X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=net%2Fsctp%2Fprotocol.c;h=25be8f04de6e6dd0ee2cfcaffa3b076603800d05;hb=577f99c1d08cf9cbdafd4e858dd13ff04d855090;hp=c20f05b3308fca4260ce2c7466ce94059b4eaa16;hpb=6244be4e063075f6077f05e70f8fa1bf7f4a968e;p=safe%2Fjmp%2Flinux-2.6 diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index c20f05b..25be8f0 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1,4 +1,4 @@ -/* SCTP kernel reference Implementation +/* SCTP kernel implementation * (C) Copyright IBM Corp. 2001, 2004 * Copyright (c) 1999-2000 Cisco, Inc. * Copyright (c) 1999-2001 Motorola, Inc. @@ -6,17 +6,17 @@ * Copyright (c) 2001 Nokia, Inc. * Copyright (c) 2001 La Monte H.P. Yarroll * - * This file is part of the SCTP kernel reference Implementation + * This file is part of the SCTP kernel implementation * * Initialization/cleanup for SCTP protocol support. * - * The SCTP reference implementation is free software; + * This SCTP implementation is free software; * you can redistribute it and/or modify it under the terms of * the GNU General Public License as published by * the Free Software Foundation; either version 2, or (at your option) * any later version. * - * The SCTP reference implementation is distributed in the hope that it + * This SCTP implementation is distributed in the hope that it * will be useful, but WITHOUT ANY WARRANTY; without even the implied * ************************ * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. @@ -51,6 +51,8 @@ #include #include #include +#include +#include #include #include #include @@ -79,8 +81,12 @@ 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 __read_mostly; -kmem_cache_t *sctp_bucket_cachep __read_mostly; +struct kmem_cache *sctp_chunk_cachep __read_mostly; +struct kmem_cache *sctp_bucket_cachep __read_mostly; + +int sysctl_sctp_mem[3]; +int sysctl_sctp_rmem[3]; +int sysctl_sctp_wmem[3]; /* Return the address of the control sock. */ struct sock *sctp_get_ctl_sock(void) @@ -93,7 +99,7 @@ static __init int sctp_proc_init(void) { if (!proc_net_sctp) { struct proc_dir_entry *ent; - ent = proc_mkdir("net/sctp", NULL); + ent = proc_mkdir("sctp", init_net.proc_net); if (ent) { ent->owner = THIS_MODULE; proc_net_sctp = ent; @@ -102,11 +108,11 @@ static __init int sctp_proc_init(void) } if (sctp_snmp_proc_init()) - goto out_nomem; + goto out_nomem; if (sctp_eps_proc_init()) - goto out_nomem; + goto out_nomem; if (sctp_assocs_proc_init()) - goto out_nomem; + goto out_nomem; return 0; @@ -114,7 +120,7 @@ out_nomem: return -ENOMEM; } -/* Clean up the proc fs entry for the SCTP protocol. +/* 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. */ @@ -126,7 +132,7 @@ static void sctp_proc_exit(void) if (proc_net_sctp) { proc_net_sctp = NULL; - remove_proc_entry("net/sctp", NULL); + remove_proc_entry("sctp", init_net.proc_net); } } @@ -153,7 +159,9 @@ static void sctp_v4_copy_addrlist(struct list_head *addrlist, addr->a.v4.sin_family = AF_INET; addr->a.v4.sin_port = 0; addr->a.v4.sin_addr.s_addr = ifa->ifa_local; - addr->a_h = addr->a; + addr->valid = 1; + INIT_LIST_HEAD(&addr->list); + INIT_RCU_HEAD(&addr->rcu); list_add_tail(&addr->list, addrlist); } } @@ -164,14 +172,14 @@ static void sctp_v4_copy_addrlist(struct list_head *addrlist, /* Extract our IP addresses from the system and stash them in the * protocol structure. */ -static void __sctp_get_local_addr_list(void) +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) { + for_each_netdev(&init_net, dev) { __list_for_each(pos, &sctp_address_families) { af = list_entry(pos, struct sctp_af, list); af->copy_addrlist(&sctp_local_addr_list, dev); @@ -180,17 +188,8 @@ static void __sctp_get_local_addr_list(void) 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) +static void sctp_free_local_addr_list(void) { struct sctp_sockaddr_entry *addr; struct list_head *pos, *temp; @@ -202,14 +201,11 @@ static void __sctp_free_local_addr_list(void) } } -/* Free the existing local addresses. */ -static void sctp_free_local_addr_list(void) +void sctp_local_addr_free(struct rcu_head *head) { - 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); + struct sctp_sockaddr_entry *e = container_of(head, + struct sctp_sockaddr_entry, rcu); + kfree(e); } /* Copy the local addresses which are valid for 'scope' into 'bp'. */ @@ -218,12 +214,11 @@ int sctp_copy_local_addr_list(struct sctp_bind_addr *bp, sctp_scope_t scope, { 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); + rcu_read_lock(); + list_for_each_entry_rcu(addr, &sctp_local_addr_list, list) { + if (!addr->valid) + continue; 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 @@ -234,8 +229,8 @@ int sctp_copy_local_addr_list(struct sctp_bind_addr *bp, sctp_scope_t scope, (((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, 1, - GFP_ATOMIC); + error = sctp_add_bind_addr(bp, &addr->a, + SCTP_ADDR_SRC, GFP_ATOMIC); if (error) goto end_copy; } @@ -243,7 +238,7 @@ int sctp_copy_local_addr_list(struct sctp_bind_addr *bp, sctp_scope_t scope, } end_copy: - sctp_spin_unlock_irqrestore(&sctp_local_addr_lock, flags); + rcu_read_unlock(); return error; } @@ -258,13 +253,13 @@ static void sctp_v4_from_skb(union sctp_addr *addr, struct sk_buff *skb, port = &addr->v4.sin_port; addr->v4.sin_family = AF_INET; - sh = (struct sctphdr *) skb->h.raw; + sh = sctp_hdr(skb); if (is_saddr) { *port = sh->source; - from = &skb->nh.iph->saddr; + from = &ip_hdr(skb)->saddr; } else { *port = sh->dest; - from = &skb->nh.iph->daddr; + from = &ip_hdr(skb)->daddr; } memcpy(&addr->v4.sin_addr.s_addr, from, sizeof(struct in_addr)); } @@ -273,7 +268,7 @@ static void sctp_v4_from_skb(union sctp_addr *addr, struct sk_buff *skb, 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_port = 0; addr->v4.sin_addr.s_addr = inet_sk(sk)->rcv_saddr; } @@ -309,7 +304,7 @@ static int sctp_v4_to_addr_param(const union sctp_addr *addr, param->v4.param_hdr.type = SCTP_PARAM_IPV4_ADDRESS; param->v4.param_hdr.length = htons(length); - param->v4.addr.s_addr = addr->v4.sin_addr.s_addr; + param->v4.addr.s_addr = addr->v4.sin_addr.s_addr; return length; } @@ -339,17 +334,17 @@ static int sctp_v4_cmp_addr(const union sctp_addr *addr1, } /* Initialize addr struct to INADDR_ANY. */ -static void sctp_v4_inaddr_any(union sctp_addr *addr, unsigned short port) +static void sctp_v4_inaddr_any(union sctp_addr *addr, __be16 port) { addr->v4.sin_family = AF_INET; - addr->v4.sin_addr.s_addr = INADDR_ANY; + addr->v4.sin_addr.s_addr = htonl(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; + return htonl(INADDR_ANY) == addr->v4.sin_addr.s_addr; } /* This function checks if the address is a valid address to be used for @@ -364,12 +359,12 @@ static int sctp_v4_addr_valid(union sctp_addr *addr, const struct sk_buff *skb) { /* Is this a non-unicast address or a unusable SCTP address? */ - if (IS_IPV4_UNUSABLE_ADDRESS(&addr->v4.sin_addr.s_addr)) + if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) return 0; - /* Is this a broadcast address? */ - if (skb && ((struct rtable *)skb->dst)->rt_flags & RTCF_BROADCAST) - return 0; + /* Is this a broadcast address? */ + if (skb && skb->rtable->rt_flags & RTCF_BROADCAST) + return 0; return 1; } @@ -377,10 +372,10 @@ static int sctp_v4_addr_valid(union sctp_addr *addr, /* 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); + int ret = inet_addr_type(&init_net, addr->v4.sin_addr.s_addr); - if (addr->v4.sin_addr.s_addr != INADDR_ANY && + if (addr->v4.sin_addr.s_addr != htonl(INADDR_ANY) && ret != RTN_LOCAL && !sp->inet.freebind && !sysctl_ip_nonlocal_bind) @@ -413,13 +408,15 @@ static sctp_scope_t sctp_v4_scope(union sctp_addr *addr) */ /* Check for unusable SCTP addresses. */ - if (IS_IPV4_UNUSABLE_ADDRESS(&addr->v4.sin_addr.s_addr)) { + if (IS_IPV4_UNUSABLE_ADDRESS(addr->v4.sin_addr.s_addr)) { retval = SCTP_SCOPE_UNUSABLE; - } else if (LOOPBACK(addr->v4.sin_addr.s_addr)) { + } else if (ipv4_is_loopback(addr->v4.sin_addr.s_addr)) { retval = SCTP_SCOPE_LOOPBACK; - } else if (IS_IPV4_LINK_ADDRESS(&addr->v4.sin_addr.s_addr)) { + } else if (ipv4_is_linklocal_169(addr->v4.sin_addr.s_addr)) { retval = SCTP_SCOPE_LINK; - } else if (IS_IPV4_PRIVATE_ADDRESS(&addr->v4.sin_addr.s_addr)) { + } else if (ipv4_is_private_10(addr->v4.sin_addr.s_addr) || + ipv4_is_private_172(addr->v4.sin_addr.s_addr) || + ipv4_is_private_192(addr->v4.sin_addr.s_addr)) { retval = SCTP_SCOPE_PRIVATE; } else { retval = SCTP_SCOPE_GLOBAL; @@ -439,9 +436,7 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, 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; @@ -456,10 +451,10 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, 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), + __func__, NIPQUAD(fl.fl4_dst), NIPQUAD(fl.fl4_src)); - if (!ip_route_output_key(&rt, &fl)) { + if (!ip_route_output_key(&init_net, &rt, &fl)) { dst = &rt->u.dst; } @@ -470,23 +465,20 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, 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); - if (!laddr->use_as_src) + rcu_read_lock(); + list_for_each_entry_rcu(laddr, &bp->address_list, list) { + if (!laddr->valid || (laddr->state != SCTP_ADDR_SRC)) continue; sctp_v4_dst_saddr(&dst_saddr, dst, htons(bp->port)); if (sctp_v4_cmp_addr(&dst_saddr, &laddr->a)) goto out_unlock; } - sctp_read_unlock(addr_lock); + rcu_read_unlock(); /* None of the bound addresses match the source address of the * dst. So release it. @@ -498,14 +490,14 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, /* 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 ((laddr->use_as_src) && + rcu_read_lock(); + list_for_each_entry_rcu(laddr, &bp->address_list, list) { + if (!laddr->valid) + continue; + if ((laddr->state == SCTP_ADDR_SRC) && (AF_INET == laddr->a.sa.sa_family)) { fl.fl4_src = laddr->a.v4.sin_addr.s_addr; - if (!ip_route_output_key(&rt, &fl)) { + if (!ip_route_output_key(&init_net, &rt, &fl)) { dst = &rt->u.dst; goto out_unlock; } @@ -513,11 +505,11 @@ static struct dst_entry *sctp_v4_get_dst(struct sctp_association *asoc, } out_unlock: - sctp_read_unlock(addr_lock); + rcu_read_unlock(); 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)); + NIPQUAD(rt->rt_dst), NIPQUAD(rt->rt_src)); else SCTP_DEBUG_PRINTK("NO ROUTE\n"); @@ -540,20 +532,20 @@ static void sctp_v4_get_saddr(struct sctp_association *asoc, if (rt) { saddr->v4.sin_family = AF_INET; saddr->v4.sin_port = htons(asoc->base.bind_addr.port); - saddr->v4.sin_addr.s_addr = rt->rt_src; + 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; + return skb->rtable->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); + return INET_ECN_is_ce(ip_hdr(skb)->tos); } /* Create and initialize a new sk for the socket returned by accept(). */ @@ -562,7 +554,8 @@ static struct sock *sctp_v4_create_accept_sk(struct sock *sk, { struct inet_sock *inet = inet_sk(sk); struct inet_sock *newinet; - struct sock *newsk = sk_alloc(PF_INET, GFP_KERNEL, sk->sk_prot, 1); + struct sock *newsk = sk_alloc(sk->sk_net, PF_INET, GFP_KERNEL, + sk->sk_prot); if (!newsk) goto out; @@ -592,7 +585,7 @@ static struct sock *sctp_v4_create_accept_sk(struct sock *sk, newinet->dport = htons(asoc->peer.port); newinet->daddr = asoc->peer.primary_addr.v4.sin_addr.s_addr; newinet->pmtudisc = inet->pmtudisc; - newinet->id = asoc->next_tsn ^ jiffies; + newinet->id = asoc->next_tsn ^ jiffies; newinet->uc_ttl = -1; newinet->mc_loop = 1; @@ -624,17 +617,51 @@ static void sctp_v4_seq_dump_addr(struct seq_file *seq, union sctp_addr *addr) } /* Event handler for inet address addition/deletion events. - * Basically, whenever there is an event, we re-build our local address list. + * The sctp_local_addr_list needs to be protocted by a spin lock since + * multiple notifiers (say IPv4 and IPv6) may be running at the same + * time and thus corrupt the list. + * The reader side is protected with RCU. */ -int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev, - void *ptr) +static int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev, + void *ptr) { - unsigned long flags; + struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; + struct sctp_sockaddr_entry *addr = NULL; + struct sctp_sockaddr_entry *temp; + int found = 0; + + if (ifa->ifa_dev->dev->nd_net != &init_net) + return NOTIFY_DONE; - 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); + switch (ev) { + case NETDEV_UP: + addr = kmalloc(sizeof(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; + addr->valid = 1; + spin_lock_bh(&sctp_local_addr_lock); + list_add_tail_rcu(&addr->list, &sctp_local_addr_list); + spin_unlock_bh(&sctp_local_addr_lock); + } + break; + case NETDEV_DOWN: + spin_lock_bh(&sctp_local_addr_lock); + list_for_each_entry_safe(addr, temp, + &sctp_local_addr_list, list) { + if (addr->a.v4.sin_addr.s_addr == ifa->ifa_local) { + found = 1; + addr->valid = 0; + list_del_rcu(&addr->list); + break; + } + } + spin_unlock_bh(&sctp_local_addr_lock); + if (found) + call_rcu(&addr->rcu, sctp_local_addr_free); + break; + } return NOTIFY_DONE; } @@ -736,15 +763,13 @@ static void sctp_inet_event_msgname(struct sctp_ulpevent *event, char *msgname, /* 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) { + struct sctphdr *sh = sctp_hdr(skb); + struct sockaddr_in *sin = (struct sockaddr_in *)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; + sin->sin_addr.s_addr = ip_hdr(skb)->saddr; } } @@ -763,8 +788,8 @@ static int sctp_inet_cmp_addr(const union sctp_addr *addr1, /* 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) + if (htonl(INADDR_ANY) == addr1->v4.sin_addr.s_addr || + htonl(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; @@ -792,7 +817,7 @@ static int sctp_inet_send_verify(struct sctp_sock *opt, union sctp_addr *addr) * chunks. Returns number of addresses supported. */ static int sctp_inet_supported_addrs(const struct sctp_sock *opt, - __u16 *types) + __be16 *types) { types[0] = SCTP_PARAM_IPV4_ADDRESS; return 1; @@ -804,15 +829,15 @@ static inline int sctp_v4_xmit(struct sk_buff *skb, { 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)); + __func__, skb, skb->len, + NIPQUAD(skb->rtable->rt_src), + NIPQUAD(skb->rtable->rt_dst)); SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS); - return ip_queue_xmit(skb, skb->sk, ipfragok); + return ip_queue_xmit(skb, ipfragok); } -static struct sctp_af sctp_ipv4_specific; +static struct sctp_af sctp_af_inet; static struct sctp_pf sctp_pf_inet = { .event_msgname = sctp_inet_event_msgname, @@ -824,7 +849,7 @@ static struct sctp_pf sctp_pf_inet = { .supported_addrs = sctp_inet_supported_addrs, .create_accept_sk = sctp_v4_create_accept_sk, .addr_v4map = sctp_v4_addr_v4map, - .af = &sctp_ipv4_specific, + .af = &sctp_af_inet }; /* Notifier for inetaddr addition/deletion events. */ @@ -886,7 +911,7 @@ static struct net_protocol sctp_protocol = { }; /* IPv4 address related functions. */ -static struct sctp_af sctp_ipv4_specific = { +static struct sctp_af sctp_af_inet = { .sa_family = AF_INET, .sctp_xmit = sctp_v4_xmit, .setsockopt = ip_setsockopt, @@ -976,39 +1001,27 @@ SCTP_STATIC __init int sctp_init(void) int i; int status = -EINVAL; unsigned long goal; + unsigned long limit; + int max_share; 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. */ + /* Allocate bind_bucket and chunk caches. */ status = -ENOBUFS; sctp_bucket_cachep = kmem_cache_create("sctp_bind_bucket", sizeof(struct sctp_bind_bucket), 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); - + NULL); if (!sctp_bucket_cachep) - goto err_bucket_cachep; + goto out; sctp_chunk_cachep = kmem_cache_create("sctp_chunk", sizeof(struct sctp_chunk), 0, SLAB_HWCACHE_ALIGN, - NULL, NULL); + NULL); if (!sctp_chunk_cachep) goto err_chunk_cachep; @@ -1049,7 +1062,7 @@ SCTP_STATIC __init int sctp_init(void) sctp_cookie_preserve_enable = 1; /* Max.Burst - 4 */ - sctp_max_burst = SCTP_MAX_BURST; + sctp_max_burst = SCTP_DEFAULT_MAX_BURST; /* Association.Max.Retrans - 10 attempts * Path.Max.Retrans - 5 attempts (per destination address) @@ -1080,6 +1093,31 @@ SCTP_STATIC __init int sctp_init(void) /* Initialize handle used for association ids. */ idr_init(&sctp_assocs_id); + /* Set the pressure threshold to be a fraction of global memory that + * is up to 1/2 at 256 MB, decreasing toward zero with the amount of + * memory, with a floor of 128 pages. + * Note this initalizes the data in sctpv6_prot too + * Unabashedly stolen from tcp_init + */ + limit = min(num_physpages, 1UL<<(28-PAGE_SHIFT)) >> (20-PAGE_SHIFT); + limit = (limit * (num_physpages >> (20-PAGE_SHIFT))) >> (PAGE_SHIFT-11); + limit = max(limit, 128UL); + sysctl_sctp_mem[0] = limit / 4 * 3; + sysctl_sctp_mem[1] = limit; + sysctl_sctp_mem[2] = sysctl_sctp_mem[0] * 2; + + /* Set per-socket limits to no more than 1/128 the pressure threshold*/ + limit = (sysctl_sctp_mem[1]) << (PAGE_SHIFT - 7); + max_share = min(4UL*1024*1024, limit); + + sysctl_sctp_rmem[0] = PAGE_SIZE; /* give each asoc 1 page min */ + sysctl_sctp_rmem[1] = (1500 *(sizeof(struct sk_buff) + 1)); + sysctl_sctp_rmem[2] = max(sysctl_sctp_rmem[1], max_share); + + sysctl_sctp_wmem[0] = SK_MEM_QUANTUM; + sysctl_sctp_wmem[1] = 16*1024; + sysctl_sctp_wmem[2] = max(64*1024, max_share); + /* Size and allocate the association hash table. * The methodology is similar to that of the tcp hash tables. */ @@ -1106,7 +1144,7 @@ SCTP_STATIC __init int sctp_init(void) } for (i = 0; i < sctp_assoc_hashsize; i++) { rwlock_init(&sctp_assoc_hashtable[i].lock); - sctp_assoc_hashtable[i].chain = NULL; + INIT_HLIST_HEAD(&sctp_assoc_hashtable[i].chain); } /* Allocate and initialize the endpoint hash table. */ @@ -1120,7 +1158,7 @@ SCTP_STATIC __init int sctp_init(void) } for (i = 0; i < sctp_ep_hashsize; i++) { rwlock_init(&sctp_ep_hashtable[i].lock); - sctp_ep_hashtable[i].chain = NULL; + INIT_HLIST_HEAD(&sctp_ep_hashtable[i].chain); } /* Allocate and initialize the SCTP port hash table. */ @@ -1139,26 +1177,35 @@ SCTP_STATIC __init int sctp_init(void) } for (i = 0; i < sctp_port_hashsize; i++) { spin_lock_init(&sctp_port_hashtable[i].lock); - sctp_port_hashtable[i].chain = NULL; + INIT_HLIST_HEAD(&sctp_port_hashtable[i].chain); } - 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; + sctp_addip_noauth = 0; /* Enable PR-SCTP by default. */ sctp_prsctp_enable = 1; + /* Disable AUTH by default. */ + sctp_auth_enable = 0; + sctp_sysctl_register(); INIT_LIST_HEAD(&sctp_address_families); - sctp_register_af(&sctp_ipv4_specific); + sctp_register_af(&sctp_af_inet); + + status = proto_register(&sctp_prot, 1); + if (status) + goto err_proto_register; + + /* Register SCTP(UDP and TCP style) with socket layer. */ + inet_register_protosw(&sctp_seqpacket_protosw); + inet_register_protosw(&sctp_stream_protosw); status = sctp_v6_init(); if (status) @@ -1174,21 +1221,40 @@ SCTP_STATIC __init int sctp_init(void) /* Initialize the local address list. */ INIT_LIST_HEAD(&sctp_local_addr_list); spin_lock_init(&sctp_local_addr_lock); + sctp_get_local_addr_list(); /* Register notifier for inet address additions/deletions. */ register_inetaddr_notifier(&sctp_inetaddr_notifier); - sctp_get_local_addr_list(); + /* Register SCTP with inet layer. */ + if (inet_add_protocol(&sctp_protocol, IPPROTO_SCTP) < 0) { + status = -EAGAIN; + goto err_add_protocol; + } + + /* Register SCTP with inet6 layer. */ + status = sctp_v6_add_protocol(); + if (status) + goto err_v6_add_protocol; - __unsafe(THIS_MODULE); status = 0; out: return status; +err_v6_add_protocol: + inet_del_protocol(&sctp_protocol, IPPROTO_SCTP); + unregister_inetaddr_notifier(&sctp_inetaddr_notifier); +err_add_protocol: + sctp_free_local_addr_list(); + sock_release(sctp_ctl_socket); err_ctl_sock_init: sctp_v6_exit(); err_v6_init: + inet_unregister_protosw(&sctp_stream_protosw); + inet_unregister_protosw(&sctp_seqpacket_protosw); + proto_unregister(&sctp_prot); +err_proto_register: sctp_sysctl_unregister(); - list_del(&sctp_ipv4_specific.list); + list_del(&sctp_af_inet.list); free_pages((unsigned long)sctp_port_hashtable, get_order(sctp_port_hashsize * sizeof(struct sctp_bind_hashbucket))); @@ -1200,19 +1266,13 @@ err_ehash_alloc: sizeof(struct sctp_hashbucket))); err_ahash_alloc: sctp_dbg_objcnt_exit(); -err_init_proc: sctp_proc_exit(); +err_init_proc: 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); -err_add_protocol: - proto_unregister(&sctp_prot); goto out; } @@ -1223,6 +1283,10 @@ SCTP_STATIC __exit void sctp_exit(void) * up all the remaining associations and all that memory. */ + /* Unregister with inet6/inet layers. */ + sctp_v6_del_protocol(); + inet_del_protocol(&sctp_protocol, IPPROTO_SCTP); + /* Unregister notifier for inet address additions/deletions. */ unregister_inetaddr_notifier(&sctp_inetaddr_notifier); @@ -1232,9 +1296,15 @@ SCTP_STATIC __exit void sctp_exit(void) /* Free the control endpoint. */ sock_release(sctp_ctl_socket); + /* Cleanup v6 initializations. */ sctp_v6_exit(); + + /* Unregister with socket layer. */ + inet_unregister_protosw(&sctp_stream_protosw); + inet_unregister_protosw(&sctp_seqpacket_protosw); + sctp_sysctl_unregister(); - list_del(&sctp_ipv4_specific.list); + list_del(&sctp_af_inet.list); free_pages((unsigned long)sctp_assoc_hashtable, get_order(sctp_assoc_hashsize * @@ -1244,16 +1314,13 @@ SCTP_STATIC __exit void sctp_exit(void) 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); + kmem_cache_destroy(sctp_chunk_cachep); + kmem_cache_destroy(sctp_bucket_cachep); + proto_unregister(&sctp_prot); } @@ -1264,6 +1331,7 @@ module_exit(sctp_exit); * __stringify doesn't likes enums, so use IPPROTO_SCTP value (132) directly. */ MODULE_ALIAS("net-pf-" __stringify(PF_INET) "-proto-132"); +MODULE_ALIAS("net-pf-" __stringify(PF_INET6) "-proto-132"); MODULE_AUTHOR("Linux Kernel SCTP developers "); MODULE_DESCRIPTION("Support for the SCTP protocol (RFC2960)"); MODULE_LICENSE("GPL");