)) { + event->msg_flags |= MSG_EOR; + return event; + } + + sctp_ulpq_store_reasm(ulpq, event); + if (!ulpq->pd_mode) + retval = sctp_ulpq_retrieve_reassembled(ulpq); + else { + __u32 ctsn, ctsnap; + + /* Do not even bother unless this is the next tsn to + * be delivered. + */ + ctsn = event->tsn; + ctsnap = sctp_tsnmap_get_ctsn(&ulpq->asoc->peer.tsn_map); + if (TSN_lte(ctsn, ctsnap)) + retval = sctp_ulpq_retrieve_partial(ulpq); + } + + return retval; +} + +/* Retrieve the first part (sequential fragments) for partial delivery. */ +static inline struct sctp_ulpevent *sctp_ulpq_retrieve_first(struct sctp_ulpq *ulpq) +{ + struct sk_buff *pos, *last_frag, *first_frag; + struct sctp_ulpevent *cevent; + __u32 ctsn, next_tsn; + struct sctp_ulpevent *retval; + + /* The chunks are held in the reasm queue sorted by TSN. + * Walk through the queue sequentially and look for a sequence of + * fragmented chunks that start a datagram. + */ + + if (skb_queue_empty(&ulpq->reasm)) + return NULL; + + last_frag = first_frag = NULL; + retval = NULL; + next_tsn = 0; + + skb_queue_walk(&ulpq->reasm, pos) { + cevent = sctp_skb2event(pos); + ctsn = cevent->tsn; + + switch (cevent->msg_flags & SCTP_DATA_FRAG_MASK) { + case SCTP_DATA_FIRST_FRAG: + if (!first_frag) { + first_frag = pos; + next_tsn = ctsn + 1; + last_frag = pos; + } else + goto done; + break; + + case SCTP_DATA_MIDDLE_FRAG: + if (!first_frag) + return NULL; + if (ctsn == next_tsn) { + next_tsn++; + last_frag = pos; + } else + goto done; + break; + default: + return NULL; + }; + } + + /* We have the reassembled event. There is no need to look + * further. + */ +done: + retval = sctp_make_reassembled_event(first_frag, last_frag); + return retval; +} + +/* Helper function to gather skbs that have possibly become + * ordered by an an incoming chunk. + */ +static inline void sctp_ulpq_retrieve_ordered(struct sctp_ulpq *ulpq, + struct sctp_ulpevent *event) +{ + struct sk_buff *pos, *tmp; + struct sctp_ulpevent *cevent; + struct sctp_stream *in; + __u16 sid, csid; + __u16 ssn, cssn; + + sid = event->stream; + ssn = event->ssn; + in = &ulpq->asoc->ssnmap->in; + + /* We are holding the chunks by stream, by SSN. */ + sctp_skb_for_each(pos, &ulpq->lobby, tmp) { + cevent = (struct sctp_ulpevent *) pos->cb; + csid = cevent->stream; + cssn = cevent->ssn; + + /* Have we gone too far? */ + if (csid > sid) + break; + + /* Have we not gone far enough? */ + if (csid < sid) + continue; + + if (cssn != sctp_ssn_peek(in, sid)) + break; + + /* Found it, so mark in the ssnmap. */ + sctp_ssn_next(in, sid); + + __skb_unlink(pos, pos->list); + + /* Attach all gathered skbs to the event. */ + __skb_queue_tail(sctp_event2skb(event)->list, pos); + } +} + +/* Helper function to store chunks needing ordering. */ +static inline void sctp_ulpq_store_ordered(struct sctp_ulpq *ulpq, + struct sctp_ulpevent *event) +{ + struct sk_buff *pos; + struct sctp_ulpevent *cevent; + __u16 sid, csid; + __u16 ssn, cssn; + + pos = skb_peek_tail(&ulpq->lobby); + if (!pos) { + __skb_queue_tail(&ulpq->lobby, sctp_event2skb(event)); + return; + } + + sid = event->stream; + ssn = event->ssn; + + cevent = (struct sctp_ulpevent *) pos->cb; + csid = cevent->stream; + cssn = cevent->ssn; + if (sid > csid) { + __skb_queue_tail(&ulpq->lobby, sctp_event2skb(event)); + return; + } + + if ((sid == csid) && SSN_lt(cssn, ssn)) { + __skb_queue_tail(&ulpq->lobby, sctp_event2skb(event)); + return; + } + + /* Find the right place in this list. We store them by + * stream ID and then by SSN. + */ + skb_queue_walk(&ulpq->lobby, pos) { + cevent = (struct sctp_ulpevent *) pos->cb; + csid = cevent->stream; + cssn = cevent->ssn; + + if (csid > sid) + break; + if (csid == sid && SSN_lt(ssn, cssn)) + break; + } + + + /* Insert before pos. */ + __skb_insert(sctp_event2skb(event), pos->prev, pos, &ulpq->lobby); + +} + +static struct sctp_ulpevent *sctp_ulpq_order(struct sctp_ulpq *ulpq, + struct sctp_ulpevent *event) +{ + __u16 sid, ssn; + struct sctp_stream *in; + + /* Check if this message needs ordering. */ + if (SCTP_DATA_UNORDERED & event->msg_flags) + return event; + + /* Note: The stream ID must be verified before this routine. */ + sid = event->stream; + ssn = event->ssn; + in = &ulpq->asoc->ssnmap->in; + + /* Is this the expected SSN for this stream ID? */ + if (ssn != sctp_ssn_peek(in, sid)) { + /* We've received something out of order, so find where it + * needs to be placed. We order by stream and then by SSN. + */ + sctp_ulpq_store_ordered(ulpq, event); + return NULL; + } + + /* Mark that the next chunk has been found. */ + sctp_ssn_next(in, sid); + + /* Go find any other chunks that were waiting for + * ordering. + */ + sctp_ulpq_retrieve_ordered(ulpq, event); + + return event; +} + +/* Helper function to gather skbs that have possibly become + * ordered by forward tsn skipping their dependencies. + */ +static inline void sctp_ulpq_reap_ordered(struct sctp_ulpq *ulpq) +{ + struct sk_buff *pos, *tmp; + struct sctp_ulpevent *cevent; + struct sctp_ulpevent *event = NULL; + struct sctp_stream *in; + struct sk_buff_head temp; + __u16 csid, cssn; + + in = &ulpq->asoc->ssnmap->in; + + /* We are holding the chunks by stream, by SSN. */ + sctp_skb_for_each(pos, &ulpq->lobby, tmp) { + cevent = (struct sctp_ulpevent *) pos->cb; + csid = cevent->stream; + cssn = cevent->ssn; + + if (cssn != sctp_ssn_peek(in, csid)) + break; + + /* Found it, so mark in the ssnmap. */ + sctp_ssn_next(in, csid); + + __skb_unlink(pos, pos->list); + if (!event) { + /* Create a temporary list to collect chunks on. */ + event = sctp_skb2event(pos); + skb_queue_head_init(&temp); + __skb_queue_tail(&temp, sctp_event2skb(event)); + } else { + /* Attach all gathered skbs to the event. */ + __skb_queue_tail(sctp_event2skb(event)->list, pos); + } + } + + /* Send event to the ULP. */ + if (event) + sctp_ulpq_tail_event(ulpq, event); +} + +/* Skip over an SSN. */ +void sctp_ulpq_skip(struct sctp_ulpq *ulpq, __u16 sid, __u16 ssn) +{ + struct sctp_stream *in; + + /* Note: The stream ID must be verified before this routine. */ + in = &ulpq->asoc->ssnmap->in; + + /* Is this an old SSN? If so ignore. */ + if (SSN_lt(ssn, sctp_ssn_peek(in, sid))) + return; + + /* Mark that we are no longer expecting this SSN or lower. */ + sctp_ssn_skip(in, sid, ssn); + + /* Go find any other chunks that were waiting for + * ordering and deliver them if needed. + */ + sctp_ulpq_reap_ordered(ulpq); + return; +} + +/* Renege 'needed' bytes from the ordering queue. */ +static __u16 sctp_ulpq_renege_order(struct sctp_ulpq *ulpq, __u16 needed) +{ + __u16 freed = 0; + __u32 tsn; + struct sk_buff *skb; + struct sctp_ulpevent *event; + struct sctp_tsnmap *tsnmap; + + tsnmap = &ulpq->asoc->peer.tsn_map; + + while ((skb = __skb_dequeue_tail(&ulpq->lobby)) != NULL) { + freed += skb_headlen(skb); + event = sctp_skb2event(skb); + tsn = event->tsn; + + sctp_ulpevent_free(event); + sctp_tsnmap_renege(tsnmap, tsn); + if (freed >= needed) + return freed; + } + + return freed; +} + +/* Renege 'needed' bytes from the reassembly queue. */ +static __u16 sctp_ulpq_renege_frags(struct sctp_ulpq *ulpq, __u16 needed) +{ + __u16 freed = 0; + __u32 tsn; + struct sk_buff *skb; + struct sctp_ulpevent *event; + struct sctp_tsnmap *tsnmap; + + tsnmap = &ulpq->asoc->peer.tsn_map; + + /* Walk backwards through the list, reneges the newest tsns. */ + while ((skb = __skb_dequeue_tail(&ulpq->reasm)) != NULL) { + freed += skb_headlen(skb); + event = sctp_skb2event(skb); + tsn = event->tsn; + + sctp_ulpevent_free(event); + sctp_tsnmap_renege(tsnmap, tsn); + if (freed >= needed) + return freed; + } + + return freed; +} + +/* Partial deliver the first message as there is pressure on rwnd. */ +void sctp_ulpq_partial_delivery(struct sctp_ulpq *ulpq, + struct sctp_chunk *chunk, int gfp) +{ + struct sctp_ulpevent *event; + struct sctp_association *asoc; + + asoc = ulpq->asoc; + + /* Are we already in partial delivery mode? */ + if (!sctp_sk(asoc->base.sk)->pd_mode) { + + /* Is partial delivery possible? */ + event = sctp_ulpq_retrieve_first(ulpq); + /* Send event to the ULP. */ + if (event) { + sctp_ulpq_tail_event(ulpq, event); + sctp_sk(asoc->base.sk)->pd_mode = 1; + ulpq->pd_mode = 1; + return; + } + } +} + +/* Renege some packets to make room for an incoming chunk. */ +void sctp_ulpq_renege(struct sctp_ulpq *ulpq, struct sctp_chunk *chunk, + int gfp) +{ + struct sctp_association *asoc; + __u16 needed, freed; + + asoc = ulpq->asoc; + + if (chunk) { + needed = ntohs(chunk->chunk_hdr->length); + needed -= sizeof(sctp_data_chunk_t); + } else + needed = SCTP_DEFAULT_MAXWINDOW; + + freed = 0; + + if (skb_queue_empty(&asoc->base.sk->sk_receive_queue)) { + freed = sctp_ulpq_renege_order(ulpq, needed); + if (freed < needed) { + freed += sctp_ulpq_renege_frags(ulpq, needed - freed); + } + } + /* If able to free enough room, accept this chunk. */ + if (chunk && (freed >= needed)) { + __u32 tsn; + tsn = ntohl(chunk->subh.data_hdr->tsn); + sctp_tsnmap_mark(&asoc->peer.tsn_map, tsn); + sctp_ulpq_tail_data(ulpq, chunk, gfp); + + sctp_ulpq_partial_delivery(ulpq, chunk, gfp); + } + + return; +} + + + +/* Notify the application if an association is aborted and in + * partial delivery mode. Send up any pending received messages. + */ +void sctp_ulpq_abort_pd(struct sctp_ulpq *ulpq, int gfp) +{ + struct sctp_ulpevent *ev = NULL; + struct sock *sk; + + if (!ulpq->pd_mode) + return; + + sk = ulpq->asoc->base.sk; + if (sctp_ulpevent_type_enabled(SCTP_PARTIAL_DELIVERY_EVENT, + &sctp_sk(sk)->subscribe)) + ev = sctp_ulpevent_make_pdapi(ulpq->asoc, + SCTP_PARTIAL_DELIVERY_ABORTED, + gfp); + if (ev) + __skb_queue_tail(&sk->sk_receive_queue, sctp_event2skb(ev)); + + /* If there is data waiting, send it up the socket now. */ + if (sctp_ulpq_clear_pd(ulpq) || ev) + sk->sk_data_ready(sk, 0); +} |