From e5b13cb10de209f924fdf9478214bcf7e4008d6d Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 28 Feb 2008 20:51:43 -0800 Subject: [NETNS]: Process devinet ioctl in the correct namespace. Add namespace parameter to devinet_ioctl and locate device inside it for state changes. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/linux/inetdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index fc4e3db649e8..da05ab47ff2f 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -129,7 +129,7 @@ extern int unregister_inetaddr_notifier(struct notifier_block *nb); extern struct net_device *ip_dev_find(struct net *net, __be32 addr); extern int inet_addr_onlink(struct in_device *in_dev, __be32 a, __be32 b); -extern int devinet_ioctl(unsigned int cmd, void __user *); +extern int devinet_ioctl(struct net *net, unsigned int cmd, void __user *); extern void devinet_init(void); extern struct in_device *inetdev_by_index(struct net *, int); extern __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope); -- cgit v1.2.3 From 99cd07a537bf4c7c954f31611e30064407104410 Mon Sep 17 00:00:00 2001 From: Juha-Matti Tapio Date: Thu, 28 Feb 2008 20:55:46 -0800 Subject: [IPV6]: Fix source address selection for ORCHID addresses Skip the prefix length matching in source address selection for orchid -> non-orchid addresses. Overlay Routable Cryptographic Hash IDentifiers (RFC 4843, 2001:10::/28) are currenty not globally reachable. Without this check a host with an ORCHID address can end up preferring those over regular addresses when talking to other regular hosts in the 2001::/16 range thus breaking non-orchid connections. Signed-off-by: Juha-Matti Tapio Signed-off-by: David S. Miller --- include/net/ipv6.h | 10 ++++++++++ net/ipv6/addrconf.c | 5 +++++ 2 files changed, 15 insertions(+) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index c0c019f72ba9..8b05c65415cb 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -383,6 +383,16 @@ static inline int ipv6_addr_v4mapped(const struct in6_addr *a) a->s6_addr32[2] == htonl(0x0000ffff)); } +/* + * Check for a RFC 4843 ORCHID address + * (Overlay Routable Cryptographic Hash Identifiers) + */ +static inline int ipv6_addr_orchid(const struct in6_addr *a) +{ + return ((a->s6_addr32[0] & htonl(0xfffffff0)) + == htonl(0x20010010)); +} + /* * find the first different bit between two addresses * length of address must be a multiple of 32bits diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 101e0e70ba27..18e3a9825d81 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1125,6 +1125,11 @@ int ipv6_dev_get_saddr(struct net_device *daddr_dev, if (hiscore.rule < 7) hiscore.rule++; #endif + + /* Skip rule 8 for orchid -> non-orchid address pairs. */ + if (ipv6_addr_orchid(&ifa->addr) && !ipv6_addr_orchid(daddr)) + continue; + /* Rule 8: Use longest matching prefix */ if (hiscore.rule < 8) { hiscore.matchlen = ipv6_addr_diff(&ifa_result->addr, daddr); -- cgit v1.2.3 From 4c563f7669c10a12354b72b518c2287ffc6ebfb3 Mon Sep 17 00:00:00 2001 From: Timo Teras Date: Thu, 28 Feb 2008 21:31:08 -0800 Subject: [XFRM]: Speed up xfrm_policy and xfrm_state walking Change xfrm_policy and xfrm_state walking algorithm from O(n^2) to O(n). This is achieved adding the entries to one more list which is used solely for walking the entries. This also fixes some races where the dump can have duplicate or missing entries when the SPD/SADB is modified during an ongoing dump. Dumping SADB with 20000 entries using "time ip xfrm state" the sys time dropped from 1.012s to 0.080s. Signed-off-by: Timo Teras Signed-off-by: David S. Miller --- include/linux/xfrm.h | 3 +- include/net/xfrm.h | 52 +++++++++++++++++++++++++++++++-- net/key/af_key.c | 24 ++++++++++++--- net/xfrm/xfrm_policy.c | 79 +++++++++++++++++++++++++++++--------------------- net/xfrm/xfrm_state.c | 53 ++++++++++++++++++++++----------- net/xfrm/xfrm_user.c | 71 +++++++++++++++++++++++++++------------------ 6 files changed, 197 insertions(+), 85 deletions(-) (limited to 'include') diff --git a/include/linux/xfrm.h b/include/linux/xfrm.h index e31b8c84f2c9..0c82c80b277f 100644 --- a/include/linux/xfrm.h +++ b/include/linux/xfrm.h @@ -113,7 +113,8 @@ enum { XFRM_POLICY_TYPE_MAIN = 0, XFRM_POLICY_TYPE_SUB = 1, - XFRM_POLICY_TYPE_MAX = 2 + XFRM_POLICY_TYPE_MAX = 2, + XFRM_POLICY_TYPE_ANY = 255 }; enum diff --git a/include/net/xfrm.h b/include/net/xfrm.h index eea7785cc757..9b6205665190 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -121,6 +121,7 @@ extern struct mutex xfrm_cfg_mutex; struct xfrm_state { /* Note: bydst is re-used during gc */ + struct list_head all; struct hlist_node bydst; struct hlist_node bysrc; struct hlist_node byspi; @@ -424,6 +425,7 @@ struct xfrm_tmpl struct xfrm_policy { struct xfrm_policy *next; + struct list_head bytype; struct hlist_node bydst; struct hlist_node byidx; @@ -1160,6 +1162,18 @@ struct xfrm6_tunnel { int priority; }; +struct xfrm_state_walk { + struct xfrm_state *state; + int count; + u8 proto; +}; + +struct xfrm_policy_walk { + struct xfrm_policy *policy; + int count; + u8 type, cur_type; +}; + extern void xfrm_init(void); extern void xfrm4_init(void); extern void xfrm_state_init(void); @@ -1184,7 +1198,23 @@ static inline void xfrm6_fini(void) extern int xfrm_proc_init(void); #endif -extern int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), void *); +static inline void xfrm_state_walk_init(struct xfrm_state_walk *walk, u8 proto) +{ + walk->proto = proto; + walk->state = NULL; + walk->count = 0; +} + +static inline void xfrm_state_walk_done(struct xfrm_state_walk *walk) +{ + if (walk->state != NULL) { + xfrm_state_put(walk->state); + walk->state = NULL; + } +} + +extern int xfrm_state_walk(struct xfrm_state_walk *walk, + int (*func)(struct xfrm_state *, int, void*), void *); extern struct xfrm_state *xfrm_state_alloc(void); extern struct xfrm_state *xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr, struct flowi *fl, struct xfrm_tmpl *tmpl, @@ -1306,7 +1336,25 @@ static inline int xfrm4_udp_encap_rcv(struct sock *sk, struct sk_buff *skb) #endif struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp); -extern int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*), void *); + +static inline void xfrm_policy_walk_init(struct xfrm_policy_walk *walk, u8 type) +{ + walk->cur_type = XFRM_POLICY_TYPE_MAIN; + walk->type = type; + walk->policy = NULL; + walk->count = 0; +} + +static inline void xfrm_policy_walk_done(struct xfrm_policy_walk *walk) +{ + if (walk->policy != NULL) { + xfrm_pol_put(walk->policy); + walk->policy = NULL; + } +} + +extern int xfrm_policy_walk(struct xfrm_policy_walk *walk, + int (*func)(struct xfrm_policy *, int, int, void*), void *); int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl); struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir, struct xfrm_selector *sel, diff --git a/net/key/af_key.c b/net/key/af_key.c index 8b5f486ac80f..7cb6f1213360 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -1742,12 +1742,18 @@ static int pfkey_dump(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr { u8 proto; struct pfkey_dump_data data = { .skb = skb, .hdr = hdr, .sk = sk }; + struct xfrm_state_walk walk; + int rc; proto = pfkey_satype2proto(hdr->sadb_msg_satype); if (proto == 0) return -EINVAL; - return xfrm_state_walk(proto, dump_sa, &data); + xfrm_state_walk_init(&walk, proto); + rc = xfrm_state_walk(&walk, dump_sa, &data); + xfrm_state_walk_done(&walk); + + return rc; } static int pfkey_promisc(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) @@ -1780,7 +1786,9 @@ static int check_reqid(struct xfrm_policy *xp, int dir, int count, void *ptr) static u32 gen_reqid(void) { + struct xfrm_policy_walk walk; u32 start; + int rc; static u32 reqid = IPSEC_MANUAL_REQID_MAX; start = reqid; @@ -1788,8 +1796,10 @@ static u32 gen_reqid(void) ++reqid; if (reqid == 0) reqid = IPSEC_MANUAL_REQID_MAX+1; - if (xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, check_reqid, - (void*)&reqid) != -EEXIST) + xfrm_policy_walk_init(&walk, XFRM_POLICY_TYPE_MAIN); + rc = xfrm_policy_walk(&walk, check_reqid, (void*)&reqid); + xfrm_policy_walk_done(&walk); + if (rc != -EEXIST) return reqid; } while (reqid != start); return 0; @@ -2665,8 +2675,14 @@ static int dump_sp(struct xfrm_policy *xp, int dir, int count, void *ptr) static int pfkey_spddump(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr, void **ext_hdrs) { struct pfkey_dump_data data = { .skb = skb, .hdr = hdr, .sk = sk }; + struct xfrm_policy_walk walk; + int rc; + + xfrm_policy_walk_init(&walk, XFRM_POLICY_TYPE_MAIN); + rc = xfrm_policy_walk(&walk, dump_sp, &data); + xfrm_policy_walk_done(&walk); - return xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, dump_sp, &data); + return rc; } static int key_notify_policy_flush(struct km_event *c) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 9fc4c315f6cd..bae94a8031a2 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -46,6 +46,7 @@ EXPORT_SYMBOL(xfrm_cfg_mutex); static DEFINE_RWLOCK(xfrm_policy_lock); +static struct list_head xfrm_policy_bytype[XFRM_POLICY_TYPE_MAX]; unsigned int xfrm_policy_count[XFRM_POLICY_MAX*2]; EXPORT_SYMBOL(xfrm_policy_count); @@ -208,6 +209,7 @@ struct xfrm_policy *xfrm_policy_alloc(gfp_t gfp) policy = kzalloc(sizeof(struct xfrm_policy), gfp); if (policy) { + INIT_LIST_HEAD(&policy->bytype); INIT_HLIST_NODE(&policy->bydst); INIT_HLIST_NODE(&policy->byidx); rwlock_init(&policy->lock); @@ -230,6 +232,10 @@ void xfrm_policy_destroy(struct xfrm_policy *policy) if (del_timer(&policy->timer)) BUG(); + write_lock_bh(&xfrm_policy_lock); + list_del(&policy->bytype); + write_unlock_bh(&xfrm_policy_lock); + security_xfrm_policy_free(policy); kfree(policy); } @@ -584,6 +590,7 @@ int xfrm_policy_insert(int dir, struct xfrm_policy *policy, int excl) policy->curlft.use_time = 0; if (!mod_timer(&policy->timer, jiffies + HZ)) xfrm_pol_hold(policy); + list_add_tail(&policy->bytype, &xfrm_policy_bytype[policy->type]); write_unlock_bh(&xfrm_policy_lock); if (delpol) @@ -822,57 +829,60 @@ out: } EXPORT_SYMBOL(xfrm_policy_flush); -int xfrm_policy_walk(u8 type, int (*func)(struct xfrm_policy *, int, int, void*), +int xfrm_policy_walk(struct xfrm_policy_walk *walk, + int (*func)(struct xfrm_policy *, int, int, void*), void *data) { - struct xfrm_policy *pol, *last = NULL; - struct hlist_node *entry; - int dir, last_dir = 0, count, error; + struct xfrm_policy *old, *pol, *last = NULL; + int error = 0; + + if (walk->type >= XFRM_POLICY_TYPE_MAX && + walk->type != XFRM_POLICY_TYPE_ANY) + return -EINVAL; + if (walk->policy == NULL && walk->count != 0) + return 0; + + old = pol = walk->policy; + walk->policy = NULL; read_lock_bh(&xfrm_policy_lock); - count = 0; - for (dir = 0; dir < 2*XFRM_POLICY_MAX; dir++) { - struct hlist_head *table = xfrm_policy_bydst[dir].table; - int i; + for (; walk->cur_type < XFRM_POLICY_TYPE_MAX; walk->cur_type++) { + if (walk->type != walk->cur_type && + walk->type != XFRM_POLICY_TYPE_ANY) + continue; - hlist_for_each_entry(pol, entry, - &xfrm_policy_inexact[dir], bydst) { - if (pol->type != type) + if (pol == NULL) { + pol = list_first_entry(&xfrm_policy_bytype[walk->cur_type], + struct xfrm_policy, bytype); + } + list_for_each_entry_from(pol, &xfrm_policy_bytype[walk->cur_type], bytype) { + if (pol->dead) continue; if (last) { - error = func(last, last_dir % XFRM_POLICY_MAX, - count, data); - if (error) + error = func(last, xfrm_policy_id2dir(last->index), + walk->count, data); + if (error) { + xfrm_pol_hold(last); + walk->policy = last; goto out; - } - last = pol; - last_dir = dir; - count++; - } - for (i = xfrm_policy_bydst[dir].hmask; i >= 0; i--) { - hlist_for_each_entry(pol, entry, table + i, bydst) { - if (pol->type != type) - continue; - if (last) { - error = func(last, last_dir % XFRM_POLICY_MAX, - count, data); - if (error) - goto out; } - last = pol; - last_dir = dir; - count++; } + last = pol; + walk->count++; } + pol = NULL; } - if (count == 0) { + if (walk->count == 0) { error = -ENOENT; goto out; } - error = func(last, last_dir % XFRM_POLICY_MAX, 0, data); + if (last) + error = func(last, xfrm_policy_id2dir(last->index), 0, data); out: read_unlock_bh(&xfrm_policy_lock); + if (old != NULL) + xfrm_pol_put(old); return error; } EXPORT_SYMBOL(xfrm_policy_walk); @@ -2365,6 +2375,9 @@ static void __init xfrm_policy_init(void) panic("XFRM: failed to allocate bydst hash\n"); } + for (dir = 0; dir < XFRM_POLICY_TYPE_MAX; dir++) + INIT_LIST_HEAD(&xfrm_policy_bytype[dir]); + INIT_WORK(&xfrm_policy_gc_work, xfrm_policy_gc_task); register_netdevice_notifier(&xfrm_dev_notifier); } diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 7ba65e82941c..9880b792e6a5 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -50,6 +50,7 @@ static DEFINE_SPINLOCK(xfrm_state_lock); * Main use is finding SA after policy selected tunnel or transport mode. * Also, it can be used by ah/esp icmp error handler to find offending SA. */ +static LIST_HEAD(xfrm_state_all); static struct hlist_head *xfrm_state_bydst __read_mostly; static struct hlist_head *xfrm_state_bysrc __read_mostly; static struct hlist_head *xfrm_state_byspi __read_mostly; @@ -510,6 +511,7 @@ struct xfrm_state *xfrm_state_alloc(void) if (x) { atomic_set(&x->refcnt, 1); atomic_set(&x->tunnel_users, 0); + INIT_LIST_HEAD(&x->all); INIT_HLIST_NODE(&x->bydst); INIT_HLIST_NODE(&x->bysrc); INIT_HLIST_NODE(&x->byspi); @@ -533,6 +535,10 @@ void __xfrm_state_destroy(struct xfrm_state *x) { BUG_TRAP(x->km.state == XFRM_STATE_DEAD); + spin_lock_bh(&xfrm_state_lock); + list_del(&x->all); + spin_unlock_bh(&xfrm_state_lock); + spin_lock_bh(&xfrm_state_gc_lock); hlist_add_head(&x->bydst, &xfrm_state_gc_list); spin_unlock_bh(&xfrm_state_gc_lock); @@ -909,6 +915,8 @@ static void __xfrm_state_insert(struct xfrm_state *x) x->genid = ++xfrm_state_genid; + list_add_tail(&x->all, &xfrm_state_all); + h = xfrm_dst_hash(&x->id.daddr, &x->props.saddr, x->props.reqid, x->props.family); hlist_add_head(&x->bydst, xfrm_state_bydst+h); @@ -1518,36 +1526,47 @@ unlock: } EXPORT_SYMBOL(xfrm_alloc_spi); -int xfrm_state_walk(u8 proto, int (*func)(struct xfrm_state *, int, void*), +int xfrm_state_walk(struct xfrm_state_walk *walk, + int (*func)(struct xfrm_state *, int, void*), void *data) { - int i; - struct xfrm_state *x, *last = NULL; - struct hlist_node *entry; - int count = 0; + struct xfrm_state *old, *x, *last = NULL; int err = 0; + if (walk->state == NULL && walk->count != 0) + return 0; + + old = x = walk->state; + walk->state = NULL; spin_lock_bh(&xfrm_state_lock); - for (i = 0; i <= xfrm_state_hmask; i++) { - hlist_for_each_entry(x, entry, xfrm_state_bydst+i, bydst) { - if (!xfrm_id_proto_match(x->id.proto, proto)) - continue; - if (last) { - err = func(last, count, data); - if (err) - goto out; + if (x == NULL) + x = list_first_entry(&xfrm_state_all, struct xfrm_state, all); + list_for_each_entry_from(x, &xfrm_state_all, all) { + if (x->km.state == XFRM_STATE_DEAD) + continue; + if (!xfrm_id_proto_match(x->id.proto, walk->proto)) + continue; + if (last) { + err = func(last, walk->count, data); + if (err) { + xfrm_state_hold(last); + walk->state = last; + goto out; } - last = x; - count++; } + last = x; + walk->count++; } - if (count == 0) { + if (walk->count == 0) { err = -ENOENT; goto out; } - err = func(last, 0, data); + if (last) + err = func(last, 0, data); out: spin_unlock_bh(&xfrm_state_lock); + if (old != NULL) + xfrm_state_put(old); return err; } EXPORT_SYMBOL(xfrm_state_walk); diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index f971ca5645f8..f5fd5b3147cc 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -532,8 +532,6 @@ struct xfrm_dump_info { struct sk_buff *out_skb; u32 nlmsg_seq; u16 nlmsg_flags; - int start_idx; - int this_idx; }; static int copy_sec_ctx(struct xfrm_sec_ctx *s, struct sk_buff *skb) @@ -600,9 +598,6 @@ static int dump_one_state(struct xfrm_state *x, int count, void *ptr) struct nlmsghdr *nlh; int err; - if (sp->this_idx < sp->start_idx) - goto out; - nlh = nlmsg_put(skb, NETLINK_CB(in_skb).pid, sp->nlmsg_seq, XFRM_MSG_NEWSA, sizeof(*p), sp->nlmsg_flags); if (nlh == NULL) @@ -615,8 +610,6 @@ static int dump_one_state(struct xfrm_state *x, int count, void *ptr) goto nla_put_failure; nlmsg_end(skb, nlh); -out: - sp->this_idx++; return 0; nla_put_failure: @@ -624,18 +617,32 @@ nla_put_failure: return err; } +static int xfrm_dump_sa_done(struct netlink_callback *cb) +{ + struct xfrm_state_walk *walk = (struct xfrm_state_walk *) &cb->args[1]; + xfrm_state_walk_done(walk); + return 0; +} + static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) { + struct xfrm_state_walk *walk = (struct xfrm_state_walk *) &cb->args[1]; struct xfrm_dump_info info; + BUILD_BUG_ON(sizeof(struct xfrm_state_walk) > + sizeof(cb->args) - sizeof(cb->args[0])); + info.in_skb = cb->skb; info.out_skb = skb; info.nlmsg_seq = cb->nlh->nlmsg_seq; info.nlmsg_flags = NLM_F_MULTI; - info.this_idx = 0; - info.start_idx = cb->args[0]; - (void) xfrm_state_walk(0, dump_one_state, &info); - cb->args[0] = info.this_idx; + + if (!cb->args[0]) { + cb->args[0] = 1; + xfrm_state_walk_init(walk, 0); + } + + (void) xfrm_state_walk(walk, dump_one_state, &info); return skb->len; } @@ -654,7 +661,6 @@ static struct sk_buff *xfrm_state_netlink(struct sk_buff *in_skb, info.out_skb = skb; info.nlmsg_seq = seq; info.nlmsg_flags = 0; - info.this_idx = info.start_idx = 0; if (dump_one_state(x, 0, &info)) { kfree_skb(skb); @@ -1232,9 +1238,6 @@ static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr struct sk_buff *skb = sp->out_skb; struct nlmsghdr *nlh; - if (sp->this_idx < sp->start_idx) - goto out; - nlh = nlmsg_put(skb, NETLINK_CB(in_skb).pid, sp->nlmsg_seq, XFRM_MSG_NEWPOLICY, sizeof(*p), sp->nlmsg_flags); if (nlh == NULL) @@ -1250,8 +1253,6 @@ static int dump_one_policy(struct xfrm_policy *xp, int dir, int count, void *ptr goto nlmsg_failure; nlmsg_end(skb, nlh); -out: - sp->this_idx++; return 0; nlmsg_failure: @@ -1259,21 +1260,33 @@ nlmsg_failure: return -EMSGSIZE; } +static int xfrm_dump_policy_done(struct netlink_callback *cb) +{ + struct xfrm_policy_walk *walk = (struct xfrm_policy_walk *) &cb->args[1]; + + xfrm_policy_walk_done(walk); + return 0; +} + static int xfrm_dump_policy(struct sk_buff *skb, struct netlink_callback *cb) { + struct xfrm_policy_walk *walk = (struct xfrm_policy_walk *) &cb->args[1]; struct xfrm_dump_info info; + BUILD_BUG_ON(sizeof(struct xfrm_policy_walk) > + sizeof(cb->args) - sizeof(cb->args[0])); + info.in_skb = cb->skb; info.out_skb = skb; info.nlmsg_seq = cb->nlh->nlmsg_seq; info.nlmsg_flags = NLM_F_MULTI; - info.this_idx = 0; - info.start_idx = cb->args[0]; - (void) xfrm_policy_walk(XFRM_POLICY_TYPE_MAIN, dump_one_policy, &info); -#ifdef CONFIG_XFRM_SUB_POLICY - (void) xfrm_policy_walk(XFRM_POLICY_TYPE_SUB, dump_one_policy, &info); -#endif - cb->args[0] = info.this_idx; + + if (!cb->args[0]) { + cb->args[0] = 1; + xfrm_policy_walk_init(walk, XFRM_POLICY_TYPE_ANY); + } + + (void) xfrm_policy_walk(walk, dump_one_policy, &info); return skb->len; } @@ -1293,7 +1306,6 @@ static struct sk_buff *xfrm_policy_netlink(struct sk_buff *in_skb, info.out_skb = skb; info.nlmsg_seq = seq; info.nlmsg_flags = 0; - info.this_idx = info.start_idx = 0; if (dump_one_policy(xp, dir, 0, &info) < 0) { kfree_skb(skb); @@ -1891,15 +1903,18 @@ static const struct nla_policy xfrma_policy[XFRMA_MAX+1] = { static struct xfrm_link { int (*doit)(struct sk_buff *, struct nlmsghdr *, struct nlattr **); int (*dump)(struct sk_buff *, struct netlink_callback *); + int (*done)(struct netlink_callback *); } xfrm_dispatch[XFRM_NR_MSGTYPES] = { [XFRM_MSG_NEWSA - XFRM_MSG_BASE] = { .doit = xfrm_add_sa }, [XFRM_MSG_DELSA - XFRM_MSG_BASE] = { .doit = xfrm_del_sa }, [XFRM_MSG_GETSA - XFRM_MSG_BASE] = { .doit = xfrm_get_sa, - .dump = xfrm_dump_sa }, + .dump = xfrm_dump_sa, + .done = xfrm_dump_sa_done }, [XFRM_MSG_NEWPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_add_policy }, [XFRM_MSG_DELPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_get_policy }, [XFRM_MSG_GETPOLICY - XFRM_MSG_BASE] = { .doit = xfrm_get_policy, - .dump = xfrm_dump_policy }, + .dump = xfrm_dump_policy, + .done = xfrm_dump_policy_done }, [XFRM_MSG_ALLOCSPI - XFRM_MSG_BASE] = { .doit = xfrm_alloc_userspi }, [XFRM_MSG_ACQUIRE - XFRM_MSG_BASE] = { .doit = xfrm_add_acquire }, [XFRM_MSG_EXPIRE - XFRM_MSG_BASE] = { .doit = xfrm_add_sa_expire }, @@ -1938,7 +1953,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (link->dump == NULL) return -EINVAL; - return netlink_dump_start(xfrm_nl, skb, nlh, link->dump, NULL); + return netlink_dump_start(xfrm_nl, skb, nlh, link->dump, link->done); } err = nlmsg_parse(nlh, xfrm_msg_min[type], attrs, XFRMA_MAX, -- cgit v1.2.3 From 9b0f976f27f00a81cf47643d90854659626795b4 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 29 Feb 2008 11:13:15 -0800 Subject: [INET]: Remove struct net_proto_family* from _init calls. struct net_proto_family* is not used in icmp[v6]_init, ndisc_init, igmp_init and tcp_v4_init. Remove it. Signed-off-by: Denis V. Lunev Acked-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 2 +- include/net/icmp.h | 2 +- include/net/ndisc.h | 4 ++-- include/net/tcp.h | 2 +- net/ipv4/af_inet.c | 4 ++-- net/ipv4/icmp.c | 2 +- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/af_inet6.c | 6 +++--- net/ipv6/icmp.c | 2 +- net/ipv6/mcast.c | 2 +- net/ipv6/ndisc.c | 2 +- 11 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 7c5e9817e998..8f86d6b621c8 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -176,7 +176,7 @@ extern void icmpv6_send(struct sk_buff *skb, __u32 info, struct net_device *dev); -extern int icmpv6_init(struct net_proto_family *ops); +extern int icmpv6_init(void); extern int icmpv6_err_convert(int type, int code, int *err); extern void icmpv6_cleanup(void); diff --git a/include/net/icmp.h b/include/net/icmp.h index 9f7ef3c8baef..7bf714d9d7c7 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -48,7 +48,7 @@ struct sk_buff; extern void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info); extern int icmp_rcv(struct sk_buff *skb); extern int icmp_ioctl(struct sock *sk, int cmd, unsigned long arg); -extern void icmp_init(struct net_proto_family *ops); +extern void icmp_init(void); extern void icmp_out_count(unsigned char type); /* Move into dst.h ? */ diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 59b70624b056..5aedf324de66 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -77,7 +77,7 @@ struct nd_opt_hdr { } __attribute__((__packed__)); -extern int ndisc_init(struct net_proto_family *ops); +extern int ndisc_init(void); extern void ndisc_cleanup(void); @@ -107,7 +107,7 @@ extern int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *d /* * IGMP */ -extern int igmp6_init(struct net_proto_family *ops); +extern int igmp6_init(void); extern void igmp6_cleanup(void); diff --git a/include/net/tcp.h b/include/net/tcp.h index 7de4ea3a04d9..ae9774b478f5 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1373,7 +1373,7 @@ struct tcp_request_sock_ops { #endif }; -extern void tcp_v4_init(struct net_proto_family *ops); +extern void tcp_v4_init(void); extern void tcp_init(void); #endif /* _TCP_H */ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index c270080f370e..a7a99ac856dc 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1415,7 +1415,7 @@ static int __init inet_init(void) ip_init(); - tcp_v4_init(&inet_family_ops); + tcp_v4_init(); /* Setup TCP slab cache for open requests. */ tcp_init(); @@ -1430,7 +1430,7 @@ static int __init inet_init(void) * Set the ICMP layer up */ - icmp_init(&inet_family_ops); + icmp_init(); /* * Initialise the multicast router diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index a13c074dac09..98372db66c66 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -1139,7 +1139,7 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = { }, }; -void __init icmp_init(struct net_proto_family *ops) +void __init icmp_init(void) { struct inet_sock *inet; int i; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 00156bf421ca..256032a41069 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2443,7 +2443,7 @@ struct proto tcp_prot = { REF_PROTO_INUSE(tcp) }; -void __init tcp_v4_init(struct net_proto_family *ops) +void __init tcp_v4_init(void) { if (inet_csk_ctl_sock_create(&tcp_socket, PF_INET, SOCK_RAW, IPPROTO_TCP) < 0) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index f0aa97738746..9869f87243cf 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -808,13 +808,13 @@ static int __init inet6_init(void) if (err) goto sysctl_fail; #endif - err = icmpv6_init(&inet6_family_ops); + err = icmpv6_init(); if (err) goto icmp_fail; - err = ndisc_init(&inet6_family_ops); + err = ndisc_init(); if (err) goto ndisc_fail; - err = igmp6_init(&inet6_family_ops); + err = igmp6_init(); if (err) goto igmp_fail; err = ipv6_netfilter_init(); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 121d517bf91c..b9b13a77ba30 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -780,7 +780,7 @@ drop_no_count: */ static struct lock_class_key icmpv6_socket_sk_dst_lock_key; -int __init icmpv6_init(struct net_proto_family *ops) +int __init icmpv6_init(void) { struct sock *sk; int err, i, j; diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index ab228d1ea114..8ce894d90063 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -2597,7 +2597,7 @@ static const struct file_operations igmp6_mcf_seq_fops = { }; #endif -int __init igmp6_init(struct net_proto_family *ops) +int __init igmp6_init(void) { struct ipv6_pinfo *np; struct sock *sk; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 0d33a7d32125..1fc33c8c7232 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1733,7 +1733,7 @@ static int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name, #endif -int __init ndisc_init(struct net_proto_family *ops) +int __init ndisc_init(void) { struct ipv6_pinfo *np; struct sock *sk; -- cgit v1.2.3 From a5710d6582868a96cf0fe02eac11d34cfbc783c9 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 29 Feb 2008 11:14:50 -0800 Subject: [ICMP]: Add return code to icmp_init. icmp_init could fail and this is normal for namespace other than initial. So, the panic should be triggered only on init_net initialization path. Additionally create rollback path for icmp_init as a separate function. It will also be used later during namespace destruction. Signed-off-by: Denis V. Lunev Acked-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/icmp.h | 2 +- net/ipv4/af_inet.c | 3 ++- net/ipv4/icmp.c | 26 ++++++++++++++++++++++---- 3 files changed, 25 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/icmp.h b/include/net/icmp.h index 7bf714d9d7c7..faba64db8ff0 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -48,7 +48,7 @@ struct sk_buff; extern void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info); extern int icmp_rcv(struct sk_buff *skb); extern int icmp_ioctl(struct sock *sk, int cmd, unsigned long arg); -extern void icmp_init(void); +extern int icmp_init(void); extern void icmp_out_count(unsigned char type); /* Move into dst.h ? */ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a7a99ac856dc..4f539bd48718 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1430,7 +1430,8 @@ static int __init inet_init(void) * Set the ICMP layer up */ - icmp_init(); + if (icmp_init() < 0) + panic("Failed to create the ICMP control socket.\n"); /* * Initialise the multicast router diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 98372db66c66..b345b3d9bcaa 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -1139,19 +1139,32 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = { }, }; -void __init icmp_init(void) +static void __exit icmp_exit(void) { - struct inet_sock *inet; int i; for_each_possible_cpu(i) { - int err; + struct socket *sock; + + sock = per_cpu(__icmp_socket, i); + if (sock == NULL) + continue; + per_cpu(__icmp_socket, i) = NULL; + sock_release(sock); + } +} +int __init icmp_init(void) +{ + struct inet_sock *inet; + int i, err; + + for_each_possible_cpu(i) { err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP, &per_cpu(__icmp_socket, i)); if (err < 0) - panic("Failed to create the ICMP control socket.\n"); + goto fail; per_cpu(__icmp_socket, i)->sk->sk_allocation = GFP_ATOMIC; @@ -1171,6 +1184,11 @@ void __init icmp_init(void) */ per_cpu(__icmp_socket, i)->sk->sk_prot->unhash(per_cpu(__icmp_socket, i)->sk); } + return 0; + +fail: + icmp_exit(); + return err; } EXPORT_SYMBOL(icmp_err_convert); -- cgit v1.2.3 From edf0208702007ec1f6a36756fdd005f771a4cf17 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 29 Feb 2008 11:18:32 -0800 Subject: [NET]: Make netlink_kernel_release publically available as sk_release_kernel. This staff will be needed for non-netlink kernel sockets, which should also not pin a namespace like tcp_socket and icmp_socket. Signed-off-by: Denis V. Lunev Acked-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/sock.h | 13 +++++++++++++ net/core/sock.c | 18 ++++++++++++++++++ net/netlink/af_netlink.c | 18 ++---------------- 3 files changed, 33 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index fd9876087651..39112e75411c 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -850,6 +850,7 @@ extern struct sock *sk_alloc(struct net *net, int family, gfp_t priority, struct proto *prot); extern void sk_free(struct sock *sk); +extern void sk_release_kernel(struct sock *sk); extern struct sock *sk_clone(const struct sock *sk, const gfp_t priority); @@ -1333,6 +1334,18 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_e } #endif +/* + * Kernel sockets, f.e. rtnl or icmp_socket, are a part of a namespace. + * They should not hold a referrence to a namespace in order to allow + * to stop it. + * Sockets after sk_change_net should be released using sk_release_kernel + */ +static inline void sk_change_net(struct sock *sk, struct net *net) +{ + put_net(sk->sk_net); + sk->sk_net = net; +} + extern void sock_enable_timestamp(struct sock *sk); extern int sock_get_timestamp(struct sock *, struct timeval __user *); extern int sock_get_timestampns(struct sock *, struct timespec __user *); diff --git a/net/core/sock.c b/net/core/sock.c index 09cb3a74de7f..c71b645a78f0 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -987,6 +987,24 @@ void sk_free(struct sock *sk) sk_prot_free(sk->sk_prot_creator, sk); } +/* + * Last sock_put should drop referrence to sk->sk_net. It has already + * been dropped in sk_change_net. Taking referrence to stopping namespace + * is not an option. + * Take referrence to a socket to remove it from hash _alive_ and after that + * destroy it in the context of init_net. + */ +void sk_release_kernel(struct sock *sk) +{ + if (sk == NULL || sk->sk_socket == NULL) + return; + + sock_hold(sk); + sock_release(sk->sk_socket); + sk->sk_net = get_net(&init_net); + sock_put(sk); +} + struct sock *sk_clone(const struct sock *sk, const gfp_t priority) { struct sock *newsk; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index e6b636d53633..524e826bb976 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1372,8 +1372,7 @@ netlink_kernel_create(struct net *net, int unit, unsigned int groups, goto out_sock_release_nosk; sk = sock->sk; - put_net(sk->sk_net); - sk->sk_net = net; + sk_change_net(sk, net); if (groups < 32) groups = 32; @@ -1421,20 +1420,7 @@ EXPORT_SYMBOL(netlink_kernel_create); void netlink_kernel_release(struct sock *sk) { - /* - * Last sock_put should drop referrence to sk->sk_net. It has already - * been dropped in netlink_kernel_create. Taking referrence to stopping - * namespace is not an option. - * Take referrence to a socket to remove it from netlink lookup table - * _alive_ and after that destroy it in the context of init_net. - */ - if (sk == NULL || sk->sk_socket == NULL) - return; - - sock_hold(sk); - sock_release(sk->sk_socket); - sk->sk_net = get_net(&init_net); - sock_put(sk); + sk_release_kernel(sk); } EXPORT_SYMBOL(netlink_kernel_release); -- cgit v1.2.3 From 4a6ad7a141cbee2cf074e6cf8dc527b231b69ece Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 29 Feb 2008 11:19:58 -0800 Subject: [NETNS]: Make icmp_sk per namespace. All preparations are done. Now just add a hook to perform an initialization on namespace startup and replace icmp_sk macro with proper inline call. Signed-off-by: Denis V. Lunev Acked-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 2 ++ net/ipv4/icmp.c | 49 +++++++++++++++++++++++++++++++----------------- 2 files changed, 34 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index a9b4f6086294..504fde174525 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -26,6 +26,8 @@ struct netns_ipv4 { struct hlist_head *fib_table_hash; struct sock *fibnl; + struct sock **icmp_sk; + struct netns_frags frags; #ifdef CONFIG_NETFILTER struct xt_table *iptable_filter; diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 97d97ada4ce6..b51f4b0a3264 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -229,8 +229,10 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES+1]; * * On SMP we have one ICMP socket per-cpu. */ -static struct sock **__icmp_sk = NULL; -#define icmp_sk (__icmp_sk[smp_processor_id()]) +static struct sock *icmp_sk(struct net *net) +{ + return net->ipv4.icmp_sk[smp_processor_id()]; +} static inline int icmp_xmit_lock(struct sock *sk) { @@ -349,7 +351,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param, struct sock *sk; struct sk_buff *skb; - sk = icmp_sk; + sk = icmp_sk(rt->u.dst.dev->nd_net); if (ip_append_data(sk, icmp_glue_bits, icmp_param, icmp_param->data_len+icmp_param->head_len, icmp_param->head_len, @@ -378,10 +380,11 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param, static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) { - struct sock *sk = icmp_sk; - struct inet_sock *inet = inet_sk(sk); struct ipcm_cookie ipc; struct rtable *rt = (struct rtable *)skb->dst; + struct net *net = rt->u.dst.dev->nd_net; + struct sock *sk = icmp_sk(net); + struct inet_sock *inet = inet_sk(sk); __be32 daddr; if (ip_options_echo(&icmp_param->replyopts, skb)) @@ -407,7 +410,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) .tos = RT_TOS(ip_hdr(skb)->tos) } }, .proto = IPPROTO_ICMP }; security_skb_classify_flow(skb, &fl); - if (ip_route_output_key(rt->u.dst.dev->nd_net, &rt, &fl)) + if (ip_route_output_key(net, &rt, &fl)) goto out_unlock; } if (icmpv4_xrlim_allow(rt, icmp_param->data.icmph.type, @@ -440,11 +443,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) __be32 saddr; u8 tos; struct net *net; - struct sock *sk = icmp_sk; + struct sock *sk; if (!rt) goto out; net = rt->u.dst.dev->nd_net; + sk = icmp_sk(net); /* * Find the original header. It is expected to be valid, of course. @@ -1142,22 +1146,23 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES + 1] = { }, }; -static void __exit icmp_exit(void) +static void __net_exit icmp_sk_exit(struct net *net) { int i; for_each_possible_cpu(i) - sk_release_kernel(__icmp_sk[i]); - kfree(__icmp_sk); - __icmp_sk = NULL; + sk_release_kernel(net->ipv4.icmp_sk[i]); + kfree(net->ipv4.icmp_sk); + net->ipv4.icmp_sk = NULL; } -int __init icmp_init(void) +int __net_init icmp_sk_init(struct net *net) { int i, err; - __icmp_sk = kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL); - if (__icmp_sk == NULL) + net->ipv4.icmp_sk = + kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL); + if (net->ipv4.icmp_sk == NULL) return -ENOMEM; for_each_possible_cpu(i) { @@ -1169,8 +1174,8 @@ int __init icmp_init(void) if (err < 0) goto fail; - __icmp_sk[i] = sk = sock->sk; - sk_change_net(sk, &init_net); + net->ipv4.icmp_sk[i] = sk = sock->sk; + sk_change_net(sk, net); sk->sk_allocation = GFP_ATOMIC; @@ -1193,10 +1198,20 @@ int __init icmp_init(void) return 0; fail: - icmp_exit(); + icmp_sk_exit(net); return err; } +static struct pernet_operations __net_initdata icmp_sk_ops = { + .init = icmp_sk_init, + .exit = icmp_sk_exit, +}; + +int __init icmp_init(void) +{ + return register_pernet_device(&icmp_sk_ops); +} + EXPORT_SYMBOL(icmp_err_convert); EXPORT_SYMBOL(icmp_send); EXPORT_SYMBOL(icmp_statistics); -- cgit v1.2.3 From 98c6d1b261e74750e6c56ede26dc295201c4a8b2 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 29 Feb 2008 11:21:22 -0800 Subject: [NETNS]: Make icmpv6_sk per namespace. All preparations are done. Now just add a hook to perform an initialization on namespace startup and replace icmpv6_sk macro with proper inline call. Actual namespace the packet belongs too will be passed later along with the one for the routing. Signed-off-by: Denis V. Lunev Acked-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 1 + net/ipv6/icmp.c | 68 +++++++++++++++++++++++++++++++++--------------- 2 files changed, 48 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 1dd7de4e4195..82623d3a8e35 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -36,5 +36,6 @@ struct netns_ipv6 { struct xt_table *ip6table_mangle; struct xt_table *ip6table_raw; #endif + struct sock **icmp_sk; }; #endif diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 3eb594d54172..9f55a965c884 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -80,8 +80,10 @@ EXPORT_SYMBOL(icmpv6msg_statistics); * * On SMP we have one ICMP socket per-cpu. */ -static struct sock **__icmpv6_sk = NULL; -#define icmpv6_sk (__icmpv6_sk[smp_processor_id()]) +static inline struct sock *icmpv6_sk(struct net *net) +{ + return net->ipv6.icmp_sk[smp_processor_id()]; +} static int icmpv6_rcv(struct sk_buff *skb); @@ -389,7 +391,7 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, fl.fl_icmp_code = code; security_skb_classify_flow(skb, &fl); - sk = icmpv6_sk; + sk = icmpv6_sk(&init_net); np = inet6_sk(sk); if (icmpv6_xmit_lock(sk)) @@ -535,7 +537,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) fl.fl_icmp_type = ICMPV6_ECHO_REPLY; security_skb_classify_flow(skb, &fl); - sk = icmpv6_sk; + sk = icmpv6_sk(&init_net); np = inet6_sk(sk); if (icmpv6_xmit_lock(sk)) @@ -780,13 +782,14 @@ drop_no_count: */ static struct lock_class_key icmpv6_socket_sk_dst_lock_key; -int __init icmpv6_init(void) +static int __net_init icmpv6_sk_init(struct net *net) { struct sock *sk; int err, i, j; - __icmpv6_sk = kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL); - if (__icmpv6_sk == NULL) + net->ipv6.icmp_sk = + kzalloc(nr_cpu_ids * sizeof(struct sock *), GFP_KERNEL); + if (net->ipv6.icmp_sk == NULL) return -ENOMEM; for_each_possible_cpu(i) { @@ -801,8 +804,8 @@ int __init icmpv6_init(void) goto fail; } - __icmpv6_sk[i] = sk = sock->sk; - sk_change_net(sk, &init_net); + net->ipv6.icmp_sk[i] = sk = sock->sk; + sk_change_net(sk, net); sk->sk_allocation = GFP_ATOMIC; /* @@ -822,33 +825,56 @@ int __init icmpv6_init(void) sk->sk_prot->unhash(sk); } - - - if (inet6_add_protocol(&icmpv6_protocol, IPPROTO_ICMPV6) < 0) { - printk(KERN_ERR "Failed to register ICMP6 protocol\n"); - err = -EAGAIN; - goto fail; - } - return 0; fail: for (j = 0; j < i; j++) - sk_release_kernel(__icmpv6_sk[j]); - + sk_release_kernel(net->ipv6.icmp_sk[j]); + kfree(net->ipv6.icmp_sk); return err; } -void icmpv6_cleanup(void) +static void __net_exit icmpv6_sk_exit(struct net *net) { int i; for_each_possible_cpu(i) { - sk_release_kernel(__icmpv6_sk[i]); + sk_release_kernel(net->ipv6.icmp_sk[i]); } + kfree(net->ipv6.icmp_sk); +} + +static struct pernet_operations __net_initdata icmpv6_sk_ops = { + .init = icmpv6_sk_init, + .exit = icmpv6_sk_exit, +}; + +int __init icmpv6_init(void) +{ + int err; + + err = register_pernet_subsys(&icmpv6_sk_ops); + if (err < 0) + return err; + + err = -EAGAIN; + if (inet6_add_protocol(&icmpv6_protocol, IPPROTO_ICMPV6) < 0) + goto fail; + return 0; + +fail: + printk(KERN_ERR "Failed to register ICMP6 protocol\n"); + unregister_pernet_subsys(&icmpv6_sk_ops); + return err; +} + +void __exit icmpv6_cleanup(void) +{ + unregister_pernet_subsys(&icmpv6_sk_ops); inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6); } + static const struct icmp6_err { int err; int fatal; -- cgit v1.2.3 From 58fbbed4fbc0094fc808a568fe99a915f85402ee Mon Sep 17 00:00:00 2001 From: Neil Horman Date: Fri, 29 Feb 2008 11:40:56 -0800 Subject: [SCTP]: extend exported data in /proc/net/sctp/assoc RFC 3873 specifies several MIB objects that can't be obtained by the current data set exported by /proc/sys/net/sctp/assoc. This patch adds the missing pieces of data that allow us to compute all the objects in the sctpAssocTable object. Signed-off-by: Neil Horman Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 8 ++++++++ net/sctp/outqueue.c | 3 +++ net/sctp/proc.c | 20 ++++++++++++++------ net/sctp/sm_statefuns.c | 2 ++ 4 files changed, 27 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 9c827a749b6f..8966599ddb9f 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1661,6 +1661,9 @@ struct sctp_association { /* Transport to which SHUTDOWN chunk was last sent. */ struct sctp_transport *shutdown_last_sent_to; + /* How many times have we resent a SHUTDOWN */ + int shutdown_retries; + /* Transport to which INIT chunk was last sent. */ struct sctp_transport *init_last_sent_to; @@ -1695,6 +1698,11 @@ struct sctp_association { */ __u16 unack_data; + /* The total number of data chunks that we've had to retransmit + * as the result of a T3 timer expiration + */ + __u32 rtx_data_chunks; + /* This is the association's receive buffer space. This value is used * to set a_rwnd field in an INIT or a SACK chunk. */ diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 1bb3c5c35d2a..fd4deefab3cf 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -494,6 +494,8 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport, */ if (transport == transport->asoc->peer.retran_path) sctp_assoc_update_retran_path(transport->asoc); + transport->asoc->rtx_data_chunks += + transport->asoc->unack_data; break; case SCTP_RTXR_FAST_RTX: SCTP_INC_STATS(SCTP_MIB_FAST_RETRANSMITS); @@ -504,6 +506,7 @@ void sctp_retransmit(struct sctp_outq *q, struct sctp_transport *transport, break; case SCTP_RTXR_T1_RTX: SCTP_INC_STATS(SCTP_MIB_T1_RETRANSMITS); + transport->asoc->init_retries++; break; default: BUG(); diff --git a/net/sctp/proc.c b/net/sctp/proc.c index 9e214da82d9e..82bea300308d 100644 --- a/net/sctp/proc.c +++ b/net/sctp/proc.c @@ -281,8 +281,10 @@ static void * sctp_assocs_seq_start(struct seq_file *seq, loff_t *pos) *pos = 0; if (*pos == 0) - seq_printf(seq, " ASSOC SOCK STY SST ST HBKT ASSOC-ID TX_QUEUE RX_QUEUE UID INODE LPORT " - "RPORT LADDRS <-> RADDRS\n"); + seq_printf(seq, " ASSOC SOCK STY SST ST HBKT " + "ASSOC-ID TX_QUEUE RX_QUEUE UID INODE LPORT " + "RPORT LADDRS <-> RADDRS " + "HBINT INS OUTS MAXRT T1X T2X RTXC\n"); return (void *)pos; } @@ -321,15 +323,21 @@ static int sctp_assocs_seq_show(struct seq_file *seq, void *v) assoc = sctp_assoc(epb); sk = epb->sk; seq_printf(seq, - "%8p %8p %-3d %-3d %-2d %-4d %4d %8d %8d %7d %5lu %-5d %5d ", + "%8p %8p %-3d %-3d %-2d %-4d " + "%4d %8d %8d %7d %5lu %-5d %5d " + "%8d %5d %5d %4d %4d %4d %8d ", assoc, sk, sctp_sk(sk)->type, sk->sk_state, - assoc->state, hash, assoc->assoc_id, + assoc->state, hash, + assoc->assoc_id, assoc->sndbuf_used, atomic_read(&assoc->rmem_alloc), sock_i_uid(sk), sock_i_ino(sk), epb->bind_addr.port, - assoc->peer.port); - + assoc->peer.port, + assoc->hbinterval, assoc->c.sinit_max_instreams, + assoc->c.sinit_num_ostreams, assoc->max_retrans, + assoc->init_retries, assoc->shutdown_retries, + assoc->rtx_data_chunks); seq_printf(seq, " "); sctp_seq_dump_local_addrs(seq, epb); seq_printf(seq, "<-> "); diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index f2ed6473feef..ade0cbd3a52b 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -5312,6 +5312,8 @@ sctp_disposition_t sctp_sf_t2_timer_expire(const struct sctp_endpoint *ep, SCTP_DEBUG_PRINTK("Timer T2 expired.\n"); SCTP_INC_STATS(SCTP_MIB_T2_SHUTDOWN_EXPIREDS); + ((struct sctp_association *)asoc)->shutdown_retries++; + if (asoc->overall_error_count >= asoc->max_retrans) { sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, SCTP_ERROR(ETIMEDOUT)); -- cgit v1.2.3 From fd80eb942ad9761f241c9b287b3b9a342b20690d Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 29 Feb 2008 11:43:03 -0800 Subject: [INET]: Remove struct dst_entry *dst from request_sock_ops.rtx_syn_ack. It looks like dst parameter is used in this API due to historical reasons. Actually, it is really used in the direct call to tcp_v4_send_synack only. So, create a wrapper for tcp_v4_send_synack and remove dst from rtx_syn_ack. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/request_sock.h | 3 +-- net/dccp/ipv4.c | 11 +++++------ net/dccp/ipv6.c | 38 ++++++++++++++++++-------------------- net/dccp/minisocks.c | 2 +- net/ipv4/inet_connection_sock.c | 2 +- net/ipv4/tcp_ipv4.c | 14 +++++++++----- net/ipv4/tcp_minisocks.c | 2 +- net/ipv6/tcp_ipv6.c | 36 +++++++++++++++++------------------- 8 files changed, 53 insertions(+), 55 deletions(-) (limited to 'include') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index cff4608179c1..040780add355 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -31,8 +31,7 @@ struct request_sock_ops { int obj_size; struct kmem_cache *slab; int (*rtx_syn_ack)(struct sock *sk, - struct request_sock *req, - struct dst_entry *dst); + struct request_sock *req); void (*send_ack)(struct sk_buff *skb, struct request_sock *req); void (*send_reset)(struct sock *sk, diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 474075adbde4..514a40b7fc7f 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -471,15 +471,14 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk, return &rt->u.dst; } -static int dccp_v4_send_response(struct sock *sk, struct request_sock *req, - struct dst_entry *dst) +static int dccp_v4_send_response(struct sock *sk, struct request_sock *req) { int err = -1; struct sk_buff *skb; + struct dst_entry *dst; - /* First, grab a route. */ - - if (dst == NULL && (dst = inet_csk_route_req(sk, req)) == NULL) + dst = inet_csk_route_req(sk, req); + if (dst == NULL) goto out; skb = dccp_make_response(sk, dst, req); @@ -620,7 +619,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) dreq->dreq_iss = dccp_v4_init_sequence(skb); dreq->dreq_service = service; - if (dccp_v4_send_response(sk, req, NULL)) + if (dccp_v4_send_response(sk, req)) goto drop_and_free; inet_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 490333d47c7b..1a5e50b90677 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -224,8 +224,7 @@ out: } -static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, - struct dst_entry *dst) +static int dccp_v6_send_response(struct sock *sk, struct request_sock *req) { struct inet6_request_sock *ireq6 = inet6_rsk(req); struct ipv6_pinfo *np = inet6_sk(sk); @@ -234,6 +233,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, struct in6_addr *final_p = NULL, final; struct flowi fl; int err = -1; + struct dst_entry *dst; memset(&fl, 0, sizeof(fl)); fl.proto = IPPROTO_DCCP; @@ -245,28 +245,26 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req, fl.fl_ip_sport = inet_sk(sk)->sport; security_req_classify_flow(req, &fl); - if (dst == NULL) { - opt = np->opt; + opt = np->opt; - if (opt != NULL && opt->srcrt != NULL) { - const struct rt0_hdr *rt0 = (struct rt0_hdr *)opt->srcrt; + if (opt != NULL && opt->srcrt != NULL) { + const struct rt0_hdr *rt0 = (struct rt0_hdr *)opt->srcrt; - ipv6_addr_copy(&final, &fl.fl6_dst); - ipv6_addr_copy(&fl.fl6_dst, rt0->addr); - final_p = &final; - } + ipv6_addr_copy(&final, &fl.fl6_dst); + ipv6_addr_copy(&fl.fl6_dst, rt0->addr); + final_p = &final; + } - err = ip6_dst_lookup(sk, &dst, &fl); - if (err) - goto done; + err = ip6_dst_lookup(sk, &dst, &fl); + if (err) + goto done; - if (final_p) - ipv6_addr_copy(&fl.fl6_dst, final_p); + if (final_p) + ipv6_addr_copy(&fl.fl6_dst, final_p); - err = xfrm_lookup(&dst, &fl, sk, 0); - if (err < 0) - goto done; - } + err = xfrm_lookup(&dst, &fl, sk, 0); + if (err < 0) + goto done; skb = dccp_make_response(sk, dst, req); if (skb != NULL) { @@ -448,7 +446,7 @@ static int dccp_v6_conn_request(struct sock *sk, struct sk_buff *skb) dreq->dreq_iss = dccp_v6_init_sequence(skb); dreq->dreq_service = service; - if (dccp_v6_send_response(sk, req, NULL)) + if (dccp_v6_send_response(sk, req)) goto drop_and_free; inet6_csk_reqsk_queue_hash_add(sk, req, DCCP_TIMEOUT_INIT); diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index 027d1814e1ab..33ad48321b08 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -216,7 +216,7 @@ struct sock *dccp_check_req(struct sock *sk, struct sk_buff *skb, * counter (backoff, monitored by dccp_response_timer). */ req->retrans++; - req->rsk_ops->rtx_syn_ack(sk, req, NULL); + req->rsk_ops->rtx_syn_ack(sk, req); } /* Network Duplicate, discard packet */ return NULL; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index b189278c7bc1..c0e0fa03fce1 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -463,7 +463,7 @@ void inet_csk_reqsk_queue_prune(struct sock *parent, if (time_after_eq(now, req->expires)) { if ((req->retrans < thresh || (inet_rsk(req)->acked && req->retrans < max_retries)) - && !req->rsk_ops->rtx_syn_ack(parent, req, NULL)) { + && !req->rsk_ops->rtx_syn_ack(parent, req)) { unsigned long timeo; if (req->retrans++ == 0) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 256032a41069..3b26f9586dcb 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -723,8 +723,8 @@ static void tcp_v4_reqsk_send_ack(struct sk_buff *skb, * This still operates on a request_sock only, not on a big * socket. */ -static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, - struct dst_entry *dst) +static int __tcp_v4_send_synack(struct sock *sk, struct request_sock *req, + struct dst_entry *dst) { const struct inet_request_sock *ireq = inet_rsk(req); int err = -1; @@ -732,7 +732,7 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, /* First, grab a route. */ if (!dst && (dst = inet_csk_route_req(sk, req)) == NULL) - goto out; + return -1; skb = tcp_make_synack(sk, dst, req); @@ -751,11 +751,15 @@ static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req, err = net_xmit_eval(err); } -out: dst_release(dst); return err; } +static int tcp_v4_send_synack(struct sock *sk, struct request_sock *req) +{ + return __tcp_v4_send_synack(sk, req, NULL); +} + /* * IPv4 request_sock destructor. */ @@ -1380,7 +1384,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) } tcp_rsk(req)->snt_isn = isn; - if (tcp_v4_send_synack(sk, req, dst)) + if (__tcp_v4_send_synack(sk, req, dst)) goto drop_and_free; if (want_cookie) { diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index b61b76847ad9..0fdd1db641ac 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -536,7 +536,7 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, * Enforce "SYN-ACK" according to figure 8, figure 6 * of RFC793, fixed by RFC1122. */ - req->rsk_ops->rtx_syn_ack(sk, req, NULL); + req->rsk_ops->rtx_syn_ack(sk, req); return NULL; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 12750f2b05ab..1cbbb87dbad2 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -455,8 +455,7 @@ out: } -static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, - struct dst_entry *dst) +static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req) { struct inet6_request_sock *treq = inet6_rsk(req); struct ipv6_pinfo *np = inet6_sk(sk); @@ -464,6 +463,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, struct ipv6_txoptions *opt = NULL; struct in6_addr * final_p = NULL, final; struct flowi fl; + struct dst_entry *dst; int err = -1; memset(&fl, 0, sizeof(fl)); @@ -476,24 +476,22 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req, fl.fl_ip_sport = inet_sk(sk)->sport; security_req_classify_flow(req, &fl); - if (dst == NULL) { - opt = np->opt; - if (opt && opt->srcrt) { - struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; - ipv6_addr_copy(&final, &fl.fl6_dst); - ipv6_addr_copy(&fl.fl6_dst, rt0->addr); - final_p = &final; - } - - err = ip6_dst_lookup(sk, &dst, &fl); - if (err) - goto done; - if (final_p) - ipv6_addr_copy(&fl.fl6_dst, final_p); - if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) - goto done; + opt = np->opt; + if (opt && opt->srcrt) { + struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; + ipv6_addr_copy(&final, &fl.fl6_dst); + ipv6_addr_copy(&fl.fl6_dst, rt0->addr); + final_p = &final; } + err = ip6_dst_lookup(sk, &dst, &fl); + if (err) + goto done; + if (final_p) + ipv6_addr_copy(&fl.fl6_dst, final_p); + if ((err = xfrm_lookup(&dst, &fl, sk, 0)) < 0) + goto done; + skb = tcp_make_synack(sk, dst, req); if (skb) { struct tcphdr *th = tcp_hdr(skb); @@ -1294,7 +1292,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) security_inet_conn_request(sk, skb, req); - if (tcp_v6_send_synack(sk, req, NULL)) + if (tcp_v6_send_synack(sk, req)) goto drop; inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT); -- cgit v1.2.3 From a90bcbd651b453d8259243115696c11b864b30fe Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Fri, 29 Feb 2008 11:45:34 -0800 Subject: [SCTP]: Kill unused static inline sctp_sysctl_jiffies_ms MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After the patch: $ git-grep sctp_sysctl_jiffies_ms | wc -l 0 Signed-off-by: Ilpo Järvinen Acked-by: Vlad Yasevich Signed-off-by: David S. Miller --- include/net/sctp/sctp.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include') diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index 57df27f19588..a653eb3e1e70 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -368,11 +368,6 @@ void sctp_sysctl_unregister(void); #else static inline void sctp_sysctl_register(void) { return; } static inline void sctp_sysctl_unregister(void) { return; } -static inline 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) { - return -ENOSYS; -} #endif /* Size of Supported Address Parameter for 'x' address types. */ -- cgit v1.2.3 From 03a64c93b68e1eff299b9bbbb0d13105171cddc4 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Fri, 29 Feb 2008 11:46:17 -0800 Subject: [LLC]: Kill static inline llc_addrany MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After the patch: $ git-grep llc_addrany | wc -l 0 Signed-off-by: Ilpo Järvinen Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/llc_if.h | 5 ----- 1 file changed, 5 deletions(-) (limited to 'include') diff --git a/include/net/llc_if.h b/include/net/llc_if.h index c608812a8e89..b595a004d31b 100644 --- a/include/net/llc_if.h +++ b/include/net/llc_if.h @@ -74,11 +74,6 @@ static inline int llc_mac_null(const u8 *mac) return is_zero_ether_addr(mac); } -static inline int llc_addrany(const struct llc_addr *addr) -{ - return llc_mac_null(addr->mac) && !addr->lsap; -} - static inline int llc_mac_multicast(const u8 *mac) { return is_multicast_ether_addr(mac); -- cgit v1.2.3 From 0df3ef45a3d7b59cc53ce4e3611033c6e3b51a1b Mon Sep 17 00:00:00 2001 From: Ron Rindjunsky Date: Mon, 28 Jan 2008 14:07:15 +0200 Subject: mac80211: A-MPDU Tx add session's and low level driver's API This patch adds the API for 3 stages in A-MPDU Tx session flow: - request mac80211 to start/stop A-MPDU Tx session for specific TID. such a request should be issued by a load aware element, either mac80211 itself or external element. - requests by mac80211 to low-level driver to start/stop Tx aggregation. notice that low level driver responds now with Starting Sequence Number. - async feedback by low-level to mac80211 to inform that HW is ready for next A-MPDU Tx state. Changes in API to Rx A-MPDU were also made, reflected in iwlwifi changes as well. Signed-off-by: Ron Rindjunsky Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-4965.c | 4 +- drivers/net/wireless/iwlwifi/iwl-4965.h | 2 +- include/net/mac80211.h | 86 ++++++++++++++++++++++++++++++++- net/mac80211/ieee80211_sta.c | 4 +- 4 files changed, 89 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c index d727de8b96fe..cd4f5e383e60 100644 --- a/drivers/net/wireless/iwlwifi/iwl-4965.c +++ b/drivers/net/wireless/iwlwifi/iwl-4965.c @@ -4732,7 +4732,7 @@ static void iwl4965_sta_modify_del_ba_tid(struct iwl4965_priv *priv, int iwl4965_mac_ampdu_action(struct ieee80211_hw *hw, enum ieee80211_ampdu_mlme_action action, - const u8 *addr, u16 tid, u16 ssn) + const u8 *addr, u16 tid, u16 *ssn) { struct iwl4965_priv *priv = hw->priv; int sta_id; @@ -4744,7 +4744,7 @@ int iwl4965_mac_ampdu_action(struct ieee80211_hw *hw, switch (action) { case IEEE80211_AMPDU_RX_START: IWL_DEBUG_HT("start Rx\n"); - iwl4965_sta_modify_add_ba_tid(priv, sta_id, tid, ssn); + iwl4965_sta_modify_add_ba_tid(priv, sta_id, tid, *ssn); break; case IEEE80211_AMPDU_RX_STOP: IWL_DEBUG_HT("stop Rx\n"); diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.h b/drivers/net/wireless/iwlwifi/iwl-4965.h index 9cb82be0ff80..6edf869d4d54 100644 --- a/drivers/net/wireless/iwlwifi/iwl-4965.h +++ b/drivers/net/wireless/iwlwifi/iwl-4965.h @@ -777,7 +777,7 @@ extern void iwl4965_set_ht_add_station(struct iwl4965_priv *priv, u8 index, struct ieee80211_ht_info *sta_ht_inf); extern int iwl4965_mac_ampdu_action(struct ieee80211_hw *hw, enum ieee80211_ampdu_mlme_action action, - const u8 *addr, u16 tid, u16 ssn); + const u8 *addr, u16 tid, u16 *ssn); #ifdef CONFIG_IWL4965_HT_AGG extern int iwl4965_mac_ht_tx_agg_start(struct ieee80211_hw *hw, u8 *da, u16 tid, u16 *start_seq_num); diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 9083bafb63ca..3bbc33cc2b96 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -967,10 +967,14 @@ enum ieee80211_filter_flags { * &struct ieee80211_ops to indicate which action is needed. * @IEEE80211_AMPDU_RX_START: start Rx aggregation * @IEEE80211_AMPDU_RX_STOP: stop Rx aggregation + * @IEEE80211_AMPDU_TX_START: start Tx aggregation + * @IEEE80211_AMPDU_TX_STOP: stop Tx aggregation */ enum ieee80211_ampdu_mlme_action { IEEE80211_AMPDU_RX_START, IEEE80211_AMPDU_RX_STOP, + IEEE80211_AMPDU_TX_START, + IEEE80211_AMPDU_TX_STOP, }; /** @@ -1111,7 +1115,8 @@ enum ieee80211_ampdu_mlme_action { * The RA/TID combination determines the destination and TID we want * the ampdu action to be performed for. The action is defined through * ieee80211_ampdu_mlme_action. Starting sequence number (@ssn) - * is the first frame we expect to perform the action on. + * is the first frame we expect to perform the action on. notice + * that TX/RX_STOP can pass NULL for this parameter. */ struct ieee80211_ops { int (*tx)(struct ieee80211_hw *hw, struct sk_buff *skb, @@ -1162,7 +1167,7 @@ struct ieee80211_ops { int (*conf_ht)(struct ieee80211_hw *hw, struct ieee80211_conf *conf); int (*ampdu_action)(struct ieee80211_hw *hw, enum ieee80211_ampdu_mlme_action action, - const u8 *ra, u16 tid, u16 ssn); + const u8 *addr, u16 tid, u16 *ssn); }; /** @@ -1574,4 +1579,81 @@ void ieee80211_iterate_active_interfaces(struct ieee80211_hw *hw, struct ieee80211_vif *vif), void *data); +/** + * ieee80211_start_tx_ba_session - Start a tx Block Ack session. + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @ra: receiver address of the BA session recipient + * @tid: the TID to BA on. + * @return: success if addBA request was sent, failure otherwise + * + * Although mac80211/low level driver/user space application can estimate + * the need to start aggregation on a certain RA/TID, the session level + * will be managed by the mac80211. + */ +int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid); + +/** + * ieee80211_start_tx_ba_cb - low level driver ready to aggregate. + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @ra: receiver address of the BA session recipient. + * @tid: the TID to BA on. + * + * This function must be called by low level driver once it has + * finished with preparations for the BA session. + */ +void ieee80211_start_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u16 tid); + +/** + * ieee80211_start_tx_ba_cb_irqsafe - low level driver ready to aggregate. + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @ra: receiver address of the BA session recipient. + * @tid: the TID to BA on. + * + * This function must be called by low level driver once it has + * finished with preparations for the BA session. + * This version of the function is irq safe. + */ +void ieee80211_start_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra, + u16 tid); + +/** + * ieee80211_stop_tx_ba_session - Stop a Block Ack session. + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @ra: receiver address of the BA session recipient + * @tid: the TID to stop BA. + * @initiator: if indicates initiator DELBA frame will be sent. + * @return: error if no sta with matching da found, success otherwise + * + * Although mac80211/low level driver/user space application can estimate + * the need to stop aggregation on a certain RA/TID, the session level + * will be managed by the mac80211. + */ +int ieee80211_stop_tx_ba_session(struct ieee80211_hw *hw, + u8 *ra, u16 tid, + enum ieee80211_back_parties initiator); + +/** + * ieee80211_stop_tx_ba_cb - low level driver ready to stop aggregate. + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @ra: receiver address of the BA session recipient. + * @tid: the desired TID to BA on. + * + * This function must be called by low level driver once it has + * finished with preparations for the BA session tear down. + */ +void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid); + +/** + * ieee80211_stop_tx_ba_cb_irqsafe - low level driver ready to stop aggregate. + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @ra: receiver address of the BA session recipient. + * @tid: the desired TID to BA on. + * + * This function must be called by low level driver once it has + * finished with preparations for the BA session tear down. + * This version of the function is irq safe. + */ +void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra, + u16 tid); + #endif /* MAC80211_H */ diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index 79dadd2976cd..d5d5610db1bd 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -1128,7 +1128,7 @@ static void ieee80211_sta_process_addba_request(struct net_device *dev, if (local->ops->ampdu_action) ret = local->ops->ampdu_action(hw, IEEE80211_AMPDU_RX_START, - sta->addr, tid, start_seq_num); + sta->addr, tid, &start_seq_num); #ifdef CONFIG_MAC80211_HT_DEBUG printk(KERN_DEBUG "Rx A-MPDU on tid %d result %d", tid, ret); #endif /* CONFIG_MAC80211_HT_DEBUG */ @@ -1230,7 +1230,7 @@ void ieee80211_sta_stop_rx_ba_session(struct net_device *dev, u8 *ra, u16 tid, BUG_ON(!local->ops->ampdu_action); ret = local->ops->ampdu_action(hw, IEEE80211_AMPDU_RX_STOP, - ra, tid, EINVAL); + ra, tid, NULL); if (ret) printk(KERN_DEBUG "HW problem - can not stop rx " "aggergation for tid %d\n", tid); -- cgit v1.2.3 From 9e7234923789897858e1a475c579b5e2e6ad5b74 Mon Sep 17 00:00:00 2001 From: Ron Rindjunsky Date: Mon, 28 Jan 2008 14:07:18 +0200 Subject: mac80211: A-MPDU Tx adding qdisc support This patch allows qdisc support in A-MPDU Tx. a method to handle QoS <-> TID switches is present in this patch. Signed-off-by: Ron Rindjunsky Signed-off-by: John W. Linville --- include/net/mac80211.h | 8 ++- net/mac80211/ieee80211.c | 8 +-- net/mac80211/ieee80211_i.h | 5 +- net/mac80211/tx.c | 2 + net/mac80211/wme.c | 135 +++++++++++++++++++++++++++++++++++++++++---- net/mac80211/wme.h | 23 +++++++- 6 files changed, 161 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 3bbc33cc2b96..0ce2e94dc844 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -246,6 +246,7 @@ struct ieee80211_tx_queue_stats_data { * @IEEE80211_TX_QUEUE_AFTER_BEACON: transmit queue for frames to be * sent after a beacon * @IEEE80211_TX_QUEUE_BEACON: transmit queue for beacon frames + * @NUM_TX_DATA_QUEUES_AMPDU: adding more queues for A-MPDU */ enum ieee80211_tx_queue { IEEE80211_TX_QUEUE_DATA0, @@ -261,11 +262,12 @@ enum ieee80211_tx_queue { * this struct need to have fixed values. As soon as it is removed, we can * fix these entries. */ IEEE80211_TX_QUEUE_AFTER_BEACON = 6, - IEEE80211_TX_QUEUE_BEACON = 7 + IEEE80211_TX_QUEUE_BEACON = 7, + NUM_TX_DATA_QUEUES_AMPDU = 16 }; struct ieee80211_tx_queue_stats { - struct ieee80211_tx_queue_stats_data data[NUM_TX_DATA_QUEUES]; + struct ieee80211_tx_queue_stats_data data[NUM_TX_DATA_QUEUES_AMPDU]; }; struct ieee80211_low_level_stats { @@ -348,6 +350,8 @@ struct ieee80211_tx_control { #define IEEE80211_TXCTL_EAPOL_FRAME (1<<11) /* internal to mac80211 */ #define IEEE80211_TXCTL_SEND_AFTER_DTIM (1<<12) /* send this frame after DTIM * beacon */ +#define IEEE80211_TXCTL_AMPDU (1<<13) /* this frame should be sent + * as part of an A-MPDU */ u32 flags; /* tx control flags defined * above */ u8 key_idx; /* keyidx from hw->set_key(), undefined if diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index c323c9af9b8f..f8e734f0da1a 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -462,7 +462,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid) spin_lock_bh(&local->mdev->queue_lock); /* create a new queue for this aggregation */ - /* ret = ieee80211_ht_agg_queue_add(local, sta, tid); */ + ret = ieee80211_ht_agg_queue_add(local, sta, tid); /* case no queue is available to aggregation * don't switch to aggregation */ @@ -488,7 +488,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid) /* No need to requeue the packets in the agg queue, since we * held the tx lock: no packet could be enqueued to the newly * allocated queue */ - /* ieee80211_ht_agg_queue_remove(local, sta, tid, 0); */ + ieee80211_ht_agg_queue_remove(local, sta, tid, 0); #ifdef CONFIG_MAC80211_HT_DEBUG printk(KERN_DEBUG "BA request denied - HW or queue unavailable" " for tid %d\n", tid); @@ -499,7 +499,7 @@ int ieee80211_start_tx_ba_session(struct ieee80211_hw *hw, u8 *ra, u16 tid) } /* Will put all the packets in the new SW queue */ - /* ieee80211_requeue(local, ieee802_1d_to_ac[tid]); */ + ieee80211_requeue(local, ieee802_1d_to_ac[tid]); spin_unlock_bh(&local->mdev->queue_lock); /* We have most probably almost emptied the legacy queue */ @@ -675,7 +675,7 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid) * the content of the qdiscs */ spin_lock_bh(&local->mdev->queue_lock); /* remove the queue for this aggregation */ - /* ieee80211_ht_agg_queue_remove(local, sta, tid, 1); */ + ieee80211_ht_agg_queue_remove(local, sta, tid, 1); spin_unlock_bh(&local->mdev->queue_lock); /* we just requeued the all the frames that were in the removed diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 8a24c2c6ebc1..cfd0717c0033 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -165,6 +165,7 @@ struct ieee80211_txrx_data { #define IEEE80211_TXPD_DO_NOT_ENCRYPT BIT(1) #define IEEE80211_TXPD_REQUEUE BIT(2) #define IEEE80211_TXPD_EAPOL_FRAME BIT(3) +#define IEEE80211_TXPD_AMPDU BIT(4) /* Stored in sk_buff->cb */ struct ieee80211_tx_packet_data { int ifindex; @@ -452,8 +453,8 @@ struct ieee80211_local { struct sta_info *sta_hash[STA_HASH_SIZE]; struct timer_list sta_cleanup; - unsigned long state[NUM_TX_DATA_QUEUES]; - struct ieee80211_tx_stored_packet pending_packet[NUM_TX_DATA_QUEUES]; + unsigned long state[NUM_TX_DATA_QUEUES_AMPDU]; + struct ieee80211_tx_stored_packet pending_packet[NUM_TX_DATA_QUEUES_AMPDU]; struct tasklet_struct tx_pending_tasklet; /* number of interfaces with corresponding IFF_ flags */ diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 85d01646abf5..38e1b2bd8245 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1260,6 +1260,8 @@ int ieee80211_master_start_xmit(struct sk_buff *skb, control.flags |= IEEE80211_TXCTL_REQUEUE; if (pkt_data->flags & IEEE80211_TXPD_EAPOL_FRAME) control.flags |= IEEE80211_TXCTL_EAPOL_FRAME; + if (pkt_data->flags & IEEE80211_TXPD_AMPDU) + control.flags |= IEEE80211_TXCTL_AMPDU; control.queue = pkt_data->queue; ret = ieee80211_tx(odev, skb, &control); diff --git a/net/mac80211/wme.c b/net/mac80211/wme.c index 4e236599dd31..425aa8588ea0 100644 --- a/net/mac80211/wme.c +++ b/net/mac80211/wme.c @@ -19,10 +19,13 @@ #include "wme.h" /* maximum number of hardware queues we support. */ -#define TC_80211_MAX_QUEUES 8 +#define TC_80211_MAX_QUEUES 16 + +const int ieee802_1d_to_ac[8] = { 2, 3, 3, 2, 1, 1, 0, 0 }; struct ieee80211_sched_data { + unsigned long qdisc_pool; struct tcf_proto *filter_list; struct Qdisc *queues[TC_80211_MAX_QUEUES]; struct sk_buff_head requeued[TC_80211_MAX_QUEUES]; @@ -98,7 +101,6 @@ static inline int classify80211(struct sk_buff *skb, struct Qdisc *qd) struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; unsigned short fc = le16_to_cpu(hdr->frame_control); int qos; - const int ieee802_1d_to_ac[8] = { 2, 3, 3, 2, 1, 1, 0, 0 }; /* see if frame is data or non data frame */ if (unlikely((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA)) { @@ -146,9 +148,25 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd) unsigned short fc = le16_to_cpu(hdr->frame_control); struct Qdisc *qdisc; int err, queue; + struct sta_info *sta; + u8 tid; if (pkt_data->flags & IEEE80211_TXPD_REQUEUE) { - skb_queue_tail(&q->requeued[pkt_data->queue], skb); + queue = pkt_data->queue; + sta = sta_info_get(local, hdr->addr1); + tid = skb->priority & QOS_CONTROL_TAG1D_MASK; + if (sta) { + int ampdu_queue = sta->tid_to_tx_q[tid]; + if ((ampdu_queue < local->hw.queues) && + test_bit(ampdu_queue, &q->qdisc_pool)) { + queue = ampdu_queue; + pkt_data->flags |= IEEE80211_TXPD_AMPDU; + } else { + pkt_data->flags &= ~IEEE80211_TXPD_AMPDU; + } + sta_info_put(sta); + } + skb_queue_tail(&q->requeued[queue], skb); qd->q.qlen++; return 0; } @@ -159,14 +177,28 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd) */ if (WLAN_FC_IS_QOS_DATA(fc)) { u8 *p = skb->data + ieee80211_get_hdrlen(fc) - 2; - u8 qos_hdr = skb->priority & QOS_CONTROL_TAG1D_MASK; + u8 ack_policy = 0; + tid = skb->priority & QOS_CONTROL_TAG1D_MASK; if (local->wifi_wme_noack_test) - qos_hdr |= QOS_CONTROL_ACK_POLICY_NOACK << + ack_policy |= QOS_CONTROL_ACK_POLICY_NOACK << QOS_CONTROL_ACK_POLICY_SHIFT; /* qos header is 2 bytes, second reserved */ - *p = qos_hdr; + *p = ack_policy | tid; p++; *p = 0; + + sta = sta_info_get(local, hdr->addr1); + if (sta) { + int ampdu_queue = sta->tid_to_tx_q[tid]; + if ((ampdu_queue < local->hw.queues) && + test_bit(ampdu_queue, &q->qdisc_pool)) { + queue = ampdu_queue; + pkt_data->flags |= IEEE80211_TXPD_AMPDU; + } else { + pkt_data->flags &= ~IEEE80211_TXPD_AMPDU; + } + sta_info_put(sta); + } } if (unlikely(queue >= local->hw.queues)) { @@ -184,6 +216,7 @@ static int wme_qdiscop_enqueue(struct sk_buff *skb, struct Qdisc* qd) kfree_skb(skb); err = NET_XMIT_DROP; } else { + tid = skb->priority & QOS_CONTROL_TAG1D_MASK; pkt_data->queue = (unsigned int) queue; qdisc = q->queues[queue]; err = qdisc->enqueue(skb, qdisc); @@ -235,10 +268,11 @@ static struct sk_buff *wme_qdiscop_dequeue(struct Qdisc* qd) /* check all the h/w queues in numeric/priority order */ for (queue = 0; queue < hw->queues; queue++) { /* see if there is room in this hardware queue */ - if (test_bit(IEEE80211_LINK_STATE_XOFF, - &local->state[queue]) || - test_bit(IEEE80211_LINK_STATE_PENDING, - &local->state[queue])) + if ((test_bit(IEEE80211_LINK_STATE_XOFF, + &local->state[queue])) || + (test_bit(IEEE80211_LINK_STATE_PENDING, + &local->state[queue])) || + (!test_bit(queue, &q->qdisc_pool))) continue; /* there is space - try and get a frame */ @@ -360,6 +394,10 @@ static int wme_qdiscop_init(struct Qdisc *qd, struct nlattr *opt) } } + /* reserve all legacy QoS queues */ + for (i = 0; i < min(IEEE80211_TX_QUEUE_DATA4, queues); i++) + set_bit(i, &q->qdisc_pool); + return err; } @@ -605,3 +643,80 @@ void ieee80211_wme_unregister(void) { unregister_qdisc(&wme_qdisc_ops); } + +int ieee80211_ht_agg_queue_add(struct ieee80211_local *local, + struct sta_info *sta, u16 tid) +{ + int i; + struct ieee80211_sched_data *q = + qdisc_priv(local->mdev->qdisc_sleeping); + DECLARE_MAC_BUF(mac); + + /* prepare the filter and save it for the SW queue + * matching the recieved HW queue */ + + /* try to get a Qdisc from the pool */ + for (i = IEEE80211_TX_QUEUE_BEACON; i < local->hw.queues; i++) + if (!test_and_set_bit(i, &q->qdisc_pool)) { + ieee80211_stop_queue(local_to_hw(local), i); + sta->tid_to_tx_q[tid] = i; + + /* IF there are already pending packets + * on this tid first we need to drain them + * on the previous queue + * since HT is strict in order */ +#ifdef CONFIG_MAC80211_HT_DEBUG + if (net_ratelimit()) + printk(KERN_DEBUG "allocated aggregation queue" + " %d tid %d addr %s pool=0x%lX\n", + i, tid, print_mac(mac, sta->addr), + q->qdisc_pool); +#endif /* CONFIG_MAC80211_HT_DEBUG */ + return 0; + } + + return -EAGAIN; +} + +/** + * the caller needs to hold local->mdev->queue_lock + */ +void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local, + struct sta_info *sta, u16 tid, + u8 requeue) +{ + struct ieee80211_sched_data *q = + qdisc_priv(local->mdev->qdisc_sleeping); + int agg_queue = sta->tid_to_tx_q[tid]; + + /* return the qdisc to the pool */ + clear_bit(agg_queue, &q->qdisc_pool); + sta->tid_to_tx_q[tid] = local->hw.queues; + + if (requeue) + ieee80211_requeue(local, agg_queue); + else + q->queues[agg_queue]->ops->reset(q->queues[agg_queue]); +} + +void ieee80211_requeue(struct ieee80211_local *local, int queue) +{ + struct Qdisc *root_qd = local->mdev->qdisc_sleeping; + struct ieee80211_sched_data *q = qdisc_priv(root_qd); + struct Qdisc *qdisc = q->queues[queue]; + struct sk_buff *skb = NULL; + u32 len = qdisc->q.qlen; + + if (!qdisc || !qdisc->dequeue) + return; + + printk(KERN_DEBUG "requeue: qlen = %d\n", qdisc->q.qlen); + for (len = qdisc->q.qlen; len > 0; len--) { + skb = qdisc->dequeue(qdisc); + root_qd->q.qlen--; + /* packet will be classified again and */ + /* skb->packet_data->queue will be overridden if needed */ + if (skb) + wme_qdiscop_enqueue(skb, root_qd); + } +} diff --git a/net/mac80211/wme.h b/net/mac80211/wme.h index 76c713a6450c..fcc6b05508cc 100644 --- a/net/mac80211/wme.h +++ b/net/mac80211/wme.h @@ -24,6 +24,8 @@ #define QOS_CONTROL_TAG1D_MASK 0x07 +extern const int ieee802_1d_to_ac[8]; + static inline int WLAN_FC_IS_QOS_DATA(u16 fc) { return (fc & 0x8C) == 0x88; @@ -32,7 +34,12 @@ static inline int WLAN_FC_IS_QOS_DATA(u16 fc) #ifdef CONFIG_NET_SCHED void ieee80211_install_qdisc(struct net_device *dev); int ieee80211_qdisc_installed(struct net_device *dev); - +int ieee80211_ht_agg_queue_add(struct ieee80211_local *local, + struct sta_info *sta, u16 tid); +void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local, + struct sta_info *sta, u16 tid, + u8 requeue); +void ieee80211_requeue(struct ieee80211_local *local, int queue); int ieee80211_wme_register(void); void ieee80211_wme_unregister(void); #else @@ -43,7 +50,19 @@ static inline int ieee80211_qdisc_installed(struct net_device *dev) { return 0; } - +static inline int ieee80211_ht_agg_queue_add(struct ieee80211_local *local, + struct sta_info *sta, u16 tid) +{ + return -EAGAIN; +} +static inline void ieee80211_ht_agg_queue_remove(struct ieee80211_local *local, + struct sta_info *sta, u16 tid, + u8 requeue) +{ +} +static inline void ieee80211_requeue(struct ieee80211_local *local, int queue) +{ +} static inline int ieee80211_wme_register(void) { return 0; -- cgit v1.2.3 From 483fdcecc564ae6b011148a758517cf561f65678 Mon Sep 17 00:00:00 2001 From: Ron Rindjunsky Date: Mon, 28 Jan 2008 14:07:21 +0200 Subject: mac80211: A-MPDU Tx change tx_status to support Block Ack data This patch adds fields to ieee80211_tx_status in order to allow block ack information exchange between low-level driver,mac80211 and rate scaling module. Signed-off-by: Ron Rindjunsky Signed-off-by: John W. Linville --- include/net/mac80211.h | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 0ce2e94dc844..277488176a44 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -445,12 +445,14 @@ struct ieee80211_rx_status { * * @IEEE80211_TX_STATUS_TX_FILTERED: The frame was not transmitted * because the destination STA was in powersave mode. - * * @IEEE80211_TX_STATUS_ACK: Frame was acknowledged + * @IEEE80211_TX_STATUS_AMPDU: The frame was aggregated, so status + * is for the whole aggregation. */ enum ieee80211_tx_status_flags { IEEE80211_TX_STATUS_TX_FILTERED = 1<<0, IEEE80211_TX_STATUS_ACK = 1<<1, + IEEE80211_TX_STATUS_AMPDU = 1<<2, }; /** @@ -461,24 +463,25 @@ enum ieee80211_tx_status_flags { * * @control: a copy of the &struct ieee80211_tx_control passed to the driver * in the tx() callback. - * * @flags: transmit status flags, defined above - * - * @ack_signal: signal strength of the ACK frame - * + * @retry_count: number of retries * @excessive_retries: set to 1 if the frame was retried many times * but not acknowledged - * - * @retry_count: number of retries - * + * @ampdu_ack_len: number of aggregated frames. + * relevant only if IEEE80211_TX_STATUS_AMPDU was set. + * @ampdu_ack_map: block ack bit map for the aggregation. + * relevant only if IEEE80211_TX_STATUS_AMPDU was set. + * @ack_signal: signal strength of the ACK frame * @queue_length: ?? REMOVE * @queue_number: ?? REMOVE */ struct ieee80211_tx_status { struct ieee80211_tx_control control; u8 flags; - bool excessive_retries; u8 retry_count; + bool excessive_retries; + u8 ampdu_ack_len; + u64 ampdu_ack_map; int ack_signal; int queue_length; int queue_number; -- cgit v1.2.3 From 8318d78a44d49ac1edf2bdec7299de3617c4232e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 24 Jan 2008 19:38:38 +0100 Subject: cfg80211 API for channels/bitrates, mac80211 and driver conversion This patch creates new cfg80211 wiphy API for channel and bitrate registration and converts mac80211 and drivers to the new API. The old mac80211 API is completely ripped out. All drivers (except ath5k) are updated to the new API, in many cases I expect that optimisations can be done. Along with the regulatory code I've also ripped out the IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED flag, I believe it to be unnecessary if the hardware simply gives us whatever channels it wants to support and we then enable/disable them as required, which is pretty much required for travelling. Additionally, the patch adds proper "basic" rate handling for STA mode interface, AP mode interface will have to have new API added to allow userspace to set the basic rate set, currently it'll be empty... However, the basic rate handling will need to be moved to the BSS conf stuff. I do expect there to be bugs in this, especially wrt. transmit power handling where I'm basically clueless about how it should work. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/Kconfig | 1 + drivers/net/wireless/adm8211.c | 80 ++--- drivers/net/wireless/adm8211.h | 65 +--- drivers/net/wireless/b43/b43.h | 5 - drivers/net/wireless/b43/main.c | 119 +++---- drivers/net/wireless/b43/sysfs.c | 89 +----- drivers/net/wireless/b43/xmit.c | 81 ++--- drivers/net/wireless/b43legacy/b43legacy.h | 4 - drivers/net/wireless/b43legacy/main.c | 159 +++++----- drivers/net/wireless/b43legacy/xmit.c | 64 ++-- drivers/net/wireless/iwlwifi/iwl-3945-rs.c | 88 +++--- drivers/net/wireless/iwlwifi/iwl-3945.c | 35 +-- drivers/net/wireless/iwlwifi/iwl-3945.h | 13 +- drivers/net/wireless/iwlwifi/iwl-4965-rs.c | 82 ++--- drivers/net/wireless/iwlwifi/iwl-4965.c | 43 ++- drivers/net/wireless/iwlwifi/iwl-4965.h | 16 +- drivers/net/wireless/iwlwifi/iwl3945-base.c | 445 ++++++++------------------- drivers/net/wireless/iwlwifi/iwl4965-base.c | 438 +++++++------------------- drivers/net/wireless/p54.h | 4 - drivers/net/wireless/p54common.c | 74 +++-- drivers/net/wireless/p54common.h | 75 ----- drivers/net/wireless/rt2x00/rt2x00.h | 11 +- drivers/net/wireless/rt2x00/rt2x00config.c | 33 +- drivers/net/wireless/rt2x00/rt2x00dev.c | 149 ++++----- drivers/net/wireless/rt2x00/rt61pci.c | 23 +- drivers/net/wireless/rt2x00/rt73usb.c | 23 +- drivers/net/wireless/rtl8180.h | 2 +- drivers/net/wireless/rtl8180_dev.c | 93 ++++-- drivers/net/wireless/rtl8180_grf5101.c | 5 +- drivers/net/wireless/rtl8180_max2820.c | 5 +- drivers/net/wireless/rtl8180_rtl8225.c | 15 +- drivers/net/wireless/rtl8180_sa2400.c | 5 +- drivers/net/wireless/rtl8187.h | 2 +- drivers/net/wireless/rtl8187_dev.c | 82 +++-- drivers/net/wireless/rtl8187_rtl8225.c | 15 +- drivers/net/wireless/rtl818x.h | 70 ----- drivers/net/wireless/zd1211rw/zd_chip.c | 15 +- drivers/net/wireless/zd1211rw/zd_ieee80211.c | 11 +- drivers/net/wireless/zd1211rw/zd_mac.c | 170 +++++----- drivers/net/wireless/zd1211rw/zd_mac.h | 2 +- include/net/mac80211.h | 197 ++---------- include/net/wireless.h | 168 ++++++++++ net/mac80211/Makefile | 1 - net/mac80211/cfg.c | 11 +- net/mac80211/debugfs.c | 47 +-- net/mac80211/debugfs_sta.c | 18 -- net/mac80211/ieee80211.c | 112 +++---- net/mac80211/ieee80211_i.h | 63 +--- net/mac80211/ieee80211_iface.c | 2 + net/mac80211/ieee80211_ioctl.c | 127 ++++---- net/mac80211/ieee80211_rate.c | 15 +- net/mac80211/ieee80211_rate.h | 28 +- net/mac80211/ieee80211_sta.c | 384 ++++++++++++----------- net/mac80211/rc80211_pid_algo.c | 76 ++--- net/mac80211/rc80211_simple.c | 66 ++-- net/mac80211/regdomain.c | 152 --------- net/mac80211/rx.c | 82 ++--- net/mac80211/sta_info.c | 24 -- net/mac80211/sta_info.h | 10 +- net/mac80211/tx.c | 164 ++++++---- net/mac80211/util.c | 142 ++------- net/wireless/Makefile | 2 +- net/wireless/core.c | 41 +++ net/wireless/core.h | 3 + net/wireless/reg.c | 153 +++++++++ net/wireless/util.c | 98 ++++++ 66 files changed, 2116 insertions(+), 2781 deletions(-) delete mode 100644 net/mac80211/regdomain.c create mode 100644 net/wireless/reg.c create mode 100644 net/wireless/util.c (limited to 'include') diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig index 714a6ca30ad2..cd5fcc67e954 100644 --- a/drivers/net/wireless/Kconfig +++ b/drivers/net/wireless/Kconfig @@ -735,6 +735,7 @@ config P54_PCI config ATH5K tristate "Atheros 5xxx wireless cards support" depends on PCI && MAC80211 && WLAN_80211 && EXPERIMENTAL + depends on BROKEN ---help--- This module adds support for wireless adapters based on Atheros 5xxx chipset. diff --git a/drivers/net/wireless/adm8211.c b/drivers/net/wireless/adm8211.c index 79796186713e..7d4218206c47 100644 --- a/drivers/net/wireless/adm8211.c +++ b/drivers/net/wireless/adm8211.c @@ -48,6 +48,32 @@ static struct pci_device_id adm8211_pci_id_table[] __devinitdata = { { 0 } }; +static struct ieee80211_rate adm8211_rates[] = { + { .bitrate = 10, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 20, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 55, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 110, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 220, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, /* XX ?? */ +}; + +static const struct ieee80211_channel adm8211_channels[] = { + { .center_freq = 2412}, + { .center_freq = 2417}, + { .center_freq = 2422}, + { .center_freq = 2427}, + { .center_freq = 2432}, + { .center_freq = 2437}, + { .center_freq = 2442}, + { .center_freq = 2447}, + { .center_freq = 2452}, + { .center_freq = 2457}, + { .center_freq = 2462}, + { .center_freq = 2467}, + { .center_freq = 2472}, + { .center_freq = 2484}, +}; + + static void adm8211_eeprom_register_read(struct eeprom_93cx6 *eeprom) { struct adm8211_priv *priv = eeprom->data; @@ -155,17 +181,17 @@ static int adm8211_read_eeprom(struct ieee80211_hw *dev) printk(KERN_DEBUG "%s (adm8211): Channel range: %d - %d\n", pci_name(priv->pdev), (int)chan_range.min, (int)chan_range.max); - priv->modes[0].num_channels = chan_range.max - chan_range.min + 1; - priv->modes[0].channels = priv->channels; + BUILD_BUG_ON(sizeof(priv->channels) != sizeof(adm8211_channels)); - memcpy(priv->channels, adm8211_channels, sizeof(adm8211_channels)); + memcpy(priv->channels, adm8211_channels, sizeof(priv->channels)); + priv->band.channels = priv->channels; + priv->band.n_channels = ARRAY_SIZE(adm8211_channels); + priv->band.bitrates = adm8211_rates; + priv->band.n_bitrates = ARRAY_SIZE(adm8211_rates); for (i = 1; i <= ARRAY_SIZE(adm8211_channels); i++) - if (i >= chan_range.min && i <= chan_range.max) - priv->channels[i - 1].flag = - IEEE80211_CHAN_W_SCAN | - IEEE80211_CHAN_W_ACTIVE_SCAN | - IEEE80211_CHAN_W_IBSS; + if (i < chan_range.min || i > chan_range.max) + priv->channels[i - 1].flags |= IEEE80211_CHAN_DISABLED; switch (priv->eeprom->specific_bbptype) { case ADM8211_BBP_RFMD3000: @@ -347,7 +373,6 @@ static void adm8211_interrupt_rci(struct ieee80211_hw *dev) unsigned int pktlen; struct sk_buff *skb, *newskb; unsigned int limit = priv->rx_ring_size; - static const u8 rate_tbl[] = {10, 20, 55, 110, 220}; u8 rssi, rate; while (!(priv->rx_ring[entry].status & cpu_to_le32(RDES0_STATUS_OWN))) { @@ -425,12 +450,10 @@ static void adm8211_interrupt_rci(struct ieee80211_hw *dev) else rx_status.ssi = 100 - rssi; - if (rate <= 4) - rx_status.rate = rate_tbl[rate]; + rx_status.rate_idx = rate; - rx_status.channel = priv->channel; - rx_status.freq = adm8211_channels[priv->channel - 1].freq; - rx_status.phymode = MODE_IEEE80211B; + rx_status.freq = adm8211_channels[priv->channel - 1].center_freq; + rx_status.band = IEEE80211_BAND_2GHZ; ieee80211_rx_irqsafe(dev, skb, &rx_status); } @@ -1054,7 +1077,7 @@ static int adm8211_set_rate(struct ieee80211_hw *dev) if (priv->pdev->revision != ADM8211_REV_BA) { rate_buf[0] = ARRAY_SIZE(adm8211_rates); for (i = 0; i < ARRAY_SIZE(adm8211_rates); i++) - rate_buf[i + 1] = (adm8211_rates[i].rate / 5) | 0x80; + rate_buf[i + 1] = (adm8211_rates[i].bitrate / 5) | 0x80; } else { /* workaround for rev BA specific bug */ rate_buf[0] = 0x04; @@ -1303,9 +1326,10 @@ static int adm8211_set_ssid(struct ieee80211_hw *dev, u8 *ssid, size_t ssid_len) static int adm8211_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { struct adm8211_priv *priv = dev->priv; + int channel = ieee80211_frequency_to_channel(conf->channel->center_freq); - if (conf->channel != priv->channel) { - priv->channel = conf->channel; + if (channel != priv->channel) { + priv->channel = channel; adm8211_rf_set_channel(dev, priv->channel); } @@ -1680,10 +1704,10 @@ static int adm8211_tx(struct ieee80211_hw *dev, struct sk_buff *skb, if (control->tx_rate < 0) { short_preamble = 1; - plcp_signal = -control->tx_rate; + plcp_signal = -control->tx_rate->bitrate; } else { short_preamble = 0; - plcp_signal = control->tx_rate; + plcp_signal = control->tx_rate->bitrate; } hdr = (struct ieee80211_hdr *)skb->data; @@ -1880,18 +1904,11 @@ static int __devinit adm8211_probe(struct pci_dev *pdev, SET_IEEE80211_PERM_ADDR(dev, perm_addr); dev->extra_tx_headroom = sizeof(struct adm8211_tx_hdr); - dev->flags = IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED; - /* IEEE80211_HW_RX_INCLUDES_FCS in promisc mode */ + /* dev->flags = IEEE80211_HW_RX_INCLUDES_FCS in promisc mode */ dev->channel_change_time = 1000; dev->max_rssi = 100; /* FIXME: find better value */ - priv->modes[0].mode = MODE_IEEE80211B; - /* channel info filled in by adm8211_read_eeprom */ - memcpy(priv->rates, adm8211_rates, sizeof(adm8211_rates)); - priv->modes[0].num_rates = ARRAY_SIZE(adm8211_rates); - priv->modes[0].rates = priv->rates; - dev->queues = 1; /* ADM8211C supports more, maybe ADM8211B too */ priv->retry_limit = 3; @@ -1917,14 +1934,7 @@ static int __devinit adm8211_probe(struct pci_dev *pdev, goto err_free_desc; } - priv->channel = priv->modes[0].channels[0].chan; - - err = ieee80211_register_hwmode(dev, &priv->modes[0]); - if (err) { - printk(KERN_ERR "%s (adm8211): Can't register hwmode\n", - pci_name(pdev)); - goto err_free_desc; - } + priv->channel = 1; err = ieee80211_register_hw(dev); if (err) { diff --git a/drivers/net/wireless/adm8211.h b/drivers/net/wireless/adm8211.h index ef326fed42e4..8d7c564b3b04 100644 --- a/drivers/net/wireless/adm8211.h +++ b/drivers/net/wireless/adm8211.h @@ -534,61 +534,6 @@ struct adm8211_eeprom { u8 cis_data[0]; /* 0x80, 384 bytes */ } __attribute__ ((packed)); -static const struct ieee80211_rate adm8211_rates[] = { - { .rate = 10, - .val = 10, - .val2 = -10, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 20, - .val = 20, - .val2 = -20, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 55, - .val = 55, - .val2 = -55, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 110, - .val = 110, - .val2 = -110, - .flags = IEEE80211_RATE_CCK_2 } -}; - -struct ieee80211_chan_range { - u8 min; - u8 max; -}; - -static const struct ieee80211_channel adm8211_channels[] = { - { .chan = 1, - .freq = 2412}, - { .chan = 2, - .freq = 2417}, - { .chan = 3, - .freq = 2422}, - { .chan = 4, - .freq = 2427}, - { .chan = 5, - .freq = 2432}, - { .chan = 6, - .freq = 2437}, - { .chan = 7, - .freq = 2442}, - { .chan = 8, - .freq = 2447}, - { .chan = 9, - .freq = 2452}, - { .chan = 10, - .freq = 2457}, - { .chan = 11, - .freq = 2462}, - { .chan = 12, - .freq = 2467}, - { .chan = 13, - .freq = 2472}, - { .chan = 14, - .freq = 2484}, -}; - struct adm8211_priv { struct pci_dev *pdev; spinlock_t lock; @@ -603,9 +548,8 @@ struct adm8211_priv { unsigned int cur_tx, dirty_tx, cur_rx; struct ieee80211_low_level_stats stats; - struct ieee80211_hw_mode modes[1]; - struct ieee80211_channel channels[ARRAY_SIZE(adm8211_channels)]; - struct ieee80211_rate rates[ARRAY_SIZE(adm8211_rates)]; + struct ieee80211_supported_band band; + struct ieee80211_channel channels[14]; int mode; int channel; @@ -643,6 +587,11 @@ struct adm8211_priv { } transceiver_type; }; +struct ieee80211_chan_range { + u8 min; + u8 max; +}; + static const struct ieee80211_chan_range cranges[] = { {1, 11}, /* FCC */ {1, 11}, /* IC */ diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h index f13346ba9dd2..3e40323cd43f 100644 --- a/drivers/net/wireless/b43/b43.h +++ b/drivers/net/wireless/b43/b43.h @@ -468,10 +468,6 @@ struct b43_phy { u8 possible_phymodes; /* GMODE bit enabled? */ bool gmode; - /* Possible ieee80211 subsystem hwmodes for this PHY. - * Which mode is selected, depends on thr GMODE enabled bit */ -#define B43_MAX_PHYHWMODES 2 - struct ieee80211_hw_mode hwmodes[B43_MAX_PHYHWMODES]; /* Analog Type */ u8 analog; @@ -727,7 +723,6 @@ struct b43_wldev { bool bad_frames_preempt; /* Use "Bad Frames Preemption" (default off) */ bool dfq_valid; /* Directed frame queue valid (IBSS PS mode, ATIM) */ - bool short_preamble; /* TRUE, if short preamble is enabled. */ bool short_slot; /* TRUE, if short slot timing is enabled. */ bool radio_hw_enable; /* saved state of radio hardware enabled state */ bool suspend_in_progress; /* TRUE, if we are in a suspend/resume cycle */ diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c index 51dfce16178a..017a041d07d0 100644 --- a/drivers/net/wireless/b43/main.c +++ b/drivers/net/wireless/b43/main.c @@ -96,25 +96,29 @@ MODULE_DEVICE_TABLE(ssb, b43_ssb_tbl); * data in there. This data is the same for all devices, so we don't * get concurrency issues */ #define RATETAB_ENT(_rateid, _flags) \ - { \ - .rate = B43_RATE_TO_BASE100KBPS(_rateid), \ - .val = (_rateid), \ - .val2 = (_rateid), \ - .flags = (_flags), \ + { \ + .bitrate = B43_RATE_TO_BASE100KBPS(_rateid), \ + .hw_value = (_rateid), \ + .flags = (_flags), \ } + +/* + * NOTE: When changing this, sync with xmit.c's + * b43_plcp_get_bitrate_idx_* functions! + */ static struct ieee80211_rate __b43_ratetable[] = { - RATETAB_ENT(B43_CCK_RATE_1MB, IEEE80211_RATE_CCK), - RATETAB_ENT(B43_CCK_RATE_2MB, IEEE80211_RATE_CCK_2), - RATETAB_ENT(B43_CCK_RATE_5MB, IEEE80211_RATE_CCK_2), - RATETAB_ENT(B43_CCK_RATE_11MB, IEEE80211_RATE_CCK_2), - RATETAB_ENT(B43_OFDM_RATE_6MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_9MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_12MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_18MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_24MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_36MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_48MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43_OFDM_RATE_54MB, IEEE80211_RATE_OFDM), + RATETAB_ENT(B43_CCK_RATE_1MB, 0), + RATETAB_ENT(B43_CCK_RATE_2MB, IEEE80211_RATE_SHORT_PREAMBLE), + RATETAB_ENT(B43_CCK_RATE_5MB, IEEE80211_RATE_SHORT_PREAMBLE), + RATETAB_ENT(B43_CCK_RATE_11MB, IEEE80211_RATE_SHORT_PREAMBLE), + RATETAB_ENT(B43_OFDM_RATE_6MB, 0), + RATETAB_ENT(B43_OFDM_RATE_9MB, 0), + RATETAB_ENT(B43_OFDM_RATE_12MB, 0), + RATETAB_ENT(B43_OFDM_RATE_18MB, 0), + RATETAB_ENT(B43_OFDM_RATE_24MB, 0), + RATETAB_ENT(B43_OFDM_RATE_36MB, 0), + RATETAB_ENT(B43_OFDM_RATE_48MB, 0), + RATETAB_ENT(B43_OFDM_RATE_54MB, 0), }; #define b43_a_ratetable (__b43_ratetable + 4) @@ -126,14 +130,8 @@ static struct ieee80211_rate __b43_ratetable[] = { #define CHANTAB_ENT(_chanid, _freq) \ { \ - .chan = (_chanid), \ - .freq = (_freq), \ - .val = (_chanid), \ - .flag = IEEE80211_CHAN_W_SCAN | \ - IEEE80211_CHAN_W_ACTIVE_SCAN | \ - IEEE80211_CHAN_W_IBSS, \ - .power_level = 0xFF, \ - .antenna_max = 0xFF, \ + .center_freq = (_freq), \ + .hw_value = (_chanid), \ } static struct ieee80211_channel b43_2ghz_chantable[] = { CHANTAB_ENT(1, 2412), @@ -151,9 +149,8 @@ static struct ieee80211_channel b43_2ghz_chantable[] = { CHANTAB_ENT(13, 2472), CHANTAB_ENT(14, 2484), }; -#define b43_2ghz_chantable_size ARRAY_SIZE(b43_2ghz_chantable) -#if 0 +#ifdef NOTYET static struct ieee80211_channel b43_5ghz_chantable[] = { CHANTAB_ENT(36, 5180), CHANTAB_ENT(40, 5200), @@ -169,9 +166,22 @@ static struct ieee80211_channel b43_5ghz_chantable[] = { CHANTAB_ENT(161, 5805), CHANTAB_ENT(165, 5825), }; -#define b43_5ghz_chantable_size ARRAY_SIZE(b43_5ghz_chantable) + +static struct ieee80211_supported_band b43_band_5GHz = { + .channels = b43_5ghz_chantable, + .n_channels = ARRAY_SIZE(b43_5ghz_chantable), + .bitrates = b43_a_ratetable, + .n_bitrates = b43_a_ratetable_size, +}; #endif +static struct ieee80211_supported_band b43_band_2GHz = { + .channels = b43_2ghz_chantable, + .n_channels = ARRAY_SIZE(b43_2ghz_chantable), + .bitrates = b43_g_ratetable, + .n_bitrates = b43_g_ratetable_size, +}; + static void b43_wireless_core_exit(struct b43_wldev *dev); static int b43_wireless_core_init(struct b43_wldev *dev); static void b43_wireless_core_stop(struct b43_wldev *dev); @@ -1222,17 +1232,18 @@ static void b43_write_beacon_template(struct b43_wldev *dev, } static void b43_write_probe_resp_plcp(struct b43_wldev *dev, - u16 shm_offset, u16 size, u8 rate) + u16 shm_offset, u16 size, + struct ieee80211_rate *rate) { struct b43_plcp_hdr4 plcp; u32 tmp; __le16 dur; plcp.data = 0; - b43_generate_plcp_hdr(&plcp, size + FCS_LEN, rate); + b43_generate_plcp_hdr(&plcp, size + FCS_LEN, rate->hw_value); dur = ieee80211_generic_frame_duration(dev->wl->hw, dev->wl->vif, size, - B43_RATE_TO_BASE100KBPS(rate)); + rate); /* Write PLCP in two parts and timing for packet transfer */ tmp = le32_to_cpu(plcp.data); b43_shm_write16(dev, B43_SHM_SHARED, shm_offset, tmp & 0xFFFF); @@ -1247,7 +1258,8 @@ static void b43_write_probe_resp_plcp(struct b43_wldev *dev, * 3) Stripping TIM */ static const u8 * b43_generate_probe_resp(struct b43_wldev *dev, - u16 *dest_size, u8 rate) + u16 *dest_size, + struct ieee80211_rate *rate) { const u8 *src_data; u8 *dest_data; @@ -1292,7 +1304,7 @@ static const u8 * b43_generate_probe_resp(struct b43_wldev *dev, IEEE80211_STYPE_PROBE_RESP); dur = ieee80211_generic_frame_duration(dev->wl->hw, dev->wl->vif, *dest_size, - B43_RATE_TO_BASE100KBPS(rate)); + rate); hdr->duration_id = dur; return dest_data; @@ -1300,7 +1312,8 @@ static const u8 * b43_generate_probe_resp(struct b43_wldev *dev, static void b43_write_probe_resp_template(struct b43_wldev *dev, u16 ram_offset, - u16 shm_size_offset, u8 rate) + u16 shm_size_offset, + struct ieee80211_rate *rate) { const u8 *probe_resp_data; u16 size; @@ -1313,14 +1326,15 @@ static void b43_write_probe_resp_template(struct b43_wldev *dev, /* Looks like PLCP headers plus packet timings are stored for * all possible basic rates */ - b43_write_probe_resp_plcp(dev, 0x31A, size, B43_CCK_RATE_1MB); - b43_write_probe_resp_plcp(dev, 0x32C, size, B43_CCK_RATE_2MB); - b43_write_probe_resp_plcp(dev, 0x33E, size, B43_CCK_RATE_5MB); - b43_write_probe_resp_plcp(dev, 0x350, size, B43_CCK_RATE_11MB); + b43_write_probe_resp_plcp(dev, 0x31A, size, &b43_b_ratetable[0]); + b43_write_probe_resp_plcp(dev, 0x32C, size, &b43_b_ratetable[1]); + b43_write_probe_resp_plcp(dev, 0x33E, size, &b43_b_ratetable[2]); + b43_write_probe_resp_plcp(dev, 0x350, size, &b43_b_ratetable[3]); size = min((size_t) size, 0x200 - sizeof(struct b43_plcp_hdr6)); b43_write_template_common(dev, probe_resp_data, - size, ram_offset, shm_size_offset, rate); + size, ram_offset, shm_size_offset, + rate->hw_value); kfree(probe_resp_data); } @@ -1388,7 +1402,7 @@ static void handle_irq_beacon(struct b43_wldev *dev) b43_write_beacon_template(dev, 0x68, 0x18, B43_CCK_RATE_1MB); b43_write_probe_resp_template(dev, 0x268, 0x4A, - B43_CCK_RATE_11MB); + &__b43_ratetable[3]); wl->beacon0_uploaded = 1; } cmd |= B43_MACCMD_BEACON0_VALID; @@ -2830,14 +2844,11 @@ static int b43_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf) mutex_lock(&wl->mutex); /* Switch the PHY mode (if necessary). */ - switch (conf->phymode) { - case MODE_IEEE80211A: + switch (conf->channel->band) { + case IEEE80211_BAND_5GHZ: new_phymode = B43_PHYMODE_A; break; - case MODE_IEEE80211B: - new_phymode = B43_PHYMODE_B; - break; - case MODE_IEEE80211G: + case IEEE80211_BAND_2GHZ: new_phymode = B43_PHYMODE_G; break; default: @@ -2863,8 +2874,8 @@ static int b43_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf) /* Switch to the requested channel. * The firmware takes care of races with the TX handler. */ - if (conf->channel_val != phy->channel) - b43_radio_selectchannel(dev, conf->channel_val, 0); + if (conf->channel->hw_value != phy->channel) + b43_radio_selectchannel(dev, conf->channel->hw_value, 0); /* Enable/Disable ShortSlot timing. */ if ((!!(conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME)) != @@ -3810,9 +3821,7 @@ static int b43_setup_modes(struct b43_wldev *dev, bool have_2ghz_phy, bool have_5ghz_phy) { struct ieee80211_hw *hw = dev->wl->hw; - struct ieee80211_hw_mode *mode; struct b43_phy *phy = &dev->phy; - int err; /* XXX: This function will go away soon, when mac80211 * band stuff is rewritten. So this is just a hack. @@ -3821,15 +3830,7 @@ static int b43_setup_modes(struct b43_wldev *dev, * This assumption is OK, as any B, N or A PHY will already * have died a horrible sanity check death earlier. */ - mode = &phy->hwmodes[0]; - mode->mode = MODE_IEEE80211G; - mode->num_channels = b43_2ghz_chantable_size; - mode->channels = b43_2ghz_chantable; - mode->num_rates = b43_g_ratetable_size; - mode->rates = b43_g_ratetable; - err = ieee80211_register_hwmode(hw, mode); - if (err) - return err; + hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &b43_band_2GHz; phy->possible_phymodes |= B43_PHYMODE_G; return 0; diff --git a/drivers/net/wireless/b43/sysfs.c b/drivers/net/wireless/b43/sysfs.c index f4faff6a7d6c..275095b8cbe7 100644 --- a/drivers/net/wireless/b43/sysfs.c +++ b/drivers/net/wireless/b43/sysfs.c @@ -47,29 +47,6 @@ static int get_integer(const char *buf, size_t count) return ret; } -static int get_boolean(const char *buf, size_t count) -{ - if (count != 0) { - if (buf[0] == '1') - return 1; - if (buf[0] == '0') - return 0; - if (count >= 4 && memcmp(buf, "true", 4) == 0) - return 1; - if (count >= 5 && memcmp(buf, "false", 5) == 0) - return 0; - if (count >= 3 && memcmp(buf, "yes", 3) == 0) - return 1; - if (count >= 2 && memcmp(buf, "no", 2) == 0) - return 0; - if (count >= 2 && memcmp(buf, "on", 2) == 0) - return 1; - if (count >= 3 && memcmp(buf, "off", 3) == 0) - return 0; - } - return -EINVAL; -} - static ssize_t b43_attr_interfmode_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -155,82 +132,18 @@ static ssize_t b43_attr_interfmode_store(struct device *dev, static DEVICE_ATTR(interference, 0644, b43_attr_interfmode_show, b43_attr_interfmode_store); -static ssize_t b43_attr_preamble_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct b43_wldev *wldev = dev_to_b43_wldev(dev); - ssize_t count; - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - mutex_lock(&wldev->wl->mutex); - - if (wldev->short_preamble) - count = - snprintf(buf, PAGE_SIZE, "1 (Short Preamble enabled)\n"); - else - count = - snprintf(buf, PAGE_SIZE, "0 (Short Preamble disabled)\n"); - - mutex_unlock(&wldev->wl->mutex); - - return count; -} - -static ssize_t b43_attr_preamble_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct b43_wldev *wldev = dev_to_b43_wldev(dev); - unsigned long flags; - int value; - - if (!capable(CAP_NET_ADMIN)) - return -EPERM; - - value = get_boolean(buf, count); - if (value < 0) - return value; - mutex_lock(&wldev->wl->mutex); - spin_lock_irqsave(&wldev->wl->irq_lock, flags); - - wldev->short_preamble = !!value; - - spin_unlock_irqrestore(&wldev->wl->irq_lock, flags); - mutex_unlock(&wldev->wl->mutex); - - return count; -} - -static DEVICE_ATTR(shortpreamble, 0644, - b43_attr_preamble_show, b43_attr_preamble_store); - int b43_sysfs_register(struct b43_wldev *wldev) { struct device *dev = wldev->dev->dev; - int err; B43_WARN_ON(b43_status(wldev) != B43_STAT_INITIALIZED); - err = device_create_file(dev, &dev_attr_interference); - if (err) - goto out; - err = device_create_file(dev, &dev_attr_shortpreamble); - if (err) - goto err_remove_interfmode; - - out: - return err; - err_remove_interfmode: - device_remove_file(dev, &dev_attr_interference); - goto out; + return device_create_file(dev, &dev_attr_interference); } void b43_sysfs_unregister(struct b43_wldev *wldev) { struct device *dev = wldev->dev->dev; - device_remove_file(dev, &dev_attr_shortpreamble); device_remove_file(dev, &dev_attr_interference); } diff --git a/drivers/net/wireless/b43/xmit.c b/drivers/net/wireless/b43/xmit.c index 7caa26eb4105..4014b6c8272b 100644 --- a/drivers/net/wireless/b43/xmit.c +++ b/drivers/net/wireless/b43/xmit.c @@ -32,46 +32,48 @@ #include "dma.h" -/* Extract the bitrate out of a CCK PLCP header. */ -static u8 b43_plcp_get_bitrate_cck(struct b43_plcp_hdr6 *plcp) +/* Extract the bitrate index out of a CCK PLCP header. */ +static int b43_plcp_get_bitrate_idx_cck(struct b43_plcp_hdr6 *plcp) { switch (plcp->raw[0]) { case 0x0A: - return B43_CCK_RATE_1MB; + return 0; case 0x14: - return B43_CCK_RATE_2MB; + return 1; case 0x37: - return B43_CCK_RATE_5MB; + return 2; case 0x6E: - return B43_CCK_RATE_11MB; + return 3; } B43_WARN_ON(1); - return 0; + return -1; } -/* Extract the bitrate out of an OFDM PLCP header. */ -static u8 b43_plcp_get_bitrate_ofdm(struct b43_plcp_hdr6 *plcp) +/* Extract the bitrate index out of an OFDM PLCP header. */ +static u8 b43_plcp_get_bitrate_idx_ofdm(struct b43_plcp_hdr6 *plcp, bool aphy) { + int base = aphy ? 0 : 4; + switch (plcp->raw[0] & 0xF) { case 0xB: - return B43_OFDM_RATE_6MB; + return base + 0; case 0xF: - return B43_OFDM_RATE_9MB; + return base + 1; case 0xA: - return B43_OFDM_RATE_12MB; + return base + 2; case 0xE: - return B43_OFDM_RATE_18MB; + return base + 3; case 0x9: - return B43_OFDM_RATE_24MB; + return base + 4; case 0xD: - return B43_OFDM_RATE_36MB; + return base + 5; case 0x8: - return B43_OFDM_RATE_48MB; + return base + 6; case 0xC: - return B43_OFDM_RATE_54MB; + return base + 7; } B43_WARN_ON(1); - return 0; + return -1; } u8 b43_plcp_get_ratecode_cck(const u8 bitrate) @@ -191,6 +193,7 @@ int b43_generate_txhdr(struct b43_wldev *dev, (const struct ieee80211_hdr *)fragment_data; int use_encryption = (!(txctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT)); u16 fctl = le16_to_cpu(wlhdr->frame_control); + struct ieee80211_rate *fbrate; u8 rate, rate_fb; int rate_ofdm, rate_fb_ofdm; unsigned int plcp_fragment_len; @@ -200,9 +203,11 @@ int b43_generate_txhdr(struct b43_wldev *dev, memset(txhdr, 0, sizeof(*txhdr)); - rate = txctl->tx_rate; + WARN_ON(!txctl->tx_rate); + rate = txctl->tx_rate ? txctl->tx_rate->hw_value : B43_CCK_RATE_1MB; rate_ofdm = b43_is_ofdm_rate(rate); - rate_fb = (txctl->alt_retry_rate == -1) ? rate : txctl->alt_retry_rate; + fbrate = txctl->alt_retry_rate ? : txctl->tx_rate; + rate_fb = fbrate->hw_value; rate_fb_ofdm = b43_is_ofdm_rate(rate_fb); if (rate_ofdm) @@ -221,11 +226,10 @@ int b43_generate_txhdr(struct b43_wldev *dev, * use the original dur_id field. */ txhdr->dur_fb = wlhdr->duration_id; } else { - int fbrate_base100kbps = B43_RATE_TO_BASE100KBPS(rate_fb); txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw, txctl->vif, fragment_len, - fbrate_base100kbps); + fbrate); } plcp_fragment_len = fragment_len + FCS_LEN; @@ -287,7 +291,7 @@ int b43_generate_txhdr(struct b43_wldev *dev, phy_ctl |= B43_TXH_PHY_ENC_OFDM; else phy_ctl |= B43_TXH_PHY_ENC_CCK; - if (dev->short_preamble) + if (txctl->flags & IEEE80211_TXCTL_SHORT_PREAMBLE) phy_ctl |= B43_TXH_PHY_SHORTPRMBL; switch (b43_ieee80211_antenna_sanitize(dev, txctl->antenna_sel_tx)) { @@ -332,7 +336,8 @@ int b43_generate_txhdr(struct b43_wldev *dev, int rts_rate_ofdm, rts_rate_fb_ofdm; struct b43_plcp_hdr6 *plcp; - rts_rate = txctl->rts_cts_rate; + WARN_ON(!txctl->rts_cts_rate); + rts_rate = txctl->rts_cts_rate ? txctl->rts_cts_rate->hw_value : B43_CCK_RATE_1MB; rts_rate_ofdm = b43_is_ofdm_rate(rts_rate); rts_rate_fb = b43_calc_fallback_rate(rts_rate); rts_rate_fb_ofdm = b43_is_ofdm_rate(rts_rate_fb); @@ -506,6 +511,7 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) u16 phystat0, phystat3, chanstat, mactime; u32 macstat; u16 chanid; + u16 phytype; u8 jssi; int padding; @@ -518,6 +524,7 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) macstat = le32_to_cpu(rxhdr->mac_status); mactime = le16_to_cpu(rxhdr->mac_time); chanstat = le16_to_cpu(rxhdr->channel); + phytype = chanstat & B43_RX_CHAN_PHYTYPE; if (macstat & B43_RX_MAC_FCSERR) dev->wl->ieee_stats.dot11FCSErrorCount++; @@ -575,9 +582,10 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) /* the next line looks wrong, but is what mac80211 wants */ status.signal = (jssi * 100) / B43_RX_MAX_SSI; if (phystat0 & B43_RX_PHYST0_OFDM) - status.rate = b43_plcp_get_bitrate_ofdm(plcp); + status.rate_idx = b43_plcp_get_bitrate_idx_ofdm(plcp, + phytype == B43_PHYTYPE_A); else - status.rate = b43_plcp_get_bitrate_cck(plcp); + status.rate_idx = b43_plcp_get_bitrate_idx_cck(plcp); status.antenna = !!(phystat0 & B43_RX_PHYST0_ANT); /* @@ -601,29 +609,28 @@ void b43_rx(struct b43_wldev *dev, struct sk_buff *skb, const void *_rxhdr) chanid = (chanstat & B43_RX_CHAN_ID) >> B43_RX_CHAN_ID_SHIFT; switch (chanstat & B43_RX_CHAN_PHYTYPE) { case B43_PHYTYPE_A: - status.phymode = MODE_IEEE80211A; + status.band = IEEE80211_BAND_5GHZ; B43_WARN_ON(1); /* FIXME: We don't really know which value the "chanid" contains. * So the following assignment might be wrong. */ - status.channel = chanid; - status.freq = b43_channel_to_freq_5ghz(status.channel); + status.freq = b43_channel_to_freq_5ghz(chanid); break; case B43_PHYTYPE_G: - status.phymode = MODE_IEEE80211G; + status.band = IEEE80211_BAND_2GHZ; /* chanid is the radio channel cookie value as used * to tune the radio. */ status.freq = chanid + 2400; - status.channel = b43_freq_to_channel_2ghz(status.freq); break; case B43_PHYTYPE_N: - status.phymode = 0xDEAD /*FIXME MODE_IEEE80211N*/; /* chanid is the SHM channel cookie. Which is the plain * channel number in b43. */ - status.channel = chanid; - if (chanstat & B43_RX_CHAN_5GHZ) - status.freq = b43_freq_to_channel_5ghz(status.freq); - else - status.freq = b43_freq_to_channel_2ghz(status.freq); + if (chanstat & B43_RX_CHAN_5GHZ) { + status.band = IEEE80211_BAND_5GHZ; + status.freq = b43_freq_to_channel_5ghz(chanid); + } else { + status.band = IEEE80211_BAND_2GHZ; + status.freq = b43_freq_to_channel_2ghz(chanid); + } break; default: B43_WARN_ON(1); diff --git a/drivers/net/wireless/b43legacy/b43legacy.h b/drivers/net/wireless/b43legacy/b43legacy.h index 93d45b71799a..5f217d6d2e6a 100644 --- a/drivers/net/wireless/b43legacy/b43legacy.h +++ b/drivers/net/wireless/b43legacy/b43legacy.h @@ -392,10 +392,6 @@ struct b43legacy_phy { u8 possible_phymodes; /* GMODE bit enabled in MACCTL? */ bool gmode; - /* Possible ieee80211 subsystem hwmodes for this PHY. - * Which mode is selected, depends on thr GMODE enabled bit */ -#define B43legacy_MAX_PHYHWMODES 2 - struct ieee80211_hw_mode hwmodes[B43legacy_MAX_PHYHWMODES]; /* Analog Type */ u8 analog; diff --git a/drivers/net/wireless/b43legacy/main.c b/drivers/net/wireless/b43legacy/main.c index c39de422e220..d2a72a2cd178 100644 --- a/drivers/net/wireless/b43legacy/main.c +++ b/drivers/net/wireless/b43legacy/main.c @@ -95,28 +95,29 @@ MODULE_DEVICE_TABLE(ssb, b43legacy_ssb_tbl); * data in there. This data is the same for all devices, so we don't * get concurrency issues */ #define RATETAB_ENT(_rateid, _flags) \ - { \ - .rate = B43legacy_RATE_TO_100KBPS(_rateid), \ - .val = (_rateid), \ - .val2 = (_rateid), \ - .flags = (_flags), \ + { \ + .bitrate = B43legacy_RATE_TO_100KBPS(_rateid), \ + .hw_value = (_rateid), \ + .flags = (_flags), \ } +/* + * NOTE: When changing this, sync with xmit.c's + * b43legacy_plcp_get_bitrate_idx_* functions! + */ static struct ieee80211_rate __b43legacy_ratetable[] = { - RATETAB_ENT(B43legacy_CCK_RATE_1MB, IEEE80211_RATE_CCK), - RATETAB_ENT(B43legacy_CCK_RATE_2MB, IEEE80211_RATE_CCK_2), - RATETAB_ENT(B43legacy_CCK_RATE_5MB, IEEE80211_RATE_CCK_2), - RATETAB_ENT(B43legacy_CCK_RATE_11MB, IEEE80211_RATE_CCK_2), - RATETAB_ENT(B43legacy_OFDM_RATE_6MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_9MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_12MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_18MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_24MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_36MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_48MB, IEEE80211_RATE_OFDM), - RATETAB_ENT(B43legacy_OFDM_RATE_54MB, IEEE80211_RATE_OFDM), + RATETAB_ENT(B43legacy_CCK_RATE_1MB, 0), + RATETAB_ENT(B43legacy_CCK_RATE_2MB, IEEE80211_RATE_SHORT_PREAMBLE), + RATETAB_ENT(B43legacy_CCK_RATE_5MB, IEEE80211_RATE_SHORT_PREAMBLE), + RATETAB_ENT(B43legacy_CCK_RATE_11MB, IEEE80211_RATE_SHORT_PREAMBLE), + RATETAB_ENT(B43legacy_OFDM_RATE_6MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_9MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_12MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_18MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_24MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_36MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_48MB, 0), + RATETAB_ENT(B43legacy_OFDM_RATE_54MB, 0), }; -#define b43legacy_a_ratetable (__b43legacy_ratetable + 4) -#define b43legacy_a_ratetable_size 8 #define b43legacy_b_ratetable (__b43legacy_ratetable + 0) #define b43legacy_b_ratetable_size 4 #define b43legacy_g_ratetable (__b43legacy_ratetable + 0) @@ -124,14 +125,8 @@ static struct ieee80211_rate __b43legacy_ratetable[] = { #define CHANTAB_ENT(_chanid, _freq) \ { \ - .chan = (_chanid), \ - .freq = (_freq), \ - .val = (_chanid), \ - .flag = IEEE80211_CHAN_W_SCAN | \ - IEEE80211_CHAN_W_ACTIVE_SCAN | \ - IEEE80211_CHAN_W_IBSS, \ - .power_level = 0x0A, \ - .antenna_max = 0xFF, \ + .center_freq = (_freq), \ + .hw_value = (_chanid), \ } static struct ieee80211_channel b43legacy_bg_chantable[] = { CHANTAB_ENT(1, 2412), @@ -149,7 +144,20 @@ static struct ieee80211_channel b43legacy_bg_chantable[] = { CHANTAB_ENT(13, 2472), CHANTAB_ENT(14, 2484), }; -#define b43legacy_bg_chantable_size ARRAY_SIZE(b43legacy_bg_chantable) + +static struct ieee80211_supported_band b43legacy_band_2GHz_BPHY = { + .channels = b43legacy_bg_chantable, + .n_channels = ARRAY_SIZE(b43legacy_bg_chantable), + .bitrates = b43legacy_b_ratetable, + .n_bitrates = b43legacy_b_ratetable_size, +}; + +static struct ieee80211_supported_band b43legacy_band_2GHz_GPHY = { + .channels = b43legacy_bg_chantable, + .n_channels = ARRAY_SIZE(b43legacy_bg_chantable), + .bitrates = b43legacy_g_ratetable, + .n_bitrates = b43legacy_g_ratetable_size, +}; static void b43legacy_wireless_core_exit(struct b43legacy_wldev *dev); static int b43legacy_wireless_core_init(struct b43legacy_wldev *dev); @@ -969,18 +977,18 @@ static void b43legacy_write_beacon_template(struct b43legacy_wldev *dev, static void b43legacy_write_probe_resp_plcp(struct b43legacy_wldev *dev, u16 shm_offset, u16 size, - u8 rate) + struct ieee80211_rate *rate) { struct b43legacy_plcp_hdr4 plcp; u32 tmp; __le16 dur; plcp.data = 0; - b43legacy_generate_plcp_hdr(&plcp, size + FCS_LEN, rate); + b43legacy_generate_plcp_hdr(&plcp, size + FCS_LEN, rate->bitrate); dur = ieee80211_generic_frame_duration(dev->wl->hw, dev->wl->vif, size, - B43legacy_RATE_TO_100KBPS(rate)); + rate); /* Write PLCP in two parts and timing for packet transfer */ tmp = le32_to_cpu(plcp.data); b43legacy_shm_write16(dev, B43legacy_SHM_SHARED, shm_offset, @@ -998,7 +1006,8 @@ static void b43legacy_write_probe_resp_plcp(struct b43legacy_wldev *dev, * 3) Stripping TIM */ static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev, - u16 *dest_size, u8 rate) + u16 *dest_size, + struct ieee80211_rate *rate) { const u8 *src_data; u8 *dest_data; @@ -1046,7 +1055,7 @@ static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev, dur = ieee80211_generic_frame_duration(dev->wl->hw, dev->wl->vif, *dest_size, - B43legacy_RATE_TO_100KBPS(rate)); + rate); hdr->duration_id = dur; return dest_data; @@ -1054,7 +1063,8 @@ static u8 *b43legacy_generate_probe_resp(struct b43legacy_wldev *dev, static void b43legacy_write_probe_resp_template(struct b43legacy_wldev *dev, u16 ram_offset, - u16 shm_size_offset, u8 rate) + u16 shm_size_offset, + struct ieee80211_rate *rate) { u8 *probe_resp_data; u16 size; @@ -1069,19 +1079,19 @@ static void b43legacy_write_probe_resp_template(struct b43legacy_wldev *dev, * all possible basic rates */ b43legacy_write_probe_resp_plcp(dev, 0x31A, size, - B43legacy_CCK_RATE_1MB); + &b43legacy_b_ratetable[0]); b43legacy_write_probe_resp_plcp(dev, 0x32C, size, - B43legacy_CCK_RATE_2MB); + &b43legacy_b_ratetable[1]); b43legacy_write_probe_resp_plcp(dev, 0x33E, size, - B43legacy_CCK_RATE_5MB); + &b43legacy_b_ratetable[2]); b43legacy_write_probe_resp_plcp(dev, 0x350, size, - B43legacy_CCK_RATE_11MB); + &b43legacy_b_ratetable[3]); size = min((size_t)size, 0x200 - sizeof(struct b43legacy_plcp_hdr6)); b43legacy_write_template_common(dev, probe_resp_data, size, ram_offset, - shm_size_offset, rate); + shm_size_offset, rate->bitrate); kfree(probe_resp_data); } @@ -1106,7 +1116,7 @@ static void b43legacy_update_templates(struct b43legacy_wldev *dev) b43legacy_write_beacon_template(dev, 0x468, 0x1A, B43legacy_CCK_RATE_1MB); b43legacy_write_probe_resp_template(dev, 0x268, 0x4A, - B43legacy_CCK_RATE_11MB); + &b43legacy_b_ratetable[0]); status = b43legacy_read32(dev, B43legacy_MMIO_MACCMD); status |= 0x03; @@ -2550,14 +2560,16 @@ static int b43legacy_op_dev_config(struct ieee80211_hw *hw, antenna_rx = b43legacy_antenna_from_ieee80211(conf->antenna_sel_rx); mutex_lock(&wl->mutex); + dev = wl->current_dev; + phy = &dev->phy; /* Switch the PHY mode (if necessary). */ - switch (conf->phymode) { - case MODE_IEEE80211B: - new_phymode = B43legacy_PHYMODE_B; - break; - case MODE_IEEE80211G: - new_phymode = B43legacy_PHYMODE_G; + switch (conf->channel->band) { + case IEEE80211_BAND_2GHZ: + if (phy->type == B43legacy_PHYTYPE_B) + new_phymode = B43legacy_PHYMODE_B; + else + new_phymode = B43legacy_PHYMODE_G; break; default: B43legacy_WARN_ON(1); @@ -2565,8 +2577,6 @@ static int b43legacy_op_dev_config(struct ieee80211_hw *hw, err = b43legacy_switch_phymode(wl, new_phymode); if (err) goto out_unlock_mutex; - dev = wl->current_dev; - phy = &dev->phy; /* Disable IRQs while reconfiguring the device. * This makes it possible to drop the spinlock throughout @@ -2582,8 +2592,8 @@ static int b43legacy_op_dev_config(struct ieee80211_hw *hw, /* Switch to the requested channel. * The firmware takes care of races with the TX handler. */ - if (conf->channel_val != phy->channel) - b43legacy_radio_selectchannel(dev, conf->channel_val, 0); + if (conf->channel->hw_value != phy->channel) + b43legacy_radio_selectchannel(dev, conf->channel->hw_value, 0); /* Enable/Disable ShortSlot timing. */ if ((!!(conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME)) @@ -3398,48 +3408,19 @@ static int b43legacy_setup_modes(struct b43legacy_wldev *dev, int have_gphy) { struct ieee80211_hw *hw = dev->wl->hw; - struct ieee80211_hw_mode *mode; struct b43legacy_phy *phy = &dev->phy; - int cnt = 0; - int err; phy->possible_phymodes = 0; - for (; 1; cnt++) { - if (have_bphy) { - B43legacy_WARN_ON(cnt >= B43legacy_MAX_PHYHWMODES); - mode = &phy->hwmodes[cnt]; - - mode->mode = MODE_IEEE80211B; - mode->num_channels = b43legacy_bg_chantable_size; - mode->channels = b43legacy_bg_chantable; - mode->num_rates = b43legacy_b_ratetable_size; - mode->rates = b43legacy_b_ratetable; - err = ieee80211_register_hwmode(hw, mode); - if (err) - return err; - - phy->possible_phymodes |= B43legacy_PHYMODE_B; - have_bphy = 0; - continue; - } - if (have_gphy) { - B43legacy_WARN_ON(cnt >= B43legacy_MAX_PHYHWMODES); - mode = &phy->hwmodes[cnt]; - - mode->mode = MODE_IEEE80211G; - mode->num_channels = b43legacy_bg_chantable_size; - mode->channels = b43legacy_bg_chantable; - mode->num_rates = b43legacy_g_ratetable_size; - mode->rates = b43legacy_g_ratetable; - err = ieee80211_register_hwmode(hw, mode); - if (err) - return err; - - phy->possible_phymodes |= B43legacy_PHYMODE_G; - have_gphy = 0; - continue; - } - break; + if (have_bphy) { + hw->wiphy->bands[IEEE80211_BAND_2GHZ] = + &b43legacy_band_2GHz_BPHY; + phy->possible_phymodes |= B43legacy_PHYMODE_B; + } + + if (have_gphy) { + hw->wiphy->bands[IEEE80211_BAND_2GHZ] = + &b43legacy_band_2GHz_GPHY; + phy->possible_phymodes |= B43legacy_PHYMODE_G; } return 0; diff --git a/drivers/net/wireless/b43legacy/xmit.c b/drivers/net/wireless/b43legacy/xmit.c index d84408a82db9..47e130e9fdf2 100644 --- a/drivers/net/wireless/b43legacy/xmit.c +++ b/drivers/net/wireless/b43legacy/xmit.c @@ -37,45 +37,48 @@ /* Extract the bitrate out of a CCK PLCP header. */ -static u8 b43legacy_plcp_get_bitrate_cck(struct b43legacy_plcp_hdr6 *plcp) +static u8 b43legacy_plcp_get_bitrate_idx_cck(struct b43legacy_plcp_hdr6 *plcp) { switch (plcp->raw[0]) { case 0x0A: - return B43legacy_CCK_RATE_1MB; + return 0; case 0x14: - return B43legacy_CCK_RATE_2MB; + return 1; case 0x37: - return B43legacy_CCK_RATE_5MB; + return 2; case 0x6E: - return B43legacy_CCK_RATE_11MB; + return 3; } B43legacy_BUG_ON(1); - return 0; + return -1; } /* Extract the bitrate out of an OFDM PLCP header. */ -static u8 b43legacy_plcp_get_bitrate_ofdm(struct b43legacy_plcp_hdr6 *plcp) +static u8 b43legacy_plcp_get_bitrate_idx_ofdm(struct b43legacy_plcp_hdr6 *plcp, + bool aphy) { + int base = aphy ? 0 : 4; + switch (plcp->raw[0] & 0xF) { case 0xB: - return B43legacy_OFDM_RATE_6MB; + return base + 0; case 0xF: - return B43legacy_OFDM_RATE_9MB; + return base + 1; case 0xA: - return B43legacy_OFDM_RATE_12MB; + return base + 2; case 0xE: - return B43legacy_OFDM_RATE_18MB; + return base + 3; case 0x9: - return B43legacy_OFDM_RATE_24MB; + return base + 4; case 0xD: - return B43legacy_OFDM_RATE_36MB; + return base + 5; case 0x8: - return B43legacy_OFDM_RATE_48MB; + return base + 6; case 0xC: - return B43legacy_OFDM_RATE_54MB; + return base + 7; } B43legacy_BUG_ON(1); - return 0; + return -1; } u8 b43legacy_plcp_get_ratecode_cck(const u8 bitrate) @@ -192,7 +195,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev, int use_encryption = (!(txctl->flags & IEEE80211_TXCTL_DO_NOT_ENCRYPT)); u16 fctl; u8 rate; - u8 rate_fb; + struct ieee80211_rate *rate_fb; int rate_ofdm; int rate_fb_ofdm; unsigned int plcp_fragment_len; @@ -204,16 +207,16 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev, memset(txhdr, 0, sizeof(*txhdr)); - rate = txctl->tx_rate; + rate = txctl->tx_rate->hw_value; rate_ofdm = b43legacy_is_ofdm_rate(rate); - rate_fb = (txctl->alt_retry_rate == -1) ? rate : txctl->alt_retry_rate; - rate_fb_ofdm = b43legacy_is_ofdm_rate(rate_fb); + rate_fb = txctl->alt_retry_rate ? : txctl->tx_rate; + rate_fb_ofdm = b43legacy_is_ofdm_rate(rate_fb->hw_value); txhdr->mac_frame_ctl = wlhdr->frame_control; memcpy(txhdr->tx_receiver, wlhdr->addr1, 6); /* Calculate duration for fallback rate */ - if ((rate_fb == rate) || + if ((rate_fb->hw_value == rate) || (wlhdr->duration_id & cpu_to_le16(0x8000)) || (wlhdr->duration_id == cpu_to_le16(0))) { /* If the fallback rate equals the normal rate or the @@ -221,11 +224,10 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev, * use the original dur_id field. */ txhdr->dur_fb = wlhdr->duration_id; } else { - int fbrate_base100kbps = B43legacy_RATE_TO_100KBPS(rate_fb); txhdr->dur_fb = ieee80211_generic_frame_duration(dev->wl->hw, txctl->vif, fragment_len, - fbrate_base100kbps); + rate_fb); } plcp_fragment_len = fragment_len + FCS_LEN; @@ -266,7 +268,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev, rate); b43legacy_generate_plcp_hdr((struct b43legacy_plcp_hdr4 *) (&txhdr->plcp_fb), plcp_fragment_len, - rate_fb); + rate_fb->hw_value); /* PHY TX Control word */ if (rate_ofdm) @@ -310,7 +312,7 @@ static int generate_txhdr_fw3(struct b43legacy_wldev *dev, int rts_rate_ofdm; int rts_rate_fb_ofdm; - rts_rate = txctl->rts_cts_rate; + rts_rate = txctl->rts_cts_rate->hw_value; rts_rate_ofdm = b43legacy_is_ofdm_rate(rts_rate); rts_rate_fb = b43legacy_calc_fallback_rate(rts_rate); rts_rate_fb_ofdm = b43legacy_is_ofdm_rate(rts_rate_fb); @@ -536,10 +538,11 @@ void b43legacy_rx(struct b43legacy_wldev *dev, (phystat3 & B43legacy_RX_PHYST3_TRSTATE)); status.noise = dev->stats.link_noise; status.signal = (jssi * 100) / B43legacy_RX_MAX_SSI; + /* change to support A PHY */ if (phystat0 & B43legacy_RX_PHYST0_OFDM) - status.rate = b43legacy_plcp_get_bitrate_ofdm(plcp); + status.rate_idx = b43legacy_plcp_get_bitrate_idx_ofdm(plcp, false); else - status.rate = b43legacy_plcp_get_bitrate_cck(plcp); + status.rate_idx = b43legacy_plcp_get_bitrate_idx_cck(plcp); status.antenna = !!(phystat0 & B43legacy_RX_PHYST0_ANT); /* @@ -564,14 +567,9 @@ void b43legacy_rx(struct b43legacy_wldev *dev, B43legacy_RX_CHAN_ID_SHIFT; switch (chanstat & B43legacy_RX_CHAN_PHYTYPE) { case B43legacy_PHYTYPE_B: - status.phymode = MODE_IEEE80211B; - status.freq = chanid + 2400; - status.channel = b43legacy_freq_to_channel_bg(chanid + 2400); - break; case B43legacy_PHYTYPE_G: - status.phymode = MODE_IEEE80211G; + status.band = IEEE80211_BAND_2GHZ; status.freq = chanid + 2400; - status.channel = b43legacy_freq_to_channel_bg(chanid + 2400); break; default: b43legacywarn(dev->wl, "Unexpected value for chanstat (0x%X)\n", diff --git a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c index 80d31ae51e77..f018ce464d97 100644 --- a/drivers/net/wireless/iwlwifi/iwl-3945-rs.c +++ b/drivers/net/wireless/iwlwifi/iwl-3945-rs.c @@ -100,14 +100,6 @@ static struct iwl3945_tpt_entry iwl3945_tpt_table_a[] = { {-89, IWL_RATE_6M_INDEX} }; -static struct iwl3945_tpt_entry iwl3945_tpt_table_b[] = { - {-86, IWL_RATE_11M_INDEX}, - {-88, IWL_RATE_5M_INDEX}, - {-90, IWL_RATE_2M_INDEX}, - {-92, IWL_RATE_1M_INDEX} - -}; - static struct iwl3945_tpt_entry iwl3945_tpt_table_g[] = { {-60, IWL_RATE_54M_INDEX}, {-64, IWL_RATE_48M_INDEX}, @@ -129,7 +121,7 @@ static struct iwl3945_tpt_entry iwl3945_tpt_table_g[] = { #define IWL_RATE_MIN_SUCCESS_TH 8 #define IWL_RATE_DECREASE_TH 1920 -static u8 iwl3945_get_rate_index_by_rssi(s32 rssi, u8 mode) +static u8 iwl3945_get_rate_index_by_rssi(s32 rssi, enum ieee80211_band band) { u32 index = 0; u32 table_size = 0; @@ -138,21 +130,19 @@ static u8 iwl3945_get_rate_index_by_rssi(s32 rssi, u8 mode) if ((rssi < IWL_MIN_RSSI_VAL) || (rssi > IWL_MAX_RSSI_VAL)) rssi = IWL_MIN_RSSI_VAL; - switch (mode) { - case MODE_IEEE80211G: + switch (band) { + case IEEE80211_BAND_2GHZ: tpt_table = iwl3945_tpt_table_g; table_size = ARRAY_SIZE(iwl3945_tpt_table_g); break; - case MODE_IEEE80211A: + case IEEE80211_BAND_5GHZ: tpt_table = iwl3945_tpt_table_a; table_size = ARRAY_SIZE(iwl3945_tpt_table_a); break; default: - case MODE_IEEE80211B: - tpt_table = iwl3945_tpt_table_b; - table_size = ARRAY_SIZE(iwl3945_tpt_table_b); + BUG(); break; } @@ -340,17 +330,17 @@ static void rs_rate_init(void *priv_rate, void *priv_sta, * after assoc.. */ for (i = IWL_RATE_COUNT - 1; i >= 0; i--) { - if (sta->supp_rates & (1 << i)) { - sta->txrate = i; + if (sta->supp_rates[local->hw.conf.channel->band] & (1 << i)) { + sta->txrate_idx = i; break; } } - sta->last_txrate = sta->txrate; + sta->last_txrate_idx = sta->txrate_idx; - /* For MODE_IEEE80211A mode it start at IWL_FIRST_OFDM_RATE */ - if (local->hw.conf.phymode == MODE_IEEE80211A) - sta->last_txrate += IWL_FIRST_OFDM_RATE; + /* For 5 GHz band it start at IWL_FIRST_OFDM_RATE */ + if (local->hw.conf.channel->band == IEEE80211_BAND_5GHZ) + sta->last_txrate_idx += IWL_FIRST_OFDM_RATE; IWL_DEBUG_RATE("leave\n"); } @@ -429,17 +419,19 @@ static int rs_adjust_next_rate(struct iwl3945_priv *priv, int rate) { int next_rate = iwl3945_get_prev_ieee_rate(rate); - switch (priv->phymode) { - case MODE_IEEE80211A: + switch (priv->band) { + case IEEE80211_BAND_5GHZ: if (rate == IWL_RATE_12M_INDEX) next_rate = IWL_RATE_9M_INDEX; else if (rate == IWL_RATE_6M_INDEX) next_rate = IWL_RATE_6M_INDEX; break; +/* XXX cannot be invoked in current mac80211 so not a regression case MODE_IEEE80211B: if (rate == IWL_RATE_11M_INDEX_TABLE) next_rate = IWL_RATE_5M_INDEX_TABLE; break; + */ default: break; } @@ -465,15 +457,17 @@ static void rs_tx_status(void *priv_rate, struct iwl3945_priv *priv = (struct iwl3945_priv *)priv_rate; struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); struct iwl3945_rs_sta *rs_sta; + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; IWL_DEBUG_RATE("enter\n"); retries = tx_resp->retry_count; - first_index = tx_resp->control.tx_rate; + first_index = &sband->bitrates[0] - tx_resp->control.tx_rate; if ((first_index < 0) || (first_index >= IWL_RATE_COUNT)) { - IWL_DEBUG_RATE("leave: Rate out of bounds: %0x for %d\n", - tx_resp->control.tx_rate, first_index); + IWL_DEBUG_RATE("leave: Rate out of bounds: %d\n", first_index); return; } @@ -561,14 +555,14 @@ static void rs_tx_status(void *priv_rate, } static u16 iwl3945_get_adjacent_rate(struct iwl3945_rs_sta *rs_sta, - u8 index, u16 rate_mask, int phymode) + u8 index, u16 rate_mask, enum ieee80211_band band) { u8 high = IWL_RATE_INVALID; u8 low = IWL_RATE_INVALID; /* 802.11A walks to the next literal adjacent rate in * the rate table */ - if (unlikely(phymode == MODE_IEEE80211A)) { + if (unlikely(band == IEEE80211_BAND_5GHZ)) { int i; u32 mask; @@ -639,7 +633,8 @@ static u16 iwl3945_get_adjacent_rate(struct iwl3945_rs_sta *rs_sta, * */ static void rs_get_rate(void *priv_rate, struct net_device *dev, - struct ieee80211_hw_mode *mode, struct sk_buff *skb, + struct ieee80211_supported_band *band, + struct sk_buff *skb, struct rate_selection *sel) { u8 low = IWL_RATE_INVALID; @@ -672,16 +667,16 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev, is_multicast_ether_addr(hdr->addr1) || !sta || !sta->rate_ctrl_priv) { IWL_DEBUG_RATE("leave: No STA priv data to update!\n"); - sel->rate = rate_lowest(local, local->oper_hw_mode, sta); + sel->rate = rate_lowest(local, band, sta); if (sta) sta_info_put(sta); return; } - rate_mask = sta->supp_rates; - index = min(sta->last_txrate & 0xffff, IWL_RATE_COUNT - 1); + rate_mask = sta->supp_rates[band->band]; + index = min(sta->last_txrate_idx & 0xffff, IWL_RATE_COUNT - 1); - if (priv->phymode == (u8) MODE_IEEE80211A) + if (priv->band == IEEE80211_BAND_5GHZ) rate_mask = rate_mask << IWL_FIRST_OFDM_RATE; rs_sta = (void *)sta->rate_ctrl_priv; @@ -732,7 +727,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev, current_tpt = window->average_tpt; high_low = iwl3945_get_adjacent_rate(rs_sta, index, rate_mask, - local->hw.conf.phymode); + band->band); low = high_low & 0xff; high = (high_low >> 8) & 0xff; @@ -810,11 +805,11 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev, out: - sta->last_txrate = index; - if (priv->phymode == (u8) MODE_IEEE80211A) - sta->txrate = sta->last_txrate - IWL_FIRST_OFDM_RATE; + sta->last_txrate_idx = index; + if (priv->band == IEEE80211_BAND_5GHZ) + sta->txrate_idx = sta->last_txrate_idx - IWL_FIRST_OFDM_RATE; else - sta->txrate = sta->last_txrate; + sta->txrate_idx = sta->last_txrate_idx; sta_info_put(sta); @@ -945,8 +940,9 @@ void iwl3945_rate_scale_init(struct ieee80211_hw *hw, s32 sta_id) spin_lock_irqsave(&rs_sta->lock, flags); rs_sta->tgg = 0; - switch (priv->phymode) { - case MODE_IEEE80211G: + switch (priv->band) { + case IEEE80211_BAND_2GHZ: + /* TODO: this always does G, not a regression */ if (priv->active_rxon.flags & RXON_FLG_TGG_PROTECT_MSK) { rs_sta->tgg = 1; rs_sta->expected_tpt = iwl3945_expected_tpt_g_prot; @@ -954,14 +950,11 @@ void iwl3945_rate_scale_init(struct ieee80211_hw *hw, s32 sta_id) rs_sta->expected_tpt = iwl3945_expected_tpt_g; break; - case MODE_IEEE80211A: + case IEEE80211_BAND_5GHZ: rs_sta->expected_tpt = iwl3945_expected_tpt_a; break; - - default: - IWL_WARNING("Invalid phymode. Defaulting to 802.11b\n"); - case MODE_IEEE80211B: - rs_sta->expected_tpt = iwl3945_expected_tpt_b; + case IEEE80211_NUM_BANDS: + BUG(); break; } @@ -974,8 +967,7 @@ void iwl3945_rate_scale_init(struct ieee80211_hw *hw, s32 sta_id) IWL_DEBUG(IWL_DL_INFO | IWL_DL_RATE, "Network RSSI: %d\n", rssi); - rs_sta->start_rate = - iwl3945_get_rate_index_by_rssi(rssi, priv->phymode); + rs_sta->start_rate = iwl3945_get_rate_index_by_rssi(rssi, priv->band); IWL_DEBUG_RATE("leave: rssi %d assign rate index: " "%d (plcp 0x%x)\n", rssi, rs_sta->start_rate, diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c index 8d4d91d35fd2..50d927bb2170 100644 --- a/drivers/net/wireless/iwlwifi/iwl-3945.c +++ b/drivers/net/wireless/iwlwifi/iwl-3945.c @@ -247,7 +247,7 @@ static void iwl3945_add_radiotap(struct iwl3945_priv *priv, * the information provided in the skb from the hardware */ s8 signal = stats->ssi; s8 noise = 0; - int rate = stats->rate; + int rate = stats->rate_idx; u64 tsf = stats->mactime; __le16 phy_flags_hw = rx_hdr->phy_flags; @@ -315,7 +315,6 @@ static void iwl3945_add_radiotap(struct iwl3945_priv *priv, IEEE80211_CHAN_2GHZ), &iwl3945_rt->rt_chbitmask); - rate = iwl3945_rate_index_from_plcp(rate); if (rate == -1) iwl3945_rt->rt_rate = 0; else @@ -387,11 +386,10 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv, struct ieee80211_rx_status stats = { .mactime = le64_to_cpu(rx_end->timestamp), .freq = ieee80211chan2mhz(le16_to_cpu(rx_hdr->channel)), - .channel = le16_to_cpu(rx_hdr->channel), - .phymode = (rx_hdr->phy_flags & RX_RES_PHY_FLAGS_BAND_24_MSK) ? - MODE_IEEE80211G : MODE_IEEE80211A, + .band = (rx_hdr->phy_flags & RX_RES_PHY_FLAGS_BAND_24_MSK) ? + IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ, .antenna = 0, - .rate = rx_hdr->rate, + .rate_idx = iwl3945_rate_index_from_plcp(rx_hdr->rate), .flag = 0, }; u8 network_packet; @@ -450,8 +448,6 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv, stats.ssi, stats.noise, stats.signal, rx_stats_sig_avg, rx_stats_noise_diff); - stats.freq = ieee80211chan2mhz(stats.channel); - /* can be covered by iwl3945_report_frame() in most cases */ /* IWL_DEBUG_RX("RX status: 0x%08X\n", rx_end->status); */ @@ -464,8 +460,9 @@ static void iwl3945_rx_reply_rx(struct iwl3945_priv *priv, IWL_DEBUG_STATS ("[%c] %d RSSI: %d Signal: %u, Noise: %u, Rate: %u\n", network_packet ? '*' : ' ', - stats.channel, stats.ssi, stats.ssi, - stats.ssi, stats.rate); + le16_to_cpu(rx_hdr->channel), + stats.ssi, stats.ssi, + stats.ssi, stats.rate_idx); if (iwl3945_debug_level & (IWL_DL_RX)) /* Set "1" to report good data frames in groups of 100 */ @@ -689,7 +686,7 @@ void iwl3945_hw_build_tx_cmd_rate(struct iwl3945_priv *priv, struct ieee80211_hdr *hdr, int sta_id, int tx_id) { unsigned long flags; - u16 rate_index = min(ctrl->tx_rate & 0xffff, IWL_RATE_COUNT - 1); + u16 rate_index = min(ctrl->tx_rate->hw_value & 0xffff, IWL_RATE_COUNT - 1); u16 rate_mask; int rate; u8 rts_retry_limit; @@ -1552,14 +1549,14 @@ int iwl3945_hw_reg_send_txpower(struct iwl3945_priv *priv) .channel = priv->active_rxon.channel, }; - txpower.band = (priv->phymode == MODE_IEEE80211A) ? 0 : 1; + txpower.band = (priv->band == IEEE80211_BAND_5GHZ) ? 0 : 1; ch_info = iwl3945_get_channel_info(priv, - priv->phymode, + priv->band, le16_to_cpu(priv->active_rxon.channel)); if (!ch_info) { IWL_ERROR ("Failed to get channel info for channel %d [%d]\n", - le16_to_cpu(priv->active_rxon.channel), priv->phymode); + le16_to_cpu(priv->active_rxon.channel), priv->band); return -EINVAL; } @@ -2241,8 +2238,8 @@ int iwl3945_init_hw_rate_table(struct iwl3945_priv *priv) table[index].next_rate_index = iwl3945_rates[prev_index].table_rs_index; } - switch (priv->phymode) { - case MODE_IEEE80211A: + switch (priv->band) { + case IEEE80211_BAND_5GHZ: IWL_DEBUG_RATE("Select A mode rate scale\n"); /* If one of the following CCK rates is used, * have it fall back to the 6M OFDM rate */ @@ -2257,8 +2254,8 @@ int iwl3945_init_hw_rate_table(struct iwl3945_priv *priv) iwl3945_rates[IWL_FIRST_OFDM_RATE].table_rs_index; break; - case MODE_IEEE80211B: - IWL_DEBUG_RATE("Select B mode rate scale\n"); + case IEEE80211_BAND_2GHZ: + IWL_DEBUG_RATE("Select B/G mode rate scale\n"); /* If an OFDM rate is used, have it fall back to the * 1M CCK rates */ for (i = IWL_RATE_6M_INDEX_TABLE; i <= IWL_RATE_54M_INDEX_TABLE; i++) @@ -2269,7 +2266,7 @@ int iwl3945_init_hw_rate_table(struct iwl3945_priv *priv) break; default: - IWL_DEBUG_RATE("Select G mode rate scale\n"); + WARN_ON(1); break; } diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.h b/drivers/net/wireless/iwlwifi/iwl-3945.h index 1da14f9bbe0f..1beb5b676797 100644 --- a/drivers/net/wireless/iwlwifi/iwl-3945.h +++ b/drivers/net/wireless/iwlwifi/iwl-3945.h @@ -195,7 +195,7 @@ struct iwl3945_channel_info { u8 group_index; /* 0-4, maps channel to group1/2/3/4/5 */ u8 band_index; /* 0-4, maps channel to band1/2/3/4/5 */ - u8 phymode; /* MODE_IEEE80211{A,B,G} */ + enum ieee80211_band band; /* Radio/DSP gain settings for each "normal" data Tx rate. * These include, in addition to RF and DSP gain, a few fields for @@ -699,14 +699,14 @@ struct iwl3945_priv { struct list_head free_frames; int frames_count; - u8 phymode; + enum ieee80211_band band; int alloc_rxb_skb; bool add_radiotap; void (*rx_handlers[REPLY_MAX])(struct iwl3945_priv *priv, struct iwl3945_rx_mem_buffer *rxb); - const struct ieee80211_hw_mode *modes; + struct ieee80211_supported_band bands[IEEE80211_NUM_BANDS]; #ifdef CONFIG_IWL3945_SPECTRUM_MEASUREMENT /* spectrum measurement report caching */ @@ -937,13 +937,12 @@ static inline int is_channel_radar(const struct iwl3945_channel_info *ch_info) static inline u8 is_channel_a_band(const struct iwl3945_channel_info *ch_info) { - return ch_info->phymode == MODE_IEEE80211A; + return ch_info->band == IEEE80211_BAND_5GHZ; } static inline u8 is_channel_bg_band(const struct iwl3945_channel_info *ch_info) { - return ((ch_info->phymode == MODE_IEEE80211B) || - (ch_info->phymode == MODE_IEEE80211G)); + return ch_info->band == IEEE80211_BAND_2GHZ; } static inline int is_channel_passive(const struct iwl3945_channel_info *ch) @@ -967,7 +966,7 @@ static inline int iwl3945_rate_index_from_plcp(int plcp) } extern const struct iwl3945_channel_info *iwl3945_get_channel_info( - const struct iwl3945_priv *priv, int phymode, u16 channel); + const struct iwl3945_priv *priv, enum ieee80211_band band, u16 channel); /* Requires full declaration of iwl3945_priv before including */ #include "iwl-3945-io.h" diff --git a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c index 660671f17a3b..48a6a85355ec 100644 --- a/drivers/net/wireless/iwlwifi/iwl-4965-rs.c +++ b/drivers/net/wireless/iwlwifi/iwl-4965-rs.c @@ -139,7 +139,7 @@ struct iwl4965_lq_sta { u8 valid_antenna; u8 is_green; u8 is_dup; - u8 phymode; + enum ieee80211_band band; u8 ibss_sta_added; /* The following are bitmaps of rates; IWL_RATE_6M_MASK, etc. */ @@ -563,7 +563,8 @@ static void rs_mcs_from_tbl(struct iwl4965_rate *mcs_rate, * fill "search" or "active" tx mode table. */ static int rs_get_tbl_info_from_mcs(const struct iwl4965_rate *mcs_rate, - int phymode, struct iwl4965_scale_tbl_info *tbl, + enum ieee80211_band band, + struct iwl4965_scale_tbl_info *tbl, int *rate_idx) { int index; @@ -588,7 +589,7 @@ static int rs_get_tbl_info_from_mcs(const struct iwl4965_rate *mcs_rate, tbl->lq_type = LQ_NONE; else { - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) tbl->lq_type = LQ_A; else tbl->lq_type = LQ_G; @@ -766,7 +767,7 @@ static void rs_get_lower_rate(struct iwl4965_lq_sta *lq_sta, if (!is_legacy(tbl->lq_type) && (!ht_possible || !scale_index)) { switch_to_legacy = 1; scale_index = rs_ht_to_legacy[scale_index]; - if (lq_sta->phymode == MODE_IEEE80211A) + if (lq_sta->band == IEEE80211_BAND_5GHZ) tbl->lq_type = LQ_A; else tbl->lq_type = LQ_G; @@ -784,7 +785,7 @@ static void rs_get_lower_rate(struct iwl4965_lq_sta *lq_sta, /* Mask with station rate restriction */ if (is_legacy(tbl->lq_type)) { /* supp_rates has no CCK bits in A mode */ - if (lq_sta->phymode == (u8) MODE_IEEE80211A) + if (lq_sta->band == IEEE80211_BAND_5GHZ) rate_mask = (u16)(rate_mask & (lq_sta->supp_rates << IWL_FIRST_OFDM_RATE)); else @@ -883,9 +884,9 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev, search_win = (struct iwl4965_rate_scale_data *) &(search_tbl->win[0]); - tx_mcs.rate_n_flags = tx_resp->control.tx_rate; + tx_mcs.rate_n_flags = tx_resp->control.tx_rate->hw_value; - rs_get_tbl_info_from_mcs(&tx_mcs, priv->phymode, + rs_get_tbl_info_from_mcs(&tx_mcs, priv->band, &tbl_type, &rs_index); if ((rs_index < 0) || (rs_index >= IWL_RATE_COUNT)) { IWL_DEBUG_RATE("bad rate index at: %d rate 0x%X\n", @@ -918,7 +919,7 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev, * Each tx attempt steps one entry deeper in the rate table. */ tx_mcs.rate_n_flags = le32_to_cpu(table->rs_table[index].rate_n_flags); - rs_get_tbl_info_from_mcs(&tx_mcs, priv->phymode, + rs_get_tbl_info_from_mcs(&tx_mcs, priv->band, &tbl_type, &rs_index); /* If type matches "search" table, @@ -959,12 +960,12 @@ static void rs_tx_status(void *priv_rate, struct net_device *dev, * else look up the rate that was, finally, successful. */ if (!tx_resp->retry_count) - tx_mcs.rate_n_flags = tx_resp->control.tx_rate; + tx_mcs.rate_n_flags = tx_resp->control.tx_rate->hw_value; else tx_mcs.rate_n_flags = le32_to_cpu(table->rs_table[index].rate_n_flags); - rs_get_tbl_info_from_mcs(&tx_mcs, priv->phymode, + rs_get_tbl_info_from_mcs(&tx_mcs, priv->band, &tbl_type, &rs_index); /* Update frame history window with "success" if Tx got ACKed ... */ @@ -1801,7 +1802,7 @@ static void rs_rate_scale_perform(struct iwl4965_priv *priv, is_green = lq_sta->is_green; /* current tx rate */ - index = sta->last_txrate; + index = sta->last_txrate_idx; IWL_DEBUG_RATE("Rate scale index %d for type %d\n", index, tbl->lq_type); @@ -1814,7 +1815,7 @@ static void rs_rate_scale_perform(struct iwl4965_priv *priv, /* mask with station rate restriction */ if (is_legacy(tbl->lq_type)) { - if (lq_sta->phymode == (u8) MODE_IEEE80211A) + if (lq_sta->band == IEEE80211_BAND_5GHZ) /* supp_rates has no CCK bits in A mode */ rate_scale_index_msk = (u16) (rate_mask & (lq_sta->supp_rates << IWL_FIRST_OFDM_RATE)); @@ -2134,15 +2135,15 @@ static void rs_rate_scale_perform(struct iwl4965_priv *priv, out: rs_mcs_from_tbl(&tbl->current_rate, tbl, index, is_green); i = index; - sta->last_txrate = i; + sta->last_txrate_idx = i; - /* sta->txrate is an index to A mode rates which start + /* sta->txrate_idx is an index to A mode rates which start * at IWL_FIRST_OFDM_RATE */ - if (lq_sta->phymode == (u8) MODE_IEEE80211A) - sta->txrate = i - IWL_FIRST_OFDM_RATE; + if (lq_sta->band == IEEE80211_BAND_5GHZ) + sta->txrate_idx = i - IWL_FIRST_OFDM_RATE; else - sta->txrate = i; + sta->txrate_idx = i; return; } @@ -2164,7 +2165,7 @@ static void rs_initialize_lq(struct iwl4965_priv *priv, goto out; lq_sta = (struct iwl4965_lq_sta *)sta->rate_ctrl_priv; - i = sta->last_txrate; + i = sta->last_txrate_idx; if ((lq_sta->lq.sta_id == 0xff) && (priv->iw_mode == IEEE80211_IF_TYPE_IBSS)) @@ -2188,7 +2189,7 @@ static void rs_initialize_lq(struct iwl4965_priv *priv, mcs_rate.rate_n_flags |= RATE_MCS_CCK_MSK; tbl->antenna_type = ANT_AUX; - rs_get_tbl_info_from_mcs(&mcs_rate, priv->phymode, tbl, &rate_idx); + rs_get_tbl_info_from_mcs(&mcs_rate, priv->band, tbl, &rate_idx); if (!rs_is_ant_connected(priv->valid_antenna, tbl->antenna_type)) rs_toggle_antenna(&mcs_rate, tbl); @@ -2202,7 +2203,8 @@ static void rs_initialize_lq(struct iwl4965_priv *priv, } static void rs_get_rate(void *priv_rate, struct net_device *dev, - struct ieee80211_hw_mode *mode, struct sk_buff *skb, + struct ieee80211_supported_band *sband, + struct sk_buff *skb, struct rate_selection *sel) { @@ -2224,14 +2226,14 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev, fc = le16_to_cpu(hdr->frame_control); if (!ieee80211_is_data(fc) || is_multicast_ether_addr(hdr->addr1) || !sta || !sta->rate_ctrl_priv) { - sel->rate = rate_lowest(local, local->oper_hw_mode, sta); + sel->rate = rate_lowest(local, sband, sta); if (sta) sta_info_put(sta); return; } lq_sta = (struct iwl4965_lq_sta *)sta->rate_ctrl_priv; - i = sta->last_txrate; + i = sta->last_txrate_idx; if ((priv->iw_mode == IEEE80211_IF_TYPE_IBSS) && !lq_sta->ibss_sta_added) { @@ -2256,7 +2258,7 @@ static void rs_get_rate(void *priv_rate, struct net_device *dev, done: if ((i < 0) || (i > IWL_RATE_COUNT)) { - sel->rate = rate_lowest(local, local->oper_hw_mode, sta); + sel->rate = rate_lowest(local, sband, sta); return; } sta_info_put(sta); @@ -2291,13 +2293,15 @@ static void rs_rate_init(void *priv_rate, void *priv_sta, { int i, j; struct ieee80211_conf *conf = &local->hw.conf; - struct ieee80211_hw_mode *mode = local->oper_hw_mode; + struct ieee80211_supported_band *sband; struct iwl4965_priv *priv = (struct iwl4965_priv *)priv_rate; struct iwl4965_lq_sta *lq_sta = priv_sta; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + lq_sta->flush_timer = 0; - lq_sta->supp_rates = sta->supp_rates; - sta->txrate = 3; + lq_sta->supp_rates = sta->supp_rates[sband->band]; + sta->txrate_idx = 3; for (j = 0; j < LQ_SIZE; j++) for (i = 0; i < IWL_RATE_COUNT; i++) rs_rate_scale_clear_window(&(lq_sta->lq_info[j].win[i])); @@ -2332,15 +2336,15 @@ static void rs_rate_init(void *priv_rate, void *priv_sta, } /* Find highest tx rate supported by hardware and destination station */ - for (i = 0; i < mode->num_rates; i++) { - if ((sta->supp_rates & BIT(i)) && - (mode->rates[i].flags & IEEE80211_RATE_SUPPORTED)) - sta->txrate = i; - } - sta->last_txrate = sta->txrate; + for (i = 0; i < sband->n_bitrates; i++) + if (sta->supp_rates[sband->band] & BIT(i)) + sta->txrate_idx = i; + + sta->last_txrate_idx = sta->txrate_idx; + /* WTF is with this bogus comment? A doesn't have cck rates */ /* For MODE_IEEE80211A, cck rates are at end of rate table */ - if (local->hw.conf.phymode == MODE_IEEE80211A) - sta->last_txrate += IWL_FIRST_OFDM_RATE; + if (local->hw.conf.channel->band == IEEE80211_BAND_5GHZ) + sta->last_txrate_idx += IWL_FIRST_OFDM_RATE; lq_sta->is_dup = 0; lq_sta->valid_antenna = priv->valid_antenna; @@ -2349,7 +2353,7 @@ static void rs_rate_init(void *priv_rate, void *priv_sta, lq_sta->active_rate = priv->active_rate; lq_sta->active_rate &= ~(0x1000); lq_sta->active_rate_basic = priv->active_rate_basic; - lq_sta->phymode = priv->phymode; + lq_sta->band = priv->band; #ifdef CONFIG_IWL4965_HT /* * active_siso_rate mask includes 9 MBits (bit 5), and CCK (bits 0-3), @@ -2401,7 +2405,7 @@ static void rs_fill_link_cmd(struct iwl4965_lq_sta *lq_sta, rs_dbgfs_set_mcs(lq_sta, tx_mcs, index); /* Interpret rate_n_flags */ - rs_get_tbl_info_from_mcs(tx_mcs, lq_sta->phymode, + rs_get_tbl_info_from_mcs(tx_mcs, lq_sta->band, &tbl_type, &rate_idx); /* How many times should we repeat the initial rate? */ @@ -2455,7 +2459,7 @@ static void rs_fill_link_cmd(struct iwl4965_lq_sta *lq_sta, index++; } - rs_get_tbl_info_from_mcs(&new_rate, lq_sta->phymode, &tbl_type, + rs_get_tbl_info_from_mcs(&new_rate, lq_sta->band, &tbl_type, &rate_idx); /* Indicate to uCode which entries might be MIMO. @@ -2542,7 +2546,7 @@ static void rs_dbgfs_set_mcs(struct iwl4965_lq_sta *lq_sta, { u32 base_rate; - if (lq_sta->phymode == (u8) MODE_IEEE80211A) + if (lq_sta->band == IEEE80211_BAND_5GHZ) base_rate = 0x800D; else base_rate = 0x820A; @@ -2802,7 +2806,7 @@ int iwl4965_fill_rs_info(struct ieee80211_hw *hw, char *buf, u8 sta_id) cnt += sprintf(&buf[cnt], "\nrate scale type %d antenna %d " "active_search %d rate index %d\n", lq_type, antenna, - lq_sta->search_better_tbl, sta->last_txrate); + lq_sta->search_better_tbl, sta->last_txrate_idx); sta_info_put(sta); return cnt; diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.c b/drivers/net/wireless/iwlwifi/iwl-4965.c index 0bded8570275..a89439320498 100644 --- a/drivers/net/wireless/iwlwifi/iwl-4965.c +++ b/drivers/net/wireless/iwlwifi/iwl-4965.c @@ -339,14 +339,15 @@ static int iwl4965_kw_alloc(struct iwl4965_priv *priv) * * Does not set up a command, or touch hardware. */ -int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv, int phymode, u16 channel, +int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv, + enum ieee80211_band band, u16 channel, const struct iwl4965_eeprom_channel *eeprom_ch, u8 fat_extension_channel) { struct iwl4965_channel_info *ch_info; ch_info = (struct iwl4965_channel_info *) - iwl4965_get_channel_info(priv, phymode, channel); + iwl4965_get_channel_info(priv, band, channel); if (!is_channel_valid(ch_info)) return -1; @@ -1939,11 +1940,12 @@ static s32 iwl4965_get_voltage_compensation(s32 eeprom_voltage, } static const struct iwl4965_channel_info * -iwl4965_get_channel_txpower_info(struct iwl4965_priv *priv, u8 phymode, u16 channel) +iwl4965_get_channel_txpower_info(struct iwl4965_priv *priv, + enum ieee80211_band band, u16 channel) { const struct iwl4965_channel_info *ch_info; - ch_info = iwl4965_get_channel_info(priv, phymode, channel); + ch_info = iwl4965_get_channel_info(priv, band, channel); if (!is_channel_valid(ch_info)) return NULL; @@ -2392,7 +2394,7 @@ static int iwl4965_fill_txpower_tbl(struct iwl4965_priv *priv, u8 band, u16 chan /* Get current (RXON) channel, band, width */ ch_info = - iwl4965_get_channel_txpower_info(priv, priv->phymode, channel); + iwl4965_get_channel_txpower_info(priv, priv->band, channel); IWL_DEBUG_TXPOWER("chan %d band %d is_fat %d\n", channel, band, is_fat); @@ -2619,8 +2621,7 @@ int iwl4965_hw_reg_send_txpower(struct iwl4965_priv *priv) return -EAGAIN; } - band = ((priv->phymode == MODE_IEEE80211B) || - (priv->phymode == MODE_IEEE80211G)); + band = priv->band == IEEE80211_BAND_2GHZ; is_fat = is_fat_channel(priv->active_rxon.flags); @@ -2650,10 +2651,9 @@ int iwl4965_hw_channel_switch(struct iwl4965_priv *priv, u16 channel) struct iwl4965_channel_switch_cmd cmd = { 0 }; const struct iwl4965_channel_info *ch_info; - band = ((priv->phymode == MODE_IEEE80211B) || - (priv->phymode == MODE_IEEE80211G)); + band = priv->band == IEEE80211_BAND_2GHZ; - ch_info = iwl4965_get_channel_info(priv, priv->phymode, channel); + ch_info = iwl4965_get_channel_info(priv, priv->band, channel); is_fat = is_fat_channel(priv->staging_rxon.flags); @@ -2698,7 +2698,7 @@ void iwl4965_hw_build_tx_cmd_rate(struct iwl4965_priv *priv, u16 fc = le16_to_cpu(hdr->frame_control); u8 rate_plcp; u16 rate_flags = 0; - int rate_idx = min(ctrl->tx_rate & 0xffff, IWL_RATE_COUNT - 1); + int rate_idx = min(ctrl->tx_rate->hw_value & 0xffff, IWL_RATE_COUNT - 1); rate_plcp = iwl4965_rates[rate_idx].plcp; @@ -3178,7 +3178,7 @@ static void iwl4965_add_radiotap(struct iwl4965_priv *priv, { s8 signal = stats->ssi; s8 noise = 0; - int rate = stats->rate; + int rate = stats->rate_idx; u64 tsf = stats->mactime; __le16 phy_flags_hw = rx_start->phy_flags; struct iwl4965_rt_rx_hdr { @@ -3246,7 +3246,6 @@ static void iwl4965_add_radiotap(struct iwl4965_priv *priv, IEEE80211_CHAN_2GHZ), &iwl4965_rt->rt_chbitmask); - rate = iwl4965_rate_index_from_plcp(rate); if (rate == -1) iwl4965_rt->rt_rate = 0; else @@ -3542,12 +3541,13 @@ static void iwl4965_rx_reply_rx(struct iwl4965_priv *priv, u16 fc; struct ieee80211_rx_status stats = { .mactime = le64_to_cpu(rx_start->timestamp), - .channel = le16_to_cpu(rx_start->channel), - .phymode = + .freq = ieee80211chan2mhz(le16_to_cpu(rx_start->channel)), + .band = (rx_start->phy_flags & RX_RES_PHY_FLAGS_BAND_24_MSK) ? - MODE_IEEE80211G : MODE_IEEE80211A, + IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ, .antenna = 0, - .rate = iwl4965_hw_get_rate(rx_start->rate_n_flags), + .rate_idx = iwl4965_hw_get_rate( + le32_to_cpu(rx_start->rate_n_flags)), .flag = 0, }; u8 network_packet; @@ -3598,8 +3598,6 @@ static void iwl4965_rx_reply_rx(struct iwl4965_priv *priv, priv->ucode_beacon_time = le32_to_cpu(rx_start->beacon_time_stamp); - stats.freq = ieee80211chan2mhz(stats.channel); - /* Find max signal strength (dBm) among 3 antenna/receiver chains */ stats.ssi = iwl4965_calc_rssi(rx_start); @@ -4185,7 +4183,7 @@ void iwl4965_add_station(struct iwl4965_priv *priv, const u8 *addr, int is_ap) * all the way down to 1M in IEEE order, and then spin on 1M */ if (is_ap) r = IWL_RATE_54M_INDEX; - else if (priv->phymode == MODE_IEEE80211A) + else if (priv->band == IEEE80211_BAND_5GHZ) r = IWL_RATE_6M_INDEX; else r = IWL_RATE_1M_INDEX; @@ -4218,12 +4216,13 @@ void iwl4965_add_station(struct iwl4965_priv *priv, const u8 *addr, int is_ap) #ifdef CONFIG_IWL4965_HT -static u8 iwl4965_is_channel_extension(struct iwl4965_priv *priv, int phymode, +static u8 iwl4965_is_channel_extension(struct iwl4965_priv *priv, + enum ieee80211_band band, u16 channel, u8 extension_chan_offset) { const struct iwl4965_channel_info *ch_info; - ch_info = iwl4965_get_channel_info(priv, phymode, channel); + ch_info = iwl4965_get_channel_info(priv, band, channel); if (!is_channel_valid(ch_info)) return 0; diff --git a/drivers/net/wireless/iwlwifi/iwl-4965.h b/drivers/net/wireless/iwlwifi/iwl-4965.h index de5c1bf8fc42..cb8f7f2a8d48 100644 --- a/drivers/net/wireless/iwlwifi/iwl-4965.h +++ b/drivers/net/wireless/iwlwifi/iwl-4965.h @@ -206,7 +206,7 @@ struct iwl4965_channel_info { u8 group_index; /* 0-4, maps channel to group1/2/3/4/5 */ u8 band_index; /* 0-4, maps channel to band1/2/3/4/5 */ - u8 phymode; /* MODE_IEEE80211{A,B,G} */ + enum ieee80211_band band; /* Radio/DSP gain settings for each "normal" data Tx rate. * These include, in addition to RF and DSP gain, a few fields for @@ -764,7 +764,8 @@ extern void iwl4965_update_rate_scaling(struct iwl4965_priv *priv, u8 mode); extern void iwl4965_chain_noise_reset(struct iwl4965_priv *priv); extern void iwl4965_init_sensitivity(struct iwl4965_priv *priv, u8 flags, u8 force); -extern int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv, int phymode, +extern int iwl4965_set_fat_chan_info(struct iwl4965_priv *priv, + enum ieee80211_band band, u16 channel, const struct iwl4965_eeprom_channel *eeprom_ch, u8 fat_extension_channel); @@ -977,14 +978,14 @@ struct iwl4965_priv { struct list_head free_frames; int frames_count; - u8 phymode; + enum ieee80211_band band; int alloc_rxb_skb; bool add_radiotap; void (*rx_handlers[REPLY_MAX])(struct iwl4965_priv *priv, struct iwl4965_rx_mem_buffer *rxb); - const struct ieee80211_hw_mode *modes; + struct ieee80211_supported_band bands[IEEE80211_NUM_BANDS]; #ifdef CONFIG_IWL4965_SPECTRUM_MEASUREMENT /* spectrum measurement report caching */ @@ -1243,13 +1244,12 @@ static inline int is_channel_radar(const struct iwl4965_channel_info *ch_info) static inline u8 is_channel_a_band(const struct iwl4965_channel_info *ch_info) { - return ch_info->phymode == MODE_IEEE80211A; + return ch_info->band == IEEE80211_BAND_5GHZ; } static inline u8 is_channel_bg_band(const struct iwl4965_channel_info *ch_info) { - return ((ch_info->phymode == MODE_IEEE80211B) || - (ch_info->phymode == MODE_IEEE80211G)); + return ch_info->band == IEEE80211_BAND_2GHZ; } static inline int is_channel_passive(const struct iwl4965_channel_info *ch) @@ -1263,7 +1263,7 @@ static inline int is_channel_ibss(const struct iwl4965_channel_info *ch) } extern const struct iwl4965_channel_info *iwl4965_get_channel_info( - const struct iwl4965_priv *priv, int phymode, u16 channel); + const struct iwl4965_priv *priv, enum ieee80211_band band, u16 channel); /* Requires full declaration of iwl4965_priv before including */ #include "iwl-4965-io.h" diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c index 40b71bc2c4a4..57a1d70f2abf 100644 --- a/drivers/net/wireless/iwlwifi/iwl3945-base.c +++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c @@ -116,16 +116,10 @@ static __le16 *ieee80211_get_qos_ctrl(struct ieee80211_hdr *hdr) return NULL; } -static const struct ieee80211_hw_mode *iwl3945_get_hw_mode( - struct iwl3945_priv *priv, int mode) +static const struct ieee80211_supported_band *iwl3945_get_band( + struct iwl3945_priv *priv, enum ieee80211_band band) { - int i; - - for (i = 0; i < 3; i++) - if (priv->modes[i].mode == mode) - return &priv->modes[i]; - - return NULL; + return priv->hw->wiphy->bands[band]; } static int iwl3945_is_empty_essid(const char *essid, int essid_len) @@ -547,7 +541,7 @@ u8 iwl3945_add_station(struct iwl3945_priv *priv, const u8 *addr, int is_ap, u8 station->sta.sta.sta_id = index; station->sta.station_flags = 0; - if (priv->phymode == MODE_IEEE80211A) + if (priv->band == IEEE80211_BAND_5GHZ) rate = IWL_RATE_6M_PLCP; else rate = IWL_RATE_1M_PLCP; @@ -894,35 +888,37 @@ int iwl3945_send_statistics_request(struct iwl3945_priv *priv) /** * iwl3945_set_rxon_channel - Set the phymode and channel values in staging RXON - * @phymode: MODE_IEEE80211A sets to 5.2GHz; all else set to 2.4GHz - * @channel: Any channel valid for the requested phymode + * @band: 2.4 or 5 GHz band + * @channel: Any channel valid for the requested band - * In addition to setting the staging RXON, priv->phymode is also set. + * In addition to setting the staging RXON, priv->band is also set. * * NOTE: Does not commit to the hardware; it sets appropriate bit fields - * in the staging RXON flag structure based on the phymode + * in the staging RXON flag structure based on the band */ -static int iwl3945_set_rxon_channel(struct iwl3945_priv *priv, u8 phymode, u16 channel) +static int iwl3945_set_rxon_channel(struct iwl3945_priv *priv, + enum ieee80211_band band, + u16 channel) { - if (!iwl3945_get_channel_info(priv, phymode, channel)) { + if (!iwl3945_get_channel_info(priv, band, channel)) { IWL_DEBUG_INFO("Could not set channel to %d [%d]\n", - channel, phymode); + channel, band); return -EINVAL; } if ((le16_to_cpu(priv->staging_rxon.channel) == channel) && - (priv->phymode == phymode)) + (priv->band == band)) return 0; priv->staging_rxon.channel = cpu_to_le16(channel); - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) priv->staging_rxon.flags &= ~RXON_FLG_BAND_24G_MSK; else priv->staging_rxon.flags |= RXON_FLG_BAND_24G_MSK; - priv->phymode = phymode; + priv->band = band; - IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, phymode); + IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, band); return 0; } @@ -1210,8 +1206,7 @@ static int iwl3945_commit_rxon(struct iwl3945_priv *priv) return -EIO; } - /* Init the hardware's rate fallback order based on the - * phymode */ + /* Init the hardware's rate fallback order based on the band */ rc = iwl3945_init_hw_rate_table(priv); if (rc) { IWL_ERROR("Error setting HW rate table: %02X\n", rc); @@ -2461,9 +2456,10 @@ static int iwl3945_set_rxon_hwcrypto(struct iwl3945_priv *priv, int hw_decrypt) return 0; } -static void iwl3945_set_flags_for_phymode(struct iwl3945_priv *priv, u8 phymode) +static void iwl3945_set_flags_for_phymode(struct iwl3945_priv *priv, + enum ieee80211_band band) { - if (phymode == MODE_IEEE80211A) { + if (band == IEEE80211_BAND_5GHZ) { priv->staging_rxon.flags &= ~(RXON_FLG_BAND_24G_MSK | RXON_FLG_AUTO_DETECT_MSK | RXON_FLG_CCK_MSK); @@ -2526,7 +2522,7 @@ static void iwl3945_connection_init_rx_config(struct iwl3945_priv *priv) priv->staging_rxon.flags |= RXON_FLG_SHORT_PREAMBLE_MSK; #endif - ch_info = iwl3945_get_channel_info(priv, priv->phymode, + ch_info = iwl3945_get_channel_info(priv, priv->band, le16_to_cpu(priv->staging_rxon.channel)); if (!ch_info) @@ -2542,11 +2538,11 @@ static void iwl3945_connection_init_rx_config(struct iwl3945_priv *priv) priv->staging_rxon.channel = cpu_to_le16(ch_info->channel); if (is_channel_a_band(ch_info)) - priv->phymode = MODE_IEEE80211A; + priv->band = IEEE80211_BAND_5GHZ; else - priv->phymode = MODE_IEEE80211G; + priv->band = IEEE80211_BAND_2GHZ; - iwl3945_set_flags_for_phymode(priv, priv->phymode); + iwl3945_set_flags_for_phymode(priv, priv->band); priv->staging_rxon.ofdm_basic_rates = (IWL_OFDM_RATES_MASK >> IWL_FIRST_OFDM_RATE) & 0xFF; @@ -2560,7 +2556,7 @@ static int iwl3945_set_mode(struct iwl3945_priv *priv, int mode) const struct iwl3945_channel_info *ch_info; ch_info = iwl3945_get_channel_info(priv, - priv->phymode, + priv->band, le16_to_cpu(priv->staging_rxon.channel)); if (!ch_info || !is_channel_ibss(ch_info)) { @@ -2792,7 +2788,7 @@ static int iwl3945_tx_skb(struct iwl3945_priv *priv, goto drop_unlock; } - if ((ctl->tx_rate & 0xFF) == IWL_INVALID_RATE) { + if ((ctl->tx_rate->hw_value & 0xFF) == IWL_INVALID_RATE) { IWL_ERROR("ERROR: No TX rate available.\n"); goto drop_unlock; } @@ -2992,12 +2988,12 @@ drop: static void iwl3945_set_rate(struct iwl3945_priv *priv) { - const struct ieee80211_hw_mode *hw = NULL; + const struct ieee80211_supported_band *sband = NULL; struct ieee80211_rate *rate; int i; - hw = iwl3945_get_hw_mode(priv, priv->phymode); - if (!hw) { + sband = iwl3945_get_band(priv, priv->band); + if (!sband) { IWL_ERROR("Failed to set rate: unable to get hw mode\n"); return; } @@ -3005,24 +3001,17 @@ static void iwl3945_set_rate(struct iwl3945_priv *priv) priv->active_rate = 0; priv->active_rate_basic = 0; - IWL_DEBUG_RATE("Setting rates for 802.11%c\n", - hw->mode == MODE_IEEE80211A ? - 'a' : ((hw->mode == MODE_IEEE80211B) ? 'b' : 'g')); - - for (i = 0; i < hw->num_rates; i++) { - rate = &(hw->rates[i]); - if ((rate->val < IWL_RATE_COUNT) && - (rate->flags & IEEE80211_RATE_SUPPORTED)) { - IWL_DEBUG_RATE("Adding rate index %d (plcp %d)%s\n", - rate->val, iwl3945_rates[rate->val].plcp, - (rate->flags & IEEE80211_RATE_BASIC) ? - "*" : ""); - priv->active_rate |= (1 << rate->val); - if (rate->flags & IEEE80211_RATE_BASIC) - priv->active_rate_basic |= (1 << rate->val); - } else - IWL_DEBUG_RATE("Not adding rate %d (plcp %d)\n", - rate->val, iwl3945_rates[rate->val].plcp); + IWL_DEBUG_RATE("Setting rates for %s GHz\n", + sband->band == IEEE80211_BAND_2GHZ ? "2.4" : "5"); + + for (i = 0; i < sband->n_bitrates; i++) { + rate = &sband->bitrates[i]; + if ((rate->hw_value < IWL_RATE_COUNT) && + !(rate->flags & IEEE80211_CHAN_DISABLED)) { + IWL_DEBUG_RATE("Adding rate index %d (plcp %d)\n", + rate->hw_value, iwl3945_rates[rate->hw_value].plcp); + priv->active_rate |= (1 << rate->hw_value); + } } IWL_DEBUG_RATE("Set active_rate = %0x, active_rate_basic = %0x\n", @@ -3436,8 +3425,6 @@ static void iwl3945_rx_reply_tx(struct iwl3945_priv *priv, tx_status->flags = iwl3945_is_tx_success(status) ? IEEE80211_TX_STATUS_ACK : 0; - tx_status->control.tx_rate = iwl3945_rate_index_from_plcp(tx_resp->rate); - IWL_DEBUG_TX("Tx queue %d Status %s (0x%08x) plcp rate %d retries %d\n", txq_id, iwl3945_get_tx_fail_reason(status), status, tx_resp->rate, tx_resp->failure_frame); @@ -5026,24 +5013,24 @@ static void iwl3945_init_band_reference(const struct iwl3945_priv *priv, int ban * Based on band and channel number. */ const struct iwl3945_channel_info *iwl3945_get_channel_info(const struct iwl3945_priv *priv, - int phymode, u16 channel) + enum ieee80211_band band, u16 channel) { int i; - switch (phymode) { - case MODE_IEEE80211A: + switch (band) { + case IEEE80211_BAND_5GHZ: for (i = 14; i < priv->channel_count; i++) { if (priv->channel_info[i].channel == channel) return &priv->channel_info[i]; } break; - case MODE_IEEE80211B: - case MODE_IEEE80211G: + case IEEE80211_BAND_2GHZ: if (channel >= 1 && channel <= 14) return &priv->channel_info[channel - 1]; break; - + case IEEE80211_NUM_BANDS: + WARN_ON(1); } return NULL; @@ -5106,8 +5093,8 @@ static int iwl3945_init_channel_map(struct iwl3945_priv *priv) /* Loop through each band adding each of the channels */ for (ch = 0; ch < eeprom_ch_count; ch++) { ch_info->channel = eeprom_ch_index[ch]; - ch_info->phymode = (band == 1) ? MODE_IEEE80211B : - MODE_IEEE80211A; + ch_info->band = (band == 1) ? IEEE80211_BAND_2GHZ : + IEEE80211_BAND_5GHZ; /* permanently store EEPROM's channel regulatory flags * and max power in channel info database. */ @@ -5203,18 +5190,20 @@ static void iwl3945_free_channel_map(struct iwl3945_priv *priv) #define IWL_PASSIVE_DWELL_BASE (100) #define IWL_CHANNEL_TUNE_TIME 5 -static inline u16 iwl3945_get_active_dwell_time(struct iwl3945_priv *priv, int phymode) +static inline u16 iwl3945_get_active_dwell_time(struct iwl3945_priv *priv, + enum ieee80211_band band) { - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) return IWL_ACTIVE_DWELL_TIME_52; else return IWL_ACTIVE_DWELL_TIME_24; } -static u16 iwl3945_get_passive_dwell_time(struct iwl3945_priv *priv, int phymode) +static u16 iwl3945_get_passive_dwell_time(struct iwl3945_priv *priv, + enum ieee80211_band band) { - u16 active = iwl3945_get_active_dwell_time(priv, phymode); - u16 passive = (phymode != MODE_IEEE80211A) ? + u16 active = iwl3945_get_active_dwell_time(priv, band); + u16 passive = (band == IEEE80211_BAND_2GHZ) ? IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_24 : IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_52; @@ -5234,28 +5223,29 @@ static u16 iwl3945_get_passive_dwell_time(struct iwl3945_priv *priv, int phymode return passive; } -static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode, +static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, + enum ieee80211_band band, u8 is_active, u8 direct_mask, struct iwl3945_scan_channel *scan_ch) { const struct ieee80211_channel *channels = NULL; - const struct ieee80211_hw_mode *hw_mode; + const struct ieee80211_supported_band *sband; const struct iwl3945_channel_info *ch_info; u16 passive_dwell = 0; u16 active_dwell = 0; int added, i; - hw_mode = iwl3945_get_hw_mode(priv, phymode); - if (!hw_mode) + sband = iwl3945_get_band(priv, band); + if (!sband) return 0; - channels = hw_mode->channels; + channels = sband->channels; - active_dwell = iwl3945_get_active_dwell_time(priv, phymode); - passive_dwell = iwl3945_get_passive_dwell_time(priv, phymode); + active_dwell = iwl3945_get_active_dwell_time(priv, band); + passive_dwell = iwl3945_get_passive_dwell_time(priv, band); - for (i = 0, added = 0; i < hw_mode->num_channels; i++) { - if (channels[i].chan == + for (i = 0, added = 0; i < sband->n_channels; i++) { + if (channels[i].hw_value == le16_to_cpu(priv->active_rxon.channel)) { if (iwl3945_is_associated(priv)) { IWL_DEBUG_SCAN @@ -5266,9 +5256,9 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode, } else if (priv->only_active_channel) continue; - scan_ch->channel = channels[i].chan; + scan_ch->channel = channels[i].hw_value; - ch_info = iwl3945_get_channel_info(priv, phymode, scan_ch->channel); + ch_info = iwl3945_get_channel_info(priv, band, scan_ch->channel); if (!is_channel_valid(ch_info)) { IWL_DEBUG_SCAN("Channel %d is INVALID for this SKU.\n", scan_ch->channel); @@ -5276,7 +5266,7 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode, } if (!is_active || is_channel_passive(ch_info) || - !(channels[i].flag & IEEE80211_CHAN_W_ACTIVE_SCAN)) + (channels[i].flags & IEEE80211_CHAN_PASSIVE_SCAN)) scan_ch->type = 0; /* passive */ else scan_ch->type = 1; /* active */ @@ -5295,7 +5285,7 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode, /* scan_pwr_info->tpc.dsp_atten; */ /*scan_pwr_info->tpc.tx_gain; */ - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) scan_ch->tpc.tx_gain = ((1 << 5) | (3 << 3)) | 3; else { scan_ch->tpc.tx_gain = ((1 << 5) | (5 << 3)); @@ -5319,41 +5309,23 @@ static int iwl3945_get_channels_for_scan(struct iwl3945_priv *priv, int phymode, return added; } -static void iwl3945_reset_channel_flag(struct iwl3945_priv *priv) -{ - int i, j; - for (i = 0; i < 3; i++) { - struct ieee80211_hw_mode *hw_mode = (void *)&priv->modes[i]; - for (j = 0; j < hw_mode->num_channels; j++) - hw_mode->channels[j].flag = hw_mode->channels[j].val; - } -} - static void iwl3945_init_hw_rates(struct iwl3945_priv *priv, struct ieee80211_rate *rates) { int i; for (i = 0; i < IWL_RATE_COUNT; i++) { - rates[i].rate = iwl3945_rates[i].ieee * 5; - rates[i].val = i; /* Rate scaling will work on indexes */ - rates[i].val2 = i; - rates[i].flags = IEEE80211_RATE_SUPPORTED; - /* Only OFDM have the bits-per-symbol set */ - if ((i <= IWL_LAST_OFDM_RATE) && (i >= IWL_FIRST_OFDM_RATE)) - rates[i].flags |= IEEE80211_RATE_OFDM; - else { + rates[i].bitrate = iwl3945_rates[i].ieee * 5; + rates[i].hw_value = i; /* Rate scaling will work on indexes */ + rates[i].hw_value_short = i; + rates[i].flags = 0; + if ((i > IWL_LAST_OFDM_RATE) || (i < IWL_FIRST_OFDM_RATE)) { /* - * If CCK 1M then set rate flag to CCK else CCK_2 - * which is CCK | PREAMBLE2 + * If CCK != 1M then set short preamble rate flag. */ rates[i].flags |= (iwl3945_rates[i].plcp == 10) ? - IEEE80211_RATE_CCK : IEEE80211_RATE_CCK_2; + 0 : IEEE80211_RATE_SHORT_PREAMBLE; } - - /* Set up which ones are basic rates... */ - if (IWL_BASIC_RATES_MASK & (1 << i)) - rates[i].flags |= IEEE80211_RATE_BASIC; } } @@ -5363,67 +5335,41 @@ static void iwl3945_init_hw_rates(struct iwl3945_priv *priv, static int iwl3945_init_geos(struct iwl3945_priv *priv) { struct iwl3945_channel_info *ch; - struct ieee80211_hw_mode *modes; + struct ieee80211_supported_band *band; struct ieee80211_channel *channels; struct ieee80211_channel *geo_ch; struct ieee80211_rate *rates; int i = 0; - enum { - A = 0, - B = 1, - G = 2, - }; - int mode_count = 3; - if (priv->modes) { + if (priv->bands[IEEE80211_BAND_2GHZ].n_bitrates || + priv->bands[IEEE80211_BAND_5GHZ].n_bitrates) { IWL_DEBUG_INFO("Geography modes already initialized.\n"); set_bit(STATUS_GEO_CONFIGURED, &priv->status); return 0; } - modes = kzalloc(sizeof(struct ieee80211_hw_mode) * mode_count, - GFP_KERNEL); - if (!modes) - return -ENOMEM; - channels = kzalloc(sizeof(struct ieee80211_channel) * priv->channel_count, GFP_KERNEL); - if (!channels) { - kfree(modes); + if (!channels) return -ENOMEM; - } rates = kzalloc((sizeof(struct ieee80211_rate) * (IWL_MAX_RATES + 1)), GFP_KERNEL); if (!rates) { - kfree(modes); kfree(channels); return -ENOMEM; } - /* 0 = 802.11a - * 1 = 802.11b - * 2 = 802.11g - */ - /* 5.2GHz channels start after the 2.4GHz channels */ - modes[A].mode = MODE_IEEE80211A; - modes[A].channels = &channels[ARRAY_SIZE(iwl3945_eeprom_band_1)]; - modes[A].rates = &rates[4]; - modes[A].num_rates = 8; /* just OFDM */ - modes[A].num_channels = 0; - - modes[B].mode = MODE_IEEE80211B; - modes[B].channels = channels; - modes[B].rates = rates; - modes[B].num_rates = 4; /* just CCK */ - modes[B].num_channels = 0; - - modes[G].mode = MODE_IEEE80211G; - modes[G].channels = channels; - modes[G].rates = rates; - modes[G].num_rates = 12; /* OFDM & CCK */ - modes[G].num_channels = 0; + band = &priv->bands[IEEE80211_BAND_5GHZ]; + band->channels = &channels[ARRAY_SIZE(iwl3945_eeprom_band_1)]; + band->bitrates = &rates[4]; + band->n_bitrates = 8; /* just OFDM */ + + band = &priv->bands[IEEE80211_BAND_2GHZ]; + band->channels = channels; + band->bitrates = rates; + band->n_bitrates = 12; /* OFDM & CCK */ priv->ieee_channels = channels; priv->ieee_rates = rates; @@ -5442,37 +5388,32 @@ static int iwl3945_init_geos(struct iwl3945_priv *priv) } if (is_channel_a_band(ch)) - geo_ch = &modes[A].channels[modes[A].num_channels++]; - else { - geo_ch = &modes[B].channels[modes[B].num_channels++]; - modes[G].num_channels++; - } + geo_ch = &priv->bands[IEEE80211_BAND_5GHZ].channels[priv->bands[IEEE80211_BAND_5GHZ].n_channels++]; + else + geo_ch = &priv->bands[IEEE80211_BAND_2GHZ].channels[priv->bands[IEEE80211_BAND_2GHZ].n_channels++]; - geo_ch->freq = ieee80211chan2mhz(ch->channel); - geo_ch->chan = ch->channel; - geo_ch->power_level = ch->max_power_avg; - geo_ch->antenna_max = 0xff; + geo_ch->center_freq = ieee80211chan2mhz(ch->channel); + geo_ch->max_power = ch->max_power_avg; + geo_ch->max_antenna_gain = 0xff; if (is_channel_valid(ch)) { - geo_ch->flag = IEEE80211_CHAN_W_SCAN; - if (ch->flags & EEPROM_CHANNEL_IBSS) - geo_ch->flag |= IEEE80211_CHAN_W_IBSS; + if (!(ch->flags & EEPROM_CHANNEL_IBSS)) + geo_ch->flags |= IEEE80211_CHAN_NO_IBSS; - if (ch->flags & EEPROM_CHANNEL_ACTIVE) - geo_ch->flag |= IEEE80211_CHAN_W_ACTIVE_SCAN; + if (!(ch->flags & EEPROM_CHANNEL_ACTIVE)) + geo_ch->flags |= IEEE80211_CHAN_PASSIVE_SCAN; if (ch->flags & EEPROM_CHANNEL_RADAR) - geo_ch->flag |= IEEE80211_CHAN_W_RADAR_DETECT; + geo_ch->flags |= IEEE80211_CHAN_RADAR; if (ch->max_power_avg > priv->max_channel_txpower_limit) priv->max_channel_txpower_limit = ch->max_power_avg; - } - - geo_ch->val = geo_ch->flag; + } else + geo_ch->flags |= IEEE80211_CHAN_DISABLED; } - if ((modes[A].num_channels == 0) && priv->is_abg) { + if ((priv->bands[IEEE80211_BAND_5GHZ].n_channels == 0) && priv->is_abg) { printk(KERN_INFO DRV_NAME ": Incorrectly detected BG card as ABG. Please send " "your PCI ID 0x%04X:0x%04X to maintainer.\n", @@ -5482,24 +5423,12 @@ static int iwl3945_init_geos(struct iwl3945_priv *priv) printk(KERN_INFO DRV_NAME ": Tunable channels: %d 802.11bg, %d 802.11a channels\n", - modes[G].num_channels, modes[A].num_channels); + priv->bands[IEEE80211_BAND_2GHZ].n_channels, + priv->bands[IEEE80211_BAND_5GHZ].n_channels); - /* - * NOTE: We register these in preference of order -- the - * stack doesn't currently (as of 7.0.6 / Apr 24 '07) pick - * a phymode based on rates or AP capabilities but seems to - * configure it purely on if the channel being configured - * is supported by a mode -- and the first match is taken - */ - - if (modes[G].num_channels) - ieee80211_register_hwmode(priv->hw, &modes[G]); - if (modes[B].num_channels) - ieee80211_register_hwmode(priv->hw, &modes[B]); - if (modes[A].num_channels) - ieee80211_register_hwmode(priv->hw, &modes[A]); + priv->hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->bands[IEEE80211_BAND_2GHZ]; + priv->hw->wiphy->bands[IEEE80211_BAND_5GHZ] = &priv->bands[IEEE80211_BAND_5GHZ]; - priv->modes = modes; set_bit(STATUS_GEO_CONFIGURED, &priv->status); return 0; @@ -5510,7 +5439,6 @@ static int iwl3945_init_geos(struct iwl3945_priv *priv) */ static void iwl3945_free_geos(struct iwl3945_priv *priv) { - kfree(priv->modes); kfree(priv->ieee_channels); kfree(priv->ieee_rates); clear_bit(STATUS_GEO_CONFIGURED, &priv->status); @@ -6519,7 +6447,7 @@ static void iwl3945_bg_request_scan(struct work_struct *data) struct iwl3945_scan_cmd *scan; struct ieee80211_conf *conf = NULL; u8 direct_mask; - int phymode; + enum ieee80211_band band; conf = ieee80211_get_hw_conf(priv->hw); @@ -6651,13 +6579,13 @@ static void iwl3945_bg_request_scan(struct work_struct *data) scan->flags = RXON_FLG_BAND_24G_MSK | RXON_FLG_AUTO_DETECT_MSK; scan->tx_cmd.rate = IWL_RATE_1M_PLCP; scan->good_CRC_th = 0; - phymode = MODE_IEEE80211G; + band = IEEE80211_BAND_2GHZ; break; case 1: scan->tx_cmd.rate = IWL_RATE_6M_PLCP; scan->good_CRC_th = IWL_GOOD_CRC_TH; - phymode = MODE_IEEE80211A; + band = IEEE80211_BAND_5GHZ; break; default: @@ -6680,7 +6608,7 @@ static void iwl3945_bg_request_scan(struct work_struct *data) scan->channel_count = iwl3945_get_channels_for_scan( - priv, phymode, 1, /* active */ + priv, band, 1, /* active */ direct_mask, (void *)&scan->data[le16_to_cpu(scan->tx_cmd.len)]); @@ -6825,7 +6753,7 @@ static void iwl3945_bg_post_associate(struct work_struct *data) iwl3945_add_station(priv, iwl3945_broadcast_addr, 0, 0); iwl3945_add_station(priv, priv->bssid, 0, 0); iwl3945_sync_sta(priv, IWL_STA_ID, - (priv->phymode == MODE_IEEE80211A)? + (priv->band == IEEE80211_BAND_5GHZ) ? IWL_RATE_6M_PLCP : IWL_RATE_1M_PLCP, CMD_ASYNC); iwl3945_rate_scale_init(priv->hw, IWL_STA_ID); @@ -7020,7 +6948,7 @@ static int iwl3945_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb, } IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len, - ctl->tx_rate); + ctl->tx_rate->bitrate); if (iwl3945_tx_skb(priv, skb, ctl)) dev_kfree_skb_any(skb); @@ -7079,7 +7007,7 @@ static int iwl3945_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co int ret = 0; mutex_lock(&priv->mutex); - IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel); + IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel->hw_value); priv->add_radiotap = !!(conf->flags & IEEE80211_CONF_RADIOTAP); @@ -7099,19 +7027,20 @@ static int iwl3945_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co spin_lock_irqsave(&priv->lock, flags); - ch_info = iwl3945_get_channel_info(priv, conf->phymode, conf->channel); + ch_info = iwl3945_get_channel_info(priv, conf->channel->band, + conf->channel->hw_value); if (!is_channel_valid(ch_info)) { IWL_DEBUG_SCAN("Channel %d [%d] is INVALID for this SKU.\n", - conf->channel, conf->phymode); + conf->channel->hw_value, conf->channel->band); IWL_DEBUG_MAC80211("leave - invalid channel\n"); spin_unlock_irqrestore(&priv->lock, flags); ret = -EINVAL; goto out; } - iwl3945_set_rxon_channel(priv, conf->phymode, conf->channel); + iwl3945_set_rxon_channel(priv, conf->channel->band, conf->channel->hw_value); - iwl3945_set_flags_for_phymode(priv, conf->phymode); + iwl3945_set_flags_for_phymode(priv, conf->channel->band); /* The list of supported rates and rate mask can be different * for each phymode; since the phymode may have changed, reset @@ -7892,65 +7821,6 @@ static ssize_t store_filter_flags(struct device *d, static DEVICE_ATTR(filter_flags, S_IWUSR | S_IRUGO, show_filter_flags, store_filter_flags); -static ssize_t show_tune(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct iwl3945_priv *priv = (struct iwl3945_priv *)d->driver_data; - - return sprintf(buf, "0x%04X\n", - (priv->phymode << 8) | - le16_to_cpu(priv->active_rxon.channel)); -} - -static void iwl3945_set_flags_for_phymode(struct iwl3945_priv *priv, u8 phymode); - -static ssize_t store_tune(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct iwl3945_priv *priv = (struct iwl3945_priv *)d->driver_data; - char *p = (char *)buf; - u16 tune = simple_strtoul(p, &p, 0); - u8 phymode = (tune >> 8) & 0xff; - u16 channel = tune & 0xff; - - IWL_DEBUG_INFO("Tune request to:%d channel:%d\n", phymode, channel); - - mutex_lock(&priv->mutex); - if ((le16_to_cpu(priv->staging_rxon.channel) != channel) || - (priv->phymode != phymode)) { - const struct iwl3945_channel_info *ch_info; - - ch_info = iwl3945_get_channel_info(priv, phymode, channel); - if (!ch_info) { - IWL_WARNING("Requested invalid phymode/channel " - "combination: %d %d\n", phymode, channel); - mutex_unlock(&priv->mutex); - return -EINVAL; - } - - /* Cancel any currently running scans... */ - if (iwl3945_scan_cancel_timeout(priv, 100)) - IWL_WARNING("Could not cancel scan.\n"); - else { - IWL_DEBUG_INFO("Committing phymode and " - "rxon.channel = %d %d\n", - phymode, channel); - - iwl3945_set_rxon_channel(priv, phymode, channel); - iwl3945_set_flags_for_phymode(priv, phymode); - - iwl3945_set_rate(priv); - iwl3945_commit_rxon(priv); - } - } - mutex_unlock(&priv->mutex); - - return count; -} - -static DEVICE_ATTR(tune, S_IWUSR | S_IRUGO, show_tune, store_tune); - #ifdef CONFIG_IWL3945_SPECTRUM_MEASUREMENT static ssize_t show_measurement(struct device *d, @@ -8165,73 +8035,8 @@ static DEVICE_ATTR(power_level, S_IWUSR | S_IRUSR, show_power_level, static ssize_t show_channels(struct device *d, struct device_attribute *attr, char *buf) { - struct iwl3945_priv *priv = dev_get_drvdata(d); - int len = 0, i; - struct ieee80211_channel *channels = NULL; - const struct ieee80211_hw_mode *hw_mode = NULL; - int count = 0; - - if (!iwl3945_is_ready(priv)) - return -EAGAIN; - - hw_mode = iwl3945_get_hw_mode(priv, MODE_IEEE80211G); - if (!hw_mode) - hw_mode = iwl3945_get_hw_mode(priv, MODE_IEEE80211B); - if (hw_mode) { - channels = hw_mode->channels; - count = hw_mode->num_channels; - } - - len += - sprintf(&buf[len], - "Displaying %d channels in 2.4GHz band " - "(802.11bg):\n", count); - - for (i = 0; i < count; i++) - len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n", - channels[i].chan, - channels[i].power_level, - channels[i]. - flag & IEEE80211_CHAN_W_RADAR_DETECT ? - " (IEEE 802.11h required)" : "", - (!(channels[i].flag & IEEE80211_CHAN_W_IBSS) - || (channels[i]. - flag & - IEEE80211_CHAN_W_RADAR_DETECT)) ? "" : - ", IBSS", - channels[i]. - flag & IEEE80211_CHAN_W_ACTIVE_SCAN ? - "active/passive" : "passive only"); - - hw_mode = iwl3945_get_hw_mode(priv, MODE_IEEE80211A); - if (hw_mode) { - channels = hw_mode->channels; - count = hw_mode->num_channels; - } else { - channels = NULL; - count = 0; - } - - len += sprintf(&buf[len], "Displaying %d channels in 5.2GHz band " - "(802.11a):\n", count); - - for (i = 0; i < count; i++) - len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n", - channels[i].chan, - channels[i].power_level, - channels[i]. - flag & IEEE80211_CHAN_W_RADAR_DETECT ? - " (IEEE 802.11h required)" : "", - (!(channels[i].flag & IEEE80211_CHAN_W_IBSS) - || (channels[i]. - flag & - IEEE80211_CHAN_W_RADAR_DETECT)) ? "" : - ", IBSS", - channels[i]. - flag & IEEE80211_CHAN_W_ACTIVE_SCAN ? - "active/passive" : "passive only"); - - return len; + /* all this shit doesn't belong into sysfs anyway */ + return 0; } static DEVICE_ATTR(channels, S_IRUSR, show_channels, NULL); @@ -8411,7 +8216,6 @@ static struct attribute *iwl3945_sysfs_entries[] = { &dev_attr_statistics.attr, &dev_attr_status.attr, &dev_attr_temperature.attr, - &dev_attr_tune.attr, &dev_attr_tx_power.attr, NULL @@ -8532,7 +8336,7 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e priv->data_retry_limit = -1; priv->ieee_channels = NULL; priv->ieee_rates = NULL; - priv->phymode = -1; + priv->band = IEEE80211_BAND_2GHZ; err = pci_set_dma_mask(pdev, DMA_32BIT_MASK); if (!err) @@ -8614,7 +8418,7 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e priv->qos_data.qos_cap.val = 0; #endif /* CONFIG_IWL3945_QOS */ - iwl3945_set_rxon_channel(priv, MODE_IEEE80211G, 6); + iwl3945_set_rxon_channel(priv, IEEE80211_BAND_2GHZ, 6); iwl3945_setup_deferred_work(priv); iwl3945_setup_rx_handlers(priv); @@ -8665,7 +8469,6 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e IWL_ERROR("initializing geos failed: %d\n", err); goto out_free_channel_map; } - iwl3945_reset_channel_flag(priv); iwl3945_rate_control_register(priv->hw); err = ieee80211_register_hw(priv->hw); diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c index 5f38fc585eda..6de969de4c84 100644 --- a/drivers/net/wireless/iwlwifi/iwl4965-base.c +++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c @@ -115,16 +115,10 @@ __le16 *ieee80211_get_qos_ctrl(struct ieee80211_hdr *hdr) return NULL; } -static const struct ieee80211_hw_mode *iwl4965_get_hw_mode( - struct iwl4965_priv *priv, int mode) +static const struct ieee80211_supported_band *iwl4965_get_hw_mode( + struct iwl4965_priv *priv, enum ieee80211_band band) { - int i; - - for (i = 0; i < 3; i++) - if (priv->modes[i].mode == mode) - return &priv->modes[i]; - - return NULL; + return priv->hw->wiphy->bands[band]; } static int iwl4965_is_empty_essid(const char *essid, int essid_len) @@ -937,28 +931,29 @@ static int iwl4965_rxon_add_station(struct iwl4965_priv *priv, * NOTE: Does not commit to the hardware; it sets appropriate bit fields * in the staging RXON flag structure based on the phymode */ -static int iwl4965_set_rxon_channel(struct iwl4965_priv *priv, u8 phymode, +static int iwl4965_set_rxon_channel(struct iwl4965_priv *priv, + enum ieee80211_band band, u16 channel) { - if (!iwl4965_get_channel_info(priv, phymode, channel)) { + if (!iwl4965_get_channel_info(priv, band, channel)) { IWL_DEBUG_INFO("Could not set channel to %d [%d]\n", - channel, phymode); + channel, band); return -EINVAL; } if ((le16_to_cpu(priv->staging_rxon.channel) == channel) && - (priv->phymode == phymode)) + (priv->band == band)) return 0; priv->staging_rxon.channel = cpu_to_le16(channel); - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) priv->staging_rxon.flags &= ~RXON_FLG_BAND_24G_MSK; else priv->staging_rxon.flags |= RXON_FLG_BAND_24G_MSK; - priv->phymode = phymode; + priv->band = band; - IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, phymode); + IWL_DEBUG_INFO("Staging channel set to %d [%d]\n", channel, band); return 0; } @@ -2571,9 +2566,10 @@ static int iwl4965_set_rxon_hwcrypto(struct iwl4965_priv *priv, int hw_decrypt) return 0; } -static void iwl4965_set_flags_for_phymode(struct iwl4965_priv *priv, u8 phymode) +static void iwl4965_set_flags_for_phymode(struct iwl4965_priv *priv, + enum ieee80211_band band) { - if (phymode == MODE_IEEE80211A) { + if (band == IEEE80211_BAND_5GHZ) { priv->staging_rxon.flags &= ~(RXON_FLG_BAND_24G_MSK | RXON_FLG_AUTO_DETECT_MSK | RXON_FLG_CCK_MSK); @@ -2636,7 +2632,7 @@ static void iwl4965_connection_init_rx_config(struct iwl4965_priv *priv) priv->staging_rxon.flags |= RXON_FLG_SHORT_PREAMBLE_MSK; #endif - ch_info = iwl4965_get_channel_info(priv, priv->phymode, + ch_info = iwl4965_get_channel_info(priv, priv->band, le16_to_cpu(priv->staging_rxon.channel)); if (!ch_info) @@ -2651,12 +2647,9 @@ static void iwl4965_connection_init_rx_config(struct iwl4965_priv *priv) ch_info = &priv->channel_info[0]; priv->staging_rxon.channel = cpu_to_le16(ch_info->channel); - if (is_channel_a_band(ch_info)) - priv->phymode = MODE_IEEE80211A; - else - priv->phymode = MODE_IEEE80211G; + priv->band = ch_info->band; - iwl4965_set_flags_for_phymode(priv, priv->phymode); + iwl4965_set_flags_for_phymode(priv, priv->band); priv->staging_rxon.ofdm_basic_rates = (IWL_OFDM_RATES_MASK >> IWL_FIRST_OFDM_RATE) & 0xFF; @@ -2678,7 +2671,7 @@ static int iwl4965_set_mode(struct iwl4965_priv *priv, int mode) const struct iwl4965_channel_info *ch_info; ch_info = iwl4965_get_channel_info(priv, - priv->phymode, + priv->band, le16_to_cpu(priv->staging_rxon.channel)); if (!ch_info || !is_channel_ibss(ch_info)) { @@ -2918,7 +2911,7 @@ static int iwl4965_tx_skb(struct iwl4965_priv *priv, goto drop_unlock; } - if ((ctl->tx_rate & 0xFF) == IWL_INVALID_RATE) { + if ((ctl->tx_rate->hw_value & 0xFF) == IWL_INVALID_RATE) { IWL_ERROR("ERROR: No TX rate available.\n"); goto drop_unlock; } @@ -3125,11 +3118,11 @@ drop: static void iwl4965_set_rate(struct iwl4965_priv *priv) { - const struct ieee80211_hw_mode *hw = NULL; + const struct ieee80211_supported_band *hw = NULL; struct ieee80211_rate *rate; int i; - hw = iwl4965_get_hw_mode(priv, priv->phymode); + hw = iwl4965_get_hw_mode(priv, priv->band); if (!hw) { IWL_ERROR("Failed to set rate: unable to get hw mode\n"); return; @@ -3138,24 +3131,10 @@ static void iwl4965_set_rate(struct iwl4965_priv *priv) priv->active_rate = 0; priv->active_rate_basic = 0; - IWL_DEBUG_RATE("Setting rates for 802.11%c\n", - hw->mode == MODE_IEEE80211A ? - 'a' : ((hw->mode == MODE_IEEE80211B) ? 'b' : 'g')); - - for (i = 0; i < hw->num_rates; i++) { - rate = &(hw->rates[i]); - if ((rate->val < IWL_RATE_COUNT) && - (rate->flags & IEEE80211_RATE_SUPPORTED)) { - IWL_DEBUG_RATE("Adding rate index %d (plcp %d)%s\n", - rate->val, iwl4965_rates[rate->val].plcp, - (rate->flags & IEEE80211_RATE_BASIC) ? - "*" : ""); - priv->active_rate |= (1 << rate->val); - if (rate->flags & IEEE80211_RATE_BASIC) - priv->active_rate_basic |= (1 << rate->val); - } else - IWL_DEBUG_RATE("Not adding rate %d (plcp %d)\n", - rate->val, iwl4965_rates[rate->val].plcp); + for (i = 0; i < hw->n_bitrates; i++) { + rate = &(hw->bitrates[i]); + if (rate->hw_value < IWL_RATE_COUNT) + priv->active_rate |= (1 << rate->hw_value); } IWL_DEBUG_RATE("Set active_rate = %0x, active_rate_basic = %0x\n", @@ -3775,9 +3754,6 @@ static void iwl4965_rx_reply_tx(struct iwl4965_priv *priv, tx_status->flags = iwl4965_is_tx_success(status) ? IEEE80211_TX_STATUS_ACK : 0; - tx_status->control.tx_rate = - iwl4965_hw_get_rate_n_flags(tx_resp->rate_n_flags); - IWL_DEBUG_TX("Tx queue %d Status %s (0x%08x) rate_n_flags 0x%x " "retries %d\n", txq_id, iwl4965_get_tx_fail_reason(status), status, le32_to_cpu(tx_resp->rate_n_flags), @@ -5419,24 +5395,23 @@ static void iwl4965_init_band_reference(const struct iwl4965_priv *priv, * Based on band and channel number. */ const struct iwl4965_channel_info *iwl4965_get_channel_info(const struct iwl4965_priv *priv, - int phymode, u16 channel) + enum ieee80211_band band, u16 channel) { int i; - switch (phymode) { - case MODE_IEEE80211A: + switch (band) { + case IEEE80211_BAND_5GHZ: for (i = 14; i < priv->channel_count; i++) { if (priv->channel_info[i].channel == channel) return &priv->channel_info[i]; } break; - - case MODE_IEEE80211B: - case MODE_IEEE80211G: + case IEEE80211_BAND_2GHZ: if (channel >= 1 && channel <= 14) return &priv->channel_info[channel - 1]; break; - + default: + BUG(); } return NULL; @@ -5499,8 +5474,8 @@ static int iwl4965_init_channel_map(struct iwl4965_priv *priv) /* Loop through each band adding each of the channels */ for (ch = 0; ch < eeprom_ch_count; ch++) { ch_info->channel = eeprom_ch_index[ch]; - ch_info->phymode = (band == 1) ? MODE_IEEE80211B : - MODE_IEEE80211A; + ch_info->band = (band == 1) ? IEEE80211_BAND_2GHZ : + IEEE80211_BAND_5GHZ; /* permanently store EEPROM's channel regulatory flags * and max power in channel info database. */ @@ -5559,14 +5534,14 @@ static int iwl4965_init_channel_map(struct iwl4965_priv *priv) /* Two additional EEPROM bands for 2.4 and 5 GHz FAT channels */ for (band = 6; band <= 7; band++) { - int phymode; + enum ieee80211_band ieeeband; u8 fat_extension_chan; iwl4965_init_band_reference(priv, band, &eeprom_ch_count, &eeprom_ch_info, &eeprom_ch_index); /* EEPROM band 6 is 2.4, band 7 is 5 GHz */ - phymode = (band == 6) ? MODE_IEEE80211B : MODE_IEEE80211A; + ieeeband = (band == 6) ? IEEE80211_BAND_2GHZ : IEEE80211_BAND_5GHZ; /* Loop through each band adding each of the channels */ for (ch = 0; ch < eeprom_ch_count; ch++) { @@ -5580,13 +5555,13 @@ static int iwl4965_init_channel_map(struct iwl4965_priv *priv) fat_extension_chan = HT_IE_EXT_CHANNEL_ABOVE; /* Set up driver's info for lower half */ - iwl4965_set_fat_chan_info(priv, phymode, + iwl4965_set_fat_chan_info(priv, ieeeband, eeprom_ch_index[ch], &(eeprom_ch_info[ch]), fat_extension_chan); /* Set up driver's info for upper half */ - iwl4965_set_fat_chan_info(priv, phymode, + iwl4965_set_fat_chan_info(priv, ieeeband, (eeprom_ch_index[ch] + 4), &(eeprom_ch_info[ch]), HT_IE_EXT_CHANNEL_BELOW); @@ -5628,18 +5603,20 @@ static void iwl4965_free_channel_map(struct iwl4965_priv *priv) #define IWL_PASSIVE_DWELL_BASE (100) #define IWL_CHANNEL_TUNE_TIME 5 -static inline u16 iwl4965_get_active_dwell_time(struct iwl4965_priv *priv, int phymode) +static inline u16 iwl4965_get_active_dwell_time(struct iwl4965_priv *priv, + enum ieee80211_band band) { - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) return IWL_ACTIVE_DWELL_TIME_52; else return IWL_ACTIVE_DWELL_TIME_24; } -static u16 iwl4965_get_passive_dwell_time(struct iwl4965_priv *priv, int phymode) +static u16 iwl4965_get_passive_dwell_time(struct iwl4965_priv *priv, + enum ieee80211_band band) { - u16 active = iwl4965_get_active_dwell_time(priv, phymode); - u16 passive = (phymode != MODE_IEEE80211A) ? + u16 active = iwl4965_get_active_dwell_time(priv, band); + u16 passive = (band != IEEE80211_BAND_5GHZ) ? IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_24 : IWL_PASSIVE_DWELL_BASE + IWL_PASSIVE_DWELL_TIME_52; @@ -5659,28 +5636,29 @@ static u16 iwl4965_get_passive_dwell_time(struct iwl4965_priv *priv, int phymode return passive; } -static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode, +static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, + enum ieee80211_band band, u8 is_active, u8 direct_mask, struct iwl4965_scan_channel *scan_ch) { const struct ieee80211_channel *channels = NULL; - const struct ieee80211_hw_mode *hw_mode; + const struct ieee80211_supported_band *sband; const struct iwl4965_channel_info *ch_info; u16 passive_dwell = 0; u16 active_dwell = 0; int added, i; - hw_mode = iwl4965_get_hw_mode(priv, phymode); - if (!hw_mode) + sband = iwl4965_get_hw_mode(priv, band); + if (!sband) return 0; - channels = hw_mode->channels; + channels = sband->channels; - active_dwell = iwl4965_get_active_dwell_time(priv, phymode); - passive_dwell = iwl4965_get_passive_dwell_time(priv, phymode); + active_dwell = iwl4965_get_active_dwell_time(priv, band); + passive_dwell = iwl4965_get_passive_dwell_time(priv, band); - for (i = 0, added = 0; i < hw_mode->num_channels; i++) { - if (channels[i].chan == + for (i = 0, added = 0; i < sband->n_channels; i++) { + if (ieee80211_frequency_to_channel(channels[i].center_freq) == le16_to_cpu(priv->active_rxon.channel)) { if (iwl4965_is_associated(priv)) { IWL_DEBUG_SCAN @@ -5691,9 +5669,9 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode, } else if (priv->only_active_channel) continue; - scan_ch->channel = channels[i].chan; + scan_ch->channel = ieee80211_frequency_to_channel(channels[i].center_freq); - ch_info = iwl4965_get_channel_info(priv, phymode, + ch_info = iwl4965_get_channel_info(priv, band, scan_ch->channel); if (!is_channel_valid(ch_info)) { IWL_DEBUG_SCAN("Channel %d is INVALID for this SKU.\n", @@ -5702,7 +5680,7 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode, } if (!is_active || is_channel_passive(ch_info) || - !(channels[i].flag & IEEE80211_CHAN_W_ACTIVE_SCAN)) + (channels[i].flags & IEEE80211_CHAN_PASSIVE_SCAN)) scan_ch->type = 0; /* passive */ else scan_ch->type = 1; /* active */ @@ -5721,7 +5699,7 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode, /* scan_pwr_info->tpc.dsp_atten; */ /*scan_pwr_info->tpc.tx_gain; */ - if (phymode == MODE_IEEE80211A) + if (band == IEEE80211_BAND_5GHZ) scan_ch->tpc.tx_gain = ((1 << 5) | (3 << 3)) | 3; else { scan_ch->tpc.tx_gain = ((1 << 5) | (5 << 3)); @@ -5745,41 +5723,23 @@ static int iwl4965_get_channels_for_scan(struct iwl4965_priv *priv, int phymode, return added; } -static void iwl4965_reset_channel_flag(struct iwl4965_priv *priv) -{ - int i, j; - for (i = 0; i < 3; i++) { - struct ieee80211_hw_mode *hw_mode = (void *)&priv->modes[i]; - for (j = 0; j < hw_mode->num_channels; j++) - hw_mode->channels[j].flag = hw_mode->channels[j].val; - } -} - static void iwl4965_init_hw_rates(struct iwl4965_priv *priv, struct ieee80211_rate *rates) { int i; for (i = 0; i < IWL_RATE_COUNT; i++) { - rates[i].rate = iwl4965_rates[i].ieee * 5; - rates[i].val = i; /* Rate scaling will work on indexes */ - rates[i].val2 = i; - rates[i].flags = IEEE80211_RATE_SUPPORTED; - /* Only OFDM have the bits-per-symbol set */ - if ((i <= IWL_LAST_OFDM_RATE) && (i >= IWL_FIRST_OFDM_RATE)) - rates[i].flags |= IEEE80211_RATE_OFDM; - else { + rates[i].bitrate = iwl4965_rates[i].ieee * 5; + rates[i].hw_value = i; /* Rate scaling will work on indexes */ + rates[i].hw_value_short = i; + rates[i].flags = 0; + if ((i > IWL_LAST_OFDM_RATE) || (i < IWL_FIRST_OFDM_RATE)) { /* - * If CCK 1M then set rate flag to CCK else CCK_2 - * which is CCK | PREAMBLE2 + * If CCK != 1M then set short preamble rate flag. */ rates[i].flags |= (iwl4965_rates[i].plcp == 10) ? - IEEE80211_RATE_CCK : IEEE80211_RATE_CCK_2; + 0 : IEEE80211_RATE_SHORT_PREAMBLE; } - - /* Set up which ones are basic rates... */ - if (IWL_BASIC_RATES_MASK & (1 << i)) - rates[i].flags |= IEEE80211_RATE_BASIC; } } @@ -5789,74 +5749,47 @@ static void iwl4965_init_hw_rates(struct iwl4965_priv *priv, static int iwl4965_init_geos(struct iwl4965_priv *priv) { struct iwl4965_channel_info *ch; - struct ieee80211_hw_mode *modes; + struct ieee80211_supported_band *band; struct ieee80211_channel *channels; struct ieee80211_channel *geo_ch; struct ieee80211_rate *rates; int i = 0; - enum { - A = 0, - B = 1, - G = 2, - }; - int mode_count = 3; - if (priv->modes) { + if (priv->bands[IEEE80211_BAND_2GHZ].n_bitrates || + priv->bands[IEEE80211_BAND_5GHZ].n_bitrates) { IWL_DEBUG_INFO("Geography modes already initialized.\n"); set_bit(STATUS_GEO_CONFIGURED, &priv->status); return 0; } - modes = kzalloc(sizeof(struct ieee80211_hw_mode) * mode_count, - GFP_KERNEL); - if (!modes) - return -ENOMEM; - channels = kzalloc(sizeof(struct ieee80211_channel) * priv->channel_count, GFP_KERNEL); - if (!channels) { - kfree(modes); + if (!channels) return -ENOMEM; - } rates = kzalloc((sizeof(struct ieee80211_rate) * (IWL_MAX_RATES + 1)), GFP_KERNEL); if (!rates) { - kfree(modes); kfree(channels); return -ENOMEM; } - /* 0 = 802.11a - * 1 = 802.11b - * 2 = 802.11g - */ - /* 5.2GHz channels start after the 2.4GHz channels */ - modes[A].mode = MODE_IEEE80211A; - modes[A].channels = &channels[ARRAY_SIZE(iwl4965_eeprom_band_1)]; - modes[A].rates = rates; - modes[A].num_rates = 8; /* just OFDM */ - modes[A].rates = &rates[4]; - modes[A].num_channels = 0; #ifdef CONFIG_IWL4965_HT iwl4965_init_ht_hw_capab(&modes[A].ht_info, MODE_IEEE80211A); #endif - - modes[B].mode = MODE_IEEE80211B; - modes[B].channels = channels; - modes[B].rates = rates; - modes[B].num_rates = 4; /* just CCK */ - modes[B].num_channels = 0; - - modes[G].mode = MODE_IEEE80211G; - modes[G].channels = channels; - modes[G].rates = rates; - modes[G].num_rates = 12; /* OFDM & CCK */ - modes[G].num_channels = 0; #ifdef CONFIG_IWL4965_HT iwl4965_init_ht_hw_capab(&modes[G].ht_info, MODE_IEEE80211G); #endif + band = &priv->bands[IEEE80211_BAND_5GHZ]; + band->channels = &channels[ARRAY_SIZE(iwl4965_eeprom_band_1)]; + band->bitrates = &rates[4]; + band->n_bitrates = 8; /* just OFDM */ + + band = &priv->bands[IEEE80211_BAND_2GHZ]; + band->channels = channels; + band->bitrates = rates; + band->n_bitrates = 12; /* OFDM & CCK */ priv->ieee_channels = channels; priv->ieee_rates = rates; @@ -5875,37 +5808,32 @@ static int iwl4965_init_geos(struct iwl4965_priv *priv) } if (is_channel_a_band(ch)) { - geo_ch = &modes[A].channels[modes[A].num_channels++]; - } else { - geo_ch = &modes[B].channels[modes[B].num_channels++]; - modes[G].num_channels++; - } + geo_ch = &priv->bands[IEEE80211_BAND_5GHZ].channels[priv->bands[IEEE80211_BAND_5GHZ].n_channels++]; + } else + geo_ch = &priv->bands[IEEE80211_BAND_2GHZ].channels[priv->bands[IEEE80211_BAND_2GHZ].n_channels++]; - geo_ch->freq = ieee80211chan2mhz(ch->channel); - geo_ch->chan = ch->channel; - geo_ch->power_level = ch->max_power_avg; - geo_ch->antenna_max = 0xff; + geo_ch->center_freq = ieee80211chan2mhz(ch->channel); + geo_ch->max_power = ch->max_power_avg; + geo_ch->max_antenna_gain = 0xff; if (is_channel_valid(ch)) { - geo_ch->flag = IEEE80211_CHAN_W_SCAN; - if (ch->flags & EEPROM_CHANNEL_IBSS) - geo_ch->flag |= IEEE80211_CHAN_W_IBSS; + if (!(ch->flags & EEPROM_CHANNEL_IBSS)) + geo_ch->flags |= IEEE80211_CHAN_NO_IBSS; - if (ch->flags & EEPROM_CHANNEL_ACTIVE) - geo_ch->flag |= IEEE80211_CHAN_W_ACTIVE_SCAN; + if (!(ch->flags & EEPROM_CHANNEL_ACTIVE)) + geo_ch->flags |= IEEE80211_CHAN_PASSIVE_SCAN; if (ch->flags & EEPROM_CHANNEL_RADAR) - geo_ch->flag |= IEEE80211_CHAN_W_RADAR_DETECT; + geo_ch->flags |= IEEE80211_CHAN_RADAR; if (ch->max_power_avg > priv->max_channel_txpower_limit) priv->max_channel_txpower_limit = ch->max_power_avg; - } - - geo_ch->val = geo_ch->flag; + } else + geo_ch->flags |= IEEE80211_CHAN_DISABLED; } - if ((modes[A].num_channels == 0) && priv->is_abg) { + if ((priv->bands[IEEE80211_BAND_5GHZ].n_channels == 0) && priv->is_abg) { printk(KERN_INFO DRV_NAME ": Incorrectly detected BG card as ABG. Please send " "your PCI ID 0x%04X:0x%04X to maintainer.\n", @@ -5915,24 +5843,12 @@ static int iwl4965_init_geos(struct iwl4965_priv *priv) printk(KERN_INFO DRV_NAME ": Tunable channels: %d 802.11bg, %d 802.11a channels\n", - modes[G].num_channels, modes[A].num_channels); - - /* - * NOTE: We register these in preference of order -- the - * stack doesn't currently (as of 7.0.6 / Apr 24 '07) pick - * a phymode based on rates or AP capabilities but seems to - * configure it purely on if the channel being configured - * is supported by a mode -- and the first match is taken - */ + priv->bands[IEEE80211_BAND_2GHZ].n_channels, + priv->bands[IEEE80211_BAND_5GHZ].n_channels); - if (modes[G].num_channels) - ieee80211_register_hwmode(priv->hw, &modes[G]); - if (modes[B].num_channels) - ieee80211_register_hwmode(priv->hw, &modes[B]); - if (modes[A].num_channels) - ieee80211_register_hwmode(priv->hw, &modes[A]); + priv->hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->bands[IEEE80211_BAND_2GHZ]; + priv->hw->wiphy->bands[IEEE80211_BAND_5GHZ] = &priv->bands[IEEE80211_BAND_5GHZ]; - priv->modes = modes; set_bit(STATUS_GEO_CONFIGURED, &priv->status); return 0; @@ -5943,7 +5859,6 @@ static int iwl4965_init_geos(struct iwl4965_priv *priv) */ static void iwl4965_free_geos(struct iwl4965_priv *priv) { - kfree(priv->modes); kfree(priv->ieee_channels); kfree(priv->ieee_rates); clear_bit(STATUS_GEO_CONFIGURED, &priv->status); @@ -6945,7 +6860,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data) struct iwl4965_scan_cmd *scan; struct ieee80211_conf *conf = NULL; u8 direct_mask; - int phymode; + enum ieee80211_band band; conf = ieee80211_get_hw_conf(priv->hw); @@ -7075,7 +6990,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data) RATE_MCS_ANT_B_MSK|RATE_MCS_CCK_MSK); scan->good_CRC_th = 0; - phymode = MODE_IEEE80211G; + band = IEEE80211_BAND_2GHZ; break; case 1: @@ -7083,7 +6998,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data) iwl4965_hw_set_rate_n_flags(IWL_RATE_6M_PLCP, RATE_MCS_ANT_B_MSK); scan->good_CRC_th = IWL_GOOD_CRC_TH; - phymode = MODE_IEEE80211A; + band = IEEE80211_BAND_5GHZ; break; default: @@ -7113,7 +7028,7 @@ static void iwl4965_bg_request_scan(struct work_struct *data) scan->channel_count = iwl4965_get_channels_for_scan( - priv, phymode, 1, /* active */ + priv, band, 1, /* active */ direct_mask, (void *)&scan->data[le16_to_cpu(scan->tx_cmd.len)]); @@ -7463,7 +7378,7 @@ static int iwl4965_mac_tx(struct ieee80211_hw *hw, struct sk_buff *skb, } IWL_DEBUG_TX("dev->xmit(%d bytes) at rate 0x%02x\n", skb->len, - ctl->tx_rate); + ctl->tx_rate->bitrate); if (iwl4965_tx_skb(priv, skb, ctl)) dev_kfree_skb_any(skb); @@ -7522,7 +7437,7 @@ static int iwl4965_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co int ret = 0; mutex_lock(&priv->mutex); - IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel); + IWL_DEBUG_MAC80211("enter to channel %d\n", conf->channel->hw_value); priv->add_radiotap = !!(conf->flags & IEEE80211_CONF_RADIOTAP); @@ -7542,10 +7457,9 @@ static int iwl4965_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co spin_lock_irqsave(&priv->lock, flags); - ch_info = iwl4965_get_channel_info(priv, conf->phymode, conf->channel); + ch_info = iwl4965_get_channel_info(priv, conf->channel->band, + ieee80211_frequency_to_channel(conf->channel->center_freq)); if (!is_channel_valid(ch_info)) { - IWL_DEBUG_SCAN("Channel %d [%d] is INVALID for this SKU.\n", - conf->channel, conf->phymode); IWL_DEBUG_MAC80211("leave - invalid channel\n"); spin_unlock_irqrestore(&priv->lock, flags); ret = -EINVAL; @@ -7564,12 +7478,13 @@ static int iwl4965_mac_config(struct ieee80211_hw *hw, struct ieee80211_conf *co priv->staging_rxon.flags = 0; #endif /* CONFIG_IWL4965_HT */ - iwl4965_set_rxon_channel(priv, conf->phymode, conf->channel); + iwl4965_set_rxon_channel(priv, conf->channel->band, + ieee80211_frequency_to_channel(conf->channel->center_freq)); - iwl4965_set_flags_for_phymode(priv, conf->phymode); + iwl4965_set_flags_for_phymode(priv, conf->channel->band); /* The list of supported rates and rate mask can be different - * for each phymode; since the phymode may have changed, reset + * for each band; since the band may have changed, reset * the rate mask to what mac80211 lists */ iwl4965_set_rate(priv); @@ -7839,7 +7754,7 @@ static void iwl4965_bss_info_changed(struct ieee80211_hw *hw, } if (changes & BSS_CHANGED_ERP_CTS_PROT) { - if (bss_conf->use_cts_prot && (priv->phymode != MODE_IEEE80211A)) + if (bss_conf->use_cts_prot && (priv->band != IEEE80211_BAND_5GHZ)) priv->staging_rxon.flags |= RXON_FLG_TGG_PROTECT_MSK; else priv->staging_rxon.flags &= ~RXON_FLG_TGG_PROTECT_MSK; @@ -8277,7 +8192,6 @@ static void iwl4965_set_ht_capab(struct ieee80211_hw *hw, u8 use_current_config) { struct ieee80211_conf *conf = &hw->conf; - struct ieee80211_hw_mode *mode = conf->mode; if (use_current_config) { ht_cap->cap_info = cpu_to_le16(conf->ht_conf.cap); @@ -8488,65 +8402,6 @@ static ssize_t store_filter_flags(struct device *d, static DEVICE_ATTR(filter_flags, S_IWUSR | S_IRUGO, show_filter_flags, store_filter_flags); -static ssize_t show_tune(struct device *d, - struct device_attribute *attr, char *buf) -{ - struct iwl4965_priv *priv = (struct iwl4965_priv *)d->driver_data; - - return sprintf(buf, "0x%04X\n", - (priv->phymode << 8) | - le16_to_cpu(priv->active_rxon.channel)); -} - -static void iwl4965_set_flags_for_phymode(struct iwl4965_priv *priv, u8 phymode); - -static ssize_t store_tune(struct device *d, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct iwl4965_priv *priv = (struct iwl4965_priv *)d->driver_data; - char *p = (char *)buf; - u16 tune = simple_strtoul(p, &p, 0); - u8 phymode = (tune >> 8) & 0xff; - u16 channel = tune & 0xff; - - IWL_DEBUG_INFO("Tune request to:%d channel:%d\n", phymode, channel); - - mutex_lock(&priv->mutex); - if ((le16_to_cpu(priv->staging_rxon.channel) != channel) || - (priv->phymode != phymode)) { - const struct iwl4965_channel_info *ch_info; - - ch_info = iwl4965_get_channel_info(priv, phymode, channel); - if (!ch_info) { - IWL_WARNING("Requested invalid phymode/channel " - "combination: %d %d\n", phymode, channel); - mutex_unlock(&priv->mutex); - return -EINVAL; - } - - /* Cancel any currently running scans... */ - if (iwl4965_scan_cancel_timeout(priv, 100)) - IWL_WARNING("Could not cancel scan.\n"); - else { - IWL_DEBUG_INFO("Committing phymode and " - "rxon.channel = %d %d\n", - phymode, channel); - - iwl4965_set_rxon_channel(priv, phymode, channel); - iwl4965_set_flags_for_phymode(priv, phymode); - - iwl4965_set_rate(priv); - iwl4965_commit_rxon(priv); - } - } - mutex_unlock(&priv->mutex); - - return count; -} - -static DEVICE_ATTR(tune, S_IWUSR | S_IRUGO, show_tune, store_tune); - #ifdef CONFIG_IWL4965_SPECTRUM_MEASUREMENT static ssize_t show_measurement(struct device *d, @@ -8736,73 +8591,8 @@ static DEVICE_ATTR(power_level, S_IWUSR | S_IRUSR, show_power_level, static ssize_t show_channels(struct device *d, struct device_attribute *attr, char *buf) { - struct iwl4965_priv *priv = dev_get_drvdata(d); - int len = 0, i; - struct ieee80211_channel *channels = NULL; - const struct ieee80211_hw_mode *hw_mode = NULL; - int count = 0; - - if (!iwl4965_is_ready(priv)) - return -EAGAIN; - - hw_mode = iwl4965_get_hw_mode(priv, MODE_IEEE80211G); - if (!hw_mode) - hw_mode = iwl4965_get_hw_mode(priv, MODE_IEEE80211B); - if (hw_mode) { - channels = hw_mode->channels; - count = hw_mode->num_channels; - } - - len += - sprintf(&buf[len], - "Displaying %d channels in 2.4GHz band " - "(802.11bg):\n", count); - - for (i = 0; i < count; i++) - len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n", - channels[i].chan, - channels[i].power_level, - channels[i]. - flag & IEEE80211_CHAN_W_RADAR_DETECT ? - " (IEEE 802.11h required)" : "", - (!(channels[i].flag & IEEE80211_CHAN_W_IBSS) - || (channels[i]. - flag & - IEEE80211_CHAN_W_RADAR_DETECT)) ? "" : - ", IBSS", - channels[i]. - flag & IEEE80211_CHAN_W_ACTIVE_SCAN ? - "active/passive" : "passive only"); - - hw_mode = iwl4965_get_hw_mode(priv, MODE_IEEE80211A); - if (hw_mode) { - channels = hw_mode->channels; - count = hw_mode->num_channels; - } else { - channels = NULL; - count = 0; - } - - len += sprintf(&buf[len], "Displaying %d channels in 5.2GHz band " - "(802.11a):\n", count); - - for (i = 0; i < count; i++) - len += sprintf(&buf[len], "%d: %ddBm: BSS%s%s, %s.\n", - channels[i].chan, - channels[i].power_level, - channels[i]. - flag & IEEE80211_CHAN_W_RADAR_DETECT ? - " (IEEE 802.11h required)" : "", - (!(channels[i].flag & IEEE80211_CHAN_W_IBSS) - || (channels[i]. - flag & - IEEE80211_CHAN_W_RADAR_DETECT)) ? "" : - ", IBSS", - channels[i]. - flag & IEEE80211_CHAN_W_ACTIVE_SCAN ? - "active/passive" : "passive only"); - - return len; + /* all this shit doesn't belong into sysfs anyway */ + return 0; } static DEVICE_ATTR(channels, S_IRUSR, show_channels, NULL); @@ -8981,7 +8771,6 @@ static struct attribute *iwl4965_sysfs_entries[] = { &dev_attr_statistics.attr, &dev_attr_status.attr, &dev_attr_temperature.attr, - &dev_attr_tune.attr, &dev_attr_tx_power.attr, NULL @@ -9109,7 +8898,7 @@ static int iwl4965_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e priv->data_retry_limit = -1; priv->ieee_channels = NULL; priv->ieee_rates = NULL; - priv->phymode = -1; + priv->band = IEEE80211_BAND_2GHZ; err = pci_set_dma_mask(pdev, DMA_32BIT_MASK); if (!err) @@ -9175,7 +8964,7 @@ static int iwl4965_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e priv->qos_data.qos_cap.val = 0; #endif /* CONFIG_IWL4965_QOS */ - iwl4965_set_rxon_channel(priv, MODE_IEEE80211G, 6); + iwl4965_set_rxon_channel(priv, IEEE80211_BAND_2GHZ, 6); iwl4965_setup_deferred_work(priv); iwl4965_setup_rx_handlers(priv); @@ -9226,7 +9015,6 @@ static int iwl4965_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e IWL_ERROR("initializing geos failed: %d\n", err); goto out_free_channel_map; } - iwl4965_reset_channel_flag(priv); iwl4965_rate_control_register(priv->hw); err = ieee80211_register_hw(priv->hw); diff --git a/drivers/net/wireless/p54.h b/drivers/net/wireless/p54.h index 744c866066c5..06d2c67f4c81 100644 --- a/drivers/net/wireless/p54.h +++ b/drivers/net/wireless/p54.h @@ -64,10 +64,6 @@ struct p54_common { unsigned int tx_hdr_len; void *cached_vdcf; unsigned int fw_var; - /* FIXME: this channels/modes/rates stuff sucks */ - struct ieee80211_channel channels[14]; - struct ieee80211_rate rates[12]; - struct ieee80211_hw_mode modes[2]; struct ieee80211_tx_queue_stats tx_stats; }; diff --git a/drivers/net/wireless/p54common.c b/drivers/net/wireless/p54common.c index 5cda49aff3a8..218ff7770ef6 100644 --- a/drivers/net/wireless/p54common.c +++ b/drivers/net/wireless/p54common.c @@ -27,6 +27,46 @@ MODULE_DESCRIPTION("Softmac Prism54 common code"); MODULE_LICENSE("GPL"); MODULE_ALIAS("prism54common"); +static struct ieee80211_rate p54_rates[] = { + { .bitrate = 10, .hw_value = 0, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 20, .hw_value = 1, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 55, .hw_value = 2, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 110, .hw_value = 3, .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 60, .hw_value = 4, }, + { .bitrate = 90, .hw_value = 5, }, + { .bitrate = 120, .hw_value = 6, }, + { .bitrate = 180, .hw_value = 7, }, + { .bitrate = 240, .hw_value = 8, }, + { .bitrate = 360, .hw_value = 9, }, + { .bitrate = 480, .hw_value = 10, }, + { .bitrate = 540, .hw_value = 11, }, +}; + +static struct ieee80211_channel p54_channels[] = { + { .center_freq = 2412, .hw_value = 1, }, + { .center_freq = 2417, .hw_value = 2, }, + { .center_freq = 2422, .hw_value = 3, }, + { .center_freq = 2427, .hw_value = 4, }, + { .center_freq = 2432, .hw_value = 5, }, + { .center_freq = 2437, .hw_value = 6, }, + { .center_freq = 2442, .hw_value = 7, }, + { .center_freq = 2447, .hw_value = 8, }, + { .center_freq = 2452, .hw_value = 9, }, + { .center_freq = 2457, .hw_value = 10, }, + { .center_freq = 2462, .hw_value = 11, }, + { .center_freq = 2467, .hw_value = 12, }, + { .center_freq = 2472, .hw_value = 13, }, + { .center_freq = 2484, .hw_value = 14, }, +}; + +struct ieee80211_supported_band band_2GHz = { + .channels = p54_channels, + .n_channels = ARRAY_SIZE(p54_channels), + .bitrates = p54_rates, + .n_bitrates = ARRAY_SIZE(p54_rates), +}; + + void p54_parse_firmware(struct ieee80211_hw *dev, const struct firmware *fw) { struct p54_common *priv = dev->priv; @@ -308,10 +348,10 @@ static void p54_rx_data(struct ieee80211_hw *dev, struct sk_buff *skb) u16 freq = le16_to_cpu(hdr->freq); rx_status.ssi = hdr->rssi; - rx_status.rate = hdr->rate & 0x1f; /* report short preambles & CCK too */ - rx_status.channel = freq == 2484 ? 14 : (freq - 2407)/5; + /* XX correct? */ + rx_status.rate_idx = hdr->rate & 0xf; rx_status.freq = freq; - rx_status.phymode = MODE_IEEE80211G; + rx_status.band = IEEE80211_BAND_2GHZ; rx_status.antenna = hdr->antenna; rx_status.mactime = le64_to_cpu(hdr->timestamp); rx_status.flag |= RX_FLAG_TSFT; @@ -547,7 +587,9 @@ static int p54_tx(struct ieee80211_hw *dev, struct sk_buff *skb, txhdr->padding2 = 0; /* TODO: add support for alternate retry TX rates */ - rate = control->tx_rate; + rate = control->tx_rate->hw_value; + if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE) + rate |= 0x10; if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) rate |= 0x40; else if (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) @@ -849,7 +891,7 @@ static int p54_config(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { int ret; - ret = p54_set_freq(dev, cpu_to_le16(conf->freq)); + ret = p54_set_freq(dev, cpu_to_le16(conf->channel->center_freq)); p54_set_vdcf(dev); return ret; } @@ -944,7 +986,6 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len) { struct ieee80211_hw *dev; struct p54_common *priv; - int i; dev = ieee80211_alloc_hw(priv_data_len, &p54_ops); if (!dev) @@ -953,18 +994,7 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len) priv = dev->priv; priv->mode = IEEE80211_IF_TYPE_INVALID; skb_queue_head_init(&priv->tx_queue); - memcpy(priv->channels, p54_channels, sizeof(p54_channels)); - memcpy(priv->rates, p54_rates, sizeof(p54_rates)); - priv->modes[1].mode = MODE_IEEE80211B; - priv->modes[1].num_rates = 4; - priv->modes[1].rates = priv->rates; - priv->modes[1].num_channels = ARRAY_SIZE(p54_channels); - priv->modes[1].channels = priv->channels; - priv->modes[0].mode = MODE_IEEE80211G; - priv->modes[0].num_rates = ARRAY_SIZE(p54_rates); - priv->modes[0].rates = priv->rates; - priv->modes[0].num_channels = ARRAY_SIZE(p54_channels); - priv->modes[0].channels = priv->channels; + dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &band_2GHz; dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING | /* not sure */ IEEE80211_HW_RX_INCLUDES_FCS; dev->channel_change_time = 1000; /* TODO: find actual value */ @@ -986,14 +1016,6 @@ struct ieee80211_hw *p54_init_common(size_t priv_data_len) p54_init_vdcf(dev); - for (i = 0; i < 2; i++) { - if (ieee80211_register_hwmode(dev, &priv->modes[i])) { - kfree(priv->cached_vdcf); - ieee80211_free_hw(dev); - return NULL; - } - } - return dev; } EXPORT_SYMBOL_GPL(p54_init_common); diff --git a/drivers/net/wireless/p54common.h b/drivers/net/wireless/p54common.h index a721334e20d9..dc9f4cef585e 100644 --- a/drivers/net/wireless/p54common.h +++ b/drivers/net/wireless/p54common.h @@ -251,79 +251,4 @@ struct p54_tx_control_vdcf { __le16 frameburst; } __attribute__ ((packed)); -static const struct ieee80211_rate p54_rates[] = { - { .rate = 10, - .val = 0, - .val2 = 0x10, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 20, - .val = 1, - .val2 = 0x11, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 55, - .val = 2, - .val2 = 0x12, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 110, - .val = 3, - .val2 = 0x13, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 60, - .val = 4, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 90, - .val = 5, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 120, - .val = 6, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 180, - .val = 7, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 240, - .val = 8, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 360, - .val = 9, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 480, - .val = 10, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 540, - .val = 11, - .flags = IEEE80211_RATE_OFDM }, -}; - -// TODO: just generate this.. -static const struct ieee80211_channel p54_channels[] = { - { .chan = 1, - .freq = 2412}, - { .chan = 2, - .freq = 2417}, - { .chan = 3, - .freq = 2422}, - { .chan = 4, - .freq = 2427}, - { .chan = 5, - .freq = 2432}, - { .chan = 6, - .freq = 2437}, - { .chan = 7, - .freq = 2442}, - { .chan = 8, - .freq = 2447}, - { .chan = 9, - .freq = 2452}, - { .chan = 10, - .freq = 2457}, - { .chan = 11, - .freq = 2462}, - { .chan = 12, - .freq = 2467}, - { .chan = 13, - .freq = 2472}, - { .chan = 14, - .freq = 2484} -}; - #endif /* PRISM54COMMON_H */ diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h index b0e4ea7c9dca..4fa762bdb734 100644 --- a/drivers/net/wireless/rt2x00/rt2x00.h +++ b/drivers/net/wireless/rt2x00/rt2x00.h @@ -390,6 +390,10 @@ static inline struct rt2x00_intf* vif_to_intf(struct ieee80211_vif *vif) return (struct rt2x00_intf *)vif->drv_priv; } +#define HWMODE_B 0 +#define HWMODE_G 1 +#define HWMODE_A 2 + /* * Details about the supported modes, rates and channels * of a particular chipset. This is used by rt2x00lib @@ -644,11 +648,8 @@ struct rt2x00_dev { * IEEE80211 control structure. */ struct ieee80211_hw *hw; - struct ieee80211_hw_mode *hwmodes; - unsigned int curr_hwmode; -#define HWMODE_B 0 -#define HWMODE_G 1 -#define HWMODE_A 2 + struct ieee80211_supported_band bands[IEEE80211_NUM_BANDS]; + enum ieee80211_band curr_band; /* * rfkill structure for RF state switching support. diff --git a/drivers/net/wireless/rt2x00/rt2x00config.c b/drivers/net/wireless/rt2x00/rt2x00config.c index 20231e0c53fa..9fba485a40ac 100644 --- a/drivers/net/wireless/rt2x00/rt2x00config.c +++ b/drivers/net/wireless/rt2x00/rt2x00config.c @@ -152,7 +152,7 @@ void rt2x00lib_config(struct rt2x00_dev *rt2x00dev, struct ieee80211_conf *conf, const int force_config) { struct rt2x00lib_conf libconf; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *band; struct ieee80211_rate *rate; struct antenna_setup *default_ant = &rt2x00dev->default_ant; struct antenna_setup *active_ant = &rt2x00dev->link.ant.active; @@ -172,9 +172,9 @@ void rt2x00lib_config(struct rt2x00_dev *rt2x00dev, * Check which configuration options have been * updated and should be send to the device. */ - if (rt2x00dev->rx_status.phymode != conf->phymode) + if (rt2x00dev->rx_status.band != conf->channel->band) flags |= CONFIG_UPDATE_PHYMODE; - if (rt2x00dev->rx_status.channel != conf->channel) + if (rt2x00dev->rx_status.freq != conf->channel->center_freq) flags |= CONFIG_UPDATE_CHANNEL; if (rt2x00dev->tx_power != conf->power_level) flags |= CONFIG_UPDATE_TXPOWER; @@ -229,33 +229,31 @@ config: memset(&libconf, 0, sizeof(libconf)); if (flags & CONFIG_UPDATE_PHYMODE) { - switch (conf->phymode) { - case MODE_IEEE80211A: + switch (conf->channel->band) { + case IEEE80211_BAND_5GHZ: libconf.phymode = HWMODE_A; break; - case MODE_IEEE80211B: - libconf.phymode = HWMODE_B; - break; - case MODE_IEEE80211G: + case IEEE80211_BAND_2GHZ: + /* Uh oh. what about B? */ libconf.phymode = HWMODE_G; break; default: ERROR(rt2x00dev, "Attempt to configure unsupported mode (%d)" - "Defaulting to 802.11b", conf->phymode); + "Defaulting to 802.11b", conf->channel->band); libconf.phymode = HWMODE_B; } - mode = &rt2x00dev->hwmodes[libconf.phymode]; - rate = &mode->rates[mode->num_rates - 1]; + band = &rt2x00dev->bands[conf->channel->band]; + rate = &band->bitrates[band->n_bitrates - 1]; libconf.basic_rates = - DEVICE_GET_RATE_FIELD(rate->val, RATEMASK) & DEV_BASIC_RATEMASK; + DEVICE_GET_RATE_FIELD(rate->hw_value, RATEMASK) & DEV_BASIC_RATEMASK; } if (flags & CONFIG_UPDATE_CHANNEL) { memcpy(&libconf.rf, - &rt2x00dev->spec.channels[conf->channel_val], + &rt2x00dev->spec.channels[conf->channel->hw_value], sizeof(libconf.rf)); } @@ -301,12 +299,11 @@ config: rt2x00lib_reset_link_tuner(rt2x00dev); if (flags & CONFIG_UPDATE_PHYMODE) { - rt2x00dev->curr_hwmode = libconf.phymode; - rt2x00dev->rx_status.phymode = conf->phymode; + rt2x00dev->curr_band = conf->channel->band; + rt2x00dev->rx_status.band = conf->channel->band; } - rt2x00dev->rx_status.freq = conf->freq; - rt2x00dev->rx_status.channel = conf->channel; + rt2x00dev->rx_status.freq = conf->channel->center_freq; rt2x00dev->tx_power = conf->power_level; if (flags & CONFIG_UPDATE_ANTENNA) { diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c index 0df8062b1a8e..83a72ae36638 100644 --- a/drivers/net/wireless/rt2x00/rt2x00dev.c +++ b/drivers/net/wireless/rt2x00/rt2x00dev.c @@ -550,19 +550,19 @@ void rt2x00lib_rxdone(struct queue_entry *entry, { struct rt2x00_dev *rt2x00dev = entry->queue->rt2x00dev; struct ieee80211_rx_status *rx_status = &rt2x00dev->rx_status; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; struct ieee80211_rate *rate; struct ieee80211_hdr *hdr; unsigned int i; - int val = 0; + int val = 0, idx = -1; u16 fc; /* * Update RX statistics. */ - mode = &rt2x00dev->hwmodes[rt2x00dev->curr_hwmode]; - for (i = 0; i < mode->num_rates; i++) { - rate = &mode->rates[i]; + sband = &rt2x00dev->bands[rt2x00dev->curr_band]; + for (i = 0; i < sband->n_bitrates; i++) { + rate = &sband->bitrates[i]; /* * When frame was received with an OFDM bitrate, @@ -570,12 +570,12 @@ void rt2x00lib_rxdone(struct queue_entry *entry, * a CCK bitrate the signal is the rate in 0.5kbit/s. */ if (!rxdesc->ofdm) - val = DEVICE_GET_RATE_FIELD(rate->val, RATE); + val = DEVICE_GET_RATE_FIELD(rate->hw_value, RATE); else - val = DEVICE_GET_RATE_FIELD(rate->val, PLCP); + val = DEVICE_GET_RATE_FIELD(rate->hw_value, PLCP); if (val == rxdesc->signal) { - val = rate->val; + idx = i; break; } } @@ -590,7 +590,7 @@ void rt2x00lib_rxdone(struct queue_entry *entry, rt2x00dev->link.qual.rx_success++; - rx_status->rate = val; + rx_status->rate_idx = idx; rx_status->signal = rt2x00lib_calculate_link_signal(rt2x00dev, rxdesc->rssi); rx_status->ssi = rxdesc->rssi; @@ -639,7 +639,7 @@ void rt2x00lib_write_tx_desc(struct rt2x00_dev *rt2x00dev, frame_control = le16_to_cpu(ieee80211hdr->frame_control); seq_ctrl = le16_to_cpu(ieee80211hdr->seq_ctrl); - tx_rate = control->tx_rate; + tx_rate = control->tx_rate->hw_value; /* * Check whether this frame is to be acked @@ -658,7 +658,7 @@ void rt2x00lib_write_tx_desc(struct rt2x00_dev *rt2x00dev, } else __clear_bit(ENTRY_TXD_ACK, &txdesc.flags); if (control->rts_cts_rate) - tx_rate = control->rts_cts_rate; + tx_rate = control->rts_cts_rate->hw_value; } /* @@ -760,54 +760,45 @@ static void rt2x00lib_channel(struct ieee80211_channel *entry, const int channel, const int tx_power, const int value) { - entry->chan = channel; if (channel <= 14) - entry->freq = 2407 + (5 * channel); + entry->center_freq = 2407 + (5 * channel); else - entry->freq = 5000 + (5 * channel); - entry->val = value; - entry->flag = - IEEE80211_CHAN_W_IBSS | - IEEE80211_CHAN_W_ACTIVE_SCAN | - IEEE80211_CHAN_W_SCAN; - entry->power_level = tx_power; - entry->antenna_max = 0xff; + entry->center_freq = 5000 + (5 * channel); + entry->hw_value = value; + entry->max_power = tx_power; + entry->max_antenna_gain = 0xff; } static void rt2x00lib_rate(struct ieee80211_rate *entry, const int rate, const int mask, const int plcp, const int flags) { - entry->rate = rate; - entry->val = + entry->bitrate = rate; + entry->hw_value = DEVICE_SET_RATE_FIELD(rate, RATE) | DEVICE_SET_RATE_FIELD(mask, RATEMASK) | DEVICE_SET_RATE_FIELD(plcp, PLCP); entry->flags = flags; - entry->val2 = entry->val; - if (entry->flags & IEEE80211_RATE_PREAMBLE2) - entry->val2 |= DEVICE_SET_RATE_FIELD(1, PREAMBLE); - entry->min_rssi_ack = 0; - entry->min_rssi_ack_delta = 0; + entry->hw_value_short = entry->hw_value; + if (entry->flags & IEEE80211_RATE_SHORT_PREAMBLE) + entry->hw_value_short |= DEVICE_SET_RATE_FIELD(1, PREAMBLE); } static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev, struct hw_mode_spec *spec) { struct ieee80211_hw *hw = rt2x00dev->hw; - struct ieee80211_hw_mode *hwmodes; + struct ieee80211_supported_band *sbands; struct ieee80211_channel *channels; struct ieee80211_rate *rates; unsigned int i; unsigned char tx_power; - hwmodes = kzalloc(sizeof(*hwmodes) * spec->num_modes, GFP_KERNEL); - if (!hwmodes) - goto exit; + sbands = &rt2x00dev->bands[0]; channels = kzalloc(sizeof(*channels) * spec->num_channels, GFP_KERNEL); if (!channels) - goto exit_free_modes; + return -ENOMEM; rates = kzalloc(sizeof(*rates) * spec->num_rates, GFP_KERNEL); if (!rates) @@ -817,31 +808,31 @@ static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev, * Initialize Rate list. */ rt2x00lib_rate(&rates[0], 10, DEV_RATEMASK_1MB, - 0x00, IEEE80211_RATE_CCK); + 0x00, 0); rt2x00lib_rate(&rates[1], 20, DEV_RATEMASK_2MB, - 0x01, IEEE80211_RATE_CCK_2); + 0x01, IEEE80211_RATE_SHORT_PREAMBLE); rt2x00lib_rate(&rates[2], 55, DEV_RATEMASK_5_5MB, - 0x02, IEEE80211_RATE_CCK_2); + 0x02, IEEE80211_RATE_SHORT_PREAMBLE); rt2x00lib_rate(&rates[3], 110, DEV_RATEMASK_11MB, - 0x03, IEEE80211_RATE_CCK_2); + 0x03, IEEE80211_RATE_SHORT_PREAMBLE); if (spec->num_rates > 4) { rt2x00lib_rate(&rates[4], 60, DEV_RATEMASK_6MB, - 0x0b, IEEE80211_RATE_OFDM); + 0x0b, 0); rt2x00lib_rate(&rates[5], 90, DEV_RATEMASK_9MB, - 0x0f, IEEE80211_RATE_OFDM); + 0x0f, 0); rt2x00lib_rate(&rates[6], 120, DEV_RATEMASK_12MB, - 0x0a, IEEE80211_RATE_OFDM); + 0x0a, 0); rt2x00lib_rate(&rates[7], 180, DEV_RATEMASK_18MB, - 0x0e, IEEE80211_RATE_OFDM); + 0x0e, 0); rt2x00lib_rate(&rates[8], 240, DEV_RATEMASK_24MB, - 0x09, IEEE80211_RATE_OFDM); + 0x09, 0); rt2x00lib_rate(&rates[9], 360, DEV_RATEMASK_36MB, - 0x0d, IEEE80211_RATE_OFDM); + 0x0d, 0); rt2x00lib_rate(&rates[10], 480, DEV_RATEMASK_48MB, - 0x08, IEEE80211_RATE_OFDM); + 0x08, 0); rt2x00lib_rate(&rates[11], 540, DEV_RATEMASK_54MB, - 0x0c, IEEE80211_RATE_OFDM); + 0x0c, 0); } /* @@ -862,27 +853,27 @@ static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev, /* * Intitialize 802.11b * Rates: CCK. - * Channels: OFDM. + * Channels: 2.4 GHz */ if (spec->num_modes > HWMODE_B) { - hwmodes[HWMODE_B].mode = MODE_IEEE80211B; - hwmodes[HWMODE_B].num_channels = 14; - hwmodes[HWMODE_B].num_rates = 4; - hwmodes[HWMODE_B].channels = channels; - hwmodes[HWMODE_B].rates = rates; + sbands[IEEE80211_BAND_2GHZ].n_channels = 14; + sbands[IEEE80211_BAND_2GHZ].n_bitrates = 4; + sbands[IEEE80211_BAND_2GHZ].channels = channels; + sbands[IEEE80211_BAND_2GHZ].bitrates = rates; + hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &rt2x00dev->bands[IEEE80211_BAND_2GHZ]; } /* * Intitialize 802.11g * Rates: CCK, OFDM. - * Channels: OFDM. + * Channels: 2.4 GHz */ if (spec->num_modes > HWMODE_G) { - hwmodes[HWMODE_G].mode = MODE_IEEE80211G; - hwmodes[HWMODE_G].num_channels = 14; - hwmodes[HWMODE_G].num_rates = spec->num_rates; - hwmodes[HWMODE_G].channels = channels; - hwmodes[HWMODE_G].rates = rates; + sbands[IEEE80211_BAND_2GHZ].n_channels = 14; + sbands[IEEE80211_BAND_2GHZ].n_bitrates = spec->num_rates; + sbands[IEEE80211_BAND_2GHZ].channels = channels; + sbands[IEEE80211_BAND_2GHZ].bitrates = rates; + hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &rt2x00dev->bands[IEEE80211_BAND_2GHZ]; } /* @@ -891,39 +882,17 @@ static int rt2x00lib_probe_hw_modes(struct rt2x00_dev *rt2x00dev, * Channels: OFDM, UNII, HiperLAN2. */ if (spec->num_modes > HWMODE_A) { - hwmodes[HWMODE_A].mode = MODE_IEEE80211A; - hwmodes[HWMODE_A].num_channels = spec->num_channels - 14; - hwmodes[HWMODE_A].num_rates = spec->num_rates - 4; - hwmodes[HWMODE_A].channels = &channels[14]; - hwmodes[HWMODE_A].rates = &rates[4]; + sbands[IEEE80211_BAND_5GHZ].n_channels = spec->num_channels - 14; + sbands[IEEE80211_BAND_5GHZ].n_bitrates = spec->num_rates - 4; + sbands[IEEE80211_BAND_5GHZ].channels = &channels[14]; + sbands[IEEE80211_BAND_5GHZ].bitrates = &rates[4]; + hw->wiphy->bands[IEEE80211_BAND_5GHZ] = &rt2x00dev->bands[IEEE80211_BAND_5GHZ]; } - if (spec->num_modes > HWMODE_G && - ieee80211_register_hwmode(hw, &hwmodes[HWMODE_G])) - goto exit_free_rates; - - if (spec->num_modes > HWMODE_B && - ieee80211_register_hwmode(hw, &hwmodes[HWMODE_B])) - goto exit_free_rates; - - if (spec->num_modes > HWMODE_A && - ieee80211_register_hwmode(hw, &hwmodes[HWMODE_A])) - goto exit_free_rates; - - rt2x00dev->hwmodes = hwmodes; - return 0; -exit_free_rates: - kfree(rates); - -exit_free_channels: + exit_free_channels: kfree(channels); - -exit_free_modes: - kfree(hwmodes); - -exit: ERROR(rt2x00dev, "Allocation ieee80211 modes failed.\n"); return -ENOMEM; } @@ -933,11 +902,11 @@ static void rt2x00lib_remove_hw(struct rt2x00_dev *rt2x00dev) if (test_bit(DEVICE_REGISTERED_HW, &rt2x00dev->flags)) ieee80211_unregister_hw(rt2x00dev->hw); - if (likely(rt2x00dev->hwmodes)) { - kfree(rt2x00dev->hwmodes->channels); - kfree(rt2x00dev->hwmodes->rates); - kfree(rt2x00dev->hwmodes); - rt2x00dev->hwmodes = NULL; + if (likely(rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ])) { + kfree(rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ]->channels); + kfree(rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ]->bitrates); + rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_2GHZ] = NULL; + rt2x00dev->hw->wiphy->bands[IEEE80211_BAND_5GHZ] = NULL; } } diff --git a/drivers/net/wireless/rt2x00/rt61pci.c b/drivers/net/wireless/rt2x00/rt61pci.c index 59e87a1d96a4..1dd30510ed1e 100644 --- a/drivers/net/wireless/rt2x00/rt61pci.c +++ b/drivers/net/wireless/rt2x00/rt61pci.c @@ -426,12 +426,12 @@ static void rt61pci_config_antenna_5x(struct rt2x00_dev *rt2x00dev, case ANTENNA_HW_DIVERSITY: rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 2); rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, - (rt2x00dev->curr_hwmode != HWMODE_A)); + (rt2x00dev->curr_band != IEEE80211_BAND_5GHZ)); break; case ANTENNA_A: rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1); rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0); - if (rt2x00dev->curr_hwmode == HWMODE_A) + if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0); else rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3); @@ -446,7 +446,7 @@ static void rt61pci_config_antenna_5x(struct rt2x00_dev *rt2x00dev, case ANTENNA_B: rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1); rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0); - if (rt2x00dev->curr_hwmode == HWMODE_A) + if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3); else rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0); @@ -602,7 +602,7 @@ static void rt61pci_config_antenna(struct rt2x00_dev *rt2x00dev, unsigned int i; u32 reg; - if (rt2x00dev->curr_hwmode == HWMODE_A) { + if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) { sel = antenna_sel_a; lna = test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags); } else { @@ -616,10 +616,9 @@ static void rt61pci_config_antenna(struct rt2x00_dev *rt2x00dev, rt2x00pci_register_read(rt2x00dev, PHY_CSR0, ®); rt2x00_set_field32(®, PHY_CSR0_PA_PE_BG, - (rt2x00dev->curr_hwmode == HWMODE_B || - rt2x00dev->curr_hwmode == HWMODE_G)); + rt2x00dev->curr_band == IEEE80211_BAND_2GHZ); rt2x00_set_field32(®, PHY_CSR0_PA_PE_A, - (rt2x00dev->curr_hwmode == HWMODE_A)); + rt2x00dev->curr_band == IEEE80211_BAND_5GHZ); rt2x00pci_register_write(rt2x00dev, PHY_CSR0, reg); @@ -698,9 +697,9 @@ static void rt61pci_enable_led(struct rt2x00_dev *rt2x00dev) rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_RADIO_STATUS, 1); rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_A_STATUS, - (rt2x00dev->rx_status.phymode == MODE_IEEE80211A)); + rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ); rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_BG_STATUS, - (rt2x00dev->rx_status.phymode != MODE_IEEE80211A)); + rt2x00dev->rx_status.band != IEEE80211_BAND_5GHZ); arg0 = rt2x00dev->led_reg & 0xff; arg1 = (rt2x00dev->led_reg >> 8) & 0xff; @@ -798,7 +797,7 @@ static void rt61pci_link_tuner(struct rt2x00_dev *rt2x00dev) /* * Determine r17 bounds. */ - if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) { + if (rt2x00dev->rx_status.band == IEEE80211_BAND_2GHZ) { low_bound = 0x28; up_bound = 0x48; if (test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags)) { @@ -1544,8 +1543,10 @@ static void rt61pci_write_tx_desc(struct rt2x00_dev *rt2x00dev, rt2x00_desc_write(txd, 2, word); rt2x00_desc_read(txd, 5, &word); +/* XXX: removed for now rt2x00_set_field32(&word, TXD_W5_TX_POWER, TXPOWER_TO_DEV(control->power_level)); + */ rt2x00_set_field32(&word, TXD_W5_WAITING_DMA_DONE_INT, 1); rt2x00_desc_write(txd, 5, word); @@ -1637,7 +1638,7 @@ static int rt61pci_agc_to_rssi(struct rt2x00_dev *rt2x00dev, int rxd_w1) return 0; } - if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) { + if (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ) { if (test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags)) offset += 14; diff --git a/drivers/net/wireless/rt2x00/rt73usb.c b/drivers/net/wireless/rt2x00/rt73usb.c index 4b5bde8b53de..9cbc879da037 100644 --- a/drivers/net/wireless/rt2x00/rt73usb.c +++ b/drivers/net/wireless/rt2x00/rt73usb.c @@ -439,13 +439,13 @@ static void rt73usb_config_antenna_5x(struct rt2x00_dev *rt2x00dev, case ANTENNA_HW_DIVERSITY: rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 2); temp = !test_bit(CONFIG_FRAME_TYPE, &rt2x00dev->flags) - && (rt2x00dev->curr_hwmode != HWMODE_A); + && (rt2x00dev->curr_band != IEEE80211_BAND_5GHZ); rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, temp); break; case ANTENNA_A: rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1); rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0); - if (rt2x00dev->curr_hwmode == HWMODE_A) + if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0); else rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3); @@ -460,7 +460,7 @@ static void rt73usb_config_antenna_5x(struct rt2x00_dev *rt2x00dev, case ANTENNA_B: rt2x00_set_field8(&r4, BBP_R4_RX_ANTENNA_CONTROL, 1); rt2x00_set_field8(&r4, BBP_R4_RX_FRAME_END, 0); - if (rt2x00dev->curr_hwmode == HWMODE_A) + if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 3); else rt2x00_set_field8(&r77, BBP_R77_RX_ANTENNA, 0); @@ -555,7 +555,7 @@ static void rt73usb_config_antenna(struct rt2x00_dev *rt2x00dev, unsigned int i; u32 reg; - if (rt2x00dev->curr_hwmode == HWMODE_A) { + if (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ) { sel = antenna_sel_a; lna = test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags); } else { @@ -569,10 +569,9 @@ static void rt73usb_config_antenna(struct rt2x00_dev *rt2x00dev, rt73usb_register_read(rt2x00dev, PHY_CSR0, ®); rt2x00_set_field32(®, PHY_CSR0_PA_PE_BG, - (rt2x00dev->curr_hwmode == HWMODE_B || - rt2x00dev->curr_hwmode == HWMODE_G)); + (rt2x00dev->curr_band == IEEE80211_BAND_2GHZ)); rt2x00_set_field32(®, PHY_CSR0_PA_PE_A, - (rt2x00dev->curr_hwmode == HWMODE_A)); + (rt2x00dev->curr_band == IEEE80211_BAND_5GHZ)); rt73usb_register_write(rt2x00dev, PHY_CSR0, reg); @@ -644,9 +643,9 @@ static void rt73usb_enable_led(struct rt2x00_dev *rt2x00dev) rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_RADIO_STATUS, 1); rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_A_STATUS, - (rt2x00dev->rx_status.phymode == MODE_IEEE80211A)); + (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ)); rt2x00_set_field16(&rt2x00dev->led_reg, MCU_LEDCS_LINK_BG_STATUS, - (rt2x00dev->rx_status.phymode != MODE_IEEE80211A)); + (rt2x00dev->rx_status.band != IEEE80211_BAND_5GHZ)); rt2x00usb_vendor_request_sw(rt2x00dev, USB_LED_CONTROL, 0x0000, rt2x00dev->led_reg, REGISTER_TIMEOUT); @@ -736,7 +735,7 @@ static void rt73usb_link_tuner(struct rt2x00_dev *rt2x00dev) /* * Determine r17 bounds. */ - if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) { + if (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ) { low_bound = 0x28; up_bound = 0x48; @@ -1278,8 +1277,10 @@ static void rt73usb_write_tx_desc(struct rt2x00_dev *rt2x00dev, rt2x00_desc_write(txd, 2, word); rt2x00_desc_read(txd, 5, &word); +/* XXX: removed for now rt2x00_set_field32(&word, TXD_W5_TX_POWER, TXPOWER_TO_DEV(control->power_level)); + */ rt2x00_set_field32(&word, TXD_W5_WAITING_DMA_DONE_INT, 1); rt2x00_desc_write(txd, 5, word); @@ -1370,7 +1371,7 @@ static int rt73usb_agc_to_rssi(struct rt2x00_dev *rt2x00dev, int rxd_w1) return 0; } - if (rt2x00dev->rx_status.phymode == MODE_IEEE80211A) { + if (rt2x00dev->rx_status.band == IEEE80211_BAND_5GHZ) { if (test_bit(CONFIG_EXTERNAL_LNA_A, &rt2x00dev->flags)) { if (lna == 3 || lna == 2) offset += 10; diff --git a/drivers/net/wireless/rtl8180.h b/drivers/net/wireless/rtl8180.h index 2cbfe3c8081f..082a11f93beb 100644 --- a/drivers/net/wireless/rtl8180.h +++ b/drivers/net/wireless/rtl8180.h @@ -102,7 +102,7 @@ struct rtl8180_priv { struct rtl8180_tx_ring tx_ring[4]; struct ieee80211_channel channels[14]; struct ieee80211_rate rates[12]; - struct ieee80211_hw_mode modes[2]; + struct ieee80211_supported_band band; struct pci_dev *pdev; u32 rx_conf; diff --git a/drivers/net/wireless/rtl8180_dev.c b/drivers/net/wireless/rtl8180_dev.c index 5e9a8ace0d81..d0928c91c21c 100644 --- a/drivers/net/wireless/rtl8180_dev.c +++ b/drivers/net/wireless/rtl8180_dev.c @@ -49,6 +49,41 @@ static struct pci_device_id rtl8180_table[] __devinitdata = { MODULE_DEVICE_TABLE(pci, rtl8180_table); +static const struct ieee80211_rate rtl818x_rates[] = { + { .bitrate = 10, .hw_value = 0, }, + { .bitrate = 20, .hw_value = 1, }, + { .bitrate = 55, .hw_value = 2, }, + { .bitrate = 110, .hw_value = 3, }, + { .bitrate = 60, .hw_value = 4, }, + { .bitrate = 90, .hw_value = 5, }, + { .bitrate = 120, .hw_value = 6, }, + { .bitrate = 180, .hw_value = 7, }, + { .bitrate = 240, .hw_value = 8, }, + { .bitrate = 360, .hw_value = 9, }, + { .bitrate = 480, .hw_value = 10, }, + { .bitrate = 540, .hw_value = 11, }, +}; + +static const struct ieee80211_channel rtl818x_channels[] = { + { .center_freq = 2412 }, + { .center_freq = 2417 }, + { .center_freq = 2422 }, + { .center_freq = 2427 }, + { .center_freq = 2432 }, + { .center_freq = 2437 }, + { .center_freq = 2442 }, + { .center_freq = 2447 }, + { .center_freq = 2452 }, + { .center_freq = 2457 }, + { .center_freq = 2462 }, + { .center_freq = 2467 }, + { .center_freq = 2472 }, + { .center_freq = 2484 }, +}; + + + + void rtl8180_write_phy(struct ieee80211_hw *dev, u8 addr, u32 data) { struct rtl8180_priv *priv = dev->priv; @@ -99,10 +134,10 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev) /* TODO: improve signal/rssi reporting */ rx_status.signal = flags2 & 0xFF; rx_status.ssi = (flags2 >> 8) & 0x7F; - rx_status.rate = (flags >> 20) & 0xF; - rx_status.freq = dev->conf.freq; - rx_status.channel = dev->conf.channel; - rx_status.phymode = dev->conf.phymode; + /* XXX: is this correct? */ + rx_status.rate_idx = (flags >> 20) & 0xF; + rx_status.freq = dev->conf.channel->center_freq; + rx_status.band = dev->conf.channel->band; rx_status.mactime = le64_to_cpu(entry->tsft); rx_status.flag |= RX_FLAG_TSFT; if (flags & RTL8180_RX_DESC_FLAG_CRC32_ERR) @@ -223,8 +258,9 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb, skb->len, PCI_DMA_TODEVICE); tx_flags = RTL8180_TX_DESC_FLAG_OWN | RTL8180_TX_DESC_FLAG_FS | - RTL8180_TX_DESC_FLAG_LS | (control->tx_rate << 24) | - (control->rts_cts_rate << 19) | skb->len; + RTL8180_TX_DESC_FLAG_LS | + (control->tx_rate->hw_value << 24) | + (control->rts_cts_rate->hw_value << 19) | skb->len; if (priv->r8185) tx_flags |= RTL8180_TX_DESC_FLAG_DMA | @@ -246,9 +282,9 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb, unsigned int remainder; plcp_len = DIV_ROUND_UP(16 * (skb->len + 4), - (control->rate->rate * 2) / 10); + (control->tx_rate->bitrate * 2) / 10); remainder = (16 * (skb->len + 4)) % - ((control->rate->rate * 2) / 10); + ((control->tx_rate->bitrate * 2) / 10); if (remainder > 0 && remainder <= 6) plcp_len |= 1 << 15; } @@ -261,8 +297,8 @@ static int rtl8180_tx(struct ieee80211_hw *dev, struct sk_buff *skb, entry->plcp_len = cpu_to_le16(plcp_len); entry->tx_buf = cpu_to_le32(mapping); entry->frame_len = cpu_to_le32(skb->len); - entry->flags2 = control->alt_retry_rate != -1 ? - control->alt_retry_rate << 4 : 0; + entry->flags2 = control->alt_retry_rate != NULL ? + control->alt_retry_rate->bitrate << 4 : 0; entry->retry_limit = control->retry_limit; entry->flags = cpu_to_le32(tx_flags); __skb_queue_tail(&ring->queue, skb); @@ -838,19 +874,19 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev, goto err_free_dev; } + BUILD_BUG_ON(sizeof(priv->channels) != sizeof(rtl818x_channels)); + BUILD_BUG_ON(sizeof(priv->rates) != sizeof(rtl818x_rates)); + memcpy(priv->channels, rtl818x_channels, sizeof(rtl818x_channels)); memcpy(priv->rates, rtl818x_rates, sizeof(rtl818x_rates)); - priv->modes[0].mode = MODE_IEEE80211G; - priv->modes[0].num_rates = ARRAY_SIZE(rtl818x_rates); - priv->modes[0].rates = priv->rates; - priv->modes[0].num_channels = ARRAY_SIZE(rtl818x_channels); - priv->modes[0].channels = priv->channels; - priv->modes[1].mode = MODE_IEEE80211B; - priv->modes[1].num_rates = 4; - priv->modes[1].rates = priv->rates; - priv->modes[1].num_channels = ARRAY_SIZE(rtl818x_channels); - priv->modes[1].channels = priv->channels; - priv->mode = IEEE80211_IF_TYPE_INVALID; + + priv->band.band = IEEE80211_BAND_2GHZ; + priv->band.channels = priv->channels; + priv->band.n_channels = ARRAY_SIZE(rtl818x_channels); + priv->band.bitrates = priv->rates; + priv->band.n_bitrates = 4; + dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->band; + dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING | IEEE80211_HW_RX_INCLUDES_FCS; dev->queues = 1; @@ -879,15 +915,10 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev, priv->r8185 = reg & RTL818X_TX_CONF_R8185_ABC; if (priv->r8185) { - if ((err = ieee80211_register_hwmode(dev, &priv->modes[0]))) - goto err_iounmap; - + priv->band.n_bitrates = ARRAY_SIZE(rtl818x_rates); pci_try_set_mwi(pdev); } - if ((err = ieee80211_register_hwmode(dev, &priv->modes[1]))) - goto err_iounmap; - eeprom.data = dev; eeprom.register_read = rtl8180_eeprom_register_read; eeprom.register_write = rtl8180_eeprom_register_write; @@ -950,8 +981,8 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev, for (i = 0; i < 14; i += 2) { u16 txpwr; eeprom_93cx6_read(&eeprom, 0x10 + (i >> 1), &txpwr); - priv->channels[i].val = txpwr & 0xFF; - priv->channels[i + 1].val = txpwr >> 8; + priv->channels[i].hw_value = txpwr & 0xFF; + priv->channels[i + 1].hw_value = txpwr >> 8; } /* OFDM TX power */ @@ -959,8 +990,8 @@ static int __devinit rtl8180_probe(struct pci_dev *pdev, for (i = 0; i < 14; i += 2) { u16 txpwr; eeprom_93cx6_read(&eeprom, 0x20 + (i >> 1), &txpwr); - priv->channels[i].val |= (txpwr & 0xFF) << 8; - priv->channels[i + 1].val |= txpwr & 0xFF00; + priv->channels[i].hw_value |= (txpwr & 0xFF) << 8; + priv->channels[i + 1].hw_value |= txpwr & 0xFF00; } } diff --git a/drivers/net/wireless/rtl8180_grf5101.c b/drivers/net/wireless/rtl8180_grf5101.c index 8293e19c4c59..5d47935dbac3 100644 --- a/drivers/net/wireless/rtl8180_grf5101.c +++ b/drivers/net/wireless/rtl8180_grf5101.c @@ -73,8 +73,9 @@ static void grf5101_rf_set_channel(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { struct rtl8180_priv *priv = dev->priv; - u32 txpw = priv->channels[conf->channel - 1].val & 0xFF; - u32 chan = conf->channel - 1; + int channel = ieee80211_frequency_to_channel(conf->channel->center_freq); + u32 txpw = priv->channels[channel - 1].hw_value & 0xFF; + u32 chan = channel - 1; /* set TX power */ write_grf5101(dev, 0x15, 0x0); diff --git a/drivers/net/wireless/rtl8180_max2820.c b/drivers/net/wireless/rtl8180_max2820.c index 98fe9fd64968..a34dfd382b6d 100644 --- a/drivers/net/wireless/rtl8180_max2820.c +++ b/drivers/net/wireless/rtl8180_max2820.c @@ -78,8 +78,9 @@ static void max2820_rf_set_channel(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { struct rtl8180_priv *priv = dev->priv; - unsigned int chan_idx = conf ? conf->channel - 1 : 0; - u32 txpw = priv->channels[chan_idx].val & 0xFF; + int channel = ieee80211_frequency_to_channel(conf->channel->center_freq); + unsigned int chan_idx = channel - 1; + u32 txpw = priv->channels[chan_idx].hw_value & 0xFF; u32 chan = max2820_chan[chan_idx]; /* While philips SA2400 drive the PA bias from diff --git a/drivers/net/wireless/rtl8180_rtl8225.c b/drivers/net/wireless/rtl8180_rtl8225.c index ef3832bee85c..cd22781728a9 100644 --- a/drivers/net/wireless/rtl8180_rtl8225.c +++ b/drivers/net/wireless/rtl8180_rtl8225.c @@ -261,8 +261,8 @@ static void rtl8225_rf_set_tx_power(struct ieee80211_hw *dev, int channel) u32 reg; int i; - cck_power = priv->channels[channel - 1].val & 0xFF; - ofdm_power = priv->channels[channel - 1].val >> 8; + cck_power = priv->channels[channel - 1].hw_value & 0xFF; + ofdm_power = priv->channels[channel - 1].hw_value >> 8; cck_power = min(cck_power, (u8)35); ofdm_power = min(ofdm_power, (u8)35); @@ -476,8 +476,8 @@ static void rtl8225z2_rf_set_tx_power(struct ieee80211_hw *dev, int channel) const u8 *tmp; int i; - cck_power = priv->channels[channel - 1].val & 0xFF; - ofdm_power = priv->channels[channel - 1].val >> 8; + cck_power = priv->channels[channel - 1].hw_value & 0xFF; + ofdm_power = priv->channels[channel - 1].hw_value >> 8; if (channel == 14) tmp = rtl8225z2_tx_power_cck_ch14; @@ -716,13 +716,14 @@ static void rtl8225_rf_set_channel(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { struct rtl8180_priv *priv = dev->priv; + int chan = ieee80211_frequency_to_channel(conf->channel->center_freq); if (priv->rf->init == rtl8225_rf_init) - rtl8225_rf_set_tx_power(dev, conf->channel); + rtl8225_rf_set_tx_power(dev, chan); else - rtl8225z2_rf_set_tx_power(dev, conf->channel); + rtl8225z2_rf_set_tx_power(dev, chan); - rtl8225_write(dev, 0x7, rtl8225_chan[conf->channel - 1]); + rtl8225_write(dev, 0x7, rtl8225_chan[chan - 1]); msleep(10); if (conf->flags & IEEE80211_CONF_SHORT_SLOT_TIME) { diff --git a/drivers/net/wireless/rtl8180_sa2400.c b/drivers/net/wireless/rtl8180_sa2400.c index e08ace7b1cb7..0311b4ea124c 100644 --- a/drivers/net/wireless/rtl8180_sa2400.c +++ b/drivers/net/wireless/rtl8180_sa2400.c @@ -80,8 +80,9 @@ static void sa2400_rf_set_channel(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { struct rtl8180_priv *priv = dev->priv; - u32 txpw = priv->channels[conf->channel - 1].val & 0xFF; - u32 chan = sa2400_chan[conf->channel - 1]; + int channel = ieee80211_frequency_to_channel(conf->channel->center_freq); + u32 txpw = priv->channels[channel - 1].hw_value & 0xFF; + u32 chan = sa2400_chan[channel - 1]; write_sa2400(dev, 7, txpw); diff --git a/drivers/net/wireless/rtl8187.h b/drivers/net/wireless/rtl8187.h index 8680a0b6433c..076d88b6db0e 100644 --- a/drivers/net/wireless/rtl8187.h +++ b/drivers/net/wireless/rtl8187.h @@ -71,7 +71,7 @@ struct rtl8187_priv { /* rtl8187 specific */ struct ieee80211_channel channels[14]; struct ieee80211_rate rates[12]; - struct ieee80211_hw_mode modes[2]; + struct ieee80211_supported_band band; struct usb_device *udev; u32 rx_conf; u16 txpwr_base; diff --git a/drivers/net/wireless/rtl8187_dev.c b/drivers/net/wireless/rtl8187_dev.c index f44505994a0e..6ef67990692d 100644 --- a/drivers/net/wireless/rtl8187_dev.c +++ b/drivers/net/wireless/rtl8187_dev.c @@ -45,6 +45,38 @@ static struct usb_device_id rtl8187_table[] __devinitdata = { MODULE_DEVICE_TABLE(usb, rtl8187_table); +static const struct ieee80211_rate rtl818x_rates[] = { + { .bitrate = 10, .hw_value = 0, }, + { .bitrate = 20, .hw_value = 1, }, + { .bitrate = 55, .hw_value = 2, }, + { .bitrate = 110, .hw_value = 3, }, + { .bitrate = 60, .hw_value = 4, }, + { .bitrate = 90, .hw_value = 5, }, + { .bitrate = 120, .hw_value = 6, }, + { .bitrate = 180, .hw_value = 7, }, + { .bitrate = 240, .hw_value = 8, }, + { .bitrate = 360, .hw_value = 9, }, + { .bitrate = 480, .hw_value = 10, }, + { .bitrate = 540, .hw_value = 11, }, +}; + +static const struct ieee80211_channel rtl818x_channels[] = { + { .center_freq = 2412 }, + { .center_freq = 2417 }, + { .center_freq = 2422 }, + { .center_freq = 2427 }, + { .center_freq = 2432 }, + { .center_freq = 2437 }, + { .center_freq = 2442 }, + { .center_freq = 2447 }, + { .center_freq = 2452 }, + { .center_freq = 2457 }, + { .center_freq = 2462 }, + { .center_freq = 2467 }, + { .center_freq = 2472 }, + { .center_freq = 2484 }, +}; + static void rtl8187_iowrite_async_cb(struct urb *urb) { kfree(urb->context); @@ -146,8 +178,8 @@ static int rtl8187_tx(struct ieee80211_hw *dev, struct sk_buff *skb, flags = skb->len; flags |= RTL8187_TX_FLAG_NO_ENCRYPT; - flags |= control->rts_cts_rate << 19; - flags |= control->tx_rate << 24; + flags |= control->rts_cts_rate->hw_value << 19; + flags |= control->tx_rate->hw_value << 24; if (ieee80211_get_morefrag((struct ieee80211_hdr *)skb->data)) flags |= RTL8187_TX_FLAG_MORE_FRAG; if (control->flags & IEEE80211_TXCTL_USE_RTS_CTS) { @@ -225,10 +257,9 @@ static void rtl8187_rx_cb(struct urb *urb) rx_status.antenna = (hdr->signal >> 7) & 1; rx_status.signal = 64 - min(hdr->noise, (u8)64); rx_status.ssi = signal; - rx_status.rate = rate; - rx_status.freq = dev->conf.freq; - rx_status.channel = dev->conf.channel; - rx_status.phymode = dev->conf.phymode; + rx_status.rate_idx = rate; + rx_status.freq = dev->conf.channel->center_freq; + rx_status.band = dev->conf.channel->band; rx_status.mactime = le64_to_cpu(hdr->mac_time); rx_status.flag |= RX_FLAG_TSFT; if (flags & (1 << 13)) @@ -682,19 +713,22 @@ static int __devinit rtl8187_probe(struct usb_interface *intf, usb_get_dev(udev); skb_queue_head_init(&priv->rx_queue); + + BUILD_BUG_ON(sizeof(priv->channels) != sizeof(rtl818x_channels)); + BUILD_BUG_ON(sizeof(priv->rates) != sizeof(rtl818x_rates)); + memcpy(priv->channels, rtl818x_channels, sizeof(rtl818x_channels)); memcpy(priv->rates, rtl818x_rates, sizeof(rtl818x_rates)); priv->map = (struct rtl818x_csr *)0xFF00; - priv->modes[0].mode = MODE_IEEE80211G; - priv->modes[0].num_rates = ARRAY_SIZE(rtl818x_rates); - priv->modes[0].rates = priv->rates; - priv->modes[0].num_channels = ARRAY_SIZE(rtl818x_channels); - priv->modes[0].channels = priv->channels; - priv->modes[1].mode = MODE_IEEE80211B; - priv->modes[1].num_rates = 4; - priv->modes[1].rates = priv->rates; - priv->modes[1].num_channels = ARRAY_SIZE(rtl818x_channels); - priv->modes[1].channels = priv->channels; + + priv->band.band = IEEE80211_BAND_2GHZ; + priv->band.channels = priv->channels; + priv->band.n_channels = ARRAY_SIZE(rtl818x_channels); + priv->band.bitrates = priv->rates; + priv->band.n_bitrates = ARRAY_SIZE(rtl818x_rates); + dev->wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->band; + + priv->mode = IEEE80211_IF_TYPE_MNTR; dev->flags = IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING | IEEE80211_HW_RX_INCLUDES_FCS; @@ -703,10 +737,6 @@ static int __devinit rtl8187_probe(struct usb_interface *intf, dev->max_rssi = 65; dev->max_signal = 64; - for (i = 0; i < 2; i++) - if ((err = ieee80211_register_hwmode(dev, &priv->modes[i]))) - goto err_free_dev; - eeprom.data = dev; eeprom.register_read = rtl8187_eeprom_register_read; eeprom.register_write = rtl8187_eeprom_register_write; @@ -730,20 +760,20 @@ static int __devinit rtl8187_probe(struct usb_interface *intf, for (i = 0; i < 3; i++) { eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_CHAN_1 + i, &txpwr); - (*channel++).val = txpwr & 0xFF; - (*channel++).val = txpwr >> 8; + (*channel++).hw_value = txpwr & 0xFF; + (*channel++).hw_value = txpwr >> 8; } for (i = 0; i < 2; i++) { eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_CHAN_4 + i, &txpwr); - (*channel++).val = txpwr & 0xFF; - (*channel++).val = txpwr >> 8; + (*channel++).hw_value = txpwr & 0xFF; + (*channel++).hw_value = txpwr >> 8; } for (i = 0; i < 2; i++) { eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_CHAN_6 + i, &txpwr); - (*channel++).val = txpwr & 0xFF; - (*channel++).val = txpwr >> 8; + (*channel++).hw_value = txpwr & 0xFF; + (*channel++).hw_value = txpwr >> 8; } eeprom_93cx6_read(&eeprom, RTL8187_EEPROM_TXPWR_BASE, diff --git a/drivers/net/wireless/rtl8187_rtl8225.c b/drivers/net/wireless/rtl8187_rtl8225.c index b713de17ba0a..9146387b4c5e 100644 --- a/drivers/net/wireless/rtl8187_rtl8225.c +++ b/drivers/net/wireless/rtl8187_rtl8225.c @@ -283,8 +283,8 @@ static void rtl8225_rf_set_tx_power(struct ieee80211_hw *dev, int channel) u32 reg; int i; - cck_power = priv->channels[channel - 1].val & 0xF; - ofdm_power = priv->channels[channel - 1].val >> 4; + cck_power = priv->channels[channel - 1].hw_value & 0xF; + ofdm_power = priv->channels[channel - 1].hw_value >> 4; cck_power = min(cck_power, (u8)11); ofdm_power = min(ofdm_power, (u8)35); @@ -500,8 +500,8 @@ static void rtl8225z2_rf_set_tx_power(struct ieee80211_hw *dev, int channel) u32 reg; int i; - cck_power = priv->channels[channel - 1].val & 0xF; - ofdm_power = priv->channels[channel - 1].val >> 4; + cck_power = priv->channels[channel - 1].hw_value & 0xF; + ofdm_power = priv->channels[channel - 1].hw_value >> 4; cck_power = min(cck_power, (u8)15); cck_power += priv->txpwr_base & 0xF; @@ -735,13 +735,14 @@ static void rtl8225_rf_set_channel(struct ieee80211_hw *dev, struct ieee80211_conf *conf) { struct rtl8187_priv *priv = dev->priv; + int chan = ieee80211_frequency_to_channel(conf->channel->center_freq); if (priv->rf->init == rtl8225_rf_init) - rtl8225_rf_set_tx_power(dev, conf->channel); + rtl8225_rf_set_tx_power(dev, chan); else - rtl8225z2_rf_set_tx_power(dev, conf->channel); + rtl8225z2_rf_set_tx_power(dev, chan); - rtl8225_write(dev, 0x7, rtl8225_chan[conf->channel - 1]); + rtl8225_write(dev, 0x7, rtl8225_chan[chan - 1]); msleep(10); } diff --git a/drivers/net/wireless/rtl818x.h b/drivers/net/wireless/rtl818x.h index 1e7d6f8278d7..4f7d38f506eb 100644 --- a/drivers/net/wireless/rtl818x.h +++ b/drivers/net/wireless/rtl818x.h @@ -175,74 +175,4 @@ struct rtl818x_rf_ops { void (*set_chan)(struct ieee80211_hw *, struct ieee80211_conf *); }; -static const struct ieee80211_rate rtl818x_rates[] = { - { .rate = 10, - .val = 0, - .flags = IEEE80211_RATE_CCK }, - { .rate = 20, - .val = 1, - .flags = IEEE80211_RATE_CCK }, - { .rate = 55, - .val = 2, - .flags = IEEE80211_RATE_CCK }, - { .rate = 110, - .val = 3, - .flags = IEEE80211_RATE_CCK }, - { .rate = 60, - .val = 4, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 90, - .val = 5, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 120, - .val = 6, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 180, - .val = 7, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 240, - .val = 8, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 360, - .val = 9, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 480, - .val = 10, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 540, - .val = 11, - .flags = IEEE80211_RATE_OFDM }, -}; - -static const struct ieee80211_channel rtl818x_channels[] = { - { .chan = 1, - .freq = 2412}, - { .chan = 2, - .freq = 2417}, - { .chan = 3, - .freq = 2422}, - { .chan = 4, - .freq = 2427}, - { .chan = 5, - .freq = 2432}, - { .chan = 6, - .freq = 2437}, - { .chan = 7, - .freq = 2442}, - { .chan = 8, - .freq = 2447}, - { .chan = 9, - .freq = 2452}, - { .chan = 10, - .freq = 2457}, - { .chan = 11, - .freq = 2462}, - { .chan = 12, - .freq = 2467}, - { .chan = 13, - .freq = 2472}, - { .chan = 14, - .freq = 2484} -}; - #endif /* RTL818X_H */ diff --git a/drivers/net/wireless/zd1211rw/zd_chip.c b/drivers/net/wireless/zd1211rw/zd_chip.c index 99e5b03b3f51..e3fba6f09455 100644 --- a/drivers/net/wireless/zd1211rw/zd_chip.c +++ b/drivers/net/wireless/zd1211rw/zd_chip.c @@ -986,7 +986,7 @@ static int print_fw_version(struct zd_chip *chip) return 0; } -static int set_mandatory_rates(struct zd_chip *chip, int mode) +static int set_mandatory_rates(struct zd_chip *chip, int gmode) { u32 rates; ZD_ASSERT(mutex_is_locked(&chip->mutex)); @@ -994,17 +994,12 @@ static int set_mandatory_rates(struct zd_chip *chip, int mode) * that the device is supporting. Until further notice we should try * to support 802.11g also for full speed USB. */ - switch (mode) { - case MODE_IEEE80211B: + if (!gmode) rates = CR_RATE_1M|CR_RATE_2M|CR_RATE_5_5M|CR_RATE_11M; - break; - case MODE_IEEE80211G: + else rates = CR_RATE_1M|CR_RATE_2M|CR_RATE_5_5M|CR_RATE_11M| CR_RATE_6M|CR_RATE_12M|CR_RATE_24M; - break; - default: - return -EINVAL; - } + return zd_iowrite32_locked(chip, rates, CR_MANDATORY_RATE_TBL); } @@ -1108,7 +1103,7 @@ int zd_chip_init_hw(struct zd_chip *chip) * It might be discussed, whether we should suppport pure b mode for * full speed USB. */ - r = set_mandatory_rates(chip, MODE_IEEE80211G); + r = set_mandatory_rates(chip, 1); if (r) goto out; /* Disabling interrupts is certainly a smart thing here. diff --git a/drivers/net/wireless/zd1211rw/zd_ieee80211.c b/drivers/net/wireless/zd1211rw/zd_ieee80211.c index 7c277ec43f79..d8dc41ec0e5d 100644 --- a/drivers/net/wireless/zd1211rw/zd_ieee80211.c +++ b/drivers/net/wireless/zd1211rw/zd_ieee80211.c @@ -65,16 +65,14 @@ static const struct channel_range *zd_channel_range(u8 regdomain) static void unmask_bg_channels(struct ieee80211_hw *hw, const struct channel_range *range, - struct ieee80211_hw_mode *mode) + struct ieee80211_supported_band *sband) { u8 channel; for (channel = range->start; channel < range->end; channel++) { struct ieee80211_channel *chan = - &mode->channels[CHAN_TO_IDX(channel)]; - chan->flag |= IEEE80211_CHAN_W_SCAN | - IEEE80211_CHAN_W_ACTIVE_SCAN | - IEEE80211_CHAN_W_IBSS; + &sband->channels[CHAN_TO_IDX(channel)]; + chan->flags = 0; } } @@ -97,7 +95,6 @@ void zd_geo_init(struct ieee80211_hw *hw, u8 regdomain) range = zd_channel_range(ZD_REGDOMAIN_FCC); } - unmask_bg_channels(hw, range, &mac->modes[0]); - unmask_bg_channels(hw, range, &mac->modes[1]); + unmask_bg_channels(hw, range, &mac->band); } diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c index 76ef2d83919d..21b653458831 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.c +++ b/drivers/net/wireless/zd1211rw/zd_mac.c @@ -34,76 +34,61 @@ /* This table contains the hardware specific values for the modulation rates. */ static const struct ieee80211_rate zd_rates[] = { - { .rate = 10, - .val = ZD_CCK_RATE_1M, - .flags = IEEE80211_RATE_CCK }, - { .rate = 20, - .val = ZD_CCK_RATE_2M, - .val2 = ZD_CCK_RATE_2M | ZD_CCK_PREA_SHORT, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 55, - .val = ZD_CCK_RATE_5_5M, - .val2 = ZD_CCK_RATE_5_5M | ZD_CCK_PREA_SHORT, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 110, - .val = ZD_CCK_RATE_11M, - .val2 = ZD_CCK_RATE_11M | ZD_CCK_PREA_SHORT, - .flags = IEEE80211_RATE_CCK_2 }, - { .rate = 60, - .val = ZD_OFDM_RATE_6M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 90, - .val = ZD_OFDM_RATE_9M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 120, - .val = ZD_OFDM_RATE_12M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 180, - .val = ZD_OFDM_RATE_18M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 240, - .val = ZD_OFDM_RATE_24M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 360, - .val = ZD_OFDM_RATE_36M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 480, - .val = ZD_OFDM_RATE_48M, - .flags = IEEE80211_RATE_OFDM }, - { .rate = 540, - .val = ZD_OFDM_RATE_54M, - .flags = IEEE80211_RATE_OFDM }, + { .bitrate = 10, + .hw_value = ZD_CCK_RATE_1M, }, + { .bitrate = 20, + .hw_value = ZD_CCK_RATE_2M, + .hw_value_short = ZD_CCK_RATE_2M | ZD_CCK_PREA_SHORT, + .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 55, + .hw_value = ZD_CCK_RATE_5_5M, + .hw_value_short = ZD_CCK_RATE_5_5M | ZD_CCK_PREA_SHORT, + .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 110, + .hw_value = ZD_CCK_RATE_11M, + .hw_value_short = ZD_CCK_RATE_11M | ZD_CCK_PREA_SHORT, + .flags = IEEE80211_RATE_SHORT_PREAMBLE }, + { .bitrate = 60, + .hw_value = ZD_OFDM_RATE_6M, + .flags = 0 }, + { .bitrate = 90, + .hw_value = ZD_OFDM_RATE_9M, + .flags = 0 }, + { .bitrate = 120, + .hw_value = ZD_OFDM_RATE_12M, + .flags = 0 }, + { .bitrate = 180, + .hw_value = ZD_OFDM_RATE_18M, + .flags = 0 }, + { .bitrate = 240, + .hw_value = ZD_OFDM_RATE_24M, + .flags = 0 }, + { .bitrate = 360, + .hw_value = ZD_OFDM_RATE_36M, + .flags = 0 }, + { .bitrate = 480, + .hw_value = ZD_OFDM_RATE_48M, + .flags = 0 }, + { .bitrate = 540, + .hw_value = ZD_OFDM_RATE_54M, + .flags = 0 }, }; static const struct ieee80211_channel zd_channels[] = { - { .chan = 1, - .freq = 2412}, - { .chan = 2, - .freq = 2417}, - { .chan = 3, - .freq = 2422}, - { .chan = 4, - .freq = 2427}, - { .chan = 5, - .freq = 2432}, - { .chan = 6, - .freq = 2437}, - { .chan = 7, - .freq = 2442}, - { .chan = 8, - .freq = 2447}, - { .chan = 9, - .freq = 2452}, - { .chan = 10, - .freq = 2457}, - { .chan = 11, - .freq = 2462}, - { .chan = 12, - .freq = 2467}, - { .chan = 13, - .freq = 2472}, - { .chan = 14, - .freq = 2484} + { .center_freq = 2412, .hw_value = 1 }, + { .center_freq = 2417, .hw_value = 2 }, + { .center_freq = 2422, .hw_value = 3 }, + { .center_freq = 2427, .hw_value = 4 }, + { .center_freq = 2432, .hw_value = 5 }, + { .center_freq = 2437, .hw_value = 6 }, + { .center_freq = 2442, .hw_value = 7 }, + { .center_freq = 2447, .hw_value = 8 }, + { .center_freq = 2452, .hw_value = 9 }, + { .center_freq = 2457, .hw_value = 10 }, + { .center_freq = 2462, .hw_value = 11 }, + { .center_freq = 2467, .hw_value = 12 }, + { .center_freq = 2472, .hw_value = 13 }, + { .center_freq = 2484, .hw_value = 14 }, }; static void housekeeping_init(struct zd_mac *mac); @@ -503,7 +488,9 @@ static int fill_ctrlset(struct zd_mac *mac, ZD_ASSERT(frag_len <= 0xffff); - cs->modulation = control->tx_rate; + cs->modulation = control->tx_rate->hw_value; + if (control->flags & IEEE80211_TXCTL_SHORT_PREAMBLE) + cs->modulation = control->tx_rate->hw_value_short; cs->tx_length = cpu_to_le16(frag_len); @@ -631,6 +618,8 @@ int zd_mac_rx(struct ieee80211_hw *hw, const u8 *buffer, unsigned int length) int bad_frame = 0; u16 fc; bool is_qos, is_4addr, need_padding; + int i; + u8 rate; if (length < ZD_PLCP_HEADER_SIZE + 10 /* IEEE80211_1ADDR_LEN */ + FCS_LEN + sizeof(struct rx_status)) @@ -660,14 +649,19 @@ int zd_mac_rx(struct ieee80211_hw *hw, const u8 *buffer, unsigned int length) } } - stats.channel = _zd_chip_get_channel(&mac->chip); - stats.freq = zd_channels[stats.channel - 1].freq; - stats.phymode = MODE_IEEE80211G; + stats.freq = zd_channels[_zd_chip_get_channel(&mac->chip) - 1].center_freq; + stats.band = IEEE80211_BAND_2GHZ; stats.ssi = status->signal_strength; stats.signal = zd_rx_qual_percent(buffer, length - sizeof(struct rx_status), status); - stats.rate = zd_rx_rate(buffer, status); + + rate = zd_rx_rate(buffer, status); + + /* todo: return index in the big switches in zd_rx_rate instead */ + for (i = 0; i < mac->band.n_bitrates; i++) + if (rate == mac->band.bitrates[i].hw_value) + stats.rate_idx = i; length -= ZD_PLCP_HEADER_SIZE + sizeof(struct rx_status); buffer += ZD_PLCP_HEADER_SIZE; @@ -736,7 +730,7 @@ static void zd_op_remove_interface(struct ieee80211_hw *hw, static int zd_op_config(struct ieee80211_hw *hw, struct ieee80211_conf *conf) { struct zd_mac *mac = zd_hw_mac(hw); - return zd_chip_set_channel(&mac->chip, conf->channel); + return zd_chip_set_channel(&mac->chip, conf->channel->hw_value); } static int zd_op_config_interface(struct ieee80211_hw *hw, @@ -894,7 +888,6 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf) { struct zd_mac *mac; struct ieee80211_hw *hw; - int i; hw = ieee80211_alloc_hw(sizeof(struct zd_mac), &zd_ops); if (!hw) { @@ -912,19 +905,14 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf) memcpy(mac->channels, zd_channels, sizeof(zd_channels)); memcpy(mac->rates, zd_rates, sizeof(zd_rates)); - mac->modes[0].mode = MODE_IEEE80211G; - mac->modes[0].num_rates = ARRAY_SIZE(zd_rates); - mac->modes[0].rates = mac->rates; - mac->modes[0].num_channels = ARRAY_SIZE(zd_channels); - mac->modes[0].channels = mac->channels; - mac->modes[1].mode = MODE_IEEE80211B; - mac->modes[1].num_rates = 4; - mac->modes[1].rates = mac->rates; - mac->modes[1].num_channels = ARRAY_SIZE(zd_channels); - mac->modes[1].channels = mac->channels; - - hw->flags = IEEE80211_HW_RX_INCLUDES_FCS | - IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED; + mac->band.n_bitrates = ARRAY_SIZE(zd_rates); + mac->band.bitrates = mac->rates; + mac->band.n_channels = ARRAY_SIZE(zd_channels); + mac->band.channels = mac->channels; + + hw->wiphy->bands[IEEE80211_BAND_2GHZ] = &mac->band; + + hw->flags = IEEE80211_HW_RX_INCLUDES_FCS; hw->max_rssi = 100; hw->max_signal = 100; @@ -933,14 +921,6 @@ struct ieee80211_hw *zd_mac_alloc_hw(struct usb_interface *intf) skb_queue_head_init(&mac->ack_wait_queue); - for (i = 0; i < 2; i++) { - if (ieee80211_register_hwmode(hw, &mac->modes[i])) { - dev_dbg_f(&intf->dev, "cannot register hwmode\n"); - ieee80211_free_hw(hw); - return NULL; - } - } - zd_chip_init(&mac->chip, hw, intf); housekeeping_init(mac); INIT_WORK(&mac->set_multicast_hash_work, set_multicast_hash_handler); diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h index 2dde108df767..67dea9739c8f 100644 --- a/drivers/net/wireless/zd1211rw/zd_mac.h +++ b/drivers/net/wireless/zd1211rw/zd_mac.h @@ -185,7 +185,7 @@ struct zd_mac { struct sk_buff_head ack_wait_queue; struct ieee80211_channel channels[14]; struct ieee80211_rate rates[12]; - struct ieee80211_hw_mode modes[2]; + struct ieee80211_supported_band band; /* Short preamble (used for RTS/CTS) */ unsigned int short_preamble:1; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 277488176a44..460da54a0019 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -69,95 +69,6 @@ * not do so then mac80211 may add this under certain circumstances. */ -#define IEEE80211_CHAN_W_SCAN 0x00000001 -#define IEEE80211_CHAN_W_ACTIVE_SCAN 0x00000002 -#define IEEE80211_CHAN_W_IBSS 0x00000004 - -/* Channel information structure. Low-level driver is expected to fill in chan, - * freq, and val fields. Other fields will be filled in by 80211.o based on - * hostapd information and low-level driver does not need to use them. The - * limits for each channel will be provided in 'struct ieee80211_conf' when - * configuring the low-level driver with hw->config callback. If a device has - * a default regulatory domain, IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED - * can be set to let the driver configure all fields */ -struct ieee80211_channel { - short chan; /* channel number (IEEE 802.11) */ - short freq; /* frequency in MHz */ - int val; /* hw specific value for the channel */ - int flag; /* flag for hostapd use (IEEE80211_CHAN_*) */ - unsigned char power_level; - unsigned char antenna_max; -}; - -#define IEEE80211_RATE_ERP 0x00000001 -#define IEEE80211_RATE_BASIC 0x00000002 -#define IEEE80211_RATE_PREAMBLE2 0x00000004 -#define IEEE80211_RATE_SUPPORTED 0x00000010 -#define IEEE80211_RATE_OFDM 0x00000020 -#define IEEE80211_RATE_CCK 0x00000040 -#define IEEE80211_RATE_MANDATORY 0x00000100 - -#define IEEE80211_RATE_CCK_2 (IEEE80211_RATE_CCK | IEEE80211_RATE_PREAMBLE2) -#define IEEE80211_RATE_MODULATION(f) \ - (f & (IEEE80211_RATE_CCK | IEEE80211_RATE_OFDM)) - -/* Low-level driver should set PREAMBLE2, OFDM and CCK flags. - * BASIC, SUPPORTED, ERP, and MANDATORY flags are set in 80211.o based on the - * configuration. */ -struct ieee80211_rate { - int rate; /* rate in 100 kbps */ - int val; /* hw specific value for the rate */ - int flags; /* IEEE80211_RATE_ flags */ - int val2; /* hw specific value for the rate when using short preamble - * (only when IEEE80211_RATE_PREAMBLE2 flag is set, i.e., for - * 2, 5.5, and 11 Mbps) */ - signed char min_rssi_ack; - unsigned char min_rssi_ack_delta; - - /* following fields are set by 80211.o and need not be filled by the - * low-level driver */ - int rate_inv; /* inverse of the rate (LCM(all rates) / rate) for - * optimizing channel utilization estimates */ -}; - -/** - * enum ieee80211_phymode - PHY modes - * - * @MODE_IEEE80211A: 5GHz as defined by 802.11a/802.11h - * @MODE_IEEE80211B: 2.4 GHz as defined by 802.11b - * @MODE_IEEE80211G: 2.4 GHz as defined by 802.11g (with OFDM), - * backwards compatible with 11b mode - * @NUM_IEEE80211_MODES: internal - */ -enum ieee80211_phymode { - MODE_IEEE80211A, - MODE_IEEE80211B, - MODE_IEEE80211G, - - /* keep last */ - NUM_IEEE80211_MODES -}; - -/** - * struct ieee80211_ht_info - describing STA's HT capabilities - * - * This structure describes most essential parameters needed - * to describe 802.11n HT capabilities for an STA. - * - * @ht_supported: is HT supported by STA, 0: no, 1: yes - * @cap: HT capabilities map as described in 802.11n spec - * @ampdu_factor: Maximum A-MPDU length factor - * @ampdu_density: Minimum A-MPDU spacing - * @supp_mcs_set: Supported MCS set as described in 802.11n spec - */ -struct ieee80211_ht_info { - u8 ht_supported; - u16 cap; /* use IEEE80211_HT_CAP_ */ - u8 ampdu_factor; - u8 ampdu_density; - u8 supp_mcs_set[16]; -}; - /** * struct ieee80211_ht_bss_info - describing BSS's HT characteristics * @@ -174,30 +85,6 @@ struct ieee80211_ht_bss_info { u8 bss_op_mode; /* use IEEE80211_HT_IE_ */ }; -/** - * struct ieee80211_hw_mode - PHY mode definition - * - * This structure describes the capabilities supported by the device - * in a single PHY mode. - * - * @list: internal - * @channels: pointer to array of supported channels - * @rates: pointer to array of supported bitrates - * @mode: the PHY mode for this definition - * @num_channels: number of supported channels - * @num_rates: number of supported bitrates - * @ht_info: PHY's 802.11n HT abilities for this mode - */ -struct ieee80211_hw_mode { - struct list_head list; - struct ieee80211_channel *channels; - struct ieee80211_rate *rates; - enum ieee80211_phymode mode; - int num_channels; - int num_rates; - struct ieee80211_ht_info ht_info; -}; - /** * struct ieee80211_tx_queue_params - transmit queue configuration * @@ -320,11 +207,13 @@ struct ieee80211_bss_conf { struct ieee80211_tx_control { struct ieee80211_vif *vif; - int tx_rate; /* Transmit rate, given as the hw specific value for the - * rate (from struct ieee80211_rate) */ - int rts_cts_rate; /* Transmit rate for RTS/CTS frame, given as the hw - * specific value for the rate (from - * struct ieee80211_rate) */ + struct ieee80211_rate *tx_rate; + + /* Transmit rate for RTS/CTS frame */ + struct ieee80211_rate *rts_cts_rate; + + /* retry rate for the last retries */ + struct ieee80211_rate *alt_retry_rate; #define IEEE80211_TXCTL_REQ_TX_STATUS (1<<0)/* request TX status callback for * this frame */ @@ -343,6 +232,7 @@ struct ieee80211_tx_control { #define IEEE80211_TXCTL_REQUEUE (1<<7) #define IEEE80211_TXCTL_FIRST_FRAGMENT (1<<8) /* this is a first fragment of * the frame */ +#define IEEE80211_TXCTL_SHORT_PREAMBLE (1<<9) #define IEEE80211_TXCTL_LONG_RETRY_LIMIT (1<<10) /* this frame should be send * using the through * set_retry_limit configured @@ -359,20 +249,11 @@ struct ieee80211_tx_control { u8 retry_limit; /* 1 = only first attempt, 2 = one retry, .. * This could be used when set_retry_limit * is not implemented by the driver */ - u8 power_level; /* per-packet transmit power level, in dBm */ u8 antenna_sel_tx; /* 0 = default/diversity, 1 = Ant0, 2 = Ant1 */ u8 icv_len; /* length of the ICV/MIC field in octets */ u8 iv_len; /* length of the IV field in octets */ u8 queue; /* hardware queue to use for this frame; * 0 = highest, hw->queues-1 = lowest */ - struct ieee80211_rate *rate; /* internal 80211.o rate */ - struct ieee80211_rate *rts_rate; /* internal 80211.o rate - * for RTS/CTS */ - int alt_retry_rate; /* retry rate for the last retries, given as the - * hw specific value for the rate (from - * struct ieee80211_rate). To be used to limit - * packet dropping when probing higher rates, if hw - * supports multiple retry rates. -1 = not used */ int type; /* internal */ }; @@ -415,26 +296,24 @@ enum mac80211_rx_flags { * supported by hardware) to the 802.11 code with each received * frame. * @mactime: MAC timestamp as defined by 802.11 + * @band: the active band when this frame was received * @freq: frequency the radio was tuned to when receiving this frame, in MHz - * @channel: channel the radio was tuned to - * @phymode: active PHY mode * @ssi: signal strength when receiving this frame * @signal: used as 'qual' in statistics reporting * @noise: PHY noise when receiving this frame * @antenna: antenna used - * @rate: data rate + * @rate_idx: index of data rate into band's supported rates * @flag: %RX_FLAG_* */ struct ieee80211_rx_status { u64 mactime; + enum ieee80211_band band; int freq; - int channel; - enum ieee80211_phymode phymode; int ssi; int signal; int noise; int antenna; - int rate; + int rate_idx; int flag; }; @@ -509,41 +388,30 @@ enum ieee80211_conf_flags { * * @radio_enabled: when zero, driver is required to switch off the radio. * TODO make a flag - * @channel: IEEE 802.11 channel number - * @freq: frequency in MHz - * @channel_val: hardware specific channel value for the channel - * @phymode: PHY mode to activate (REMOVE) - * @chan: channel to switch to, pointer to the channel information - * @mode: pointer to mode definition - * @regulatory_domain: ?? * @beacon_int: beacon interval (TODO make interface config) * @flags: configuration flags defined above - * @power_level: transmit power limit for current regulatory domain in dBm - * @antenna_max: maximum antenna gain + * @power_level: requested transmit power (in dBm) + * @max_antenna_gain: maximum antenna gain (in dBi) * @antenna_sel_tx: transmit antenna selection, 0: default/diversity, * 1/2: antenna 0/1 * @antenna_sel_rx: receive antenna selection, like @antenna_sel_tx * @ht_conf: describes current self configuration of 802.11n HT capabilies * @ht_bss_conf: describes current BSS configuration of 802.11n HT parameters + * @channel: the channel to tune to */ struct ieee80211_conf { - int channel; /* IEEE 802.11 channel number */ - int freq; /* MHz */ - int channel_val; /* hw specific value for the channel */ - - enum ieee80211_phymode phymode; - struct ieee80211_channel *chan; - struct ieee80211_hw_mode *mode; unsigned int regulatory_domain; int radio_enabled; int beacon_int; u32 flags; - u8 power_level; - u8 antenna_max; + int power_level; + int max_antenna_gain; u8 antenna_sel_tx; u8 antenna_sel_rx; + struct ieee80211_channel *channel; + struct ieee80211_ht_info ht_conf; struct ieee80211_ht_bss_info ht_bss_conf; }; @@ -764,15 +632,19 @@ enum sta_notify_cmd { * %IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE is also not set because * otherwise the stack will not know when the DTIM beacon was sent. * - * @IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED: - * Channels are already configured to the default regulatory domain - * specified in the device's EEPROM + * @IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE: + * Hardware is not capable of short slot operation on the 2.4 GHz band. + * + * @IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE: + * Hardware is not capable of receiving frames with short preamble on + * the 2.4 GHz band. */ enum ieee80211_hw_flags { IEEE80211_HW_HOST_GEN_BEACON_TEMPLATE = 1<<0, IEEE80211_HW_RX_INCLUDES_FCS = 1<<1, IEEE80211_HW_HOST_BROADCAST_PS_BUFFERING = 1<<2, - IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED = 1<<3, + IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE = 1<<3, + IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE = 1<<4, }; /** @@ -784,7 +656,8 @@ enum ieee80211_hw_flags { * @wiphy: This points to the &struct wiphy allocated for this * 802.11 PHY. You must fill in the @perm_addr and @dev * members of this structure using SET_IEEE80211_DEV() - * and SET_IEEE80211_PERM_ADDR(). + * and SET_IEEE80211_PERM_ADDR(). Additionally, all supported + * bands (with channels, bitrates) are registered here. * * @conf: &struct ieee80211_conf, device configuration, don't use. * @@ -1062,7 +935,9 @@ enum ieee80211_ampdu_mlme_action { * given local_address is enabled. * * @hw_scan: Ask the hardware to service the scan request, no need to start - * the scan state machine in stack. + * the scan state machine in stack. The scan must honour the channel + * configuration done by the regulatory agent in the wiphy's registered + * bands. * * @get_stats: return low-level statistics * @@ -1284,10 +1159,6 @@ static inline char *ieee80211_get_radio_led_name(struct ieee80211_hw *hw) #endif } -/* Register a new hardware PHYMODE capability to the stack. */ -int ieee80211_register_hwmode(struct ieee80211_hw *hw, - struct ieee80211_hw_mode *mode); - /** * ieee80211_unregister_hw - Unregister a hardware device * @@ -1461,7 +1332,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, * @hw: pointer obtained from ieee80211_alloc_hw(). * @vif: &struct ieee80211_vif pointer from &struct ieee80211_if_init_conf. * @frame_len: the length of the frame. - * @rate: the rate (in 100kbps) at which the frame is going to be transmitted. + * @rate: the rate at which the frame is going to be transmitted. * * Calculate the duration field of some generic frame, given its * length and transmission rate (in 100kbps). @@ -1469,7 +1340,7 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw, struct ieee80211_vif *vif, size_t frame_len, - int rate); + struct ieee80211_rate *rate); /** * ieee80211_get_buffered_bc - accessing buffered broadcast and multicast frames diff --git a/include/net/wireless.h b/include/net/wireless.h index d30c4ba8fd99..c7f805ee5545 100644 --- a/include/net/wireless.h +++ b/include/net/wireless.h @@ -12,6 +12,162 @@ #include #include +/** + * enum ieee80211_band - supported frequency bands + * + * The bands are assigned this way because the supported + * bitrates differ in these bands. + * + * @IEEE80211_BAND_2GHZ: 2.4GHz ISM band + * @IEEE80211_BAND_5GHZ: around 5GHz band (4.9-5.7) + */ +enum ieee80211_band { + IEEE80211_BAND_2GHZ, + IEEE80211_BAND_5GHZ, + + /* keep last */ + IEEE80211_NUM_BANDS +}; + +/** + * enum ieee80211_channel_flags - channel flags + * + * Channel flags set by the regulatory control code. + * + * @IEEE80211_CHAN_DISABLED: This channel is disabled. + * @IEEE80211_CHAN_PASSIVE_SCAN: Only passive scanning is permitted + * on this channel. + * @IEEE80211_CHAN_NO_IBSS: IBSS is not allowed on this channel. + * @IEEE80211_CHAN_RADAR: Radar detection is required on this channel. + */ +enum ieee80211_channel_flags { + IEEE80211_CHAN_DISABLED = 1<<0, + IEEE80211_CHAN_PASSIVE_SCAN = 1<<1, + IEEE80211_CHAN_NO_IBSS = 1<<2, + IEEE80211_CHAN_RADAR = 1<<3, +}; + +/** + * struct ieee80211_channel - channel definition + * + * This structure describes a single channel for use + * with cfg80211. + * + * @center_freq: center frequency in MHz + * @hw_value: hardware-specific value for the channel + * @flags: channel flags from &enum ieee80211_channel_flags. + * @orig_flags: channel flags at registration time, used by regulatory + * code to support devices with additional restrictions + * @band: band this channel belongs to. + * @max_antenna_gain: maximum antenna gain in dBi + * @max_power: maximum transmission power (in dBm) + * @orig_mag: internal use + * @orig_mpwr: internal use + */ +struct ieee80211_channel { + enum ieee80211_band band; + u16 center_freq; + u16 hw_value; + u32 flags; + int max_antenna_gain; + int max_power; + u32 orig_flags; + int orig_mag, orig_mpwr; +}; + +/** + * enum ieee80211_rate_flags - rate flags + * + * Hardware/specification flags for rates. These are structured + * in a way that allows using the same bitrate structure for + * different bands/PHY modes. + * + * @IEEE80211_RATE_SHORT_PREAMBLE: Hardware can send with short + * preamble on this bitrate; only relevant in 2.4GHz band and + * with CCK rates. + * @IEEE80211_RATE_MANDATORY_A: This bitrate is a mandatory rate + * when used with 802.11a (on the 5 GHz band); filled by the + * core code when registering the wiphy. + * @IEEE80211_RATE_MANDATORY_B: This bitrate is a mandatory rate + * when used with 802.11b (on the 2.4 GHz band); filled by the + * core code when registering the wiphy. + * @IEEE80211_RATE_MANDATORY_G: This bitrate is a mandatory rate + * when used with 802.11g (on the 2.4 GHz band); filled by the + * core code when registering the wiphy. + * @IEEE80211_RATE_ERP_G: This is an ERP rate in 802.11g mode. + */ +enum ieee80211_rate_flags { + IEEE80211_RATE_SHORT_PREAMBLE = 1<<0, + IEEE80211_RATE_MANDATORY_A = 1<<1, + IEEE80211_RATE_MANDATORY_B = 1<<2, + IEEE80211_RATE_MANDATORY_G = 1<<3, + IEEE80211_RATE_ERP_G = 1<<4, +}; + +/** + * struct ieee80211_rate - bitrate definition + * + * This structure describes a bitrate that an 802.11 PHY can + * operate with. The two values @hw_value and @hw_value_short + * are only for driver use when pointers to this structure are + * passed around. + * + * @flags: rate-specific flags + * @bitrate: bitrate in units of 100 Kbps + * @hw_value: driver/hardware value for this rate + * @hw_value_short: driver/hardware value for this rate when + * short preamble is used + */ +struct ieee80211_rate { + u32 flags; + u16 bitrate; + u16 hw_value, hw_value_short; +}; + +/** + * struct ieee80211_ht_info - describing STA's HT capabilities + * + * This structure describes most essential parameters needed + * to describe 802.11n HT capabilities for an STA. + * + * @ht_supported: is HT supported by STA, 0: no, 1: yes + * @cap: HT capabilities map as described in 802.11n spec + * @ampdu_factor: Maximum A-MPDU length factor + * @ampdu_density: Minimum A-MPDU spacing + * @supp_mcs_set: Supported MCS set as described in 802.11n spec + */ +struct ieee80211_ht_info { + u16 cap; /* use IEEE80211_HT_CAP_ */ + u8 ht_supported; + u8 ampdu_factor; + u8 ampdu_density; + u8 supp_mcs_set[16]; +}; + +/** + * struct ieee80211_supported_band - frequency band definition + * + * This structure describes a frequency band a wiphy + * is able to operate in. + * + * @channels: Array of channels the hardware can operate in + * in this band. + * @band: the band this structure represents + * @n_channels: Number of channels in @channels + * @bitrates: Array of bitrates the hardware can operate with + * in this band. Must be sorted to give a valid "supported + * rates" IE, i.e. CCK rates first, then OFDM. + * @n_bitrates: Number of bitrates in @bitrates + */ +struct ieee80211_supported_band { + struct ieee80211_channel *channels; + struct ieee80211_rate *bitrates; + enum ieee80211_band band; + int n_channels; + int n_bitrates; + struct ieee80211_ht_info ht_info; +}; + /** * struct wiphy - wireless hardware description * @idx: the wiphy index assigned to this item @@ -30,6 +186,8 @@ struct wiphy { * help determine whether you own this wiphy or not. */ void *privid; + struct ieee80211_supported_band *bands[IEEE80211_NUM_BANDS]; + /* fields below are read-only, assigned by cfg80211 */ /* the item in /sys/class/ieee80211/ points to this, @@ -136,4 +294,14 @@ extern void wiphy_unregister(struct wiphy *wiphy); */ extern void wiphy_free(struct wiphy *wiphy); +/** + * ieee80211_channel_to_frequency - convert channel number to frequency + */ +extern int ieee80211_channel_to_frequency(int chan); + +/** + * ieee80211_frequency_to_channel - convert frequency to channel number + */ +extern int ieee80211_frequency_to_channel(int freq); + #endif /* __NET_WIRELESS_H */ diff --git a/net/mac80211/Makefile b/net/mac80211/Makefile index 54f46bc80cfe..9d7a19581a29 100644 --- a/net/mac80211/Makefile +++ b/net/mac80211/Makefile @@ -19,7 +19,6 @@ mac80211-y := \ ieee80211_iface.o \ ieee80211_rate.o \ michael.o \ - regdomain.o \ tkip.o \ aes_ccm.o \ cfg.o \ diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index 22c9619ba776..15b8cf94f510 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -498,7 +498,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, { u32 rates; int i, j; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; if (params->station_flags & STATION_FLAG_CHANGED) { sta->flags &= ~WLAN_STA_AUTHORIZED; @@ -525,15 +525,16 @@ static void sta_apply_parameters(struct ieee80211_local *local, if (params->supported_rates) { rates = 0; - mode = local->oper_hw_mode; + sband = local->hw.wiphy->bands[local->oper_channel->band]; + for (i = 0; i < params->supported_rates_len; i++) { int rate = (params->supported_rates[i] & 0x7f) * 5; - for (j = 0; j < mode->num_rates; j++) { - if (mode->rates[j].rate == rate) + for (j = 0; j < sband->n_bitrates; j++) { + if (sband->bitrates[j].bitrate == rate) rates |= BIT(j); } } - sta->supp_rates = rates; + sta->supp_rates[local->oper_channel->band] = rates; } } diff --git a/net/mac80211/debugfs.c b/net/mac80211/debugfs.c index 60514b2c97b9..4736c64937b4 100644 --- a/net/mac80211/debugfs.c +++ b/net/mac80211/debugfs.c @@ -19,41 +19,6 @@ int mac80211_open_file_generic(struct inode *inode, struct file *file) return 0; } -static const char *ieee80211_mode_str(int mode) -{ - switch (mode) { - case MODE_IEEE80211A: - return "IEEE 802.11a"; - case MODE_IEEE80211B: - return "IEEE 802.11b"; - case MODE_IEEE80211G: - return "IEEE 802.11g"; - default: - return "UNKNOWN"; - } -} - -static ssize_t modes_read(struct file *file, char __user *userbuf, - size_t count, loff_t *ppos) -{ - struct ieee80211_local *local = file->private_data; - struct ieee80211_hw_mode *mode; - char buf[150], *p = buf; - - /* FIXME: locking! */ - list_for_each_entry(mode, &local->modes_list, list) { - p += scnprintf(p, sizeof(buf)+buf-p, - "%s\n", ieee80211_mode_str(mode->mode)); - } - - return simple_read_from_buffer(userbuf, count, ppos, buf, p-buf); -} - -static const struct file_operations modes_ops = { - .read = modes_read, - .open = mac80211_open_file_generic, -}; - #define DEBUGFS_READONLY_FILE(name, buflen, fmt, value...) \ static ssize_t name## _read(struct file *file, char __user *userbuf, \ size_t count, loff_t *ppos) \ @@ -80,10 +45,8 @@ static const struct file_operations name## _ops = { \ local->debugfs.name = NULL; -DEBUGFS_READONLY_FILE(channel, 20, "%d", - local->hw.conf.channel); DEBUGFS_READONLY_FILE(frequency, 20, "%d", - local->hw.conf.freq); + local->hw.conf.channel->center_freq); DEBUGFS_READONLY_FILE(antenna_sel_tx, 20, "%d", local->hw.conf.antenna_sel_tx); DEBUGFS_READONLY_FILE(antenna_sel_rx, 20, "%d", @@ -100,8 +63,6 @@ DEBUGFS_READONLY_FILE(long_retry_limit, 20, "%d", local->long_retry_limit); DEBUGFS_READONLY_FILE(total_ps_buffered, 20, "%d", local->total_ps_buffered); -DEBUGFS_READONLY_FILE(mode, 20, "%s", - ieee80211_mode_str(local->hw.conf.phymode)); DEBUGFS_READONLY_FILE(wep_iv, 20, "%#06x", local->wep_iv & 0xffffff); DEBUGFS_READONLY_FILE(rate_ctrl_alg, 100, "%s", @@ -294,7 +255,6 @@ void debugfs_hw_add(struct ieee80211_local *local) local->debugfs.stations = debugfs_create_dir("stations", phyd); local->debugfs.keys = debugfs_create_dir("keys", phyd); - DEBUGFS_ADD(channel); DEBUGFS_ADD(frequency); DEBUGFS_ADD(antenna_sel_tx); DEBUGFS_ADD(antenna_sel_rx); @@ -304,9 +264,7 @@ void debugfs_hw_add(struct ieee80211_local *local) DEBUGFS_ADD(short_retry_limit); DEBUGFS_ADD(long_retry_limit); DEBUGFS_ADD(total_ps_buffered); - DEBUGFS_ADD(mode); DEBUGFS_ADD(wep_iv); - DEBUGFS_ADD(modes); statsd = debugfs_create_dir("statistics", phyd); local->debugfs.statistics = statsd; @@ -356,7 +314,6 @@ void debugfs_hw_add(struct ieee80211_local *local) void debugfs_hw_del(struct ieee80211_local *local) { - DEBUGFS_DEL(channel); DEBUGFS_DEL(frequency); DEBUGFS_DEL(antenna_sel_tx); DEBUGFS_DEL(antenna_sel_rx); @@ -366,9 +323,7 @@ void debugfs_hw_del(struct ieee80211_local *local) DEBUGFS_DEL(short_retry_limit); DEBUGFS_DEL(long_retry_limit); DEBUGFS_DEL(total_ps_buffered); - DEBUGFS_DEL(mode); DEBUGFS_DEL(wep_iv); - DEBUGFS_DEL(modes); DEBUGFS_STATS_DEL(transmitted_fragment_count); DEBUGFS_STATS_DEL(multicast_transmitted_frame_count); diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c index df25abf63137..49660f4e845d 100644 --- a/net/mac80211/debugfs_sta.c +++ b/net/mac80211/debugfs_sta.c @@ -33,22 +33,6 @@ static ssize_t sta_ ##name## _read(struct file *file, \ #define STA_READ_LU(name, field) STA_READ(name, 20, field, "%lu\n") #define STA_READ_S(name, field) STA_READ(name, 20, field, "%s\n") -#define STA_READ_RATE(name, field) \ -static ssize_t sta_##name##_read(struct file *file, \ - char __user *userbuf, \ - size_t count, loff_t *ppos) \ -{ \ - struct sta_info *sta = file->private_data; \ - struct ieee80211_local *local = wdev_priv(sta->dev->ieee80211_ptr);\ - struct ieee80211_hw_mode *mode = local->oper_hw_mode; \ - char buf[20]; \ - int res = scnprintf(buf, sizeof(buf), "%d\n", \ - (sta->field >= 0 && \ - sta->field < mode->num_rates) ? \ - mode->rates[sta->field].rate : -1); \ - return simple_read_from_buffer(userbuf, count, ppos, buf, res); \ -} - #define STA_OPS(name) \ static const struct file_operations sta_ ##name## _ops = { \ .read = sta_##name##_read, \ @@ -77,8 +61,6 @@ STA_FILE(rx_fragments, rx_fragments, LU); STA_FILE(rx_dropped, rx_dropped, LU); STA_FILE(tx_fragments, tx_fragments, LU); STA_FILE(tx_filtered, tx_filtered_count, LU); -STA_FILE(txrate, txrate, RATE); -STA_FILE(last_txrate, last_txrate, RATE); STA_FILE(tx_retry_failed, tx_retry_failed, LU); STA_FILE(tx_retry_count, tx_retry_count, LU); STA_FILE(last_rssi, last_rssi, D); diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index 3961d4c4320c..de894b61a23c 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -876,37 +876,28 @@ int ieee80211_if_config_beacon(struct net_device *dev) int ieee80211_hw_config(struct ieee80211_local *local) { - struct ieee80211_hw_mode *mode; struct ieee80211_channel *chan; int ret = 0; - if (local->sta_sw_scanning) { + if (local->sta_sw_scanning) chan = local->scan_channel; - mode = local->scan_hw_mode; - } else { + else chan = local->oper_channel; - mode = local->oper_hw_mode; - } - local->hw.conf.channel = chan->chan; - local->hw.conf.channel_val = chan->val; - if (!local->hw.conf.power_level) { - local->hw.conf.power_level = chan->power_level; - } else { - local->hw.conf.power_level = min(chan->power_level, - local->hw.conf.power_level); - } - local->hw.conf.freq = chan->freq; - local->hw.conf.phymode = mode->mode; - local->hw.conf.antenna_max = chan->antenna_max; - local->hw.conf.chan = chan; - local->hw.conf.mode = mode; + local->hw.conf.channel = chan; + + if (!local->hw.conf.power_level) + local->hw.conf.power_level = chan->max_power; + else + local->hw.conf.power_level = min(chan->max_power, + local->hw.conf.power_level); + + local->hw.conf.max_antenna_gain = chan->max_antenna_gain; #ifdef CONFIG_MAC80211_VERBOSE_DEBUG - printk(KERN_DEBUG "HW CONFIG: channel=%d freq=%d " - "phymode=%d\n", local->hw.conf.channel, local->hw.conf.freq, - local->hw.conf.phymode); -#endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ + printk(KERN_DEBUG "%s: HW CONFIG: freq=%d\n", + wiphy_name(local->hw.wiphy), chan->center_freq); +#endif if (local->open_count) ret = local->ops->config(local_to_hw(local), &local->hw.conf); @@ -924,11 +915,13 @@ int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht, struct ieee80211_ht_bss_info *req_bss_cap) { struct ieee80211_conf *conf = &local->hw.conf; - struct ieee80211_hw_mode *mode = conf->mode; + struct ieee80211_supported_band *sband; int i; + sband = local->hw.wiphy->bands[conf->channel->band]; + /* HT is not supported */ - if (!mode->ht_info.ht_supported) { + if (!sband->ht_info.ht_supported) { conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE; return -EOPNOTSUPP; } @@ -938,17 +931,17 @@ int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht, conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE; } else { conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE; - conf->ht_conf.cap = req_ht_cap->cap & mode->ht_info.cap; + conf->ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap; conf->ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS); conf->ht_conf.cap |= - mode->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS; + sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS; conf->ht_bss_conf.primary_channel = req_bss_cap->primary_channel; conf->ht_bss_conf.bss_cap = req_bss_cap->bss_cap; conf->ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode; for (i = 0; i < SUPP_MCS_SET_LEN; i++) conf->ht_conf.supp_mcs_set[i] = - mode->ht_info.supp_mcs_set[i] & + sband->ht_info.supp_mcs_set[i] & req_ht_cap->supp_mcs_set[i]; /* In STA mode, this gives us indication @@ -1418,10 +1411,6 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, local->long_retry_limit = 4; local->hw.conf.radio_enabled = 1; - local->enabled_modes = ~0; - - INIT_LIST_HEAD(&local->modes_list); - INIT_LIST_HEAD(&local->interfaces); INIT_DELAYED_WORK(&local->scan_work, ieee80211_sta_scan_work); @@ -1466,6 +1455,25 @@ int ieee80211_register_hw(struct ieee80211_hw *hw) struct ieee80211_local *local = hw_to_local(hw); const char *name; int result; + enum ieee80211_band band; + + /* + * generic code guarantees at least one band, + * set this very early because much code assumes + * that hw.conf.channel is assigned + */ + for (band = 0; band < IEEE80211_NUM_BANDS; band++) { + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[band]; + if (sband) { + /* init channel we're on */ + local->hw.conf.channel = + local->oper_channel = + local->scan_channel = &sband->channels[0]; + break; + } + } result = wiphy_register(local->hw.wiphy); if (result < 0) @@ -1567,44 +1575,10 @@ fail_workqueue: } EXPORT_SYMBOL(ieee80211_register_hw); -int ieee80211_register_hwmode(struct ieee80211_hw *hw, - struct ieee80211_hw_mode *mode) -{ - struct ieee80211_local *local = hw_to_local(hw); - struct ieee80211_rate *rate; - int i; - - INIT_LIST_HEAD(&mode->list); - list_add_tail(&mode->list, &local->modes_list); - - local->hw_modes |= (1 << mode->mode); - for (i = 0; i < mode->num_rates; i++) { - rate = &(mode->rates[i]); - rate->rate_inv = CHAN_UTIL_RATE_LCM / rate->rate; - } - ieee80211_prepare_rates(local, mode); - - if (!local->oper_hw_mode) { - /* Default to this mode */ - local->hw.conf.phymode = mode->mode; - local->oper_hw_mode = local->scan_hw_mode = mode; - local->oper_channel = local->scan_channel = &mode->channels[0]; - local->hw.conf.mode = local->oper_hw_mode; - local->hw.conf.chan = local->oper_channel; - } - - if (!(hw->flags & IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED)) - ieee80211_set_default_regdomain(mode); - - return 0; -} -EXPORT_SYMBOL(ieee80211_register_hwmode); - void ieee80211_unregister_hw(struct ieee80211_hw *hw) { struct ieee80211_local *local = hw_to_local(hw); struct ieee80211_sub_if_data *sdata, *tmp; - int i; tasklet_kill(&local->tx_pending_tasklet); tasklet_kill(&local->tasklet); @@ -1645,11 +1619,6 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw) rate_control_deinitialize(local); debugfs_hw_del(local); - for (i = 0; i < NUM_IEEE80211_MODES; i++) { - kfree(local->supp_rates[i]); - kfree(local->basic_rates[i]); - } - if (skb_queue_len(&local->skb_queue) || skb_queue_len(&local->skb_queue_unreliable)) printk(KERN_WARNING "%s: skb_queue not empty\n", @@ -1696,7 +1665,6 @@ static int __init ieee80211_init(void) } ieee80211_debugfs_netdev_init(); - ieee80211_regdomain_init(); return 0; diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 9d09ba8cc02b..54eea5f24474 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -79,8 +79,7 @@ struct ieee80211_sta_bss { u8 ssid[IEEE80211_MAX_SSID_LEN]; size_t ssid_len; u16 capability; /* host byte order */ - int hw_mode; - int channel; + enum ieee80211_band band; int freq; int rssi, signal, noise; u8 *wpa_ie; @@ -136,13 +135,12 @@ struct ieee80211_txrx_data { union { struct { struct ieee80211_tx_control *control; - struct ieee80211_hw_mode *mode; + struct ieee80211_channel *channel; struct ieee80211_rate *rate; /* use this rate (if set) for last fragment; rate can * be set to lower rate for the first fragments, e.g., * when using CTS protection with IEEE 802.11g. */ struct ieee80211_rate *last_frag_rate; - int last_frag_hwrate; /* Extra fragments (in addition to the first fragment * in skb) */ @@ -151,6 +149,7 @@ struct ieee80211_txrx_data { } tx; struct { struct ieee80211_rx_status *status; + struct ieee80211_rate *rate; int sent_ps_buffered; int queue; int load; @@ -179,8 +178,6 @@ struct ieee80211_tx_stored_packet { struct sk_buff *skb; int num_extra_frag; struct sk_buff **extra_frag; - int last_frag_rateidx; - int last_frag_hwrate; struct ieee80211_rate *last_frag_rate; unsigned int last_frag_rate_ctrl_probe; }; @@ -283,7 +280,7 @@ struct ieee80211_if_sta { unsigned long ibss_join_req; struct sk_buff *probe_resp; /* ProbeResp template for IBSS */ - u32 supp_rates_bits; + u32 supp_rates_bits[IEEE80211_NUM_BANDS]; int wmm_last_param_set; }; @@ -293,6 +290,7 @@ struct ieee80211_if_sta { #define IEEE80211_SDATA_ALLMULTI BIT(0) #define IEEE80211_SDATA_PROMISC BIT(1) #define IEEE80211_SDATA_USERSPACE_MLME BIT(2) +#define IEEE80211_SDATA_OPERATING_GMODE BIT(3) struct ieee80211_sub_if_data { struct list_head list; @@ -313,6 +311,11 @@ struct ieee80211_sub_if_data { */ int ieee802_1x_pac; + /* + * basic rates of this AP or the AP we're associated to + */ + u64 basic_rates; + u16 sequence; /* Fragment table for host-based reassembly */ @@ -420,9 +423,6 @@ struct ieee80211_local { const struct ieee80211_ops *ops; - /* List of registered struct ieee80211_hw_mode */ - struct list_head modes_list; - struct net_device *mdev; /* wmaster# - "master" 802.11 device */ int open_count; int monitors; @@ -462,11 +462,6 @@ struct ieee80211_local { struct rate_control_ref *rate_ctrl; - /* Supported and basic rate filters for different modes. These are - * pointers to -1 terminated lists and rates in 100 kbps units. */ - int *supp_rates[NUM_IEEE80211_MODES]; - int *basic_rates[NUM_IEEE80211_MODES]; - int rts_threshold; int fragmentation_threshold; int short_retry_limit; /* dot11ShortRetryLimit */ @@ -488,12 +483,13 @@ struct ieee80211_local { bool sta_sw_scanning; bool sta_hw_scanning; int scan_channel_idx; + enum ieee80211_band scan_band; + enum { SCAN_SET_CHANNEL, SCAN_SEND_PROBE } scan_state; unsigned long last_scan_completed; struct delayed_work scan_work; struct net_device *scan_dev; struct ieee80211_channel *oper_channel, *scan_channel; - struct ieee80211_hw_mode *oper_hw_mode, *scan_hw_mode; u8 scan_ssid[IEEE80211_MAX_SSID_LEN]; size_t scan_ssid_len; struct list_head sta_bss_list; @@ -562,14 +558,8 @@ struct ieee80211_local { int wifi_wme_noack_test; unsigned int wmm_acm; /* bit field of ACM bits (BIT(802.1D tag)) */ - unsigned int enabled_modes; /* bitfield of allowed modes; - * (1 << MODE_*) */ - unsigned int hw_modes; /* bitfield of supported hardware modes; - * (1 << MODE_*) */ - #ifdef CONFIG_MAC80211_DEBUGFS struct local_debugfsdentries { - struct dentry *channel; struct dentry *frequency; struct dentry *antenna_sel_tx; struct dentry *antenna_sel_rx; @@ -579,9 +569,7 @@ struct ieee80211_local { struct dentry *short_retry_limit; struct dentry *long_retry_limit; struct dentry *total_ps_buffered; - struct dentry *mode; struct dentry *wep_iv; - struct dentry *modes; struct dentry *statistics; struct local_debugfsdentries_statsdentries { struct dentry *transmitted_fragment_count; @@ -692,23 +680,6 @@ static inline void bss_tim_clear(struct ieee80211_local *local, read_unlock_bh(&local->sta_lock); } -/** - * ieee80211_is_erp_rate - Check if a rate is an ERP rate - * @phymode: The PHY-mode for this rate (MODE_IEEE80211...) - * @rate: Transmission rate to check, in 100 kbps - * - * Check if a given rate is an Extended Rate PHY (ERP) rate. - */ -static inline int ieee80211_is_erp_rate(int phymode, int rate) -{ - if (phymode == MODE_IEEE80211G) { - if (rate != 10 && rate != 20 && - rate != 55 && rate != 110) - return 1; - } - return 0; -} - static inline int ieee80211_bssid_match(const u8 *raddr, const u8 *addr) { return compare_ether_addr(raddr, addr) == 0 || @@ -720,13 +691,9 @@ static inline int ieee80211_bssid_match(const u8 *raddr, const u8 *addr) int ieee80211_hw_config(struct ieee80211_local *local); int ieee80211_if_config(struct net_device *dev); int ieee80211_if_config_beacon(struct net_device *dev); -void ieee80211_prepare_rates(struct ieee80211_local *local, - struct ieee80211_hw_mode *mode); void ieee80211_tx_set_iswep(struct ieee80211_txrx_data *tx); int ieee80211_if_update_wds(struct net_device *dev, u8 *remote_addr); void ieee80211_if_setup(struct net_device *dev); -struct ieee80211_rate *ieee80211_get_rate(struct ieee80211_local *local, - int phymode, int hwrate); int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht, struct ieee80211_ht_info *req_ht_cap, struct ieee80211_ht_bss_info *req_bss_cap); @@ -757,7 +724,7 @@ extern const struct iw_handler_def ieee80211_iw_handler_def; /* ieee80211_ioctl.c */ int ieee80211_set_compression(struct ieee80211_local *local, struct net_device *dev, struct sta_info *sta); -int ieee80211_set_channel(struct ieee80211_local *local, int channel, int freq); +int ieee80211_set_freq(struct ieee80211_local *local, int freq); /* ieee80211_sta.c */ void ieee80211_sta_timer(unsigned long data); void ieee80211_sta_work(struct work_struct *work); @@ -810,10 +777,6 @@ int ieee80211_if_remove(struct net_device *dev, const char *name, int id); void ieee80211_if_free(struct net_device *dev); void ieee80211_if_sdata_init(struct ieee80211_sub_if_data *sdata); -/* regdomain.c */ -void ieee80211_regdomain_init(void); -void ieee80211_set_default_regdomain(struct ieee80211_hw_mode *mode); - /* rx handling */ extern ieee80211_rx_handler ieee80211_rx_handlers[]; diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c index 92f1eb2da311..27cee580f9f1 100644 --- a/net/mac80211/ieee80211_iface.c +++ b/net/mac80211/ieee80211_iface.c @@ -118,6 +118,8 @@ void ieee80211_if_set_type(struct net_device *dev, int type) sdata->bss = NULL; sdata->vif.type = type; + sdata->basic_rates = 0; + switch (type) { case IEEE80211_IF_TYPE_WDS: /* nothing special */ diff --git a/net/mac80211/ieee80211_ioctl.c b/net/mac80211/ieee80211_ioctl.c index 5024d3733834..54ad07aafe2d 100644 --- a/net/mac80211/ieee80211_ioctl.c +++ b/net/mac80211/ieee80211_ioctl.c @@ -129,22 +129,7 @@ static int ieee80211_ioctl_giwname(struct net_device *dev, struct iw_request_info *info, char *name, char *extra) { - struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - - switch (local->hw.conf.phymode) { - case MODE_IEEE80211A: - strcpy(name, "IEEE 802.11a"); - break; - case MODE_IEEE80211B: - strcpy(name, "IEEE 802.11b"); - break; - case MODE_IEEE80211G: - strcpy(name, "IEEE 802.11g"); - break; - default: - strcpy(name, "IEEE 802.11"); - break; - } + strcpy(name, "IEEE 802.11"); return 0; } @@ -156,7 +141,7 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev, { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); struct iw_range *range = (struct iw_range *) extra; - struct ieee80211_hw_mode *mode = NULL; + enum ieee80211_band band; int c = 0; data->length = sizeof(struct iw_range); @@ -191,24 +176,27 @@ static int ieee80211_ioctl_giwrange(struct net_device *dev, range->enc_capa = IW_ENC_CAPA_WPA | IW_ENC_CAPA_WPA2 | IW_ENC_CAPA_CIPHER_TKIP | IW_ENC_CAPA_CIPHER_CCMP; - list_for_each_entry(mode, &local->modes_list, list) { - int i = 0; - if (!(local->enabled_modes & (1 << mode->mode)) || - (local->hw_modes & local->enabled_modes & - (1 << MODE_IEEE80211G) && mode->mode == MODE_IEEE80211B)) + for (band = 0; band < IEEE80211_NUM_BANDS; band ++) { + int i; + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[band]; + + if (!sband) continue; - while (i < mode->num_channels && c < IW_MAX_FREQUENCIES) { - struct ieee80211_channel *chan = &mode->channels[i]; + for (i = 0; i < sband->n_channels && c < IW_MAX_FREQUENCIES; i++) { + struct ieee80211_channel *chan = &sband->channels[i]; - if (chan->flag & IEEE80211_CHAN_W_SCAN) { - range->freq[c].i = chan->chan; - range->freq[c].m = chan->freq * 100000; - range->freq[c].e = 1; + if (!(chan->flags & IEEE80211_CHAN_DISABLED)) { + range->freq[c].i = + ieee80211_frequency_to_channel( + chan->center_freq); + range->freq[c].m = chan->center_freq; + range->freq[c].e = 6; c++; } - i++; } } range->num_channels = c; @@ -294,22 +282,29 @@ static int ieee80211_ioctl_giwmode(struct net_device *dev, return 0; } -int ieee80211_set_channel(struct ieee80211_local *local, int channel, int freq) +int ieee80211_set_freq(struct ieee80211_local *local, int freqMHz) { - struct ieee80211_hw_mode *mode; - int c, set = 0; + int set = 0; int ret = -EINVAL; + enum ieee80211_band band; + struct ieee80211_supported_band *sband; + int i; + + for (band = 0; band < IEEE80211_NUM_BANDS; band ++) { + sband = local->hw.wiphy->bands[band]; - list_for_each_entry(mode, &local->modes_list, list) { - if (!(local->enabled_modes & (1 << mode->mode))) + if (!sband) continue; - for (c = 0; c < mode->num_channels; c++) { - struct ieee80211_channel *chan = &mode->channels[c]; - if (chan->flag & IEEE80211_CHAN_W_SCAN && - ((chan->chan == channel) || (chan->freq == freq))) { - local->oper_channel = chan; - local->oper_hw_mode = mode; + + for (i = 0; i < sband->n_channels; i++) { + struct ieee80211_channel *chan = &sband->channels[i]; + + if (chan->flags & IEEE80211_CHAN_DISABLED) + continue; + + if (chan->center_freq == freqMHz) { set = 1; + local->oper_channel = chan; break; } } @@ -347,13 +342,14 @@ static int ieee80211_ioctl_siwfreq(struct net_device *dev, IEEE80211_STA_AUTO_CHANNEL_SEL; return 0; } else - return ieee80211_set_channel(local, freq->m, -1); + return ieee80211_set_freq(local, + ieee80211_channel_to_frequency(freq->m)); } else { int i, div = 1000000; for (i = 0; i < freq->e; i++) div /= 10; if (div > 0) - return ieee80211_set_channel(local, -1, freq->m / div); + return ieee80211_set_freq(local, freq->m / div); else return -EINVAL; } @@ -366,10 +362,7 @@ static int ieee80211_ioctl_giwfreq(struct net_device *dev, { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - /* TODO: in station mode (Managed/Ad-hoc) might need to poll low-level - * driver for the current channel with firmware-based management */ - - freq->m = local->hw.conf.freq; + freq->m = local->hw.conf.channel->center_freq; freq->e = 6; return 0; @@ -566,15 +559,17 @@ static int ieee80211_ioctl_siwrate(struct net_device *dev, struct iw_param *rate, char *extra) { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - struct ieee80211_hw_mode *mode; - int i; + int i, err = -EINVAL; u32 target_rate = rate->value / 100000; struct ieee80211_sub_if_data *sdata; + struct ieee80211_supported_band *sband; sdata = IEEE80211_DEV_TO_SUB_IF(dev); if (!sdata->bss) return -ENODEV; - mode = local->oper_hw_mode; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + /* target_rate = -1, rate->fixed = 0 means auto only, so use all rates * target_rate = X, rate->fixed = 1 means only rate X * target_rate = X, rate->fixed = 0 means all rates <= X */ @@ -582,18 +577,20 @@ static int ieee80211_ioctl_siwrate(struct net_device *dev, sdata->bss->force_unicast_rateidx = -1; if (rate->value < 0) return 0; - for (i=0; i < mode->num_rates; i++) { - struct ieee80211_rate *rates = &mode->rates[i]; - int this_rate = rates->rate; + + for (i=0; i< sband->n_bitrates; i++) { + struct ieee80211_rate *brate = &sband->bitrates[i]; + int this_rate = brate->bitrate; if (target_rate == this_rate) { sdata->bss->max_ratectrl_rateidx = i; if (rate->fixed) sdata->bss->force_unicast_rateidx = i; - return 0; + err = 0; + break; } } - return -EINVAL; + return err; } static int ieee80211_ioctl_giwrate(struct net_device *dev, @@ -603,18 +600,24 @@ static int ieee80211_ioctl_giwrate(struct net_device *dev, struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); struct sta_info *sta; struct ieee80211_sub_if_data *sdata; + struct ieee80211_supported_band *sband; sdata = IEEE80211_DEV_TO_SUB_IF(dev); + if (sdata->vif.type == IEEE80211_IF_TYPE_STA) sta = sta_info_get(local, sdata->u.sta.bssid); else return -EOPNOTSUPP; if (!sta) return -ENODEV; - if (sta->txrate < local->oper_hw_mode->num_rates) - rate->value = local->oper_hw_mode->rates[sta->txrate].rate * 100000; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + + if (sta->txrate_idx < sband->n_bitrates) + rate->value = sband->bitrates[sta->txrate_idx].bitrate; else rate->value = 0; + rate->value *= 100000; sta_info_put(sta); return 0; } @@ -625,7 +628,7 @@ static int ieee80211_ioctl_siwtxpower(struct net_device *dev, { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); bool need_reconfig = 0; - u8 new_power_level; + int new_power_level; if ((data->txpower.flags & IW_TXPOW_TYPE) != IW_TXPOW_DBM) return -EINVAL; @@ -635,13 +638,15 @@ static int ieee80211_ioctl_siwtxpower(struct net_device *dev, if (data->txpower.fixed) { new_power_level = data->txpower.value; } else { - /* Automatic power level. Get the px power from the current - * channel. */ - struct ieee80211_channel* chan = local->oper_channel; + /* + * Automatic power level. Use maximum power for the current + * channel. Should be part of rate control. + */ + struct ieee80211_channel* chan = local->hw.conf.channel; if (!chan) return -EINVAL; - new_power_level = chan->power_level; + new_power_level = chan->max_power; } if (local->hw.conf.power_level != new_power_level) { diff --git a/net/mac80211/ieee80211_rate.c b/net/mac80211/ieee80211_rate.c index b957e67c5fba..ebe29b716b27 100644 --- a/net/mac80211/ieee80211_rate.c +++ b/net/mac80211/ieee80211_rate.c @@ -163,7 +163,8 @@ static void rate_control_release(struct kref *kref) } void rate_control_get_rate(struct net_device *dev, - struct ieee80211_hw_mode *mode, struct sk_buff *skb, + struct ieee80211_supported_band *sband, + struct sk_buff *skb, struct rate_selection *sel) { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); @@ -174,17 +175,17 @@ void rate_control_get_rate(struct net_device *dev, memset(sel, 0, sizeof(struct rate_selection)); - ref->ops->get_rate(ref->priv, dev, mode, skb, sel); + ref->ops->get_rate(ref->priv, dev, sband, skb, sel); /* Select a non-ERP backup rate. */ if (!sel->nonerp) { - for (i = 0; i < mode->num_rates - 1; i++) { - struct ieee80211_rate *rate = &mode->rates[i]; - if (sel->rate->rate < rate->rate) + for (i = 0; i < sband->n_bitrates; i++) { + struct ieee80211_rate *rate = &sband->bitrates[i]; + if (sel->rate->bitrate < rate->bitrate) break; - if (rate_supported(sta, mode, i) && - !(rate->flags & IEEE80211_RATE_ERP)) + if (rate_supported(sta, sband->band, i) && + !(rate->flags & IEEE80211_RATE_ERP_G)) sel->nonerp = rate; } } diff --git a/net/mac80211/ieee80211_rate.h b/net/mac80211/ieee80211_rate.h index 73f19e8aa51c..5f9a2ca49a57 100644 --- a/net/mac80211/ieee80211_rate.h +++ b/net/mac80211/ieee80211_rate.h @@ -18,6 +18,7 @@ #include "ieee80211_i.h" #include "sta_info.h" +/* TODO: kdoc */ struct rate_selection { /* Selected transmission rate */ struct ieee80211_rate *rate; @@ -34,7 +35,8 @@ struct rate_control_ops { struct sk_buff *skb, struct ieee80211_tx_status *status); void (*get_rate)(void *priv, struct net_device *dev, - struct ieee80211_hw_mode *mode, struct sk_buff *skb, + struct ieee80211_supported_band *band, + struct sk_buff *skb, struct rate_selection *sel); void (*rate_init)(void *priv, void *priv_sta, struct ieee80211_local *local, struct sta_info *sta); @@ -66,7 +68,8 @@ void ieee80211_rate_control_unregister(struct rate_control_ops *ops); struct rate_control_ref *rate_control_alloc(const char *name, struct ieee80211_local *local); void rate_control_get_rate(struct net_device *dev, - struct ieee80211_hw_mode *mode, struct sk_buff *skb, + struct ieee80211_supported_band *sband, + struct sk_buff *skb, struct rate_selection *sel); struct rate_control_ref *rate_control_get(struct rate_control_ref *ref); void rate_control_put(struct rate_control_ref *ref); @@ -127,23 +130,23 @@ static inline void rate_control_remove_sta_debugfs(struct sta_info *sta) #endif } -static inline int -rate_supported(struct sta_info *sta, struct ieee80211_hw_mode *mode, int index) +static inline int rate_supported(struct sta_info *sta, + enum ieee80211_band band, + int index) { - return (sta == NULL || sta->supp_rates & BIT(index)) && - (mode->rates[index].flags & IEEE80211_RATE_SUPPORTED); + return (sta == NULL || sta->supp_rates[band] & BIT(index)); } static inline int -rate_lowest_index(struct ieee80211_local *local, struct ieee80211_hw_mode *mode, +rate_lowest_index(struct ieee80211_local *local, + struct ieee80211_supported_band *sband, struct sta_info *sta) { int i; - for (i = 0; i < mode->num_rates; i++) { - if (rate_supported(sta, mode, i)) + for (i = 0; i < sband->n_bitrates; i++) + if (rate_supported(sta, sband->band, i)) return i; - } /* warn when we cannot find a rate. */ WARN_ON(1); @@ -152,10 +155,11 @@ rate_lowest_index(struct ieee80211_local *local, struct ieee80211_hw_mode *mode, } static inline struct ieee80211_rate * -rate_lowest(struct ieee80211_local *local, struct ieee80211_hw_mode *mode, +rate_lowest(struct ieee80211_local *local, + struct ieee80211_supported_band *sband, struct sta_info *sta) { - return &mode->rates[rate_lowest_index(local, mode, sta)]; + return &sband->bitrates[rate_lowest_index(local, sband, sta)]; } diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index d0273ccbdbae..2628222a5085 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -74,7 +74,7 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst, u8 *ssid, size_t ssid_len); static struct ieee80211_sta_bss * -ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int channel, +ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int freq, u8 *ssid, u8 ssid_len); static void ieee80211_rx_bss_put(struct net_device *dev, struct ieee80211_sta_bss *bss); @@ -466,7 +466,7 @@ static void ieee80211_set_associated(struct net_device *dev, return; bss = ieee80211_rx_bss_get(dev, ifsta->bssid, - local->hw.conf.channel, + local->hw.conf.channel->center_freq, ifsta->ssid, ifsta->ssid_len); if (bss) { if (bss->has_erp_value) @@ -593,7 +593,6 @@ static void ieee80211_send_assoc(struct net_device *dev, struct ieee80211_if_sta *ifsta) { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - struct ieee80211_hw_mode *mode; struct sk_buff *skb; struct ieee80211_mgmt *mgmt; u8 *pos, *ies; @@ -601,6 +600,7 @@ static void ieee80211_send_assoc(struct net_device *dev, u16 capab; struct ieee80211_sta_bss *bss; int wmm = 0; + struct ieee80211_supported_band *sband; skb = dev_alloc_skb(local->hw.extra_tx_headroom + sizeof(*mgmt) + 200 + ifsta->extra_ie_len + @@ -612,13 +612,19 @@ static void ieee80211_send_assoc(struct net_device *dev, } skb_reserve(skb, local->hw.extra_tx_headroom); - mode = local->oper_hw_mode; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + capab = ifsta->capab; - if (mode->mode == MODE_IEEE80211G) { - capab |= WLAN_CAPABILITY_SHORT_SLOT_TIME | - WLAN_CAPABILITY_SHORT_PREAMBLE; + + if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ) { + if (!(local->hw.flags & IEEE80211_HW_2GHZ_SHORT_SLOT_INCAPABLE)) + capab |= WLAN_CAPABILITY_SHORT_SLOT_TIME; + if (!(local->hw.flags & IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE)) + capab |= WLAN_CAPABILITY_SHORT_PREAMBLE; } - bss = ieee80211_rx_bss_get(dev, ifsta->bssid, local->hw.conf.channel, + + bss = ieee80211_rx_bss_get(dev, ifsta->bssid, + local->hw.conf.channel->center_freq, ifsta->ssid, ifsta->ssid_len); if (bss) { if (bss->capability & WLAN_CAPABILITY_PRIVACY) @@ -657,23 +663,23 @@ static void ieee80211_send_assoc(struct net_device *dev, *pos++ = ifsta->ssid_len; memcpy(pos, ifsta->ssid, ifsta->ssid_len); - len = mode->num_rates; + len = sband->n_bitrates; if (len > 8) len = 8; pos = skb_put(skb, len + 2); *pos++ = WLAN_EID_SUPP_RATES; *pos++ = len; for (i = 0; i < len; i++) { - int rate = mode->rates[i].rate; + int rate = sband->bitrates[i].bitrate; *pos++ = (u8) (rate / 5); } - if (mode->num_rates > len) { - pos = skb_put(skb, mode->num_rates - len + 2); + if (sband->n_bitrates > len) { + pos = skb_put(skb, sband->n_bitrates - len + 2); *pos++ = WLAN_EID_EXT_SUPP_RATES; - *pos++ = mode->num_rates - len; - for (i = len; i < mode->num_rates; i++) { - int rate = mode->rates[i].rate; + *pos++ = sband->n_bitrates - len; + for (i = len; i < sband->n_bitrates; i++) { + int rate = sband->bitrates[i].bitrate; *pos++ = (u8) (rate / 5); } } @@ -696,17 +702,18 @@ static void ieee80211_send_assoc(struct net_device *dev, *pos++ = 0; } /* wmm support is a must to HT */ - if (wmm && mode->ht_info.ht_supported) { - __le16 tmp = cpu_to_le16(mode->ht_info.cap); + if (wmm && sband->ht_info.ht_supported) { + __le16 tmp = cpu_to_le16(sband->ht_info.cap); pos = skb_put(skb, sizeof(struct ieee80211_ht_cap)+2); *pos++ = WLAN_EID_HT_CAPABILITY; *pos++ = sizeof(struct ieee80211_ht_cap); memset(pos, 0, sizeof(struct ieee80211_ht_cap)); memcpy(pos, &tmp, sizeof(u16)); pos += sizeof(u16); - *pos++ = (mode->ht_info.ampdu_factor | - (mode->ht_info.ampdu_density << 2)); - memcpy(pos, mode->ht_info.supp_mcs_set, 16); + /* TODO: needs a define here for << 2 */ + *pos++ = sband->ht_info.ampdu_factor | + (sband->ht_info.ampdu_density << 2); + memcpy(pos, sband->ht_info.supp_mcs_set, 16); } kfree(ifsta->assocreq_ies); @@ -789,7 +796,8 @@ static int ieee80211_privacy_mismatch(struct net_device *dev, if (!ifsta || (ifsta->flags & IEEE80211_STA_MIXED_CELL)) return 0; - bss = ieee80211_rx_bss_get(dev, ifsta->bssid, local->hw.conf.channel, + bss = ieee80211_rx_bss_get(dev, ifsta->bssid, + local->hw.conf.channel->center_freq, ifsta->ssid, ifsta->ssid_len); if (!bss) return 0; @@ -899,7 +907,7 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst, u8 *ssid, size_t ssid_len) { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; struct sk_buff *skb; struct ieee80211_mgmt *mgmt; u8 *pos, *supp_rates, *esupp_rates = NULL; @@ -933,11 +941,10 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst, supp_rates = skb_put(skb, 2); supp_rates[0] = WLAN_EID_SUPP_RATES; supp_rates[1] = 0; - mode = local->oper_hw_mode; - for (i = 0; i < mode->num_rates; i++) { - struct ieee80211_rate *rate = &mode->rates[i]; - if (!(rate->flags & IEEE80211_RATE_SUPPORTED)) - continue; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + + for (i = 0; i < sband->n_bitrates; i++) { + struct ieee80211_rate *rate = &sband->bitrates[i]; if (esupp_rates) { pos = skb_put(skb, 1); esupp_rates[1]++; @@ -950,7 +957,7 @@ static void ieee80211_send_probe_req(struct net_device *dev, u8 *dst, pos = skb_put(skb, 1); supp_rates[1]++; } - *pos = rate->rate / 5; + *pos = rate->bitrate / 5; } ieee80211_sta_tx(dev, skb, 0); @@ -1146,9 +1153,11 @@ static void ieee80211_sta_process_addba_request(struct net_device *dev, } /* determine default buffer size */ if (buf_size == 0) { - struct ieee80211_hw_mode *mode = conf->mode; + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[conf->channel->band]; buf_size = IEEE80211_MIN_AMPDU_BUF; - buf_size = buf_size << mode->ht_info.ampdu_factor; + buf_size = buf_size << sband->ht_info.ampdu_factor; } tid_agg_rx = &sta->ampdu_mlme.tid_rx[tid]; @@ -1718,15 +1727,16 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, { struct ieee80211_local *local = sdata->local; struct net_device *dev = sdata->dev; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; struct sta_info *sta; - u32 rates; + u64 rates, basic_rates; u16 capab_info, status_code, aid; struct ieee802_11_elems elems; struct ieee80211_bss_conf *bss_conf = &sdata->bss_conf; u8 *pos; int i, j; DECLARE_MAC_BUF(mac); + bool have_higher_than_11mbit = false; /* AssocResp and ReassocResp have identical structure, so process both * of them in this function. */ @@ -1796,10 +1806,6 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, if (ifsta->assocresp_ies) memcpy(ifsta->assocresp_ies, pos, ifsta->assocresp_ies_len); - /* set AID, ieee80211_set_associated() will tell the driver */ - bss_conf->aid = aid; - ieee80211_set_associated(dev, ifsta, 1); - /* Add STA entry for the AP */ sta = sta_info_get(local, ifsta->bssid); if (!sta) { @@ -1811,7 +1817,7 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, return; } bss = ieee80211_rx_bss_get(dev, ifsta->bssid, - local->hw.conf.channel, + local->hw.conf.channel->center_freq, ifsta->ssid, ifsta->ssid_len); if (bss) { sta->last_rssi = bss->rssi; @@ -1825,20 +1831,46 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, sta->flags |= WLAN_STA_AUTH | WLAN_STA_ASSOC | WLAN_STA_ASSOC_AP; rates = 0; - mode = local->oper_hw_mode; + basic_rates = 0; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + for (i = 0; i < elems.supp_rates_len; i++) { int rate = (elems.supp_rates[i] & 0x7f) * 5; - for (j = 0; j < mode->num_rates; j++) - if (mode->rates[j].rate == rate) + + if (rate > 110) + have_higher_than_11mbit = true; + + for (j = 0; j < sband->n_bitrates; j++) { + if (sband->bitrates[j].bitrate == rate) rates |= BIT(j); + if (elems.supp_rates[i] & 0x80) + basic_rates |= BIT(j); + } } + for (i = 0; i < elems.ext_supp_rates_len; i++) { int rate = (elems.ext_supp_rates[i] & 0x7f) * 5; - for (j = 0; j < mode->num_rates; j++) - if (mode->rates[j].rate == rate) + + if (rate > 110) + have_higher_than_11mbit = true; + + for (j = 0; j < sband->n_bitrates; j++) { + if (sband->bitrates[j].bitrate == rate) rates |= BIT(j); + if (elems.ext_supp_rates[i] & 0x80) + basic_rates |= BIT(j); + } } - sta->supp_rates = rates; + + sta->supp_rates[local->hw.conf.channel->band] = rates; + sdata->basic_rates = basic_rates; + + /* cf. IEEE 802.11 9.2.12 */ + if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ && + have_higher_than_11mbit) + sdata->flags |= IEEE80211_SDATA_OPERATING_GMODE; + else + sdata->flags &= ~IEEE80211_SDATA_OPERATING_GMODE; if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param && local->ops->conf_ht) { @@ -1861,6 +1893,9 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, elems.wmm_param_len); } + /* set AID, ieee80211_set_associated() will tell the driver */ + bss_conf->aid = aid; + ieee80211_set_associated(dev, ifsta, 1); sta_info_put(sta); @@ -1901,7 +1936,7 @@ static void __ieee80211_rx_bss_hash_del(struct net_device *dev, static struct ieee80211_sta_bss * -ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int channel, +ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int freq, u8 *ssid, u8 ssid_len) { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); @@ -1913,7 +1948,7 @@ ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int channel, atomic_inc(&bss->users); atomic_inc(&bss->users); memcpy(bss->bssid, bssid, ETH_ALEN); - bss->channel = channel; + bss->freq = freq; if (ssid && ssid_len <= IEEE80211_MAX_SSID_LEN) { memcpy(bss->ssid, ssid, ssid_len); bss->ssid_len = ssid_len; @@ -1929,7 +1964,7 @@ ieee80211_rx_bss_add(struct net_device *dev, u8 *bssid, int channel, static struct ieee80211_sta_bss * -ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int channel, +ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int freq, u8 *ssid, u8 ssid_len) { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); @@ -1939,7 +1974,7 @@ ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int channel, bss = local->sta_bss_hash[STA_HASH(bssid)]; while (bss) { if (!memcmp(bss->bssid, bssid, ETH_ALEN) && - bss->channel == channel && + bss->freq == freq && bss->ssid_len == ssid_len && (ssid_len == 0 || !memcmp(bss->ssid, ssid, ssid_len))) { atomic_inc(&bss->users); @@ -2004,7 +2039,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev, struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); struct ieee802_11_elems elems; size_t baselen; - int channel, clen; + int freq, clen; struct ieee80211_sta_bss *bss; struct sta_info *sta; struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); @@ -2055,26 +2090,22 @@ static void ieee80211_rx_bss_info(struct net_device *dev, if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates && memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0 && (sta = sta_info_get(local, mgmt->sa))) { - struct ieee80211_hw_mode *mode; - struct ieee80211_rate *rates; + struct ieee80211_supported_band *sband; + struct ieee80211_rate *bitrates; size_t num_rates; - u32 supp_rates, prev_rates; + u64 supp_rates, prev_rates; int i, j; - mode = local->sta_sw_scanning ? - local->scan_hw_mode : local->oper_hw_mode; - - if (local->sta_hw_scanning) { - /* search for the correct mode matches the beacon */ - list_for_each_entry(mode, &local->modes_list, list) - if (mode->mode == rx_status->phymode) - break; + sband = local->hw.wiphy->bands[rx_status->band]; - if (mode == NULL) - mode = local->oper_hw_mode; + if (!sband) { + WARN_ON(1); + sband = local->hw.wiphy->bands[ + local->hw.conf.channel->band]; } - rates = mode->rates; - num_rates = mode->num_rates; + + bitrates = sband->bitrates; + num_rates = sband->n_bitrates; supp_rates = 0; for (i = 0; i < elems.supp_rates_len + @@ -2088,24 +2119,27 @@ static void ieee80211_rx_bss_info(struct net_device *dev, [i - elems.supp_rates_len]; own_rate = 5 * (rate & 0x7f); for (j = 0; j < num_rates; j++) - if (rates[j].rate == own_rate) + if (bitrates[j].bitrate == own_rate) supp_rates |= BIT(j); } - prev_rates = sta->supp_rates; - sta->supp_rates &= supp_rates; - if (sta->supp_rates == 0) { + prev_rates = sta->supp_rates[rx_status->band]; + sta->supp_rates[rx_status->band] &= supp_rates; + if (sta->supp_rates[rx_status->band] == 0) { /* No matching rates - this should not really happen. * Make sure that at least one rate is marked * supported to avoid issues with TX rate ctrl. */ - sta->supp_rates = sdata->u.sta.supp_rates_bits; + sta->supp_rates[rx_status->band] = + sdata->u.sta.supp_rates_bits[rx_status->band]; } - if (sta->supp_rates != prev_rates) { + if (sta->supp_rates[rx_status->band] != prev_rates) { printk(KERN_DEBUG "%s: updated supp_rates set for " - "%s based on beacon info (0x%x & 0x%x -> " - "0x%x)\n", - dev->name, print_mac(mac, sta->addr), prev_rates, - supp_rates, sta->supp_rates); + "%s based on beacon info (0x%llx & 0x%llx -> " + "0x%llx)\n", + dev->name, print_mac(mac, sta->addr), + (unsigned long long) prev_rates, + (unsigned long long) supp_rates, + (unsigned long long) sta->supp_rates[rx_status->band]); } sta_info_put(sta); } @@ -2114,14 +2148,14 @@ static void ieee80211_rx_bss_info(struct net_device *dev, return; if (elems.ds_params && elems.ds_params_len == 1) - channel = elems.ds_params[0]; + freq = ieee80211_channel_to_frequency(elems.ds_params[0]); else - channel = rx_status->channel; + freq = rx_status->freq; - bss = ieee80211_rx_bss_get(dev, mgmt->bssid, channel, + bss = ieee80211_rx_bss_get(dev, mgmt->bssid, freq, elems.ssid, elems.ssid_len); if (!bss) { - bss = ieee80211_rx_bss_add(dev, mgmt->bssid, channel, + bss = ieee80211_rx_bss_add(dev, mgmt->bssid, freq, elems.ssid, elems.ssid_len); if (!bss) return; @@ -2134,6 +2168,8 @@ static void ieee80211_rx_bss_info(struct net_device *dev, #endif } + bss->band = rx_status->band; + if (bss->probe_resp && beacon) { /* Do not allow beacon to override data from Probe Response. */ ieee80211_rx_bss_put(dev, bss); @@ -2232,20 +2268,6 @@ static void ieee80211_rx_bss_info(struct net_device *dev, bss->ht_ie_len = 0; } - bss->hw_mode = rx_status->phymode; - bss->freq = rx_status->freq; - if (channel != rx_status->channel && - (bss->hw_mode == MODE_IEEE80211G || - bss->hw_mode == MODE_IEEE80211B) && - channel >= 1 && channel <= 14) { - static const int freq_list[] = { - 2412, 2417, 2422, 2427, 2432, 2437, 2442, - 2447, 2452, 2457, 2462, 2467, 2472, 2484 - }; - /* IEEE 802.11g/b mode can receive packets from neighboring - * channels, so map the channel into frequency. */ - bss->freq = freq_list[channel - 1]; - } bss->timestamp = timestamp; bss->last_update = jiffies; bss->rssi = rx_status->ssi; @@ -2817,7 +2839,7 @@ static int ieee80211_sta_config_auth(struct net_device *dev, } spin_lock_bh(&local->sta_bss_lock); - freq = local->oper_channel->freq; + freq = local->oper_channel->center_freq; list_for_each_entry(bss, &local->sta_bss_list, list) { if (!(bss->capability & WLAN_CAPABILITY_ESS)) continue; @@ -2848,7 +2870,7 @@ static int ieee80211_sta_config_auth(struct net_device *dev, spin_unlock_bh(&local->sta_bss_lock); if (selected) { - ieee80211_set_channel(local, -1, selected->freq); + ieee80211_set_freq(local, selected->freq); if (!(ifsta->flags & IEEE80211_STA_SSID_SET)) ieee80211_sta_set_ssid(dev, selected->ssid, selected->ssid_len); @@ -2881,10 +2903,12 @@ static int ieee80211_sta_join_ibss(struct net_device *dev, struct sk_buff *skb; struct ieee80211_mgmt *mgmt; struct ieee80211_tx_control control; - struct ieee80211_hw_mode *mode; struct rate_selection ratesel; u8 *pos; struct ieee80211_sub_if_data *sdata; + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; /* Remove possible STA entries from other IBSS networks. */ sta_info_flush(local, NULL); @@ -2904,12 +2928,11 @@ static int ieee80211_sta_join_ibss(struct net_device *dev, sdata->drop_unencrypted = bss->capability & WLAN_CAPABILITY_PRIVACY ? 1 : 0; - res = ieee80211_set_channel(local, -1, bss->freq); + res = ieee80211_set_freq(local, bss->freq); - if (!(local->oper_channel->flag & IEEE80211_CHAN_W_IBSS)) { - printk(KERN_DEBUG "%s: IBSS not allowed on channel %d " - "(%d MHz)\n", dev->name, local->hw.conf.channel, - local->hw.conf.freq); + if (local->oper_channel->flags & IEEE80211_CHAN_NO_IBSS) { + printk(KERN_DEBUG "%s: IBSS not allowed on frequency " + "%d MHz\n", dev->name, local->oper_channel->center_freq); return -1; } @@ -2946,10 +2969,12 @@ static int ieee80211_sta_join_ibss(struct net_device *dev, *pos++ = rates; memcpy(pos, bss->supp_rates, rates); - pos = skb_put(skb, 2 + 1); - *pos++ = WLAN_EID_DS_PARAMS; - *pos++ = 1; - *pos++ = bss->channel; + if (bss->band == IEEE80211_BAND_2GHZ) { + pos = skb_put(skb, 2 + 1); + *pos++ = WLAN_EID_DS_PARAMS; + *pos++ = 1; + *pos++ = ieee80211_frequency_to_channel(bss->freq); + } pos = skb_put(skb, 2 + 2); *pos++ = WLAN_EID_IBSS_PARAMS; @@ -2967,19 +2992,18 @@ static int ieee80211_sta_join_ibss(struct net_device *dev, } memset(&control, 0, sizeof(control)); - rate_control_get_rate(dev, local->oper_hw_mode, skb, &ratesel); + rate_control_get_rate(dev, sband, skb, &ratesel); if (!ratesel.rate) { printk(KERN_DEBUG "%s: Failed to determine TX rate " "for IBSS beacon\n", dev->name); break; } control.vif = &sdata->vif; - control.tx_rate = - (sdata->bss_conf.use_short_preamble && - (ratesel.rate->flags & IEEE80211_RATE_PREAMBLE2)) ? - ratesel.rate->val2 : ratesel.rate->val; + control.tx_rate = ratesel.rate; + if (sdata->bss_conf.use_short_preamble && + ratesel.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) + control.flags |= IEEE80211_TXCTL_SHORT_PREAMBLE; control.antenna_sel_tx = local->hw.conf.antenna_sel_tx; - control.power_level = local->hw.conf.power_level; control.flags |= IEEE80211_TXCTL_NO_ACK; control.retry_limit = 1; @@ -3004,14 +3028,14 @@ static int ieee80211_sta_join_ibss(struct net_device *dev, } rates = 0; - mode = local->oper_hw_mode; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; for (i = 0; i < bss->supp_rates_len; i++) { int bitrate = (bss->supp_rates[i] & 0x7f) * 5; - for (j = 0; j < mode->num_rates; j++) - if (mode->rates[j].rate == bitrate) + for (j = 0; j < sband->n_bitrates; j++) + if (sband->bitrates[j].bitrate == bitrate) rates |= BIT(j); } - ifsta->supp_rates_bits = rates; + ifsta->supp_rates_bits[local->hw.conf.channel->band] = rates; } while (0); if (skb) { @@ -3035,7 +3059,7 @@ static int ieee80211_sta_create_ibss(struct net_device *dev, struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); struct ieee80211_sta_bss *bss; struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; u8 bssid[ETH_ALEN], *pos; int i; DECLARE_MAC_BUF(mac); @@ -3057,28 +3081,28 @@ static int ieee80211_sta_create_ibss(struct net_device *dev, printk(KERN_DEBUG "%s: Creating new IBSS network, BSSID %s\n", dev->name, print_mac(mac, bssid)); - bss = ieee80211_rx_bss_add(dev, bssid, local->hw.conf.channel, + bss = ieee80211_rx_bss_add(dev, bssid, + local->hw.conf.channel->center_freq, sdata->u.sta.ssid, sdata->u.sta.ssid_len); if (!bss) return -ENOMEM; - mode = local->oper_hw_mode; + bss->band = local->hw.conf.channel->band; + sband = local->hw.wiphy->bands[bss->band]; if (local->hw.conf.beacon_int == 0) local->hw.conf.beacon_int = 100; bss->beacon_int = local->hw.conf.beacon_int; - bss->hw_mode = local->hw.conf.phymode; - bss->freq = local->hw.conf.freq; bss->last_update = jiffies; bss->capability = WLAN_CAPABILITY_IBSS; if (sdata->default_key) { bss->capability |= WLAN_CAPABILITY_PRIVACY; } else sdata->drop_unencrypted = 0; - bss->supp_rates_len = mode->num_rates; + bss->supp_rates_len = sband->n_bitrates; pos = bss->supp_rates; - for (i = 0; i < mode->num_rates; i++) { - int rate = mode->rates[i].rate; + for (i = 0; i < sband->n_bitrates; i++) { + int rate = sband->bitrates[i].bitrate; *pos++ = (u8) (rate / 5); } @@ -3127,7 +3151,8 @@ static int ieee80211_sta_find_ibss(struct net_device *dev, "%s\n", print_mac(mac, bssid), print_mac(mac2, ifsta->bssid)); #endif /* CONFIG_MAC80211_IBSS_DEBUG */ if (found && memcmp(ifsta->bssid, bssid, ETH_ALEN) != 0 && - (bss = ieee80211_rx_bss_get(dev, bssid, local->hw.conf.channel, + (bss = ieee80211_rx_bss_get(dev, bssid, + local->hw.conf.channel->center_freq, ifsta->ssid, ifsta->ssid_len))) { printk(KERN_DEBUG "%s: Selected IBSS BSSID %s" " based on configured SSID\n", @@ -3155,13 +3180,13 @@ static int ieee80211_sta_find_ibss(struct net_device *dev, if (time_after(jiffies, ifsta->ibss_join_req + IEEE80211_IBSS_JOIN_TIMEOUT)) { if ((ifsta->flags & IEEE80211_STA_CREATE_IBSS) && - local->oper_channel->flag & IEEE80211_CHAN_W_IBSS) + (!(local->oper_channel->flags & + IEEE80211_CHAN_NO_IBSS))) return ieee80211_sta_create_ibss(dev, ifsta); if (ifsta->flags & IEEE80211_STA_CREATE_IBSS) { - printk(KERN_DEBUG "%s: IBSS not allowed on the" - " configured channel %d (%d MHz)\n", - dev->name, local->hw.conf.channel, - local->hw.conf.freq); + printk(KERN_DEBUG "%s: IBSS not allowed on" + " %d MHz\n", dev->name, + local->hw.conf.channel->center_freq); } /* No IBSS found - decrease scan interval and continue @@ -3180,7 +3205,7 @@ static int ieee80211_sta_find_ibss(struct net_device *dev, int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len) { - struct ieee80211_sub_if_data *sdata; + struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_if_sta *ifsta; struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); @@ -3194,18 +3219,23 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len) int i; memset(&qparam, 0, sizeof(qparam)); - /* TODO: are these ok defaults for all hw_modes? */ + qparam.aifs = 2; - qparam.cw_min = - local->hw.conf.phymode == MODE_IEEE80211B ? 31 : 15; + + if (local->hw.conf.channel->band == IEEE80211_BAND_2GHZ && + !(sdata->flags & IEEE80211_SDATA_OPERATING_GMODE)) + qparam.cw_min = 31; + else + qparam.cw_min = 15; + qparam.cw_max = 1023; qparam.burst_time = 0; + for (i = IEEE80211_TX_QUEUE_DATA0; i < NUM_TX_DATA_QUEUES; i++) - { local->ops->conf_tx(local_to_hw(local), i + IEEE80211_TX_QUEUE_DATA0, &qparam); - } + /* IBSS uses different parameters for Beacon sending */ qparam.cw_min++; qparam.cw_min *= 2; @@ -3214,7 +3244,6 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len) IEEE80211_TX_QUEUE_BEACON, &qparam); } - sdata = IEEE80211_DEV_TO_SUB_IF(dev); ifsta = &sdata->u.sta; if (ifsta->ssid_len != len || memcmp(ifsta->ssid, ssid, len) != 0) @@ -3373,7 +3402,7 @@ void ieee80211_sta_scan_work(struct work_struct *work) container_of(work, struct ieee80211_local, scan_work.work); struct net_device *dev = local->scan_dev; struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; struct ieee80211_channel *chan; int skip; unsigned long next_delay = 0; @@ -3383,44 +3412,47 @@ void ieee80211_sta_scan_work(struct work_struct *work) switch (local->scan_state) { case SCAN_SET_CHANNEL: - mode = local->scan_hw_mode; - if (local->scan_hw_mode->list.next == &local->modes_list && - local->scan_channel_idx >= mode->num_channels) { + /* get current scan band */ + if (local->scan_band < IEEE80211_NUM_BANDS) + sband = local->hw.wiphy->bands[local->scan_band]; + else + sband = NULL; + + /* if we started at an unsupported one, advance */ + while (!sband && local->scan_band < IEEE80211_NUM_BANDS) { + local->scan_band++; + sband = local->hw.wiphy->bands[local->scan_band]; + local->scan_channel_idx = 0; + } + + if (!sband || + (local->scan_channel_idx >= sband->n_channels && + local->scan_band >= IEEE80211_NUM_BANDS)) { ieee80211_scan_completed(local_to_hw(local)); return; } - skip = !(local->enabled_modes & (1 << mode->mode)); - chan = &mode->channels[local->scan_channel_idx]; - if (!(chan->flag & IEEE80211_CHAN_W_SCAN) || + skip = 0; + chan = &sband->channels[local->scan_channel_idx]; + + if (chan->flags & IEEE80211_CHAN_DISABLED || (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && - !(chan->flag & IEEE80211_CHAN_W_IBSS)) || - (local->hw_modes & local->enabled_modes & - (1 << MODE_IEEE80211G) && mode->mode == MODE_IEEE80211B)) + chan->flags & IEEE80211_CHAN_NO_IBSS)) skip = 1; if (!skip) { -#if 0 - printk(KERN_DEBUG "%s: scan channel %d (%d MHz)\n", - dev->name, chan->chan, chan->freq); -#endif - local->scan_channel = chan; if (ieee80211_hw_config(local)) { - printk(KERN_DEBUG "%s: failed to set channel " - "%d (%d MHz) for scan\n", dev->name, - chan->chan, chan->freq); + printk(KERN_DEBUG "%s: failed to set freq to " + "%d MHz for scan\n", dev->name, + chan->center_freq); skip = 1; } } local->scan_channel_idx++; - if (local->scan_channel_idx >= local->scan_hw_mode->num_channels) { - if (local->scan_hw_mode->list.next != &local->modes_list) { - local->scan_hw_mode = list_entry(local->scan_hw_mode->list.next, - struct ieee80211_hw_mode, - list); - local->scan_channel_idx = 0; - } + if (local->scan_channel_idx >= sband->n_channels) { + local->scan_band++; + local->scan_channel_idx = 0; } if (skip) @@ -3431,13 +3463,14 @@ void ieee80211_sta_scan_work(struct work_struct *work) local->scan_state = SCAN_SEND_PROBE; break; case SCAN_SEND_PROBE: - if (local->scan_channel->flag & IEEE80211_CHAN_W_ACTIVE_SCAN) { - ieee80211_send_probe_req(dev, NULL, local->scan_ssid, - local->scan_ssid_len); - next_delay = IEEE80211_CHANNEL_TIME; - } else - next_delay = IEEE80211_PASSIVE_CHANNEL_TIME; + next_delay = IEEE80211_PASSIVE_CHANNEL_TIME; local->scan_state = SCAN_SET_CHANNEL; + + if (local->scan_channel->flags & IEEE80211_CHAN_PASSIVE_SCAN) + break; + ieee80211_send_probe_req(dev, NULL, local->scan_ssid, + local->scan_ssid_len); + next_delay = IEEE80211_CHANNEL_TIME; break; } @@ -3512,10 +3545,8 @@ static int ieee80211_sta_start_scan(struct net_device *dev, } else local->scan_ssid_len = 0; local->scan_state = SCAN_SET_CHANNEL; - local->scan_hw_mode = list_entry(local->modes_list.next, - struct ieee80211_hw_mode, - list); local->scan_channel_idx = 0; + local->scan_band = IEEE80211_BAND_2GHZ; local->scan_dev = dev; netif_tx_lock_bh(local->mdev); @@ -3570,9 +3601,6 @@ ieee80211_sta_scan_result(struct net_device *dev, bss->last_update + IEEE80211_SCAN_RESULT_EXPIRE)) return current_ev; - if (!(local->enabled_modes & (1 << bss->hw_mode))) - return current_ev; - memset(&iwe, 0, sizeof(iwe)); iwe.cmd = SIOCGIWAP; iwe.u.ap_addr.sa_family = ARPHRD_ETHER; @@ -3600,12 +3628,15 @@ ieee80211_sta_scan_result(struct net_device *dev, memset(&iwe, 0, sizeof(iwe)); iwe.cmd = SIOCGIWFREQ; - iwe.u.freq.m = bss->channel; - iwe.u.freq.e = 0; + iwe.u.freq.m = bss->freq; + iwe.u.freq.e = 6; current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN); - iwe.u.freq.m = bss->freq * 100000; - iwe.u.freq.e = 1; + + memset(&iwe, 0, sizeof(iwe)); + iwe.cmd = SIOCGIWFREQ; + iwe.u.freq.m = ieee80211_frequency_to_channel(bss->freq); + iwe.u.freq.e = 0; current_ev = iwe_stream_add_event(current_ev, end_buf, &iwe, IW_EV_FREQ_LEN); @@ -3748,7 +3779,8 @@ struct sta_info * ieee80211_ibss_add_sta(struct net_device *dev, if (!sta) return NULL; - sta->supp_rates = sdata->u.sta.supp_rates_bits; + sta->supp_rates[local->hw.conf.channel->band] = + sdata->u.sta.supp_rates_bits[local->hw.conf.channel->band]; rate_control_rate_init(sta, local); diff --git a/net/mac80211/rc80211_pid_algo.c b/net/mac80211/rc80211_pid_algo.c index c339571632b2..c5a607ca8440 100644 --- a/net/mac80211/rc80211_pid_algo.c +++ b/net/mac80211/rc80211_pid_algo.c @@ -102,23 +102,23 @@ static void rate_control_pid_adjust_rate(struct ieee80211_local *local, struct rc_pid_rateinfo *rinfo) { struct ieee80211_sub_if_data *sdata; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; int newidx; int maxrate; int back = (adj > 0) ? 1 : -1; sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); - mode = local->oper_hw_mode; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; maxrate = sdata->bss ? sdata->bss->max_ratectrl_rateidx : -1; - newidx = rate_control_pid_shift_adjust(rinfo, adj, sta->txrate, - mode->num_rates); + newidx = rate_control_pid_shift_adjust(rinfo, adj, sta->txrate_idx, + sband->n_bitrates); - while (newidx != sta->txrate) { - if (rate_supported(sta, mode, newidx) && + while (newidx != sta->txrate_idx) { + if (rate_supported(sta, sband->band, newidx) && (maxrate < 0 || newidx <= maxrate)) { - sta->txrate = newidx; + sta->txrate_idx = newidx; break; } @@ -128,7 +128,7 @@ static void rate_control_pid_adjust_rate(struct ieee80211_local *local, #ifdef CONFIG_MAC80211_DEBUGFS rate_control_pid_event_rate_change( &((struct rc_pid_sta_info *)sta->rate_ctrl_priv)->events, - newidx, mode->rates[newidx].rate); + newidx, sband->bitrates[newidx].bitrate); #endif } @@ -155,7 +155,7 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo, { struct rc_pid_sta_info *spinfo = sta->rate_ctrl_priv; struct rc_pid_rateinfo *rinfo = pinfo->rinfo; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; u32 pf; s32 err_avg; u32 err_prop; @@ -164,7 +164,7 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo, int adj, i, j, tmp; unsigned long period; - mode = local->oper_hw_mode; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; spinfo = sta->rate_ctrl_priv; /* In case nothing happened during the previous control interval, turn @@ -190,18 +190,18 @@ static void rate_control_pid_sample(struct rc_pid_info *pinfo, spinfo->tx_num_failed = 0; /* If we just switched rate, update the rate behaviour info. */ - if (pinfo->oldrate != sta->txrate) { + if (pinfo->oldrate != sta->txrate_idx) { i = rinfo[pinfo->oldrate].rev_index; - j = rinfo[sta->txrate].rev_index; + j = rinfo[sta->txrate_idx].rev_index; tmp = (pf - spinfo->last_pf); tmp = RC_PID_DO_ARITH_RIGHT_SHIFT(tmp, RC_PID_ARITH_SHIFT); rinfo[j].diff = rinfo[i].diff + tmp; - pinfo->oldrate = sta->txrate; + pinfo->oldrate = sta->txrate_idx; } - rate_control_pid_normalize(pinfo, mode->num_rates); + rate_control_pid_normalize(pinfo, sband->n_bitrates); /* Compute the proportional, integral and derivative errors. */ err_prop = (pinfo->target << RC_PID_ARITH_SHIFT) - pf; @@ -242,8 +242,10 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev, struct sta_info *sta; struct rc_pid_sta_info *spinfo; unsigned long period; + struct ieee80211_supported_band *sband; sta = sta_info_get(local, hdr->addr1); + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; if (!sta) return; @@ -251,13 +253,13 @@ static void rate_control_pid_tx_status(void *priv, struct net_device *dev, /* Don't update the state if we're not controlling the rate. */ sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); if (sdata->bss && sdata->bss->force_unicast_rateidx > -1) { - sta->txrate = sdata->bss->max_ratectrl_rateidx; + sta->txrate_idx = sdata->bss->max_ratectrl_rateidx; return; } /* Ignore all frames that were sent with a different rate than the rate * we currently advise mac80211 to use. */ - if (status->control.rate != &local->oper_hw_mode->rates[sta->txrate]) + if (status->control.tx_rate != &sband->bitrates[sta->txrate_idx]) goto ignore; spinfo = sta->rate_ctrl_priv; @@ -304,7 +306,7 @@ ignore: } static void rate_control_pid_get_rate(void *priv, struct net_device *dev, - struct ieee80211_hw_mode *mode, + struct ieee80211_supported_band *sband, struct sk_buff *skb, struct rate_selection *sel) { @@ -322,7 +324,7 @@ static void rate_control_pid_get_rate(void *priv, struct net_device *dev, fc = le16_to_cpu(hdr->frame_control); if ((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA || is_multicast_ether_addr(hdr->addr1) || !sta) { - sel->rate = rate_lowest(local, mode, sta); + sel->rate = rate_lowest(local, sband, sta); if (sta) sta_info_put(sta); return; @@ -331,23 +333,23 @@ static void rate_control_pid_get_rate(void *priv, struct net_device *dev, /* If a forced rate is in effect, select it. */ sdata = IEEE80211_DEV_TO_SUB_IF(dev); if (sdata->bss && sdata->bss->force_unicast_rateidx > -1) - sta->txrate = sdata->bss->force_unicast_rateidx; + sta->txrate_idx = sdata->bss->force_unicast_rateidx; - rateidx = sta->txrate; + rateidx = sta->txrate_idx; - if (rateidx >= mode->num_rates) - rateidx = mode->num_rates - 1; + if (rateidx >= sband->n_bitrates) + rateidx = sband->n_bitrates - 1; - sta->last_txrate = rateidx; + sta->last_txrate_idx = rateidx; sta_info_put(sta); - sel->rate = &mode->rates[rateidx]; + sel->rate = &sband->bitrates[rateidx]; #ifdef CONFIG_MAC80211_DEBUGFS rate_control_pid_event_tx_rate( &((struct rc_pid_sta_info *) sta->rate_ctrl_priv)->events, - rateidx, mode->rates[rateidx].rate); + rateidx, sband->bitrates[rateidx].bitrate); #endif } @@ -359,28 +361,32 @@ static void rate_control_pid_rate_init(void *priv, void *priv_sta, * as we need to have IEEE 802.1X auth succeed immediately after assoc.. * Until that method is implemented, we will use the lowest supported * rate as a workaround. */ - sta->txrate = rate_lowest_index(local, local->oper_hw_mode, sta); + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + sta->txrate_idx = rate_lowest_index(local, sband, sta); } static void *rate_control_pid_alloc(struct ieee80211_local *local) { struct rc_pid_info *pinfo; struct rc_pid_rateinfo *rinfo; - struct ieee80211_hw_mode *mode; + struct ieee80211_supported_band *sband; int i, j, tmp; bool s; #ifdef CONFIG_MAC80211_DEBUGFS struct rc_pid_debugfs_entries *de; #endif + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + pinfo = kmalloc(sizeof(*pinfo), GFP_ATOMIC); if (!pinfo) return NULL; - /* We can safely assume that oper_hw_mode won't change unless we get + /* We can safely assume that sband won't change unless we get * reinitialized. */ - mode = local->oper_hw_mode; - rinfo = kmalloc(sizeof(*rinfo) * mode->num_rates, GFP_ATOMIC); + rinfo = kmalloc(sizeof(*rinfo) * sband->n_bitrates, GFP_ATOMIC); if (!rinfo) { kfree(pinfo); return NULL; @@ -389,7 +395,7 @@ static void *rate_control_pid_alloc(struct ieee80211_local *local) /* Sort the rates. This is optimized for the most common case (i.e. * almost-sorted CCK+OFDM rates). Kind of bubble-sort with reversed * mapping too. */ - for (i = 0; i < mode->num_rates; i++) { + for (i = 0; i < sband->n_bitrates; i++) { rinfo[i].index = i; rinfo[i].rev_index = i; if (pinfo->fast_start) @@ -397,11 +403,11 @@ static void *rate_control_pid_alloc(struct ieee80211_local *local) else rinfo[i].diff = i * pinfo->norm_offset; } - for (i = 1; i < mode->num_rates; i++) { + for (i = 1; i < sband->n_bitrates; i++) { s = 0; - for (j = 0; j < mode->num_rates - i; j++) - if (unlikely(mode->rates[rinfo[j].index].rate > - mode->rates[rinfo[j + 1].index].rate)) { + for (j = 0; j < sband->n_bitrates - i; j++) + if (unlikely(sband->bitrates[rinfo[j].index].bitrate > + sband->bitrates[rinfo[j + 1].index].bitrate)) { tmp = rinfo[j].index; rinfo[j].index = rinfo[j + 1].index; rinfo[j + 1].index = tmp; diff --git a/net/mac80211/rc80211_simple.c b/net/mac80211/rc80211_simple.c index 9a78b116acff..c4678905a142 100644 --- a/net/mac80211/rc80211_simple.c +++ b/net/mac80211/rc80211_simple.c @@ -35,8 +35,8 @@ static void rate_control_rate_inc(struct ieee80211_local *local, struct sta_info *sta) { struct ieee80211_sub_if_data *sdata; - struct ieee80211_hw_mode *mode; - int i = sta->txrate; + struct ieee80211_supported_band *sband; + int i = sta->txrate_idx; int maxrate; sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); @@ -45,18 +45,17 @@ static void rate_control_rate_inc(struct ieee80211_local *local, return; } - mode = local->oper_hw_mode; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; maxrate = sdata->bss ? sdata->bss->max_ratectrl_rateidx : -1; - if (i > mode->num_rates) - i = mode->num_rates - 2; + if (i > sband->n_bitrates) + i = sband->n_bitrates - 2; - while (i + 1 < mode->num_rates) { + while (i + 1 < sband->n_bitrates) { i++; - if (sta->supp_rates & BIT(i) && - mode->rates[i].flags & IEEE80211_RATE_SUPPORTED && + if (rate_supported(sta, sband->band, i) && (maxrate < 0 || i <= maxrate)) { - sta->txrate = i; + sta->txrate_idx = i; break; } } @@ -67,8 +66,8 @@ static void rate_control_rate_dec(struct ieee80211_local *local, struct sta_info *sta) { struct ieee80211_sub_if_data *sdata; - struct ieee80211_hw_mode *mode; - int i = sta->txrate; + struct ieee80211_supported_band *sband; + int i = sta->txrate_idx; sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); if (sdata->bss && sdata->bss->force_unicast_rateidx > -1) { @@ -76,15 +75,14 @@ static void rate_control_rate_dec(struct ieee80211_local *local, return; } - mode = local->oper_hw_mode; - if (i > mode->num_rates) - i = mode->num_rates; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + if (i > sband->n_bitrates) + i = sband->n_bitrates; while (i > 0) { i--; - if (sta->supp_rates & BIT(i) && - mode->rates[i].flags & IEEE80211_RATE_SUPPORTED) { - sta->txrate = i; + if (rate_supported(sta, sband->band, i)) { + sta->txrate_idx = i; break; } } @@ -168,7 +166,7 @@ static void rate_control_simple_tx_status(void *priv, struct net_device *dev, } else if (per_failed < RATE_CONTROL_NUM_UP) { rate_control_rate_inc(local, sta); } - srctrl->tx_avg_rate_sum += status->control.rate->rate; + srctrl->tx_avg_rate_sum += status->control.tx_rate->bitrate; srctrl->tx_avg_rate_num++; srctrl->tx_num_failures = 0; srctrl->tx_num_xmit = 0; @@ -201,7 +199,7 @@ static void rate_control_simple_tx_status(void *priv, struct net_device *dev, static void rate_control_simple_get_rate(void *priv, struct net_device *dev, - struct ieee80211_hw_mode *mode, + struct ieee80211_supported_band *sband, struct sk_buff *skb, struct rate_selection *sel) { @@ -219,7 +217,7 @@ rate_control_simple_get_rate(void *priv, struct net_device *dev, fc = le16_to_cpu(hdr->frame_control); if ((fc & IEEE80211_FCTL_FTYPE) != IEEE80211_FTYPE_DATA || is_multicast_ether_addr(hdr->addr1) || !sta) { - sel->rate = rate_lowest(local, mode, sta); + sel->rate = rate_lowest(local, sband, sta); if (sta) sta_info_put(sta); return; @@ -228,18 +226,18 @@ rate_control_simple_get_rate(void *priv, struct net_device *dev, /* If a forced rate is in effect, select it. */ sdata = IEEE80211_DEV_TO_SUB_IF(dev); if (sdata->bss && sdata->bss->force_unicast_rateidx > -1) - sta->txrate = sdata->bss->force_unicast_rateidx; + sta->txrate_idx = sdata->bss->force_unicast_rateidx; - rateidx = sta->txrate; + rateidx = sta->txrate_idx; - if (rateidx >= mode->num_rates) - rateidx = mode->num_rates - 1; + if (rateidx >= sband->n_bitrates) + rateidx = sband->n_bitrates - 1; - sta->last_txrate = rateidx; + sta->last_txrate_idx = rateidx; sta_info_put(sta); - sel->rate = &mode->rates[rateidx]; + sel->rate = &sband->bitrates[rateidx]; } @@ -247,21 +245,15 @@ static void rate_control_simple_rate_init(void *priv, void *priv_sta, struct ieee80211_local *local, struct sta_info *sta) { - struct ieee80211_hw_mode *mode; - int i; - sta->txrate = 0; - mode = local->oper_hw_mode; + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + /* TODO: This routine should consider using RSSI from previous packets * as we need to have IEEE 802.1X auth succeed immediately after assoc.. * Until that method is implemented, we will use the lowest supported rate * as a workaround, */ - for (i = 0; i < mode->num_rates; i++) { - if ((sta->supp_rates & BIT(i)) && - (mode->rates[i].flags & IEEE80211_RATE_SUPPORTED)) { - sta->txrate = i; - break; - } - } + sta->txrate_idx = rate_lowest_index(local, sband, sta); } diff --git a/net/mac80211/regdomain.c b/net/mac80211/regdomain.c deleted file mode 100644 index f42678fa62d1..000000000000 --- a/net/mac80211/regdomain.c +++ /dev/null @@ -1,152 +0,0 @@ -/* - * Copyright 2002-2005, Instant802 Networks, Inc. - * Copyright 2005-2006, Devicescape Software, Inc. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -/* - * This regulatory domain control implementation is known to be incomplete - * and confusing. mac80211 regulatory domain control will be significantly - * reworked in the not-too-distant future. - * - * For now, drivers wishing to control which channels are and aren't available - * are advised as follows: - * - set the IEEE80211_HW_DEFAULT_REG_DOMAIN_CONFIGURED flag - * - continue to include *ALL* possible channels in the modes registered - * through ieee80211_register_hwmode() - * - for each allowable ieee80211_channel structure registered in the above - * call, set the flag member to some meaningful value such as - * IEEE80211_CHAN_W_SCAN | IEEE80211_CHAN_W_ACTIVE_SCAN | - * IEEE80211_CHAN_W_IBSS. - * - leave flag as 0 for non-allowable channels - * - * The usual implementation is for a driver to read a device EEPROM to - * determine which regulatory domain it should be operating under, then - * looking up the allowable channels in a driver-local table, then performing - * the above. - */ - -#include -#include -#include -#include "ieee80211_i.h" - -static int ieee80211_regdom = 0x10; /* FCC */ -module_param(ieee80211_regdom, int, 0444); -MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain; 64=MKK"); - -/* - * If firmware is upgraded by the vendor, additional channels can be used based - * on the new Japanese regulatory rules. This is indicated by setting - * ieee80211_japan_5ghz module parameter to one when loading the 80211 kernel - * module. - */ -static int ieee80211_japan_5ghz /* = 0 */; -module_param(ieee80211_japan_5ghz, int, 0444); -MODULE_PARM_DESC(ieee80211_japan_5ghz, "Vendor-updated firmware for 5 GHz"); - - -struct ieee80211_channel_range { - short start_freq; - short end_freq; - unsigned char power_level; - unsigned char antenna_max; -}; - -static const struct ieee80211_channel_range ieee80211_fcc_channels[] = { - { 2412, 2462, 27, 6 } /* IEEE 802.11b/g, channels 1..11 */, - { 5180, 5240, 17, 6 } /* IEEE 802.11a, channels 36..48 */, - { 5260, 5320, 23, 6 } /* IEEE 802.11a, channels 52..64 */, - { 5745, 5825, 30, 6 } /* IEEE 802.11a, channels 149..165, outdoor */, - { 0 } -}; - -static const struct ieee80211_channel_range ieee80211_mkk_channels[] = { - { 2412, 2472, 20, 6 } /* IEEE 802.11b/g, channels 1..13 */, - { 5170, 5240, 20, 6 } /* IEEE 802.11a, channels 34..48 */, - { 5260, 5320, 20, 6 } /* IEEE 802.11a, channels 52..64 */, - { 0 } -}; - - -static const struct ieee80211_channel_range *channel_range = - ieee80211_fcc_channels; - - -static void ieee80211_unmask_channel(int mode, struct ieee80211_channel *chan) -{ - int i; - - chan->flag = 0; - - for (i = 0; channel_range[i].start_freq; i++) { - const struct ieee80211_channel_range *r = &channel_range[i]; - if (r->start_freq <= chan->freq && r->end_freq >= chan->freq) { - if (ieee80211_regdom == 64 && !ieee80211_japan_5ghz && - chan->freq >= 5260 && chan->freq <= 5320) { - /* - * Skip new channels in Japan since the - * firmware was not marked having been upgraded - * by the vendor. - */ - continue; - } - - if (ieee80211_regdom == 0x10 && - (chan->freq == 5190 || chan->freq == 5210 || - chan->freq == 5230)) { - /* Skip MKK channels when in FCC domain. */ - continue; - } - - chan->flag |= IEEE80211_CHAN_W_SCAN | - IEEE80211_CHAN_W_ACTIVE_SCAN | - IEEE80211_CHAN_W_IBSS; - chan->power_level = r->power_level; - chan->antenna_max = r->antenna_max; - - if (ieee80211_regdom == 64 && - (chan->freq == 5170 || chan->freq == 5190 || - chan->freq == 5210 || chan->freq == 5230)) { - /* - * New regulatory rules in Japan have backwards - * compatibility with old channels in 5.15-5.25 - * GHz band, but the station is not allowed to - * use active scan on these old channels. - */ - chan->flag &= ~IEEE80211_CHAN_W_ACTIVE_SCAN; - } - - if (ieee80211_regdom == 64 && - (chan->freq == 5260 || chan->freq == 5280 || - chan->freq == 5300 || chan->freq == 5320)) { - /* - * IBSS is not allowed on 5.25-5.35 GHz band - * due to radar detection requirements. - */ - chan->flag &= ~IEEE80211_CHAN_W_IBSS; - } - - break; - } - } -} - - -void ieee80211_set_default_regdomain(struct ieee80211_hw_mode *mode) -{ - int c; - for (c = 0; c < mode->num_channels; c++) - ieee80211_unmask_channel(mode->mode, &mode->channels[c]); -} - - -void ieee80211_regdomain_init(void) -{ - if (ieee80211_regdom == 0x40) - channel_range = ieee80211_mkk_channels; -} - diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 3aae8e9e4e0a..c9ff98a93211 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -82,10 +82,10 @@ static inline int should_drop_frame(struct ieee80211_rx_status *status, */ static struct sk_buff * ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb, - struct ieee80211_rx_status *status) + struct ieee80211_rx_status *status, + struct ieee80211_rate *rate) { struct ieee80211_sub_if_data *sdata; - struct ieee80211_rate *rate; int needed_headroom = 0; struct ieee80211_radiotap_header *rthdr; __le64 *rttsft = NULL; @@ -194,14 +194,11 @@ ieee80211_rx_monitor(struct ieee80211_local *local, struct sk_buff *origskb, rtfixed->rx_flags |= cpu_to_le16(IEEE80211_RADIOTAP_F_RX_BADFCS); - rate = ieee80211_get_rate(local, status->phymode, - status->rate); - if (rate) - rtfixed->rate = rate->rate / 5; + rtfixed->rate = rate->bitrate / 5; rtfixed->chan_freq = cpu_to_le16(status->freq); - if (status->phymode == MODE_IEEE80211A) + if (status->band == IEEE80211_BAND_5GHZ) rtfixed->chan_flags = cpu_to_le16(IEEE80211_CHAN_OFDM | IEEE80211_CHAN_5GHZ); @@ -320,34 +317,21 @@ static void ieee80211_verify_ip_alignment(struct ieee80211_txrx_data *rx) static u32 ieee80211_rx_load_stats(struct ieee80211_local *local, - struct sk_buff *skb, - struct ieee80211_rx_status *status) + struct sk_buff *skb, + struct ieee80211_rx_status *status, + struct ieee80211_rate *rate) { struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; u32 load = 0, hdrtime; - struct ieee80211_rate *rate; - struct ieee80211_hw_mode *mode = local->hw.conf.mode; - int i; /* Estimate total channel use caused by this frame */ - if (unlikely(mode->num_rates < 0)) - return TXRX_CONTINUE; - - rate = &mode->rates[0]; - for (i = 0; i < mode->num_rates; i++) { - if (mode->rates[i].val == status->rate) { - rate = &mode->rates[i]; - break; - } - } - /* 1 bit at 1 Mbit/s takes 1 usec; in channel_use values, * 1 usec = 1/8 * (1080 / 10) = 13.5 */ - if (mode->mode == MODE_IEEE80211A || - (mode->mode == MODE_IEEE80211G && - rate->flags & IEEE80211_RATE_ERP)) + if (status->band == IEEE80211_BAND_5GHZ || + (status->band == IEEE80211_BAND_5GHZ && + rate->flags & IEEE80211_RATE_ERP_G)) hdrtime = CHAN_UTIL_HDR_SHORT; else hdrtime = CHAN_UTIL_HDR_LONG; @@ -356,7 +340,8 @@ static u32 ieee80211_rx_load_stats(struct ieee80211_local *local, if (!is_multicast_ether_addr(hdr->addr1)) load += hdrtime; - load += skb->len * rate->rate_inv; + /* TODO: optimise again */ + load += skb->len * CHAN_UTIL_RATE_LCM / rate->bitrate; /* Divide channel_use by 8 to avoid wrapping around the counter */ load >>= CHAN_UTIL_SHIFT; @@ -1685,7 +1670,8 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, struct sk_buff *skb, struct ieee80211_rx_status *status, - u32 load) + u32 load, + struct ieee80211_rate *rate) { struct ieee80211_local *local = hw_to_local(hw); struct ieee80211_sub_if_data *sdata; @@ -1705,6 +1691,7 @@ static void __ieee80211_rx_handle_packet(struct ieee80211_hw *hw, rx.u.rx.status = status; rx.u.rx.load = load; + rx.u.rx.rate = rate; rx.fc = le16_to_cpu(hdr->frame_control); type = rx.fc & IEEE80211_FCTL_FTYPE; @@ -1837,6 +1824,8 @@ u8 ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw, u16 head_seq_num, buf_size; int index; u32 pkt_load; + struct ieee80211_supported_band *sband; + struct ieee80211_rate *rate; buf_size = tid_agg_rx->buf_size; head_seq_num = tid_agg_rx->head_seq_num; @@ -1867,12 +1856,14 @@ u8 ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw, memcpy(&status, tid_agg_rx->reorder_buf[index]->cb, sizeof(status)); + sband = local->hw.wiphy->bands[status.band]; + rate = &sband->bitrates[status.rate_idx]; pkt_load = ieee80211_rx_load_stats(local, tid_agg_rx->reorder_buf[index], - &status); + &status, rate); __ieee80211_rx_handle_packet(hw, tid_agg_rx->reorder_buf[index], - &status, pkt_load); + &status, pkt_load, rate); tid_agg_rx->stored_mpdu_num--; tid_agg_rx->reorder_buf[index] = NULL; } @@ -1912,11 +1903,13 @@ u8 ieee80211_sta_manage_reorder_buf(struct ieee80211_hw *hw, /* release the reordered frame back to stack */ memcpy(&status, tid_agg_rx->reorder_buf[index]->cb, sizeof(status)); + sband = local->hw.wiphy->bands[status.band]; + rate = &sband->bitrates[status.rate_idx]; pkt_load = ieee80211_rx_load_stats(local, tid_agg_rx->reorder_buf[index], - &status); + &status, rate); __ieee80211_rx_handle_packet(hw, tid_agg_rx->reorder_buf[index], - &status, pkt_load); + &status, pkt_load, rate); tid_agg_rx->stored_mpdu_num--; tid_agg_rx->reorder_buf[index] = NULL; tid_agg_rx->head_seq_num = seq_inc(tid_agg_rx->head_seq_num); @@ -1997,6 +1990,25 @@ void __ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb, { struct ieee80211_local *local = hw_to_local(hw); u32 pkt_load; + struct ieee80211_rate *rate = NULL; + struct ieee80211_supported_band *sband; + + if (status->band < 0 || + status->band > IEEE80211_NUM_BANDS) { + WARN_ON(1); + return; + } + + sband = local->hw.wiphy->bands[status->band]; + + if (!sband || + status->rate_idx < 0 || + status->rate_idx >= sband->n_bitrates) { + WARN_ON(1); + return; + } + + rate = &sband->bitrates[status->rate_idx]; /* * key references and virtual interfaces are protected using RCU @@ -2011,17 +2023,17 @@ void __ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb, * if it was previously present. * Also, frames with less than 16 bytes are dropped. */ - skb = ieee80211_rx_monitor(local, skb, status); + skb = ieee80211_rx_monitor(local, skb, status, rate); if (!skb) { rcu_read_unlock(); return; } - pkt_load = ieee80211_rx_load_stats(local, skb, status); + pkt_load = ieee80211_rx_load_stats(local, skb, status, rate); local->channel_use_raw += pkt_load; if (!ieee80211_rx_reorder_ampdu(local, skb)) - __ieee80211_rx_handle_packet(hw, skb, status, pkt_load); + __ieee80211_rx_handle_packet(hw, skb, status, pkt_load, rate); rcu_read_unlock(); } diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index ddc1f47194dd..746bbdea6b4c 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -74,30 +74,6 @@ struct sta_info *sta_info_get(struct ieee80211_local *local, u8 *addr) } EXPORT_SYMBOL(sta_info_get); -int sta_info_min_txrate_get(struct ieee80211_local *local) -{ - struct sta_info *sta; - struct ieee80211_hw_mode *mode; - int min_txrate = 9999999; - int i; - - read_lock_bh(&local->sta_lock); - mode = local->oper_hw_mode; - for (i = 0; i < STA_HASH_SIZE; i++) { - sta = local->sta_hash[i]; - while (sta) { - if (sta->txrate < min_txrate) - min_txrate = sta->txrate; - sta = sta->hnext; - } - } - read_unlock_bh(&local->sta_lock); - if (min_txrate == 9999999) - min_txrate = 0; - - return mode->rates[min_txrate].rate; -} - static void sta_info_release(struct kref *kref) { diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 75573dc79d7a..3573743dfa59 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -133,10 +133,11 @@ struct sta_info { unsigned int wep_weak_iv_count; /* number of RX frames with weak IV */ unsigned long last_rx; - u32 supp_rates; /* bitmap of supported rates in local->curr_rates */ - int txrate; /* index in local->curr_rates */ - int last_txrate; /* last rate used to send a frame to this STA */ - int last_nonerp_idx; + /* bitmap of supported rates per band */ + u64 supp_rates[IEEE80211_NUM_BANDS]; + int txrate_idx; + /* last rates used to send a frame to this STA */ + int last_txrate_idx, last_nonerp_txrate_idx; struct net_device *dev; /* which net device is this station associated * to */ @@ -222,7 +223,6 @@ static inline void __sta_info_get(struct sta_info *sta) } struct sta_info * sta_info_get(struct ieee80211_local *local, u8 *addr); -int sta_info_min_txrate_get(struct ieee80211_local *local); void sta_info_put(struct sta_info *sta); struct sta_info * sta_info_add(struct ieee80211_local *local, struct net_device *dev, u8 *addr, gfp_t gfp); diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 38e1b2bd8245..9e5359991985 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -92,9 +92,13 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr, int rate, mrate, erp, dur, i; struct ieee80211_rate *txrate = tx->u.tx.rate; struct ieee80211_local *local = tx->local; - struct ieee80211_hw_mode *mode = tx->u.tx.mode; + struct ieee80211_supported_band *sband; - erp = txrate->flags & IEEE80211_RATE_ERP; + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; + + erp = 0; + if (tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) + erp = txrate->flags & IEEE80211_RATE_ERP_G; /* * data and mgmt (except PS Poll): @@ -150,20 +154,36 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr, * Mandatory rates for IEEE 802.11g PHY: 1, 2, 5.5, 11, 6, 12, 24 Mbps */ rate = -1; - mrate = 10; /* use 1 Mbps if everything fails */ - for (i = 0; i < mode->num_rates; i++) { - struct ieee80211_rate *r = &mode->rates[i]; - if (r->rate > txrate->rate) - break; + /* use lowest available if everything fails */ + mrate = sband->bitrates[0].bitrate; + for (i = 0; i < sband->n_bitrates; i++) { + struct ieee80211_rate *r = &sband->bitrates[i]; - if (IEEE80211_RATE_MODULATION(txrate->flags) != - IEEE80211_RATE_MODULATION(r->flags)) - continue; + if (r->bitrate > txrate->bitrate) + break; - if (r->flags & IEEE80211_RATE_BASIC) - rate = r->rate; - else if (r->flags & IEEE80211_RATE_MANDATORY) - mrate = r->rate; + if (tx->sdata->basic_rates & BIT(i)) + rate = r->bitrate; + + switch (sband->band) { + case IEEE80211_BAND_2GHZ: { + u32 flag; + if (tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) + flag = IEEE80211_RATE_MANDATORY_G; + else + flag = IEEE80211_RATE_MANDATORY_B; + if (r->flags & flag) + mrate = r->bitrate; + break; + } + case IEEE80211_BAND_5GHZ: + if (r->flags & IEEE80211_RATE_MANDATORY_A) + mrate = r->bitrate; + break; + case IEEE80211_NUM_BANDS: + WARN_ON(1); + break; + } } if (rate == -1) { /* No matching basic rate found; use highest suitable mandatory @@ -184,7 +204,7 @@ static u16 ieee80211_duration(struct ieee80211_txrx_data *tx, int group_addr, dur *= 2; /* ACK + SIFS */ /* next fragment */ dur += ieee80211_frame_duration(local, next_frag_len, - txrate->rate, erp, + txrate->bitrate, erp, tx->sdata->bss_conf.use_short_preamble); } @@ -585,26 +605,28 @@ static ieee80211_txrx_result ieee80211_tx_h_rate_ctrl(struct ieee80211_txrx_data *tx) { struct rate_selection rsel; + struct ieee80211_supported_band *sband; + + sband = tx->local->hw.wiphy->bands[tx->local->hw.conf.channel->band]; if (likely(!tx->u.tx.rate)) { - rate_control_get_rate(tx->dev, tx->u.tx.mode, tx->skb, &rsel); + rate_control_get_rate(tx->dev, sband, tx->skb, &rsel); tx->u.tx.rate = rsel.rate; - if (unlikely(rsel.probe != NULL)) { + if (unlikely(rsel.probe)) { tx->u.tx.control->flags |= IEEE80211_TXCTL_RATE_CTRL_PROBE; tx->flags |= IEEE80211_TXRXD_TXPROBE_LAST_FRAG; - tx->u.tx.control->alt_retry_rate = tx->u.tx.rate->val; + tx->u.tx.control->alt_retry_rate = tx->u.tx.rate; tx->u.tx.rate = rsel.probe; } else - tx->u.tx.control->alt_retry_rate = -1; + tx->u.tx.control->alt_retry_rate = NULL; if (!tx->u.tx.rate) return TXRX_DROP; } else - tx->u.tx.control->alt_retry_rate = -1; + tx->u.tx.control->alt_retry_rate = NULL; - if (tx->u.tx.mode->mode == MODE_IEEE80211G && - tx->sdata->bss_conf.use_cts_prot && + if (tx->sdata->bss_conf.use_cts_prot && (tx->flags & IEEE80211_TXRXD_FRAGMENTED) && rsel.nonerp) { tx->u.tx.last_frag_rate = tx->u.tx.rate; if (rsel.probe) @@ -612,13 +634,13 @@ ieee80211_tx_h_rate_ctrl(struct ieee80211_txrx_data *tx) else tx->flags |= IEEE80211_TXRXD_TXPROBE_LAST_FRAG; tx->u.tx.rate = rsel.nonerp; - tx->u.tx.control->rate = rsel.nonerp; + tx->u.tx.control->tx_rate = rsel.nonerp; tx->u.tx.control->flags &= ~IEEE80211_TXCTL_RATE_CTRL_PROBE; } else { tx->u.tx.last_frag_rate = tx->u.tx.rate; - tx->u.tx.control->rate = tx->u.tx.rate; + tx->u.tx.control->tx_rate = tx->u.tx.rate; } - tx->u.tx.control->tx_rate = tx->u.tx.rate->val; + tx->u.tx.control->tx_rate = tx->u.tx.rate; return TXRX_CONTINUE; } @@ -630,7 +652,6 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx) u16 fc = le16_to_cpu(hdr->frame_control); u16 dur; struct ieee80211_tx_control *control = tx->u.tx.control; - struct ieee80211_hw_mode *mode = tx->u.tx.mode; if (!control->retry_limit) { if (!is_multicast_ether_addr(hdr->addr1)) { @@ -657,14 +678,14 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx) * frames. * TODO: The last fragment could still use multiple retry * rates. */ - control->alt_retry_rate = -1; + control->alt_retry_rate = NULL; } /* Use CTS protection for unicast frames sent using extended rates if * there are associated non-ERP stations and RTS/CTS is not configured * for the frame. */ - if (mode->mode == MODE_IEEE80211G && - (tx->u.tx.rate->flags & IEEE80211_RATE_ERP) && + if ((tx->sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) && + (tx->u.tx.rate->flags & IEEE80211_RATE_ERP_G) && (tx->flags & IEEE80211_TXRXD_TXUNICAST) && tx->sdata->bss_conf.use_cts_prot && !(control->flags & IEEE80211_TXCTL_USE_RTS_CTS)) @@ -674,10 +695,10 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx) * short preambles at the selected rate and short preambles are * available on the network at the current point in time. */ if (((fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA) && - (tx->u.tx.rate->flags & IEEE80211_RATE_PREAMBLE2) && + (tx->u.tx.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) && tx->sdata->bss_conf.use_short_preamble && (!tx->sta || (tx->sta->flags & WLAN_STA_SHORT_PREAMBLE))) { - tx->u.tx.control->tx_rate = tx->u.tx.rate->val2; + tx->u.tx.control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE; } /* Setup duration field for the first fragment of the frame. Duration @@ -690,19 +711,33 @@ ieee80211_tx_h_misc(struct ieee80211_txrx_data *tx) if ((control->flags & IEEE80211_TXCTL_USE_RTS_CTS) || (control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT)) { - struct ieee80211_rate *rate; + struct ieee80211_supported_band *sband; + struct ieee80211_rate *rate, *baserate; + int idx; + + sband = tx->local->hw.wiphy->bands[ + tx->local->hw.conf.channel->band]; /* Do not use multiple retry rates when using RTS/CTS */ - control->alt_retry_rate = -1; + control->alt_retry_rate = NULL; /* Use min(data rate, max base rate) as CTS/RTS rate */ rate = tx->u.tx.rate; - while (rate > mode->rates && - !(rate->flags & IEEE80211_RATE_BASIC)) - rate--; + baserate = NULL; + + for (idx = 0; idx < sband->n_bitrates; idx++) { + if (sband->bitrates[idx].bitrate > rate->bitrate) + continue; + if (tx->sdata->basic_rates & BIT(idx) && + (!baserate || + (baserate->bitrate < sband->bitrates[idx].bitrate))) + baserate = &sband->bitrates[idx]; + } - control->rts_cts_rate = rate->val; - control->rts_rate = rate; + if (baserate) + control->rts_cts_rate = baserate; + else + control->rts_cts_rate = &sband->bitrates[0]; } if (tx->sta) { @@ -726,10 +761,10 @@ static ieee80211_txrx_result ieee80211_tx_h_load_stats(struct ieee80211_txrx_data *tx) { struct ieee80211_local *local = tx->local; - struct ieee80211_hw_mode *mode = tx->u.tx.mode; struct sk_buff *skb = tx->skb; struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; u32 load = 0, hdrtime; + struct ieee80211_rate *rate = tx->u.tx.rate; /* TODO: this could be part of tx_status handling, so that the number * of retries would be known; TX rate should in that case be stored @@ -740,9 +775,9 @@ ieee80211_tx_h_load_stats(struct ieee80211_txrx_data *tx) /* 1 bit at 1 Mbit/s takes 1 usec; in channel_use values, * 1 usec = 1/8 * (1080 / 10) = 13.5 */ - if (mode->mode == MODE_IEEE80211A || - (mode->mode == MODE_IEEE80211G && - tx->u.tx.rate->flags & IEEE80211_RATE_ERP)) + if (tx->u.tx.channel->band == IEEE80211_BAND_5GHZ || + (tx->u.tx.channel->band == IEEE80211_BAND_2GHZ && + rate->flags & IEEE80211_RATE_ERP_G)) hdrtime = CHAN_UTIL_HDR_SHORT; else hdrtime = CHAN_UTIL_HDR_LONG; @@ -756,14 +791,15 @@ ieee80211_tx_h_load_stats(struct ieee80211_txrx_data *tx) else if (tx->u.tx.control->flags & IEEE80211_TXCTL_USE_CTS_PROTECT) load += hdrtime; - load += skb->len * tx->u.tx.rate->rate_inv; + /* TODO: optimise again */ + load += skb->len * CHAN_UTIL_RATE_LCM / rate->bitrate; if (tx->u.tx.extra_frag) { int i; for (i = 0; i < tx->u.tx.num_extra_frag; i++) { load += 2 * hdrtime; load += tx->u.tx.extra_frag[i]->len * - tx->u.tx.rate->rate; + tx->u.tx.rate->bitrate; } } @@ -816,10 +852,12 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_txrx_data *tx, struct ieee80211_radiotap_iterator iterator; struct ieee80211_radiotap_header *rthdr = (struct ieee80211_radiotap_header *) skb->data; - struct ieee80211_hw_mode *mode = tx->local->hw.conf.mode; + struct ieee80211_supported_band *sband; int ret = ieee80211_radiotap_iterator_init(&iterator, rthdr, skb->len); struct ieee80211_tx_control *control = tx->u.tx.control; + sband = tx->local->hw.wiphy->bands[tx->local->hw.conf.channel->band]; + control->flags |= IEEE80211_TXCTL_DO_NOT_ENCRYPT; tx->flags |= IEEE80211_TXRXD_TX_INJECTED; tx->flags &= ~IEEE80211_TXRXD_FRAGMENTED; @@ -852,10 +890,12 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_txrx_data *tx, * ieee80211 rate int is in 100kbps units eg, 0x0a=1Mbps */ target_rate = (*iterator.this_arg) * 5; - for (i = 0; i < mode->num_rates; i++) { - struct ieee80211_rate *r = &mode->rates[i]; + for (i = 0; i < sband->n_bitrates; i++) { + struct ieee80211_rate *r; - if (r->rate == target_rate) { + r = &sband->bitrates[i]; + + if (r->bitrate == target_rate) { tx->u.tx.rate = r; break; } @@ -870,9 +910,11 @@ __ieee80211_parse_tx_radiotap(struct ieee80211_txrx_data *tx, control->antenna_sel_tx = (*iterator.this_arg) + 1; break; +#if 0 case IEEE80211_RADIOTAP_DBM_TX_POWER: control->power_level = *iterator.this_arg; break; +#endif case IEEE80211_RADIOTAP_FLAGS: if (*iterator.this_arg & IEEE80211_RADIOTAP_F_FCS) { @@ -1054,8 +1096,8 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb, if (__ieee80211_queue_stopped(local, control->queue)) return IEEE80211_TX_FRAG_AGAIN; if (i == tx->u.tx.num_extra_frag) { - control->tx_rate = tx->u.tx.last_frag_hwrate; - control->rate = tx->u.tx.last_frag_rate; + control->tx_rate = tx->u.tx.last_frag_rate; + if (tx->flags & IEEE80211_TXRXD_TXPROBE_LAST_FRAG) control->flags |= IEEE80211_TXCTL_RATE_CTRL_PROBE; @@ -1114,7 +1156,7 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb, rcu_read_lock(); sta = tx.sta; - tx.u.tx.mode = local->hw.conf.mode; + tx.u.tx.channel = local->hw.conf.channel; for (handler = local->tx_handlers; *handler != NULL; handler++) { @@ -1151,7 +1193,6 @@ static int ieee80211_tx(struct net_device *dev, struct sk_buff *skb, } else { next_len = 0; tx.u.tx.rate = tx.u.tx.last_frag_rate; - tx.u.tx.last_frag_hwrate = tx.u.tx.rate->val; } dur = ieee80211_duration(&tx, 0, next_len); hdr->duration_id = cpu_to_le16(dur); @@ -1188,7 +1229,6 @@ retry: store->skb = skb; store->extra_frag = tx.u.tx.extra_frag; store->num_extra_frag = tx.u.tx.num_extra_frag; - store->last_frag_hwrate = tx.u.tx.last_frag_hwrate; store->last_frag_rate = tx.u.tx.last_frag_rate; store->last_frag_rate_ctrl_probe = !!(tx.flags & IEEE80211_TXRXD_TXPROBE_LAST_FRAG); @@ -1609,7 +1649,6 @@ void ieee80211_tx_pending(unsigned long data) tx.u.tx.control = &store->control; tx.u.tx.extra_frag = store->extra_frag; tx.u.tx.num_extra_frag = store->num_extra_frag; - tx.u.tx.last_frag_hwrate = store->last_frag_hwrate; tx.u.tx.last_frag_rate = store->last_frag_rate; tx.flags = 0; if (store->last_frag_rate_ctrl_probe) @@ -1712,6 +1751,9 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, struct ieee80211_if_ap *ap = NULL; struct rate_selection rsel; struct beacon_data *beacon; + struct ieee80211_supported_band *sband; + + sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; rcu_read_lock(); @@ -1750,8 +1792,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, beacon->tail_len); if (control) { - rate_control_get_rate(local->mdev, local->oper_hw_mode, skb, - &rsel); + rate_control_get_rate(local->mdev, sband, skb, &rsel); if (!rsel.rate) { if (net_ratelimit()) { printk(KERN_DEBUG "%s: ieee80211_beacon_get: " @@ -1764,12 +1805,11 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, } control->vif = vif; - control->tx_rate = - (sdata->bss_conf.use_short_preamble && - (rsel.rate->flags & IEEE80211_RATE_PREAMBLE2)) ? - rsel.rate->val2 : rsel.rate->val; + control->tx_rate = rsel.rate; + if (sdata->bss_conf.use_short_preamble && + rsel.rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) + control->flags |= IEEE80211_TXCTL_SHORT_PREAMBLE; control->antenna_sel_tx = local->hw.conf.antenna_sel_tx; - control->power_level = local->hw.conf.power_level; control->flags |= IEEE80211_TXCTL_NO_ACK; control->retry_limit = 1; control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK; @@ -1874,7 +1914,7 @@ ieee80211_get_buffered_bc(struct ieee80211_hw *hw, } sta = tx.sta; tx.flags |= IEEE80211_TXRXD_TXPS_BUFFERED; - tx.u.tx.mode = local->hw.conf.mode; + tx.u.tx.channel = local->hw.conf.channel; for (handler = local->tx_handlers; *handler != NULL; handler++) { res = (*handler)(&tx); diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 5e631ce98d7e..f64804fed0a9 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -41,92 +41,6 @@ const unsigned char bridge_tunnel_header[] = { 0xaa, 0xaa, 0x03, 0x00, 0x00, 0xf8 }; -static int rate_list_match(const int *rate_list, int rate) -{ - int i; - - if (!rate_list) - return 0; - - for (i = 0; rate_list[i] >= 0; i++) - if (rate_list[i] == rate) - return 1; - - return 0; -} - -void ieee80211_prepare_rates(struct ieee80211_local *local, - struct ieee80211_hw_mode *mode) -{ - int i; - - for (i = 0; i < mode->num_rates; i++) { - struct ieee80211_rate *rate = &mode->rates[i]; - - rate->flags &= ~(IEEE80211_RATE_SUPPORTED | - IEEE80211_RATE_BASIC); - - if (local->supp_rates[mode->mode]) { - if (!rate_list_match(local->supp_rates[mode->mode], - rate->rate)) - continue; - } - - rate->flags |= IEEE80211_RATE_SUPPORTED; - - /* Use configured basic rate set if it is available. If not, - * use defaults that are sane for most cases. */ - if (local->basic_rates[mode->mode]) { - if (rate_list_match(local->basic_rates[mode->mode], - rate->rate)) - rate->flags |= IEEE80211_RATE_BASIC; - } else switch (mode->mode) { - case MODE_IEEE80211A: - if (rate->rate == 60 || rate->rate == 120 || - rate->rate == 240) - rate->flags |= IEEE80211_RATE_BASIC; - break; - case MODE_IEEE80211B: - if (rate->rate == 10 || rate->rate == 20) - rate->flags |= IEEE80211_RATE_BASIC; - break; - case MODE_IEEE80211G: - if (rate->rate == 10 || rate->rate == 20 || - rate->rate == 55 || rate->rate == 110) - rate->flags |= IEEE80211_RATE_BASIC; - break; - case NUM_IEEE80211_MODES: - /* not useful */ - break; - } - - /* Set ERP and MANDATORY flags based on phymode */ - switch (mode->mode) { - case MODE_IEEE80211A: - if (rate->rate == 60 || rate->rate == 120 || - rate->rate == 240) - rate->flags |= IEEE80211_RATE_MANDATORY; - break; - case MODE_IEEE80211B: - if (rate->rate == 10) - rate->flags |= IEEE80211_RATE_MANDATORY; - break; - case MODE_IEEE80211G: - if (rate->rate == 10 || rate->rate == 20 || - rate->rate == 55 || rate->rate == 110 || - rate->rate == 60 || rate->rate == 120 || - rate->rate == 240) - rate->flags |= IEEE80211_RATE_MANDATORY; - break; - case NUM_IEEE80211_MODES: - /* not useful */ - break; - } - if (ieee80211_is_erp_rate(mode->mode, rate->rate)) - rate->flags |= IEEE80211_RATE_ERP; - } -} - u8 *ieee80211_get_bssid(struct ieee80211_hdr *hdr, size_t len, enum ieee80211_if_types type) { @@ -262,7 +176,7 @@ int ieee80211_frame_duration(struct ieee80211_local *local, size_t len, * DIV_ROUND_UP() operations. */ - if (local->hw.conf.phymode == MODE_IEEE80211A || erp) { + if (local->hw.conf.channel->band == IEEE80211_BAND_5GHZ || erp) { /* * OFDM: * @@ -304,15 +218,19 @@ int ieee80211_frame_duration(struct ieee80211_local *local, size_t len, /* Exported duration function for driver use */ __le16 ieee80211_generic_frame_duration(struct ieee80211_hw *hw, struct ieee80211_vif *vif, - size_t frame_len, int rate) + size_t frame_len, + struct ieee80211_rate *rate) { struct ieee80211_local *local = hw_to_local(hw); struct ieee80211_sub_if_data *sdata = vif_to_sdata(vif); u16 dur; int erp; - erp = ieee80211_is_erp_rate(hw->conf.phymode, rate); - dur = ieee80211_frame_duration(local, frame_len, rate, erp, + erp = 0; + if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) + erp = rate->flags & IEEE80211_RATE_ERP_G; + + dur = ieee80211_frame_duration(local, frame_len, rate->bitrate, erp, sdata->bss_conf.use_short_preamble); return cpu_to_le16(dur); @@ -332,17 +250,20 @@ __le16 ieee80211_rts_duration(struct ieee80211_hw *hw, short_preamble = sdata->bss_conf.use_short_preamble; - rate = frame_txctl->rts_rate; - erp = !!(rate->flags & IEEE80211_RATE_ERP); + rate = frame_txctl->rts_cts_rate; + + erp = 0; + if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) + erp = rate->flags & IEEE80211_RATE_ERP_G; /* CTS duration */ - dur = ieee80211_frame_duration(local, 10, rate->rate, + dur = ieee80211_frame_duration(local, 10, rate->bitrate, erp, short_preamble); /* Data frame duration */ - dur += ieee80211_frame_duration(local, frame_len, rate->rate, + dur += ieee80211_frame_duration(local, frame_len, rate->bitrate, erp, short_preamble); /* ACK duration */ - dur += ieee80211_frame_duration(local, 10, rate->rate, + dur += ieee80211_frame_duration(local, 10, rate->bitrate, erp, short_preamble); return cpu_to_le16(dur); @@ -363,15 +284,17 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, short_preamble = sdata->bss_conf.use_short_preamble; - rate = frame_txctl->rts_rate; - erp = !!(rate->flags & IEEE80211_RATE_ERP); + rate = frame_txctl->rts_cts_rate; + erp = 0; + if (sdata->flags & IEEE80211_SDATA_OPERATING_GMODE) + erp = rate->flags & IEEE80211_RATE_ERP_G; /* Data frame duration */ - dur = ieee80211_frame_duration(local, frame_len, rate->rate, + dur = ieee80211_frame_duration(local, frame_len, rate->bitrate, erp, short_preamble); if (!(frame_txctl->flags & IEEE80211_TXCTL_NO_ACK)) { /* ACK duration */ - dur += ieee80211_frame_duration(local, 10, rate->rate, + dur += ieee80211_frame_duration(local, 10, rate->bitrate, erp, short_preamble); } @@ -379,27 +302,6 @@ __le16 ieee80211_ctstoself_duration(struct ieee80211_hw *hw, } EXPORT_SYMBOL(ieee80211_ctstoself_duration); -struct ieee80211_rate * -ieee80211_get_rate(struct ieee80211_local *local, int phymode, int hw_rate) -{ - struct ieee80211_hw_mode *mode; - int r; - - list_for_each_entry(mode, &local->modes_list, list) { - if (mode->mode != phymode) - continue; - for (r = 0; r < mode->num_rates; r++) { - struct ieee80211_rate *rate = &mode->rates[r]; - if (rate->val == hw_rate || - (rate->flags & IEEE80211_RATE_PREAMBLE2 && - rate->val2 == hw_rate)) - return rate; - } - } - - return NULL; -} - void ieee80211_wake_queue(struct ieee80211_hw *hw, int queue) { struct ieee80211_local *local = hw_to_local(hw); diff --git a/net/wireless/Makefile b/net/wireless/Makefile index 65710a42e5a7..b9f943c45f3b 100644 --- a/net/wireless/Makefile +++ b/net/wireless/Makefile @@ -1,5 +1,5 @@ obj-$(CONFIG_WIRELESS_EXT) += wext.o obj-$(CONFIG_CFG80211) += cfg80211.o -cfg80211-y += core.o sysfs.o radiotap.o +cfg80211-y += core.o sysfs.o radiotap.o util.o reg.o cfg80211-$(CONFIG_NL80211) += nl80211.o diff --git a/net/wireless/core.c b/net/wireless/core.c index cfc5fc5f9e75..80afacdae46c 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -232,6 +232,47 @@ int wiphy_register(struct wiphy *wiphy) { struct cfg80211_registered_device *drv = wiphy_to_dev(wiphy); int res; + enum ieee80211_band band; + struct ieee80211_supported_band *sband; + bool have_band = false; + int i; + + /* sanity check supported bands/channels */ + for (band = 0; band < IEEE80211_NUM_BANDS; band++) { + sband = wiphy->bands[band]; + if (!sband) + continue; + + sband->band = band; + + if (!sband->n_channels || !sband->n_bitrates) { + WARN_ON(1); + return -EINVAL; + } + + for (i = 0; i < sband->n_channels; i++) { + sband->channels[i].orig_flags = + sband->channels[i].flags; + sband->channels[i].orig_mag = + sband->channels[i].max_antenna_gain; + sband->channels[i].orig_mpwr = + sband->channels[i].max_power; + sband->channels[i].band = band; + } + + have_band = true; + } + + if (!have_band) { + WARN_ON(1); + return -EINVAL; + } + + /* check and set up bitrates */ + ieee80211_set_bitrate_flags(wiphy); + + /* set up regulatory info */ + wiphy_update_regulatory(wiphy); mutex_lock(&cfg80211_drv_mutex); diff --git a/net/wireless/core.h b/net/wireless/core.h index eb0f846b40df..7a02c356d63d 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -78,4 +78,7 @@ extern void cfg80211_dev_free(struct cfg80211_registered_device *drv); extern int cfg80211_dev_rename(struct cfg80211_registered_device *drv, char *newname); +void ieee80211_set_bitrate_flags(struct wiphy *wiphy); +void wiphy_update_regulatory(struct wiphy *wiphy); + #endif /* __NET_WIRELESS_CORE_H */ diff --git a/net/wireless/reg.c b/net/wireless/reg.c new file mode 100644 index 000000000000..2b63c96dcf19 --- /dev/null +++ b/net/wireless/reg.c @@ -0,0 +1,153 @@ +/* + * Copyright 2002-2005, Instant802 Networks, Inc. + * Copyright 2005-2006, Devicescape Software, Inc. + * Copyright 2007 Johannes Berg + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +/* + * This regulatory domain control implementation is highly incomplete, it + * only exists for the purpose of not regressing mac80211. + * + * For now, drivers can restrict the set of allowed channels by either + * not registering those channels or setting the IEEE80211_CHAN_DISABLED + * flag; that flag will only be *set* by this code, never *cleared. + * + * The usual implementation is for a driver to read a device EEPROM to + * determine which regulatory domain it should be operating under, then + * looking up the allowable channels in a driver-local table and finally + * registering those channels in the wiphy structure. + * + * Alternatively, drivers that trust the regulatory domain control here + * will register a complete set of capabilities and the control code + * will restrict the set by setting the IEEE80211_CHAN_* flags. + */ +#include +#include +#include "core.h" + +static char *ieee80211_regdom = "US"; +module_param(ieee80211_regdom, charp, 0444); +MODULE_PARM_DESC(ieee80211_regdom, "IEEE 802.11 regulatory domain code"); + +struct ieee80211_channel_range { + short start_freq; + short end_freq; + int max_power; + int max_antenna_gain; + u32 flags; +}; + +struct ieee80211_regdomain { + const char *code; + const struct ieee80211_channel_range *ranges; + int n_ranges; +}; + +#define RANGE_PWR(_start, _end, _pwr, _ag, _flags) \ + { _start, _end, _pwr, _ag, _flags } + + +/* + * Ideally, in the future, these definitions will be loaded from a + * userspace table via some daemon. + */ +static const struct ieee80211_channel_range ieee80211_US_channels[] = { + /* IEEE 802.11b/g, channels 1..11 */ + RANGE_PWR(2412, 2462, 27, 6, 0), + /* IEEE 802.11a, channels 52..64 */ + RANGE_PWR(5260, 5320, 23, 6, 0), + /* IEEE 802.11a, channels 149..165, outdoor */ + RANGE_PWR(5745, 5825, 30, 6, 0), +}; + +static const struct ieee80211_channel_range ieee80211_JP_channels[] = { + /* IEEE 802.11b/g, channels 1..14 */ + RANGE_PWR(2412, 2484, 20, 6, 0), + /* IEEE 802.11a, channels 34..48 */ + RANGE_PWR(5170, 5240, 20, 6, IEEE80211_CHAN_PASSIVE_SCAN), + /* IEEE 802.11a, channels 52..64 */ + RANGE_PWR(5260, 5320, 20, 6, IEEE80211_CHAN_NO_IBSS | + IEEE80211_CHAN_RADAR), +}; + +#define REGDOM(_code) \ + { \ + .code = __stringify(_code), \ + .ranges = ieee80211_ ##_code## _channels, \ + .n_ranges = ARRAY_SIZE(ieee80211_ ##_code## _channels), \ + } + +static const struct ieee80211_regdomain ieee80211_regdoms[] = { + REGDOM(US), + REGDOM(JP), +}; + + +static const struct ieee80211_regdomain *get_regdom(void) +{ + static const struct ieee80211_channel_range + ieee80211_world_channels[] = { + /* IEEE 802.11b/g, channels 1..11 */ + RANGE_PWR(2412, 2462, 27, 6, 0), + }; + static const struct ieee80211_regdomain regdom_world = REGDOM(world); + int i; + + for (i = 0; i < ARRAY_SIZE(ieee80211_regdoms); i++) + if (strcmp(ieee80211_regdom, ieee80211_regdoms[i].code) == 0) + return &ieee80211_regdoms[i]; + + return ®dom_world; +} + + +static void handle_channel(struct ieee80211_channel *chan, + const struct ieee80211_regdomain *rd) +{ + int i; + u32 flags = chan->orig_flags; + const struct ieee80211_channel_range *rg = NULL; + + for (i = 0; i < rd->n_ranges; i++) { + if (rd->ranges[i].start_freq <= chan->center_freq && + chan->center_freq <= rd->ranges[i].end_freq) { + rg = &rd->ranges[i]; + break; + } + } + + if (!rg) { + /* not found */ + flags |= IEEE80211_CHAN_DISABLED; + chan->flags = flags; + return; + } + + chan->flags = flags; + chan->max_antenna_gain = min(chan->orig_mag, + rg->max_antenna_gain); + chan->max_power = min(chan->orig_mpwr, rg->max_power); +} + +static void handle_band(struct ieee80211_supported_band *sband, + const struct ieee80211_regdomain *rd) +{ + int i; + + for (i = 0; i < sband->n_channels; i++) + handle_channel(&sband->channels[i], rd); +} + +void wiphy_update_regulatory(struct wiphy *wiphy) +{ + enum ieee80211_band band; + const struct ieee80211_regdomain *rd = get_regdom(); + + for (band = 0; band < IEEE80211_NUM_BANDS; band++) + if (wiphy->bands[band]) + handle_band(wiphy->bands[band], rd); +} diff --git a/net/wireless/util.c b/net/wireless/util.c new file mode 100644 index 000000000000..0dcccbf3eb57 --- /dev/null +++ b/net/wireless/util.c @@ -0,0 +1,98 @@ +/* + * Wireless utility functions + * + * Copyright 2007 Johannes Berg + */ +#include +#include +#include "core.h" + +int ieee80211_channel_to_frequency(int chan) +{ + if (chan < 14) + return 2407 + chan * 5; + + if (chan == 14) + return 2484; + + /* FIXME: 802.11j 17.3.8.3.2 */ + return (chan + 1000) * 5; +} +EXPORT_SYMBOL(ieee80211_channel_to_frequency); + +int ieee80211_frequency_to_channel(int freq) +{ + if (freq == 2484) + return 14; + + if (freq < 2484) + return (freq - 2407) / 5; + + /* FIXME: 802.11j 17.3.8.3.2 */ + return freq/5 - 1000; +} +EXPORT_SYMBOL(ieee80211_frequency_to_channel); + +static void set_mandatory_flags_band(struct ieee80211_supported_band *sband, + enum ieee80211_band band) +{ + int i, want; + + switch (band) { + case IEEE80211_BAND_5GHZ: + want = 3; + for (i = 0; i < sband->n_bitrates; i++) { + if (sband->bitrates[i].bitrate == 60 || + sband->bitrates[i].bitrate == 120 || + sband->bitrates[i].bitrate == 240) { + sband->bitrates[i].flags |= + IEEE80211_RATE_MANDATORY_A; + want--; + } + } + WARN_ON(want); + break; + case IEEE80211_BAND_2GHZ: + want = 7; + for (i = 0; i < sband->n_bitrates; i++) { + if (sband->bitrates[i].bitrate == 10) { + sband->bitrates[i].flags |= + IEEE80211_RATE_MANDATORY_B | + IEEE80211_RATE_MANDATORY_G; + want--; + } + + if (sband->bitrates[i].bitrate == 20 || + sband->bitrates[i].bitrate == 55 || + sband->bitrates[i].bitrate == 110 || + sband->bitrates[i].bitrate == 60 || + sband->bitrates[i].bitrate == 120 || + sband->bitrates[i].bitrate == 240) { + sband->bitrates[i].flags |= + IEEE80211_RATE_MANDATORY_G; + want--; + } + + if (sband->bitrates[i].bitrate == 10 || + sband->bitrates[i].bitrate == 20 || + sband->bitrates[i].bitrate == 55 || + sband->bitrates[i].bitrate == 110) + sband->bitrates[i].flags |= + IEEE80211_RATE_ERP_G; + } + WARN_ON(want != 0 && want != 6); + break; + case IEEE80211_NUM_BANDS: + WARN_ON(1); + break; + } +} + +void ieee80211_set_bitrate_flags(struct wiphy *wiphy) +{ + enum ieee80211_band band; + + for (band = 0; band < IEEE80211_NUM_BANDS; band++) + if (wiphy->bands[band]) + set_mandatory_flags_band(wiphy->bands[band], band); +} -- cgit v1.2.3 From ee688b000d35f413f33561ec9c7d3355be561e2f Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Thu, 24 Jan 2008 19:38:39 +0100 Subject: nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 64 ++++++++++++++++++++++++++++++++++++++++++ net/wireless/nl80211.c | 74 +++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 138 insertions(+) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 9fecf902419c..63695060db9f 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -161,6 +161,9 @@ enum nl80211_commands { * given for %NL80211_CMD_GET_STATION, nested attribute containing * info as possible, see &enum nl80211_sta_stats. * + * @NL80211_ATTR_WIPHY_BANDS: Information about an operating bands, + * consisting of a nested array. + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -195,6 +198,8 @@ enum nl80211_attrs { NL80211_ATTR_STA_VLAN, NL80211_ATTR_STA_STATS, + NL80211_ATTR_WIPHY_BANDS, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -280,4 +285,63 @@ enum nl80211_sta_stats { NL80211_STA_STAT_MAX = __NL80211_STA_STAT_AFTER_LAST - 1 }; +/** + * enum nl80211_band_attr - band attributes + * @__NL80211_BAND_ATTR_INVALID: attribute number 0 is reserved + * @NL80211_BAND_ATTR_FREQS: supported frequencies in this band, + * an array of nested frequency attributes + * @NL80211_BAND_ATTR_RATES: supported bitrates in this band, + * an array of nested bitrate attributes + */ +enum nl80211_band_attr { + __NL80211_BAND_ATTR_INVALID, + NL80211_BAND_ATTR_FREQS, + NL80211_BAND_ATTR_RATES, + + /* keep last */ + __NL80211_BAND_ATTR_AFTER_LAST, + NL80211_BAND_ATTR_MAX = __NL80211_BAND_ATTR_AFTER_LAST - 1 +}; + +/** + * enum nl80211_frequency_attr - frequency attributes + * @NL80211_FREQUENCY_ATTR_FREQ: Frequency in MHz + * @NL80211_FREQUENCY_ATTR_DISABLED: Channel is disabled in current + * regulatory domain. + * @NL80211_FREQUENCY_ATTR_PASSIVE_SCAN: Only passive scanning is + * permitted on this channel in current regulatory domain. + * @NL80211_FREQUENCY_ATTR_NO_IBSS: IBSS networks are not permitted + * on this channel in current regulatory domain. + * @NL80211_FREQUENCY_ATTR_RADAR: Radar detection is mandatory + * on this channel in current regulatory domain. + */ +enum nl80211_frequency_attr { + __NL80211_FREQUENCY_ATTR_INVALID, + NL80211_FREQUENCY_ATTR_FREQ, + NL80211_FREQUENCY_ATTR_DISABLED, + NL80211_FREQUENCY_ATTR_PASSIVE_SCAN, + NL80211_FREQUENCY_ATTR_NO_IBSS, + NL80211_FREQUENCY_ATTR_RADAR, + + /* keep last */ + __NL80211_FREQUENCY_ATTR_AFTER_LAST, + NL80211_FREQUENCY_ATTR_MAX = __NL80211_FREQUENCY_ATTR_AFTER_LAST - 1 +}; + +/** + * enum nl80211_bitrate_attr - bitrate attributes + * @NL80211_BITRATE_ATTR_RATE: Bitrate in units of 100 kbps + * @NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE: Short preamble supported + * in 2.4 GHz band. + */ +enum nl80211_bitrate_attr { + __NL80211_BITRATE_ATTR_INVALID, + NL80211_BITRATE_ATTR_RATE, + NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE, + + /* keep last */ + __NL80211_BITRATE_ATTR_AFTER_LAST, + NL80211_BITRATE_ATTR_MAX = __NL80211_BITRATE_ATTR_AFTER_LAST - 1 +}; + #endif /* __LINUX_NL80211_H */ diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index e3a214f63f91..b123f58d3909 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -98,6 +98,13 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, struct cfg80211_registered_device *dev) { void *hdr; + struct nlattr *nl_bands, *nl_band; + struct nlattr *nl_freqs, *nl_freq; + struct nlattr *nl_rates, *nl_rate; + enum ieee80211_band band; + struct ieee80211_channel *chan; + struct ieee80211_rate *rate; + int i; hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_WIPHY); if (!hdr) @@ -105,6 +112,73 @@ static int nl80211_send_wiphy(struct sk_buff *msg, u32 pid, u32 seq, int flags, NLA_PUT_U32(msg, NL80211_ATTR_WIPHY, dev->idx); NLA_PUT_STRING(msg, NL80211_ATTR_WIPHY_NAME, wiphy_name(&dev->wiphy)); + + nl_bands = nla_nest_start(msg, NL80211_ATTR_WIPHY_BANDS); + if (!nl_bands) + goto nla_put_failure; + + for (band = 0; band < IEEE80211_NUM_BANDS; band++) { + if (!dev->wiphy.bands[band]) + continue; + + nl_band = nla_nest_start(msg, band); + if (!nl_band) + goto nla_put_failure; + + /* add frequencies */ + nl_freqs = nla_nest_start(msg, NL80211_BAND_ATTR_FREQS); + if (!nl_freqs) + goto nla_put_failure; + + for (i = 0; i < dev->wiphy.bands[band]->n_channels; i++) { + nl_freq = nla_nest_start(msg, i); + if (!nl_freq) + goto nla_put_failure; + + chan = &dev->wiphy.bands[band]->channels[i]; + NLA_PUT_U32(msg, NL80211_FREQUENCY_ATTR_FREQ, + chan->center_freq); + + if (chan->flags & IEEE80211_CHAN_DISABLED) + NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_DISABLED); + if (chan->flags & IEEE80211_CHAN_PASSIVE_SCAN) + NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_PASSIVE_SCAN); + if (chan->flags & IEEE80211_CHAN_NO_IBSS) + NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_NO_IBSS); + if (chan->flags & IEEE80211_CHAN_RADAR) + NLA_PUT_FLAG(msg, NL80211_FREQUENCY_ATTR_RADAR); + + nla_nest_end(msg, nl_freq); + } + + nla_nest_end(msg, nl_freqs); + + /* add bitrates */ + nl_rates = nla_nest_start(msg, NL80211_BAND_ATTR_RATES); + if (!nl_rates) + goto nla_put_failure; + + for (i = 0; i < dev->wiphy.bands[band]->n_bitrates; i++) { + nl_rate = nla_nest_start(msg, i); + if (!nl_rate) + goto nla_put_failure; + + rate = &dev->wiphy.bands[band]->bitrates[i]; + NLA_PUT_U32(msg, NL80211_BITRATE_ATTR_RATE, + rate->bitrate); + if (rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) + NLA_PUT_FLAG(msg, + NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE); + + nla_nest_end(msg, nl_rate); + } + + nla_nest_end(msg, nl_rates); + + nla_nest_end(msg, nl_band); + } + nla_nest_end(msg, nl_bands); + return genlmsg_end(msg, hdr); nla_put_failure: -- cgit v1.2.3 From 66f7ac50ed7cc5c19a62bc97e8f6e7891004a03a Mon Sep 17 00:00:00 2001 From: Michael Wu Date: Thu, 31 Jan 2008 19:48:22 +0100 Subject: nl80211: Add monitor interface configuration flags This allows precise control over what a monitor interface shows. Signed-off-by: Michael Wu Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 35 +++++++++++++++++++++++++++++++++++ include/net/cfg80211.h | 24 ++++++++++++++++++++++-- net/mac80211/cfg.c | 4 ++-- net/wireless/nl80211.c | 44 ++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 101 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index 63695060db9f..a9f0b93324a2 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -164,6 +164,9 @@ enum nl80211_commands { * @NL80211_ATTR_WIPHY_BANDS: Information about an operating bands, * consisting of a nested array. * + * @NL80211_ATTR_MNTR_FLAGS: flags, nested element with NLA_FLAG attributes of + * &enum nl80211_mntr_flags. + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -200,6 +203,8 @@ enum nl80211_attrs { NL80211_ATTR_WIPHY_BANDS, + NL80211_ATTR_MNTR_FLAGS, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -344,4 +349,34 @@ enum nl80211_bitrate_attr { NL80211_BITRATE_ATTR_MAX = __NL80211_BITRATE_ATTR_AFTER_LAST - 1 }; +/** + * enum nl80211_mntr_flags - monitor configuration flags + * + * Monitor configuration flags. + * + * @__NL80211_MNTR_FLAG_INVALID: reserved + * + * @NL80211_MNTR_FLAG_FCSFAIL: pass frames with bad FCS + * @NL80211_MNTR_FLAG_PLCPFAIL: pass frames with bad PLCP + * @NL80211_MNTR_FLAG_CONTROL: pass control frames + * @NL80211_MNTR_FLAG_OTHER_BSS: disable BSSID filtering + * @NL80211_MNTR_FLAG_COOK_FRAMES: report frames after processing. + * overrides all other flags. + * + * @__NL80211_MNTR_FLAG_AFTER_LAST: internal use + * @NL80211_MNTR_FLAG_MAX: highest possible monitor flag + */ +enum nl80211_mntr_flags { + __NL80211_MNTR_FLAG_INVALID, + NL80211_MNTR_FLAG_FCSFAIL, + NL80211_MNTR_FLAG_PLCPFAIL, + NL80211_MNTR_FLAG_CONTROL, + NL80211_MNTR_FLAG_OTHER_BSS, + NL80211_MNTR_FLAG_COOK_FRAMES, + + /* keep last */ + __NL80211_MNTR_FLAG_AFTER_LAST, + NL80211_MNTR_FLAG_MAX = __NL80211_MNTR_FLAG_AFTER_LAST - 1 +}; + #endif /* __LINUX_NL80211_H */ diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index bcc480b8892a..ab4caf63954f 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -163,6 +163,26 @@ struct station_stats { u32 tx_bytes; }; +/** + * enum monitor_flags - monitor flags + * + * Monitor interface configuration flags. Note that these must be the bits + * according to the nl80211 flags. + * + * @MONITOR_FLAG_FCSFAIL: pass frames with bad FCS + * @MONITOR_FLAG_PLCPFAIL: pass frames with bad PLCP + * @MONITOR_FLAG_CONTROL: pass control frames + * @MONITOR_FLAG_OTHER_BSS: disable BSSID filtering + * @MONITOR_FLAG_COOK_FRAMES: report frames after processing + */ +enum monitor_flags { + MONITOR_FLAG_FCSFAIL = 1<attrs[NL80211_ATTR_IFTYPE]) { type = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]); @@ -362,7 +393,11 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) } rtnl_lock(); - err = drv->ops->change_virtual_intf(&drv->wiphy, ifindex, type); + err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ? + info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL, + &flags); + err = drv->ops->change_virtual_intf(&drv->wiphy, ifindex, + type, err ? NULL : &flags); rtnl_unlock(); unlock: @@ -375,6 +410,7 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) struct cfg80211_registered_device *drv; int err; enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED; + u32 flags; if (!info->attrs[NL80211_ATTR_IFNAME]) return -EINVAL; @@ -395,8 +431,12 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) } rtnl_lock(); + err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ? + info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL, + &flags); err = drv->ops->add_virtual_intf(&drv->wiphy, - nla_data(info->attrs[NL80211_ATTR_IFNAME]), type); + nla_data(info->attrs[NL80211_ATTR_IFNAME]), + type, err ? NULL : &flags); rtnl_unlock(); unlock: -- cgit v1.2.3 From 3330d7be7008fa8e213648750fc13613eecc54bb Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 10 Feb 2008 16:49:38 +0100 Subject: mac80211: give burst time in txop rather than 0.1msec units This changes mac80211 to pass the burst time to conf_tx in txop units rather than 0.1msec units. 0.1msec units are only required by atheros hardware (according to current driver support), all other drivers do other calculations or require the txop value. Therefore, it results in fewer calculations and more precision if we just pass the txop value through to the driver. Signed-off-by: Johannes Berg Acked-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl3945-base.c | 2 +- drivers/net/wireless/iwlwifi/iwl4965-base.c | 2 +- drivers/net/wireless/p54common.c | 21 +++++++++++++-------- include/net/mac80211.h | 12 ++++++------ net/mac80211/ieee80211_sta.c | 11 ++++++----- 5 files changed, 27 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c index 4ad26f7613c5..bb3a5f633227 100644 --- a/drivers/net/wireless/iwlwifi/iwl3945-base.c +++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c @@ -7442,7 +7442,7 @@ static int iwl3945_mac_conf_tx(struct ieee80211_hw *hw, int queue, priv->qos_data.def_qos_parm.ac[q].cw_max = cpu_to_le16(params->cw_max); priv->qos_data.def_qos_parm.ac[q].aifsn = params->aifs; priv->qos_data.def_qos_parm.ac[q].edca_txop = - cpu_to_le16((params->burst_time * 100)); + cpu_to_le16((params->txop * 32)); priv->qos_data.def_qos_parm.ac[q].reserved1 = 0; priv->qos_data.qos_active = 1; diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c index f8931993ab70..fb4c84e31932 100644 --- a/drivers/net/wireless/iwlwifi/iwl4965-base.c +++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c @@ -7916,7 +7916,7 @@ static int iwl4965_mac_conf_tx(struct ieee80211_hw *hw, int queue, priv->qos_data.def_qos_parm.ac[q].cw_max = cpu_to_le16(params->cw_max); priv->qos_data.def_qos_parm.ac[q].aifsn = params->aifs; priv->qos_data.def_qos_parm.ac[q].edca_txop = - cpu_to_le16((params->burst_time * 100)); + cpu_to_le16((params->txop * 32)); priv->qos_data.def_qos_parm.ac[q].reserved1 = 0; priv->qos_data.qos_active = 1; diff --git a/drivers/net/wireless/p54common.c b/drivers/net/wireless/p54common.c index 218ff7770ef6..0e9f887178ce 100644 --- a/drivers/net/wireless/p54common.c +++ b/drivers/net/wireless/p54common.c @@ -759,13 +759,12 @@ static int p54_set_leds(struct ieee80211_hw *dev, int mode, int link, int act) return 0; } -#define P54_SET_QUEUE(queue, ai_fs, cw_min, cw_max, burst) \ +#define P54_SET_QUEUE(queue, ai_fs, cw_min, cw_max, _txop) \ do { \ queue.aifs = cpu_to_le16(ai_fs); \ queue.cwmin = cpu_to_le16(cw_min); \ queue.cwmax = cpu_to_le16(cw_max); \ - queue.txop = (burst == 0) ? \ - 0 : cpu_to_le16((burst * 100) / 32 + 1); \ + queue.txop = cpu_to_le16(_txop); \ } while(0) static void p54_init_vdcf(struct ieee80211_hw *dev) @@ -783,10 +782,16 @@ static void p54_init_vdcf(struct ieee80211_hw *dev) vdcf = (struct p54_tx_control_vdcf *) hdr->data; - P54_SET_QUEUE(vdcf->queue[0], 0x0002, 0x0003, 0x0007, 0x000f); - P54_SET_QUEUE(vdcf->queue[1], 0x0002, 0x0007, 0x000f, 0x001e); - P54_SET_QUEUE(vdcf->queue[2], 0x0002, 0x000f, 0x03ff, 0x0014); - P54_SET_QUEUE(vdcf->queue[3], 0x0007, 0x000f, 0x03ff, 0x0000); + /* + * FIXME: The default values in the spec (IEEE 802.11 + * 7.3.2.19 Table 37) are 47, 94, 0, 0, why use + * 47, 94, 63, 0 here? Also, the default AIFS + * values (second parameter) are 2, 2, 3, 7... + */ + P54_SET_QUEUE(vdcf->queue[0], 0x0002, 0x0003, 0x0007, 47); + P54_SET_QUEUE(vdcf->queue[1], 0x0002, 0x0007, 0x000f, 94); + P54_SET_QUEUE(vdcf->queue[2], 0x0002, 0x000f, 0x03ff, 63); + P54_SET_QUEUE(vdcf->queue[3], 0x0007, 0x000f, 0x03ff, 0); } static void p54_set_vdcf(struct ieee80211_hw *dev) @@ -939,7 +944,7 @@ static int p54_conf_tx(struct ieee80211_hw *dev, int queue, if ((params) && !((queue < 0) || (queue > 4))) { P54_SET_QUEUE(vdcf->queue[queue], params->aifs, - params->cw_min, params->cw_max, params->burst_time); + params->cw_min, params->cw_max, params->txop); } else return -EINVAL; diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 460da54a0019..5ecf3cc8d977 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -89,19 +89,19 @@ struct ieee80211_ht_bss_info { * struct ieee80211_tx_queue_params - transmit queue configuration * * The information provided in this structure is required for QoS - * transmit queue configuration. + * transmit queue configuration. Cf. IEEE 802.11 7.3.2.29. * * @aifs: arbitration interface space [0..255, -1: use default] * @cw_min: minimum contention window [will be a value of the form * 2^n-1 in the range 1..1023; 0: use default] * @cw_max: maximum contention window [like @cw_min] - * @burst_time: maximum burst time in units of 0.1ms, 0 meaning disabled + * @txop: maximum burst time in units of 32 usecs, 0 meaning disabled */ struct ieee80211_tx_queue_params { - int aifs; - int cw_min; - int cw_max; - int burst_time; + s16 aifs; + u16 cw_min; + u16 cw_max; + u16 txop; }; /** diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index a8251a23917b..11f2ee6b2269 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -297,12 +297,13 @@ static void ieee80211_sta_wmm_params(struct net_device *dev, params.aifs = pos[0] & 0x0f; params.cw_max = ecw2cw((pos[1] & 0xf0) >> 4); params.cw_min = ecw2cw(pos[1] & 0x0f); - /* TXOP is in units of 32 usec; burst_time in 0.1 ms */ - params.burst_time = (pos[2] | (pos[3] << 8)) * 32 / 100; + params.txop = pos[2] | (pos[3] << 8); +#ifdef CONFIG_MAC80211_DEBUG printk(KERN_DEBUG "%s: WMM queue=%d aci=%d acm=%d aifs=%d " - "cWmin=%d cWmax=%d burst=%d\n", + "cWmin=%d cWmax=%d txop=%d\n", dev->name, queue, aci, acm, params.aifs, params.cw_min, - params.cw_max, params.burst_time); + params.cw_max, params.txop); +#endif /* TODO: handle ACM (block TX, fallback to next lowest allowed * AC for now) */ if (local->ops->conf_tx(local_to_hw(local), queue, ¶ms)) { @@ -3230,7 +3231,7 @@ int ieee80211_sta_set_ssid(struct net_device *dev, char *ssid, size_t len) qparam.cw_min = 15; qparam.cw_max = 1023; - qparam.burst_time = 0; + qparam.txop = 0; for (i = IEEE80211_TX_QUEUE_DATA0; i < NUM_TX_DATA_QUEUES; i++) local->ops->conf_tx(local_to_hw(local), -- cgit v1.2.3 From d0f5afbe6de6b8c06f94a8a0b370252e3863afe7 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Tue, 12 Feb 2008 20:12:45 +0100 Subject: mac80211: Extend filter flag documentation about unsupported flags This extends the filter flags documentation to make it clear what clearing a flag really means. Signed-off-by: Michael Buesch Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 5ecf3cc8d977..027d51f876ee 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -793,8 +793,18 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr) * parameter to see whether multicast frames should be accepted * or dropped. * - * All unsupported flags in @total_flags must be cleared, i.e. you - * should clear all bits except those you honoured. + * All unsupported flags in @total_flags must be cleared. + * Hardware does not support a flag if it is incapable of _passing_ + * the frame to the stack. Otherwise the driver must ignore + * the flag, but not clear it. + * You must _only_ clear the flag (announce no support for the + * flag to mac80211) if you are not able to pass the packet type + * to the stack (so the hardware always filters it). + * So for example, you should clear @FIF_CONTROL, if your hardware + * always filters control frames. If your hardware always passes + * control frames to the kernel and is incapable of filtering them, + * you do _not_ clear the @FIF_CONTROL flag. + * This rule applies to all other FIF flags as well. */ /** -- cgit v1.2.3 From c132bec33c2eb5e46d8e4b80cfa5a9656d8e57e7 Mon Sep 17 00:00:00 2001 From: Bruno Randolf Date: Mon, 18 Feb 2008 11:20:51 +0900 Subject: mac80211: better definition of mactime define mactime as the time when the first data symbol arrived at the HW. the old definition was questionable because 802.11 defines timestamp only for beacon and probe response frames, and there it means the timestamp field. a stricter definition of mactime is necessary for correct merging of IBSS. note that it is up to the driver to convert whatever its hardware returns to this definition. unfortunately we don't know for example when atheros hardware takes its rx timestamp exactly :( Signed-off-by: Bruno Randolf Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 027d51f876ee..1a1d0d8203aa 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -295,7 +295,8 @@ enum mac80211_rx_flags { * The low-level driver should provide this information (the subset * supported by hardware) to the 802.11 code with each received * frame. - * @mactime: MAC timestamp as defined by 802.11 + * @mactime: value in microseconds of the 64-bit Time Synchronization Function + * (TSF) timer when the first data symbol (MPDU) arrived at the hardware. * @band: the active band when this frame was received * @freq: frequency the radio was tuned to when receiving this frame, in MHz * @ssi: signal strength when receiving this frame -- cgit v1.2.3 From 9d9bf77d16ba527f6f63846ca18cf20ae6e8d697 Mon Sep 17 00:00:00 2001 From: Bruno Randolf Date: Mon, 18 Feb 2008 11:21:36 +0900 Subject: mac80211: enable IBSS merging enable IBSS cell merging. if an IBSS beacon with the same channel, same ESSID and a TSF higher than the local TSF (mactime) is received, we have to join its BSSID. while this might not be immediately apparent from reading the 802.11 standard it is compliant and necessary to make IBSS mode functional in many cases. most drivers have a similar behaviour. * move the relevant code section (previously only containing debug code) down to the end of the function, so we can reuse the bss structure. * we have to compare the mactime (TSF at the time of packet receive) rather than the current TSF. since mactime is defined as the time the first data symbol arrived we add the time until byte 24 where the timestamp resides, since this is how the beacon timestamp is defined. as some some drivers are not able to give a reliable mactime we fall back to use the current TSF, which will be enough to catch most (but not all) cases where an IBSS merge is necessary. * in IBSS mode we want to allow beacons to override probe response info so we can correctly do merges. * we don't only configure beacons based on scan results, so change that message. * to enable this we have to let all beacons thru in IBSS mode, even if they have a different BSSID. Signed-off-by: Bruno Randolf Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 3 +- net/mac80211/ieee80211_sta.c | 90 ++++++++++++++++++++++++++++++-------------- net/mac80211/rx.c | 5 ++- 3 files changed, 67 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 1a1d0d8203aa..784ab769b00b 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -276,7 +276,8 @@ struct ieee80211_tx_control { * @RX_FLAG_FAILED_PLCP_CRC: Set this flag if the PCLP check failed on * the frame. * @RX_FLAG_TSFT: The timestamp passed in the RX status (@mactime field) - * is valid. + * is valid. This is useful in monitor mode and necessary for beacon frames + * to enable IBSS merging. */ enum mac80211_rx_flags { RX_FLAG_MMIC_ERROR = 1<<0, diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index a84ead49b1e6..6e5b57dccc50 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -2199,7 +2199,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev, struct ieee80211_sta_bss *bss; struct sta_info *sta; struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); - u64 timestamp; + u64 beacon_timestamp, rx_timestamp; DECLARE_MAC_BUF(mac); DECLARE_MAC_BUF(mac2); @@ -2216,31 +2216,7 @@ static void ieee80211_rx_bss_info(struct net_device *dev, if (baselen > len) return; - timestamp = le64_to_cpu(mgmt->u.beacon.timestamp); - - if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && beacon && - memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0) { -#ifdef CONFIG_MAC80211_IBSS_DEBUG - static unsigned long last_tsf_debug = 0; - u64 tsf; - if (local->ops->get_tsf) - tsf = local->ops->get_tsf(local_to_hw(local)); - else - tsf = -1LLU; - if (time_after(jiffies, last_tsf_debug + 5 * HZ)) { - printk(KERN_DEBUG "RX beacon SA=%s BSSID=" - "%s TSF=0x%llx BCN=0x%llx diff=%lld " - "@%lu\n", - print_mac(mac, mgmt->sa), print_mac(mac2, mgmt->bssid), - (unsigned long long)tsf, - (unsigned long long)timestamp, - (unsigned long long)(tsf - timestamp), - jiffies); - last_tsf_debug = jiffies; - } -#endif /* CONFIG_MAC80211_IBSS_DEBUG */ - } - + beacon_timestamp = le64_to_cpu(mgmt->u.beacon.timestamp); ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems); if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates && @@ -2326,8 +2302,10 @@ static void ieee80211_rx_bss_info(struct net_device *dev, bss->band = rx_status->band; - if (bss->probe_resp && beacon) { - /* Do not allow beacon to override data from Probe Response. */ + if (sdata->vif.type != IEEE80211_IF_TYPE_IBSS && + bss->probe_resp && beacon) { + /* STA mode: + * Do not allow beacon to override data from Probe Response. */ ieee80211_rx_bss_put(dev, bss); return; } @@ -2424,13 +2402,67 @@ static void ieee80211_rx_bss_info(struct net_device *dev, bss->ht_ie_len = 0; } - bss->timestamp = timestamp; + bss->timestamp = beacon_timestamp; bss->last_update = jiffies; bss->rssi = rx_status->ssi; bss->signal = rx_status->signal; bss->noise = rx_status->noise; if (!beacon) bss->probe_resp++; + + /* check if we need to merge IBSS */ + if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && beacon && + !local->sta_sw_scanning && !local->sta_hw_scanning && + mgmt->u.beacon.capab_info & WLAN_CAPABILITY_IBSS && + bss->freq == local->oper_channel->center_freq && + elems.ssid_len == sdata->u.sta.ssid_len && + memcmp(elems.ssid, sdata->u.sta.ssid, sdata->u.sta.ssid_len) == 0) { + if (rx_status->flag & RX_FLAG_TSFT) { + /* in order for correct IBSS merging we need mactime + * + * since mactime is defined as the time the first data + * symbol of the frame hits the PHY, and the timestamp + * of the beacon is defined as "the time that the data + * symbol containing the first bit of the timestamp is + * transmitted to the PHY plus the transmitting STA’s + * delays through its local PHY from the MAC-PHY + * interface to its interface with the WM" + * (802.11 11.1.2) - equals the time this bit arrives at + * the receiver - we have to take into account the + * offset between the two. + * e.g: at 1 MBit that means mactime is 192 usec earlier + * (=24 bytes * 8 usecs/byte) than the beacon timestamp. + */ + int rate = local->hw.wiphy->bands[rx_status->band]-> + bitrates[rx_status->rate_idx].bitrate; + rx_timestamp = rx_status->mactime + (24 * 8 * 10 / rate); + } else if (local && local->ops && local->ops->get_tsf) + /* second best option: get current TSF */ + rx_timestamp = local->ops->get_tsf(local_to_hw(local)); + else + /* can't merge without knowing the TSF */ + rx_timestamp = -1LLU; +#ifdef CONFIG_MAC80211_IBSS_DEBUG + printk(KERN_DEBUG "RX beacon SA=%s BSSID=" + "%s TSF=0x%llx BCN=0x%llx diff=%lld @%lu\n", + print_mac(mac, mgmt->sa), + print_mac(mac2, mgmt->bssid), + (unsigned long long)rx_timestamp, + (unsigned long long)beacon_timestamp, + (unsigned long long)(rx_timestamp - beacon_timestamp), + jiffies); +#endif /* CONFIG_MAC80211_IBSS_DEBUG */ + if (beacon_timestamp > rx_timestamp) { + if (CONFIG_MAC80211_IBSS_DEBUG || net_ratelimit()) + printk(KERN_DEBUG "%s: beacon TSF higher than " + "local TSF - IBSS merge with BSSID %s\n", + dev->name, print_mac(mac, mgmt->bssid)); + ieee80211_sta_join_ibss(dev, &sdata->u.sta, bss); + ieee80211_ibss_add_sta(dev, NULL, + mgmt->bssid, mgmt->sa); + } + } + ieee80211_rx_bss_put(dev, bss); } diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 472b19fc9143..279ee493520d 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1695,7 +1695,10 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, case IEEE80211_IF_TYPE_IBSS: if (!bssid) return 0; - if (!ieee80211_bssid_match(bssid, sdata->u.sta.bssid)) { + if ((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_MGMT && + (rx->fc & IEEE80211_FCTL_STYPE) == IEEE80211_STYPE_BEACON) + return 1; + else if (!ieee80211_bssid_match(bssid, sdata->u.sta.bssid)) { if (!(rx->flags & IEEE80211_TXRXD_RXIN_SCAN)) return 0; rx->flags &= ~IEEE80211_TXRXD_RXRA_MATCH; -- cgit v1.2.3 From ffc7689ddae5cbe12bde437ae0f2b386d568b5cd Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Wed, 20 Feb 2008 19:08:10 +0100 Subject: ssb: Add support for 8bit register access This adds support for 8bit wide register reads/writes. This is needed in order to support the gigabit ethernet core. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/ssb/main.c | 18 ++++++++++++++++++ drivers/ssb/pci.c | 28 ++++++++++++++++++++++++++++ drivers/ssb/pcmcia.c | 32 ++++++++++++++++++++++++++++++++ include/linux/ssb/ssb.h | 10 ++++++++++ 4 files changed, 88 insertions(+) (limited to 'include') diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c index bedb2b4ee9d2..8db40c4b86e9 100644 --- a/drivers/ssb/main.c +++ b/drivers/ssb/main.c @@ -505,6 +505,14 @@ error: return err; } +static u8 ssb_ssb_read8(struct ssb_device *dev, u16 offset) +{ + struct ssb_bus *bus = dev->bus; + + offset += dev->core_index * SSB_CORE_SIZE; + return readb(bus->mmio + offset); +} + static u16 ssb_ssb_read16(struct ssb_device *dev, u16 offset) { struct ssb_bus *bus = dev->bus; @@ -521,6 +529,14 @@ static u32 ssb_ssb_read32(struct ssb_device *dev, u16 offset) return readl(bus->mmio + offset); } +static void ssb_ssb_write8(struct ssb_device *dev, u16 offset, u8 value) +{ + struct ssb_bus *bus = dev->bus; + + offset += dev->core_index * SSB_CORE_SIZE; + writeb(value, bus->mmio + offset); +} + static void ssb_ssb_write16(struct ssb_device *dev, u16 offset, u16 value) { struct ssb_bus *bus = dev->bus; @@ -539,8 +555,10 @@ static void ssb_ssb_write32(struct ssb_device *dev, u16 offset, u32 value) /* Ops for the plain SSB bus without a host-device (no PCI or PCMCIA). */ static const struct ssb_bus_ops ssb_ssb_ops = { + .read8 = ssb_ssb_read8, .read16 = ssb_ssb_read16, .read32 = ssb_ssb_read32, + .write8 = ssb_ssb_write8, .write16 = ssb_ssb_write16, .write32 = ssb_ssb_write32, }; diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c index b434df75047f..1facc7620fc8 100644 --- a/drivers/ssb/pci.c +++ b/drivers/ssb/pci.c @@ -572,6 +572,19 @@ static inline int ssb_pci_assert_buspower(struct ssb_bus *bus) } #endif /* DEBUG */ +static u8 ssb_pci_read8(struct ssb_device *dev, u16 offset) +{ + struct ssb_bus *bus = dev->bus; + + if (unlikely(ssb_pci_assert_buspower(bus))) + return 0xFF; + if (unlikely(bus->mapped_device != dev)) { + if (unlikely(ssb_pci_switch_core(bus, dev))) + return 0xFF; + } + return ioread8(bus->mmio + offset); +} + static u16 ssb_pci_read16(struct ssb_device *dev, u16 offset) { struct ssb_bus *bus = dev->bus; @@ -598,6 +611,19 @@ static u32 ssb_pci_read32(struct ssb_device *dev, u16 offset) return ioread32(bus->mmio + offset); } +static void ssb_pci_write8(struct ssb_device *dev, u16 offset, u8 value) +{ + struct ssb_bus *bus = dev->bus; + + if (unlikely(ssb_pci_assert_buspower(bus))) + return; + if (unlikely(bus->mapped_device != dev)) { + if (unlikely(ssb_pci_switch_core(bus, dev))) + return; + } + iowrite8(value, bus->mmio + offset); +} + static void ssb_pci_write16(struct ssb_device *dev, u16 offset, u16 value) { struct ssb_bus *bus = dev->bus; @@ -626,8 +652,10 @@ static void ssb_pci_write32(struct ssb_device *dev, u16 offset, u32 value) /* Not "static", as it's used in main.c */ const struct ssb_bus_ops ssb_pci_ops = { + .read8 = ssb_pci_read8, .read16 = ssb_pci_read16, .read32 = ssb_pci_read32, + .write8 = ssb_pci_write8, .write16 = ssb_pci_write16, .write32 = ssb_pci_write32, }; diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c index 46816cda8b98..84b3a845a8a8 100644 --- a/drivers/ssb/pcmcia.c +++ b/drivers/ssb/pcmcia.c @@ -172,6 +172,22 @@ static int select_core_and_segment(struct ssb_device *dev, return 0; } +static u8 ssb_pcmcia_read8(struct ssb_device *dev, u16 offset) +{ + struct ssb_bus *bus = dev->bus; + unsigned long flags; + int err; + u8 value = 0xFF; + + spin_lock_irqsave(&bus->bar_lock, flags); + err = select_core_and_segment(dev, &offset); + if (likely(!err)) + value = readb(bus->mmio + offset); + spin_unlock_irqrestore(&bus->bar_lock, flags); + + return value; +} + static u16 ssb_pcmcia_read16(struct ssb_device *dev, u16 offset) { struct ssb_bus *bus = dev->bus; @@ -206,6 +222,20 @@ static u32 ssb_pcmcia_read32(struct ssb_device *dev, u16 offset) return (lo | (hi << 16)); } +static void ssb_pcmcia_write8(struct ssb_device *dev, u16 offset, u8 value) +{ + struct ssb_bus *bus = dev->bus; + unsigned long flags; + int err; + + spin_lock_irqsave(&bus->bar_lock, flags); + err = select_core_and_segment(dev, &offset); + if (likely(!err)) + writeb(value, bus->mmio + offset); + mmiowb(); + spin_unlock_irqrestore(&bus->bar_lock, flags); +} + static void ssb_pcmcia_write16(struct ssb_device *dev, u16 offset, u16 value) { struct ssb_bus *bus = dev->bus; @@ -238,8 +268,10 @@ static void ssb_pcmcia_write32(struct ssb_device *dev, u16 offset, u32 value) /* Not "static", as it's used in main.c */ const struct ssb_bus_ops ssb_pcmcia_ops = { + .read8 = ssb_pcmcia_read8, .read16 = ssb_pcmcia_read16, .read32 = ssb_pcmcia_read32, + .write8 = ssb_pcmcia_write8, .write16 = ssb_pcmcia_write16, .write32 = ssb_pcmcia_write32, }; diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h index 20add65215af..860d28c6d149 100644 --- a/include/linux/ssb/ssb.h +++ b/include/linux/ssb/ssb.h @@ -72,8 +72,10 @@ struct ssb_device; /* Lowlevel read/write operations on the device MMIO. * Internal, don't use that outside of ssb. */ struct ssb_bus_ops { + u8 (*read8)(struct ssb_device *dev, u16 offset); u16 (*read16)(struct ssb_device *dev, u16 offset); u32 (*read32)(struct ssb_device *dev, u16 offset); + void (*write8)(struct ssb_device *dev, u16 offset, u8 value); void (*write16)(struct ssb_device *dev, u16 offset, u16 value); void (*write32)(struct ssb_device *dev, u16 offset, u32 value); }; @@ -344,6 +346,10 @@ void ssb_device_disable(struct ssb_device *dev, u32 core_specific_flags); /* Device MMIO register read/write functions. */ +static inline u8 ssb_read8(struct ssb_device *dev, u16 offset) +{ + return dev->ops->read8(dev, offset); +} static inline u16 ssb_read16(struct ssb_device *dev, u16 offset) { return dev->ops->read16(dev, offset); @@ -352,6 +358,10 @@ static inline u32 ssb_read32(struct ssb_device *dev, u16 offset) { return dev->ops->read32(dev, offset); } +static inline void ssb_write8(struct ssb_device *dev, u16 offset, u8 value) +{ + dev->ops->write8(dev, offset, value); +} static inline void ssb_write16(struct ssb_device *dev, u16 offset, u16 value) { dev->ops->write16(dev, offset, value); -- cgit v1.2.3 From d46e144b65bf053b25d134ec9f52a38e63e04bb4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 20 Feb 2008 23:59:33 +0100 Subject: mac80211: rework TX filtered frame code This reworks the code for TX filtered frames, splitting it out to a new function to handle those cases, making the clear instruction a flag and renaming a few things to be easier to understand and less Atheros hardware specific. Finally, it also makes the comments explain more. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 3 +- net/mac80211/ieee80211.c | 121 +++++++++++++++++++++++++++++++---------------- net/mac80211/sta_info.h | 6 ++- net/mac80211/tx.c | 12 ++--- 4 files changed, 91 insertions(+), 51 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 784ab769b00b..cd4b1c1a0d48 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -228,7 +228,8 @@ struct ieee80211_tx_control { #define IEEE80211_TXCTL_NO_ACK (1<<4) /* tell the low level not to * wait for an ack */ #define IEEE80211_TXCTL_RATE_CTRL_PROBE (1<<5) -#define IEEE80211_TXCTL_CLEAR_DST_MASK (1<<6) +#define IEEE80211_TXCTL_CLEAR_PS_FILT (1<<6) /* clear powersave filter + * for destination station */ #define IEEE80211_TXCTL_REQUEUE (1<<7) #define IEEE80211_TXCTL_FIRST_FRAGMENT (1<<8) /* this is a first fragment of * the frame */ diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index 7df14799e38b..a00858dbab18 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -1178,6 +1178,77 @@ no_key: } } +static void ieee80211_handle_filtered_frame(struct ieee80211_local *local, + struct sta_info *sta, + struct sk_buff *skb, + struct ieee80211_tx_status *status) +{ + sta->tx_filtered_count++; + + /* + * Clear the TX filter mask for this STA when sending the next + * packet. If the STA went to power save mode, this will happen + * happen when it wakes up for the next time. + */ + sta->flags |= WLAN_STA_CLEAR_PS_FILT; + + /* + * This code races in the following way: + * + * (1) STA sends frame indicating it will go to sleep and does so + * (2) hardware/firmware adds STA to filter list, passes frame up + * (3) hardware/firmware processes TX fifo and suppresses a frame + * (4) we get TX status before having processed the frame and + * knowing that the STA has gone to sleep. + * + * This is actually quite unlikely even when both those events are + * processed from interrupts coming in quickly after one another or + * even at the same time because we queue both TX status events and + * RX frames to be processed by a tasklet and process them in the + * same order that they were received or TX status last. Hence, there + * is no race as long as the frame RX is processed before the next TX + * status, which drivers can ensure, see below. + * + * Note that this can only happen if the hardware or firmware can + * actually add STAs to the filter list, if this is done by the + * driver in response to set_tim() (which will only reduce the race + * this whole filtering tries to solve, not completely solve it) + * this situation cannot happen. + * + * To completely solve this race drivers need to make sure that they + * (a) don't mix the irq-safe/not irq-safe TX status/RX processing + * functions and + * (b) always process RX events before TX status events if ordering + * can be unknown, for example with different interrupt status + * bits. + */ + if (sta->flags & WLAN_STA_PS && + skb_queue_len(&sta->tx_filtered) < STA_MAX_TX_BUFFER) { + ieee80211_remove_tx_extra(local, sta->key, skb, + &status->control); + skb_queue_tail(&sta->tx_filtered, skb); + return; + } + + if (!(sta->flags & WLAN_STA_PS) && + !(status->control.flags & IEEE80211_TXCTL_REQUEUE)) { + /* Software retry the packet once */ + status->control.flags |= IEEE80211_TXCTL_REQUEUE; + ieee80211_remove_tx_extra(local, sta->key, skb, + &status->control); + dev_queue_xmit(skb); + return; + } + + if (net_ratelimit()) + printk(KERN_DEBUG "%s: dropped TX filtered frame, " + "queue_len=%d PS=%d @%lu\n", + wiphy_name(local->hw.wiphy), + skb_queue_len(&sta->tx_filtered), + !!(sta->flags & WLAN_STA_PS), jiffies); + dev_kfree_skb(skb); +} + void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb, struct ieee80211_tx_status *status) { @@ -1202,11 +1273,16 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb, sta = sta_info_get(local, hdr->addr1); if (sta) { if (sta->flags & WLAN_STA_PS) { - /* The STA is in power save mode, so assume + /* + * The STA is in power save mode, so assume * that this TX packet failed because of that. */ status->excessive_retries = 0; status->flags |= IEEE80211_TX_STATUS_TX_FILTERED; + ieee80211_handle_filtered_frame(local, sta, + skb, status); + sta_info_put(sta); + return; } sta_info_put(sta); } @@ -1216,47 +1292,8 @@ void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb, struct sta_info *sta; sta = sta_info_get(local, hdr->addr1); if (sta) { - sta->tx_filtered_count++; - - /* Clear the TX filter mask for this STA when sending - * the next packet. If the STA went to power save mode, - * this will happen when it is waking up for the next - * time. */ - sta->clear_dst_mask = 1; - - /* TODO: Is the WLAN_STA_PS flag always set here or is - * the race between RX and TX status causing some - * packets to be filtered out before 80211.o gets an - * update for PS status? This seems to be the case, so - * no changes are likely to be needed. */ - if (sta->flags & WLAN_STA_PS && - skb_queue_len(&sta->tx_filtered) < - STA_MAX_TX_BUFFER) { - ieee80211_remove_tx_extra(local, sta->key, - skb, - &status->control); - skb_queue_tail(&sta->tx_filtered, skb); - } else if (!(sta->flags & WLAN_STA_PS) && - !(status->control.flags & IEEE80211_TXCTL_REQUEUE)) { - /* Software retry the packet once */ - status->control.flags |= IEEE80211_TXCTL_REQUEUE; - ieee80211_remove_tx_extra(local, sta->key, - skb, - &status->control); - dev_queue_xmit(skb); - } else { - if (net_ratelimit()) { - printk(KERN_DEBUG "%s: dropped TX " - "filtered frame queue_len=%d " - "PS=%d @%lu\n", - wiphy_name(local->hw.wiphy), - skb_queue_len( - &sta->tx_filtered), - !!(sta->flags & WLAN_STA_PS), - jiffies); - } - dev_kfree_skb(skb); - } + ieee80211_handle_filtered_frame(local, sta, skb, + status); sta_info_put(sta); return; } diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 4099ece143ef..f7e65fa3f9ed 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -32,6 +32,9 @@ * @WLAN_STA_WME: Station is a QoS-STA. * @WLAN_STA_WDS: Station is one of our WDS peers. * @WLAN_STA_PSPOLL: Station has just PS-polled us. + * @WLAN_STA_CLEAR_PS_FILT: Clear PS filter in hardware (using the + * IEEE80211_TXCTL_CLEAR_PS_FILT control flag) when the next + * frame to this station is transmitted. */ enum ieee80211_sta_info_flags { WLAN_STA_AUTH = 1<<0, @@ -43,6 +46,7 @@ enum ieee80211_sta_info_flags { WLAN_STA_WME = 1<<6, WLAN_STA_WDS = 1<<7, WLAN_STA_PSPOLL = 1<<8, + WLAN_STA_CLEAR_PS_FILT = 1<<9, }; #define STA_TID_NUM 16 @@ -136,8 +140,6 @@ struct sta_info { struct sk_buff_head tx_filtered; /* buffer of TX frames that were * already given to low-level driver, * but were filtered */ - int clear_dst_mask; - unsigned long rx_packets, tx_packets; /* number of RX/TX MSDUs */ unsigned long rx_bytes, tx_bytes; unsigned long tx_retry_failed, tx_retry_count; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 69fdb763aa8b..1cd58e01f1ee 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -1020,10 +1020,10 @@ __ieee80211_tx_prepare(struct ieee80211_txrx_data *tx, } if (!tx->sta) - control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK; - else if (tx->sta->clear_dst_mask) { - control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK; - tx->sta->clear_dst_mask = 0; + control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT; + else if (tx->sta->flags & WLAN_STA_CLEAR_PS_FILT) { + control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT; + tx->sta->flags &= ~WLAN_STA_CLEAR_PS_FILT; } hdrlen = ieee80211_get_hdrlen(tx->fc); @@ -1084,7 +1084,7 @@ static int __ieee80211_tx(struct ieee80211_local *local, struct sk_buff *skb, if (tx->u.tx.extra_frag) { control->flags &= ~(IEEE80211_TXCTL_USE_RTS_CTS | IEEE80211_TXCTL_USE_CTS_PROTECT | - IEEE80211_TXCTL_CLEAR_DST_MASK | + IEEE80211_TXCTL_CLEAR_PS_FILT | IEEE80211_TXCTL_FIRST_FRAGMENT); for (i = 0; i < tx->u.tx.num_extra_frag; i++) { if (!tx->u.tx.extra_frag[i]) @@ -1806,7 +1806,7 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, control->antenna_sel_tx = local->hw.conf.antenna_sel_tx; control->flags |= IEEE80211_TXCTL_NO_ACK; control->retry_limit = 1; - control->flags |= IEEE80211_TXCTL_CLEAR_DST_MASK; + control->flags |= IEEE80211_TXCTL_CLEAR_PS_FILT; } ap->num_beacons++; -- cgit v1.2.3 From 28de57d1a9eb7e67badb731297197fcbef0cc19e Mon Sep 17 00:00:00 2001 From: Aurelien Jarno Date: Fri, 22 Feb 2008 16:14:58 +0100 Subject: ssb: Add CHIPCO IRQ access functions This patch adds functions to setup and read the CHIPCO IRQ. Signed-off-by: Aurelien Jarno Signed-off-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/ssb/driver_chipcommon.c | 10 ++++++++++ include/linux/ssb/ssb_driver_chipcommon.h | 4 ++++ 2 files changed, 14 insertions(+) (limited to 'include') diff --git a/drivers/ssb/driver_chipcommon.c b/drivers/ssb/driver_chipcommon.c index e586321a473a..45b672a69003 100644 --- a/drivers/ssb/driver_chipcommon.c +++ b/drivers/ssb/driver_chipcommon.c @@ -353,6 +353,16 @@ void ssb_chipco_watchdog_timer_set(struct ssb_chipcommon *cc, u32 ticks) chipco_write32(cc, SSB_CHIPCO_WATCHDOG, ticks); } +void ssb_chipco_irq_mask(struct ssb_chipcommon *cc, u32 mask, u32 value) +{ + chipco_write32_masked(cc, SSB_CHIPCO_IRQMASK, mask, value); +} + +u32 ssb_chipco_irq_status(struct ssb_chipcommon *cc, u32 mask) +{ + return chipco_read32(cc, SSB_CHIPCO_IRQSTAT) & mask; +} + u32 ssb_chipco_gpio_in(struct ssb_chipcommon *cc, u32 mask) { return chipco_read32(cc, SSB_CHIPCO_GPIOIN) & mask; diff --git a/include/linux/ssb/ssb_driver_chipcommon.h b/include/linux/ssb/ssb_driver_chipcommon.h index 536851b946f6..b548a54ff1f5 100644 --- a/include/linux/ssb/ssb_driver_chipcommon.h +++ b/include/linux/ssb/ssb_driver_chipcommon.h @@ -390,6 +390,10 @@ extern void ssb_chipco_set_clockmode(struct ssb_chipcommon *cc, extern void ssb_chipco_watchdog_timer_set(struct ssb_chipcommon *cc, u32 ticks); +void ssb_chipco_irq_mask(struct ssb_chipcommon *cc, u32 mask, u32 value); + +u32 ssb_chipco_irq_status(struct ssb_chipcommon *cc, u32 mask); + /* Chipcommon GPIO pin access. */ u32 ssb_chipco_gpio_in(struct ssb_chipcommon *cc, u32 mask); u32 ssb_chipco_gpio_out(struct ssb_chipcommon *cc, u32 mask, u32 value); -- cgit v1.2.3 From 2485f7105f20f85c2dbebc67be6b2cb97175fa7e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 25 Feb 2008 16:27:41 +0100 Subject: mac80211: clarify use of TX status/RX callbacks This patch clarifies the use of the irqsafe vs. non-irq-safe functions and their respective locking requirements. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 37 ++++++++++++++++++++++++++++++++++--- 1 file changed, 34 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index cd4b1c1a0d48..7a80c3981237 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -38,7 +38,11 @@ * called in hardware interrupt context. The low-level driver must not call any * other functions in hardware interrupt context. If there is a need for such * call, the low-level driver should first ACK the interrupt and perform the - * IEEE 802.11 code call after this, e.g. from a scheduled workqueue function. + * IEEE 802.11 code call after this, e.g. from a scheduled workqueue or even + * tasklet function. + * + * NOTE: If the driver opts to use the _irqsafe() functions, it may not also + * use the non-irqsafe functions! */ /** @@ -1204,7 +1208,10 @@ void __ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb, * buffer in @skb must start with an IEEE 802.11 header or a radiotap * header if %RX_FLAG_RADIOTAP is set in the @status flags. * - * This function may not be called in IRQ context. + * This function may not be called in IRQ context. Calls to this function + * for a single hardware must be synchronized against each other. Calls + * to this function and ieee80211_rx_irqsafe() may not be mixed for a + * single hardware. * * @hw: the hardware this frame came in on * @skb: the buffer to receive, owned by mac80211 after this call @@ -1221,7 +1228,10 @@ static inline void ieee80211_rx(struct ieee80211_hw *hw, struct sk_buff *skb, * ieee80211_rx_irqsafe - receive frame * * Like ieee80211_rx() but can be called in IRQ context - * (internally defers to a workqueue.) + * (internally defers to a tasklet.) + * + * Calls to this function and ieee80211_rx() may not be mixed for a + * single hardware. * * @hw: the hardware this frame came in on * @skb: the buffer to receive, owned by mac80211 after this call @@ -1240,6 +1250,11 @@ void ieee80211_rx_irqsafe(struct ieee80211_hw *hw, * transmitted. It is permissible to not call this function for * multicast frames but this can affect statistics. * + * This function may not be called in IRQ context. Calls to this function + * for a single hardware must be synchronized against each other. Calls + * to this function and ieee80211_tx_status_irqsafe() may not be mixed + * for a single hardware. + * * @hw: the hardware the frame was transmitted by * @skb: the frame that was transmitted, owned by mac80211 after this call * @status: status information for this frame; the status pointer need not @@ -1249,6 +1264,22 @@ void ieee80211_rx_irqsafe(struct ieee80211_hw *hw, void ieee80211_tx_status(struct ieee80211_hw *hw, struct sk_buff *skb, struct ieee80211_tx_status *status); + +/** + * ieee80211_tx_status_irqsafe - irq-safe transmit status callback + * + * Like ieee80211_tx_status() but can be called in IRQ context + * (internally defers to a tasklet.) + * + * Calls to this function and ieee80211_tx_status() may not be mixed for a + * single hardware. + * + * @hw: the hardware the frame was transmitted by + * @skb: the frame that was transmitted, owned by mac80211 after this call + * @status: status information for this frame; the status pointer need not + * be valid after this function returns and is not freed by mac80211, + * it is recommended that it points to a stack area + */ void ieee80211_tx_status_irqsafe(struct ieee80211_hw *hw, struct sk_buff *skb, struct ieee80211_tx_status *status); -- cgit v1.2.3 From 988b705077d8f922408913f4f521ae073256d4a1 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Mon, 3 Mar 2008 12:20:57 -0800 Subject: [ARP]: Introduce the arp_hdr_len helper. There are some place, that calculate the ARP header length. These calculations are correct, but a) some operate with "magic" constants, b) enlarge the code length (sometimes at the cost of coding style), c) are not informative from the first glance. The proposal is to introduce a helper, that includes all the good sides of these calculations. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 5 +---- include/linux/if_arp.h | 6 ++++++ net/core/netpoll.c | 6 ++---- net/ipv4/arp.c | 9 +++------ net/ipv4/ipconfig.c | 5 +---- net/ipv4/netfilter/arp_tables.c | 5 +---- 6 files changed, 14 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 966643473da7..5fc9d8d58ece 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2646,10 +2646,7 @@ static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct pack if (!slave || !slave_do_arp_validate(bond, slave)) goto out_unlock; - /* ARP header, plus 2 device addresses, plus 2 IP addresses. */ - if (!pskb_may_pull(skb, (sizeof(struct arphdr) + - (2 * dev->addr_len) + - (2 * sizeof(u32))))) + if (!pskb_may_pull(skb, arp_hdr_len(dev))) goto out_unlock; arp = arp_hdr(skb); diff --git a/include/linux/if_arp.h b/include/linux/if_arp.h index 296e8e86e91d..4d3401812e6c 100644 --- a/include/linux/if_arp.h +++ b/include/linux/if_arp.h @@ -156,6 +156,12 @@ static inline struct arphdr *arp_hdr(const struct sk_buff *skb) { return (struct arphdr *)skb_network_header(skb); } + +static inline int arp_hdr_len(struct net_device *dev) +{ + /* ARP header, plus 2 device addresses, plus 2 IP addresses. */ + return sizeof(struct arphdr) + (dev->addr_len + sizeof(u32)) * 2; +} #endif #endif /* _LINUX_IF_ARP_H */ diff --git a/net/core/netpoll.c b/net/core/netpoll.c index 6faa128a4c8e..7ae98659d79d 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -384,9 +384,7 @@ static void arp_reply(struct sk_buff *skb) if (skb->dev->flags & IFF_NOARP) return; - if (!pskb_may_pull(skb, (sizeof(struct arphdr) + - (2 * skb->dev->addr_len) + - (2 * sizeof(u32))))) + if (!pskb_may_pull(skb, arp_hdr_len(skb->dev))) return; skb_reset_network_header(skb); @@ -414,7 +412,7 @@ static void arp_reply(struct sk_buff *skb) ipv4_is_loopback(tip) || ipv4_is_multicast(tip)) return; - size = sizeof(struct arphdr) + 2 * (skb->dev->addr_len + 4); + size = arp_hdr_len(skb->dev); send_skb = find_skb(np, size + LL_RESERVED_SPACE(np->dev), LL_RESERVED_SPACE(np->dev)); diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 8e17f65f4002..69e80bd9774a 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -570,14 +570,13 @@ struct sk_buff *arp_create(int type, int ptype, __be32 dest_ip, * Allocate a buffer */ - skb = alloc_skb(sizeof(struct arphdr)+ 2*(dev->addr_len+4) - + LL_RESERVED_SPACE(dev), GFP_ATOMIC); + skb = alloc_skb(arp_hdr_len(dev) + LL_RESERVED_SPACE(dev), GFP_ATOMIC); if (skb == NULL) return NULL; skb_reserve(skb, LL_RESERVED_SPACE(dev)); skb_reset_network_header(skb); - arp = (struct arphdr *) skb_put(skb,sizeof(struct arphdr) + 2*(dev->addr_len+4)); + arp = (struct arphdr *) skb_put(skb, arp_hdr_len(dev)); skb->dev = dev; skb->protocol = htons(ETH_P_ARP); if (src_hw == NULL) @@ -916,9 +915,7 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev, goto freeskb; /* ARP header, plus 2 device addresses, plus 2 IP addresses. */ - if (!pskb_may_pull(skb, (sizeof(struct arphdr) + - (2 * dev->addr_len) + - (2 * sizeof(u32))))) + if (!pskb_may_pull(skb, arp_hdr_len(dev))) goto freeskb; arp = arp_hdr(skb); diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index c90e75a66e81..f84041d1f623 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -459,10 +459,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt if (rarp->ar_pro != htons(ETH_P_IP)) goto drop; - if (!pskb_may_pull(skb, - sizeof(struct arphdr) + - (2 * dev->addr_len) + - (2 * 4))) + if (!pskb_may_pull(skb, arp_hdr_len(dev))) goto drop; /* OK, it is all there and looks valid, process... */ diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index a7591ce344d2..9b5904486184 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -233,10 +233,7 @@ unsigned int arpt_do_table(struct sk_buff *skb, void *table_base; struct xt_table_info *private; - /* ARP header, plus 2 device addresses, plus 2 IP addresses. */ - if (!pskb_may_pull(skb, (sizeof(struct arphdr) + - (2 * skb->dev->addr_len) + - (2 * sizeof(u32))))) + if (!pskb_may_pull(skb, arp_hdr_len(skb->dev))) return NF_DROP; indev = in ? in->name : nulldevname; -- cgit v1.2.3 From c6aefafb7ec620911d46174eed514f9df639e5a4 Mon Sep 17 00:00:00 2001 From: Glenn Griffin Date: Thu, 7 Feb 2008 21:49:26 -0800 Subject: [TCP]: Add IPv6 support to TCP SYN cookies Updated to incorporate Eric's suggestion of using a per cpu buffer rather than allocating on the stack. Just a two line change, but will resend in it's entirety. Signed-off-by: Glenn Griffin Signed-off-by: YOSHIFUJI Hideaki --- include/net/tcp.h | 8 ++ net/ipv4/syncookies.c | 7 +- net/ipv4/tcp_input.c | 1 + net/ipv4/tcp_minisocks.c | 2 + net/ipv4/tcp_output.c | 1 + net/ipv6/Makefile | 1 + net/ipv6/syncookies.c | 267 +++++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/tcp_ipv6.c | 77 ++++++++++---- 8 files changed, 338 insertions(+), 26 deletions(-) create mode 100644 net/ipv6/syncookies.c (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index ae9774b478f5..11119e33acfe 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -434,11 +435,17 @@ extern int tcp_disconnect(struct sock *sk, int flags); extern void tcp_unhash(struct sock *sk); /* From syncookies.c */ +extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt); extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mss); +/* From net/ipv6/syncookies.c */ +extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb); +extern __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, + __u16 *mss); + /* tcp_output.c */ extern void __tcp_push_pending_frames(struct sock *sk, unsigned int cur_mss, @@ -1332,6 +1339,7 @@ extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo); extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo); extern struct request_sock_ops tcp_request_sock_ops; +extern struct request_sock_ops tcp6_request_sock_ops; extern int tcp_v4_destroy_sock(struct sock *sk); diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 177da14f6b14..4704f27f6c0b 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -10,8 +10,6 @@ * 2 of the License, or (at your option) any later version. * * $Id: syncookies.c,v 1.18 2002/02/01 22:01:04 davem Exp $ - * - * Missing: IPv6 support. */ #include @@ -23,14 +21,15 @@ extern int sysctl_tcp_syncookies; -static __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; +__u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; +EXPORT_SYMBOL(syncookie_secret); static __init int init_syncookies(void) { get_random_bytes(syncookie_secret, sizeof(syncookie_secret)); return 0; } -module_init(init_syncookies); +__initcall(init_syncookies); #define COOKIEBITS 24 /* Upper bits store count */ #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 19c449f62672..93e128c239c4 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5326,6 +5326,7 @@ discard: EXPORT_SYMBOL(sysctl_tcp_ecn); EXPORT_SYMBOL(sysctl_tcp_reordering); +EXPORT_SYMBOL(sysctl_tcp_adv_win_scale); EXPORT_SYMBOL(tcp_parse_options); EXPORT_SYMBOL(tcp_rcv_established); EXPORT_SYMBOL(tcp_rcv_state_process); diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 0fdd1db641ac..8245247a6ceb 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -35,6 +35,8 @@ #endif int sysctl_tcp_syncookies __read_mostly = SYNC_INIT; +EXPORT_SYMBOL(sysctl_tcp_syncookies); + int sysctl_tcp_abort_on_overflow __read_mostly; struct inet_timewait_death_row tcp_death_row = { diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index ed750f9ceb07..cbfef8b1f5e8 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2560,6 +2560,7 @@ void tcp_send_probe0(struct sock *sk) } } +EXPORT_SYMBOL(tcp_select_initial_window); EXPORT_SYMBOL(tcp_connect); EXPORT_SYMBOL(tcp_make_synack); EXPORT_SYMBOL(tcp_simple_retransmit); diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 24f3aa0f2a35..ae14617e607f 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -16,6 +16,7 @@ ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \ ipv6-$(CONFIG_NETFILTER) += netfilter.o ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o ipv6-$(CONFIG_PROC_FS) += proc.o +ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o ipv6-objs += $(ipv6-y) diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c new file mode 100644 index 000000000000..827c5aa7524c --- /dev/null +++ b/net/ipv6/syncookies.c @@ -0,0 +1,267 @@ +/* + * IPv6 Syncookies implementation for the Linux kernel + * + * Authors: + * Glenn Griffin + * + * Based on IPv4 implementation by Andi Kleen + * linux/net/ipv4/syncookies.c + * + * This program 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 of the License, or (at your option) any later version. + * + */ + +#include +#include +#include +#include +#include +#include + +extern int sysctl_tcp_syncookies; +extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; + +#define COOKIEBITS 24 /* Upper bits store count */ +#define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) + +/* + * This table has to be sorted and terminated with (__u16)-1. + * XXX generate a better table. + * Unresolved Issues: HIPPI with a 64k MSS is not well supported. + * + * Taken directly from ipv4 implementation. + * Should this list be modified for ipv6 use or is it close enough? + * rfc 2460 8.3 suggests mss values 20 bytes less than ipv4 counterpart + */ +static __u16 const msstab[] = { + 64 - 1, + 256 - 1, + 512 - 1, + 536 - 1, + 1024 - 1, + 1440 - 1, + 1460 - 1, + 4312 - 1, + (__u16)-1 +}; +/* The number doesn't include the -1 terminator */ +#define NUM_MSS (ARRAY_SIZE(msstab) - 1) + +/* + * This (misnamed) value is the age of syncookie which is permitted. + * Its ideal value should be dependent on TCP_TIMEOUT_INIT and + * sysctl_tcp_retries1. It's a rather complicated formula (exponential + * backoff) to compute at runtime so it's currently hardcoded here. + */ +#define COUNTER_TRIES 4 + +static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, + struct request_sock *req, + struct dst_entry *dst) +{ + struct inet_connection_sock *icsk = inet_csk(sk); + struct sock *child; + + child = icsk->icsk_af_ops->syn_recv_sock(sk, skb, req, dst); + if (child) + inet_csk_reqsk_queue_add(sk, req, child); + else + reqsk_free(req); + + return child; +} + +static DEFINE_PER_CPU(__u32, cookie_scratch)[16 + 5 + SHA_WORKSPACE_WORDS]; + +static u32 cookie_hash(struct in6_addr *saddr, struct in6_addr *daddr, + __be16 sport, __be16 dport, u32 count, int c) +{ + __u32 *tmp = __get_cpu_var(cookie_scratch); + + /* + * we have 320 bits of information to hash, copy in the remaining + * 192 bits required for sha_transform, from the syncookie_secret + * and overwrite the digest with the secret + */ + memcpy(tmp + 10, syncookie_secret[c], 44); + memcpy(tmp, saddr, 16); + memcpy(tmp + 4, daddr, 16); + tmp[8] = ((__force u32)sport << 16) + (__force u32)dport; + tmp[9] = count; + sha_transform(tmp + 16, (__u8 *)tmp, tmp + 16 + 5); + + return tmp[17]; +} + +static __u32 secure_tcp_syn_cookie(struct in6_addr *saddr, struct in6_addr *daddr, + __be16 sport, __be16 dport, __u32 sseq, + __u32 count, __u32 data) +{ + return (cookie_hash(saddr, daddr, sport, dport, 0, 0) + + sseq + (count << COOKIEBITS) + + ((cookie_hash(saddr, daddr, sport, dport, count, 1) + data) + & COOKIEMASK)); +} + +static __u32 check_tcp_syn_cookie(__u32 cookie, struct in6_addr *saddr, + struct in6_addr *daddr, __be16 sport, + __be16 dport, __u32 sseq, __u32 count, + __u32 maxdiff) +{ + __u32 diff; + + cookie -= cookie_hash(saddr, daddr, sport, dport, 0, 0) + sseq; + + diff = (count - (cookie >> COOKIEBITS)) & ((__u32) -1 >> COOKIEBITS); + if (diff >= maxdiff) + return (__u32)-1; + + return (cookie - + cookie_hash(saddr, daddr, sport, dport, count - diff, 1)) + & COOKIEMASK; +} + +__u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mssp) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + const struct tcphdr *th = tcp_hdr(skb); + int mssind; + const __u16 mss = *mssp; + + tcp_sk(sk)->last_synq_overflow = jiffies; + + for (mssind = 0; mss > msstab[mssind + 1]; mssind++) + ; + *mssp = msstab[mssind] + 1; + + NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESSENT); + + return secure_tcp_syn_cookie(&iph->saddr, &iph->daddr, th->source, + th->dest, ntohl(th->seq), + jiffies / (HZ * 60), mssind); +} + +static inline int cookie_check(struct sk_buff *skb, __u32 cookie) +{ + struct ipv6hdr *iph = ipv6_hdr(skb); + const struct tcphdr *th = tcp_hdr(skb); + __u32 seq = ntohl(th->seq) - 1; + __u32 mssind = check_tcp_syn_cookie(cookie, &iph->saddr, &iph->daddr, + th->source, th->dest, seq, + jiffies / (HZ * 60), COUNTER_TRIES); + + return mssind < NUM_MSS ? msstab[mssind] + 1 : 0; +} + +struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) +{ + struct inet_request_sock *ireq; + struct inet6_request_sock *ireq6; + struct tcp_request_sock *treq; + struct ipv6_pinfo *np = inet6_sk(sk); + struct tcp_sock *tp = tcp_sk(sk); + const struct tcphdr *th = tcp_hdr(skb); + __u32 cookie = ntohl(th->ack_seq) - 1; + struct sock *ret = sk; + struct request_sock *req; + int mss; + struct dst_entry *dst; + __u8 rcv_wscale; + + if (!sysctl_tcp_syncookies || !th->ack) + goto out; + + if (time_after(jiffies, tp->last_synq_overflow + TCP_TIMEOUT_INIT) || + (mss = cookie_check(skb, cookie)) == 0) { + NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESFAILED); + goto out; + } + + NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); + + ret = NULL; + req = inet6_reqsk_alloc(&tcp6_request_sock_ops); + if (!req) + goto out; + + ireq = inet_rsk(req); + ireq6 = inet6_rsk(req); + treq = tcp_rsk(req); + ireq6->pktopts = NULL; + + if (security_inet_conn_request(sk, skb, req)) { + reqsk_free(req); + goto out; + } + + req->mss = mss; + ireq->rmt_port = th->source; + ipv6_addr_copy(&ireq6->rmt_addr, &ipv6_hdr(skb)->saddr); + ipv6_addr_copy(&ireq6->loc_addr, &ipv6_hdr(skb)->daddr); + if (ipv6_opt_accepted(sk, skb) || + np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo || + np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) { + atomic_inc(&skb->users); + ireq6->pktopts = skb; + } + + ireq6->iif = sk->sk_bound_dev_if; + /* So that link locals have meaning */ + if (!sk->sk_bound_dev_if && + ipv6_addr_type(&ireq6->rmt_addr) & IPV6_ADDR_LINKLOCAL) + ireq6->iif = inet6_iif(skb); + + req->expires = 0UL; + req->retrans = 0; + ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0; + ireq->wscale_ok = ireq->sack_ok = 0; + treq->rcv_isn = ntohl(th->seq) - 1; + treq->snt_isn = cookie; + + /* + * We need to lookup the dst_entry to get the correct window size. + * This is taken from tcp_v6_syn_recv_sock. Somebody please enlighten + * me if there is a preferred way. + */ + { + struct in6_addr *final_p = NULL, final; + struct flowi fl; + memset(&fl, 0, sizeof(fl)); + fl.proto = IPPROTO_TCP; + ipv6_addr_copy(&fl.fl6_dst, &ireq6->rmt_addr); + if (np->opt && np->opt->srcrt) { + struct rt0_hdr *rt0 = (struct rt0_hdr *) np->opt->srcrt; + ipv6_addr_copy(&final, &fl.fl6_dst); + ipv6_addr_copy(&fl.fl6_dst, rt0->addr); + final_p = &final; + } + ipv6_addr_copy(&fl.fl6_src, &ireq6->loc_addr); + fl.oif = sk->sk_bound_dev_if; + fl.fl_ip_dport = inet_rsk(req)->rmt_port; + fl.fl_ip_sport = inet_sk(sk)->sport; + security_req_classify_flow(req, &fl); + if (ip6_dst_lookup(sk, &dst, &fl)) { + reqsk_free(req); + goto out; + } + if (final_p) + ipv6_addr_copy(&fl.fl6_dst, final_p); + if ((xfrm_lookup(&dst, &fl, sk, 0)) < 0) + goto out; + } + + req->window_clamp = dst_metric(dst, RTAX_WINDOW); + tcp_select_initial_window(tcp_full_space(sk), req->mss, + &req->rcv_wnd, &req->window_clamp, + 0, &rcv_wscale); + + ireq->rcv_wscale = rcv_wscale; + + ret = get_cookie_sock(sk, skb, req, dst); + +out: return ret; +} + diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 1cbbb87dbad2..fd773ac7531a 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -512,6 +512,20 @@ done: return err; } +static inline void syn_flood_warning(struct sk_buff *skb) +{ +#ifdef CONFIG_SYN_COOKIES + if (sysctl_tcp_syncookies) + printk(KERN_INFO + "TCPv6: Possible SYN flooding on port %d. " + "Sending cookies.\n", ntohs(tcp_hdr(skb)->dest)); + else +#endif + printk(KERN_INFO + "TCPv6: Possible SYN flooding on port %d. " + "Dropping request.\n", ntohs(tcp_hdr(skb)->dest)); +} + static void tcp_v6_reqsk_destructor(struct request_sock *req) { if (inet6_rsk(req)->pktopts) @@ -915,7 +929,7 @@ done_opts: } #endif -static struct request_sock_ops tcp6_request_sock_ops __read_mostly = { +struct request_sock_ops tcp6_request_sock_ops __read_mostly = { .family = AF_INET6, .obj_size = sizeof(struct tcp6_request_sock), .rtx_syn_ack = tcp_v6_send_synack, @@ -1213,9 +1227,9 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) return NULL; } -#if 0 /*def CONFIG_SYN_COOKIES*/ +#ifdef CONFIG_SYN_COOKIES if (!th->rst && !th->syn && th->ack) - sk = cookie_v6_check(sk, skb, &(IPCB(skb)->opt)); + sk = cookie_v6_check(sk, skb); #endif return sk; } @@ -1231,6 +1245,11 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) struct tcp_sock *tp = tcp_sk(sk); struct request_sock *req = NULL; __u32 isn = TCP_SKB_CB(skb)->when; +#ifdef CONFIG_SYN_COOKIES + int want_cookie = 0; +#else +#define want_cookie 0 +#endif if (skb->protocol == htons(ETH_P_IP)) return tcp_v4_conn_request(sk, skb); @@ -1238,12 +1257,14 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (!ipv6_unicast_destination(skb)) goto drop; - /* - * There are no SYN attacks on IPv6, yet... - */ if (inet_csk_reqsk_queue_is_full(sk) && !isn) { if (net_ratelimit()) - printk(KERN_INFO "TCPv6: dropping request, synflood is possible\n"); + syn_flood_warning(skb); +#ifdef CONFIG_SYN_COOKIES + if (sysctl_tcp_syncookies) + want_cookie = 1; + else +#endif goto drop; } @@ -1264,29 +1285,39 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) tcp_parse_options(skb, &tmp_opt, 0); + if (want_cookie) { + tcp_clear_options(&tmp_opt); + tmp_opt.saw_tstamp = 0; + } + tmp_opt.tstamp_ok = tmp_opt.saw_tstamp; tcp_openreq_init(req, &tmp_opt, skb); treq = inet6_rsk(req); ipv6_addr_copy(&treq->rmt_addr, &ipv6_hdr(skb)->saddr); ipv6_addr_copy(&treq->loc_addr, &ipv6_hdr(skb)->daddr); - TCP_ECN_create_request(req, tcp_hdr(skb)); treq->pktopts = NULL; - if (ipv6_opt_accepted(sk, skb) || - np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo || - np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) { - atomic_inc(&skb->users); - treq->pktopts = skb; - } - treq->iif = sk->sk_bound_dev_if; + if (!want_cookie) + TCP_ECN_create_request(req, tcp_hdr(skb)); + + if (want_cookie) { + isn = cookie_v6_init_sequence(sk, skb, &req->mss); + } else if (!isn) { + if (ipv6_opt_accepted(sk, skb) || + np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo || + np->rxopt.bits.rxhlim || np->rxopt.bits.rxohlim) { + atomic_inc(&skb->users); + treq->pktopts = skb; + } + treq->iif = sk->sk_bound_dev_if; - /* So that link locals have meaning */ - if (!sk->sk_bound_dev_if && - ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL) - treq->iif = inet6_iif(skb); + /* So that link locals have meaning */ + if (!sk->sk_bound_dev_if && + ipv6_addr_type(&treq->rmt_addr) & IPV6_ADDR_LINKLOCAL) + treq->iif = inet6_iif(skb); - if (isn == 0) isn = tcp_v6_init_sequence(skb); + } tcp_rsk(req)->snt_isn = isn; @@ -1295,8 +1326,10 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (tcp_v6_send_synack(sk, req)) goto drop; - inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT); - return 0; + if (!want_cookie) { + inet6_csk_reqsk_queue_hash_add(sk, req, TCP_TIMEOUT_INIT); + return 0; + } drop: if (req) -- cgit v1.2.3 From e898d4db2749c6052072e9bc4448e396cbdeb06a Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sat, 1 Mar 2008 01:06:47 +0900 Subject: [UDP]: Allow users to configure UDP-Lite. Let's give users an option for disabling UDP-Lite (~4K). old: | text data bss dec hex filename | 286498 12432 6072 305002 4a76a net/ipv4/built-in.o | 193830 8192 3204 205226 321aa net/ipv6/ipv6.o new (without UDP-Lite): | text data bss dec hex filename | 284086 12136 5432 301654 49a56 net/ipv4/built-in.o | 191835 7832 3076 202743 317f7 net/ipv6/ipv6.o Signed-off-by: YOSHIFUJI Hideaki --- include/linux/udp.h | 11 +++++++++++ include/net/ipv6.h | 5 +++++ include/net/transp_v6.h | 5 +++++ include/net/udplite.h | 9 +++++++-- net/ipv4/Kconfig | 10 ++++++++++ net/ipv4/Makefile | 3 ++- net/ipv4/af_inet.c | 7 ++++++- net/ipv4/proc.c | 5 ++++- net/ipv4/udp.c | 24 ++++++++++++++---------- net/ipv6/Makefile | 3 ++- net/ipv6/af_inet6.c | 14 ++++++++++++++ net/ipv6/ipv6_sockglue.c | 6 +++++- net/ipv6/proc.c | 6 ++++++ net/ipv6/udp.c | 16 ++++++++-------- 14 files changed, 99 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/include/linux/udp.h b/include/linux/udp.h index 8ec703f462da..4144664d69d9 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -70,8 +70,10 @@ struct udp_sock { #define UDPLITE_BIT 0x1 /* set by udplite proto init function */ #define UDPLITE_SEND_CC 0x2 /* set via udplite setsockopt */ #define UDPLITE_RECV_CC 0x4 /* set via udplite setsocktopt */ +#ifdef CONFIG_IP_UDPLITE __u8 pcflag; /* marks socket as UDP-Lite if > 0 */ __u8 unused[3]; +#endif /* * For encapsulation sockets. */ @@ -82,7 +84,16 @@ static inline struct udp_sock *udp_sk(const struct sock *sk) { return (struct udp_sock *)sk; } + +#ifdef CONFIG_IP_UDPLITE #define IS_UDPLITE(__sk) (udp_sk(__sk)->pcflag) +#define IS_PROTO_UDPLITE(__proto) ((__proto) == IPPROTO_UDPLITE) +#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP || (level) == SOL_UDPLITE) +#else +#define IS_UDPLITE(__sk) 0 +#define IS_PROTO_UDPLITE(__proto) 0 +#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP) +#endif #endif diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 8b05c65415cb..96b1763bfcaa 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -603,8 +603,13 @@ extern int tcp6_proc_init(void); extern void tcp6_proc_exit(void); extern int udp6_proc_init(void); extern void udp6_proc_exit(void); +#ifdef CONFIG_IP_UDPLITE extern int udplite6_proc_init(void); extern void udplite6_proc_exit(void); +#else +static inline int udplite6_proc_init(void) { return 0; } +static inline void udplite6_proc_exit(void) { } +#endif extern int ipv6_misc_proc_init(void); extern void ipv6_misc_proc_exit(void); extern int snmp6_register_dev(struct inet6_dev *idev); diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h index 27394e0447d8..902e6c6bc793 100644 --- a/include/net/transp_v6.h +++ b/include/net/transp_v6.h @@ -27,8 +27,13 @@ extern int rawv6_init(void); extern void rawv6_exit(void); extern int udpv6_init(void); extern void udpv6_exit(void); +#ifdef CONFIG_IP_UDPLITE extern int udplitev6_init(void); extern void udplitev6_exit(void); +#else +static inline int udplitev6_init(void) { return 0; } +static inline void udplitev6_exit(void) { } +#endif extern int tcpv6_init(void); extern void tcpv6_exit(void); diff --git a/include/net/udplite.h b/include/net/udplite.h index b76b2e377af4..01ddb2c20264 100644 --- a/include/net/udplite.h +++ b/include/net/udplite.h @@ -25,7 +25,9 @@ static __inline__ int udplite_getfrag(void *from, char *to, int offset, /* Designate sk as UDP-Lite socket */ static inline int udplite_sk_init(struct sock *sk) { +#ifdef CONFIG_IP_UDPLITE udp_sk(sk)->pcflag = UDPLITE_BIT; +#endif return 0; } @@ -69,7 +71,7 @@ static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh) static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) { int cscov = up->len; - +#ifdef CONFIG_IP_UDPLITE /* * Sender has set `partial coverage' option on UDP-Lite socket */ @@ -93,13 +95,15 @@ static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) * illegal, we fall back to the defaults here. */ } +#endif return cscov; } static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) { - int cscov = udplite_sender_cscov(udp_sk(sk), udp_hdr(skb)); __wsum csum = 0; +#ifdef CONFIG_IP_UDPLITE + int cscov = udplite_sender_cscov(udp_sk(sk), udp_hdr(skb)); skb->ip_summed = CHECKSUM_NONE; /* no HW support for checksumming */ @@ -112,6 +116,7 @@ static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) if ((cscov -= len) <= 0) break; } +#endif return csum; } diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 19880b086e71..efe3832c4ad8 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -632,5 +632,15 @@ config TCP_MD5SIG If unsure, say N. +config IP_UDPLITE + bool "IP: UDP-Lite Protocol (RFC 3828)" + default n + ---help--- + UDP-Lite (RFC 3828) is a UDP-like protocol with variable-length + checksum. Read for + details. + + If unsure, say N. + source "net/ipv4/ipvs/Kconfig" diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index ad40ef3f9ebc..e88cebdf3e30 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -8,7 +8,7 @@ obj-y := route.o inetpeer.o protocol.o \ inet_timewait_sock.o inet_connection_sock.o \ tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \ tcp_minisocks.o tcp_cong.o \ - datagram.o raw.o udp.o udplite.o \ + datagram.o raw.o udp.o \ arp.o icmp.o devinet.o af_inet.o igmp.o \ fib_frontend.o fib_semantics.o \ inet_fragment.o @@ -49,6 +49,7 @@ obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o +obj-$(CONFIG_IP_UDPLITE) += udplite.o obj-$(CONFIG_NETLABEL) += cipso_ipv4.o obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 4f539bd48718..67260c0eaaa8 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1317,15 +1317,18 @@ static int __init init_ipv4_mibs(void) if (snmp_mib_init((void **)udp_statistics, sizeof(struct udp_mib)) < 0) goto err_udp_mib; +#ifdef CONFIG_IP_UDPLITE if (snmp_mib_init((void **)udplite_statistics, sizeof(struct udp_mib)) < 0) goto err_udplite_mib; - +#endif tcp_mib_init(); return 0; +#ifdef CONFIG_IP_UDPLITE err_udplite_mib: +#endif snmp_mib_free((void **)udp_statistics); err_udp_mib: snmp_mib_free((void **)tcp_statistics); @@ -1423,8 +1426,10 @@ static int __init inet_init(void) /* Setup UDP memory threshold */ udp_init(); +#ifdef CONFIG_IP_UDPLITE /* Add UDP-Lite (RFC 3828) */ udplite4_register(); +#endif /* * Set the ICMP layer up diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index d63474c6b400..d75ddb7fa4b8 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -59,7 +59,9 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) atomic_read(&tcp_memory_allocated)); seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot), atomic_read(&udp_memory_allocated)); +#ifdef CONFIG_IP_UDPLITE seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot)); +#endif seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot)); seq_printf(seq, "FRAG: inuse %d memory %d\n", ip_frag_nqueues(&init_net), ip_frag_mem(&init_net)); @@ -349,6 +351,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) snmp_fold_field((void **)udp_statistics, snmp4_udp_list[i].entry)); +#ifdef CONFIG_IP_UDPLITE /* the UDP and UDP-Lite MIBs are the same */ seq_puts(seq, "\nUdpLite:"); for (i = 0; snmp4_udp_list[i].name != NULL; i++) @@ -359,7 +362,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) seq_printf(seq, " %lu", snmp_fold_field((void **)udplite_statistics, snmp4_udp_list[i].entry)); - +#endif seq_putc(seq, '\n'); return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7ea1b67b6de1..acc353aa89eb 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1127,7 +1127,7 @@ static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh, UDP_SKB_CB(skb)->partial_cov = 0; UDP_SKB_CB(skb)->cscov = skb->len; - if (proto == IPPROTO_UDPLITE) { + if (IS_PROTO_UDPLITE(proto)) { err = udplite_checksum_init(skb, uh); if (err) return err; @@ -1175,7 +1175,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], if (ulen > skb->len) goto short_packet; - if (proto == IPPROTO_UDP) { + if (IS_PROTO_UDPLITE(proto)) { /* UDP validates ulen. */ if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen)) goto short_packet; @@ -1217,7 +1217,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], if (udp_lib_checksum_complete(skb)) goto csum_error; - UDP_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); + UDP_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto)); icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); /* @@ -1229,7 +1229,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], short_packet: LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", - proto == IPPROTO_UDPLITE ? "-Lite" : "", + IS_PROTO_UDPLITE(proto) ? "-Lite" : "", NIPQUAD(saddr), ntohs(uh->source), ulen, @@ -1244,14 +1244,14 @@ csum_error: * the network is concerned, anyway) as per 4.1.3.4 (MUST). */ LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", - proto == IPPROTO_UDPLITE ? "-Lite" : "", + IS_PROTO_UDPLITE(proto) ? "-Lite" : "", NIPQUAD(saddr), ntohs(uh->source), NIPQUAD(daddr), ntohs(uh->dest), ulen); drop: - UDP_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE); + UDP_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto)); kfree_skb(skb); return 0; } @@ -1279,7 +1279,9 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, struct udp_sock *up = udp_sk(sk); int val; int err = 0; +#ifdef CONFIG_IP_UDPLITE int is_udplite = IS_UDPLITE(sk); +#endif if (optlenpcrlen = val; up->pcflag |= UDPLITE_RECV_CC; break; +#endif default: err = -ENOPROTOOPT; @@ -1352,7 +1356,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, int udp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_setsockopt(sk, level, optname, optval, optlen, udp_push_pending_frames); return ip_setsockopt(sk, level, optname, optval, optlen); @@ -1362,7 +1366,7 @@ int udp_setsockopt(struct sock *sk, int level, int optname, int compat_udp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_setsockopt(sk, level, optname, optval, optlen, udp_push_pending_frames); return compat_ip_setsockopt(sk, level, optname, optval, optlen); @@ -1416,7 +1420,7 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname, int udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_getsockopt(sk, level, optname, optval, optlen); return ip_getsockopt(sk, level, optname, optval, optlen); } @@ -1425,7 +1429,7 @@ int udp_getsockopt(struct sock *sk, int level, int optname, int compat_udp_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_getsockopt(sk, level, optname, optval, optlen); return compat_ip_getsockopt(sk, level, optname, optval, optlen); } diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index ae14617e607f..81969479955f 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_IPV6) += ipv6.o ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ addrlabel.o \ - route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \ + route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o \ raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o @@ -17,6 +17,7 @@ ipv6-$(CONFIG_NETFILTER) += netfilter.o ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o ipv6-$(CONFIG_PROC_FS) += proc.o ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o +ipv6-$(CONFIG_IP_UDPLITE) += udplite.o ipv6-objs += $(ipv6-y) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 9869f87243cf..243c42a6b80d 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -691,12 +691,16 @@ static int __init init_ipv6_mibs(void) goto err_icmpmsg_mib; if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib)) < 0) goto err_udp_mib; +#ifdef CONFIG_IP_UDPLITE if (snmp_mib_init((void **)udplite_stats_in6, sizeof (struct udp_mib)) < 0) goto err_udplite_mib; +#endif return 0; +#ifdef CONFIG_IP_UDPLITE err_udplite_mib: +#endif snmp_mib_free((void **)udp_stats_in6); err_udp_mib: snmp_mib_free((void **)icmpv6msg_statistics); @@ -715,7 +719,9 @@ static void cleanup_ipv6_mibs(void) snmp_mib_free((void **)icmpv6_statistics); snmp_mib_free((void **)icmpv6msg_statistics); snmp_mib_free((void **)udp_stats_in6); +#ifdef CONFIG_IP_UDPLITE snmp_mib_free((void **)udplite_stats_in6); +#endif } static int inet6_net_init(struct net *net) @@ -760,9 +766,11 @@ static int __init inet6_init(void) if (err) goto out_unregister_tcp_proto; +#ifdef CONFIG_IP_UDPLITE err = proto_register(&udplitev6_prot, 1); if (err) goto out_unregister_udp_proto; +#endif err = proto_register(&rawv6_prot, 1); if (err) @@ -933,8 +941,10 @@ out_sock_register_fail: out_unregister_raw_proto: proto_unregister(&rawv6_prot); out_unregister_udplite_proto: +#ifdef CONFIG_IP_UDPLITE proto_unregister(&udplitev6_prot); out_unregister_udp_proto: +#endif proto_unregister(&udpv6_prot); out_unregister_tcp_proto: proto_unregister(&tcpv6_prot); @@ -950,7 +960,9 @@ static void __exit inet6_exit(void) rtnl_unregister_all(PF_INET6); udpv6_exit(); +#ifdef CONFIG_IP_UDPLITE udplitev6_exit(); +#endif tcpv6_exit(); /* Cleanup code parts. */ @@ -982,7 +994,9 @@ static void __exit inet6_exit(void) unregister_pernet_subsys(&inet6_net_ops); cleanup_ipv6_mibs(); proto_unregister(&rawv6_prot); +#ifdef CONFIG_IP_UDPLITE proto_unregister(&udplitev6_prot); +#endif proto_unregister(&udpv6_prot); proto_unregister(&tcpv6_prot); } diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index bf2a686aa13d..0a18fecb93d1 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -239,7 +239,9 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, struct sk_buff *pktopt; if (sk->sk_protocol != IPPROTO_UDP && +#ifdef CONFIG_IP_UDPLITE sk->sk_protocol != IPPROTO_UDPLITE && +#endif sk->sk_protocol != IPPROTO_TCP) break; @@ -279,7 +281,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, } else { struct proto *prot = &udp_prot; - if (sk->sk_protocol == IPPROTO_UDPLITE) + if (IS_PROTO_UDPLITE(sk->sk_protocol)) prot = &udplite_prot; local_bh_disable(); sock_prot_inuse_add(sk->sk_prot, -1); @@ -844,7 +846,9 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case IPV6_ADDRFORM: if (sk->sk_protocol != IPPROTO_UDP && +#ifdef CONFIG_IP_UDPLITE sk->sk_protocol != IPPROTO_UDPLITE && +#endif sk->sk_protocol != IPPROTO_TCP) return -EINVAL; if (sk->sk_state != TCP_ESTABLISHED) diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 199ef379e501..5ba7ae849d04 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -39,8 +39,10 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v) sock_prot_inuse_get(&tcpv6_prot)); seq_printf(seq, "UDP6: inuse %d\n", sock_prot_inuse_get(&udpv6_prot)); +#ifdef CONFIG_IP_UDPLITE seq_printf(seq, "UDPLITE6: inuse %d\n", sock_prot_inuse_get(&udplitev6_prot)); +#endif seq_printf(seq, "RAW6: inuse %d\n", sock_prot_inuse_get(&rawv6_prot)); seq_printf(seq, "FRAG6: inuse %d memory %d\n", @@ -111,6 +113,7 @@ static struct snmp_mib snmp6_udp6_list[] = { SNMP_MIB_SENTINEL }; +#ifdef CONFIG_IP_UDPLITE static struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS), SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS), @@ -118,6 +121,7 @@ static struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS), SNMP_MIB_SENTINEL }; +#endif static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) { @@ -176,7 +180,9 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) snmp6_seq_show_item(seq, (void **)icmpv6_statistics, snmp6_icmp6_list); snmp6_seq_show_icmpv6msg(seq, (void **)icmpv6msg_statistics); snmp6_seq_show_item(seq, (void **)udp_stats_in6, snmp6_udp6_list); +#ifdef CONFIG_IP_UDPLITE snmp6_seq_show_item(seq, (void **)udplite_stats_in6, snmp6_udplite6_list); +#endif } return 0; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 53739de829db..55feac7ba717 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -400,7 +400,7 @@ static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, UDP_SKB_CB(skb)->partial_cov = 0; UDP_SKB_CB(skb)->cscov = skb->len; - if (proto == IPPROTO_UDPLITE) { + if (IS_PROTO_UDPLITE(proto)) { err = udplite_checksum_init(skb, uh); if (err) return err; @@ -489,7 +489,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], if (udp_lib_checksum_complete(skb)) goto discard; - UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); + UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto)); icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev); @@ -510,11 +510,11 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], short_packet: LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n", - proto == IPPROTO_UDPLITE ? "-Lite" : "", + IS_PROTO_UDPLITE(proto) ? "-Lite" : "", ulen, skb->len); discard: - UDP6_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE); + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto)); kfree_skb(skb); return 0; } @@ -890,7 +890,7 @@ int udpv6_destroy_sock(struct sock *sk) int udpv6_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_setsockopt(sk, level, optname, optval, optlen, udp_v6_push_pending_frames); return ipv6_setsockopt(sk, level, optname, optval, optlen); @@ -900,7 +900,7 @@ int udpv6_setsockopt(struct sock *sk, int level, int optname, int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_setsockopt(sk, level, optname, optval, optlen, udp_v6_push_pending_frames); return compat_ipv6_setsockopt(sk, level, optname, optval, optlen); @@ -910,7 +910,7 @@ int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, int udpv6_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_getsockopt(sk, level, optname, optval, optlen); return ipv6_getsockopt(sk, level, optname, optval, optlen); } @@ -919,7 +919,7 @@ int udpv6_getsockopt(struct sock *sk, int level, int optname, int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { - if (level == SOL_UDP || level == SOL_UDPLITE) + if (IS_SOL_UDPFAMILY(level)) return udp_lib_getsockopt(sk, level, optname, optval, optlen); return compat_ipv6_getsockopt(sk, level, optname, optval, optlen); } -- cgit v1.2.3 From 662397fd7aaa10afdbdc55a0bfdb7e9701454c27 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 27 Feb 2008 23:14:03 +0900 Subject: [IPV6]: Move packet_type{} related bits to af_inet6.c. Signed-off-by: YOSHIFUJI Hideaki --- include/net/ipv6.h | 4 -- net/ipv6/af_inet6.c | 123 +++++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/ipv6_sockglue.c | 122 ---------------------------------------------- 3 files changed, 123 insertions(+), 126 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 96b1763bfcaa..5dc8164e5d3b 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -555,10 +555,6 @@ extern int compat_ipv6_getsockopt(struct sock *sk, char __user *optval, int __user *optlen); -extern int ipv6_packet_init(void); - -extern void ipv6_packet_cleanup(void); - extern int ip6_datagram_connect(struct sock *sk, struct sockaddr *addr, int addr_len); diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 243c42a6b80d..73021d5baece 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -678,6 +678,129 @@ int ipv6_opt_accepted(struct sock *sk, struct sk_buff *skb) EXPORT_SYMBOL_GPL(ipv6_opt_accepted); +static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb, + int proto) +{ + struct inet6_protocol *ops = NULL; + + for (;;) { + struct ipv6_opt_hdr *opth; + int len; + + if (proto != NEXTHDR_HOP) { + ops = rcu_dereference(inet6_protos[proto]); + + if (unlikely(!ops)) + break; + + if (!(ops->flags & INET6_PROTO_GSO_EXTHDR)) + break; + } + + if (unlikely(!pskb_may_pull(skb, 8))) + break; + + opth = (void *)skb->data; + len = ipv6_optlen(opth); + + if (unlikely(!pskb_may_pull(skb, len))) + break; + + proto = opth->nexthdr; + __skb_pull(skb, len); + } + + return ops; +} + +static int ipv6_gso_send_check(struct sk_buff *skb) +{ + struct ipv6hdr *ipv6h; + struct inet6_protocol *ops; + int err = -EINVAL; + + if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h)))) + goto out; + + ipv6h = ipv6_hdr(skb); + __skb_pull(skb, sizeof(*ipv6h)); + err = -EPROTONOSUPPORT; + + rcu_read_lock(); + ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr); + if (likely(ops && ops->gso_send_check)) { + skb_reset_transport_header(skb); + err = ops->gso_send_check(skb); + } + rcu_read_unlock(); + +out: + return err; +} + +static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + struct ipv6hdr *ipv6h; + struct inet6_protocol *ops; + + if (!(features & NETIF_F_V6_CSUM)) + features &= ~NETIF_F_SG; + + if (unlikely(skb_shinfo(skb)->gso_type & + ~(SKB_GSO_UDP | + SKB_GSO_DODGY | + SKB_GSO_TCP_ECN | + SKB_GSO_TCPV6 | + 0))) + goto out; + + if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h)))) + goto out; + + ipv6h = ipv6_hdr(skb); + __skb_pull(skb, sizeof(*ipv6h)); + segs = ERR_PTR(-EPROTONOSUPPORT); + + rcu_read_lock(); + ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr); + if (likely(ops && ops->gso_segment)) { + skb_reset_transport_header(skb); + segs = ops->gso_segment(skb, features); + } + rcu_read_unlock(); + + if (unlikely(IS_ERR(segs))) + goto out; + + for (skb = segs; skb; skb = skb->next) { + ipv6h = ipv6_hdr(skb); + ipv6h->payload_len = htons(skb->len - skb->mac_len - + sizeof(*ipv6h)); + } + +out: + return segs; +} + +static struct packet_type ipv6_packet_type = { + .type = __constant_htons(ETH_P_IPV6), + .func = ipv6_rcv, + .gso_send_check = ipv6_gso_send_check, + .gso_segment = ipv6_gso_segment, +}; + +static int __init ipv6_packet_init(void) +{ + dev_add_pack(&ipv6_packet_type); + return 0; +} + +static void ipv6_packet_cleanup(void) +{ + dev_remove_pack(&ipv6_packet_type); +} + static int __init init_ipv6_mibs(void) { if (snmp_mib_init((void **)ipv6_statistics, diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 0a18fecb93d1..3bbfdff698d2 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -57,118 +57,6 @@ DEFINE_SNMP_STAT(struct ipstats_mib, ipv6_statistics) __read_mostly; -static struct inet6_protocol *ipv6_gso_pull_exthdrs(struct sk_buff *skb, - int proto) -{ - struct inet6_protocol *ops = NULL; - - for (;;) { - struct ipv6_opt_hdr *opth; - int len; - - if (proto != NEXTHDR_HOP) { - ops = rcu_dereference(inet6_protos[proto]); - - if (unlikely(!ops)) - break; - - if (!(ops->flags & INET6_PROTO_GSO_EXTHDR)) - break; - } - - if (unlikely(!pskb_may_pull(skb, 8))) - break; - - opth = (void *)skb->data; - len = opth->hdrlen * 8 + 8; - - if (unlikely(!pskb_may_pull(skb, len))) - break; - - proto = opth->nexthdr; - __skb_pull(skb, len); - } - - return ops; -} - -static int ipv6_gso_send_check(struct sk_buff *skb) -{ - struct ipv6hdr *ipv6h; - struct inet6_protocol *ops; - int err = -EINVAL; - - if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h)))) - goto out; - - ipv6h = ipv6_hdr(skb); - __skb_pull(skb, sizeof(*ipv6h)); - err = -EPROTONOSUPPORT; - - rcu_read_lock(); - ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr); - if (likely(ops && ops->gso_send_check)) { - skb_reset_transport_header(skb); - err = ops->gso_send_check(skb); - } - rcu_read_unlock(); - -out: - return err; -} - -static struct sk_buff *ipv6_gso_segment(struct sk_buff *skb, int features) -{ - struct sk_buff *segs = ERR_PTR(-EINVAL); - struct ipv6hdr *ipv6h; - struct inet6_protocol *ops; - - if (!(features & NETIF_F_V6_CSUM)) - features &= ~NETIF_F_SG; - - if (unlikely(skb_shinfo(skb)->gso_type & - ~(SKB_GSO_UDP | - SKB_GSO_DODGY | - SKB_GSO_TCP_ECN | - SKB_GSO_TCPV6 | - 0))) - goto out; - - if (unlikely(!pskb_may_pull(skb, sizeof(*ipv6h)))) - goto out; - - ipv6h = ipv6_hdr(skb); - __skb_pull(skb, sizeof(*ipv6h)); - segs = ERR_PTR(-EPROTONOSUPPORT); - - rcu_read_lock(); - ops = ipv6_gso_pull_exthdrs(skb, ipv6h->nexthdr); - if (likely(ops && ops->gso_segment)) { - skb_reset_transport_header(skb); - segs = ops->gso_segment(skb, features); - } - rcu_read_unlock(); - - if (unlikely(IS_ERR(segs))) - goto out; - - for (skb = segs; skb; skb = skb->next) { - ipv6h = ipv6_hdr(skb); - ipv6h->payload_len = htons(skb->len - skb->mac_len - - sizeof(*ipv6h)); - } - -out: - return segs; -} - -static struct packet_type ipv6_packet_type = { - .type = __constant_htons(ETH_P_IPV6), - .func = ipv6_rcv, - .gso_send_check = ipv6_gso_send_check, - .gso_segment = ipv6_gso_segment, -}; - struct ip6_ra_chain *ip6_ra_chain; DEFINE_RWLOCK(ip6_ra_lock); @@ -1132,13 +1020,3 @@ int compat_ipv6_getsockopt(struct sock *sk, int level, int optname, EXPORT_SYMBOL(compat_ipv6_getsockopt); #endif -int __init ipv6_packet_init(void) -{ - dev_add_pack(&ipv6_packet_type); - return 0; -} - -void ipv6_packet_cleanup(void) -{ - dev_remove_pack(&ipv6_packet_type); -} -- cgit v1.2.3 From 0e7b8dcd16eb91b9cd8ecc07c4094512f20d7e3c Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 28 Feb 2008 17:03:12 +0900 Subject: [IPV6]: Use htonl() instead of __constant_htonl() where appricable. Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 496503c03846..a9ff97c120c7 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -214,29 +214,25 @@ static inline void addrconf_addr_solict_mult(const struct in6_addr *addr, struct in6_addr *solicited) { ipv6_addr_set(solicited, - __constant_htonl(0xFF020000), 0, - __constant_htonl(0x1), - __constant_htonl(0xFF000000) | addr->s6_addr32[3]); + htonl(0xFF020000), 0, + htonl(0x1), + htonl(0xFF000000) | addr->s6_addr32[3]); } static inline void ipv6_addr_all_nodes(struct in6_addr *addr) { - ipv6_addr_set(addr, - __constant_htonl(0xFF020000), 0, 0, - __constant_htonl(0x1)); + ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x1)); } static inline void ipv6_addr_all_routers(struct in6_addr *addr) { - ipv6_addr_set(addr, - __constant_htonl(0xFF020000), 0, 0, - __constant_htonl(0x2)); + ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x2)); } static inline int ipv6_addr_is_multicast(const struct in6_addr *addr) { - return (addr->s6_addr32[0] & __constant_htonl(0xFF000000)) == __constant_htonl(0xFF000000); + return (addr->s6_addr32[0] & htonl(0xFF000000)) == htonl(0xFF000000); } static inline int ipv6_addr_is_ll_all_nodes(const struct in6_addr *addr) -- cgit v1.2.3 From 8082c37cdc31fb0ed178d9d706bf7568ada0edd9 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 4 Mar 2008 14:55:03 +0900 Subject: [NET] NEIGHBOUR: Remove unpopular neigh_is_connected(). neigh_is_connected() is not popular at all, and the only user drivers/net/cxgb3/l2t.c:t3_l2t_update() also have raw (expanded) expression. Let's expand it and remove the inline function. Signed-off-by: YOSHIFUJI Hideaki --- drivers/net/cxgb3/l2t.c | 2 +- include/net/neighbour.h | 6 ------ 2 files changed, 1 insertion(+), 7 deletions(-) (limited to 'include') diff --git a/drivers/net/cxgb3/l2t.c b/drivers/net/cxgb3/l2t.c index 865faee53e17..f510140885ae 100644 --- a/drivers/net/cxgb3/l2t.c +++ b/drivers/net/cxgb3/l2t.c @@ -407,7 +407,7 @@ found: } else if (neigh->nud_state & (NUD_CONNECTED|NUD_STALE)) setup_l2e_send_pending(dev, NULL, e); } else { - e->state = neigh_is_connected(neigh) ? + e->state = neigh->nud_state & NUD_CONNECTED ? L2T_STATE_VALID : L2T_STATE_STALE; if (memcmp(e->dmac, neigh->ha, 6)) setup_l2e_send_pending(dev, NULL, e); diff --git a/include/net/neighbour.h b/include/net/neighbour.h index ebbfb509822e..062281872064 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -284,12 +284,6 @@ static inline void neigh_confirm(struct neighbour *neigh) neigh->confirmed = jiffies; } -static inline int neigh_is_connected(struct neighbour *neigh) -{ - return neigh->nud_state&NUD_CONNECTED; -} - - static inline int neigh_event_send(struct neighbour *neigh, struct sk_buff *skb) { neigh->used = jiffies; -- cgit v1.2.3 From 5e5f3f0f801321078c897a5de0b4b4304f234da0 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 3 Mar 2008 21:44:34 +0900 Subject: [IPV6] ADDRCONF: Convert ipv6_get_saddr() to ipv6_dev_get_saddr(). Since most users of ipv6_get_saddr() pass non-NULL as dst argument, use ipv6_dev_get_saddr() directly. Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 3 --- net/ipv6/addrconf.c | 9 +-------- net/ipv6/fib6_rules.c | 4 ++-- net/ipv6/ip6_output.c | 3 ++- net/ipv6/route.c | 3 ++- net/ipv6/xfrm6_policy.c | 5 +++-- net/sctp/ipv6.c | 3 ++- 7 files changed, 12 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index a9ff97c120c7..89e3c53c8886 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -73,9 +73,6 @@ extern struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct net_device *dev, int strict); -extern int ipv6_get_saddr(struct dst_entry *dst, - struct in6_addr *daddr, - struct in6_addr *saddr); extern int ipv6_dev_get_saddr(struct net_device *dev, struct in6_addr *daddr, struct in6_addr *saddr); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 18e3a9825d81..9b3a2d0e4269 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1167,14 +1167,7 @@ record_it: return 0; } - -int ipv6_get_saddr(struct dst_entry *dst, - struct in6_addr *daddr, struct in6_addr *saddr) -{ - return ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, daddr, saddr); -} - -EXPORT_SYMBOL(ipv6_get_saddr); +EXPORT_SYMBOL(ipv6_dev_get_saddr); int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr, unsigned char banned_flags) diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 695c0ca8a417..157db3a1ce00 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -85,8 +85,8 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, if ((rule->flags & FIB_RULE_FIND_SADDR) && r->src.plen && !(flags & RT6_LOOKUP_F_HAS_SADDR)) { struct in6_addr saddr; - if (ipv6_get_saddr(&rt->u.dst, &flp->fl6_dst, - &saddr)) + if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, + &flp->fl6_dst, &saddr)) goto again; if (!ipv6_prefix_equal(&saddr, &r->src.addr, r->src.plen)) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a59d259cf97e..ff3971173e1e 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -919,7 +919,8 @@ static int ip6_dst_lookup_tail(struct sock *sk, goto out_err_release; if (ipv6_addr_any(&fl->fl6_src)) { - err = ipv6_get_saddr(*dst, &fl->fl6_dst, &fl->fl6_src); + err = ipv6_dev_get_saddr(ip6_dst_idev(*dst)->dev, + &fl->fl6_dst, &fl->fl6_src); if (err) goto out_err_release; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 1b15e1708409..6abe7da45ef7 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2122,7 +2122,8 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, NLA_PUT_U32(skb, RTA_IIF, iif); else if (dst) { struct in6_addr saddr_buf; - if (ipv6_get_saddr(&rt->u.dst, dst, &saddr_buf) == 0) + if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, + dst, &saddr_buf) == 0) NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); } diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 7d20199ee1f3..6ef56303e69e 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -57,8 +57,9 @@ static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr) if (IS_ERR(dst)) return -EHOSTUNREACH; - ipv6_get_saddr(dst, (struct in6_addr *)&daddr->a6, - (struct in6_addr *)&saddr->a6); + ipv6_dev_get_saddr(ip6_dst_idev(dst)->dev, + (struct in6_addr *)&daddr->a6, + (struct in6_addr *)&saddr->a6); dst_release(dst); return 0; } diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 87f940587d5f..3e4878800b36 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -313,7 +313,8 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc, __FUNCTION__, asoc, dst, NIP6(daddr->v6.sin6_addr)); if (!asoc) { - ipv6_get_saddr(dst, &daddr->v6.sin6_addr,&saddr->v6.sin6_addr); + ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, + &daddr->v6.sin6_addr, &saddr->v6.sin6_addr); SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " NIP6_FMT "\n", NIP6(saddr->v6.sin6_addr)); return; -- cgit v1.2.3 From 95e41e93e18d8e1e272ce23d96bae4f17ce11d42 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 6 Dec 2007 15:43:30 -0800 Subject: [IPV6]: Make ndisc_flow_init() common for later use. For later use, this patch is renaming ndisc_flow_init() to icmpv6_flow_init() and putting it in common place. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/icmpv6.h | 8 ++++++++ net/ipv6/icmp.c | 16 ++++++++++++++++ net/ipv6/ndisc.c | 23 ++++------------------- 3 files changed, 28 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 8f86d6b621c8..e4d4300d768f 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -182,6 +182,14 @@ extern int icmpv6_err_convert(int type, int code, extern void icmpv6_cleanup(void); extern void icmpv6_param_prob(struct sk_buff *skb, int code, int pos); + +struct flowi; +extern void icmpv6_flow_init(struct sock *sk, + struct flowi *fl, + u8 type, + const struct in6_addr *saddr, + const struct in6_addr *daddr, + int oif); #endif #endif diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 12c0b85d6c46..cff74127ea32 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -777,6 +777,22 @@ drop_no_count: return 0; } +void icmpv6_flow_init(struct sock *sk, struct flowi *fl, + u8 type, + const struct in6_addr *saddr, + const struct in6_addr *daddr, + int oif) +{ + memset(fl, 0, sizeof(*fl)); + ipv6_addr_copy(&fl->fl6_src, saddr); + ipv6_addr_copy(&fl->fl6_dst, daddr); + fl->proto = IPPROTO_ICMPV6; + fl->fl_icmp_type = type; + fl->fl_icmp_code = 0; + fl->oif = oif; + security_sk_classify_flow(sk, fl); +} + /* * Special lock-class for __icmpv6_sk: */ diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 1fc33c8c7232..8db5f4a419aa 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -441,21 +441,6 @@ static void pndisc_destructor(struct pneigh_entry *n) /* * Send a Neighbour Advertisement */ - -static inline void ndisc_flow_init(struct flowi *fl, u8 type, - struct in6_addr *saddr, struct in6_addr *daddr, - int oif) -{ - memset(fl, 0, sizeof(*fl)); - ipv6_addr_copy(&fl->fl6_src, saddr); - ipv6_addr_copy(&fl->fl6_dst, daddr); - fl->proto = IPPROTO_ICMPV6; - fl->fl_icmp_type = type; - fl->fl_icmp_code = 0; - fl->oif = oif; - security_sk_classify_flow(ndisc_socket->sk, fl); -} - static void __ndisc_send(struct net_device *dev, struct neighbour *neigh, struct in6_addr *daddr, struct in6_addr *saddr, @@ -474,8 +459,8 @@ static void __ndisc_send(struct net_device *dev, type = icmp6h->icmp6_type; - ndisc_flow_init(&fl, type, saddr, daddr, - dev->ifindex); + icmpv6_flow_init(ndisc_socket->sk, &fl, type, + saddr, daddr, dev->ifindex); dst = ndisc_dst_alloc(dev, neigh, daddr, ip6_output); if (!dst) @@ -1439,8 +1424,8 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, return; } - ndisc_flow_init(&fl, NDISC_REDIRECT, &saddr_buf, &ipv6_hdr(skb)->saddr, - dev->ifindex); + icmpv6_flow_init(ndisc_socket->sk, &fl, NDISC_REDIRECT, + &saddr_buf, &ipv6_hdr(skb)->saddr, dev->ifindex); dst = ip6_route_output(NULL, &fl); if (dst == NULL) -- cgit v1.2.3 From 3b00944c5c73c49ef52bf17b66557c43c1d945fe Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 6 Dec 2007 16:11:48 -0800 Subject: [IPV6]: Make ndisc_dst_alloc() common for later use. For later use, this patch is renaming ndisc_dst_alloc() (and related function/structures) to icmp6_dst_alloc() (and so on). This patch also removing unused function- pointer argument for it. Signed-off-by: YOSHIFUJI Hideaki --- include/net/ip6_route.h | 8 ++++---- net/ipv6/ip6_fib.c | 2 +- net/ipv6/ndisc.c | 2 +- net/ipv6/route.c | 27 +++++++++++++-------------- 4 files changed, 19 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index f99e4f0f568f..8d155a645aba 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -73,11 +73,11 @@ extern struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, int oif, int flags); -extern struct dst_entry *ndisc_dst_alloc(struct net_device *dev, +extern struct dst_entry *icmp6_dst_alloc(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *addr, - int (*output)(struct sk_buff *)); -extern int ndisc_dst_gc(int *more); + struct in6_addr *addr); +extern int icmp6_dst_gc(int *more); + extern void fib6_force_start_gc(void); extern struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index bab72b6f1444..c70fd38b54b7 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1464,7 +1464,7 @@ void fib6_run_gc(unsigned long dummy) } gc_args.more = 0; - ndisc_dst_gc(&gc_args.more); + icmp6_dst_gc(&gc_args.more); fib6_clean_all(fib6_age, 0, NULL); if (gc_args.more) diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 8db5f4a419aa..eb322959a3e1 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -462,7 +462,7 @@ static void __ndisc_send(struct net_device *dev, icmpv6_flow_init(ndisc_socket->sk, &fl, type, saddr, daddr, dev->ifindex); - dst = ndisc_dst_alloc(dev, neigh, daddr, ip6_output); + dst = icmp6_dst_alloc(dev, neigh, daddr); if (!dst) return; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 6abe7da45ef7..cd717450fb10 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -904,13 +904,12 @@ static inline unsigned int ipv6_advmss(unsigned int mtu) return mtu; } -static struct dst_entry *ndisc_dst_gc_list; -static DEFINE_SPINLOCK(ndisc_lock); +static struct dst_entry *icmp6_dst_gc_list; +static DEFINE_SPINLOCK(icmp6_dst_lock); -struct dst_entry *ndisc_dst_alloc(struct net_device *dev, +struct dst_entry *icmp6_dst_alloc(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *addr, - int (*output)(struct sk_buff *)) + struct in6_addr *addr) { struct rt6_info *rt; struct inet6_dev *idev = in6_dev_get(dev); @@ -937,7 +936,7 @@ struct dst_entry *ndisc_dst_alloc(struct net_device *dev, rt->u.dst.metrics[RTAX_HOPLIMIT-1] = 255; rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev); rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst)); - rt->u.dst.output = output; + rt->u.dst.output = ip6_output; #if 0 /* there's no chance to use these for ndisc */ rt->u.dst.flags = ipv6_addr_type(addr) & IPV6_ADDR_UNICAST @@ -947,10 +946,10 @@ struct dst_entry *ndisc_dst_alloc(struct net_device *dev, rt->rt6i_dst.plen = 128; #endif - spin_lock_bh(&ndisc_lock); - rt->u.dst.next = ndisc_dst_gc_list; - ndisc_dst_gc_list = &rt->u.dst; - spin_unlock_bh(&ndisc_lock); + spin_lock_bh(&icmp6_dst_lock); + rt->u.dst.next = icmp6_dst_gc_list; + icmp6_dst_gc_list = &rt->u.dst; + spin_unlock_bh(&icmp6_dst_lock); fib6_force_start_gc(); @@ -958,7 +957,7 @@ out: return &rt->u.dst; } -int ndisc_dst_gc(int *more) +int icmp6_dst_gc(int *more) { struct dst_entry *dst, *next, **pprev; int freed; @@ -966,8 +965,8 @@ int ndisc_dst_gc(int *more) next = NULL; freed = 0; - spin_lock_bh(&ndisc_lock); - pprev = &ndisc_dst_gc_list; + spin_lock_bh(&icmp6_dst_lock); + pprev = &icmp6_dst_gc_list; while ((dst = *pprev) != NULL) { if (!atomic_read(&dst->__refcnt)) { @@ -980,7 +979,7 @@ int ndisc_dst_gc(int *more) } } - spin_unlock_bh(&ndisc_lock); + spin_unlock_bh(&icmp6_dst_lock); return freed; } -- cgit v1.2.3 From 58f09b78b730cf0d936597272bf35b3d615e967c Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Mon, 3 Mar 2008 23:25:27 -0800 Subject: [NETNS][IPV6] ip6_fib - make it per network namespace The fib table for ipv6 are moved to the network namespace structure. All references to them are made relatively to the network namespace. All external calls to the ip6_fib functions taking the network namespace parameter are made using the init_net variable, so the ip6_fib engine is ready for the namespaces but the callers not yet. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 9 +-- include/net/netns/ipv6.h | 5 ++ net/ipv6/fib6_rules.c | 8 +-- net/ipv6/ip6_fib.c | 161 +++++++++++++++++++++++++++-------------------- net/ipv6/route.c | 22 ++++--- 5 files changed, 119 insertions(+), 86 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 953d6040ff50..4d4c8aca8fb9 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -181,10 +181,11 @@ typedef struct rt6_info *(*pol_lookup_t)(struct fib6_table *, * exported functions */ -extern struct fib6_table * fib6_get_table(u32 id); -extern struct fib6_table * fib6_new_table(u32 id); -extern struct dst_entry * fib6_rule_lookup(struct flowi *fl, int flags, - pol_lookup_t lookup); +extern struct fib6_table *fib6_get_table(struct net *net, u32 id); +extern struct fib6_table *fib6_new_table(struct net *net, u32 id); +extern struct dst_entry *fib6_rule_lookup(struct net *net, + struct flowi *fl, int flags, + pol_lookup_t lookup); extern struct fib6_node *fib6_lookup(struct fib6_node *root, struct in6_addr *daddr, diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 82623d3a8e35..b0653261c5a3 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -35,6 +35,11 @@ struct netns_ipv6 { struct xt_table *ip6table_filter; struct xt_table *ip6table_mangle; struct xt_table *ip6table_raw; +#endif + struct hlist_head *fib_table_hash; + struct fib6_table *fib6_main_tbl; +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + struct fib6_table *fib6_local_tbl; #endif struct sock **icmp_sk; }; diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 157db3a1ce00..03ad23a5fd3c 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -31,8 +31,8 @@ struct fib6_rule static struct fib_rules_ops fib6_rules_ops; -struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags, - pol_lookup_t lookup) +struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, + int flags, pol_lookup_t lookup) { struct fib_lookup_arg arg = { .lookup_ptr = lookup, @@ -71,7 +71,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, goto discard_pkt; } - table = fib6_get_table(rule->table); + table = fib6_get_table(&init_net, rule->table); if (table) rt = lookup(table, flp, flags); @@ -151,7 +151,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, if (rule->table == RT6_TABLE_UNSPEC) goto errout; - if (fib6_new_table(rule->table) == NULL) { + if (fib6_new_table(&init_net, rule->table) == NULL) { err = -ENOBUFS; goto errout; } diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 04d774963f3c..7b549f0bc428 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -166,16 +166,13 @@ static __inline__ void rt6_release(struct rt6_info *rt) dst_free(&rt->u.dst); } -static struct fib6_table *fib6_main_tbl; - #ifdef CONFIG_IPV6_MULTIPLE_TABLES #define FIB_TABLE_HASHSZ 256 #else #define FIB_TABLE_HASHSZ 1 #endif -static struct hlist_head *fib_table_hash; -static void fib6_link_table(struct fib6_table *tb) +static void fib6_link_table(struct net *net, struct fib6_table *tb) { unsigned int h; @@ -191,13 +188,11 @@ static void fib6_link_table(struct fib6_table *tb) * No protection necessary, this is the only list mutatation * operation, tables never disappear once they exist. */ - hlist_add_head_rcu(&tb->tb6_hlist, &fib_table_hash[h]); + hlist_add_head_rcu(&tb->tb6_hlist, &net->ipv6.fib_table_hash[h]); } #ifdef CONFIG_IPV6_MULTIPLE_TABLES -static struct fib6_table *fib6_local_tbl; - static struct fib6_table *fib6_alloc_table(u32 id) { struct fib6_table *table; @@ -212,26 +207,27 @@ static struct fib6_table *fib6_alloc_table(u32 id) return table; } -struct fib6_table *fib6_new_table(u32 id) +struct fib6_table *fib6_new_table(struct net *net, u32 id) { struct fib6_table *tb; if (id == 0) id = RT6_TABLE_MAIN; - tb = fib6_get_table(id); + tb = fib6_get_table(net, id); if (tb) return tb; tb = fib6_alloc_table(id); if (tb != NULL) - fib6_link_table(tb); + fib6_link_table(net, tb); return tb; } -struct fib6_table *fib6_get_table(u32 id) +struct fib6_table *fib6_get_table(struct net *net, u32 id) { struct fib6_table *tb; + struct hlist_head *head; struct hlist_node *node; unsigned int h; @@ -239,7 +235,8 @@ struct fib6_table *fib6_get_table(u32 id) id = RT6_TABLE_MAIN; h = id & (FIB_TABLE_HASHSZ - 1); rcu_read_lock(); - hlist_for_each_entry_rcu(tb, node, &fib_table_hash[h], tb6_hlist) { + head = &net->ipv6.fib_table_hash[h]; + hlist_for_each_entry_rcu(tb, node, head, tb6_hlist) { if (tb->tb6_id == id) { rcu_read_unlock(); return tb; @@ -250,33 +247,32 @@ struct fib6_table *fib6_get_table(u32 id) return NULL; } -static void __init fib6_tables_init(void) +static void fib6_tables_init(struct net *net) { - fib6_link_table(fib6_main_tbl); - fib6_link_table(fib6_local_tbl); + fib6_link_table(net, net->ipv6.fib6_main_tbl); + fib6_link_table(net, net->ipv6.fib6_local_tbl); } - #else -struct fib6_table *fib6_new_table(u32 id) +struct fib6_table *fib6_new_table(struct net *net, u32 id) { - return fib6_get_table(id); + return fib6_get_table(net, id); } -struct fib6_table *fib6_get_table(u32 id) +struct fib6_table *fib6_get_table(struct net *net, u32 id) { - return fib6_main_tbl; + return net->ipv6.fib6_main_tbl; } -struct dst_entry *fib6_rule_lookup(struct flowi *fl, int flags, - pol_lookup_t lookup) +struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, + int flags, pol_lookup_t lookup) { - return (struct dst_entry *) lookup(fib6_main_tbl, fl, flags); + return (struct dst_entry *) lookup(net->ipv6.fib6_main_tbl, fl, flags); } -static void __init fib6_tables_init(void) +static void fib6_tables_init(struct net *net) { - fib6_link_table(fib6_main_tbl); + fib6_link_table(net, net->ipv6.fib6_main_tbl); } #endif @@ -357,11 +353,9 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) struct fib6_walker_t *w; struct fib6_table *tb; struct hlist_node *node; + struct hlist_head *head; int res = 0; - if (net != &init_net) - return 0; - s_h = cb->args[0]; s_e = cb->args[1]; @@ -390,7 +384,8 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) for (h = s_h; h < FIB_TABLE_HASHSZ; h++, s_e = 0) { e = 0; - hlist_for_each_entry(tb, node, &fib_table_hash[h], tb6_hlist) { + head = &net->ipv6.fib_table_hash[h]; + hlist_for_each_entry(tb, node, head, tb6_hlist) { if (e < s_e) goto next; res = fib6_dump_table(tb, skb, cb); @@ -1360,12 +1355,13 @@ void fib6_clean_all(int (*func)(struct rt6_info *, void *arg), { struct fib6_table *table; struct hlist_node *node; + struct hlist_head *head; unsigned int h; rcu_read_lock(); for (h = 0; h < FIB_TABLE_HASHSZ; h++) { - hlist_for_each_entry_rcu(table, node, &fib_table_hash[h], - tb6_hlist) { + head = &init_net.ipv6.fib_table_hash[h]; + hlist_for_each_entry_rcu(table, node, head, tb6_hlist) { write_lock_bh(&table->tb6_lock); fib6_clean_tree(&table->tb6_root, func, prune, arg); write_unlock_bh(&table->tb6_lock); @@ -1466,55 +1462,88 @@ void fib6_run_gc(unsigned long dummy) spin_unlock_bh(&fib6_gc_lock); } -int __init fib6_init(void) +static int fib6_net_init(struct net *net) { - int ret = -ENOMEM; - fib6_node_kmem = kmem_cache_create("fib6_nodes", - sizeof(struct fib6_node), - 0, SLAB_HWCACHE_ALIGN, - NULL); - if (!fib6_node_kmem) - goto out; + int ret; - fib_table_hash = kzalloc(sizeof(*fib_table_hash)*FIB_TABLE_HASHSZ, - GFP_KERNEL); - if (!fib_table_hash) - goto out_kmem_cache_create; + ret = -ENOMEM; + net->ipv6.fib_table_hash = + kzalloc(sizeof(*net->ipv6.fib_table_hash)*FIB_TABLE_HASHSZ, + GFP_KERNEL); + if (!net->ipv6.fib_table_hash) + goto out; - fib6_main_tbl = kzalloc(sizeof(*fib6_main_tbl), GFP_KERNEL); - if (!fib6_main_tbl) + net->ipv6.fib6_main_tbl = kzalloc(sizeof(*net->ipv6.fib6_main_tbl), + GFP_KERNEL); + if (!net->ipv6.fib6_main_tbl) goto out_fib_table_hash; - fib6_main_tbl->tb6_id = RT6_TABLE_MAIN; - fib6_main_tbl->tb6_root.leaf = &ip6_null_entry; - fib6_main_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; + net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN; + net->ipv6.fib6_main_tbl->tb6_root.leaf = &ip6_null_entry; + net->ipv6.fib6_main_tbl->tb6_root.fn_flags = + RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; #ifdef CONFIG_IPV6_MULTIPLE_TABLES - fib6_local_tbl = kzalloc(sizeof(*fib6_local_tbl), GFP_KERNEL); - if (!fib6_local_tbl) + net->ipv6.fib6_local_tbl = kzalloc(sizeof(*net->ipv6.fib6_local_tbl), + GFP_KERNEL); + if (!net->ipv6.fib6_local_tbl) goto out_fib6_main_tbl; - - fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL; - fib6_local_tbl->tb6_root.leaf = &ip6_null_entry; - fib6_local_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; + net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL; + net->ipv6.fib6_local_tbl->tb6_root.leaf = &ip6_null_entry; + net->ipv6.fib6_local_tbl->tb6_root.fn_flags = + RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; #endif + fib6_tables_init(net); - fib6_tables_init(); - - ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib); - if (ret) - goto out_fib6_local_tbl; + ret = 0; out: return ret; -out_fib6_local_tbl: #ifdef CONFIG_IPV6_MULTIPLE_TABLES - kfree(fib6_local_tbl); out_fib6_main_tbl: + kfree(net->ipv6.fib6_main_tbl); #endif - kfree(fib6_main_tbl); out_fib_table_hash: - kfree(fib_table_hash); + kfree(net->ipv6.fib_table_hash); + goto out; + } + +static void fib6_net_exit(struct net *net) +{ +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + kfree(net->ipv6.fib6_local_tbl); +#endif + kfree(net->ipv6.fib6_main_tbl); + kfree(net->ipv6.fib_table_hash); +} + +static struct pernet_operations fib6_net_ops = { + .init = fib6_net_init, + .exit = fib6_net_exit, +}; + +int __init fib6_init(void) +{ + int ret = -ENOMEM; + fib6_node_kmem = kmem_cache_create("fib6_nodes", + sizeof(struct fib6_node), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (!fib6_node_kmem) + goto out; + + ret = register_pernet_subsys(&fib6_net_ops); + if (ret) + goto out_kmem_cache_create; + + ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib); + if (ret) + goto out_unregister_subsys; +out: + return ret; + +out_unregister_subsys: + unregister_pernet_subsys(&fib6_net_ops); out_kmem_cache_create: kmem_cache_destroy(fib6_node_kmem); goto out; @@ -1523,10 +1552,6 @@ out_kmem_cache_create: void fib6_gc_cleanup(void) { del_timer(&ip6_fib_timer); -#ifdef CONFIG_IPV6_MULTIPLE_TABLES - kfree(fib6_local_tbl); -#endif - kfree(fib6_main_tbl); - kfree(fib_table_hash); + unregister_pernet_subsys(&fib6_net_ops); kmem_cache_destroy(fib6_node_kmem); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index cd717450fb10..09206f7ba525 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -571,7 +571,7 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, flags |= RT6_LOOKUP_F_HAS_SADDR; } - dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_lookup); + dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_lookup); if (dst->error == 0) return (struct rt6_info *) dst; @@ -758,7 +758,7 @@ void ip6_route_input(struct sk_buff *skb) if (rt6_need_strict(&iph->daddr)) flags |= RT6_LOOKUP_F_IFACE; - skb->dst = fib6_rule_lookup(&fl, flags, ip6_pol_route_input); + skb->dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_input); } static struct rt6_info *ip6_pol_route_output(struct fib6_table *table, @@ -777,7 +777,7 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) if (!ipv6_addr_any(&fl->fl6_src)) flags |= RT6_LOOKUP_F_HAS_SADDR; - return fib6_rule_lookup(fl, flags, ip6_pol_route_output); + return fib6_rule_lookup(&init_net, fl, flags, ip6_pol_route_output); } EXPORT_SYMBOL(ip6_route_output); @@ -1069,7 +1069,7 @@ int ip6_route_add(struct fib6_config *cfg) if (cfg->fc_metric == 0) cfg->fc_metric = IP6_RT_PRIO_USER; - table = fib6_new_table(cfg->fc_table); + table = fib6_new_table(&init_net, cfg->fc_table); if (table == NULL) { err = -ENOBUFS; goto out; @@ -1275,7 +1275,7 @@ static int ip6_route_del(struct fib6_config *cfg) struct rt6_info *rt; int err = -ESRCH; - table = fib6_get_table(cfg->fc_table); + table = fib6_get_table(&init_net, cfg->fc_table); if (table == NULL) return err; @@ -1390,7 +1390,9 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, if (rt6_need_strict(dest)) flags |= RT6_LOOKUP_F_IFACE; - return (struct rt6_info *)fib6_rule_lookup((struct flowi *)&rdfl, flags, __ip6_route_redirect); + return (struct rt6_info *)fib6_rule_lookup(&init_net, + (struct flowi *)&rdfl, + flags, __ip6_route_redirect); } void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, @@ -1589,7 +1591,7 @@ static struct rt6_info *rt6_get_route_info(struct in6_addr *prefix, int prefixle struct rt6_info *rt = NULL; struct fib6_table *table; - table = fib6_get_table(RT6_TABLE_INFO); + table = fib6_get_table(&init_net, RT6_TABLE_INFO); if (table == NULL) return NULL; @@ -1644,7 +1646,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d struct rt6_info *rt; struct fib6_table *table; - table = fib6_get_table(RT6_TABLE_DFLT); + table = fib6_get_table(&init_net, RT6_TABLE_DFLT); if (table == NULL) return NULL; @@ -1688,7 +1690,7 @@ void rt6_purge_dflt_routers(void) struct fib6_table *table; /* NOTE: Keep consistent with rt6_get_dflt_router */ - table = fib6_get_table(RT6_TABLE_DFLT); + table = fib6_get_table(&init_net, RT6_TABLE_DFLT); if (table == NULL) return; @@ -1851,7 +1853,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, ipv6_addr_copy(&rt->rt6i_dst.addr, addr); rt->rt6i_dst.plen = 128; - rt->rt6i_table = fib6_get_table(RT6_TABLE_LOCAL); + rt->rt6i_table = fib6_get_table(&init_net, RT6_TABLE_LOCAL); atomic_set(&rt->u.dst.__refcnt, 1); -- cgit v1.2.3 From f3db48517f59133610f558f29de8834d7b007691 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Mon, 3 Mar 2008 23:27:06 -0800 Subject: [NETNS][IPV6] ip6_fib - fib6_clean_all handle several network namespaces The function fib6_clean_all takes the network namespace as parameter. That allows to flush the routes related to a specific network namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 3 ++- include/net/ip6_route.h | 2 +- net/ipv6/addrconf.c | 3 ++- net/ipv6/ip6_fib.c | 7 ++++--- net/ipv6/route.c | 28 ++++++++++++++++++++-------- 5 files changed, 29 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 4d4c8aca8fb9..e54075d902db 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -195,7 +195,8 @@ struct fib6_node *fib6_locate(struct fib6_node *root, struct in6_addr *daddr, int dst_len, struct in6_addr *saddr, int src_len); -extern void fib6_clean_all(int (*func)(struct rt6_info *, void *arg), +extern void fib6_clean_all(struct net *net, + int (*func)(struct rt6_info *, void *arg), int prune, void *arg); extern int fib6_add(struct fib6_node *root, diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 8d155a645aba..a158abe71933 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -121,7 +121,7 @@ struct rt6_rtnl_dump_arg }; extern int rt6_dump_route(struct rt6_info *rt, void *p_arg); -extern void rt6_ifdown(struct net_device *dev); +extern void rt6_ifdown(struct net *net, struct net_device *dev); extern void rt6_mtu_change(struct net_device *dev, unsigned mtu); extern rwlock_t rt6_lock; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 9b3a2d0e4269..a1d872dacad6 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2431,6 +2431,7 @@ static int addrconf_ifdown(struct net_device *dev, int how) { struct inet6_dev *idev; struct inet6_ifaddr *ifa, **bifa; + struct net *net = dev->nd_net; int i; ASSERT_RTNL(); @@ -2438,7 +2439,7 @@ static int addrconf_ifdown(struct net_device *dev, int how) if (dev == init_net.loopback_dev && how == 1) how = 0; - rt6_ifdown(dev); + rt6_ifdown(net, dev); neigh_ifdown(&nd_tbl, dev); idev = __in6_dev_get(dev); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 7b549f0bc428..0f9dc81f0e61 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1350,7 +1350,7 @@ static void fib6_clean_tree(struct fib6_node *root, fib6_walk(&c.w); } -void fib6_clean_all(int (*func)(struct rt6_info *, void *arg), +void fib6_clean_all(struct net *net, int (*func)(struct rt6_info *, void *arg), int prune, void *arg) { struct fib6_table *table; @@ -1360,7 +1360,7 @@ void fib6_clean_all(int (*func)(struct rt6_info *, void *arg), rcu_read_lock(); for (h = 0; h < FIB_TABLE_HASHSZ; h++) { - head = &init_net.ipv6.fib_table_hash[h]; + head = &net->ipv6.fib_table_hash[h]; hlist_for_each_entry_rcu(table, node, head, tb6_hlist) { write_lock_bh(&table->tb6_lock); fib6_clean_tree(&table->tb6_root, func, prune, arg); @@ -1450,7 +1450,8 @@ void fib6_run_gc(unsigned long dummy) gc_args.more = 0; icmp6_dst_gc(&gc_args.more); - fib6_clean_all(fib6_age, 0, NULL); + + fib6_clean_all(&init_net, fib6_age, 0, NULL); if (gc_args.more) mod_timer(&ip6_fib_timer, jiffies + diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 09206f7ba525..2e6da2afd948 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1870,9 +1870,9 @@ static int fib6_ifdown(struct rt6_info *rt, void *arg) return 0; } -void rt6_ifdown(struct net_device *dev) +void rt6_ifdown(struct net *net, struct net_device *dev) { - fib6_clean_all(fib6_ifdown, 0, dev); + fib6_clean_all(net, fib6_ifdown, 0, dev); } struct rt6_mtu_change_arg @@ -1928,7 +1928,7 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu) .mtu = mtu, }; - fib6_clean_all(rt6_mtu_change_route, 0, &arg); + fib6_clean_all(dev->nd_net, rt6_mtu_change_route, 0, &arg); } static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = { @@ -2318,13 +2318,25 @@ static int rt6_info_route(struct rt6_info *rt, void *p_arg) static int ipv6_route_show(struct seq_file *m, void *v) { - fib6_clean_all(rt6_info_route, 0, m); + struct net *net = (struct net *)m->private; + fib6_clean_all(net, rt6_info_route, 0, m); return 0; } static int ipv6_route_open(struct inode *inode, struct file *file) { - return single_open(file, ipv6_route_show, NULL); + struct net *net = get_proc_net(inode); + if (!net) + return -ENXIO; + return single_open(file, ipv6_route_show, net); +} + +static int ipv6_route_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct net *net = seq->private; + put_net(net); + return single_release(inode, file); } static const struct file_operations ipv6_route_proc_fops = { @@ -2332,7 +2344,7 @@ static const struct file_operations ipv6_route_proc_fops = { .open = ipv6_route_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = ipv6_route_release, }; static int rt6_stats_seq_show(struct seq_file *seq, void *v) @@ -2570,7 +2582,7 @@ xfrm6_init: out_proc_init: ipv6_route_proc_fini(&init_net); out_fib6_init: - rt6_ifdown(NULL); + rt6_ifdown(&init_net, NULL); fib6_gc_cleanup(); out_kmem_cache: kmem_cache_destroy(ip6_dst_ops.kmem_cachep); @@ -2582,7 +2594,7 @@ void ip6_route_cleanup(void) fib6_rules_cleanup(); ipv6_route_proc_fini(&init_net); xfrm6_fini(); - rt6_ifdown(NULL); + rt6_ifdown(&init_net, NULL); fib6_gc_cleanup(); kmem_cache_destroy(ip6_dst_ops.kmem_cachep); } -- cgit v1.2.3 From 5b7c931dff03621ae7ac524c4fa280d4e5f187a4 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Mon, 3 Mar 2008 23:28:58 -0800 Subject: [NETNS][IPV6] ip6_fib - add net to gc timer parameter The fib tables are now relative to the network namespace. When the garbage collector timer expires, we must have a network namespace parameter in order to retrieve the tables. For now this is the init_net, but we should be able to have a timer per namespace and use the timer callback parameter to pass the network namespace from the expired timer. The timer callback, fib6_run_gc, is actually used to be called synchronously by some functions and asynchronously when the timer expires. When the timer expires, the delay specified for fib6_run_gc parameter is always zero. So, I changed fib6_run_gc to not be a timer callback but a function called by the timer callback and I added a timer callback where its work is just to retrieve from the data arg of the timer the network namespace and call fib6_run_gc with zero expiring time and the network namespace parameters. That makes the code cleaner for the fib6_run_gc callers. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 3 ++- net/ipv6/ip6_fib.c | 24 ++++++++++++++++-------- net/ipv6/ndisc.c | 5 +++-- net/ipv6/route.c | 8 +++++--- 4 files changed, 26 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index e54075d902db..fae267f65341 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -209,7 +209,8 @@ extern int fib6_del(struct rt6_info *rt, extern void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info); -extern void fib6_run_gc(unsigned long dummy); +extern void fib6_run_gc(unsigned long expires, + struct net *net); extern void fib6_gc_cleanup(void); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 0f9dc81f0e61..9b1c232a29ee 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -93,7 +93,10 @@ static int fib6_walk_continue(struct fib6_walker_t *w); static __u32 rt_sernum; -static DEFINE_TIMER(ip6_fib_timer, fib6_run_gc, 0, 0); +static void fib6_gc_timer_cb(unsigned long arg); + +static DEFINE_TIMER(ip6_fib_timer, fib6_gc_timer_cb, 0, + (unsigned long)&init_net); static struct fib6_walker_t fib6_walker_list = { .prev = &fib6_walker_list, @@ -1432,12 +1435,12 @@ static int fib6_age(struct rt6_info *rt, void *arg) static DEFINE_SPINLOCK(fib6_gc_lock); -void fib6_run_gc(unsigned long dummy) +void fib6_run_gc(unsigned long expires, struct net *net) { - if (dummy != ~0UL) { + if (expires != ~0UL) { spin_lock_bh(&fib6_gc_lock); - gc_args.timeout = dummy ? (int)dummy : - init_net.ipv6.sysctl.ip6_rt_gc_interval; + gc_args.timeout = expires ? (int)expires : + net->ipv6.sysctl.ip6_rt_gc_interval; } else { local_bh_disable(); if (!spin_trylock(&fib6_gc_lock)) { @@ -1445,17 +1448,17 @@ void fib6_run_gc(unsigned long dummy) local_bh_enable(); return; } - gc_args.timeout = init_net.ipv6.sysctl.ip6_rt_gc_interval; + gc_args.timeout = net->ipv6.sysctl.ip6_rt_gc_interval; } gc_args.more = 0; icmp6_dst_gc(&gc_args.more); - fib6_clean_all(&init_net, fib6_age, 0, NULL); + fib6_clean_all(net, fib6_age, 0, NULL); if (gc_args.more) mod_timer(&ip6_fib_timer, jiffies + - init_net.ipv6.sysctl.ip6_rt_gc_interval); + net->ipv6.sysctl.ip6_rt_gc_interval); else { del_timer(&ip6_fib_timer); ip6_fib_timer.expires = 0; @@ -1463,6 +1466,11 @@ void fib6_run_gc(unsigned long dummy) spin_unlock_bh(&fib6_gc_lock); } +static void fib6_gc_timer_cb(unsigned long arg) +{ + fib6_run_gc(0, (struct net *)arg); +} + static int fib6_net_init(struct net *net) { int ret; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index eb322959a3e1..f1c95125100d 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1598,6 +1598,7 @@ int ndisc_rcv(struct sk_buff *skb) static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; + struct net *net = dev->nd_net; if (dev->nd_net != &init_net) return NOTIFY_DONE; @@ -1605,11 +1606,11 @@ static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, switch (event) { case NETDEV_CHANGEADDR: neigh_changeaddr(&nd_tbl, dev); - fib6_run_gc(~0UL); + fib6_run_gc(~0UL, net); break; case NETDEV_DOWN: neigh_ifdown(&nd_tbl, dev); - fib6_run_gc(~0UL); + fib6_run_gc(~0UL, net); break; default: break; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 2e6da2afd948..fd44721abebb 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include #include @@ -995,7 +996,7 @@ static int ip6_dst_gc(struct dst_ops *ops) goto out; expire++; - fib6_run_gc(expire); + fib6_run_gc(expire, &init_net); last_gc = now; if (atomic_read(&ip6_dst_ops.entries) < ip6_dst_ops.gc_thresh) expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1; @@ -2413,10 +2414,11 @@ static int ipv6_sysctl_rtcache_flush(ctl_table *ctl, int write, struct file * filp, void __user *buffer, size_t *lenp, loff_t *ppos) { - int delay = init_net.ipv6.sysctl.flush_delay; + struct net *net = current->nsproxy->net_ns; + int delay = net->ipv6.sysctl.flush_delay; if (write) { proc_dointvec(ctl, write, filp, buffer, lenp, ppos); - fib6_run_gc(delay <= 0 ? ~0UL : (unsigned long)delay); + fib6_run_gc(delay <= 0 ? ~0UL : (unsigned long)delay, net); return 0; } else return -EINVAL; -- cgit v1.2.3 From 63152fc0de4dfe83da543bf133cef73d885a50fc Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Mon, 3 Mar 2008 23:31:11 -0800 Subject: [NETNS][IPV6] ip6_fib - gc timer per namespace Move the timer initialization at the network namespace creation and store the network namespace in the timer argument. That enables multiple timers (one per network namespace) to do garbage collecting. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_route.h | 2 +- include/net/netns/ipv6.h | 1 + net/ipv6/ip6_fib.c | 56 ++++++++++++++++++++++++------------------------ net/ipv6/route.c | 5 ++++- 4 files changed, 34 insertions(+), 30 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index a158abe71933..79dce496f4d4 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -78,7 +78,7 @@ extern struct dst_entry *icmp6_dst_alloc(struct net_device *dev, struct in6_addr *addr); extern int icmp6_dst_gc(int *more); -extern void fib6_force_start_gc(void); +extern void fib6_force_start_gc(struct net *net); extern struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, const struct in6_addr *addr, diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index b0653261c5a3..5279cd6a00ba 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -36,6 +36,7 @@ struct netns_ipv6 { struct xt_table *ip6table_mangle; struct xt_table *ip6table_raw; #endif + struct timer_list *ip6_fib_timer; struct hlist_head *fib_table_hash; struct fib6_table *fib6_main_tbl; #ifdef CONFIG_IPV6_MULTIPLE_TABLES diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 77ad1002c904..4aa67988f518 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -95,8 +95,6 @@ static __u32 rt_sernum; static void fib6_gc_timer_cb(unsigned long arg); -static struct timer_list *ip6_fib_timer; - static struct fib6_walker_t fib6_walker_list = { .prev = &fib6_walker_list, .next = &fib6_walker_list, @@ -663,19 +661,19 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, return 0; } -static __inline__ void fib6_start_gc(struct rt6_info *rt) +static __inline__ void fib6_start_gc(struct net *net, struct rt6_info *rt) { - if (ip6_fib_timer->expires == 0 && + if (net->ipv6.ip6_fib_timer->expires == 0 && (rt->rt6i_flags & (RTF_EXPIRES|RTF_CACHE))) - mod_timer(ip6_fib_timer, jiffies + - init_net.ipv6.sysctl.ip6_rt_gc_interval); + mod_timer(net->ipv6.ip6_fib_timer, jiffies + + net->ipv6.sysctl.ip6_rt_gc_interval); } -void fib6_force_start_gc(void) +void fib6_force_start_gc(struct net *net) { - if (ip6_fib_timer->expires == 0) - mod_timer(ip6_fib_timer, jiffies + - init_net.ipv6.sysctl.ip6_rt_gc_interval); + if (net->ipv6.ip6_fib_timer->expires == 0) + mod_timer(net->ipv6.ip6_fib_timer, jiffies + + net->ipv6.sysctl.ip6_rt_gc_interval); } /* @@ -762,7 +760,7 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info) err = fib6_add_rt2node(fn, rt, info); if (err == 0) { - fib6_start_gc(rt); + fib6_start_gc(info->nl_net, rt); if (!(rt->rt6i_flags&RTF_CACHE)) fib6_prune_clones(pn, rt); } @@ -1443,7 +1441,7 @@ void fib6_run_gc(unsigned long expires, struct net *net) } else { local_bh_disable(); if (!spin_trylock(&fib6_gc_lock)) { - mod_timer(ip6_fib_timer, jiffies + HZ); + mod_timer(net->ipv6.ip6_fib_timer, jiffies + HZ); local_bh_enable(); return; } @@ -1456,11 +1454,11 @@ void fib6_run_gc(unsigned long expires, struct net *net) fib6_clean_all(net, fib6_age, 0, NULL); if (gc_args.more) - mod_timer(ip6_fib_timer, jiffies + + mod_timer(net->ipv6.ip6_fib_timer, jiffies + net->ipv6.sysctl.ip6_rt_gc_interval); else { - del_timer(ip6_fib_timer); - ip6_fib_timer->expires = 0; + del_timer(net->ipv6.ip6_fib_timer); + net->ipv6.ip6_fib_timer->expires = 0; } spin_unlock_bh(&fib6_gc_lock); } @@ -1473,13 +1471,21 @@ static void fib6_gc_timer_cb(unsigned long arg) static int fib6_net_init(struct net *net) { int ret; + struct timer_list *timer; ret = -ENOMEM; + timer = kzalloc(sizeof(*timer), GFP_KERNEL); + if (!timer) + goto out; + + setup_timer(timer, fib6_gc_timer_cb, (unsigned long)net); + net->ipv6.ip6_fib_timer = timer; + net->ipv6.fib_table_hash = kzalloc(sizeof(*net->ipv6.fib_table_hash)*FIB_TABLE_HASHSZ, GFP_KERNEL); if (!net->ipv6.fib_table_hash) - goto out; + goto out_timer; net->ipv6.fib6_main_tbl = kzalloc(sizeof(*net->ipv6.fib6_main_tbl), GFP_KERNEL); @@ -1513,11 +1519,15 @@ out_fib6_main_tbl: #endif out_fib_table_hash: kfree(net->ipv6.fib_table_hash); +out_timer: + kfree(timer); goto out; } static void fib6_net_exit(struct net *net) { + del_timer(net->ipv6.ip6_fib_timer); + kfree(net->ipv6.ip6_fib_timer); #ifdef CONFIG_IPV6_MULTIPLE_TABLES kfree(net->ipv6.fib6_local_tbl); #endif @@ -1533,6 +1543,7 @@ static struct pernet_operations fib6_net_ops = { int __init fib6_init(void) { int ret = -ENOMEM; + fib6_node_kmem = kmem_cache_create("fib6_nodes", sizeof(struct fib6_node), 0, SLAB_HWCACHE_ALIGN, @@ -1540,16 +1551,9 @@ int __init fib6_init(void) if (!fib6_node_kmem) goto out; - ret = -ENOMEM; - ip6_fib_timer = kzalloc(sizeof(*ip6_fib_timer), GFP_KERNEL); - if (!ip6_fib_timer) - goto out_kmem_cache_create; - - setup_timer(ip6_fib_timer, fib6_gc_timer_cb, (unsigned long)&init_net); - ret = register_pernet_subsys(&fib6_net_ops); if (ret) - goto out_timer; + goto out_kmem_cache_create; ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib); if (ret) @@ -1559,8 +1563,6 @@ out: out_unregister_subsys: unregister_pernet_subsys(&fib6_net_ops); -out_timer: - kfree(ip6_fib_timer); out_kmem_cache_create: kmem_cache_destroy(fib6_node_kmem); goto out; @@ -1568,8 +1570,6 @@ out_kmem_cache_create: void fib6_gc_cleanup(void) { - del_timer(ip6_fib_timer); - kfree(ip6_fib_timer); unregister_pernet_subsys(&fib6_net_ops); kmem_cache_destroy(fib6_node_kmem); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index fd44721abebb..b13eb0111270 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -952,7 +952,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, icmp6_dst_gc_list = &rt->u.dst; spin_unlock_bh(&icmp6_dst_lock); - fib6_force_start_gc(); + fib6_force_start_gc(dev->nd_net); out: return &rt->u.dst; @@ -1230,6 +1230,9 @@ install_route: rt->u.dst.dev = dev; rt->rt6i_idev = idev; rt->rt6i_table = table; + + cfg->fc_nlinfo.nl_net = dev->nd_net; + return __ip6_ins_rt(rt, &cfg->fc_nlinfo); out: -- cgit v1.2.3 From dcabb819a6eced95ef531b001e663d0d592c8d9f Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Mon, 3 Mar 2008 23:33:08 -0800 Subject: [NETNS][IPV6] fib6_rules - handle several network namespaces The fib6_rules_ops is moved to the network namespace structure. All references are changed to have it relatively to it. Each time a network namespace is created a new fib6_rules_ops is allocated, initialized and stored into the network namespace structure. The common part of the fib rules is namespace aware, so it is quite easy to retrieve the network namespace from the rules and use it in the different callbacks. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 1 + net/ipv6/fib6_rules.c | 82 +++++++++++++++++++++++++++--------------------- 2 files changed, 47 insertions(+), 36 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 5279cd6a00ba..66bf9c0f745b 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -41,6 +41,7 @@ struct netns_ipv6 { struct fib6_table *fib6_main_tbl; #ifdef CONFIG_IPV6_MULTIPLE_TABLES struct fib6_table *fib6_local_tbl; + struct fib_rules_ops *fib6_rules_ops; #endif struct sock **icmp_sk; }; diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 60af08f12547..89cb092c9732 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -29,8 +29,6 @@ struct fib6_rule u8 tclass; }; -static struct fib_rules_ops *fib6_rules_ops; - struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, int flags, pol_lookup_t lookup) { @@ -38,7 +36,7 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, .lookup_ptr = lookup, }; - fib_rules_lookup(fib6_rules_ops, fl, flags, &arg); + fib_rules_lookup(net->ipv6.fib6_rules_ops, fl, flags, &arg); if (arg.rule) fib_rule_put(arg.rule); @@ -71,7 +69,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, goto discard_pkt; } - table = fib6_get_table(&init_net, rule->table); + table = fib6_get_table(rule->fr_net, rule->table); if (table) rt = lookup(table, flp, flags); @@ -145,13 +143,14 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct nlattr **tb) { int err = -EINVAL; + struct net *net = skb->sk->sk_net; struct fib6_rule *rule6 = (struct fib6_rule *) rule; if (rule->action == FR_ACT_TO_TBL) { if (rule->table == RT6_TABLE_UNSPEC) goto errout; - if (fib6_new_table(&init_net, rule->table) == NULL) { + if (fib6_new_table(net, rule->table) == NULL) { err = -ENOBUFS; goto errout; } @@ -251,49 +250,60 @@ static struct fib_rules_ops fib6_rules_ops_template = { .fro_net = &init_net, }; -static int __init fib6_default_rules_init(void) +static int fib6_rules_net_init(struct net *net) { - int err; + int err = -ENOMEM; - fib6_rules_ops = kmemdup(&fib6_rules_ops_template, - sizeof(*fib6_rules_ops), GFP_KERNEL); - if (!fib6_rules_ops) - return -ENOMEM; + net->ipv6.fib6_rules_ops = kmemdup(&fib6_rules_ops_template, + sizeof(*net->ipv6.fib6_rules_ops), + GFP_KERNEL); + if (!net->ipv6.fib6_rules_ops) + goto out; - INIT_LIST_HEAD(&fib6_rules_ops->rules_list); + net->ipv6.fib6_rules_ops->fro_net = net; + INIT_LIST_HEAD(&net->ipv6.fib6_rules_ops->rules_list); - err = fib_default_rule_add(fib6_rules_ops, 0, + err = fib_default_rule_add(net->ipv6.fib6_rules_ops, 0, RT6_TABLE_LOCAL, FIB_RULE_PERMANENT); - if (err < 0) - return err; - err = fib_default_rule_add(fib6_rules_ops, 0x7FFE, RT6_TABLE_MAIN, 0); - if (err < 0) - return err; - return 0; -} - -int __init fib6_rules_init(void) -{ - int ret; + if (err) + goto out_fib6_rules_ops; - ret = fib6_default_rules_init(); - if (ret) - goto out; + err = fib_default_rule_add(net->ipv6.fib6_rules_ops, + 0x7FFE, RT6_TABLE_MAIN, 0); + if (err) + goto out_fib6_default_rule_add; - ret = fib_rules_register(fib6_rules_ops); - if (ret) - goto out_default_rules_init; + err = fib_rules_register(net->ipv6.fib6_rules_ops); + if (err) + goto out_fib6_default_rule_add; out: - return ret; + return err; -out_default_rules_init: - fib_rules_cleanup_ops(fib6_rules_ops); - kfree(fib6_rules_ops); +out_fib6_default_rule_add: + fib_rules_cleanup_ops(net->ipv6.fib6_rules_ops); +out_fib6_rules_ops: + kfree(net->ipv6.fib6_rules_ops); goto out; } +static void fib6_rules_net_exit(struct net *net) +{ + fib_rules_unregister(net->ipv6.fib6_rules_ops); + kfree(net->ipv6.fib6_rules_ops); +} + +static struct pernet_operations fib6_rules_net_ops = { + .init = fib6_rules_net_init, + .exit = fib6_rules_net_exit, +}; + +int __init fib6_rules_init(void) +{ + return register_pernet_subsys(&fib6_rules_net_ops); +} + + void fib6_rules_cleanup(void) { - fib_rules_unregister(fib6_rules_ops); - kfree(fib6_rules_ops); + return unregister_pernet_subsys(&fib6_rules_net_ops); } -- cgit v1.2.3 From 6cc118bd506ae8c6436f507b838a0e1f6185fec2 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Mon, 3 Mar 2008 23:33:43 -0800 Subject: [NETNS][IPV6] rt6_stats - dynamically allocate the routes statistics This patch allocates the rt6_stats struct dynamically when the fib6 is initialized. That provides the ability to create several instances of this structure for the network namespaces. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ipv6.h | 2 +- net/ipv6/ip6_fib.c | 22 +++++++++++++++------- net/ipv6/route.c | 8 ++++---- 3 files changed, 20 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 5dc8164e5d3b..9a00f35ae773 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -611,7 +611,7 @@ extern void ipv6_misc_proc_exit(void); extern int snmp6_register_dev(struct inet6_dev *idev); extern int snmp6_unregister_dev(struct inet6_dev *idev); -extern struct rt6_statistics rt6_stats; +extern struct rt6_statistics *rt6_stats; #else static inline int snmp6_register_dev(struct inet6_dev *idev) { diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 350a970ca1e3..7d82ad540140 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -48,7 +48,7 @@ #define RT6_TRACE(x...) do { ; } while (0) #endif -struct rt6_statistics rt6_stats; +struct rt6_statistics *rt6_stats; static struct kmem_cache * fib6_node_kmem __read_mostly; @@ -653,10 +653,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, rt->rt6i_node = fn; atomic_inc(&rt->rt6i_ref); inet6_rt_notify(RTM_NEWROUTE, rt, info); - rt6_stats.fib_rt_entries++; + rt6_stats->fib_rt_entries++; if ((fn->fn_flags & RTN_RTINFO) == 0) { - rt6_stats.fib_route_nodes++; + rt6_stats->fib_route_nodes++; fn->fn_flags |= RTN_RTINFO; } @@ -1093,8 +1093,8 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, /* Unlink it */ *rtp = rt->u.dst.rt6_next; rt->rt6i_node = NULL; - rt6_stats.fib_rt_entries--; - rt6_stats.fib_discarded_routes++; + rt6_stats->fib_rt_entries--; + rt6_stats->fib_discarded_routes++; /* Reset round-robin state, if necessary */ if (fn->rr_ptr == rt) @@ -1117,7 +1117,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, /* If it was last route, expunge its radix tree node */ if (fn->leaf == NULL) { fn->fn_flags &= ~RTN_RTINFO; - rt6_stats.fib_route_nodes--; + rt6_stats->fib_route_nodes--; fn = fib6_repair_tree(fn); } @@ -1556,9 +1556,14 @@ int __init fib6_init(void) if (!fib6_node_kmem) goto out; + ret = -ENOMEM; + rt6_stats = kzalloc(sizeof(*rt6_stats), GFP_KERNEL); + if (!rt6_stats) + goto out_kmem_cache_create; + ret = register_pernet_subsys(&fib6_net_ops); if (ret) - goto out_kmem_cache_create; + goto out_rt6_stats; ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib); if (ret) @@ -1568,6 +1573,8 @@ out: out_unregister_subsys: unregister_pernet_subsys(&fib6_net_ops); +out_rt6_stats: + kfree(rt6_stats); out_kmem_cache_create: kmem_cache_destroy(fib6_node_kmem); goto out; @@ -1576,5 +1583,6 @@ out_kmem_cache_create: void fib6_gc_cleanup(void) { unregister_pernet_subsys(&fib6_net_ops); + kfree(rt6_stats); kmem_cache_destroy(fib6_node_kmem); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index b13eb0111270..370f2aef6909 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2354,11 +2354,11 @@ static const struct file_operations ipv6_route_proc_fops = { static int rt6_stats_seq_show(struct seq_file *seq, void *v) { seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n", - rt6_stats.fib_nodes, rt6_stats.fib_route_nodes, - rt6_stats.fib_rt_alloc, rt6_stats.fib_rt_entries, - rt6_stats.fib_rt_cache, + rt6_stats->fib_nodes, rt6_stats->fib_route_nodes, + rt6_stats->fib_rt_alloc, rt6_stats->fib_rt_entries, + rt6_stats->fib_rt_cache, atomic_read(&ip6_dst_ops.entries), - rt6_stats.fib_discarded_routes); + rt6_stats->fib_discarded_routes); return 0; } -- cgit v1.2.3 From c572872f89e46e38cdb35a43b81122bfb7ff43fc Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Mon, 3 Mar 2008 23:34:17 -0800 Subject: [NETNS][IPV6] rt6_stats - make the stats per network namespace The rt6_stats is now per namespace with this patch. It is allocated when a network namespace is created and freed when the network namespace exits and references are relative to the network namespace. Signed-off-by: Benjamin Thery Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/ipv6.h | 1 - include/net/netns/ipv6.h | 1 + net/ipv6/ip6_fib.c | 32 +++++++++++++++----------------- net/ipv6/route.c | 12 +++++++----- 4 files changed, 23 insertions(+), 23 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 9a00f35ae773..5f6df50a33a9 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -611,7 +611,6 @@ extern void ipv6_misc_proc_exit(void); extern int snmp6_register_dev(struct inet6_dev *idev); extern int snmp6_unregister_dev(struct inet6_dev *idev); -extern struct rt6_statistics *rt6_stats; #else static inline int snmp6_register_dev(struct inet6_dev *idev) { diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 66bf9c0f745b..c6c9afff13ef 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -36,6 +36,7 @@ struct netns_ipv6 { struct xt_table *ip6table_mangle; struct xt_table *ip6table_raw; #endif + struct rt6_statistics *rt6_stats; struct timer_list *ip6_fib_timer; struct hlist_head *fib_table_hash; struct fib6_table *fib6_main_tbl; diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 7d82ad540140..1c2566e3d392 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -48,8 +48,6 @@ #define RT6_TRACE(x...) do { ; } while (0) #endif -struct rt6_statistics *rt6_stats; - static struct kmem_cache * fib6_node_kmem __read_mostly; enum fib_walk_state_t @@ -653,10 +651,10 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, rt->rt6i_node = fn; atomic_inc(&rt->rt6i_ref); inet6_rt_notify(RTM_NEWROUTE, rt, info); - rt6_stats->fib_rt_entries++; + info->nl_net->ipv6.rt6_stats->fib_rt_entries++; if ((fn->fn_flags & RTN_RTINFO) == 0) { - rt6_stats->fib_route_nodes++; + info->nl_net->ipv6.rt6_stats->fib_route_nodes++; fn->fn_flags |= RTN_RTINFO; } @@ -1087,14 +1085,15 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, { struct fib6_walker_t *w; struct rt6_info *rt = *rtp; + struct net *net = info->nl_net; RT6_TRACE("fib6_del_route\n"); /* Unlink it */ *rtp = rt->u.dst.rt6_next; rt->rt6i_node = NULL; - rt6_stats->fib_rt_entries--; - rt6_stats->fib_discarded_routes++; + net->ipv6.rt6_stats->fib_rt_entries--; + net->ipv6.rt6_stats->fib_discarded_routes++; /* Reset round-robin state, if necessary */ if (fn->rr_ptr == rt) @@ -1117,7 +1116,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, /* If it was last route, expunge its radix tree node */ if (fn->leaf == NULL) { fn->fn_flags &= ~RTN_RTINFO; - rt6_stats->fib_route_nodes--; + net->ipv6.rt6_stats->fib_route_nodes--; fn = fib6_repair_tree(fn); } @@ -1486,11 +1485,15 @@ static int fib6_net_init(struct net *net) setup_timer(timer, fib6_gc_timer_cb, (unsigned long)net); net->ipv6.ip6_fib_timer = timer; + net->ipv6.rt6_stats = kzalloc(sizeof(*net->ipv6.rt6_stats), GFP_KERNEL); + if (!net->ipv6.rt6_stats) + goto out_timer; + net->ipv6.fib_table_hash = kzalloc(sizeof(*net->ipv6.fib_table_hash)*FIB_TABLE_HASHSZ, GFP_KERNEL); if (!net->ipv6.fib_table_hash) - goto out_timer; + goto out_rt6_stats; net->ipv6.fib6_main_tbl = kzalloc(sizeof(*net->ipv6.fib6_main_tbl), GFP_KERNEL); @@ -1524,6 +1527,8 @@ out_fib6_main_tbl: #endif out_fib_table_hash: kfree(net->ipv6.fib_table_hash); +out_rt6_stats: + kfree(net->ipv6.rt6_stats); out_timer: kfree(timer); goto out; @@ -1538,6 +1543,7 @@ static void fib6_net_exit(struct net *net) #endif kfree(net->ipv6.fib6_main_tbl); kfree(net->ipv6.fib_table_hash); + kfree(net->ipv6.rt6_stats); } static struct pernet_operations fib6_net_ops = { @@ -1556,14 +1562,9 @@ int __init fib6_init(void) if (!fib6_node_kmem) goto out; - ret = -ENOMEM; - rt6_stats = kzalloc(sizeof(*rt6_stats), GFP_KERNEL); - if (!rt6_stats) - goto out_kmem_cache_create; - ret = register_pernet_subsys(&fib6_net_ops); if (ret) - goto out_rt6_stats; + goto out_kmem_cache_create; ret = __rtnl_register(PF_INET6, RTM_GETROUTE, NULL, inet6_dump_fib); if (ret) @@ -1573,8 +1574,6 @@ out: out_unregister_subsys: unregister_pernet_subsys(&fib6_net_ops); -out_rt6_stats: - kfree(rt6_stats); out_kmem_cache_create: kmem_cache_destroy(fib6_node_kmem); goto out; @@ -1583,6 +1582,5 @@ out_kmem_cache_create: void fib6_gc_cleanup(void) { unregister_pernet_subsys(&fib6_net_ops); - kfree(rt6_stats); kmem_cache_destroy(fib6_node_kmem); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 370f2aef6909..3afc3f41f2d5 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2354,11 +2354,13 @@ static const struct file_operations ipv6_route_proc_fops = { static int rt6_stats_seq_show(struct seq_file *seq, void *v) { seq_printf(seq, "%04x %04x %04x %04x %04x %04x %04x\n", - rt6_stats->fib_nodes, rt6_stats->fib_route_nodes, - rt6_stats->fib_rt_alloc, rt6_stats->fib_rt_entries, - rt6_stats->fib_rt_cache, - atomic_read(&ip6_dst_ops.entries), - rt6_stats->fib_discarded_routes); + init_net.ipv6.rt6_stats->fib_nodes, + init_net.ipv6.rt6_stats->fib_route_nodes, + init_net.ipv6.rt6_stats->fib_rt_alloc, + init_net.ipv6.rt6_stats->fib_rt_entries, + init_net.ipv6.rt6_stats->fib_rt_cache, + atomic_read(&ip6_dst_ops.entries), + init_net.ipv6.rt6_stats->fib_discarded_routes); return 0; } -- cgit v1.2.3 From 606a2b4862d4be31fa55cad89871fe52a422d511 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 4 Mar 2008 13:45:59 -0800 Subject: [NETNS][IPV6] route6 - Pass the network namespace parameter to rt6_lookup Add a network namespace parameter to rt6_lookup(). Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_route.h | 3 ++- net/ipv6/addrconf.c | 4 ++-- net/ipv6/anycast.c | 2 +- net/ipv6/ip6_tunnel.c | 4 ++-- net/ipv6/mcast.c | 4 ++-- net/ipv6/route.c | 10 +++++----- net/ipv6/sit.c | 4 ++-- 7 files changed, 16 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 79dce496f4d4..92004c5160c7 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -69,7 +69,8 @@ extern void rt6_sndmsg(int type, struct in6_addr *dst, int dstlen, int srclen, int metric, __u32 flags); -extern struct rt6_info *rt6_lookup(struct in6_addr *daddr, +extern struct rt6_info *rt6_lookup(struct net *net, + struct in6_addr *daddr, struct in6_addr *saddr, int oif, int flags); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index a1d872dacad6..9d894e8c7b72 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -753,7 +753,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) struct rt6_info *rt; ipv6_addr_prefix(&prefix, &ifp->addr, ifp->prefix_len); - rt = rt6_lookup(&prefix, NULL, ifp->idev->dev->ifindex, 1); + rt = rt6_lookup(&init_net, &prefix, NULL, ifp->idev->dev->ifindex, 1); if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) { if (onlink == 0) { @@ -1700,7 +1700,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len) if (pinfo->onlink) { struct rt6_info *rt; - rt = rt6_lookup(&pinfo->prefix, NULL, dev->ifindex, 1); + rt = rt6_lookup(&init_net, &pinfo->prefix, NULL, dev->ifindex, 1); if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) { if (rt->rt6i_flags&RTF_EXPIRES) { diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 9c7f83fbc3a1..96868b994b37 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -101,7 +101,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr) if (ifindex == 0) { struct rt6_info *rt; - rt = rt6_lookup(addr, NULL, 0, 0); + rt = rt6_lookup(&init_net, addr, NULL, 0, 0); if (rt) { dev = rt->rt6i_dev; dev_hold(dev); diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 78f438880923..4e1981660b3c 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -602,7 +602,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, skb_reset_network_header(skb2); /* Try to guess incoming interface */ - rt = rt6_lookup(&ipv6_hdr(skb2)->saddr, NULL, 0, 0); + rt = rt6_lookup(&init_net, &ipv6_hdr(skb2)->saddr, NULL, 0, 0); if (rt && rt->rt6i_dev) skb2->dev = rt->rt6i_dev; @@ -1112,7 +1112,7 @@ static void ip6_tnl_link_config(struct ip6_tnl *t) int strict = (ipv6_addr_type(&p->raddr) & (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL)); - struct rt6_info *rt = rt6_lookup(&p->raddr, &p->laddr, + struct rt6_info *rt = rt6_lookup(&init_net, &p->raddr, &p->laddr, p->link, strict); if (rt == NULL) diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index a373b8e7f241..197ca390a15d 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -208,7 +208,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr) if (ifindex == 0) { struct rt6_info *rt; - rt = rt6_lookup(addr, NULL, 0, 0); + rt = rt6_lookup(&init_net, addr, NULL, 0, 0); if (rt) { dev = rt->rt6i_dev; dev_hold(dev); @@ -294,7 +294,7 @@ static struct inet6_dev *ip6_mc_find_dev(struct in6_addr *group, int ifindex) if (ifindex == 0) { struct rt6_info *rt; - rt = rt6_lookup(group, NULL, 0, 0); + rt = rt6_lookup(&init_net, group, NULL, 0, 0); if (rt) { dev = rt->rt6i_dev; dev_hold(dev); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 5d9d293156cd..d9d840ced1c6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -553,8 +553,8 @@ out: } -struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, - int oif, int strict) +struct rt6_info *rt6_lookup(struct net *net, struct in6_addr *daddr, + struct in6_addr *saddr, int oif, int strict) { struct flowi fl = { .oif = oif, @@ -572,7 +572,7 @@ struct rt6_info *rt6_lookup(struct in6_addr *daddr, struct in6_addr *saddr, flags |= RT6_LOOKUP_F_HAS_SADDR; } - dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_lookup); + dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_lookup); if (dst->error == 0) return (struct rt6_info *) dst; @@ -1159,7 +1159,7 @@ int ip6_route_add(struct fib6_config *cfg) if (!(gwa_type&IPV6_ADDR_UNICAST)) goto out; - grt = rt6_lookup(gw_addr, NULL, cfg->fc_ifindex, 1); + grt = rt6_lookup(&init_net, gw_addr, NULL, cfg->fc_ifindex, 1); err = -EHOSTUNREACH; if (grt == NULL) @@ -1483,7 +1483,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, struct rt6_info *rt, *nrt; int allfrag = 0; - rt = rt6_lookup(daddr, saddr, dev->ifindex, 0); + rt = rt6_lookup(dev->nd_net, daddr, saddr, dev->ifindex, 0); if (rt == NULL) return; diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 1656c003b989..68720aa63f96 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -339,11 +339,11 @@ out: skb_reset_network_header(skb2); /* Try to guess incoming interface */ - rt6i = rt6_lookup(&iph6->saddr, NULL, NULL, 0); + rt6i = rt6_lookup(&init_net, &iph6->saddr, NULL, NULL, 0); if (rt6i && rt6i->rt6i_dev) { skb2->dev = rt6i->rt6i_dev; - rt6i = rt6_lookup(&iph6->daddr, &iph6->saddr, NULL, 0); + rt6i = rt6_lookup(&init_net, &iph6->daddr, &iph6->saddr, NULL, 0); if (rt6i && rt6i->rt6i_dev && rt6i->rt6i_dev->type == ARPHRD_SIT) { struct ip_tunnel *t = netdev_priv(rt6i->rt6i_dev); -- cgit v1.2.3 From 7b4da53229bb61469bdab321384b9a13406e3485 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 4 Mar 2008 13:47:14 -0800 Subject: [NETNS][IPV6] route6 - Pass the network namespace parameter to rt6_purge_dflt_routers Add a network namespace parameter to rt6_purge_dflt_routers. This is needed to call fib6_get_table with the appropriate network namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_route.h | 2 +- net/ipv6/addrconf.c | 2 +- net/ipv6/route.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 92004c5160c7..1444d358836d 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -95,7 +95,7 @@ extern struct rt6_info * rt6_add_dflt_router(struct in6_addr *gwaddr, struct net_device *dev, unsigned int pref); -extern void rt6_purge_dflt_routers(void); +extern void rt6_purge_dflt_routers(struct net *net); extern int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 9d894e8c7b72..b37ae421b61b 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -493,7 +493,7 @@ static void addrconf_fixup_forwarding(struct ctl_table *table, int *p, int old) dev_forward_change((struct inet6_dev *)table->extra1); if (*p) - rt6_purge_dflt_routers(); + rt6_purge_dflt_routers(net); } #endif diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 4278cec522c5..ad3d684e544a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1697,13 +1697,13 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr, return rt6_get_dflt_router(gwaddr, dev); } -void rt6_purge_dflt_routers(void) +void rt6_purge_dflt_routers(struct net *net) { struct rt6_info *rt; struct fib6_table *table; /* NOTE: Keep consistent with rt6_get_dflt_router */ - table = fib6_get_table(&init_net, RT6_TABLE_DFLT); + table = fib6_get_table(net, RT6_TABLE_DFLT); if (table == NULL) return; -- cgit v1.2.3 From 5578689a4e3c04f2d43ea39736fd3fa396d80c6e Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 4 Mar 2008 13:47:47 -0800 Subject: [NETNS][IPV6] route6 - make route6 per namespace This patch makes the routing engine use the network namespaces to access routing informations: Add a network namespace parameter to ipv6_route_ioctl and propagate the network namespace value to all the routing code that have not yet been changed. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_route.h | 4 +- net/ipv6/af_inet6.c | 3 +- net/ipv6/route.c | 98 ++++++++++++++++++++++++------------------------- 3 files changed, 54 insertions(+), 51 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 1444d358836d..2bcbfb826530 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -49,7 +49,9 @@ extern struct dst_entry * ip6_route_output(struct sock *sk, extern int ip6_route_init(void); extern void ip6_route_cleanup(void); -extern int ipv6_route_ioctl(unsigned int cmd, void __user *arg); +extern int ipv6_route_ioctl(struct net *net, + unsigned int cmd, + void __user *arg); extern int ip6_route_add(struct fib6_config *cfg); extern int ip6_ins_rt(struct rt6_info *); diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 73021d5baece..60b8a2254046 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -440,6 +440,7 @@ EXPORT_SYMBOL(inet6_getname); int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { struct sock *sk = sock->sk; + struct net *net = sk->sk_net; switch(cmd) { @@ -452,7 +453,7 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCADDRT: case SIOCDELRT: - return(ipv6_route_ioctl(cmd,(void __user *)arg)); + return(ipv6_route_ioctl(net, cmd, (void __user *)arg)); case SIOCSIFADDR: return addrconf_add_ifaddr((void __user *) arg); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index ad3d684e544a..b3ac4901af86 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -609,7 +609,7 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info) int ip6_ins_rt(struct rt6_info *rt) { struct nl_info info = { - .nl_net = &init_net, + .nl_net = rt->rt6i_dev->nd_net, }; return __ip6_ins_rt(rt, &info); } @@ -746,6 +746,7 @@ static struct rt6_info *ip6_pol_route_input(struct fib6_table *table, void ip6_route_input(struct sk_buff *skb) { struct ipv6hdr *iph = ipv6_hdr(skb); + struct net *net = skb->dev->nd_net; int flags = RT6_LOOKUP_F_HAS_SADDR; struct flowi fl = { .iif = skb->dev->ifindex, @@ -763,7 +764,7 @@ void ip6_route_input(struct sk_buff *skb) if (rt6_need_strict(&iph->daddr)) flags |= RT6_LOOKUP_F_IFACE; - skb->dst = fib6_rule_lookup(&init_net, &fl, flags, ip6_pol_route_input); + skb->dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input); } static struct rt6_info *ip6_pol_route_output(struct fib6_table *table, @@ -891,12 +892,12 @@ static void ip6_rt_update_pmtu(struct dst_entry *dst, u32 mtu) static int ipv6_get_mtu(struct net_device *dev); -static inline unsigned int ipv6_advmss(unsigned int mtu) +static inline unsigned int ipv6_advmss(struct net *net, unsigned int mtu) { mtu -= sizeof(struct ipv6hdr) + sizeof(struct tcphdr); - if (mtu < init_net.ipv6.sysctl.ip6_rt_min_advmss) - mtu = init_net.ipv6.sysctl.ip6_rt_min_advmss; + if (mtu < net->ipv6.sysctl.ip6_rt_min_advmss) + mtu = net->ipv6.sysctl.ip6_rt_min_advmss; /* * Maximal non-jumbo IPv6 payload is IPV6_MAXPLEN and @@ -918,6 +919,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, { struct rt6_info *rt; struct inet6_dev *idev = in6_dev_get(dev); + struct net *net = dev->nd_net; if (unlikely(idev == NULL)) return NULL; @@ -940,7 +942,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, atomic_set(&rt->u.dst.__refcnt, 1); rt->u.dst.metrics[RTAX_HOPLIMIT-1] = 255; rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev); - rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst)); + rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst)); rt->u.dst.output = ip6_output; #if 0 /* there's no chance to use these for ndisc */ @@ -956,7 +958,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, icmp6_dst_gc_list = &rt->u.dst; spin_unlock_bh(&icmp6_dst_lock); - fib6_force_start_gc(dev->nd_net); + fib6_force_start_gc(net); out: return &rt->u.dst; @@ -1049,6 +1051,7 @@ int ipv6_get_hoplimit(struct net_device *dev) int ip6_route_add(struct fib6_config *cfg) { int err; + struct net *net = cfg->fc_nlinfo.nl_net; struct rt6_info *rt = NULL; struct net_device *dev = NULL; struct inet6_dev *idev = NULL; @@ -1063,7 +1066,7 @@ int ip6_route_add(struct fib6_config *cfg) #endif if (cfg->fc_ifindex) { err = -ENODEV; - dev = dev_get_by_index(&init_net, cfg->fc_ifindex); + dev = dev_get_by_index(net, cfg->fc_ifindex); if (!dev) goto out; idev = in6_dev_get(dev); @@ -1074,7 +1077,7 @@ int ip6_route_add(struct fib6_config *cfg) if (cfg->fc_metric == 0) cfg->fc_metric = IP6_RT_PRIO_USER; - table = fib6_new_table(&init_net, cfg->fc_table); + table = fib6_new_table(net, cfg->fc_table); if (table == NULL) { err = -ENOBUFS; goto out; @@ -1121,12 +1124,12 @@ int ip6_route_add(struct fib6_config *cfg) if ((cfg->fc_flags & RTF_REJECT) || (dev && (dev->flags&IFF_LOOPBACK) && !(addr_type&IPV6_ADDR_LOOPBACK))) { /* hold loopback dev/idev if we haven't done so. */ - if (dev != init_net.loopback_dev) { + if (dev != net->loopback_dev) { if (dev) { dev_put(dev); in6_dev_put(idev); } - dev = init_net.loopback_dev; + dev = net->loopback_dev; dev_hold(dev); idev = in6_dev_get(dev); if (!idev) { @@ -1163,7 +1166,7 @@ int ip6_route_add(struct fib6_config *cfg) if (!(gwa_type&IPV6_ADDR_UNICAST)) goto out; - grt = rt6_lookup(&init_net, gw_addr, NULL, cfg->fc_ifindex, 1); + grt = rt6_lookup(net, gw_addr, NULL, cfg->fc_ifindex, 1); err = -EHOSTUNREACH; if (grt == NULL) @@ -1230,7 +1233,7 @@ install_route: if (!rt->u.dst.metrics[RTAX_MTU-1]) rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(dev); if (!rt->u.dst.metrics[RTAX_ADVMSS-1]) - rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst)); + rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst)); rt->u.dst.dev = dev; rt->rt6i_idev = idev; rt->rt6i_table = table; @@ -1271,7 +1274,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) int ip6_del_rt(struct rt6_info *rt) { struct nl_info info = { - .nl_net = &init_net, + .nl_net = rt->rt6i_dev->nd_net, }; return __ip6_del_rt(rt, &info); } @@ -1283,7 +1286,7 @@ static int ip6_route_del(struct fib6_config *cfg) struct rt6_info *rt; int err = -ESRCH; - table = fib6_get_table(&init_net, cfg->fc_table); + table = fib6_get_table(cfg->fc_nlinfo.nl_net, cfg->fc_table); if (table == NULL) return err; @@ -1382,6 +1385,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, struct net_device *dev) { int flags = RT6_LOOKUP_F_HAS_SADDR; + struct net *net = dev->nd_net; struct ip6rd_flowi rdfl = { .fl = { .oif = dev->ifindex, @@ -1398,8 +1402,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, if (rt6_need_strict(dest)) flags |= RT6_LOOKUP_F_IFACE; - return (struct rt6_info *)fib6_rule_lookup(&init_net, - (struct flowi *)&rdfl, + return (struct rt6_info *)fib6_rule_lookup(net, (struct flowi *)&rdfl, flags, __ip6_route_redirect); } @@ -1457,7 +1460,8 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, nrt->rt6i_nexthop = neigh_clone(neigh); /* Reset pmtu, it may be better */ nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev); - nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&nrt->u.dst)); + nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(neigh->dev->nd_net, + dst_mtu(&nrt->u.dst)); if (ip6_ins_rt(nrt)) goto out; @@ -1485,9 +1489,10 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, struct net_device *dev, u32 pmtu) { struct rt6_info *rt, *nrt; + struct net *net = dev->nd_net; int allfrag = 0; - rt = rt6_lookup(dev->nd_net, daddr, saddr, dev->ifindex, 0); + rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0); if (rt == NULL) return; @@ -1520,7 +1525,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, rt->u.dst.metrics[RTAX_MTU-1] = pmtu; if (allfrag) rt->u.dst.metrics[RTAX_FEATURES-1] |= RTAX_FEATURE_ALLFRAG; - dst_set_expires(&rt->u.dst, init_net.ipv6.sysctl.ip6_rt_mtu_expires); + dst_set_expires(&rt->u.dst, net->ipv6.sysctl.ip6_rt_mtu_expires); rt->rt6i_flags |= RTF_MODIFIED|RTF_EXPIRES; goto out; } @@ -1546,7 +1551,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, * which is 10 mins. After 10 mins the decreased pmtu is expired * and detecting PMTU increase will be automatically happened. */ - dst_set_expires(&nrt->u.dst, init_net.ipv6.sysctl.ip6_rt_mtu_expires); + dst_set_expires(&nrt->u.dst, net->ipv6.sysctl.ip6_rt_mtu_expires); nrt->rt6i_flags |= RTF_DYNAMIC|RTF_EXPIRES; ip6_ins_rt(nrt); @@ -1659,7 +1664,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d struct rt6_info *rt; struct fib6_table *table; - table = fib6_get_table(&init_net, RT6_TABLE_DFLT); + table = fib6_get_table(dev->nd_net, RT6_TABLE_DFLT); if (table == NULL) return NULL; @@ -1688,6 +1693,9 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr, .fc_ifindex = dev->ifindex, .fc_flags = RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT | RTF_UP | RTF_EXPIRES | RTF_PREF(pref), + .fc_nlinfo.pid = 0, + .fc_nlinfo.nlh = NULL, + .fc_nlinfo.nl_net = dev->nd_net, }; ipv6_addr_copy(&cfg.fc_gateway, gwaddr); @@ -1720,7 +1728,8 @@ restart: read_unlock_bh(&table->tb6_lock); } -static void rtmsg_to_fib6_config(struct in6_rtmsg *rtmsg, +static void rtmsg_to_fib6_config(struct net *net, + struct in6_rtmsg *rtmsg, struct fib6_config *cfg) { memset(cfg, 0, sizeof(*cfg)); @@ -1733,14 +1742,14 @@ static void rtmsg_to_fib6_config(struct in6_rtmsg *rtmsg, cfg->fc_src_len = rtmsg->rtmsg_src_len; cfg->fc_flags = rtmsg->rtmsg_flags; - cfg->fc_nlinfo.nl_net = &init_net; + cfg->fc_nlinfo.nl_net = net; ipv6_addr_copy(&cfg->fc_dst, &rtmsg->rtmsg_dst); ipv6_addr_copy(&cfg->fc_src, &rtmsg->rtmsg_src); ipv6_addr_copy(&cfg->fc_gateway, &rtmsg->rtmsg_gateway); } -int ipv6_route_ioctl(unsigned int cmd, void __user *arg) +int ipv6_route_ioctl(struct net *net, unsigned int cmd, void __user *arg) { struct fib6_config cfg; struct in6_rtmsg rtmsg; @@ -1756,7 +1765,7 @@ int ipv6_route_ioctl(unsigned int cmd, void __user *arg) if (err) return -EFAULT; - rtmsg_to_fib6_config(&rtmsg, &cfg); + rtmsg_to_fib6_config(net, &rtmsg, &cfg); rtnl_lock(); switch (cmd) { @@ -1835,21 +1844,22 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, const struct in6_addr *addr, int anycast) { + struct net *net = idev->dev->nd_net; struct rt6_info *rt = ip6_dst_alloc(); if (rt == NULL) return ERR_PTR(-ENOMEM); - dev_hold(init_net.loopback_dev); + dev_hold(net->loopback_dev); in6_dev_hold(idev); rt->u.dst.flags = DST_HOST; rt->u.dst.input = ip6_input; rt->u.dst.output = ip6_output; - rt->rt6i_dev = init_net.loopback_dev; + rt->rt6i_dev = net->loopback_dev; rt->rt6i_idev = idev; rt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(rt->rt6i_dev); - rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dst_mtu(&rt->u.dst)); + rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, dst_mtu(&rt->u.dst)); rt->u.dst.metrics[RTAX_HOPLIMIT-1] = -1; rt->u.dst.obsolete = -1; @@ -1866,7 +1876,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, ipv6_addr_copy(&rt->rt6i_dst.addr, addr); rt->rt6i_dst.plen = 128; - rt->rt6i_table = fib6_get_table(&init_net, RT6_TABLE_LOCAL); + rt->rt6i_table = fib6_get_table(net, RT6_TABLE_LOCAL); atomic_set(&rt->u.dst.__refcnt, 1); @@ -1898,6 +1908,7 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg) { struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg; struct inet6_dev *idev; + struct net *net = arg->dev->nd_net; /* In IPv6 pmtu discovery is not optional, so that RTAX_MTU lock cannot disable it. @@ -1929,7 +1940,7 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg) (dst_mtu(&rt->u.dst) < arg->mtu && dst_mtu(&rt->u.dst) == idev->cnf.mtu6))) { rt->u.dst.metrics[RTAX_MTU-1] = arg->mtu; - rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(arg->mtu); + rt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(net, arg->mtu); } return 0; } @@ -2024,13 +2035,9 @@ errout: static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; struct fib6_config cfg; int err; - if (net != &init_net) - return -EINVAL; - err = rtm_to_fib6_config(skb, nlh, &cfg); if (err < 0) return err; @@ -2040,13 +2047,9 @@ static int inet6_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *a static int inet6_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; struct fib6_config cfg; int err; - if (net != &init_net) - return -EINVAL; - err = rtm_to_fib6_config(skb, nlh, &cfg); if (err < 0) return err; @@ -2190,9 +2193,6 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void struct flowi fl; int err, iif = 0; - if (net != &init_net) - return -EINVAL; - err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv6_policy); if (err < 0) goto errout; @@ -2222,7 +2222,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void if (iif) { struct net_device *dev; - dev = __dev_get_by_index(&init_net, iif); + dev = __dev_get_by_index(net, iif); if (!dev) { err = -ENODEV; goto errout; @@ -2252,7 +2252,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void goto errout; } - err = rtnl_unicast(skb, &init_net, NETLINK_CB(in_skb).pid); + err = rtnl_unicast(skb, net, NETLINK_CB(in_skb).pid); errout: return err; } @@ -2260,6 +2260,7 @@ errout: void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) { struct sk_buff *skb; + struct net *net = info->nl_net; u32 seq; int err; @@ -2278,11 +2279,11 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) kfree_skb(skb); goto errout; } - err = rtnl_notify(skb, &init_net, info->pid, - RTNLGRP_IPV6_ROUTE, info->nlh, gfp_any()); + err = rtnl_notify(skb, net, info->pid, RTNLGRP_IPV6_ROUTE, + info->nlh, gfp_any()); errout: if (err < 0) - rtnl_set_sk_err(&init_net, RTNLGRP_IPV6_ROUTE, err); + rtnl_set_sk_err(net, RTNLGRP_IPV6_ROUTE, err); } /* @@ -2544,6 +2545,7 @@ static void ip6_route_net_exit(struct net *net) proc_net_remove(net, "ipv6_route"); proc_net_remove(net, "rt6_stats"); #endif + rt6_ifdown(net, NULL); } static struct pernet_operations ip6_route_net_ops = { @@ -2592,7 +2594,6 @@ fib6_rules_init: xfrm6_init: xfrm6_fini(); out_fib6_init: - rt6_ifdown(&init_net, NULL); fib6_gc_cleanup(); out_kmem_cache: kmem_cache_destroy(ip6_dst_ops.kmem_cachep); @@ -2604,7 +2605,6 @@ void ip6_route_cleanup(void) unregister_pernet_subsys(&ip6_route_net_ops); fib6_rules_cleanup(); xfrm6_fini(); - rt6_ifdown(&init_net, NULL); fib6_gc_cleanup(); kmem_cache_destroy(ip6_dst_ops.kmem_cachep); } -- cgit v1.2.3 From bdb3289f739e94bcae8b51972ae844ec66c2f4df Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 4 Mar 2008 13:48:10 -0800 Subject: [NETNS][IPV6] rt6_info - make rt6_info accessed as a pointer This patch make mindless changes and prepares the code to use dynamic allocation for rt6_info structure. The code accesses the rt6_info structure as a pointer instead of a global static variable. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_route.h | 6 ++--- net/ipv6/addrconf.c | 12 ++++----- net/ipv6/fib6_rules.c | 12 ++++----- net/ipv6/ip6_fib.c | 18 ++++++------- net/ipv6/route.c | 70 ++++++++++++++++++++++++++++++++++++++----------- 5 files changed, 78 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 2bcbfb826530..e0caed25bfba 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -34,11 +34,11 @@ struct route_info { #define RT6_LOOKUP_F_REACHABLE 0x2 #define RT6_LOOKUP_F_HAS_SADDR 0x4 -extern struct rt6_info ip6_null_entry; +extern struct rt6_info *ip6_null_entry; #ifdef CONFIG_IPV6_MULTIPLE_TABLES -extern struct rt6_info ip6_prohibit_entry; -extern struct rt6_info ip6_blk_hole_entry; +extern struct rt6_info *ip6_prohibit_entry; +extern struct rt6_info *ip6_blk_hole_entry; #endif extern void ip6_route_input(struct sk_buff *skb); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index b37ae421b61b..3192a848a53a 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4301,13 +4301,13 @@ int __init addrconf_init(void) if (err) goto errlo; - ip6_null_entry.u.dst.dev = init_net.loopback_dev; - ip6_null_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev); + ip6_null_entry->u.dst.dev = init_net.loopback_dev; + ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); #ifdef CONFIG_IPV6_MULTIPLE_TABLES - ip6_prohibit_entry.u.dst.dev = init_net.loopback_dev; - ip6_prohibit_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev); - ip6_blk_hole_entry.u.dst.dev = init_net.loopback_dev; - ip6_blk_hole_entry.rt6i_idev = in6_dev_get(init_net.loopback_dev); + ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev; + ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); + ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev; + ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); #endif register_netdevice_notifier(&ipv6_dev_notf); diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 89cb092c9732..c00055f232c4 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -43,8 +43,8 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, if (arg.result) return arg.result; - dst_hold(&ip6_null_entry.u.dst); - return &ip6_null_entry.u.dst; + dst_hold(&ip6_null_entry->u.dst); + return &ip6_null_entry->u.dst; } static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, @@ -58,14 +58,14 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, case FR_ACT_TO_TBL: break; case FR_ACT_UNREACHABLE: - rt = &ip6_null_entry; + rt = ip6_null_entry; goto discard_pkt; default: case FR_ACT_BLACKHOLE: - rt = &ip6_blk_hole_entry; + rt = ip6_blk_hole_entry; goto discard_pkt; case FR_ACT_PROHIBIT: - rt = &ip6_prohibit_entry; + rt = ip6_prohibit_entry; goto discard_pkt; } @@ -73,7 +73,7 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, if (table) rt = lookup(table, flp, flags); - if (rt != &ip6_null_entry) { + if (rt != ip6_null_entry) { struct fib6_rule *r = (struct fib6_rule *)rule; /* diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 1c2566e3d392..f028f7a1a446 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -200,7 +200,7 @@ static struct fib6_table *fib6_alloc_table(u32 id) table = kzalloc(sizeof(*table), GFP_ATOMIC); if (table != NULL) { table->tb6_id = id; - table->tb6_root.leaf = &ip6_null_entry; + table->tb6_root.leaf = ip6_null_entry; table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; } @@ -717,8 +717,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info) if (sfn == NULL) goto st_failure; - sfn->leaf = &ip6_null_entry; - atomic_inc(&ip6_null_entry.rt6i_ref); + sfn->leaf = ip6_null_entry; + atomic_inc(&ip6_null_entry->rt6i_ref); sfn->fn_flags = RTN_ROOT; sfn->fn_sernum = fib6_new_sernum(); @@ -777,7 +777,7 @@ out: #if RT6_DEBUG >= 2 if (!pn->leaf) { BUG_TRAP(pn->leaf != NULL); - pn->leaf = &ip6_null_entry; + pn->leaf = ip6_null_entry; } #endif atomic_inc(&pn->leaf->rt6i_ref); @@ -962,7 +962,7 @@ struct fib6_node * fib6_locate(struct fib6_node *root, static struct rt6_info * fib6_find_prefix(struct fib6_node *fn) { if (fn->fn_flags&RTN_ROOT) - return &ip6_null_entry; + return ip6_null_entry; while(fn) { if(fn->left) @@ -1012,7 +1012,7 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn) #if RT6_DEBUG >= 2 if (fn->leaf==NULL) { BUG_TRAP(fn->leaf); - fn->leaf = &ip6_null_entry; + fn->leaf = ip6_null_entry; } #endif atomic_inc(&fn->leaf->rt6i_ref); @@ -1154,7 +1154,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info) return -ENOENT; } #endif - if (fn == NULL || rt == &ip6_null_entry) + if (fn == NULL || rt == ip6_null_entry) return -ENOENT; BUG_TRAP(fn->fn_flags&RTN_RTINFO); @@ -1501,7 +1501,7 @@ static int fib6_net_init(struct net *net) goto out_fib_table_hash; net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN; - net->ipv6.fib6_main_tbl->tb6_root.leaf = &ip6_null_entry; + net->ipv6.fib6_main_tbl->tb6_root.leaf = ip6_null_entry; net->ipv6.fib6_main_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; @@ -1511,7 +1511,7 @@ static int fib6_net_init(struct net *net) if (!net->ipv6.fib6_local_tbl) goto out_fib6_main_tbl; net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL; - net->ipv6.fib6_local_tbl->tb6_root.leaf = &ip6_null_entry; + net->ipv6.fib6_local_tbl->tb6_root.leaf = ip6_null_entry; net->ipv6.fib6_local_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; #endif diff --git a/net/ipv6/route.c b/net/ipv6/route.c index b3ac4901af86..8f954c1e961f 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -127,7 +127,7 @@ static struct dst_ops ip6_dst_blackhole_ops = { .entries = ATOMIC_INIT(0), }; -struct rt6_info ip6_null_entry = { +static struct rt6_info ip6_null_entry_template = { .u = { .dst = { .__refcnt = ATOMIC_INIT(1), @@ -138,7 +138,6 @@ struct rt6_info ip6_null_entry = { .input = ip6_pkt_discard, .output = ip6_pkt_discard_out, .ops = &ip6_dst_ops, - .path = (struct dst_entry*)&ip6_null_entry, } }, .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), @@ -146,12 +145,14 @@ struct rt6_info ip6_null_entry = { .rt6i_ref = ATOMIC_INIT(1), }; +struct rt6_info *ip6_null_entry; + #ifdef CONFIG_IPV6_MULTIPLE_TABLES static int ip6_pkt_prohibit(struct sk_buff *skb); static int ip6_pkt_prohibit_out(struct sk_buff *skb); -struct rt6_info ip6_prohibit_entry = { +struct rt6_info ip6_prohibit_entry_template = { .u = { .dst = { .__refcnt = ATOMIC_INIT(1), @@ -162,7 +163,6 @@ struct rt6_info ip6_prohibit_entry = { .input = ip6_pkt_prohibit, .output = ip6_pkt_prohibit_out, .ops = &ip6_dst_ops, - .path = (struct dst_entry*)&ip6_prohibit_entry, } }, .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), @@ -170,7 +170,9 @@ struct rt6_info ip6_prohibit_entry = { .rt6i_ref = ATOMIC_INIT(1), }; -struct rt6_info ip6_blk_hole_entry = { +struct rt6_info *ip6_prohibit_entry; + +static struct rt6_info ip6_blk_hole_entry_template = { .u = { .dst = { .__refcnt = ATOMIC_INIT(1), @@ -181,7 +183,6 @@ struct rt6_info ip6_blk_hole_entry = { .input = dst_discard, .output = dst_discard, .ops = &ip6_dst_ops, - .path = (struct dst_entry*)&ip6_blk_hole_entry, } }, .rt6i_flags = (RTF_REJECT | RTF_NONEXTHOP), @@ -189,6 +190,8 @@ struct rt6_info ip6_blk_hole_entry = { .rt6i_ref = ATOMIC_INIT(1), }; +struct rt6_info *ip6_blk_hole_entry; + #endif /* allocate dst with ip6_dst_ops */ @@ -271,7 +274,7 @@ static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt, return local; if (strict) - return &ip6_null_entry; + return ip6_null_entry; } return rt; } @@ -437,7 +440,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) RT6_TRACE("%s() => %p\n", __FUNCTION__, match); - return (match ? match : &ip6_null_entry); + return (match ? match : ip6_null_entry); } #ifdef CONFIG_IPV6_ROUTE_INFO @@ -522,7 +525,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, #define BACKTRACK(saddr) \ do { \ - if (rt == &ip6_null_entry) { \ + if (rt == ip6_null_entry) { \ struct fib6_node *pn; \ while (1) { \ if (fn->fn_flags & RTN_TL_ROOT) \ @@ -686,7 +689,7 @@ restart_2: restart: rt = rt6_select(fn, oif, strict | reachable); BACKTRACK(&fl->fl6_src); - if (rt == &ip6_null_entry || + if (rt == ip6_null_entry || rt->rt6i_flags & RTF_CACHE) goto out; @@ -704,7 +707,7 @@ restart: } dst_release(&rt->u.dst); - rt = nrt ? : &ip6_null_entry; + rt = nrt ? : ip6_null_entry; dst_hold(&rt->u.dst); if (nrt) { @@ -1257,7 +1260,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) int err; struct fib6_table *table; - if (rt == &ip6_null_entry) + if (rt == ip6_null_entry) return -ENOENT; table = rt->rt6i_table; @@ -1369,7 +1372,7 @@ restart: } if (!rt) - rt = &ip6_null_entry; + rt = ip6_null_entry; BACKTRACK(&fl->fl6_src); out: dst_hold(&rt->u.dst); @@ -1415,7 +1418,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, rt = ip6_route_redirect(dest, src, saddr, neigh->dev); - if (rt == &ip6_null_entry) { + if (rt == ip6_null_entry) { if (net_ratelimit()) printk(KERN_DEBUG "rt6_redirect: source isn't a valid nexthop " "for redirect target\n"); @@ -1886,7 +1889,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, static int fib6_ifdown(struct rt6_info *rt, void *arg) { if (((void*)rt->rt6i_dev == arg || arg == NULL) && - rt != &ip6_null_entry) { + rt != ip6_null_entry) { RT6_TRACE("deleted by ifdown %p\n", rt); return -1; } @@ -2565,9 +2568,30 @@ int __init ip6_route_init(void) ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops.kmem_cachep; + ret = -ENOMEM; + ip6_null_entry = kmemdup(&ip6_null_entry_template, + sizeof(*ip6_null_entry), GFP_KERNEL); + if (!ip6_null_entry) + goto out_kmem_cache; + ip6_null_entry->u.dst.path = (struct dst_entry *)ip6_null_entry; + +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template, + sizeof(*ip6_prohibit_entry), GFP_KERNEL); + if (!ip6_prohibit_entry) + goto out_ip6_null_entry; + ip6_prohibit_entry->u.dst.path = (struct dst_entry *)ip6_prohibit_entry; + + ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template, + sizeof(*ip6_blk_hole_entry), GFP_KERNEL); + if (!ip6_blk_hole_entry) + goto out_ip6_prohibit_entry; + ip6_blk_hole_entry->u.dst.path = (struct dst_entry *)ip6_blk_hole_entry; +#endif + ret = fib6_init(); if (ret) - goto out_kmem_cache; + goto out_ip6_blk_hole_entry; ret = xfrm6_init(); if (ret) @@ -2595,6 +2619,14 @@ xfrm6_init: xfrm6_fini(); out_fib6_init: fib6_gc_cleanup(); +out_ip6_blk_hole_entry: +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + kfree(ip6_blk_hole_entry); +out_ip6_prohibit_entry: + kfree(ip6_prohibit_entry); +out_ip6_null_entry: +#endif + kfree(ip6_null_entry); out_kmem_cache: kmem_cache_destroy(ip6_dst_ops.kmem_cachep); goto out; @@ -2607,4 +2639,10 @@ void ip6_route_cleanup(void) xfrm6_fini(); fib6_gc_cleanup(); kmem_cache_destroy(ip6_dst_ops.kmem_cachep); + + kfree(ip6_null_entry); +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + kfree(ip6_prohibit_entry); + kfree(ip6_blk_hole_entry); +#endif } -- cgit v1.2.3 From 8ed677896752fff056f6cf3d7ce462adc6c464f0 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Tue, 4 Mar 2008 13:48:30 -0800 Subject: [NETNS][IPV6] rt6_info - move rt6_info structure inside the namespace The rt6_info structures are moved inside the network namespace structure. All references to these structures are now relative to the initial network namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_fib.h | 3 +- include/net/netns/ipv6.h | 3 + net/ipv6/addrconf.c | 9 --- net/ipv6/fib6_rules.c | 17 ++-- net/ipv6/ip6_fib.c | 45 ++++++----- net/ipv6/route.c | 202 +++++++++++++++++++++++++++++++---------------- 6 files changed, 171 insertions(+), 108 deletions(-) (limited to 'include') diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index fae267f65341..7c5c0f79168a 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -174,7 +174,8 @@ struct fib6_table { #define RT6_TABLE_LOCAL RT6_TABLE_MAIN #endif -typedef struct rt6_info *(*pol_lookup_t)(struct fib6_table *, +typedef struct rt6_info *(*pol_lookup_t)(struct net *, + struct fib6_table *, struct flowi *, int); /* diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index c6c9afff13ef..311a942f36e9 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -36,11 +36,14 @@ struct netns_ipv6 { struct xt_table *ip6table_mangle; struct xt_table *ip6table_raw; #endif + struct rt6_info *ip6_null_entry; struct rt6_statistics *rt6_stats; struct timer_list *ip6_fib_timer; struct hlist_head *fib_table_hash; struct fib6_table *fib6_main_tbl; #ifdef CONFIG_IPV6_MULTIPLE_TABLES + struct rt6_info *ip6_prohibit_entry; + struct rt6_info *ip6_blk_hole_entry; struct fib6_table *fib6_local_tbl; struct fib_rules_ops *fib6_rules_ops; #endif diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 3192a848a53a..17b06d220e95 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -4301,15 +4301,6 @@ int __init addrconf_init(void) if (err) goto errlo; - ip6_null_entry->u.dst.dev = init_net.loopback_dev; - ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); -#ifdef CONFIG_IPV6_MULTIPLE_TABLES - ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev; - ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); - ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev; - ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); -#endif - register_netdevice_notifier(&ipv6_dev_notf); addrconf_verify(0); diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index c00055f232c4..55137408f054 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -43,8 +43,8 @@ struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, if (arg.result) return arg.result; - dst_hold(&ip6_null_entry->u.dst); - return &ip6_null_entry->u.dst; + dst_hold(&net->ipv6.ip6_null_entry->u.dst); + return &net->ipv6.ip6_null_entry->u.dst; } static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, @@ -52,28 +52,29 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, { struct rt6_info *rt = NULL; struct fib6_table *table; + struct net *net = rule->fr_net; pol_lookup_t lookup = arg->lookup_ptr; switch (rule->action) { case FR_ACT_TO_TBL: break; case FR_ACT_UNREACHABLE: - rt = ip6_null_entry; + rt = net->ipv6.ip6_null_entry; goto discard_pkt; default: case FR_ACT_BLACKHOLE: - rt = ip6_blk_hole_entry; + rt = net->ipv6.ip6_blk_hole_entry; goto discard_pkt; case FR_ACT_PROHIBIT: - rt = ip6_prohibit_entry; + rt = net->ipv6.ip6_prohibit_entry; goto discard_pkt; } - table = fib6_get_table(rule->fr_net, rule->table); + table = fib6_get_table(net, rule->table); if (table) - rt = lookup(table, flp, flags); + rt = lookup(net, table, flp, flags); - if (rt != ip6_null_entry) { + if (rt != net->ipv6.ip6_null_entry) { struct fib6_rule *r = (struct fib6_rule *)rule; /* diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index f028f7a1a446..b0814b0082e7 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -79,8 +79,8 @@ static DEFINE_RWLOCK(fib6_walker_lock); static void fib6_prune_clones(struct net *net, struct fib6_node *fn, struct rt6_info *rt); -static struct rt6_info * fib6_find_prefix(struct fib6_node *fn); -static struct fib6_node * fib6_repair_tree(struct fib6_node *fn); +static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn); +static struct fib6_node *fib6_repair_tree(struct net *net, struct fib6_node *fn); static int fib6_walk(struct fib6_walker_t *w); static int fib6_walk_continue(struct fib6_walker_t *w); @@ -193,14 +193,14 @@ static void fib6_link_table(struct net *net, struct fib6_table *tb) #ifdef CONFIG_IPV6_MULTIPLE_TABLES -static struct fib6_table *fib6_alloc_table(u32 id) +static struct fib6_table *fib6_alloc_table(struct net *net, u32 id) { struct fib6_table *table; table = kzalloc(sizeof(*table), GFP_ATOMIC); if (table != NULL) { table->tb6_id = id; - table->tb6_root.leaf = ip6_null_entry; + table->tb6_root.leaf = net->ipv6.ip6_null_entry; table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; } @@ -217,7 +217,7 @@ struct fib6_table *fib6_new_table(struct net *net, u32 id) if (tb) return tb; - tb = fib6_alloc_table(id); + tb = fib6_alloc_table(net, id); if (tb != NULL) fib6_link_table(net, tb); @@ -267,7 +267,7 @@ struct fib6_table *fib6_get_table(struct net *net, u32 id) struct dst_entry *fib6_rule_lookup(struct net *net, struct flowi *fl, int flags, pol_lookup_t lookup) { - return (struct dst_entry *) lookup(net->ipv6.fib6_main_tbl, fl, flags); + return (struct dst_entry *) lookup(net, net->ipv6.fib6_main_tbl, fl, flags); } static void fib6_tables_init(struct net *net) @@ -717,8 +717,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info) if (sfn == NULL) goto st_failure; - sfn->leaf = ip6_null_entry; - atomic_inc(&ip6_null_entry->rt6i_ref); + sfn->leaf = info->nl_net->ipv6.ip6_null_entry; + atomic_inc(&info->nl_net->ipv6.ip6_null_entry->rt6i_ref); sfn->fn_flags = RTN_ROOT; sfn->fn_sernum = fib6_new_sernum(); @@ -773,11 +773,11 @@ out: * super-tree leaf node we have to find a new one for it. */ if (pn != fn && !pn->leaf && !(pn->fn_flags & RTN_RTINFO)) { - pn->leaf = fib6_find_prefix(pn); + pn->leaf = fib6_find_prefix(info->nl_net, pn); #if RT6_DEBUG >= 2 if (!pn->leaf) { BUG_TRAP(pn->leaf != NULL); - pn->leaf = ip6_null_entry; + pn->leaf = info->nl_net->ipv6.ip6_null_entry; } #endif atomic_inc(&pn->leaf->rt6i_ref); @@ -793,7 +793,7 @@ out: */ st_failure: if (fn && !(fn->fn_flags & (RTN_RTINFO|RTN_ROOT))) - fib6_repair_tree(fn); + fib6_repair_tree(info->nl_net, fn); dst_free(&rt->u.dst); return err; #endif @@ -959,10 +959,10 @@ struct fib6_node * fib6_locate(struct fib6_node *root, * */ -static struct rt6_info * fib6_find_prefix(struct fib6_node *fn) +static struct rt6_info *fib6_find_prefix(struct net *net, struct fib6_node *fn) { if (fn->fn_flags&RTN_ROOT) - return ip6_null_entry; + return net->ipv6.ip6_null_entry; while(fn) { if(fn->left) @@ -981,7 +981,8 @@ static struct rt6_info * fib6_find_prefix(struct fib6_node *fn) * is the node we want to try and remove. */ -static struct fib6_node * fib6_repair_tree(struct fib6_node *fn) +static struct fib6_node *fib6_repair_tree(struct net *net, + struct fib6_node *fn) { int children; int nstate; @@ -1008,11 +1009,11 @@ static struct fib6_node * fib6_repair_tree(struct fib6_node *fn) || (children && fn->fn_flags&RTN_ROOT) #endif ) { - fn->leaf = fib6_find_prefix(fn); + fn->leaf = fib6_find_prefix(net, fn); #if RT6_DEBUG >= 2 if (fn->leaf==NULL) { BUG_TRAP(fn->leaf); - fn->leaf = ip6_null_entry; + fn->leaf = net->ipv6.ip6_null_entry; } #endif atomic_inc(&fn->leaf->rt6i_ref); @@ -1117,7 +1118,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, if (fn->leaf == NULL) { fn->fn_flags &= ~RTN_RTINFO; net->ipv6.rt6_stats->fib_route_nodes--; - fn = fib6_repair_tree(fn); + fn = fib6_repair_tree(net, fn); } if (atomic_read(&rt->rt6i_ref) != 1) { @@ -1129,7 +1130,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, */ while (fn) { if (!(fn->fn_flags&RTN_RTINFO) && fn->leaf == rt) { - fn->leaf = fib6_find_prefix(fn); + fn->leaf = fib6_find_prefix(net, fn); atomic_inc(&fn->leaf->rt6i_ref); rt6_release(rt); } @@ -1145,6 +1146,7 @@ static void fib6_del_route(struct fib6_node *fn, struct rt6_info **rtp, int fib6_del(struct rt6_info *rt, struct nl_info *info) { + struct net *net = info->nl_net; struct fib6_node *fn = rt->rt6i_node; struct rt6_info **rtp; @@ -1154,7 +1156,7 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info) return -ENOENT; } #endif - if (fn == NULL || rt == ip6_null_entry) + if (fn == NULL || rt == net->ipv6.ip6_null_entry) return -ENOENT; BUG_TRAP(fn->fn_flags&RTN_RTINFO); @@ -1501,7 +1503,7 @@ static int fib6_net_init(struct net *net) goto out_fib_table_hash; net->ipv6.fib6_main_tbl->tb6_id = RT6_TABLE_MAIN; - net->ipv6.fib6_main_tbl->tb6_root.leaf = ip6_null_entry; + net->ipv6.fib6_main_tbl->tb6_root.leaf = net->ipv6.ip6_null_entry; net->ipv6.fib6_main_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; @@ -1511,7 +1513,7 @@ static int fib6_net_init(struct net *net) if (!net->ipv6.fib6_local_tbl) goto out_fib6_main_tbl; net->ipv6.fib6_local_tbl->tb6_id = RT6_TABLE_LOCAL; - net->ipv6.fib6_local_tbl->tb6_root.leaf = ip6_null_entry; + net->ipv6.fib6_local_tbl->tb6_root.leaf = net->ipv6.ip6_null_entry; net->ipv6.fib6_local_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; #endif @@ -1536,6 +1538,7 @@ out_timer: static void fib6_net_exit(struct net *net) { + rt6_ifdown(net, NULL); del_timer(net->ipv6.ip6_fib_timer); kfree(net->ipv6.ip6_fib_timer); #ifdef CONFIG_IPV6_MULTIPLE_TABLES diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8f954c1e961f..7ff66cebe77c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -145,8 +145,6 @@ static struct rt6_info ip6_null_entry_template = { .rt6i_ref = ATOMIC_INIT(1), }; -struct rt6_info *ip6_null_entry; - #ifdef CONFIG_IPV6_MULTIPLE_TABLES static int ip6_pkt_prohibit(struct sk_buff *skb); @@ -170,8 +168,6 @@ struct rt6_info ip6_prohibit_entry_template = { .rt6i_ref = ATOMIC_INIT(1), }; -struct rt6_info *ip6_prohibit_entry; - static struct rt6_info ip6_blk_hole_entry_template = { .u = { .dst = { @@ -190,8 +186,6 @@ static struct rt6_info ip6_blk_hole_entry_template = { .rt6i_ref = ATOMIC_INIT(1), }; -struct rt6_info *ip6_blk_hole_entry; - #endif /* allocate dst with ip6_dst_ops */ @@ -245,7 +239,8 @@ static inline int rt6_need_strict(struct in6_addr *daddr) * Route lookup. Any table->tb6_lock is implied. */ -static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt, +static inline struct rt6_info *rt6_device_match(struct net *net, + struct rt6_info *rt, int oif, int strict) { @@ -274,7 +269,7 @@ static __inline__ struct rt6_info *rt6_device_match(struct rt6_info *rt, return local; if (strict) - return ip6_null_entry; + return net->ipv6.ip6_null_entry; } return rt; } @@ -415,6 +410,7 @@ static struct rt6_info *find_rr_leaf(struct fib6_node *fn, static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) { struct rt6_info *match, *rt0; + struct net *net; RT6_TRACE("%s(fn->leaf=%p, oif=%d)\n", __FUNCTION__, fn->leaf, oif); @@ -440,7 +436,8 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) RT6_TRACE("%s() => %p\n", __FUNCTION__, match); - return (match ? match : ip6_null_entry); + net = rt0->rt6i_dev->nd_net; + return (match ? match : net->ipv6.ip6_null_entry); } #ifdef CONFIG_IPV6_ROUTE_INFO @@ -523,9 +520,9 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, } #endif -#define BACKTRACK(saddr) \ +#define BACKTRACK(__net, saddr) \ do { \ - if (rt == ip6_null_entry) { \ + if (rt == __net->ipv6.ip6_null_entry) { \ struct fib6_node *pn; \ while (1) { \ if (fn->fn_flags & RTN_TL_ROOT) \ @@ -541,7 +538,8 @@ do { \ } \ } while(0) -static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table, +static struct rt6_info *ip6_pol_route_lookup(struct net *net, + struct fib6_table *table, struct flowi *fl, int flags) { struct fib6_node *fn; @@ -551,8 +549,8 @@ static struct rt6_info *ip6_pol_route_lookup(struct fib6_table *table, fn = fib6_lookup(&table->tb6_root, &fl->fl6_dst, &fl->fl6_src); restart: rt = fn->leaf; - rt = rt6_device_match(rt, fl->oif, flags); - BACKTRACK(&fl->fl6_src); + rt = rt6_device_match(net, rt, fl->oif, flags); + BACKTRACK(net, &fl->fl6_src); out: dst_use(&rt->u.dst, jiffies); read_unlock_bh(&table->tb6_lock); @@ -668,8 +666,8 @@ static struct rt6_info *rt6_alloc_clone(struct rt6_info *ort, struct in6_addr *d return rt; } -static struct rt6_info *ip6_pol_route(struct fib6_table *table, int oif, - struct flowi *fl, int flags) +static struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table, int oif, + struct flowi *fl, int flags) { struct fib6_node *fn; struct rt6_info *rt, *nrt; @@ -688,8 +686,9 @@ restart_2: restart: rt = rt6_select(fn, oif, strict | reachable); - BACKTRACK(&fl->fl6_src); - if (rt == ip6_null_entry || + + BACKTRACK(net, &fl->fl6_src); + if (rt == net->ipv6.ip6_null_entry || rt->rt6i_flags & RTF_CACHE) goto out; @@ -707,7 +706,7 @@ restart: } dst_release(&rt->u.dst); - rt = nrt ? : ip6_null_entry; + rt = nrt ? : net->ipv6.ip6_null_entry; dst_hold(&rt->u.dst); if (nrt) { @@ -740,10 +739,10 @@ out2: return rt; } -static struct rt6_info *ip6_pol_route_input(struct fib6_table *table, +static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table *table, struct flowi *fl, int flags) { - return ip6_pol_route(table, fl->iif, fl, flags); + return ip6_pol_route(net, table, fl->iif, fl, flags); } void ip6_route_input(struct sk_buff *skb) @@ -770,10 +769,10 @@ void ip6_route_input(struct sk_buff *skb) skb->dst = fib6_rule_lookup(net, &fl, flags, ip6_pol_route_input); } -static struct rt6_info *ip6_pol_route_output(struct fib6_table *table, +static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table *table, struct flowi *fl, int flags) { - return ip6_pol_route(table, fl->oif, fl, flags); + return ip6_pol_route(net, table, fl->oif, fl, flags); } struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) @@ -1259,8 +1258,9 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) { int err; struct fib6_table *table; + struct net *net = rt->rt6i_dev->nd_net; - if (rt == ip6_null_entry) + if (rt == net->ipv6.ip6_null_entry) return -ENOENT; table = rt->rt6i_table; @@ -1329,7 +1329,8 @@ struct ip6rd_flowi { struct in6_addr gateway; }; -static struct rt6_info *__ip6_route_redirect(struct fib6_table *table, +static struct rt6_info *__ip6_route_redirect(struct net *net, + struct fib6_table *table, struct flowi *fl, int flags) { @@ -1372,8 +1373,8 @@ restart: } if (!rt) - rt = ip6_null_entry; - BACKTRACK(&fl->fl6_src); + rt = net->ipv6.ip6_null_entry; + BACKTRACK(net, &fl->fl6_src); out: dst_hold(&rt->u.dst); @@ -1415,10 +1416,11 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, { struct rt6_info *rt, *nrt = NULL; struct netevent_redirect netevent; + struct net *net = neigh->dev->nd_net; rt = ip6_route_redirect(dest, src, saddr, neigh->dev); - if (rt == ip6_null_entry) { + if (rt == net->ipv6.ip6_null_entry) { if (net_ratelimit()) printk(KERN_DEBUG "rt6_redirect: source isn't a valid nexthop " "for redirect target\n"); @@ -1886,10 +1888,18 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, return rt; } +struct arg_dev_net { + struct net_device *dev; + struct net *net; +}; + static int fib6_ifdown(struct rt6_info *rt, void *arg) { - if (((void*)rt->rt6i_dev == arg || arg == NULL) && - rt != ip6_null_entry) { + struct net_device *dev = ((struct arg_dev_net *)arg)->dev; + struct net *net = ((struct arg_dev_net *)arg)->net; + + if (((void *)rt->rt6i_dev == dev || dev == NULL) && + rt != net->ipv6.ip6_null_entry) { RT6_TRACE("deleted by ifdown %p\n", rt); return -1; } @@ -1898,7 +1908,12 @@ static int fib6_ifdown(struct rt6_info *rt, void *arg) void rt6_ifdown(struct net *net, struct net_device *dev) { - fib6_clean_all(net, fib6_ifdown, 0, dev); + struct arg_dev_net adn = { + .dev = dev, + .net = net, + }; + + fib6_clean_all(net, fib6_ifdown, 0, &adn); } struct rt6_mtu_change_arg @@ -2289,6 +2304,26 @@ errout: rtnl_set_sk_err(net, RTNLGRP_IPV6_ROUTE, err); } +static int ip6_route_dev_notify(struct notifier_block *this, + unsigned long event, void *data) +{ + struct net_device *dev = (struct net_device *)data; + struct net *net = dev->nd_net; + + if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) { + net->ipv6.ip6_null_entry->u.dst.dev = dev; + net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev); +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + net->ipv6.ip6_prohibit_entry->u.dst.dev = dev; + net->ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(dev); + net->ipv6.ip6_blk_hole_entry->u.dst.dev = dev; + net->ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(dev); +#endif + } + + return NOTIFY_OK; +} + /* * /proc */ @@ -2535,11 +2570,47 @@ struct ctl_table *ipv6_route_sysctl_init(struct net *net) static int ip6_route_net_init(struct net *net) { + int ret = 0; + + ret = -ENOMEM; + net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template, + sizeof(*net->ipv6.ip6_null_entry), + GFP_KERNEL); + if (!net->ipv6.ip6_null_entry) + goto out; + net->ipv6.ip6_null_entry->u.dst.path = + (struct dst_entry *)net->ipv6.ip6_null_entry; + +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template, + sizeof(*net->ipv6.ip6_prohibit_entry), + GFP_KERNEL); + if (!net->ipv6.ip6_prohibit_entry) { + kfree(net->ipv6.ip6_null_entry); + goto out; + } + net->ipv6.ip6_prohibit_entry->u.dst.path = + (struct dst_entry *)net->ipv6.ip6_prohibit_entry; + + net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template, + sizeof(*net->ipv6.ip6_blk_hole_entry), + GFP_KERNEL); + if (!net->ipv6.ip6_blk_hole_entry) { + kfree(net->ipv6.ip6_null_entry); + kfree(net->ipv6.ip6_prohibit_entry); + goto out; + } + net->ipv6.ip6_blk_hole_entry->u.dst.path = + (struct dst_entry *)net->ipv6.ip6_blk_hole_entry; +#endif + #ifdef CONFIG_PROC_FS proc_net_fops_create(net, "ipv6_route", 0, &ipv6_route_proc_fops); proc_net_fops_create(net, "rt6_stats", S_IRUGO, &rt6_stats_seq_fops); #endif - return 0; + ret = 0; +out: + return ret; } static void ip6_route_net_exit(struct net *net) @@ -2548,7 +2619,11 @@ static void ip6_route_net_exit(struct net *net) proc_net_remove(net, "ipv6_route"); proc_net_remove(net, "rt6_stats"); #endif - rt6_ifdown(net, NULL); + kfree(net->ipv6.ip6_null_entry); +#ifdef CONFIG_IPV6_MULTIPLE_TABLES + kfree(net->ipv6.ip6_prohibit_entry); + kfree(net->ipv6.ip6_blk_hole_entry); +#endif } static struct pernet_operations ip6_route_net_ops = { @@ -2556,6 +2631,11 @@ static struct pernet_operations ip6_route_net_ops = { .exit = ip6_route_net_exit, }; +static struct notifier_block ip6_route_dev_notifier = { + .notifier_call = ip6_route_dev_notify, + .priority = 0, +}; + int __init ip6_route_init(void) { int ret; @@ -2568,30 +2648,24 @@ int __init ip6_route_init(void) ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops.kmem_cachep; - ret = -ENOMEM; - ip6_null_entry = kmemdup(&ip6_null_entry_template, - sizeof(*ip6_null_entry), GFP_KERNEL); - if (!ip6_null_entry) + ret = register_pernet_subsys(&ip6_route_net_ops); + if (ret) goto out_kmem_cache; - ip6_null_entry->u.dst.path = (struct dst_entry *)ip6_null_entry; - -#ifdef CONFIG_IPV6_MULTIPLE_TABLES - ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template, - sizeof(*ip6_prohibit_entry), GFP_KERNEL); - if (!ip6_prohibit_entry) - goto out_ip6_null_entry; - ip6_prohibit_entry->u.dst.path = (struct dst_entry *)ip6_prohibit_entry; - - ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template, - sizeof(*ip6_blk_hole_entry), GFP_KERNEL); - if (!ip6_blk_hole_entry) - goto out_ip6_prohibit_entry; - ip6_blk_hole_entry->u.dst.path = (struct dst_entry *)ip6_blk_hole_entry; -#endif + /* Registering of the loopback is done before this portion of code, + * the loopback reference in rt6_info will not be taken, do it + * manually for init_net */ + init_net.ipv6.ip6_null_entry->u.dst.dev = init_net.loopback_dev; + init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); + #ifdef CONFIG_IPV6_MULTIPLE_TABLES + init_net.ipv6.ip6_prohibit_entry->u.dst.dev = init_net.loopback_dev; + init_net.ipv6.ip6_prohibit_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); + init_net.ipv6.ip6_blk_hole_entry->u.dst.dev = init_net.loopback_dev; + init_net.ipv6.ip6_blk_hole_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev); + #endif ret = fib6_init(); if (ret) - goto out_ip6_blk_hole_entry; + goto out_register_subsys; ret = xfrm6_init(); if (ret) @@ -2607,9 +2681,10 @@ int __init ip6_route_init(void) __rtnl_register(PF_INET6, RTM_GETROUTE, inet6_rtm_getroute, NULL)) goto fib6_rules_init; - ret = register_pernet_subsys(&ip6_route_net_ops); + ret = register_netdevice_notifier(&ip6_route_dev_notifier); if (ret) goto fib6_rules_init; + out: return ret; @@ -2619,14 +2694,8 @@ xfrm6_init: xfrm6_fini(); out_fib6_init: fib6_gc_cleanup(); -out_ip6_blk_hole_entry: -#ifdef CONFIG_IPV6_MULTIPLE_TABLES - kfree(ip6_blk_hole_entry); -out_ip6_prohibit_entry: - kfree(ip6_prohibit_entry); -out_ip6_null_entry: -#endif - kfree(ip6_null_entry); +out_register_subsys: + unregister_pernet_subsys(&ip6_route_net_ops); out_kmem_cache: kmem_cache_destroy(ip6_dst_ops.kmem_cachep); goto out; @@ -2634,15 +2703,10 @@ out_kmem_cache: void ip6_route_cleanup(void) { - unregister_pernet_subsys(&ip6_route_net_ops); + unregister_netdevice_notifier(&ip6_route_dev_notifier); fib6_rules_cleanup(); xfrm6_fini(); fib6_gc_cleanup(); + unregister_pernet_subsys(&ip6_route_net_ops); kmem_cache_destroy(ip6_dst_ops.kmem_cachep); - - kfree(ip6_null_entry); -#ifdef CONFIG_IPV6_MULTIPLE_TABLES - kfree(ip6_prohibit_entry); - kfree(ip6_blk_hole_entry); -#endif } -- cgit v1.2.3 From f2fc6a54585a1be6669613a31fbaba2ecbadcd36 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Tue, 4 Mar 2008 13:49:23 -0800 Subject: [NETNS][IPV6] route6 - move ip6_dst_ops inside the network namespace The ip6_dst_ops is moved inside the network namespace structure. All references to this structure are now relative to the initial network namespace. Signed-off-by: Benjamin Thery Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 1 + net/ipv6/route.c | 62 ++++++++++++++++++++++++------------------------ 2 files changed, 32 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 311a942f36e9..53b5a283a4b3 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -41,6 +41,7 @@ struct netns_ipv6 { struct timer_list *ip6_fib_timer; struct hlist_head *fib_table_hash; struct fib6_table *fib6_main_tbl; + struct dst_ops *ip6_dst_ops; #ifdef CONFIG_IPV6_MULTIPLE_TABLES struct rt6_info *ip6_prohibit_entry; struct rt6_info *ip6_blk_hole_entry; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index fa014d701c1a..d88b6ec3c5d1 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -113,8 +113,6 @@ static struct dst_ops ip6_dst_ops_template = { .entries = ATOMIC_INIT(0), }; -static struct dst_ops *ip6_dst_ops; - static void ip6_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu) { } @@ -188,9 +186,9 @@ static struct rt6_info ip6_blk_hole_entry_template = { #endif /* allocate dst with ip6_dst_ops */ -static __inline__ struct rt6_info *ip6_dst_alloc(void) +static inline struct rt6_info *ip6_dst_alloc(struct dst_ops *ops) { - return (struct rt6_info *)dst_alloc(ip6_dst_ops); + return (struct rt6_info *)dst_alloc(ops); } static void ip6_dst_destroy(struct dst_entry *dst) @@ -925,7 +923,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, if (unlikely(idev == NULL)) return NULL; - rt = ip6_dst_alloc(); + rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); if (unlikely(rt == NULL)) { in6_dev_put(idev); goto out; @@ -999,18 +997,18 @@ static int ip6_dst_gc(struct dst_ops *ops) unsigned long now = jiffies; if (time_after(last_gc + init_net.ipv6.sysctl.ip6_rt_gc_min_interval, now) && - atomic_read(&ip6_dst_ops->entries) <= init_net.ipv6.sysctl.ip6_rt_max_size) + atomic_read(&init_net.ipv6.ip6_dst_ops->entries) <= init_net.ipv6.sysctl.ip6_rt_max_size) goto out; expire++; fib6_run_gc(expire, &init_net); last_gc = now; - if (atomic_read(&ip6_dst_ops->entries) < ip6_dst_ops->gc_thresh) + if (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) < init_net.ipv6.ip6_dst_ops->gc_thresh) expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1; out: expire -= expire>>init_net.ipv6.sysctl.ip6_rt_gc_elasticity; - return (atomic_read(&ip6_dst_ops->entries) > init_net.ipv6.sysctl.ip6_rt_max_size); + return (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) > init_net.ipv6.sysctl.ip6_rt_max_size); } /* Clean host part of a prefix. Not necessary in radix tree, @@ -1084,7 +1082,7 @@ int ip6_route_add(struct fib6_config *cfg) goto out; } - rt = ip6_dst_alloc(); + rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); if (rt == NULL) { err = -ENOMEM; @@ -1570,7 +1568,8 @@ out: static struct rt6_info * ip6_rt_copy(struct rt6_info *ort) { - struct rt6_info *rt = ip6_dst_alloc(); + struct net *net = ort->rt6i_dev->nd_net; + struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); if (rt) { rt->u.dst.input = ort->u.dst.input; @@ -1849,7 +1848,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, int anycast) { struct net *net = idev->dev->nd_net; - struct rt6_info *rt = ip6_dst_alloc(); + struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); if (rt == NULL) return ERR_PTR(-ENOMEM); @@ -2407,7 +2406,7 @@ static int rt6_stats_seq_show(struct seq_file *seq, void *v) net->ipv6.rt6_stats->fib_rt_alloc, net->ipv6.rt6_stats->fib_rt_entries, net->ipv6.rt6_stats->fib_rt_cache, - atomic_read(&ip6_dst_ops->entries), + atomic_read(&net->ipv6.ip6_dst_ops->entries), net->ipv6.rt6_stats->fib_discarded_routes); return 0; @@ -2552,7 +2551,7 @@ struct ctl_table *ipv6_route_sysctl_init(struct net *net) if (table) { table[0].data = &net->ipv6.sysctl.flush_delay; - table[1].data = &ip6_dst_ops_template.gc_thresh; + table[1].data = &net->ipv6.ip6_dst_ops->gc_thresh; table[2].data = &net->ipv6.sysctl.ip6_rt_max_size; table[3].data = &net->ipv6.sysctl.ip6_rt_gc_min_interval; table[4].data = &net->ipv6.sysctl.ip6_rt_gc_timeout; @@ -2571,14 +2570,21 @@ static int ip6_route_net_init(struct net *net) int ret = 0; ret = -ENOMEM; + net->ipv6.ip6_dst_ops = kmemdup(&ip6_dst_ops_template, + sizeof(*net->ipv6.ip6_dst_ops), + GFP_KERNEL); + if (!net->ipv6.ip6_dst_ops) + goto out; + net->ipv6.ip6_dst_ops->dst_net = net; + net->ipv6.ip6_null_entry = kmemdup(&ip6_null_entry_template, sizeof(*net->ipv6.ip6_null_entry), GFP_KERNEL); if (!net->ipv6.ip6_null_entry) - goto out; + goto out_ip6_dst_ops; net->ipv6.ip6_null_entry->u.dst.path = (struct dst_entry *)net->ipv6.ip6_null_entry; - net->ipv6.ip6_null_entry->u.dst.ops = ip6_dst_ops; + net->ipv6.ip6_null_entry->u.dst.ops = net->ipv6.ip6_dst_ops; #ifdef CONFIG_IPV6_MULTIPLE_TABLES net->ipv6.ip6_prohibit_entry = kmemdup(&ip6_prohibit_entry_template, @@ -2590,7 +2596,7 @@ static int ip6_route_net_init(struct net *net) } net->ipv6.ip6_prohibit_entry->u.dst.path = (struct dst_entry *)net->ipv6.ip6_prohibit_entry; - net->ipv6.ip6_prohibit_entry->u.dst.ops = ip6_dst_ops; + net->ipv6.ip6_prohibit_entry->u.dst.ops = net->ipv6.ip6_dst_ops; net->ipv6.ip6_blk_hole_entry = kmemdup(&ip6_blk_hole_entry_template, sizeof(*net->ipv6.ip6_blk_hole_entry), @@ -2602,7 +2608,7 @@ static int ip6_route_net_init(struct net *net) } net->ipv6.ip6_blk_hole_entry->u.dst.path = (struct dst_entry *)net->ipv6.ip6_blk_hole_entry; - net->ipv6.ip6_blk_hole_entry->u.dst.ops = ip6_dst_ops; + net->ipv6.ip6_blk_hole_entry->u.dst.ops = net->ipv6.ip6_dst_ops; #endif #ifdef CONFIG_PROC_FS @@ -2612,6 +2618,10 @@ static int ip6_route_net_init(struct net *net) ret = 0; out: return ret; + +out_ip6_dst_ops: + kfree(net->ipv6.ip6_dst_ops); + goto out; } static void ip6_route_net_exit(struct net *net) @@ -2625,6 +2635,7 @@ static void ip6_route_net_exit(struct net *net) kfree(net->ipv6.ip6_prohibit_entry); kfree(net->ipv6.ip6_blk_hole_entry); #endif + kfree(net->ipv6.ip6_dst_ops); } static struct pernet_operations ip6_route_net_ops = { @@ -2641,20 +2652,12 @@ int __init ip6_route_init(void) { int ret; - ip6_dst_ops = kmemdup(&ip6_dst_ops_template, - sizeof(*ip6_dst_ops), GFP_KERNEL); - if (!ip6_dst_ops) - return -ENOMEM; - ret = -ENOMEM; ip6_dst_ops_template.kmem_cachep = kmem_cache_create("ip6_dst_cache", sizeof(struct rt6_info), 0, SLAB_HWCACHE_ALIGN, NULL); if (!ip6_dst_ops_template.kmem_cachep) - goto out_ip6_dst_ops; - - ip6_dst_ops->kmem_cachep = ip6_dst_ops_template.kmem_cachep; - ip6_dst_blackhole_ops.kmem_cachep = ip6_dst_ops_template.kmem_cachep; + goto out;; ret = register_pernet_subsys(&ip6_route_net_ops); if (ret) @@ -2705,9 +2708,7 @@ out_fib6_init: out_register_subsys: unregister_pernet_subsys(&ip6_route_net_ops); out_kmem_cache: - kmem_cache_destroy(ip6_dst_ops->kmem_cachep); -out_ip6_dst_ops: - kfree(ip6_dst_ops); + kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep); goto out; } @@ -2718,6 +2719,5 @@ void ip6_route_cleanup(void) xfrm6_fini(); fib6_gc_cleanup(); unregister_pernet_subsys(&ip6_route_net_ops); - kmem_cache_destroy(ip6_dst_ops->kmem_cachep); - kfree(ip6_dst_ops); + kmem_cache_destroy(ip6_dst_ops_template.kmem_cachep); } -- cgit v1.2.3 From 6891a346c387bd0a64afa50f4522f5fe8ba879d8 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Tue, 4 Mar 2008 13:49:47 -0800 Subject: [NETNS][IPV6] route6 - make garbage collection work with multiple network namespaces This patch makes the necessary changes to make IPv6 dst_entry garbage collection work with multiple network namespaces. In ip6_dst_gc(), static local variables are now declared per-namespace. Signed-off-by: Benjamin Thery Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 2 ++ net/ipv6/route.c | 23 ++++++++++++----------- 2 files changed, 14 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 53b5a283a4b3..90e6e24df858 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -42,6 +42,8 @@ struct netns_ipv6 { struct hlist_head *fib_table_hash; struct fib6_table *fib6_main_tbl; struct dst_ops *ip6_dst_ops; + unsigned int ip6_rt_gc_expire; + unsigned long ip6_rt_last_gc; #ifdef CONFIG_IPV6_MULTIPLE_TABLES struct rt6_info *ip6_prohibit_entry; struct rt6_info *ip6_blk_hole_entry; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d88b6ec3c5d1..10a6cc0aca6c 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -992,23 +992,22 @@ int icmp6_dst_gc(int *more) static int ip6_dst_gc(struct dst_ops *ops) { - static unsigned expire = 30*HZ; - static unsigned long last_gc; + struct net *net = ops->dst_net; unsigned long now = jiffies; - if (time_after(last_gc + init_net.ipv6.sysctl.ip6_rt_gc_min_interval, now) && - atomic_read(&init_net.ipv6.ip6_dst_ops->entries) <= init_net.ipv6.sysctl.ip6_rt_max_size) + if (time_after(net->ipv6.ip6_rt_last_gc + net->ipv6.sysctl.ip6_rt_gc_min_interval, now) && + atomic_read(&net->ipv6.ip6_dst_ops->entries) <= net->ipv6.sysctl.ip6_rt_max_size) goto out; - expire++; - fib6_run_gc(expire, &init_net); - last_gc = now; - if (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) < init_net.ipv6.ip6_dst_ops->gc_thresh) - expire = init_net.ipv6.sysctl.ip6_rt_gc_timeout>>1; + net->ipv6.ip6_rt_gc_expire++; + fib6_run_gc(net->ipv6.ip6_rt_gc_expire, net); + net->ipv6.ip6_rt_last_gc = now; + if (atomic_read(&net->ipv6.ip6_dst_ops->entries) < net->ipv6.ip6_dst_ops->gc_thresh) + net->ipv6.ip6_rt_gc_expire = net->ipv6.sysctl.ip6_rt_gc_timeout>>1; out: - expire -= expire>>init_net.ipv6.sysctl.ip6_rt_gc_elasticity; - return (atomic_read(&init_net.ipv6.ip6_dst_ops->entries) > init_net.ipv6.sysctl.ip6_rt_max_size); + net->ipv6.ip6_rt_gc_expire -= net->ipv6.ip6_rt_gc_expire>>net->ipv6.sysctl.ip6_rt_gc_elasticity; + return (atomic_read(&net->ipv6.ip6_dst_ops->entries) > net->ipv6.sysctl.ip6_rt_max_size); } /* Clean host part of a prefix. Not necessary in radix tree, @@ -2615,6 +2614,8 @@ static int ip6_route_net_init(struct net *net) proc_net_fops_create(net, "ipv6_route", 0, &ipv6_route_proc_fops); proc_net_fops_create(net, "rt6_stats", S_IRUGO, &rt6_stats_seq_fops); #endif + net->ipv6.ip6_rt_gc_expire = 30*HZ; + ret = 0; out: return ret; -- cgit v1.2.3 From af2849377e7b70afa1274e475be50286cd0ef6eb Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 5 Mar 2008 10:46:57 -0800 Subject: [NETNS][IPV6] addrconf - Pass the proper network namespace parameters to addrconf This patch propagates the network namespace pointer to the address configuration routines which need it, which means adding a new parameter to these functions, and make them use it instead of using the initial network namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/addrconf.h | 9 ++++--- net/ipv6/addrconf.c | 66 ++++++++++++++++++++++++++++++++++++-------------- net/ipv6/af_inet6.c | 6 ++--- 3 files changed, 57 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 89e3c53c8886..232da20e7171 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -55,9 +55,12 @@ struct prefix_info { extern int addrconf_init(void); extern void addrconf_cleanup(void); -extern int addrconf_add_ifaddr(void __user *arg); -extern int addrconf_del_ifaddr(void __user *arg); -extern int addrconf_set_dstaddr(void __user *arg); +extern int addrconf_add_ifaddr(struct net *net, + void __user *arg); +extern int addrconf_del_ifaddr(struct net *net, + void __user *arg); +extern int addrconf_set_dstaddr(struct net *net, + void __user *arg); extern int ipv6_chk_addr(struct net *net, struct in6_addr *addr, diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 17b06d220e95..127021bdd180 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1866,7 +1866,7 @@ ok: * Special case for SIT interfaces where we create a new "virtual" * device. */ -int addrconf_set_dstaddr(void __user *arg) +int addrconf_set_dstaddr(struct net *net, void __user *arg) { struct in6_ifreq ireq; struct net_device *dev; @@ -1878,7 +1878,7 @@ int addrconf_set_dstaddr(void __user *arg) if (copy_from_user(&ireq, arg, sizeof(struct in6_ifreq))) goto err_exit; - dev = __dev_get_by_index(&init_net, ireq.ifr6_ifindex); + dev = __dev_get_by_index(net, ireq.ifr6_ifindex); err = -ENODEV; if (dev == NULL) @@ -1909,7 +1909,8 @@ int addrconf_set_dstaddr(void __user *arg) if (err == 0) { err = -ENOBUFS; - if ((dev = __dev_get_by_name(&init_net, p.name)) == NULL) + dev = __dev_get_by_name(net, p.name); + if (!dev) goto err_exit; err = dev_open(dev); } @@ -1924,8 +1925,9 @@ err_exit: /* * Manual configuration of address on an interface */ -static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen, - __u8 ifa_flags, __u32 prefered_lft, __u32 valid_lft) +static int inet6_addr_add(struct net *net, int ifindex, struct in6_addr *pfx, + int plen, __u8 ifa_flags, __u32 prefered_lft, + __u32 valid_lft) { struct inet6_ifaddr *ifp; struct inet6_dev *idev; @@ -1939,7 +1941,8 @@ static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen, if (!valid_lft || prefered_lft > valid_lft) return -EINVAL; - if ((dev = __dev_get_by_index(&init_net, ifindex)) == NULL) + dev = __dev_get_by_index(net, ifindex); + if (!dev) return -ENODEV; if ((idev = addrconf_add_dev(dev)) == NULL) @@ -1984,13 +1987,15 @@ static int inet6_addr_add(int ifindex, struct in6_addr *pfx, int plen, return PTR_ERR(ifp); } -static int inet6_addr_del(int ifindex, struct in6_addr *pfx, int plen) +static int inet6_addr_del(struct net *net, int ifindex, struct in6_addr *pfx, + int plen) { struct inet6_ifaddr *ifp; struct inet6_dev *idev; struct net_device *dev; - if ((dev = __dev_get_by_index(&init_net, ifindex)) == NULL) + dev = __dev_get_by_index(net, ifindex); + if (!dev) return -ENODEV; if ((idev = __in6_dev_get(dev)) == NULL) @@ -2018,7 +2023,7 @@ static int inet6_addr_del(int ifindex, struct in6_addr *pfx, int plen) } -int addrconf_add_ifaddr(void __user *arg) +int addrconf_add_ifaddr(struct net *net, void __user *arg) { struct in6_ifreq ireq; int err; @@ -2030,13 +2035,14 @@ int addrconf_add_ifaddr(void __user *arg) return -EFAULT; rtnl_lock(); - err = inet6_addr_add(ireq.ifr6_ifindex, &ireq.ifr6_addr, ireq.ifr6_prefixlen, - IFA_F_PERMANENT, INFINITY_LIFE_TIME, INFINITY_LIFE_TIME); + err = inet6_addr_add(net, ireq.ifr6_ifindex, &ireq.ifr6_addr, + ireq.ifr6_prefixlen, IFA_F_PERMANENT, + INFINITY_LIFE_TIME, INFINITY_LIFE_TIME); rtnl_unlock(); return err; } -int addrconf_del_ifaddr(void __user *arg) +int addrconf_del_ifaddr(struct net *net, void __user *arg) { struct in6_ifreq ireq; int err; @@ -2048,7 +2054,8 @@ int addrconf_del_ifaddr(void __user *arg) return -EFAULT; rtnl_lock(); - err = inet6_addr_del(ireq.ifr6_ifindex, &ireq.ifr6_addr, ireq.ifr6_prefixlen); + err = inet6_addr_del(net, ireq.ifr6_ifindex, &ireq.ifr6_addr, + ireq.ifr6_prefixlen); rtnl_unlock(); return err; } @@ -3061,7 +3068,7 @@ inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) if (pfx == NULL) return -EINVAL; - return inet6_addr_del(ifm->ifa_index, pfx, ifm->ifa_prefixlen); + return inet6_addr_del(net, ifm->ifa_index, pfx, ifm->ifa_prefixlen); } static int inet6_addr_modify(struct inet6_ifaddr *ifp, u8 ifa_flags, @@ -3137,7 +3144,7 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) valid_lft = INFINITY_LIFE_TIME; } - dev = __dev_get_by_index(&init_net, ifm->ifa_index); + dev = __dev_get_by_index(net, ifm->ifa_index); if (dev == NULL) return -ENODEV; @@ -3150,8 +3157,9 @@ inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) * It would be best to check for !NLM_F_CREATE here but * userspace alreay relies on not having to provide this. */ - return inet6_addr_add(ifm->ifa_index, pfx, ifm->ifa_prefixlen, - ifa_flags, preferred_lft, valid_lft); + return inet6_addr_add(net, ifm->ifa_index, pfx, + ifm->ifa_prefixlen, ifa_flags, + preferred_lft, valid_lft); } if (nlh->nlmsg_flags & NLM_F_EXCL || @@ -4260,6 +4268,22 @@ int unregister_inet6addr_notifier(struct notifier_block *nb) EXPORT_SYMBOL(unregister_inet6addr_notifier); + +static int addrconf_net_init(struct net *net) +{ + return 0; +} + +static void addrconf_net_exit(struct net *net) +{ + ; +} + +static struct pernet_operations addrconf_net_ops = { + .init = addrconf_net_init, + .exit = addrconf_net_exit, +}; + /* * Init / cleanup code */ @@ -4301,6 +4325,10 @@ int __init addrconf_init(void) if (err) goto errlo; + err = register_pernet_device(&addrconf_net_ops); + if (err) + return err; + register_netdevice_notifier(&ipv6_dev_notf); addrconf_verify(0); @@ -4334,6 +4362,7 @@ void addrconf_cleanup(void) int i; unregister_netdevice_notifier(&ipv6_dev_notf); + unregister_pernet_device(&addrconf_net_ops); unregister_pernet_subsys(&addrconf_ops); @@ -4370,6 +4399,7 @@ void addrconf_cleanup(void) write_unlock_bh(&addrconf_hash_lock); del_timer(&addr_chk_timer); - rtnl_unlock(); + + unregister_pernet_subsys(&addrconf_net_ops); } diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 430bbd2139c1..afe9276d0420 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -454,11 +454,11 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) return(ipv6_route_ioctl(net, cmd, (void __user *)arg)); case SIOCSIFADDR: - return addrconf_add_ifaddr((void __user *) arg); + return addrconf_add_ifaddr(net, (void __user *) arg); case SIOCDIFADDR: - return addrconf_del_ifaddr((void __user *) arg); + return addrconf_del_ifaddr(net, (void __user *) arg); case SIOCSIFDSTADDR: - return addrconf_set_dstaddr((void __user *) arg); + return addrconf_set_dstaddr(net, (void __user *) arg); default: if (!sk->sk_prot->ioctl) return -ENOIOCTLCMD; -- cgit v1.2.3 From 4591db4f37618f37a9f1f25d291c3c7a43a15a21 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 5 Mar 2008 10:48:10 -0800 Subject: [NETNS][IPV6] route6 - add netns parameter to ip6_route_output Add an netns parameter to ip6_route_output. That will allow to access to the right routing table for outgoing traffic. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/ip6_route.h | 3 ++- net/ipv6/icmp.c | 2 +- net/ipv6/ip6_output.c | 4 ++-- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/ndisc.c | 2 +- net/ipv6/netfilter.c | 4 ++-- net/ipv6/netfilter/ip6t_REJECT.c | 2 +- net/ipv6/route.c | 7 ++++--- net/ipv6/sit.c | 2 +- net/ipv6/xfrm6_policy.c | 2 +- net/sctp/ipv6.c | 2 +- 11 files changed, 17 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index e0caed25bfba..0e2895c8b270 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -43,7 +43,8 @@ extern struct rt6_info *ip6_blk_hole_entry; extern void ip6_route_input(struct sk_buff *skb); -extern struct dst_entry * ip6_route_output(struct sock *sk, +extern struct dst_entry * ip6_route_output(struct net *net, + struct sock *sk, struct flowi *fl); extern int ip6_route_init(void); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index cff74127ea32..d4c4b50420d8 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -178,7 +178,7 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type, * XXX: perhaps the expire for routing entries cloned by * this lookup should be more aggressive (not longer than timeout). */ - dst = ip6_route_output(sk, fl); + dst = ip6_route_output(&init_net, sk, fl); if (dst->error) { IP6_INC_STATS(ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index ff3971173e1e..161afd1142d0 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -913,7 +913,7 @@ static int ip6_dst_lookup_tail(struct sock *sk, int err; if (*dst == NULL) - *dst = ip6_route_output(sk, fl); + *dst = ip6_route_output(&init_net, sk, fl); if ((err = (*dst)->error)) goto out_err_release; @@ -954,7 +954,7 @@ static int ip6_dst_lookup_tail(struct sock *sk, dst_release(*dst); memcpy(&fl_gw, fl, sizeof(struct flowi)); memset(&fl_gw.fl6_dst, 0, sizeof(struct in6_addr)); - *dst = ip6_route_output(sk, &fl_gw); + *dst = ip6_route_output(&init_net, sk, &fl_gw); if ((err = (*dst)->error)) goto out_err_release; } diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index 4e1981660b3c..1e1ad1ed87e6 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -847,7 +847,7 @@ static int ip6_tnl_xmit2(struct sk_buff *skb, if ((dst = ip6_tnl_dst_check(t)) != NULL) dst_hold(dst); else { - dst = ip6_route_output(NULL, fl); + dst = ip6_route_output(&init_net, NULL, fl); if (dst->error || xfrm_lookup(&dst, fl, NULL, 0) < 0) goto tx_err_link_failure; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index f1c95125100d..b5b4fd173e98 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1427,7 +1427,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, icmpv6_flow_init(ndisc_socket->sk, &fl, NDISC_REDIRECT, &saddr_buf, &ipv6_hdr(skb)->saddr, dev->ifindex); - dst = ip6_route_output(NULL, &fl); + dst = ip6_route_output(&init_net, NULL, &fl); if (dst == NULL) return; diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 2e06724dc348..aed51bcc66b4 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -23,7 +23,7 @@ int ip6_route_me_harder(struct sk_buff *skb) .saddr = iph->saddr, } }, }; - dst = ip6_route_output(skb->sk, &fl); + dst = ip6_route_output(&init_net, skb->sk, &fl); #ifdef CONFIG_XFRM if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) && @@ -86,7 +86,7 @@ static int nf_ip6_reroute(struct sk_buff *skb, static int nf_ip6_route(struct dst_entry **dst, struct flowi *fl) { - *dst = ip6_route_output(NULL, fl); + *dst = ip6_route_output(&init_net, NULL, fl); return (*dst)->error; } diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index b23baa635fe0..831708aeab37 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -93,7 +93,7 @@ static void send_reset(struct sk_buff *oldskb) fl.fl_ip_sport = otcph.dest; fl.fl_ip_dport = otcph.source; security_skb_classify_flow(oldskb, &fl); - dst = ip6_route_output(NULL, &fl); + dst = ip6_route_output(&init_net, NULL, &fl); if (dst == NULL) return; if (dst->error || xfrm_lookup(&dst, &fl, NULL, 0)) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index ec1fedb339cc..f31d7dc11e72 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -772,7 +772,8 @@ static struct rt6_info *ip6_pol_route_output(struct net *net, struct fib6_table return ip6_pol_route(net, table, fl->oif, fl, flags); } -struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) +struct dst_entry * ip6_route_output(struct net *net, struct sock *sk, + struct flowi *fl) { int flags = 0; @@ -782,7 +783,7 @@ struct dst_entry * ip6_route_output(struct sock *sk, struct flowi *fl) if (!ipv6_addr_any(&fl->fl6_src)) flags |= RT6_LOOKUP_F_HAS_SADDR; - return fib6_rule_lookup(&init_net, fl, flags, ip6_pol_route_output); + return fib6_rule_lookup(net, fl, flags, ip6_pol_route_output); } EXPORT_SYMBOL(ip6_route_output); @@ -2260,7 +2261,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void skb_reset_mac_header(skb); skb_reserve(skb, MAX_HEADER + sizeof(struct ipv6hdr)); - rt = (struct rt6_info*) ip6_route_output(NULL, &fl); + rt = (struct rt6_info*) ip6_route_output(&init_net, NULL, &fl); skb->dst = &rt->u.dst; err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 68720aa63f96..1b8196c8d145 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -393,7 +393,7 @@ isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev) fl.oif = dev->ifindex; security_skb_classify_flow(skb, &fl); - dst = ip6_route_output(NULL, &fl); + dst = ip6_route_output(&init_net, NULL, &fl); if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) { addr6 = (struct in6_addr*)&neigh->primary_key; diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 6ef56303e69e..e96dafdc7032 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -38,7 +38,7 @@ static struct dst_entry *xfrm6_dst_lookup(int tos, xfrm_address_t *saddr, if (saddr) memcpy(&fl.fl6_src, saddr, sizeof(fl.fl6_src)); - dst = ip6_route_output(NULL, &fl); + dst = ip6_route_output(&init_net, NULL, &fl); err = dst->error; if (dst->error) { diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 3e4878800b36..4862835b0c39 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -257,7 +257,7 @@ static struct dst_entry *sctp_v6_get_dst(struct sctp_association *asoc, NIP6(fl.fl6_src)); } - dst = ip6_route_output(NULL, &fl); + dst = ip6_route_output(&init_net, NULL, &fl); if (!dst->error) { struct rt6_info *rt; rt = (struct rt6_info *)dst; -- cgit v1.2.3 From ee6b967301b4aa5d4a4b61e2f682f086266db9fb Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 5 Mar 2008 18:30:47 -0800 Subject: [IPV4]: Add 'rtable' field in struct sk_buff to alias 'dst' and avoid casts (Anonymous) unions can help us to avoid ugly casts. A common cast it the (struct rtable *)skb->dst one. Defining an union like : union { struct dst_entry *dst; struct rtable *rtable; }; permits to use skb->rtable in place. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/skbuff.h | 5 ++++- include/net/inet_sock.h | 2 +- net/bridge/br_netfilter.c | 14 +++++++------- net/dccp/ipv4.c | 7 +++---- net/ipv4/arp.c | 4 ++-- net/ipv4/icmp.c | 10 +++++----- net/ipv4/igmp.c | 2 +- net/ipv4/ip_forward.c | 2 +- net/ipv4/ip_gre.c | 4 ++-- net/ipv4/ip_input.c | 2 +- net/ipv4/ip_options.c | 16 ++++++++-------- net/ipv4/ip_output.c | 10 +++++----- net/ipv4/ip_sockglue.c | 2 +- net/ipv4/ipip.c | 2 +- net/ipv4/ipmr.c | 6 +++--- net/ipv4/netfilter/ipt_MASQUERADE.c | 2 +- net/ipv4/netfilter/nf_nat_helper.c | 4 ++-- net/ipv4/route.c | 22 +++++++++++----------- net/ipv4/tcp_ipv4.c | 5 ++--- net/ipv4/udp_ipv4.c | 2 +- net/netfilter/nf_conntrack_netbios_ns.c | 2 +- net/sched/em_meta.c | 4 ++-- net/sctp/protocol.c | 8 ++++---- 23 files changed, 69 insertions(+), 68 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index bbd8d0027e2f..7beb239d2ee0 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -256,7 +256,10 @@ struct sk_buff { ktime_t tstamp; struct net_device *dev; - struct dst_entry *dst; + union { + struct dst_entry *dst; + struct rtable *rtable; + }; struct sec_path *sp; /* diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 89cd011edb99..8660cb0fa0dd 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -195,7 +195,7 @@ static inline int inet_sk_ehashfn(const struct sock *sk) static inline int inet_iif(const struct sk_buff *skb) { - return ((struct rtable *)skb->dst)->rt_iif; + return skb->rtable->rt_iif; } #endif /* _INET_SOCK_H */ diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 1c0efd8ad9f3..0278a069c6f1 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -223,8 +223,8 @@ static int br_nf_pre_routing_finish_ipv6(struct sk_buff *skb) } nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING; - skb->dst = (struct dst_entry *)&__fake_rtable; - dst_hold(skb->dst); + skb->rtable = &__fake_rtable; + dst_hold(&__fake_rtable.u.dst); skb->dev = nf_bridge->physindev; nf_bridge_push_encap_header(skb); @@ -388,8 +388,8 @@ bridged_dnat: skb->pkt_type = PACKET_HOST; } } else { - skb->dst = (struct dst_entry *)&__fake_rtable; - dst_hold(skb->dst); + skb->rtable = &__fake_rtable; + dst_hold(&__fake_rtable.u.dst); } skb->dev = nf_bridge->physindev; @@ -608,9 +608,9 @@ static unsigned int br_nf_local_in(unsigned int hook, struct sk_buff *skb, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - if (skb->dst == (struct dst_entry *)&__fake_rtable) { - dst_release(skb->dst); - skb->dst = NULL; + if (skb->rtable == &__fake_rtable) { + dst_release(&__fake_rtable.u.dst); + skb->rtable = NULL; } return NF_ACCEPT; diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 514a40b7fc7f..17ad69e90e48 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -450,7 +450,7 @@ static struct dst_entry* dccp_v4_route_skb(struct sock *sk, struct sk_buff *skb) { struct rtable *rt; - struct flowi fl = { .oif = ((struct rtable *)skb->dst)->rt_iif, + struct flowi fl = { .oif = skb->rtable->rt_iif, .nl_u = { .ip4_u = { .daddr = ip_hdr(skb)->saddr, .saddr = ip_hdr(skb)->daddr, @@ -511,7 +511,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET) return; - if (((struct rtable *)rxskb->dst)->rt_type != RTN_LOCAL) + if (rxskb->rtable->rt_type != RTN_LOCAL) return; dst = dccp_v4_route_skb(dccp_v4_ctl_socket->sk, rxskb); @@ -563,8 +563,7 @@ int dccp_v4_conn_request(struct sock *sk, struct sk_buff *skb) struct dccp_skb_cb *dcb = DCCP_SKB_CB(skb); /* Never answer to DCCP_PKT_REQUESTs send to broadcast or multicast */ - if (((struct rtable *)skb->dst)->rt_flags & - (RTCF_BROADCAST | RTCF_MULTICAST)) + if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) return 0; /* discard, don't send a reset here */ if (dccp_bad_service_code(sk, service)) { diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 69e80bd9774a..efe01df8fc0e 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -475,7 +475,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb) return 1; } - paddr = ((struct rtable*)skb->dst)->rt_gateway; + paddr = skb->rtable->rt_gateway; if (arp_set_predefined(inet_addr_type(&init_net, paddr), haddr, paddr, dev)) return 0; @@ -814,7 +814,7 @@ static int arp_process(struct sk_buff *skb) if (arp->ar_op == htons(ARPOP_REQUEST) && ip_route_input(skb, tip, sip, 0, dev) == 0) { - rt = (struct rtable*)skb->dst; + rt = skb->rtable; addr_type = rt->rt_type; if (addr_type == RTN_LOCAL) { diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index cee77d606fbe..ff9a8e643fcc 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -381,7 +381,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param, static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) { struct ipcm_cookie ipc; - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; struct net *net = rt->u.dst.dev->nd_net; struct sock *sk = icmp_sk(net); struct inet_sock *inet = inet_sk(sk); @@ -438,7 +438,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) struct iphdr *iph; int room; struct icmp_bxm icmp_param; - struct rtable *rt = (struct rtable *)skb_in->dst; + struct rtable *rt = skb_in->rtable; struct ipcm_cookie ipc; __be32 saddr; u8 tos; @@ -616,7 +616,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) RT_TOS(tos), rt2->u.dst.dev); dst_release(&rt2->u.dst); - rt2 = (struct rtable *)skb_in->dst; + rt2 = skb_in->rtable; skb_in->dst = odst; } @@ -943,7 +943,7 @@ static void icmp_address(struct sk_buff *skb) static void icmp_address_reply(struct sk_buff *skb) { - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; struct net_device *dev = skb->dev; struct in_device *in_dev; struct in_ifaddr *ifa; @@ -988,7 +988,7 @@ static void icmp_discard(struct sk_buff *skb) int icmp_rcv(struct sk_buff *skb) { struct icmphdr *icmph; - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) { int nh; diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index d3f34a772f3b..6a4ee8da6994 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -948,7 +948,7 @@ int igmp_rcv(struct sk_buff *skb) case IGMPV2_HOST_MEMBERSHIP_REPORT: case IGMPV3_HOST_MEMBERSHIP_REPORT: /* Is it our report looped back? */ - if (((struct rtable*)skb->dst)->fl.iif == 0) + if (skb->rtable->fl.iif == 0) break; /* don't rely on MC router hearing unicast reports */ if (skb->pkt_type == PACKET_MULTICAST || diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index 0b3b328d82db..9d6d3befd854 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -80,7 +80,7 @@ int ip_forward(struct sk_buff *skb) if (!xfrm4_route_forward(skb)) goto drop; - rt = (struct rtable*)skb->dst; + rt = skb->rtable; if (opt->is_strictroute && rt->rt_dst != rt->rt_gateway) goto sr_failed; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index e7821ba7a9a0..f9ee84420cb3 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -619,7 +619,7 @@ static int ipgre_rcv(struct sk_buff *skb) #ifdef CONFIG_NET_IPGRE_BROADCAST if (ipv4_is_multicast(iph->daddr)) { /* Looped back packet, drop it! */ - if (((struct rtable*)skb->dst)->fl.iif == 0) + if (skb->rtable->fl.iif == 0) goto drop; tunnel->stat.multicast++; skb->pkt_type = PACKET_BROADCAST; @@ -699,7 +699,7 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) } if (skb->protocol == htons(ETH_P_IP)) { - rt = (struct rtable*)skb->dst; + rt = skb->rtable; if ((dst = rt->rt_gateway) == 0) goto tx_error_icmp; } diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 65631391d479..d36e310b314d 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -351,7 +351,7 @@ static int ip_rcv_finish(struct sk_buff *skb) if (iph->ihl > 5 && ip_rcv_options(skb)) goto drop; - rt = (struct rtable*)skb->dst; + rt = skb->rtable; if (rt->rt_type == RTN_MULTICAST) IP_INC_STATS_BH(IPSTATS_MIB_INMCASTPKTS); else if (rt->rt_type == RTN_BROADCAST) diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index baaedd9689a0..df93a9c2efda 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -107,7 +107,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb) sptr = skb_network_header(skb); dptr = dopt->__data; - daddr = ((struct rtable*)skb->dst)->rt_spec_dst; + daddr = skb->rtable->rt_spec_dst; if (sopt->rr) { optlen = sptr[sopt->rr+1]; @@ -258,7 +258,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb) unsigned char * optptr; int optlen; unsigned char * pp_ptr = NULL; - struct rtable *rt = skb ? (struct rtable*)skb->dst : NULL; + struct rtable *rt = skb ? skb->rtable : NULL; if (!opt) { opt = &(IPCB(skb)->opt); @@ -558,7 +558,7 @@ void ip_forward_options(struct sk_buff *skb) { struct ip_options * opt = &(IPCB(skb)->opt); unsigned char * optptr; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; unsigned char *raw = skb_network_header(skb); if (opt->rr_needaddr) { @@ -606,7 +606,7 @@ int ip_options_rcv_srr(struct sk_buff *skb) __be32 nexthop; struct iphdr *iph = ip_hdr(skb); unsigned char *optptr = skb_network_header(skb) + opt->srr; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; struct rtable *rt2; int err; @@ -631,13 +631,13 @@ int ip_options_rcv_srr(struct sk_buff *skb) } memcpy(&nexthop, &optptr[srrptr-1], 4); - rt = (struct rtable*)skb->dst; - skb->dst = NULL; + rt = skb->rtable; + skb->rtable = NULL; err = ip_route_input(skb, nexthop, iph->saddr, iph->tos, skb->dev); - rt2 = (struct rtable*)skb->dst; + rt2 = skb->rtable; if (err || (rt2->rt_type != RTN_UNICAST && rt2->rt_type != RTN_LOCAL)) { ip_rt_put(rt2); - skb->dst = &rt->u.dst; + skb->rtable = rt; return -EINVAL; } ip_rt_put(rt); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 341779e685d9..dc494ea594a7 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -142,7 +142,7 @@ int ip_build_and_send_pkt(struct sk_buff *skb, struct sock *sk, __be32 saddr, __be32 daddr, struct ip_options *opt) { struct inet_sock *inet = inet_sk(sk); - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; struct iphdr *iph; /* Build the IP header. */ @@ -240,7 +240,7 @@ static int ip_finish_output(struct sk_buff *skb) int ip_mc_output(struct sk_buff *skb) { struct sock *sk = skb->sk; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; struct net_device *dev = rt->u.dst.dev; /* @@ -321,7 +321,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok) /* Skip all of this if the packet is already routed, * f.e. by something like SCTP. */ - rt = (struct rtable *) skb->dst; + rt = skb->rtable; if (rt != NULL) goto packet_routed; @@ -441,7 +441,7 @@ int ip_fragment(struct sk_buff *skb, int (*output)(struct sk_buff*)) unsigned int mtu, hlen, left, len, ll_rs, pad; int offset; __be16 not_last_frag; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; int err = 0; dev = rt->u.dst.dev; @@ -1357,7 +1357,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar } replyopts; struct ipcm_cookie ipc; __be32 daddr; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; if (ip_options_echo(&replyopts.opt, skb)) return; diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index de0572c88859..e7c9e4e72327 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -57,7 +57,7 @@ static void ip_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb) { struct in_pktinfo info; - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; info.ipi_addr.s_addr = ip_hdr(skb)->daddr; if (rt) { diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index dbaed69de06a..894bce96284a 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -528,7 +528,7 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev) if (!dst) { /* NBMA tunnel */ - if ((rt = (struct rtable*)skb->dst) == NULL) { + if ((rt = skb->rtable) == NULL) { tunnel->stat.tx_fifo_errors++; goto tx_error; } diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index a94f52c207a7..7d63d74ef62a 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1283,7 +1283,7 @@ static int ip_mr_forward(struct sk_buff *skb, struct mfc_cache *cache, int local if (vif_table[vif].dev != skb->dev) { int true_vifi; - if (((struct rtable*)skb->dst)->fl.iif == 0) { + if (skb->rtable->fl.iif == 0) { /* It is our own packet, looped back. Very complicated situation... @@ -1357,7 +1357,7 @@ dont_forward: int ip_mr_input(struct sk_buff *skb) { struct mfc_cache *cache; - int local = ((struct rtable*)skb->dst)->rt_flags&RTCF_LOCAL; + int local = skb->rtable->rt_flags&RTCF_LOCAL; /* Packet is looped back after forward, it should not be forwarded second time, but still can be delivered locally. @@ -1594,7 +1594,7 @@ int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait) { int err; struct mfc_cache *cache; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; read_lock(&mrt_lock); cache = ipmr_cache_find(rt->rt_src, rt->rt_dst); diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index 313b3fcf387e..c6817b18366a 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -77,7 +77,7 @@ masquerade_tg(struct sk_buff *skb, const struct net_device *in, return NF_ACCEPT; mr = targinfo; - rt = (struct rtable *)skb->dst; + rt = skb->rtable; newsrc = inet_select_addr(out, rt->rt_gateway, RT_SCOPE_UNIVERSE); if (!newsrc) { printk("MASQUERADE: %s ate my IP address\n", out->name); diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c index ca57f47bbd25..2fca727aa8ba 100644 --- a/net/ipv4/netfilter/nf_nat_helper.c +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -139,7 +139,7 @@ nf_nat_mangle_tcp_packet(struct sk_buff *skb, const char *rep_buffer, unsigned int rep_len) { - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; struct iphdr *iph; struct tcphdr *tcph; int oldlen, datalen; @@ -217,7 +217,7 @@ nf_nat_mangle_udp_packet(struct sk_buff *skb, const char *rep_buffer, unsigned int rep_len) { - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; struct iphdr *iph; struct udphdr *udph; int datalen, oldlen; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 8c3e165f0034..1051326c36b2 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1289,7 +1289,7 @@ reject_redirect: static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst) { - struct rtable *rt = (struct rtable*)dst; + struct rtable *rt = (struct rtable *)dst; struct dst_entry *ret = dst; if (rt) { @@ -1330,7 +1330,7 @@ static struct dst_entry *ipv4_negative_advice(struct dst_entry *dst) void ip_rt_send_redirect(struct sk_buff *skb) { - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; struct in_device *in_dev = in_dev_get(rt->u.dst.dev); if (!in_dev) @@ -1379,7 +1379,7 @@ out: static int ip_error(struct sk_buff *skb) { - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; unsigned long now; int code; @@ -1548,7 +1548,7 @@ static void ipv4_link_failure(struct sk_buff *skb) icmp_send(skb, ICMP_DEST_UNREACH, ICMP_HOST_UNREACH, 0); - rt = (struct rtable *) skb->dst; + rt = skb->rtable; if (rt) dst_set_expires(&rt->u.dst, 0); } @@ -1708,7 +1708,7 @@ static int ip_route_input_mc(struct sk_buff *skb, __be32 daddr, __be32 saddr, in_dev_put(in_dev); hash = rt_hash(daddr, saddr, dev->ifindex); - return rt_intern_hash(hash, rth, (struct rtable**) &skb->dst); + return rt_intern_hash(hash, rth, &skb->rtable); e_nobufs: in_dev_put(in_dev); @@ -1869,7 +1869,7 @@ static inline int ip_mkroute_input(struct sk_buff *skb, /* put it into the cache */ hash = rt_hash(daddr, saddr, fl->iif); - return rt_intern_hash(hash, rth, (struct rtable**)&skb->dst); + return rt_intern_hash(hash, rth, &skb->rtable); } /* @@ -2025,7 +2025,7 @@ local_input: } rth->rt_type = res.type; hash = rt_hash(daddr, saddr, fl.iif); - err = rt_intern_hash(hash, rth, (struct rtable**)&skb->dst); + err = rt_intern_hash(hash, rth, &skb->rtable); goto done; no_route: @@ -2091,7 +2091,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(in_hit); rcu_read_unlock(); - skb->dst = (struct dst_entry*)rth; + skb->rtable = rth; return 0; } RT_CACHE_STAT_INC(in_hlist_search); @@ -2598,7 +2598,7 @@ int ip_route_output_key(struct net *net, struct rtable **rp, struct flowi *flp) static int rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq, int event, int nowait, unsigned int flags) { - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; struct rtmsg *r; struct nlmsghdr *nlh; long expires; @@ -2742,7 +2742,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void err = ip_route_input(skb, dst, src, rtm->rtm_tos, dev); local_bh_enable(); - rt = (struct rtable*) skb->dst; + rt = skb->rtable; if (err == 0 && rt->u.dst.error) err = -rt->u.dst.error; } else { @@ -2762,7 +2762,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void if (err) goto errout_free; - skb->dst = &rt->u.dst; + skb->rtable = rt; if (rtm->rtm_flags & RTM_F_NOTIFY) rt->rt_flags |= RTCF_NOTIFY; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 3873c4dbeaeb..a79e324638eb 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -552,7 +552,7 @@ static void tcp_v4_send_reset(struct sock *sk, struct sk_buff *skb) if (th->rst) return; - if (((struct rtable *)skb->dst)->rt_type != RTN_LOCAL) + if (skb->rtable->rt_type != RTN_LOCAL) return; /* Swap the send and the receive. */ @@ -1262,8 +1262,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) #endif /* Never answer to SYNs send to broadcast or multicast */ - if (((struct rtable *)skb->dst)->rt_flags & - (RTCF_BROADCAST | RTCF_MULTICAST)) + if (skb->rtable->rt_flags & (RTCF_BROADCAST | RTCF_MULTICAST)) goto drop; /* TW buckets are converted to open requests without diff --git a/net/ipv4/udp_ipv4.c b/net/ipv4/udp_ipv4.c index 40978de7fb51..fd14c2c50ed4 100644 --- a/net/ipv4/udp_ipv4.c +++ b/net/ipv4/udp_ipv4.c @@ -893,7 +893,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], struct sock *sk; struct udphdr *uh = udp_hdr(skb); unsigned short ulen; - struct rtable *rt = (struct rtable*)skb->dst; + struct rtable *rt = skb->rtable; __be32 saddr = ip_hdr(skb)->saddr; __be32 daddr = ip_hdr(skb)->daddr; diff --git a/net/netfilter/nf_conntrack_netbios_ns.c b/net/netfilter/nf_conntrack_netbios_ns.c index 9810d81e2a06..60dedaded84e 100644 --- a/net/netfilter/nf_conntrack_netbios_ns.c +++ b/net/netfilter/nf_conntrack_netbios_ns.c @@ -47,7 +47,7 @@ static int help(struct sk_buff *skb, unsigned int protoff, { struct nf_conntrack_expect *exp; struct iphdr *iph = ip_hdr(skb); - struct rtable *rt = (struct rtable *)skb->dst; + struct rtable *rt = skb->rtable; struct in_device *in_dev; __be32 mask = 0; diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c index 3da4129b89d1..72cf86e3c090 100644 --- a/net/sched/em_meta.c +++ b/net/sched/em_meta.c @@ -256,10 +256,10 @@ META_COLLECTOR(int_rtclassid) META_COLLECTOR(int_rtiif) { - if (unlikely(skb->dst == NULL)) + if (unlikely(skb->rtable == NULL)) *err = -1; else - dst->value = ((struct rtable*) skb->dst)->fl.iif; + dst->value = skb->rtable->fl.iif; } /************************************************************************** diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 8d9d929f6cea..1afef08f6c1d 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -363,7 +363,7 @@ static int sctp_v4_addr_valid(union sctp_addr *addr, return 0; /* Is this a broadcast address? */ - if (skb && ((struct rtable *)skb->dst)->rt_flags & RTCF_BROADCAST) + if (skb && skb->rtable->rt_flags & RTCF_BROADCAST) return 0; return 1; @@ -539,7 +539,7 @@ static void sctp_v4_get_saddr(struct sctp_association *asoc, /* 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? */ @@ -828,8 +828,8 @@ 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)); + NIPQUAD(skb->rtable->rt_src), + NIPQUAD(skb->rtable->rt_dst)); SCTP_INC_STATS(SCTP_MIB_OUTSCTPPACKS); return ip_queue_xmit(skb, ipfragok); -- cgit v1.2.3 From f59d43899e279c77924a7ada4bec8c70e5aeca06 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 5 Mar 2008 20:58:10 -0800 Subject: [IPV6]: Fix powerpc allmodconfig build warnings. Introduced by changeset 95e41e93e18d8e1e272ce23d96bae4f17ce11d42 ("[IPV6]: Make ndisc_flow_init() common for later use.") Reported by Stephen Rothwell. In file included from net/ipv6/netfilter/ip6_tables.c:21: include/linux/icmpv6.h:192: warning: 'struct in6_addr' declared inside parameter list include/linux/icmpv6.h:192: warning: its scope is only this definition or declaration, which is probably not what you want Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index e4d4300d768f..03067443198a 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -184,6 +184,7 @@ extern void icmpv6_param_prob(struct sk_buff *skb, int code, int pos); struct flowi; +struct in6_addr; extern void icmpv6_flow_init(struct sock *sk, struct flowi *fl, u8 type, -- cgit v1.2.3 From 37c5798968d0ce4d479f114f1d5785551b57bfa5 Mon Sep 17 00:00:00 2001 From: Luis Carlos Cobo Date: Sat, 23 Feb 2008 15:17:04 +0100 Subject: wireless: various definitions for mesh networking Signed-off-by: Luis Carlos Cobo Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/ieee80211.h | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) (limited to 'include') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index f577c8f1c66d..f27d11ab418b 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -97,6 +97,7 @@ #define IEEE80211_MAX_FRAME_LEN 2352 #define IEEE80211_MAX_SSID_LEN 32 +#define IEEE80211_MAX_MESH_ID_LEN 32 struct ieee80211_hdr { __le16 frame_control; @@ -109,6 +110,16 @@ struct ieee80211_hdr { } __attribute__ ((packed)); +struct ieee80211s_hdr { + u8 flags; + u8 ttl; + u8 seqnum[3]; + u8 eaddr1[6]; + u8 eaddr2[6]; + u8 eaddr3[6]; +} __attribute__ ((packed)); + + struct ieee80211_mgmt { __le16 frame_control; __le16 duration; @@ -206,6 +217,23 @@ struct ieee80211_mgmt { __le16 params; __le16 reason_code; } __attribute__((packed)) delba; + struct{ + u8 action_code; + /* capab_info for open and confirm, + * reason for close + */ + __le16 aux; + /* Followed in plink_confirm by status + * code, AID and supported rates, + * and directly by supported rates in + * plink_open and plink_close + */ + u8 variable[0]; + } __attribute__((packed)) plink_action; + struct{ + u8 action_code; + u8 variable[0]; + } __attribute__((packed)) mesh_action; } u; } __attribute__ ((packed)) action; } u; @@ -437,6 +465,13 @@ enum ieee80211_eid { WLAN_EID_TS_DELAY = 43, WLAN_EID_TCLAS_PROCESSING = 44, WLAN_EID_QOS_CAPA = 46, + /* 802.11s */ + WLAN_EID_MESH_CONFIG = 36, /* Pending IEEE 802.11 ANA approval */ + WLAN_EID_MESH_ID = 37, /* Pending IEEE 802.11 ANA approval */ + WLAN_EID_PEER_LINK = 40, /* Pending IEEE 802.11 ANA approval */ + WLAN_EID_PREQ = 53, /* Pending IEEE 802.11 ANA approval */ + WLAN_EID_PREP = 54, /* Pending IEEE 802.11 ANA approval */ + WLAN_EID_PERR = 55, /* Pending IEEE 802.11 ANA approval */ /* 802.11h */ WLAN_EID_PWR_CONSTRAINT = 32, WLAN_EID_PWR_CAPABILITY = 33, -- cgit v1.2.3 From cc0672a1066829be7e1b0128a13e36a2d0a15479 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 23 Feb 2008 15:17:05 +0100 Subject: WEXT: add mesh interface type This introduces a new WEXT type IW_MODE_MESH for mesh networks, used for scan results. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/wireless.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/wireless.h b/include/linux/wireless.h index 3160dfed73ca..2864b1699ecc 100644 --- a/include/linux/wireless.h +++ b/include/linux/wireless.h @@ -455,6 +455,7 @@ #define IW_MODE_REPEAT 4 /* Wireless Repeater (forwarder) */ #define IW_MODE_SECOND 5 /* Secondary master/repeater (backup) */ #define IW_MODE_MONITOR 6 /* Passive monitor (listen only) */ +#define IW_MODE_MESH 7 /* Mesh (IEEE 802.11s) network */ /* Statistics flags (bitmask in updated) */ #define IW_QUAL_QUAL_UPDATED 0x01 /* Value was updated since last read */ -- cgit v1.2.3 From 2ec600d672e74488f8d1acf67a0a2baed222564c Mon Sep 17 00:00:00 2001 From: Luis Carlos Cobo Date: Sat, 23 Feb 2008 15:17:06 +0100 Subject: nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/linux/nl80211.h | 119 +++++++++++-- include/net/cfg80211.h | 139 +++++++++++++-- net/mac80211/cfg.c | 20 ++- net/wireless/nl80211.c | 438 +++++++++++++++++++++++++++++++++++++++++++++--- 4 files changed, 650 insertions(+), 66 deletions(-) (limited to 'include') diff --git a/include/linux/nl80211.h b/include/linux/nl80211.h index a9f0b93324a2..ea6517e58b04 100644 --- a/include/linux/nl80211.h +++ b/include/linux/nl80211.h @@ -78,6 +78,18 @@ * or, if no MAC address given, all stations, on the interface identified * by %NL80211_ATTR_IFINDEX. * + * @NL80211_CMD_GET_MPATH: Get mesh path attributes for mesh path to + * destination %NL80211_ATTR_MAC on the interface identified by + * %NL80211_ATTR_IFINDEX. + * @NL80211_CMD_SET_MPATH: Set mesh path attributes for mesh path to + * destination %NL80211_ATTR_MAC on the interface identified by + * %NL80211_ATTR_IFINDEX. + * @NL80211_CMD_NEW_PATH: Add a mesh path with given attributes to the + * the interface identified by %NL80211_ATTR_IFINDEX. + * @NL80211_CMD_DEL_PATH: Remove a mesh path identified by %NL80211_ATTR_MAC + * or, if no MAC address given, all mesh paths, on the interface identified + * by %NL80211_ATTR_IFINDEX. + * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use */ @@ -112,6 +124,11 @@ enum nl80211_commands { /* add commands here */ + NL80211_CMD_GET_MPATH, + NL80211_CMD_SET_MPATH, + NL80211_CMD_NEW_MPATH, + NL80211_CMD_DEL_MPATH, + /* used to define NL80211_CMD_MAX below */ __NL80211_CMD_AFTER_LAST, NL80211_CMD_MAX = __NL80211_CMD_AFTER_LAST - 1 @@ -157,13 +174,21 @@ enum nl80211_commands { * restriction (at most %NL80211_MAX_SUPP_RATES). * @NL80211_ATTR_STA_VLAN: interface index of VLAN interface to move station * to, or the AP interface the station was originally added to to. - * @NL80211_ATTR_STA_STATS: statistics for a station, part of station info + * @NL80211_ATTR_STA_INFO: information about a station, part of station info * given for %NL80211_CMD_GET_STATION, nested attribute containing - * info as possible, see &enum nl80211_sta_stats. + * info as possible, see &enum nl80211_sta_info. * * @NL80211_ATTR_WIPHY_BANDS: Information about an operating bands, * consisting of a nested array. * + * @NL80211_ATTR_MESH_ID: mesh id (1-32 bytes). + * @NL80211_ATTR_PLINK_ACTION: action to perform on the mesh peer link. + * @NL80211_ATTR_MPATH_NEXT_HOP: MAC address of the next hop for a mesh path. + * @NL80211_ATTR_MPATH_INFO: information about a mesh_path, part of mesh path + * info given for %NL80211_CMD_GET_MPATH, nested attribute described at + * &enum nl80211_mpath_info. + * + * * @NL80211_ATTR_MNTR_FLAGS: flags, nested element with NLA_FLAG attributes of * &enum nl80211_mntr_flags. * @@ -199,7 +224,7 @@ enum nl80211_attrs { NL80211_ATTR_STA_LISTEN_INTERVAL, NL80211_ATTR_STA_SUPPORTED_RATES, NL80211_ATTR_STA_VLAN, - NL80211_ATTR_STA_STATS, + NL80211_ATTR_STA_INFO, NL80211_ATTR_WIPHY_BANDS, @@ -207,6 +232,11 @@ enum nl80211_attrs { /* add attributes here, update the policy in nl80211.c */ + NL80211_ATTR_MESH_ID, + NL80211_ATTR_STA_PLINK_ACTION, + NL80211_ATTR_MPATH_NEXT_HOP, + NL80211_ATTR_MPATH_INFO, + __NL80211_ATTR_AFTER_LAST, NL80211_ATTR_MAX = __NL80211_ATTR_AFTER_LAST - 1 }; @@ -223,6 +253,7 @@ enum nl80211_attrs { * @NL80211_IFTYPE_AP_VLAN: VLAN interface for access points * @NL80211_IFTYPE_WDS: wireless distribution interface * @NL80211_IFTYPE_MONITOR: monitor interface receiving all frames + * @NL80211_IFTYPE_MESH_POINT: mesh point * @NL80211_IFTYPE_MAX: highest interface type number currently defined * @__NL80211_IFTYPE_AFTER_LAST: internal use * @@ -238,6 +269,7 @@ enum nl80211_iftype { NL80211_IFTYPE_AP_VLAN, NL80211_IFTYPE_WDS, NL80211_IFTYPE_MONITOR, + NL80211_IFTYPE_MESH_POINT, /* keep last */ __NL80211_IFTYPE_AFTER_LAST, @@ -267,27 +299,78 @@ enum nl80211_sta_flags { }; /** - * enum nl80211_sta_stats - station statistics + * enum nl80211_sta_info - station information * - * These attribute types are used with %NL80211_ATTR_STA_STATS + * These attribute types are used with %NL80211_ATTR_STA_INFO * when getting information about a station. * - * @__NL80211_STA_STAT_INVALID: attribute number 0 is reserved - * @NL80211_STA_STAT_INACTIVE_TIME: time since last activity (u32, msecs) - * @NL80211_STA_STAT_RX_BYTES: total received bytes (u32, from this station) - * @NL80211_STA_STAT_TX_BYTES: total transmitted bytes (u32, to this station) - * @__NL80211_STA_STAT_AFTER_LAST: internal - * @NL80211_STA_STAT_MAX: highest possible station stats attribute + * @__NL80211_STA_INFO_INVALID: attribute number 0 is reserved + * @NL80211_STA_INFO_INACTIVE_TIME: time since last activity (u32, msecs) + * @NL80211_STA_INFO_RX_BYTES: total received bytes (u32, from this station) + * @NL80211_STA_INFO_TX_BYTES: total transmitted bytes (u32, to this station) + * @__NL80211_STA_INFO_AFTER_LAST: internal + * @NL80211_STA_INFO_MAX: highest possible station info attribute + */ +enum nl80211_sta_info { + __NL80211_STA_INFO_INVALID, + NL80211_STA_INFO_INACTIVE_TIME, + NL80211_STA_INFO_RX_BYTES, + NL80211_STA_INFO_TX_BYTES, + NL80211_STA_INFO_LLID, + NL80211_STA_INFO_PLID, + NL80211_STA_INFO_PLINK_STATE, + + /* keep last */ + __NL80211_STA_INFO_AFTER_LAST, + NL80211_STA_INFO_MAX = __NL80211_STA_INFO_AFTER_LAST - 1 +}; + +/** + * enum nl80211_mpath_flags - nl80211 mesh path flags + * + * @NL80211_MPATH_FLAG_ACTIVE: the mesh path is active + * @NL80211_MPATH_FLAG_RESOLVING: the mesh path discovery process is running + * @NL80211_MPATH_FLAG_DSN_VALID: the mesh path contains a valid DSN + * @NL80211_MPATH_FLAG_FIXED: the mesh path has been manually set + * @NL80211_MPATH_FLAG_RESOLVED: the mesh path discovery process succeeded + */ +enum nl80211_mpath_flags { + NL80211_MPATH_FLAG_ACTIVE = 1<<0, + NL80211_MPATH_FLAG_RESOLVING = 1<<1, + NL80211_MPATH_FLAG_DSN_VALID = 1<<2, + NL80211_MPATH_FLAG_FIXED = 1<<3, + NL80211_MPATH_FLAG_RESOLVED = 1<<4, +}; + +/** + * enum nl80211_mpath_info - mesh path information + * + * These attribute types are used with %NL80211_ATTR_MPATH_INFO when getting + * information about a mesh path. + * + * @__NL80211_MPATH_INFO_INVALID: attribute number 0 is reserved + * @NL80211_ATTR_MPATH_FRAME_QLEN: number of queued frames for this destination + * @NL80211_ATTR_MPATH_DSN: destination sequence number + * @NL80211_ATTR_MPATH_METRIC: metric (cost) of this mesh path + * @NL80211_ATTR_MPATH_EXPTIME: expiration time for the path, in msec from now + * @NL80211_ATTR_MPATH_FLAGS: mesh path flags, enumerated in + * &enum nl80211_mpath_flags; + * @NL80211_ATTR_MPATH_DISCOVERY_TIMEOUT: total path discovery timeout, in msec + * @NL80211_ATTR_MPATH_DISCOVERY_RETRIES: mesh path discovery retries */ -enum nl80211_sta_stats { - __NL80211_STA_STAT_INVALID, - NL80211_STA_STAT_INACTIVE_TIME, - NL80211_STA_STAT_RX_BYTES, - NL80211_STA_STAT_TX_BYTES, +enum nl80211_mpath_info { + __NL80211_MPATH_INFO_INVALID, + NL80211_MPATH_INFO_FRAME_QLEN, + NL80211_MPATH_INFO_DSN, + NL80211_MPATH_INFO_METRIC, + NL80211_MPATH_INFO_EXPTIME, + NL80211_MPATH_INFO_FLAGS, + NL80211_MPATH_INFO_DISCOVERY_TIMEOUT, + NL80211_MPATH_INFO_DISCOVERY_RETRIES, /* keep last */ - __NL80211_STA_STAT_AFTER_LAST, - NL80211_STA_STAT_MAX = __NL80211_STA_STAT_AFTER_LAST - 1 + __NL80211_MPATH_INFO_AFTER_LAST, + NL80211_MPATH_INFO_MAX = __NL80211_MPATH_INFO_AFTER_LAST - 1 }; /** diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index ab4caf63954f..e00750836ba5 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -12,6 +12,16 @@ * Copyright 2006, 2007 Johannes Berg */ +/** + * struct vif_params - describes virtual interface parameters + * @mesh_id: mesh ID to use + * @mesh_id_len: length of the mesh ID + */ +struct vif_params { + u8 *mesh_id; + int mesh_id_len; +}; + /* Radiotap header iteration * implemented in net/wireless/radiotap.c * docs in Documentation/networking/radiotap-headers.txt @@ -108,6 +118,19 @@ enum station_flags { STATION_FLAG_WME = 1<ieee80211_ptr); struct sta_info *sta; @@ -307,13 +309,13 @@ static int ieee80211_get_station(struct wiphy *wiphy, struct net_device *dev, /* XXX: verify sta->dev == dev */ - stats->filled = STATION_STAT_INACTIVE_TIME | - STATION_STAT_RX_BYTES | - STATION_STAT_TX_BYTES; + sinfo->filled = STATION_INFO_INACTIVE_TIME | + STATION_INFO_RX_BYTES | + STATION_INFO_TX_BYTES; - stats->inactive_time = jiffies_to_msecs(jiffies - sta->last_rx); - stats->rx_bytes = sta->rx_bytes; - stats->tx_bytes = sta->tx_bytes; + sinfo->inactive_time = jiffies_to_msecs(jiffies - sta->last_rx); + sinfo->rx_bytes = sta->rx_bytes; + sinfo->tx_bytes = sta->tx_bytes; sta_info_put(sta); diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 5b3474798b8d..64a7460af734 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -81,8 +81,12 @@ static struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] __read_mostly = { [NL80211_ATTR_STA_LISTEN_INTERVAL] = { .type = NLA_U16 }, [NL80211_ATTR_STA_SUPPORTED_RATES] = { .type = NLA_BINARY, .len = NL80211_MAX_SUPP_RATES }, + [NL80211_ATTR_STA_PLINK_ACTION] = { .type = NLA_U8 }, [NL80211_ATTR_STA_VLAN] = { .type = NLA_U32 }, [NL80211_ATTR_MNTR_FLAGS] = { .type = NLA_NESTED }, + [NL80211_ATTR_MESH_ID] = { .type = NLA_BINARY, + .len = IEEE80211_MAX_MESH_ID_LEN }, + [NL80211_ATTR_MPATH_NEXT_HOP] = { .type = NLA_U32 }, }; /* message building helper */ @@ -369,11 +373,14 @@ static int parse_monitor_flags(struct nlattr *nla, u32 *mntrflags) static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *drv; + struct vif_params params; int err, ifindex; enum nl80211_iftype type; struct net_device *dev; u32 flags; + memset(¶ms, 0, sizeof(params)); + if (info->attrs[NL80211_ATTR_IFTYPE]) { type = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]); if (type > NL80211_IFTYPE_MAX) @@ -392,12 +399,18 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) goto unlock; } + if (type == NL80211_IFTYPE_MESH_POINT && + info->attrs[NL80211_ATTR_MESH_ID]) { + params.mesh_id = nla_data(info->attrs[NL80211_ATTR_MESH_ID]); + params.mesh_id_len = nla_len(info->attrs[NL80211_ATTR_MESH_ID]); + } + rtnl_lock(); err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ? info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL, &flags); err = drv->ops->change_virtual_intf(&drv->wiphy, ifindex, - type, err ? NULL : &flags); + type, err ? NULL : &flags, ¶ms); rtnl_unlock(); unlock: @@ -408,10 +421,13 @@ static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info) static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *drv; + struct vif_params params; int err; enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED; u32 flags; + memset(¶ms, 0, sizeof(params)); + if (!info->attrs[NL80211_ATTR_IFNAME]) return -EINVAL; @@ -430,15 +446,22 @@ static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info) goto unlock; } + if (type == NL80211_IFTYPE_MESH_POINT && + info->attrs[NL80211_ATTR_MESH_ID]) { + params.mesh_id = nla_data(info->attrs[NL80211_ATTR_MESH_ID]); + params.mesh_id_len = nla_len(info->attrs[NL80211_ATTR_MESH_ID]); + } + rtnl_lock(); err = parse_monitor_flags(type == NL80211_IFTYPE_MONITOR ? info->attrs[NL80211_ATTR_MNTR_FLAGS] : NULL, &flags); err = drv->ops->add_virtual_intf(&drv->wiphy, nla_data(info->attrs[NL80211_ATTR_IFNAME]), - type, err ? NULL : &flags); + type, err ? NULL : &flags, ¶ms); rtnl_unlock(); + unlock: cfg80211_put_dev(drv); return err; @@ -866,10 +889,10 @@ static int parse_station_flags(struct nlattr *nla, u32 *staflags) static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq, int flags, struct net_device *dev, - u8 *mac_addr, struct station_stats *stats) + u8 *mac_addr, struct station_info *sinfo) { void *hdr; - struct nlattr *statsattr; + struct nlattr *sinfoattr; hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_STATION); if (!hdr) @@ -878,20 +901,29 @@ static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq, NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, dev->ifindex); NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr); - statsattr = nla_nest_start(msg, NL80211_ATTR_STA_STATS); - if (!statsattr) + sinfoattr = nla_nest_start(msg, NL80211_ATTR_STA_INFO); + if (!sinfoattr) goto nla_put_failure; - if (stats->filled & STATION_STAT_INACTIVE_TIME) - NLA_PUT_U32(msg, NL80211_STA_STAT_INACTIVE_TIME, - stats->inactive_time); - if (stats->filled & STATION_STAT_RX_BYTES) - NLA_PUT_U32(msg, NL80211_STA_STAT_RX_BYTES, - stats->rx_bytes); - if (stats->filled & STATION_STAT_TX_BYTES) - NLA_PUT_U32(msg, NL80211_STA_STAT_TX_BYTES, - stats->tx_bytes); - - nla_nest_end(msg, statsattr); + if (sinfo->filled & STATION_INFO_INACTIVE_TIME) + NLA_PUT_U32(msg, NL80211_STA_INFO_INACTIVE_TIME, + sinfo->inactive_time); + if (sinfo->filled & STATION_INFO_RX_BYTES) + NLA_PUT_U32(msg, NL80211_STA_INFO_RX_BYTES, + sinfo->rx_bytes); + if (sinfo->filled & STATION_INFO_TX_BYTES) + NLA_PUT_U32(msg, NL80211_STA_INFO_TX_BYTES, + sinfo->tx_bytes); + if (sinfo->filled & STATION_INFO_LLID) + NLA_PUT_U16(msg, NL80211_STA_INFO_LLID, + sinfo->llid); + if (sinfo->filled & STATION_INFO_PLID) + NLA_PUT_U16(msg, NL80211_STA_INFO_PLID, + sinfo->plid); + if (sinfo->filled & STATION_INFO_PLINK_STATE) + NLA_PUT_U8(msg, NL80211_STA_INFO_PLINK_STATE, + sinfo->plink_state); + + nla_nest_end(msg, sinfoattr); return genlmsg_end(msg, hdr); @@ -899,17 +931,80 @@ static int nl80211_send_station(struct sk_buff *msg, u32 pid, u32 seq, return genlmsg_cancel(msg, hdr); } +static int nl80211_dump_station(struct sk_buff *skb, + struct netlink_callback *cb) +{ + int wp_idx = 0; + int if_idx = 0; + int sta_idx = cb->args[2]; + int wp_start = cb->args[0]; + int if_start = cb->args[1]; + struct station_info sinfo; + struct cfg80211_registered_device *dev; + struct wireless_dev *wdev; + u8 mac_addr[ETH_ALEN]; + int err; + int exit = 0; + + /* TODO: filter by device */ + mutex_lock(&cfg80211_drv_mutex); + list_for_each_entry(dev, &cfg80211_drv_list, list) { + if (exit) + break; + if (++wp_idx < wp_start) + continue; + if_idx = 0; + + mutex_lock(&dev->devlist_mtx); + list_for_each_entry(wdev, &dev->netdev_list, list) { + if (exit) + break; + if (++if_idx < if_start) + continue; + if (!dev->ops->dump_station) + continue; + + for (;; ++sta_idx) { + rtnl_lock(); + err = dev->ops->dump_station(&dev->wiphy, + wdev->netdev, sta_idx, mac_addr, + &sinfo); + rtnl_unlock(); + if (err) { + sta_idx = 0; + break; + } + if (nl80211_send_station(skb, + NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + wdev->netdev, mac_addr, + &sinfo) < 0) { + exit = 1; + break; + } + } + } + mutex_unlock(&dev->devlist_mtx); + } + mutex_unlock(&cfg80211_drv_mutex); + + cb->args[0] = wp_idx; + cb->args[1] = if_idx; + cb->args[2] = sta_idx; + + return skb->len; +} static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *drv; int err; struct net_device *dev; - struct station_stats stats; + struct station_info sinfo; struct sk_buff *msg; u8 *mac_addr = NULL; - memset(&stats, 0, sizeof(stats)); + memset(&sinfo, 0, sizeof(sinfo)); if (!info->attrs[NL80211_ATTR_MAC]) return -EINVAL; @@ -926,15 +1021,18 @@ static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info) } rtnl_lock(); - err = drv->ops->get_station(&drv->wiphy, dev, mac_addr, &stats); + err = drv->ops->get_station(&drv->wiphy, dev, mac_addr, &sinfo); rtnl_unlock(); + if (err) + goto out; + msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); if (!msg) goto out; if (nl80211_send_station(msg, info->snd_pid, info->snd_seq, 0, - dev, mac_addr, &stats) < 0) + dev, mac_addr, &sinfo) < 0) goto out_free; err = genlmsg_unicast(msg, info->snd_pid); @@ -1005,6 +1103,10 @@ static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info) ¶ms.station_flags)) return -EINVAL; + if (info->attrs[NL80211_ATTR_STA_PLINK_ACTION]) + params.plink_action = + nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_ACTION]); + err = get_drv_dev_by_info_ifindex(info, &drv, &dev); if (err) return err; @@ -1119,6 +1221,273 @@ static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info) return err; } +static int nl80211_send_mpath(struct sk_buff *msg, u32 pid, u32 seq, + int flags, struct net_device *dev, + u8 *dst, u8 *next_hop, + struct mpath_info *pinfo) +{ + void *hdr; + struct nlattr *pinfoattr; + + hdr = nl80211hdr_put(msg, pid, seq, flags, NL80211_CMD_NEW_STATION); + if (!hdr) + return -1; + + NLA_PUT_U32(msg, NL80211_ATTR_IFINDEX, dev->ifindex); + NLA_PUT(msg, NL80211_ATTR_MAC, ETH_ALEN, dst); + NLA_PUT(msg, NL80211_ATTR_MPATH_NEXT_HOP, ETH_ALEN, next_hop); + + pinfoattr = nla_nest_start(msg, NL80211_ATTR_MPATH_INFO); + if (!pinfoattr) + goto nla_put_failure; + if (pinfo->filled & MPATH_INFO_FRAME_QLEN) + NLA_PUT_U32(msg, NL80211_MPATH_INFO_FRAME_QLEN, + pinfo->frame_qlen); + if (pinfo->filled & MPATH_INFO_DSN) + NLA_PUT_U32(msg, NL80211_MPATH_INFO_DSN, + pinfo->dsn); + if (pinfo->filled & MPATH_INFO_METRIC) + NLA_PUT_U32(msg, NL80211_MPATH_INFO_METRIC, + pinfo->metric); + if (pinfo->filled & MPATH_INFO_EXPTIME) + NLA_PUT_U32(msg, NL80211_MPATH_INFO_EXPTIME, + pinfo->exptime); + if (pinfo->filled & MPATH_INFO_FLAGS) + NLA_PUT_U8(msg, NL80211_MPATH_INFO_FLAGS, + pinfo->flags); + if (pinfo->filled & MPATH_INFO_DISCOVERY_TIMEOUT) + NLA_PUT_U32(msg, NL80211_MPATH_INFO_DISCOVERY_TIMEOUT, + pinfo->discovery_timeout); + if (pinfo->filled & MPATH_INFO_DISCOVERY_RETRIES) + NLA_PUT_U8(msg, NL80211_MPATH_INFO_DISCOVERY_RETRIES, + pinfo->discovery_retries); + + nla_nest_end(msg, pinfoattr); + + return genlmsg_end(msg, hdr); + + nla_put_failure: + return genlmsg_cancel(msg, hdr); +} + +static int nl80211_dump_mpath(struct sk_buff *skb, + struct netlink_callback *cb) +{ + int wp_idx = 0; + int if_idx = 0; + int sta_idx = cb->args[2]; + int wp_start = cb->args[0]; + int if_start = cb->args[1]; + struct mpath_info pinfo; + struct cfg80211_registered_device *dev; + struct wireless_dev *wdev; + u8 dst[ETH_ALEN]; + u8 next_hop[ETH_ALEN]; + int err; + int exit = 0; + + /* TODO: filter by device */ + mutex_lock(&cfg80211_drv_mutex); + list_for_each_entry(dev, &cfg80211_drv_list, list) { + if (exit) + break; + if (++wp_idx < wp_start) + continue; + if_idx = 0; + + mutex_lock(&dev->devlist_mtx); + list_for_each_entry(wdev, &dev->netdev_list, list) { + if (exit) + break; + if (++if_idx < if_start) + continue; + if (!dev->ops->dump_mpath) + continue; + + for (;; ++sta_idx) { + rtnl_lock(); + err = dev->ops->dump_mpath(&dev->wiphy, + wdev->netdev, sta_idx, dst, + next_hop, &pinfo); + rtnl_unlock(); + if (err) { + sta_idx = 0; + break; + } + if (nl80211_send_mpath(skb, + NETLINK_CB(cb->skb).pid, + cb->nlh->nlmsg_seq, NLM_F_MULTI, + wdev->netdev, dst, next_hop, + &pinfo) < 0) { + exit = 1; + break; + } + } + } + mutex_unlock(&dev->devlist_mtx); + } + mutex_unlock(&cfg80211_drv_mutex); + + cb->args[0] = wp_idx; + cb->args[1] = if_idx; + cb->args[2] = sta_idx; + + return skb->len; +} + +static int nl80211_get_mpath(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *drv; + int err; + struct net_device *dev; + struct mpath_info pinfo; + struct sk_buff *msg; + u8 *dst = NULL; + u8 next_hop[ETH_ALEN]; + + memset(&pinfo, 0, sizeof(pinfo)); + + if (!info->attrs[NL80211_ATTR_MAC]) + return -EINVAL; + + dst = nla_data(info->attrs[NL80211_ATTR_MAC]); + + err = get_drv_dev_by_info_ifindex(info, &drv, &dev); + if (err) + return err; + + if (!drv->ops->get_mpath) { + err = -EOPNOTSUPP; + goto out; + } + + rtnl_lock(); + err = drv->ops->get_mpath(&drv->wiphy, dev, dst, next_hop, &pinfo); + rtnl_unlock(); + + if (err) + goto out; + + msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!msg) + goto out; + + if (nl80211_send_mpath(msg, info->snd_pid, info->snd_seq, 0, + dev, dst, next_hop, &pinfo) < 0) + goto out_free; + + err = genlmsg_unicast(msg, info->snd_pid); + goto out; + + out_free: + nlmsg_free(msg); + + out: + cfg80211_put_dev(drv); + dev_put(dev); + return err; +} + +static int nl80211_set_mpath(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *drv; + int err; + struct net_device *dev; + u8 *dst = NULL; + u8 *next_hop = NULL; + + if (!info->attrs[NL80211_ATTR_MAC]) + return -EINVAL; + + if (!info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]) + return -EINVAL; + + dst = nla_data(info->attrs[NL80211_ATTR_MAC]); + next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]); + + err = get_drv_dev_by_info_ifindex(info, &drv, &dev); + if (err) + return err; + + if (!drv->ops->change_mpath) { + err = -EOPNOTSUPP; + goto out; + } + + rtnl_lock(); + err = drv->ops->change_mpath(&drv->wiphy, dev, dst, next_hop); + rtnl_unlock(); + + out: + cfg80211_put_dev(drv); + dev_put(dev); + return err; +} +static int nl80211_new_mpath(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *drv; + int err; + struct net_device *dev; + u8 *dst = NULL; + u8 *next_hop = NULL; + + if (!info->attrs[NL80211_ATTR_MAC]) + return -EINVAL; + + if (!info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]) + return -EINVAL; + + dst = nla_data(info->attrs[NL80211_ATTR_MAC]); + next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]); + + err = get_drv_dev_by_info_ifindex(info, &drv, &dev); + if (err) + return err; + + if (!drv->ops->add_mpath) { + err = -EOPNOTSUPP; + goto out; + } + + rtnl_lock(); + err = drv->ops->add_mpath(&drv->wiphy, dev, dst, next_hop); + rtnl_unlock(); + + out: + cfg80211_put_dev(drv); + dev_put(dev); + return err; +} + +static int nl80211_del_mpath(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *drv; + int err; + struct net_device *dev; + u8 *dst = NULL; + + if (info->attrs[NL80211_ATTR_MAC]) + dst = nla_data(info->attrs[NL80211_ATTR_MAC]); + + err = get_drv_dev_by_info_ifindex(info, &drv, &dev); + if (err) + return err; + + if (!drv->ops->del_mpath) { + err = -EOPNOTSUPP; + goto out; + } + + rtnl_lock(); + err = drv->ops->del_mpath(&drv->wiphy, dev, dst); + rtnl_unlock(); + + out: + cfg80211_put_dev(drv); + dev_put(dev); + return err; +} + static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_WIPHY, @@ -1203,7 +1572,7 @@ static struct genl_ops nl80211_ops[] = { { .cmd = NL80211_CMD_GET_STATION, .doit = nl80211_get_station, - /* TODO: implement dumpit */ + .dumpit = nl80211_dump_station, .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, }, @@ -1225,6 +1594,31 @@ static struct genl_ops nl80211_ops[] = { .policy = nl80211_policy, .flags = GENL_ADMIN_PERM, }, + { + .cmd = NL80211_CMD_GET_MPATH, + .doit = nl80211_get_mpath, + .dumpit = nl80211_dump_mpath, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = NL80211_CMD_SET_MPATH, + .doit = nl80211_set_mpath, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = NL80211_CMD_NEW_MPATH, + .doit = nl80211_new_mpath, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = NL80211_CMD_DEL_MPATH, + .doit = nl80211_del_mpath, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + }, }; /* multicast groups */ -- cgit v1.2.3 From 6032f934c818e5c3435c9f17274fe1983f53c6b4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 23 Feb 2008 15:17:07 +0100 Subject: mac80211: add mesh interface type This adds the mesh interface type. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 1 + net/mac80211/ieee80211.c | 1 + net/mac80211/ieee80211_iface.c | 1 + net/mac80211/rx.c | 10 ++++++++++ net/mac80211/util.c | 1 + 5 files changed, 14 insertions(+) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 7a80c3981237..934cc25f757a 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -443,6 +443,7 @@ enum ieee80211_if_types { IEEE80211_IF_TYPE_AP, IEEE80211_IF_TYPE_STA, IEEE80211_IF_TYPE_IBSS, + IEEE80211_IF_TYPE_MESH_POINT, IEEE80211_IF_TYPE_MNTR, IEEE80211_IF_TYPE_WDS, IEEE80211_IF_TYPE_VLAN, diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index 2133c9fd27a4..1ddb8e1b6ab6 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -258,6 +258,7 @@ static int ieee80211_open(struct net_device *dev) case IEEE80211_IF_TYPE_STA: case IEEE80211_IF_TYPE_MNTR: case IEEE80211_IF_TYPE_IBSS: + case IEEE80211_IF_TYPE_MESH_POINT: /* no special treatment */ break; case IEEE80211_IF_TYPE_INVALID: diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c index 677705046c6d..9523aeb71032 100644 --- a/net/mac80211/ieee80211_iface.c +++ b/net/mac80211/ieee80211_iface.c @@ -235,6 +235,7 @@ void ieee80211_if_reinit(struct net_device *dev) #endif /* CONFIG_MAC80211_VERBOSE_DEBUG */ } break; + case IEEE80211_IF_TYPE_MESH_POINT: case IEEE80211_IF_TYPE_STA: case IEEE80211_IF_TYPE_IBSS: kfree(sdata->u.sta.extra_ie); diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 48574f6c0e74..b7eeae0d3956 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1713,6 +1713,16 @@ static int prepare_for_handlers(struct ieee80211_sub_if_data *sdata, rx->sta = ieee80211_ibss_add_sta(sdata->dev, rx->skb, bssid, hdr->addr2); break; + case IEEE80211_IF_TYPE_MESH_POINT: + if (!multicast && + compare_ether_addr(sdata->dev->dev_addr, + hdr->addr1) != 0) { + if (!(sdata->dev->flags & IFF_PROMISC)) + return 0; + + rx->flags &= ~IEEE80211_TXRXD_RXRA_MATCH; + } + break; case IEEE80211_IF_TYPE_VLAN: case IEEE80211_IF_TYPE_AP: if (!bssid) { diff --git a/net/mac80211/util.c b/net/mac80211/util.c index f64804fed0a9..790c32f894c0 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -382,6 +382,7 @@ void ieee80211_iterate_active_interfaces( case IEEE80211_IF_TYPE_STA: case IEEE80211_IF_TYPE_IBSS: case IEEE80211_IF_TYPE_WDS: + case IEEE80211_IF_TYPE_MESH_POINT: break; } if (sdata->dev == local->mdev) -- cgit v1.2.3 From 902acc7896d7649fb30e4b22bd4e643c7f34b02c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 23 Feb 2008 15:17:19 +0100 Subject: mac80211: clean up mesh code Various cleanups, reducing the #ifdef mess and other things. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 8 +++ net/mac80211/cfg.c | 32 +++-------- net/mac80211/ieee80211.c | 9 +-- net/mac80211/ieee80211_i.h | 56 ++++++++++++++++--- net/mac80211/ieee80211_iface.c | 66 ++++------------------ net/mac80211/ieee80211_sta.c | 72 +++++++++++------------- net/mac80211/mesh.c | 67 ++++++++++++++++++++++ net/mac80211/mesh.h | 13 ++++- net/mac80211/mesh_plink.c | 28 +++++++--- net/mac80211/rc80211_pid_algo.c | 3 - net/mac80211/rx.c | 28 +++++----- net/mac80211/sta_info.c | 13 +---- net/mac80211/sta_info.h | 2 - net/mac80211/tx.c | 120 +++++++++++++++++----------------------- net/mac80211/util.c | 32 ----------- 15 files changed, 277 insertions(+), 272 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 934cc25f757a..6aca472d7a02 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -465,6 +465,14 @@ struct ieee80211_vif { u8 drv_priv[0] __attribute__((__aligned__(sizeof(void *)))); }; +static inline bool ieee80211_vif_is_mesh(struct ieee80211_vif *vif) +{ +#ifdef CONFIG_MAC80211_MESH + return vif->type == IEEE80211_IF_TYPE_MESH_POINT; +#endif + return false; +} + /** * struct ieee80211_if_init_conf - initial configuration of an interface * diff --git a/net/mac80211/cfg.c b/net/mac80211/cfg.c index b1befac1736a..6ac49231efa9 100644 --- a/net/mac80211/cfg.c +++ b/net/mac80211/cfg.c @@ -15,9 +15,7 @@ #include "ieee80211_i.h" #include "cfg.h" #include "ieee80211_rate.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif #define DEFAULT_RATES 0 @@ -119,14 +117,10 @@ static int ieee80211_change_iface(struct wiphy *wiphy, int ifindex, ieee80211_if_reinit(dev); ieee80211_if_set_type(dev, itype); -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT && - params->mesh_id_len) { - sdata->u.sta.mesh_id_len = params->mesh_id_len; - memcpy(sdata->u.sta.mesh_id, params->mesh_id, - params->mesh_id_len); - } -#endif + if (ieee80211_vif_is_mesh(&sdata->vif) && params->mesh_id_len) + ieee80211_if_sta_set_mesh_id(&sdata->u.sta, + params->mesh_id_len, + params->mesh_id); if (sdata->vif.type != IEEE80211_IF_TYPE_MNTR || !flags) return 0; @@ -317,9 +311,7 @@ static int ieee80211_config_default_key(struct wiphy *wiphy, static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo) { -#ifdef CONFIG_MAC80211_MESH struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); -#endif sinfo->filled = STATION_INFO_INACTIVE_TIME | STATION_INFO_RX_BYTES | @@ -329,8 +321,8 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo) sinfo->rx_bytes = sta->rx_bytes; sinfo->tx_bytes = sta->tx_bytes; + if (ieee80211_vif_is_mesh(&sdata->vif)) { #ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) { sinfo->filled |= STATION_INFO_LLID | STATION_INFO_PLID | STATION_INFO_PLINK_STATE; @@ -338,8 +330,8 @@ static void sta_set_sinfo(struct sta_info *sta, struct station_info *sinfo) sinfo->llid = le16_to_cpu(sta->llid); sinfo->plid = le16_to_cpu(sta->plid); sinfo->plink_state = sta->plink_state; - } #endif + } } @@ -580,9 +572,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, u32 rates; int i, j; struct ieee80211_supported_band *sband; -#ifdef CONFIG_MAC80211_MESH struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); -#endif if (params->station_flags & STATION_FLAG_CHANGED) { sta->flags &= ~WLAN_STA_AUTHORIZED; @@ -621,9 +611,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, sta->supp_rates[local->oper_channel->band] = rates; } -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT && - params->plink_action) + if (ieee80211_vif_is_mesh(&sdata->vif) && params->plink_action) { switch (params->plink_action) { case PLINK_ACTION_OPEN: mesh_plink_open(sta); @@ -632,7 +620,7 @@ static void sta_apply_parameters(struct ieee80211_local *local, mesh_plink_block(sta); break; } -#endif + } } static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev, @@ -655,11 +643,9 @@ static int ieee80211_add_station(struct wiphy *wiphy, struct net_device *dev, } else sdata = IEEE80211_DEV_TO_SUB_IF(dev); -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) + if (ieee80211_vif_is_mesh(&sdata->vif)) sta = mesh_plink_add(mac, DEFAULT_RATES, dev); else -#endif sta = sta_info_add(local, dev, mac, GFP_KERNEL); if (IS_ERR(sta)) diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index 7106d651f4f9..727af295c969 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -26,9 +26,7 @@ #include "ieee80211_i.h" #include "ieee80211_rate.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif #include "wep.h" #include "wme.h" #include "aes_ccm.h" @@ -938,11 +936,9 @@ static int __ieee80211_if_config(struct net_device *dev, conf.bssid = sdata->u.sta.bssid; conf.ssid = sdata->u.sta.ssid; conf.ssid_len = sdata->u.sta.ssid_len; -#ifdef CONFIG_MAC80211_MESH - } else if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) { + } else if (ieee80211_vif_is_mesh(&sdata->vif)) { conf.beacon = beacon; ieee80211_start_mesh(dev); -#endif } else if (sdata->vif.type == IEEE80211_IF_TYPE_AP) { conf.ssid = sdata->u.ap.ssid; conf.ssid_len = sdata->u.ap.ssid_len; @@ -1824,10 +1820,9 @@ static void __exit ieee80211_exit(void) rc80211_simple_exit(); rc80211_pid_exit(); -#ifdef CONFIG_MAC80211_MESH if (mesh_allocated) ieee80211s_stop(); -#endif + ieee80211_wme_unregister(); ieee80211_debugfs_netdev_exit(); } diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 49466b6996d1..7394c9b783b8 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -93,9 +93,8 @@ struct ieee80211_sta_bss { #ifdef CONFIG_MAC80211_MESH u8 *mesh_id; size_t mesh_id_len; -#endif - /* mesh_cfg left out the ifdef to reduce clutter on bss handling */ u8 *mesh_cfg; +#endif #define IEEE80211_MAX_SUPP_RATES 32 u8 supp_rates[IEEE80211_MAX_SUPP_RATES]; size_t supp_rates_len; @@ -113,6 +112,30 @@ struct ieee80211_sta_bss { u8 erp_value; }; +static inline u8 *bss_mesh_cfg(struct ieee80211_sta_bss *bss) +{ +#ifdef CONFIG_MAC80211_MESH + return bss->mesh_cfg; +#endif + return NULL; +} + +static inline u8 *bss_mesh_id(struct ieee80211_sta_bss *bss) +{ +#ifdef CONFIG_MAC80211_MESH + return bss->mesh_id; +#endif + return NULL; +} + +static inline u8 bss_mesh_id_len(struct ieee80211_sta_bss *bss) +{ +#ifdef CONFIG_MAC80211_MESH + return bss->mesh_id_len; +#endif + return 0; +} + typedef unsigned __bitwise__ ieee80211_tx_result; #define TX_CONTINUE ((__force ieee80211_tx_result) 0u) @@ -233,7 +256,6 @@ struct ieee80211_if_vlan { struct list_head list; }; -#ifdef CONFIG_MAC80211_MESH struct mesh_stats { __u32 fwded_frames; /* Mesh forwarded frames */ __u32 dropped_frames_ttl; /* Not transmitted since mesh_ttl == 0*/ @@ -249,7 +271,6 @@ struct mesh_preq_queue { u8 flags; }; - struct mesh_config { /* Timeouts in ms */ /* Mesh plink management parameters */ @@ -268,7 +289,7 @@ struct mesh_config { u32 path_refresh_time; u16 min_discovery_timeout; }; -#endif + /* flags used in struct ieee80211_if_sta.flags */ #define IEEE80211_STA_SSID_SET BIT(0) @@ -361,6 +382,22 @@ struct ieee80211_if_sta { int num_beacons; /* number of TXed beacon frames by this STA */ }; +static inline void ieee80211_if_sta_set_mesh_id(struct ieee80211_if_sta *ifsta, + u8 mesh_id_len, u8 *mesh_id) +{ +#ifdef CONFIG_MAC80211_MESH + ifsta->mesh_id_len = mesh_id_len; + memcpy(ifsta->mesh_id, mesh_id, mesh_id_len); +#endif +} + +#ifdef CONFIG_MAC80211_MESH +#define IEEE80211_IFSTA_MESH_CTR_INC(sta, name) \ + do { (sta)->mshstats.name++; } while (0) +#else +#define IEEE80211_IFSTA_MESH_CTR_INC(sta, name) \ + do { } while (0) +#endif /* flags used in struct ieee80211_sub_if_data.flags */ #define IEEE80211_SDATA_ALLMULTI BIT(0) @@ -472,7 +509,7 @@ struct ieee80211_sub_if_data { struct dentry *dropped_frames_ttl; struct dentry *dropped_frames_no_route; struct dentry *estab_plinks; - struct timer_list mesh_path_timer; + struct timer_list mesh_path_timer; } mesh_stats; struct dentry *mesh_config_dir; @@ -884,12 +921,17 @@ void sta_addba_resp_timer_expired(unsigned long data); u64 ieee80211_sta_get_rates(struct ieee80211_local *local, struct ieee802_11_elems *elems, enum ieee80211_band band); -void ieee80211_start_mesh(struct net_device *dev); void ieee80211_sta_tx(struct net_device *dev, struct sk_buff *skb, int encrypt); void ieee802_11_parse_elems(u8 *start, size_t len, struct ieee802_11_elems *elems); +#ifdef CONFIG_MAC80211_MESH +void ieee80211_start_mesh(struct net_device *dev); +#else +static inline void ieee80211_start_mesh(struct net_device *dev) +{} +#endif /* ieee80211_iface.c */ int ieee80211_if_add(struct net_device *dev, const char *name, diff --git a/net/mac80211/ieee80211_iface.c b/net/mac80211/ieee80211_iface.c index c2f92b78bfc9..b0f17a2b1a42 100644 --- a/net/mac80211/ieee80211_iface.c +++ b/net/mac80211/ieee80211_iface.c @@ -15,9 +15,7 @@ #include "ieee80211_i.h" #include "sta_info.h" #include "debugfs_netdev.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif void ieee80211_if_sdata_init(struct ieee80211_sub_if_data *sdata) { @@ -82,14 +80,11 @@ int ieee80211_if_add(struct net_device *dev, const char *name, ieee80211_debugfs_add_netdev(sdata); ieee80211_if_set_type(ndev, type); -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT && - params && params->mesh_id_len) { - sdata->u.sta.mesh_id_len = params->mesh_id_len; - memcpy(sdata->u.sta.mesh_id, params->mesh_id, - params->mesh_id_len); - } -#endif + if (ieee80211_vif_is_mesh(&sdata->vif) && + params && params->mesh_id_len) + ieee80211_if_sta_set_mesh_id(&sdata->u.sta, + params->mesh_id_len, + params->mesh_id); /* we're under RTNL so all this is fine */ if (unlikely(local->reg_state == IEEE80211_DEV_UNREGISTERED)) { @@ -170,47 +165,8 @@ void ieee80211_if_set_type(struct net_device *dev, int type) msdata = IEEE80211_DEV_TO_SUB_IF(sdata->local->mdev); sdata->bss = &msdata->u.ap; -#ifdef CONFIG_MAC80211_MESH - if (type == IEEE80211_IF_TYPE_MESH_POINT) { - ifsta->mshcfg.dot11MeshRetryTimeout = MESH_RET_T; - ifsta->mshcfg.dot11MeshConfirmTimeout = MESH_CONF_T; - ifsta->mshcfg.dot11MeshHoldingTimeout = MESH_HOLD_T; - ifsta->mshcfg.dot11MeshMaxRetries = MESH_MAX_RETR; - ifsta->mshcfg.dot11MeshTTL = MESH_TTL; - ifsta->mshcfg.auto_open_plinks = true; - ifsta->mshcfg.dot11MeshMaxPeerLinks = - MESH_MAX_ESTAB_PLINKS; - ifsta->mshcfg.dot11MeshHWMPactivePathTimeout = - MESH_PATH_TIMEOUT; - ifsta->mshcfg.dot11MeshHWMPpreqMinInterval = - MESH_PREQ_MIN_INT; - ifsta->mshcfg.dot11MeshHWMPnetDiameterTraversalTime = - MESH_DIAM_TRAVERSAL_TIME; - ifsta->mshcfg.dot11MeshHWMPmaxPREQretries = - MESH_MAX_PREQ_RETRIES; - ifsta->mshcfg.path_refresh_time = - MESH_PATH_REFRESH_TIME; - ifsta->mshcfg.min_discovery_timeout = - MESH_MIN_DISCOVERY_TIMEOUT; - ifsta->accepting_plinks = true; - ifsta->preq_id = 0; - ifsta->dsn = 0; - atomic_set(&ifsta->mpaths, 0); - mesh_rmc_init(dev); - ifsta->last_preq = jiffies; - /* Allocate all mesh structures when creating the first - * mesh interface. - */ - if (!mesh_allocated) - ieee80211s_init(); - mesh_ids_set_default(ifsta); - setup_timer(&ifsta->mesh_path_timer, - ieee80211_mesh_path_timer, - (unsigned long) sdata); - INIT_LIST_HEAD(&ifsta->preq_queue.list); - spin_lock_init(&ifsta->mesh_preq_queue_lock); - } -#endif + if (ieee80211_vif_is_mesh(&sdata->vif)) + ieee80211_mesh_init_sdata(sdata); break; } case IEEE80211_IF_TYPE_MNTR: @@ -240,6 +196,10 @@ void ieee80211_if_reinit(struct net_device *dev) ieee80211_if_sdata_deinit(sdata); + /* Need to handle mesh specially to allow eliding the function call */ + if (ieee80211_vif_is_mesh(&sdata->vif)) + mesh_rmc_free(dev); + switch (sdata->vif.type) { case IEEE80211_IF_TYPE_INVALID: /* cannot happen */ @@ -292,10 +252,6 @@ void ieee80211_if_reinit(struct net_device *dev) } break; case IEEE80211_IF_TYPE_MESH_POINT: -#ifdef CONFIG_MAC80211_MESH - mesh_rmc_free(dev); -#endif - /* fall through */ case IEEE80211_IF_TYPE_STA: case IEEE80211_IF_TYPE_IBSS: kfree(sdata->u.sta.extra_ie); diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index d2dedcb5a954..9f933aeca719 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -31,9 +31,7 @@ #include "ieee80211_i.h" #include "ieee80211_rate.h" #include "ieee80211_led.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif #define IEEE80211_AUTH_TIMEOUT (HZ / 5) #define IEEE80211_AUTH_MAX_TRIES 3 @@ -1897,12 +1895,13 @@ static void __ieee80211_rx_bss_hash_add(struct net_device *dev, { struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); u8 hash_idx; -#ifdef CONFIG_MAC80211_MESH - if (bss->mesh_cfg) - hash_idx = mesh_id_hash(bss->mesh_id, bss->mesh_id_len); + + if (bss_mesh_cfg(bss)) + hash_idx = mesh_id_hash(bss_mesh_id(bss), + bss_mesh_id_len(bss)); else -#endif hash_idx = STA_HASH(bss->bssid); + bss->hnext = local->sta_bss_hash[hash_idx]; local->sta_bss_hash[hash_idx] = bss; } @@ -1967,7 +1966,8 @@ ieee80211_rx_bss_get(struct net_device *dev, u8 *bssid, int freq, spin_lock_bh(&local->sta_bss_lock); bss = local->sta_bss_hash[STA_HASH(bssid)]; while (bss) { - if (!bss->mesh_cfg && !memcmp(bss->bssid, bssid, ETH_ALEN) && + if (!bss_mesh_cfg(bss) && + !memcmp(bss->bssid, bssid, ETH_ALEN) && bss->freq == freq && bss->ssid_len == ssid_len && (ssid_len == 0 || !memcmp(bss->ssid, ssid, ssid_len))) { @@ -1991,8 +1991,8 @@ ieee80211_rx_mesh_bss_get(struct net_device *dev, u8 *mesh_id, int mesh_id_len, spin_lock_bh(&local->sta_bss_lock); bss = local->sta_bss_hash[mesh_id_hash(mesh_id, mesh_id_len)]; while (bss) { - if (bss->mesh_cfg && - !memcmp(bss->mesh_cfg, mesh_cfg, MESH_CFG_CMP_LEN) && + if (bss_mesh_cfg(bss) && + !memcmp(bss_mesh_cfg(bss), mesh_cfg, MESH_CFG_CMP_LEN) && bss->freq == freq && mesh_id_len == bss->mesh_id_len && (mesh_id_len == 0 || !memcmp(bss->mesh_id, mesh_id, @@ -2053,10 +2053,8 @@ static void ieee80211_rx_bss_free(struct ieee80211_sta_bss *bss) kfree(bss->rsn_ie); kfree(bss->wmm_ie); kfree(bss->ht_ie); -#ifdef CONFIG_MAC80211_MESH - kfree(bss->mesh_id); - kfree(bss->mesh_cfg); -#endif + kfree(bss_mesh_id(bss)); + kfree(bss_mesh_cfg(bss)); kfree(bss); } @@ -2322,16 +2320,14 @@ static void ieee80211_rx_bss_info(struct net_device *dev, beacon_timestamp = le64_to_cpu(mgmt->u.beacon.timestamp); ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems); -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT && elems.mesh_id - && elems.mesh_config) - if (mesh_matches_local(&elems, dev)) { - u64 rates = ieee80211_sta_get_rates(local, &elems, - rx_status->band); - mesh_neighbour_update(mgmt->sa, rates, dev, - mesh_peer_accepts_plinks(&elems, dev)); - } -#endif + if (ieee80211_vif_is_mesh(&sdata->vif) && elems.mesh_id && + elems.mesh_config && mesh_matches_local(&elems, dev)) { + u64 rates = ieee80211_sta_get_rates(local, &elems, + rx_status->band); + + mesh_neighbour_update(mgmt->sa, rates, dev, + mesh_peer_accepts_plinks(&elems, dev)); + } if (sdata->vif.type == IEEE80211_IF_TYPE_IBSS && elems.supp_rates && memcmp(mgmt->bssid, sdata->u.sta.bssid, ETH_ALEN) == 0 && @@ -2712,9 +2708,7 @@ static void ieee80211_rx_mgmt_action(struct net_device *dev, size_t len, struct ieee80211_rx_status *rx_status) { -#ifdef CONFIG_MAC80211_MESH struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); -#endif if (len < IEEE80211_MIN_ACTION_SIZE) return; @@ -2747,17 +2741,14 @@ static void ieee80211_rx_mgmt_action(struct net_device *dev, break; } break; -#ifdef CONFIG_MAC80211_MESH case PLINK_CATEGORY: - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) + if (ieee80211_vif_is_mesh(&sdata->vif)) mesh_rx_plink_frame(dev, mgmt, len, rx_status); break; - case MESH_PATH_SEL_CATEGORY: - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) + if (ieee80211_vif_is_mesh(&sdata->vif)) mesh_rx_path_sel_frame(dev, mgmt, len); break; -#endif default: if (net_ratelimit()) printk(KERN_DEBUG "%s: Rx unknown action frame - " @@ -3027,8 +3018,9 @@ void ieee80211_sta_work(struct work_struct *work) ieee80211_sta_rx_queued_mgmt(dev, skb); #ifdef CONFIG_MAC80211_MESH - if (ifsta->preq_queue_len && time_after(jiffies, ifsta->last_preq + - msecs_to_jiffies(ifsta->mshcfg.dot11MeshHWMPpreqMinInterval))) + if (ifsta->preq_queue_len && + time_after(jiffies, + ifsta->last_preq + msecs_to_jiffies(ifsta->mshcfg.dot11MeshHWMPpreqMinInterval))) mesh_path_start_discovery(dev); #endif @@ -3810,13 +3802,11 @@ ieee80211_sta_scan_result(struct net_device *dev, memset(&iwe, 0, sizeof(iwe)); iwe.cmd = SIOCGIWESSID; - if (bss->mesh_cfg) { -#ifdef CONFIG_MAC80211_MESH - iwe.u.data.length = bss->mesh_id_len; + if (bss_mesh_cfg(bss)) { + iwe.u.data.length = bss_mesh_id_len(bss); iwe.u.data.flags = 1; current_ev = iwe_stream_add_point(current_ev, end_buf, &iwe, - bss->mesh_id); -#endif + bss_mesh_id(bss)); } else { iwe.u.data.length = bss->ssid_len; iwe.u.data.flags = 1; @@ -3825,10 +3815,10 @@ ieee80211_sta_scan_result(struct net_device *dev, } if (bss->capability & (WLAN_CAPABILITY_ESS | WLAN_CAPABILITY_IBSS - || bss->mesh_cfg)) { + || bss_mesh_cfg(bss))) { memset(&iwe, 0, sizeof(iwe)); iwe.cmd = SIOCGIWMODE; - if (bss->mesh_cfg) + if (bss_mesh_cfg(bss)) iwe.u.mode = IW_MODE_MESH; else if (bss->capability & WLAN_CAPABILITY_ESS) iwe.u.mode = IW_MODE_MASTER; @@ -3919,9 +3909,9 @@ ieee80211_sta_scan_result(struct net_device *dev, } } - if (bss->mesh_cfg) { + if (bss_mesh_cfg(bss)) { char *buf; - u8 *cfg = bss->mesh_cfg; + u8 *cfg = bss_mesh_cfg(bss); buf = kmalloc(200, GFP_ATOMIC); if (buf) { memset(&iwe, 0, sizeof(iwe)); diff --git a/net/mac80211/mesh.c b/net/mac80211/mesh.c index 8ff533005d92..ebe1a7a80bad 100644 --- a/net/mac80211/mesh.c +++ b/net/mac80211/mesh.c @@ -381,3 +381,70 @@ endgrow: else return newtbl; } + +/** + * ieee80211_new_mesh_header - create a new mesh header + * @meshhdr: uninitialized mesh header + * @sdata: mesh interface to be used + * + * Return the header length. + */ +int ieee80211_new_mesh_header(struct ieee80211s_hdr *meshhdr, + struct ieee80211_sub_if_data *sdata) +{ + meshhdr->flags = 0; + meshhdr->ttl = sdata->u.sta.mshcfg.dot11MeshTTL; + + meshhdr->seqnum[0] = sdata->u.sta.mesh_seqnum[0]++; + meshhdr->seqnum[1] = sdata->u.sta.mesh_seqnum[1]; + meshhdr->seqnum[2] = sdata->u.sta.mesh_seqnum[2]; + + if (sdata->u.sta.mesh_seqnum[0] == 0) { + sdata->u.sta.mesh_seqnum[1]++; + if (sdata->u.sta.mesh_seqnum[1] == 0) + sdata->u.sta.mesh_seqnum[2]++; + } + + return 5; +} + +void ieee80211_mesh_init_sdata(struct ieee80211_sub_if_data *sdata) +{ + struct ieee80211_if_sta *ifsta = &sdata->u.sta; + + ifsta->mshcfg.dot11MeshRetryTimeout = MESH_RET_T; + ifsta->mshcfg.dot11MeshConfirmTimeout = MESH_CONF_T; + ifsta->mshcfg.dot11MeshHoldingTimeout = MESH_HOLD_T; + ifsta->mshcfg.dot11MeshMaxRetries = MESH_MAX_RETR; + ifsta->mshcfg.dot11MeshTTL = MESH_TTL; + ifsta->mshcfg.auto_open_plinks = true; + ifsta->mshcfg.dot11MeshMaxPeerLinks = + MESH_MAX_ESTAB_PLINKS; + ifsta->mshcfg.dot11MeshHWMPactivePathTimeout = + MESH_PATH_TIMEOUT; + ifsta->mshcfg.dot11MeshHWMPpreqMinInterval = + MESH_PREQ_MIN_INT; + ifsta->mshcfg.dot11MeshHWMPnetDiameterTraversalTime = + MESH_DIAM_TRAVERSAL_TIME; + ifsta->mshcfg.dot11MeshHWMPmaxPREQretries = + MESH_MAX_PREQ_RETRIES; + ifsta->mshcfg.path_refresh_time = + MESH_PATH_REFRESH_TIME; + ifsta->mshcfg.min_discovery_timeout = + MESH_MIN_DISCOVERY_TIMEOUT; + ifsta->accepting_plinks = true; + ifsta->preq_id = 0; + ifsta->dsn = 0; + atomic_set(&ifsta->mpaths, 0); + mesh_rmc_init(sdata->dev); + ifsta->last_preq = jiffies; + /* Allocate all mesh structures when creating the first mesh interface. */ + if (!mesh_allocated) + ieee80211s_init(); + mesh_ids_set_default(ifsta); + setup_timer(&ifsta->mesh_path_timer, + ieee80211_mesh_path_timer, + (unsigned long) sdata); + INIT_LIST_HEAD(&ifsta->preq_queue.list); + spin_lock_init(&ifsta->mesh_preq_queue_lock); +} diff --git a/net/mac80211/mesh.h b/net/mac80211/mesh.h index ac8923793908..d565b3fb9e6a 100644 --- a/net/mac80211/mesh.h +++ b/net/mac80211/mesh.h @@ -11,10 +11,10 @@ #ifndef IEEE80211S_H #define IEEE80211S_H -#include "ieee80211_i.h" +#include #include +#include "ieee80211_i.h" -extern int mesh_allocated; /* Data structures */ @@ -211,6 +211,8 @@ void mesh_rmc_free(struct net_device *dev); int mesh_rmc_init(struct net_device *dev); void ieee80211s_init(void); void ieee80211s_stop(void); +void ieee80211_mesh_init_sdata(struct ieee80211_sub_if_data *sdata); + /* Mesh paths */ int mesh_nexthop_lookup(u8 *next_hop, struct sk_buff *skb, struct net_device *dev); @@ -257,6 +259,9 @@ void mesh_path_timer(unsigned long data); void mesh_path_flush_by_nexthop(struct sta_info *sta); void mesh_path_discard_frame(struct sk_buff *skb, struct net_device *dev); +#ifdef CONFIG_MAC80211_MESH +extern int mesh_allocated; + static inline int mesh_plink_free_count(struct ieee80211_sub_if_data *sdata) { return sdata->u.sta.mshcfg.dot11MeshMaxPeerLinks - @@ -278,6 +283,10 @@ static inline void mesh_path_activate(struct mesh_path *mpath) for (i = 0; i <= x->hash_mask; i++) \ hlist_for_each_entry_rcu(node, p, &x->hash_buckets[i], list) +#else +#define mesh_allocated 0 +#endif + #define MESH_PREQ(skb) (skb->cb + 30) #endif /* IEEE80211S_H */ diff --git a/net/mac80211/mesh_plink.c b/net/mac80211/mesh_plink.c index 5cd97e99be62..0b0e8d7eb9c7 100644 --- a/net/mac80211/mesh_plink.c +++ b/net/mac80211/mesh_plink.c @@ -6,11 +6,11 @@ * it under the terms of the GNU General Public License version 2 as * published by the Free Software Foundation. */ - +#include +#include #include "ieee80211_i.h" #include "ieee80211_rate.h" #include "mesh.h" -#include #ifdef CONFIG_MAC80211_VERBOSE_MPL_DEBUG #define mpl_dbg(fmt, args...) printk(KERN_DEBUG fmt, ##args) @@ -131,7 +131,7 @@ struct sta_info *mesh_plink_add(u8 *hw_addr, u64 rates, struct net_device *dev) } /** - * mesh_plink_deactivate - deactivate mesh peer link + * __mesh_plink_deactivate - deactivate mesh peer link * * @sta: mesh peer link to deactivate * @@ -139,7 +139,7 @@ struct sta_info *mesh_plink_add(u8 *hw_addr, u64 rates, struct net_device *dev) * * Locking: the caller must hold sta->plink_lock */ -void mesh_plink_deactivate(struct sta_info *sta) +static void __mesh_plink_deactivate(struct sta_info *sta) { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(sta->dev); if (sta->plink_state == ESTAB) @@ -148,6 +148,20 @@ void mesh_plink_deactivate(struct sta_info *sta) mesh_path_flush_by_nexthop(sta); } +/** + * __mesh_plink_deactivate - deactivate mesh peer link + * + * @sta: mesh peer link to deactivate + * + * All mesh paths with this peer as next hop will be flushed + */ +void mesh_plink_deactivate(struct sta_info *sta) +{ + spin_lock_bh(&sta->plink_lock); + __mesh_plink_deactivate(sta); + spin_unlock_bh(&sta->plink_lock); +} + static int mesh_plink_frame_tx(struct net_device *dev, enum plink_frame_type action, u8 *da, __le16 llid, __le16 plid, __le16 reason) { @@ -365,7 +379,7 @@ void mesh_plink_block(struct sta_info *sta) #endif spin_lock_bh(&sta->plink_lock); - mesh_plink_deactivate(sta); + __mesh_plink_deactivate(sta); sta->plink_state = BLOCKED; spin_unlock_bh(&sta->plink_lock); } @@ -390,7 +404,7 @@ int mesh_plink_close(struct sta_info *sta) sta_info_put(sta); return 0; } else if (sta->plink_state == ESTAB) { - mesh_plink_deactivate(sta); + __mesh_plink_deactivate(sta); /* The timer should not be running */ if (!mod_plink_timer(sta, dot11MeshHoldingTimeout(sdata))) __sta_info_get(sta); @@ -699,7 +713,7 @@ void mesh_rx_plink_frame(struct net_device *dev, struct ieee80211_mgmt *mgmt, case CLS_ACPT: reason = cpu_to_le16(MESH_CLOSE_RCVD); sta->reason = reason; - mesh_plink_deactivate(sta); + __mesh_plink_deactivate(sta); sta->plink_state = HOLDING; llid = sta->llid; if (!mod_plink_timer(sta, diff --git a/net/mac80211/rc80211_pid_algo.c b/net/mac80211/rc80211_pid_algo.c index 4a51647a41af..217c0f487bba 100644 --- a/net/mac80211/rc80211_pid_algo.c +++ b/net/mac80211/rc80211_pid_algo.c @@ -15,10 +15,7 @@ #include #include #include "ieee80211_rate.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif - #include "rc80211_pid.h" diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index cc4a896c617f..d0018fc40b09 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -20,9 +20,7 @@ #include "ieee80211_i.h" #include "ieee80211_led.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif #include "wep.h" #include "wpa.h" #include "tkip.h" @@ -439,6 +437,13 @@ ieee80211_rx_mesh_check(struct ieee80211_txrx_data *rx) else return RX_CONTINUE; } +#undef msh_h_get +#else +static inline ieee80211_rx_result +ieee80211_rx_mesh_check(struct ieee80211_txrx_data *rx) +{ + return RX_CONTINUE; +} #endif @@ -477,10 +482,8 @@ ieee80211_rx_h_check(struct ieee80211_txrx_data *rx) * responsible for filtering on both auth and assoc states. */ -#ifdef CONFIG_MAC80211_MESH - if (rx->sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) + if (ieee80211_vif_is_mesh(&rx->sdata->vif)) return ieee80211_rx_mesh_check(rx); -#endif if (unlikely(((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_DATA || ((rx->fc & IEEE80211_FCTL_FTYPE) == IEEE80211_FTYPE_CTL && @@ -1111,8 +1114,7 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx) hdrlen = ieee80211_get_hdrlen(fc); -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) { + if (ieee80211_vif_is_mesh(&sdata->vif)) { int meshhdrlen = ieee80211_get_mesh_hdrlen( (struct ieee80211s_hdr *) (skb->data + hdrlen)); /* Copy on cb: @@ -1126,7 +1128,6 @@ ieee80211_data_to_8023(struct ieee80211_txrx_data *rx) memcpy(MESH_PREQ(skb), hdr->addr2, ETH_ALEN); hdrlen += meshhdrlen; } -#endif /* convert IEEE 802.11 header + possible LLC headers into Ethernet * header @@ -1306,9 +1307,8 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx) } } -#ifdef CONFIG_MAC80211_MESH /* Mesh forwarding */ - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) { + if (ieee80211_vif_is_mesh(&sdata->vif)) { u8 *mesh_ttl = &((struct ieee80211s_hdr *)skb->cb)->ttl; (*mesh_ttl)--; @@ -1321,12 +1321,13 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx) else xmit_skb->pkt_type = PACKET_OTHERHOST; } else - sdata->u.sta.mshstats.dropped_frames_ttl++; - + IEEE80211_IFSTA_MESH_CTR_INC(&sdata->u.sta, + dropped_frames_ttl); } else if (skb->pkt_type != PACKET_OTHERHOST && compare_ether_addr(dev->dev_addr, skb->data) != 0) { if (*mesh_ttl == 0) { - sdata->u.sta.mshstats.dropped_frames_ttl++; + IEEE80211_IFSTA_MESH_CTR_INC(&sdata->u.sta, + dropped_frames_ttl); dev_kfree_skb(skb); skb = NULL; } else { @@ -1337,7 +1338,6 @@ ieee80211_deliver_skb(struct ieee80211_txrx_data *rx) } } } -#endif if (skb) { /* deliver to local stack */ diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 1f3c9eb98500..81c4e3392f40 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -21,9 +21,7 @@ #include "ieee80211_rate.h" #include "sta_info.h" #include "debugfs_sta.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif /* Caller must hold local->sta_lock */ static void sta_info_hash_add(struct ieee80211_local *local, @@ -309,10 +307,8 @@ void sta_info_remove(struct sta_info *sta) } local->num_sta--; -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) + if (ieee80211_vif_is_mesh(&sdata->vif)) mesh_accept_plinks_update(sdata->dev); -#endif } void sta_info_free(struct sta_info *sta) @@ -329,13 +325,8 @@ void sta_info_free(struct sta_info *sta) sta_info_remove(sta); write_unlock_bh(&local->sta_lock); -#ifdef CONFIG_MAC80211_MESH - if (sdata->vif.type == IEEE80211_IF_TYPE_MESH_POINT) { - spin_lock_bh(&sta->plink_lock); + if (ieee80211_vif_is_mesh(&sdata->vif)) mesh_plink_deactivate(sta); - spin_unlock_bh(&sta->plink_lock); - } -#endif while ((skb = skb_dequeue(&sta->ps_tx_buf)) != NULL) { local->total_ps_buffered--; diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index 9d1d7a0e3114..4ad500373d5a 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -107,7 +107,6 @@ struct tid_ampdu_rx { struct timer_list session_timer; }; -#ifdef CONFIG_MAC80211_MESH enum plink_state { LISTEN, OPN_SNT, @@ -117,7 +116,6 @@ enum plink_state { HOLDING, BLOCKED }; -#endif /** * struct sta_ampdu_mlme - STA aggregation information. diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index fc1ffb55ed5c..3b06e0d8f35c 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -26,9 +26,7 @@ #include "ieee80211_i.h" #include "ieee80211_led.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif #include "wep.h" #include "wpa.h" #include "wme.h" @@ -1460,7 +1458,7 @@ int ieee80211_subif_start_xmit(struct sk_buff *skb, goto fail; } meshhdrlen = ieee80211_new_mesh_header(&mesh_hdr, - sdata); + sdata); } hdrlen = 30; break; @@ -1778,40 +1776,6 @@ static void ieee80211_beacon_add_tim(struct ieee80211_local *local, read_unlock_bh(&local->sta_lock); } -#ifdef CONFIG_MAC80211_MESH -static struct sk_buff *ieee80211_mesh_beacon_get(struct net_device *dev) -{ - struct ieee80211_local *local = wdev_priv(dev->ieee80211_ptr); - struct sk_buff *skb = dev_alloc_skb(local->hw.extra_tx_headroom + 400); - struct ieee80211_mgmt *mgmt; - u8 *pos; - - if (!skb) - return NULL; - skb_reserve(skb, local->hw.extra_tx_headroom); - mgmt = (struct ieee80211_mgmt *) - skb_put(skb, 24 + sizeof(mgmt->u.beacon)); - memset(mgmt, 0, 24 + sizeof(mgmt->u.beacon)); - mgmt->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT, - IEEE80211_STYPE_BEACON); - memset(mgmt->da, 0xff, ETH_ALEN); - memcpy(mgmt->sa, dev->dev_addr, ETH_ALEN); - /* BSSID is left zeroed, wildcard value */ - mgmt->u.beacon.beacon_int = - cpu_to_le16(local->hw.conf.beacon_int); - mgmt->u.beacon.capab_info = 0x0; /* 0x0 for MPs */ - - pos = skb_put(skb, 2); - *pos++ = WLAN_EID_SSID; - *pos++ = 0x0; - - mesh_mgmt_ies_add(skb, dev); - - return skb; -} -#endif - - struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, struct ieee80211_vif *vif, struct ieee80211_tx_control *control) @@ -1824,8 +1788,10 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, struct rate_selection rsel; struct beacon_data *beacon; struct ieee80211_supported_band *sband; + struct ieee80211_mgmt *mgmt; int *num_beacons; - int err = 0; + bool err = true; + u8 *pos; sband = local->hw.wiphy->bands[local->hw.conf.channel->band]; @@ -1834,47 +1800,65 @@ struct sk_buff *ieee80211_beacon_get(struct ieee80211_hw *hw, sdata = vif_to_sdata(vif); bdev = sdata->dev; - switch (sdata->vif.type) { - case IEEE80211_IF_TYPE_AP: + if (sdata->vif.type == IEEE80211_IF_TYPE_AP) { ap = &sdata->u.ap; beacon = rcu_dereference(ap->beacon); - if (!ap || !beacon) { - err = -1; - break; - } + if (ap && beacon) { + /* + * headroom, head length, + * tail length and maximum TIM length + */ + skb = dev_alloc_skb(local->tx_headroom + + beacon->head_len + + beacon->tail_len + 256); + if (!skb) + goto out; - /* headroom, head length, tail length and maximum TIM length */ - skb = dev_alloc_skb(local->tx_headroom + beacon->head_len + - beacon->tail_len + 256); - if (!skb) - goto out; + skb_reserve(skb, local->tx_headroom); + memcpy(skb_put(skb, beacon->head_len), beacon->head, + beacon->head_len); - skb_reserve(skb, local->tx_headroom); - memcpy(skb_put(skb, beacon->head_len), beacon->head, - beacon->head_len); + ieee80211_include_sequence(sdata, + (struct ieee80211_hdr *)skb->data); - ieee80211_include_sequence(sdata, - (struct ieee80211_hdr *)skb->data); + ieee80211_beacon_add_tim(local, ap, skb, beacon); - ieee80211_beacon_add_tim(local, ap, skb, beacon); + if (beacon->tail) + memcpy(skb_put(skb, beacon->tail_len), + beacon->tail, beacon->tail_len); - if (beacon->tail) - memcpy(skb_put(skb, beacon->tail_len), beacon->tail, - beacon->tail_len); + num_beacons = &ap->num_beacons; - num_beacons = &ap->num_beacons; - break; + err = false; + } + } else if (ieee80211_vif_is_mesh(&sdata->vif)) { + /* headroom, head length, tail length and maximum TIM length */ + skb = dev_alloc_skb(local->tx_headroom + 400); + if (!skb) + goto out; + + skb_reserve(skb, local->hw.extra_tx_headroom); + mgmt = (struct ieee80211_mgmt *) + skb_put(skb, 24 + sizeof(mgmt->u.beacon)); + memset(mgmt, 0, 24 + sizeof(mgmt->u.beacon)); + mgmt->frame_control = IEEE80211_FC(IEEE80211_FTYPE_MGMT, + IEEE80211_STYPE_BEACON); + memset(mgmt->da, 0xff, ETH_ALEN); + memcpy(mgmt->sa, sdata->dev->dev_addr, ETH_ALEN); + /* BSSID is left zeroed, wildcard value */ + mgmt->u.beacon.beacon_int = + cpu_to_le16(local->hw.conf.beacon_int); + mgmt->u.beacon.capab_info = 0x0; /* 0x0 for MPs */ + + pos = skb_put(skb, 2); + *pos++ = WLAN_EID_SSID; + *pos++ = 0x0; + + mesh_mgmt_ies_add(skb, sdata->dev); -#ifdef CONFIG_MAC80211_MESH - case IEEE80211_IF_TYPE_MESH_POINT: - skb = ieee80211_mesh_beacon_get(bdev); num_beacons = &sdata->u.sta.num_beacons; - break; -#endif - default: - err = -1; - break; + err = false; } if (err) { diff --git a/net/mac80211/util.c b/net/mac80211/util.c index 6b50b6c12da3..b46496fa2e10 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -26,9 +26,7 @@ #include "ieee80211_i.h" #include "ieee80211_rate.h" -#ifdef CONFIG_MAC80211_MESH #include "mesh.h" -#endif #include "wme.h" /* privid for wiphys to determine whether they belong to us or not */ @@ -149,7 +147,6 @@ int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb) } EXPORT_SYMBOL(ieee80211_get_hdrlen_from_skb); -#ifdef CONFIG_MAC80211_MESH int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr) { int ae = meshhdr->flags & IEEE80211S_FLAGS_AE; @@ -167,7 +164,6 @@ int ieee80211_get_mesh_hdrlen(struct ieee80211s_hdr *meshhdr) return 5; } } -#endif void ieee80211_tx_set_iswep(struct ieee80211_txrx_data *tx) { @@ -418,31 +414,3 @@ void ieee80211_iterate_active_interfaces( rcu_read_unlock(); } EXPORT_SYMBOL_GPL(ieee80211_iterate_active_interfaces); - -#ifdef CONFIG_MAC80211_MESH -/** - * ieee80211_new_mesh_header - create a new mesh header - * @meshhdr: uninitialized mesh header - * @sdata: mesh interface to be used - * - * Return the header length. - */ -int ieee80211_new_mesh_header(struct ieee80211s_hdr *meshhdr, - struct ieee80211_sub_if_data *sdata) -{ - meshhdr->flags = 0; - meshhdr->ttl = sdata->u.sta.mshcfg.dot11MeshTTL; - - meshhdr->seqnum[0] = sdata->u.sta.mesh_seqnum[0]++; - meshhdr->seqnum[1] = sdata->u.sta.mesh_seqnum[1]; - meshhdr->seqnum[2] = sdata->u.sta.mesh_seqnum[2]; - - if (sdata->u.sta.mesh_seqnum[0] == 0) { - sdata->u.sta.mesh_seqnum[1]++; - if (sdata->u.sta.mesh_seqnum[1] == 0) - sdata->u.sta.mesh_seqnum[2]++; - } - - return 5; -} -#endif -- cgit v1.2.3 From dbbea6713d6096cd1c411cb453a6b71292c78b33 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 26 Feb 2008 14:34:06 +0100 Subject: mac80211: add documentation book Quite a while ago I started this book. The required kernel-doc patches have since gone into the tree so it is now possible to build the book in mainline. The actual documentation is still rather incomplete and not all things are linked into the book, but this enables us to edit the documentation collaboratively, hopefully driver authors can add documentation based on their experience with mac80211. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- Documentation/DocBook/Makefile | 3 +- Documentation/DocBook/mac80211.tmpl | 335 ++++++++++++++++++++++++++++++++++++ include/net/mac80211.h | 7 +- net/mac80211/key.c | 4 +- net/mac80211/sta_info.c | 16 +- 5 files changed, 351 insertions(+), 14 deletions(-) create mode 100644 Documentation/DocBook/mac80211.tmpl (limited to 'include') diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 300e1707893f..9ebd1f00c6e7 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -11,7 +11,8 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \ procfs-guide.xml writing_usb_driver.xml networking.xml \ kernel-api.xml filesystems.xml lsm.xml usb.xml \ gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml \ - genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml + genericirq.xml s390-drivers.xml uio-howto.xml scsi.xml \ + mac80211.xml ### # The build process is as follows (targets): diff --git a/Documentation/DocBook/mac80211.tmpl b/Documentation/DocBook/mac80211.tmpl new file mode 100644 index 000000000000..b651e0a4b1c0 --- /dev/null +++ b/Documentation/DocBook/mac80211.tmpl @@ -0,0 +1,335 @@ + + + + + + The mac80211 subsystem for kernel developers + + + + Johannes + Berg + +
johannes@sipsolutions.net
+
+
+
+ + + 2007 + 2008 + Johannes Berg + + + + + This documentation is free software; you can redistribute + it and/or modify it under the terms of the GNU General Public + License version 2 as published by the Free Software Foundation. + + + + This documentation 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 this documentation; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, + MA 02111-1307 USA + + + + For more details see the file COPYING in the source + distribution of Linux. + + + + +!Pinclude/net/mac80211.h Introduction +!Pinclude/net/mac80211.h Warning + +
+ + + + + + + The basic mac80211 driver interface + + + You should read and understand the information contained + within this part of the book while implementing a driver. + In some chapters, advanced usage is noted, that may be + skipped at first. + + + This part of the book only covers station and monitor mode + functionality, additional information required to implement + the other modes is covered in the second part of the book. + + + + + Basic hardware handling + TBD + + This chapter shall contain information on getting a hw + struct allocated and registered with mac80211. + + + Since it is required to allocate rates/modes before registering + a hw struct, this chapter shall also contain information on setting + up the rate/mode structs. + + + Additionally, some discussion about the callbacks and + the general programming model should be in here, including + the definition of ieee80211_ops which will be referred to + a lot. + + + Finally, a discussion of hardware capabilities should be done + with references to other parts of the book. + + +!Finclude/net/mac80211.h ieee80211_hw +!Finclude/net/mac80211.h ieee80211_hw_flags +!Finclude/net/mac80211.h SET_IEEE80211_DEV +!Finclude/net/mac80211.h SET_IEEE80211_PERM_ADDR +!Finclude/net/mac80211.h ieee80211_ops +!Finclude/net/mac80211.h ieee80211_alloc_hw +!Finclude/net/mac80211.h ieee80211_register_hw +!Finclude/net/mac80211.h ieee80211_get_tx_led_name +!Finclude/net/mac80211.h ieee80211_get_rx_led_name +!Finclude/net/mac80211.h ieee80211_get_assoc_led_name +!Finclude/net/mac80211.h ieee80211_get_radio_led_name +!Finclude/net/mac80211.h ieee80211_unregister_hw +!Finclude/net/mac80211.h ieee80211_free_hw + + + + PHY configuration + TBD + + This chapter should describe PHY handling including + start/stop callbacks and the various structures used. + +!Finclude/net/mac80211.h ieee80211_conf +!Finclude/net/mac80211.h ieee80211_conf_flags + + + + Virtual interfaces + TBD + + This chapter should describe virtual interface basics + that are relevant to the driver (VLANs, MGMT etc are not.) + It should explain the use of the add_iface/remove_iface + callbacks as well as the interface configuration callbacks. + + Things related to AP mode should be discussed there. + + Things related to supporting multiple interfaces should be + in the appropriate chapter, a BIG FAT note should be here about + this though and the recommendation to allow only a single + interface in STA mode at first! + +!Finclude/net/mac80211.h ieee80211_if_types +!Finclude/net/mac80211.h ieee80211_if_init_conf +!Finclude/net/mac80211.h ieee80211_if_conf + + + + Receive and transmit processing + + what should be here + TBD + + This should describe the receive and transmit + paths in mac80211/the drivers as well as + transmit status handling. + + + + Frame format +!Pinclude/net/mac80211.h Frame format + + + Alignment issues + TBD + + + Calling into mac80211 from interrupts +!Pinclude/net/mac80211.h Calling mac80211 from interrupts + + + functions/definitions +!Finclude/net/mac80211.h ieee80211_rx_status +!Finclude/net/mac80211.h mac80211_rx_flags +!Finclude/net/mac80211.h ieee80211_tx_control +!Finclude/net/mac80211.h ieee80211_tx_status_flags +!Finclude/net/mac80211.h ieee80211_rx +!Finclude/net/mac80211.h ieee80211_rx_irqsafe +!Finclude/net/mac80211.h ieee80211_tx_status +!Finclude/net/mac80211.h ieee80211_tx_status_irqsafe +!Finclude/net/mac80211.h ieee80211_rts_get +!Finclude/net/mac80211.h ieee80211_rts_duration +!Finclude/net/mac80211.h ieee80211_ctstoself_get +!Finclude/net/mac80211.h ieee80211_ctstoself_duration +!Finclude/net/mac80211.h ieee80211_generic_frame_duration +!Finclude/net/mac80211.h ieee80211_get_hdrlen_from_skb +!Finclude/net/mac80211.h ieee80211_get_hdrlen +!Finclude/net/mac80211.h ieee80211_wake_queue +!Finclude/net/mac80211.h ieee80211_stop_queue +!Finclude/net/mac80211.h ieee80211_start_queues +!Finclude/net/mac80211.h ieee80211_stop_queues +!Finclude/net/mac80211.h ieee80211_wake_queues + + + + + Frame filtering +!Pinclude/net/mac80211.h Frame filtering +!Finclude/net/mac80211.h ieee80211_filter_flags + + + + + Advanced driver interface + + + Information contained within this part of the book is + of interest only for advanced interaction of mac80211 + with drivers to exploit more hardware capabilities and + improve performance. + + + + + Hardware crypto acceleration +!Pinclude/net/mac80211.h Hardware crypto acceleration + +!Finclude/net/mac80211.h set_key_cmd +!Finclude/net/mac80211.h ieee80211_key_conf +!Finclude/net/mac80211.h ieee80211_key_alg +!Finclude/net/mac80211.h ieee80211_key_flags + + + + Multiple queues and QoS support + TBD +!Finclude/net/mac80211.h ieee80211_tx_queue_params +!Finclude/net/mac80211.h ieee80211_tx_queue_stats_data +!Finclude/net/mac80211.h ieee80211_tx_queue + + + + Access point mode support + TBD + Some parts of the if_conf should be discussed here instead + + Insert notes about VLAN interfaces with hw crypto here or + in the hw crypto chapter. + +!Finclude/net/mac80211.h ieee80211_get_buffered_bc +!Finclude/net/mac80211.h ieee80211_beacon_get + + + + Supporting multiple virtual interfaces + TBD + + Note: WDS with identical MAC address should almost always be OK + + + Insert notes about having multiple virtual interfaces with + different MAC addresses here, note which configurations are + supported by mac80211, add notes about supporting hw crypto + with it. + + + + + Hardware scan offload + TBD +!Finclude/net/mac80211.h ieee80211_scan_completed + + + + + Rate control interface + + TBD + + This part of the book describes the rate control algorithm + interface and how it relates to mac80211 and drivers. + + + + dummy chapter + TBD + + + + + Internals + + TBD + + This part of the book describes mac80211 internals. + + + + + Key handling + + Key handling basics +!Pnet/mac80211/key.c Key handling basics + + + MORE TBD + TBD + + + + + Receive processing + TBD + + + + Transmit processing + TBD + + + + Station info handling + + Programming information +!Fnet/mac80211/sta_info.h sta_info +!Fnet/mac80211/sta_info.h ieee80211_sta_info_flags + + + STA information lifetime rules +!Pnet/mac80211/sta_info.c STA information lifetime rules + + + + + Synchronisation + TBD + Locking, lots of RCU + + +
diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 6aca472d7a02..d002b1c6e78e 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -407,7 +407,6 @@ enum ieee80211_conf_flags { * @channel: the channel to tune to */ struct ieee80211_conf { - unsigned int regulatory_domain; int radio_enabled; int beacon_int; @@ -437,6 +436,7 @@ struct ieee80211_conf { * @IEEE80211_IF_TYPE_WDS: interface in WDS mode. * @IEEE80211_IF_TYPE_VLAN: VLAN interface bound to an AP, drivers * will never see this type. + * @IEEE80211_IF_TYPE_MESH_POINT: 802.11s mesh point */ enum ieee80211_if_types { IEEE80211_IF_TYPE_INVALID, @@ -1096,8 +1096,9 @@ struct ieee80211_hw *ieee80211_alloc_hw(size_t priv_data_len, /** * ieee80211_register_hw - Register hardware device * - * You must call this function before any other functions - * except ieee80211_register_hwmode. + * You must call this function before any other functions in + * mac80211. Note that before a hardware can be registered, you + * need to fill the contained wiphy's information. * * @hw: the device to register as returned by ieee80211_alloc_hw() */ diff --git a/net/mac80211/key.c b/net/mac80211/key.c index df0c04cedbe4..166d0f00d135 100644 --- a/net/mac80211/key.c +++ b/net/mac80211/key.c @@ -20,8 +20,8 @@ #include "aes_ccm.h" -/* - * Key handling basics +/** + * DOC: Key handling basics * * Key handling in mac80211 is done based on per-interface (sub_if_data) * keys and per-station keys. Since each station belongs to an interface, diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index a767042ec4fd..42414b441592 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -31,10 +31,10 @@ * for faster lookup and a list for iteration. They are managed using * RCU, i.e. access to the list and hash table is protected by RCU. * - * Upon allocating a STA info structure with @sta_info_alloc() or + * Upon allocating a STA info structure with sta_info_alloc() or * mesh_plink_alloc(), the caller owns that structure. It must then either - * destroy it using @sta_info_destroy() (which is pretty useless) or insert - * it into the hash table using @sta_info_insert() which demotes the reference + * destroy it using sta_info_destroy() (which is pretty useless) or insert + * it into the hash table using sta_info_insert() which demotes the reference * from ownership to a regular RCU-protected reference; if the function * is called without protection by an RCU critical section the reference * is instantly invalidated. @@ -42,19 +42,19 @@ * Because there are debugfs entries for each station, and adding those * must be able to sleep, it is also possible to "pin" a station entry, * that means it can be removed from the hash table but not be freed. - * See the comment in @__sta_info_unlink() for more information. + * See the comment in __sta_info_unlink() for more information. * * In order to remove a STA info structure, the caller needs to first - * unlink it (@sta_info_unlink()) from the list and hash tables and + * unlink it (sta_info_unlink()) from the list and hash tables and * then wait for an RCU synchronisation before it can be freed. Due to * the pinning and the possibility of multiple callers trying to remove - * the same STA info at the same time, @sta_info_unlink() can clear the + * the same STA info at the same time, sta_info_unlink() can clear the * STA info pointer it is passed to indicate that the STA info is owned * by somebody else now. * - * If @sta_info_unlink() did not clear the pointer then the caller owns + * If sta_info_unlink() did not clear the pointer then the caller owns * the STA info structure now and is responsible of destroying it with - * a call to @sta_info_destroy(), not before RCU synchronisation, of + * a call to sta_info_destroy(), not before RCU synchronisation, of * course. Note that sta_info_destroy() must be protected by the RTNL. * * In all other cases, there is no concept of ownership on a STA entry, -- cgit v1.2.3 From aab547ce0d1493d400b6468c521a0137cd8c1edf Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Fri, 29 Feb 2008 11:36:12 +0100 Subject: ssb: Add Gigabit Ethernet driver This adds the Gigabit Ethernet driver for the SSB Gigabit Ethernet core. This driver actually is a frontend to the Tigon3 driver. So the real work is done by tg3. This device is used in the Linksys WRT350N. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/ssb/Kconfig | 9 ++ drivers/ssb/Makefile | 1 + drivers/ssb/driver_gige.c | 294 ++++++++++++++++++++++++++++++++++++ drivers/ssb/driver_mipscore.c | 1 + drivers/ssb/driver_pcicore.c | 160 +++++++++++--------- drivers/ssb/embedded.c | 90 +++++++++++ drivers/ssb/main.c | 30 +++- drivers/ssb/ssb_private.h | 2 + include/linux/ssb/ssb.h | 7 + include/linux/ssb/ssb_driver_gige.h | 174 +++++++++++++++++++++ include/linux/ssb/ssb_driver_pci.h | 19 +++ 11 files changed, 715 insertions(+), 72 deletions(-) create mode 100644 drivers/ssb/driver_gige.c create mode 100644 include/linux/ssb/ssb_driver_gige.h (limited to 'include') diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig index adea792fb675..f69ef0ba2613 100644 --- a/drivers/ssb/Kconfig +++ b/drivers/ssb/Kconfig @@ -125,4 +125,13 @@ config SSB_DRIVER_EXTIF If unsure, say N +config SSB_DRIVER_GIGE + bool "SSB Broadcom Gigabit Ethernet driver" + depends on SSB_PCIHOST_POSSIBLE && SSB_EMBEDDED && MIPS + help + Driver for the Sonics Silicon Backplane attached + Broadcom Gigabit Ethernet. + + If unsure, say N + endmenu diff --git a/drivers/ssb/Makefile b/drivers/ssb/Makefile index de94c2eb7a37..910f35e32fc9 100644 --- a/drivers/ssb/Makefile +++ b/drivers/ssb/Makefile @@ -11,6 +11,7 @@ ssb-y += driver_chipcommon.o ssb-$(CONFIG_SSB_DRIVER_MIPS) += driver_mipscore.o ssb-$(CONFIG_SSB_DRIVER_EXTIF) += driver_extif.o ssb-$(CONFIG_SSB_DRIVER_PCICORE) += driver_pcicore.o +ssb-$(CONFIG_SSB_DRIVER_GIGE) += driver_gige.o # b43 pci-ssb-bridge driver # Not strictly a part of SSB, but kept here for convenience diff --git a/drivers/ssb/driver_gige.c b/drivers/ssb/driver_gige.c new file mode 100644 index 000000000000..172f90407b93 --- /dev/null +++ b/drivers/ssb/driver_gige.c @@ -0,0 +1,294 @@ +/* + * Sonics Silicon Backplane + * Broadcom Gigabit Ethernet core driver + * + * Copyright 2008, Broadcom Corporation + * Copyright 2008, Michael Buesch + * + * Licensed under the GNU/GPL. See COPYING for details. + */ + +#include +#include +#include +#include + + +/* +MODULE_DESCRIPTION("SSB Broadcom Gigabit Ethernet driver"); +MODULE_AUTHOR("Michael Buesch"); +MODULE_LICENSE("GPL"); +*/ + +static const struct ssb_device_id ssb_gige_tbl[] = { + SSB_DEVICE(SSB_VENDOR_BROADCOM, SSB_DEV_ETHERNET_GBIT, SSB_ANY_REV), + SSB_DEVTABLE_END +}; +/* MODULE_DEVICE_TABLE(ssb, ssb_gige_tbl); */ + + +static inline u8 gige_read8(struct ssb_gige *dev, u16 offset) +{ + return ssb_read8(dev->dev, offset); +} + +static inline u16 gige_read16(struct ssb_gige *dev, u16 offset) +{ + return ssb_read16(dev->dev, offset); +} + +static inline u32 gige_read32(struct ssb_gige *dev, u16 offset) +{ + return ssb_read32(dev->dev, offset); +} + +static inline void gige_write8(struct ssb_gige *dev, + u16 offset, u8 value) +{ + ssb_write8(dev->dev, offset, value); +} + +static inline void gige_write16(struct ssb_gige *dev, + u16 offset, u16 value) +{ + ssb_write16(dev->dev, offset, value); +} + +static inline void gige_write32(struct ssb_gige *dev, + u16 offset, u32 value) +{ + ssb_write32(dev->dev, offset, value); +} + +static inline +u8 gige_pcicfg_read8(struct ssb_gige *dev, unsigned int offset) +{ + BUG_ON(offset >= 256); + return gige_read8(dev, SSB_GIGE_PCICFG + offset); +} + +static inline +u16 gige_pcicfg_read16(struct ssb_gige *dev, unsigned int offset) +{ + BUG_ON(offset >= 256); + return gige_read16(dev, SSB_GIGE_PCICFG + offset); +} + +static inline +u32 gige_pcicfg_read32(struct ssb_gige *dev, unsigned int offset) +{ + BUG_ON(offset >= 256); + return gige_read32(dev, SSB_GIGE_PCICFG + offset); +} + +static inline +void gige_pcicfg_write8(struct ssb_gige *dev, + unsigned int offset, u8 value) +{ + BUG_ON(offset >= 256); + gige_write8(dev, SSB_GIGE_PCICFG + offset, value); +} + +static inline +void gige_pcicfg_write16(struct ssb_gige *dev, + unsigned int offset, u16 value) +{ + BUG_ON(offset >= 256); + gige_write16(dev, SSB_GIGE_PCICFG + offset, value); +} + +static inline +void gige_pcicfg_write32(struct ssb_gige *dev, + unsigned int offset, u32 value) +{ + BUG_ON(offset >= 256); + gige_write32(dev, SSB_GIGE_PCICFG + offset, value); +} + +static int ssb_gige_pci_read_config(struct pci_bus *bus, unsigned int devfn, + int reg, int size, u32 *val) +{ + struct ssb_gige *dev = container_of(bus->ops, struct ssb_gige, pci_ops); + unsigned long flags; + + if ((PCI_SLOT(devfn) > 0) || (PCI_FUNC(devfn) > 0)) + return PCIBIOS_DEVICE_NOT_FOUND; + if (reg >= 256) + return PCIBIOS_DEVICE_NOT_FOUND; + + spin_lock_irqsave(&dev->lock, flags); + switch (size) { + case 1: + *val = gige_pcicfg_read8(dev, reg); + break; + case 2: + *val = gige_pcicfg_read16(dev, reg); + break; + case 4: + *val = gige_pcicfg_read32(dev, reg); + break; + default: + WARN_ON(1); + } + spin_unlock_irqrestore(&dev->lock, flags); + + return PCIBIOS_SUCCESSFUL; +} + +static int ssb_gige_pci_write_config(struct pci_bus *bus, unsigned int devfn, + int reg, int size, u32 val) +{ + struct ssb_gige *dev = container_of(bus->ops, struct ssb_gige, pci_ops); + unsigned long flags; + + if ((PCI_SLOT(devfn) > 0) || (PCI_FUNC(devfn) > 0)) + return PCIBIOS_DEVICE_NOT_FOUND; + if (reg >= 256) + return PCIBIOS_DEVICE_NOT_FOUND; + + spin_lock_irqsave(&dev->lock, flags); + switch (size) { + case 1: + gige_pcicfg_write8(dev, reg, val); + break; + case 2: + gige_pcicfg_write16(dev, reg, val); + break; + case 4: + gige_pcicfg_write32(dev, reg, val); + break; + default: + WARN_ON(1); + } + spin_unlock_irqrestore(&dev->lock, flags); + + return PCIBIOS_SUCCESSFUL; +} + +static int ssb_gige_probe(struct ssb_device *sdev, const struct ssb_device_id *id) +{ + struct ssb_gige *dev; + u32 base, tmslow, tmshigh; + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) + return -ENOMEM; + dev->dev = sdev; + + spin_lock_init(&dev->lock); + dev->pci_controller.pci_ops = &dev->pci_ops; + dev->pci_controller.io_resource = &dev->io_resource; + dev->pci_controller.mem_resource = &dev->mem_resource; + dev->pci_controller.io_map_base = 0x800; + dev->pci_ops.read = ssb_gige_pci_read_config; + dev->pci_ops.write = ssb_gige_pci_write_config; + + dev->io_resource.name = SSB_GIGE_IO_RES_NAME; + dev->io_resource.start = 0x800; + dev->io_resource.end = 0x8FF; + dev->io_resource.flags = IORESOURCE_IO | IORESOURCE_PCI_FIXED; + + if (!ssb_device_is_enabled(sdev)) + ssb_device_enable(sdev, 0); + + /* Setup BAR0. This is a 64k MMIO region. */ + base = ssb_admatch_base(ssb_read32(sdev, SSB_ADMATCH1)); + gige_pcicfg_write32(dev, PCI_BASE_ADDRESS_0, base); + gige_pcicfg_write32(dev, PCI_BASE_ADDRESS_1, 0); + + dev->mem_resource.name = SSB_GIGE_MEM_RES_NAME; + dev->mem_resource.start = base; + dev->mem_resource.end = base + 0x10000 - 1; + dev->mem_resource.flags = IORESOURCE_MEM | IORESOURCE_PCI_FIXED; + + /* Enable the memory region. */ + gige_pcicfg_write16(dev, PCI_COMMAND, + gige_pcicfg_read16(dev, PCI_COMMAND) + | PCI_COMMAND_MEMORY); + + /* Write flushing is controlled by the Flush Status Control register. + * We want to flush every register write with a timeout and we want + * to disable the IRQ mask while flushing to avoid concurrency. + * Note that automatic write flushing does _not_ work from + * an IRQ handler. The driver must flush manually by reading a register. + */ + gige_write32(dev, SSB_GIGE_SHIM_FLUSHSTAT, 0x00000068); + + /* Check if we have an RGMII or GMII PHY-bus. + * On RGMII do not bypass the DLLs */ + tmslow = ssb_read32(sdev, SSB_TMSLOW); + tmshigh = ssb_read32(sdev, SSB_TMSHIGH); + if (tmshigh & SSB_GIGE_TMSHIGH_RGMII) { + tmslow &= ~SSB_GIGE_TMSLOW_TXBYPASS; + tmslow &= ~SSB_GIGE_TMSLOW_RXBYPASS; + dev->has_rgmii = 1; + } else { + tmslow |= SSB_GIGE_TMSLOW_TXBYPASS; + tmslow |= SSB_GIGE_TMSLOW_RXBYPASS; + dev->has_rgmii = 0; + } + tmslow |= SSB_GIGE_TMSLOW_DLLEN; + ssb_write32(sdev, SSB_TMSLOW, tmslow); + + ssb_set_drvdata(sdev, dev); + register_pci_controller(&dev->pci_controller); + + return 0; +} + +bool pdev_is_ssb_gige_core(struct pci_dev *pdev) +{ + if (!pdev->resource[0].name) + return 0; + return (strcmp(pdev->resource[0].name, SSB_GIGE_MEM_RES_NAME) == 0); +} +EXPORT_SYMBOL(pdev_is_ssb_gige_core); + +int ssb_gige_pcibios_plat_dev_init(struct ssb_device *sdev, + struct pci_dev *pdev) +{ + struct ssb_gige *dev = ssb_get_drvdata(sdev); + struct resource *res; + + if (pdev->bus->ops != &dev->pci_ops) { + /* The PCI device is not on this SSB GigE bridge device. */ + return -ENODEV; + } + + /* Fixup the PCI resources. */ + res = &(pdev->resource[0]); + res->flags = IORESOURCE_MEM | IORESOURCE_PCI_FIXED; + res->name = dev->mem_resource.name; + res->start = dev->mem_resource.start; + res->end = dev->mem_resource.end; + + /* Fixup interrupt lines. */ + pdev->irq = ssb_mips_irq(sdev) + 2; + pci_write_config_byte(pdev, PCI_INTERRUPT_LINE, pdev->irq); + + return 0; +} + +int ssb_gige_map_irq(struct ssb_device *sdev, + const struct pci_dev *pdev) +{ + struct ssb_gige *dev = ssb_get_drvdata(sdev); + + if (pdev->bus->ops != &dev->pci_ops) { + /* The PCI device is not on this SSB GigE bridge device. */ + return -ENODEV; + } + + return ssb_mips_irq(sdev) + 2; +} + +static struct ssb_driver ssb_gige_driver = { + .name = "BCM-GigE", + .id_table = ssb_gige_tbl, + .probe = ssb_gige_probe, +}; + +int ssb_gige_init(void) +{ + return ssb_driver_register(&ssb_gige_driver); +} diff --git a/drivers/ssb/driver_mipscore.c b/drivers/ssb/driver_mipscore.c index 3d3dd32bf3ab..e3fad3123ecb 100644 --- a/drivers/ssb/driver_mipscore.c +++ b/drivers/ssb/driver_mipscore.c @@ -209,6 +209,7 @@ void ssb_mipscore_init(struct ssb_mipscore *mcore) /* fallthrough */ case SSB_DEV_PCI: case SSB_DEV_ETHERNET: + case SSB_DEV_ETHERNET_GBIT: case SSB_DEV_80211: case SSB_DEV_USB20_HOST: /* These devices get their own IRQ line if available, the rest goes on IRQ0 */ diff --git a/drivers/ssb/driver_pcicore.c b/drivers/ssb/driver_pcicore.c index 74b9a8aea52b..33a7d5620474 100644 --- a/drivers/ssb/driver_pcicore.c +++ b/drivers/ssb/driver_pcicore.c @@ -60,77 +60,6 @@ static DEFINE_SPINLOCK(cfgspace_lock); /* Core to access the external PCI config space. Can only have one. */ static struct ssb_pcicore *extpci_core; -static u32 ssb_pcicore_pcibus_iobase = 0x100; -static u32 ssb_pcicore_pcibus_membase = SSB_PCI_DMA; - -int pcibios_plat_dev_init(struct pci_dev *d) -{ - struct resource *res; - int pos, size; - u32 *base; - - ssb_printk(KERN_INFO "PCI: Fixing up device %s\n", - pci_name(d)); - - /* Fix up resource bases */ - for (pos = 0; pos < 6; pos++) { - res = &d->resource[pos]; - if (res->flags & IORESOURCE_IO) - base = &ssb_pcicore_pcibus_iobase; - else - base = &ssb_pcicore_pcibus_membase; - res->flags |= IORESOURCE_PCI_FIXED; - if (res->end) { - size = res->end - res->start + 1; - if (*base & (size - 1)) - *base = (*base + size) & ~(size - 1); - res->start = *base; - res->end = res->start + size - 1; - *base += size; - pci_write_config_dword(d, PCI_BASE_ADDRESS_0 + (pos << 2), res->start); - } - /* Fix up PCI bridge BAR0 only */ - if (d->bus->number == 0 && PCI_SLOT(d->devfn) == 0) - break; - } - /* Fix up interrupt lines */ - d->irq = ssb_mips_irq(extpci_core->dev) + 2; - pci_write_config_byte(d, PCI_INTERRUPT_LINE, d->irq); - - return 0; -} - -static void __init ssb_fixup_pcibridge(struct pci_dev *dev) -{ - u8 lat; - - if (dev->bus->number != 0 || PCI_SLOT(dev->devfn) != 0) - return; - - ssb_printk(KERN_INFO "PCI: Fixing up bridge %s\n", pci_name(dev)); - - /* Enable PCI bridge bus mastering and memory space */ - pci_set_master(dev); - if (pcibios_enable_device(dev, ~0) < 0) { - ssb_printk(KERN_ERR "PCI: SSB bridge enable failed\n"); - return; - } - - /* Enable PCI bridge BAR1 prefetch and burst */ - pci_write_config_dword(dev, SSB_BAR1_CONTROL, 3); - - /* Make sure our latency is high enough to handle the devices behind us */ - lat = 168; - ssb_printk(KERN_INFO "PCI: Fixing latency timer of device %s to %u\n", - pci_name(dev), lat); - pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat); -} -DECLARE_PCI_FIXUP_EARLY(PCI_ANY_ID, PCI_ANY_ID, ssb_fixup_pcibridge); - -int __init pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) -{ - return ssb_mips_irq(extpci_core->dev) + 2; -} static u32 get_cfgspace_addr(struct ssb_pcicore *pc, unsigned int bus, unsigned int dev, @@ -320,6 +249,95 @@ static struct pci_controller ssb_pcicore_controller = { .mem_offset = 0x24000000, }; +static u32 ssb_pcicore_pcibus_iobase = 0x100; +static u32 ssb_pcicore_pcibus_membase = SSB_PCI_DMA; + +/* This function is called when doing a pci_enable_device(). + * We must first check if the device is a device on the PCI-core bridge. */ +int ssb_pcicore_plat_dev_init(struct pci_dev *d) +{ + struct resource *res; + int pos, size; + u32 *base; + + if (d->bus->ops != &ssb_pcicore_pciops) { + /* This is not a device on the PCI-core bridge. */ + return -ENODEV; + } + + ssb_printk(KERN_INFO "PCI: Fixing up device %s\n", + pci_name(d)); + + /* Fix up resource bases */ + for (pos = 0; pos < 6; pos++) { + res = &d->resource[pos]; + if (res->flags & IORESOURCE_IO) + base = &ssb_pcicore_pcibus_iobase; + else + base = &ssb_pcicore_pcibus_membase; + res->flags |= IORESOURCE_PCI_FIXED; + if (res->end) { + size = res->end - res->start + 1; + if (*base & (size - 1)) + *base = (*base + size) & ~(size - 1); + res->start = *base; + res->end = res->start + size - 1; + *base += size; + pci_write_config_dword(d, PCI_BASE_ADDRESS_0 + (pos << 2), res->start); + } + /* Fix up PCI bridge BAR0 only */ + if (d->bus->number == 0 && PCI_SLOT(d->devfn) == 0) + break; + } + /* Fix up interrupt lines */ + d->irq = ssb_mips_irq(extpci_core->dev) + 2; + pci_write_config_byte(d, PCI_INTERRUPT_LINE, d->irq); + + return 0; +} + +/* Early PCI fixup for a device on the PCI-core bridge. */ +static void ssb_pcicore_fixup_pcibridge(struct pci_dev *dev) +{ + u8 lat; + + if (dev->bus->ops != &ssb_pcicore_pciops) { + /* This is not a device on the PCI-core bridge. */ + return; + } + if (dev->bus->number != 0 || PCI_SLOT(dev->devfn) != 0) + return; + + ssb_printk(KERN_INFO "PCI: Fixing up bridge %s\n", pci_name(dev)); + + /* Enable PCI bridge bus mastering and memory space */ + pci_set_master(dev); + if (pcibios_enable_device(dev, ~0) < 0) { + ssb_printk(KERN_ERR "PCI: SSB bridge enable failed\n"); + return; + } + + /* Enable PCI bridge BAR1 prefetch and burst */ + pci_write_config_dword(dev, SSB_BAR1_CONTROL, 3); + + /* Make sure our latency is high enough to handle the devices behind us */ + lat = 168; + ssb_printk(KERN_INFO "PCI: Fixing latency timer of device %s to %u\n", + pci_name(dev), lat); + pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat); +} +DECLARE_PCI_FIXUP_EARLY(PCI_ANY_ID, PCI_ANY_ID, ssb_pcicore_fixup_pcibridge); + +/* PCI device IRQ mapping. */ +int ssb_pcicore_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +{ + if (dev->bus->ops != &ssb_pcicore_pciops) { + /* This is not a device on the PCI-core bridge. */ + return -ENODEV; + } + return ssb_mips_irq(extpci_core->dev) + 2; +} + static void ssb_pcicore_init_hostmode(struct ssb_pcicore *pc) { u32 val; diff --git a/drivers/ssb/embedded.c b/drivers/ssb/embedded.c index d3ade821555c..7dc3a6b41397 100644 --- a/drivers/ssb/embedded.c +++ b/drivers/ssb/embedded.c @@ -10,6 +10,9 @@ #include #include +#include +#include +#include #include "ssb_private.h" @@ -130,3 +133,90 @@ u32 ssb_gpio_polarity(struct ssb_bus *bus, u32 mask, u32 value) return res; } EXPORT_SYMBOL(ssb_gpio_polarity); + +#ifdef CONFIG_SSB_DRIVER_GIGE +static int gige_pci_init_callback(struct ssb_bus *bus, unsigned long data) +{ + struct pci_dev *pdev = (struct pci_dev *)data; + struct ssb_device *dev; + unsigned int i; + int res; + + for (i = 0; i < bus->nr_devices; i++) { + dev = &(bus->devices[i]); + if (dev->id.coreid != SSB_DEV_ETHERNET_GBIT) + continue; + if (!dev->dev || + !dev->dev->driver || + !device_is_registered(dev->dev)) + continue; + res = ssb_gige_pcibios_plat_dev_init(dev, pdev); + if (res >= 0) + return res; + } + + return -ENODEV; +} +#endif /* CONFIG_SSB_DRIVER_GIGE */ + +int ssb_pcibios_plat_dev_init(struct pci_dev *dev) +{ + int err; + + err = ssb_pcicore_plat_dev_init(dev); + if (!err) + return 0; +#ifdef CONFIG_SSB_DRIVER_GIGE + err = ssb_for_each_bus_call((unsigned long)dev, gige_pci_init_callback); + if (err >= 0) + return err; +#endif + /* This is not a PCI device on any SSB device. */ + + return -ENODEV; +} + +#ifdef CONFIG_SSB_DRIVER_GIGE +static int gige_map_irq_callback(struct ssb_bus *bus, unsigned long data) +{ + const struct pci_dev *pdev = (const struct pci_dev *)data; + struct ssb_device *dev; + unsigned int i; + int res; + + for (i = 0; i < bus->nr_devices; i++) { + dev = &(bus->devices[i]); + if (dev->id.coreid != SSB_DEV_ETHERNET_GBIT) + continue; + if (!dev->dev || + !dev->dev->driver || + !device_is_registered(dev->dev)) + continue; + res = ssb_gige_map_irq(dev, pdev); + if (res >= 0) + return res; + } + + return -ENODEV; +} +#endif /* CONFIG_SSB_DRIVER_GIGE */ + +int ssb_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +{ + int res; + + /* Check if this PCI device is a device on a SSB bus or device + * and return the IRQ number for it. */ + + res = ssb_pcicore_pcibios_map_irq(dev, slot, pin); + if (res >= 0) + return res; +#ifdef CONFIG_SSB_DRIVER_GIGE + res = ssb_for_each_bus_call((unsigned long)dev, gige_map_irq_callback); + if (res >= 0) + return res; +#endif + /* This is not a PCI device on any SSB device. */ + + return -ENODEV; +} diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c index 8db40c4b86e9..49d7bbb9bea7 100644 --- a/drivers/ssb/main.c +++ b/drivers/ssb/main.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -68,6 +69,25 @@ found: } #endif /* CONFIG_SSB_PCIHOST */ +int ssb_for_each_bus_call(unsigned long data, + int (*func)(struct ssb_bus *bus, unsigned long data)) +{ + struct ssb_bus *bus; + int res; + + ssb_buses_lock(); + list_for_each_entry(bus, &buses, list) { + res = func(bus, data); + if (res >= 0) { + ssb_buses_unlock(); + return res; + } + } + ssb_buses_unlock(); + + return -ENODEV; +} + static struct ssb_device *ssb_device_get(struct ssb_device *dev) { if (dev) @@ -1171,7 +1191,14 @@ static int __init ssb_modinit(void) err = b43_pci_ssb_bridge_init(); if (err) { ssb_printk(KERN_ERR "Broadcom 43xx PCI-SSB-bridge " - "initialization failed"); + "initialization failed\n"); + /* don't fail SSB init because of this */ + err = 0; + } + err = ssb_gige_init(); + if (err) { + ssb_printk(KERN_ERR "SSB Broadcom Gigabit Ethernet " + "driver initialization failed\n"); /* don't fail SSB init because of this */ err = 0; } @@ -1185,6 +1212,7 @@ fs_initcall(ssb_modinit); static void __exit ssb_modexit(void) { + ssb_gige_exit(); b43_pci_ssb_bridge_exit(); bus_unregister(&ssb_bustype); } diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h index 21eca2b5118b..d03b20983b1e 100644 --- a/drivers/ssb/ssb_private.h +++ b/drivers/ssb/ssb_private.h @@ -118,6 +118,8 @@ extern u32 ssb_calc_clock_rate(u32 plltype, u32 n, u32 m); extern int ssb_devices_freeze(struct ssb_bus *bus); extern int ssb_devices_thaw(struct ssb_bus *bus); extern struct ssb_bus *ssb_pci_dev_to_bus(struct pci_dev *pdev); +int ssb_for_each_bus_call(unsigned long data, + int (*func)(struct ssb_bus *bus, unsigned long data)); /* b43_pci_bridge.c */ #ifdef CONFIG_SSB_B43_PCI_BRIDGE diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h index 860d28c6d149..b7c388972fcf 100644 --- a/include/linux/ssb/ssb.h +++ b/include/linux/ssb/ssb.h @@ -422,5 +422,12 @@ extern int ssb_bus_powerup(struct ssb_bus *bus, bool dynamic_pctl); extern u32 ssb_admatch_base(u32 adm); extern u32 ssb_admatch_size(u32 adm); +/* PCI device mapping and fixup routines. + * Called from the architecture pcibios init code. + * These are only available on SSB_EMBEDDED configurations. */ +#ifdef CONFIG_SSB_EMBEDDED +int ssb_pcibios_plat_dev_init(struct pci_dev *dev); +int ssb_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin); +#endif /* CONFIG_SSB_EMBEDDED */ #endif /* LINUX_SSB_H_ */ diff --git a/include/linux/ssb/ssb_driver_gige.h b/include/linux/ssb/ssb_driver_gige.h new file mode 100644 index 000000000000..01fbdf5fef22 --- /dev/null +++ b/include/linux/ssb/ssb_driver_gige.h @@ -0,0 +1,174 @@ +#ifndef LINUX_SSB_DRIVER_GIGE_H_ +#define LINUX_SSB_DRIVER_GIGE_H_ + +#include +#include +#include + + +#ifdef CONFIG_SSB_DRIVER_GIGE + + +#define SSB_GIGE_PCIIO 0x0000 /* PCI I/O Registers (1024 bytes) */ +#define SSB_GIGE_RESERVED 0x0400 /* Reserved (1024 bytes) */ +#define SSB_GIGE_PCICFG 0x0800 /* PCI config space (256 bytes) */ +#define SSB_GIGE_SHIM_FLUSHSTAT 0x0C00 /* PCI to OCP: Flush status control (32bit) */ +#define SSB_GIGE_SHIM_FLUSHRDA 0x0C04 /* PCI to OCP: Flush read address (32bit) */ +#define SSB_GIGE_SHIM_FLUSHTO 0x0C08 /* PCI to OCP: Flush timeout counter (32bit) */ +#define SSB_GIGE_SHIM_BARRIER 0x0C0C /* PCI to OCP: Barrier register (32bit) */ +#define SSB_GIGE_SHIM_MAOCPSI 0x0C10 /* PCI to OCP: MaocpSI Control (32bit) */ +#define SSB_GIGE_SHIM_SIOCPMA 0x0C14 /* PCI to OCP: SiocpMa Control (32bit) */ + +/* TM Status High flags */ +#define SSB_GIGE_TMSHIGH_RGMII 0x00010000 /* Have an RGMII PHY-bus */ +/* TM Status Low flags */ +#define SSB_GIGE_TMSLOW_TXBYPASS 0x00080000 /* TX bypass (no delay) */ +#define SSB_GIGE_TMSLOW_RXBYPASS 0x00100000 /* RX bypass (no delay) */ +#define SSB_GIGE_TMSLOW_DLLEN 0x01000000 /* Enable DLL controls */ + +/* Boardflags (low) */ +#define SSB_GIGE_BFL_ROBOSWITCH 0x0010 + + +#define SSB_GIGE_MEM_RES_NAME "SSB Broadcom 47xx GigE memory" +#define SSB_GIGE_IO_RES_NAME "SSB Broadcom 47xx GigE I/O" + +struct ssb_gige { + struct ssb_device *dev; + + spinlock_t lock; + + /* True, if the device has an RGMII bus. + * False, if the device has a GMII bus. */ + bool has_rgmii; + + /* The PCI controller device. */ + struct pci_controller pci_controller; + struct pci_ops pci_ops; + struct resource mem_resource; + struct resource io_resource; +}; + +/* Check whether a PCI device is a SSB Gigabit Ethernet core. */ +extern bool pdev_is_ssb_gige_core(struct pci_dev *pdev); + +/* Convert a pci_dev pointer to a ssb_gige pointer. */ +static inline struct ssb_gige * pdev_to_ssb_gige(struct pci_dev *pdev) +{ + if (!pdev_is_ssb_gige_core(pdev)) + return NULL; + return container_of(pdev->bus->ops, struct ssb_gige, pci_ops); +} + +/* Returns whether the PHY is connected by an RGMII bus. */ +static inline bool ssb_gige_is_rgmii(struct pci_dev *pdev) +{ + struct ssb_gige *dev = pdev_to_ssb_gige(pdev); + return (dev ? dev->has_rgmii : 0); +} + +/* Returns whether we have a Roboswitch. */ +static inline bool ssb_gige_have_roboswitch(struct pci_dev *pdev) +{ + struct ssb_gige *dev = pdev_to_ssb_gige(pdev); + if (dev) + return !!(dev->dev->bus->sprom.boardflags_lo & + SSB_GIGE_BFL_ROBOSWITCH); + return 0; +} + +/* Returns whether we can only do one DMA at once. */ +static inline bool ssb_gige_one_dma_at_once(struct pci_dev *pdev) +{ + struct ssb_gige *dev = pdev_to_ssb_gige(pdev); + if (dev) + return ((dev->dev->bus->chip_id == 0x4785) && + (dev->dev->bus->chip_rev < 2)); + return 0; +} + +/* Returns whether we must flush posted writes. */ +static inline bool ssb_gige_must_flush_posted_writes(struct pci_dev *pdev) +{ + struct ssb_gige *dev = pdev_to_ssb_gige(pdev); + if (dev) + return (dev->dev->bus->chip_id == 0x4785); + return 0; +} + +extern char * nvram_get(const char *name); +/* Get the device MAC address */ +static inline void ssb_gige_get_macaddr(struct pci_dev *pdev, u8 *macaddr) +{ +#ifdef CONFIG_BCM947XX + char *res = nvram_get("et0macaddr"); + if (res) + memcpy(macaddr, res, 6); +#endif +} + +extern int ssb_gige_pcibios_plat_dev_init(struct ssb_device *sdev, + struct pci_dev *pdev); +extern int ssb_gige_map_irq(struct ssb_device *sdev, + const struct pci_dev *pdev); + +/* The GigE driver is not a standalone module, because we don't have support + * for unregistering the driver. So we could not unload the module anyway. */ +extern int ssb_gige_init(void); +static inline void ssb_gige_exit(void) +{ + /* Currently we can not unregister the GigE driver, + * because we can not unregister the PCI bridge. */ + BUG(); +} + + +#else /* CONFIG_SSB_DRIVER_GIGE */ +/* Gigabit Ethernet driver disabled */ + + +static inline int ssb_gige_pcibios_plat_dev_init(struct ssb_device *sdev, + struct pci_dev *pdev) +{ + return -ENOSYS; +} +static inline int ssb_gige_map_irq(struct ssb_device *sdev, + const struct pci_dev *pdev) +{ + return -ENOSYS; +} +static inline int ssb_gige_init(void) +{ + return 0; +} +static inline void ssb_gige_exit(void) +{ +} + +static inline bool pdev_is_ssb_gige_core(struct pci_dev *pdev) +{ + return 0; +} +static inline struct ssb_gige * pdev_to_ssb_gige(struct pci_dev *pdev) +{ + return NULL; +} +static inline bool ssb_gige_is_rgmii(struct pci_dev *pdev) +{ + return 0; +} +static inline bool ssb_gige_have_roboswitch(struct pci_dev *pdev) +{ + return 0; +} +static inline bool ssb_gige_one_dma_at_once(struct pci_dev *pdev) +{ + return 0; +} +static inline bool ssb_gige_must_flush_posted_writes(struct pci_dev *pdev) +{ + return 0; +} + +#endif /* CONFIG_SSB_DRIVER_GIGE */ +#endif /* LINUX_SSB_DRIVER_GIGE_H_ */ diff --git a/include/linux/ssb/ssb_driver_pci.h b/include/linux/ssb/ssb_driver_pci.h index 5e25bac4ed31..41e330e51c2a 100644 --- a/include/linux/ssb/ssb_driver_pci.h +++ b/include/linux/ssb/ssb_driver_pci.h @@ -1,6 +1,11 @@ #ifndef LINUX_SSB_PCICORE_H_ #define LINUX_SSB_PCICORE_H_ +#include + +struct pci_dev; + + #ifdef CONFIG_SSB_DRIVER_PCICORE /* PCI core registers. */ @@ -88,6 +93,9 @@ extern void ssb_pcicore_init(struct ssb_pcicore *pc); extern int ssb_pcicore_dev_irqvecs_enable(struct ssb_pcicore *pc, struct ssb_device *dev); +int ssb_pcicore_plat_dev_init(struct pci_dev *d); +int ssb_pcicore_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin); + #else /* CONFIG_SSB_DRIVER_PCICORE */ @@ -107,5 +115,16 @@ int ssb_pcicore_dev_irqvecs_enable(struct ssb_pcicore *pc, return 0; } +static inline +int ssb_pcicore_plat_dev_init(struct pci_dev *d) +{ + return -ENODEV; +} +static inline +int ssb_pcicore_pcibios_map_irq(const struct pci_dev *dev, u8 slot, u8 pin) +{ + return -ENODEV; +} + #endif /* CONFIG_SSB_DRIVER_PCICORE */ #endif /* LINUX_SSB_PCICORE_H_ */ -- cgit v1.2.3 From 8c8696553aa3895c2ad4289537e4af45a8877b62 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Thu, 6 Mar 2008 15:05:07 -0800 Subject: [TIPC]: Removal of message header option code This patch removes code associated with optional, user-specified fields of the TIPC message header. Such fields were never utilized by TIPC, and have now been removed from the protocol specification. Signed-off-by: Allan Stephens Signed-off-by: David S. Miller --- include/net/tipc/tipc_port.h | 7 ------- net/tipc/core.c | 1 - net/tipc/msg.h | 23 ----------------------- net/tipc/port.c | 7 ------- 4 files changed, 38 deletions(-) (limited to 'include') diff --git a/include/net/tipc/tipc_port.h b/include/net/tipc/tipc_port.h index cfc4ba46de8f..c9b36b77a0b9 100644 --- a/include/net/tipc/tipc_port.h +++ b/include/net/tipc/tipc_port.h @@ -86,13 +86,6 @@ u32 tipc_createport_raw(void *usr_handle, void (*wakeup)(struct tipc_port *), const u32 importance); -/* - * tipc_set_msg_option(): port must be locked. - */ -int tipc_set_msg_option(struct tipc_port *tp_ptr, - const char *opt, - const u32 len); - int tipc_reject_msg(struct sk_buff *buf, u32 err); int tipc_send_buf_fast(struct sk_buff *buf, u32 destnode); diff --git a/net/tipc/core.c b/net/tipc/core.c index d2d7d32c02c7..9a5eb97684a2 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -277,7 +277,6 @@ EXPORT_SYMBOL(tipc_register_media); /* TIPC API for external APIs (see tipc_port.h) */ EXPORT_SYMBOL(tipc_createport_raw); -EXPORT_SYMBOL(tipc_set_msg_option); EXPORT_SYMBOL(tipc_reject_msg); EXPORT_SYMBOL(tipc_send_buf_fast); EXPORT_SYMBOL(tipc_acknowledge); diff --git a/net/tipc/msg.h b/net/tipc/msg.h index e9ef6df26562..12f8945e8a54 100644 --- a/net/tipc/msg.h +++ b/net/tipc/msg.h @@ -190,18 +190,6 @@ static inline void msg_set_lookup_scope(struct tipc_msg *m, u32 n) msg_set_bits(m, 1, 19, 0x3, n); } -static inline void msg_set_options(struct tipc_msg *m, const char *opt, u32 sz) -{ - u32 hsz = msg_hdr_sz(m); - char *to = (char *)&m->hdr[hsz/4]; - - if ((hsz < DIR_MSG_H_SIZE) || ((hsz + sz) > MAX_H_SIZE)) - return; - msg_set_bits(m, 1, 16, 0x7, (hsz - 28)/4); - msg_set_hdr_sz(m, hsz + sz); - memcpy(to, opt, sz); -} - static inline u32 msg_bcast_ack(struct tipc_msg *m) { return msg_bits(m, 1, 0, 0xffff); @@ -330,17 +318,6 @@ static inline struct tipc_msg *msg_get_wrapped(struct tipc_msg *m) return (struct tipc_msg *)msg_data(m); } -static inline void msg_expand(struct tipc_msg *m, u32 destnode) -{ - if (!msg_short(m)) - return; - msg_set_hdr_sz(m, LONG_H_SIZE); - msg_set_orignode(m, msg_prevnode(m)); - msg_set_destnode(m, destnode); - memset(&m->hdr[8], 0, 12); -} - - /* TIPC internal message header format, version 2 diff --git a/net/tipc/port.c b/net/tipc/port.c index f508614ca59b..2a16ca43e953 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -413,13 +413,6 @@ static struct sk_buff *port_build_proto_msg(u32 destport, u32 destnode, return buf; } -int tipc_set_msg_option(struct tipc_port *tp_ptr, const char *opt, const u32 sz) -{ - msg_expand(&tp_ptr->phdr, msg_destnode(&tp_ptr->phdr)); - msg_set_options(&tp_ptr->phdr, opt, sz); - return TIPC_OK; -} - int tipc_reject_msg(struct sk_buff *buf, u32 err) { struct tipc_msg *msg = buf_msg(buf); -- cgit v1.2.3 From 0e0609bbd2ab39a5964a70b409a5567ebbaf3700 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Thu, 6 Mar 2008 15:06:06 -0800 Subject: [TIPC]: Eliminate "sparse" symbol warnings This patch eliminates warnings about undeclared symbols. Signed-off-by: Allan Stephens Signed-off-by: David S. Miller --- include/net/tipc/tipc_bearer.h | 9 +++++++++ net/tipc/core.c | 8 -------- net/tipc/core.h | 6 ++++++ 3 files changed, 15 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/tipc/tipc_bearer.h b/include/net/tipc/tipc_bearer.h index 2151a80cdf30..ee2f304e4919 100644 --- a/include/net/tipc/tipc_bearer.h +++ b/include/net/tipc/tipc_bearer.h @@ -99,6 +99,9 @@ struct tipc_bearer { char name[TIPC_MAX_BEARER_NAME]; }; +/* + * TIPC routines available to supported media types + */ int tipc_register_media(u32 media_type, char *media_name, @@ -123,6 +126,12 @@ void tipc_continue(struct tipc_bearer *tb_ptr); int tipc_enable_bearer(const char *bearer_name, u32 bcast_scope, u32 priority); int tipc_disable_bearer(const char *name); +/* + * Routines made available to TIPC by supported media types + */ + +int tipc_eth_media_start(void); +void tipc_eth_media_stop(void); #endif diff --git a/net/tipc/core.c b/net/tipc/core.c index 9a5eb97684a2..b41b0ac67c0b 100644 --- a/net/tipc/core.c +++ b/net/tipc/core.c @@ -48,14 +48,6 @@ #include "subscr.h" #include "config.h" -int tipc_eth_media_start(void); -void tipc_eth_media_stop(void); -int tipc_handler_start(void); -void tipc_handler_stop(void); -int tipc_socket_init(void); -void tipc_socket_stop(void); -int tipc_netlink_start(void); -void tipc_netlink_stop(void); #define TIPC_MOD_VER "1.6.2" diff --git a/net/tipc/core.h b/net/tipc/core.h index feabca580820..3fe9b70331d9 100644 --- a/net/tipc/core.h +++ b/net/tipc/core.h @@ -180,6 +180,12 @@ extern int tipc_core_start(void); extern void tipc_core_stop(void); extern int tipc_core_start_net(void); extern void tipc_core_stop_net(void); +extern int tipc_handler_start(void); +extern void tipc_handler_stop(void); +extern int tipc_netlink_start(void); +extern void tipc_netlink_stop(void); +extern int tipc_socket_init(void); +extern void tipc_socket_stop(void); static inline int delimit(int val, int min, int max) { -- cgit v1.2.3 From db8dac20d5199307dcfcf4e01dac4bda5edf9e89 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 6 Mar 2008 16:22:02 -0800 Subject: [UDP]: Revert udplite and code split. This reverts commit db1ed684f6c430c4cdad67d058688b8a1b5e607c ("[IPV6] UDP: Rename IPv6 UDP files."), commit 8be8af8fa4405652e6c0797db5465a4be8afb998 ("[IPV4] UDP: Move IPv4-specific bits to other file.") and commit e898d4db2749c6052072e9bc4448e396cbdeb06a ("[UDP]: Allow users to configure UDP-Lite."). First, udplite is of such small cost, and it is a core protocol just like TCP and normal UDP are. We spent enormous amounts of effort to make udplite share as much code with core UDP as possible. All of that work is less valuable if we're just going to slap a config option on udplite support. It is also causing build failures, as reported on linux-next, showing that the changeset was not tested very well. In fact, this is the second build failure resulting from the udplite change. Finally, the config options provided was a bool, instead of a modular option. Meaning the udplite code does not even get build tested by allmodconfig builds, and furthermore the user is not presented with a reasonable modular build option which is particularly needed by distribution vendors. Signed-off-by: David S. Miller --- include/linux/udp.h | 10 - include/net/ipv6.h | 5 - include/net/transp_v6.h | 5 - include/net/udplite.h | 9 +- net/ipv4/Kconfig | 10 - net/ipv4/Makefile | 3 +- net/ipv4/af_inet.c | 7 +- net/ipv4/proc.c | 5 +- net/ipv4/udp.c | 1090 +++++++++++++++++++++++++++++++++++++++++++- net/ipv4/udp_ipv4.c | 1134 ---------------------------------------------- net/ipv4/udplite.c | 121 +++++ net/ipv4/udplite_ipv4.c | 121 ----- net/ipv6/Makefile | 3 +- net/ipv6/af_inet6.c | 14 - net/ipv6/ipv6_sockglue.c | 6 +- net/ipv6/proc.c | 6 - net/ipv6/udp.c | 1065 +++++++++++++++++++++++++++++++++++++++++++ net/ipv6/udp_ipv6.c | 1065 ------------------------------------------- net/ipv6/udplite.c | 125 +++++ net/ipv6/udplite_ipv6.c | 125 ----- 20 files changed, 2404 insertions(+), 2525 deletions(-) delete mode 100644 net/ipv4/udp_ipv4.c create mode 100644 net/ipv4/udplite.c delete mode 100644 net/ipv4/udplite_ipv4.c create mode 100644 net/ipv6/udp.c delete mode 100644 net/ipv6/udp_ipv6.c create mode 100644 net/ipv6/udplite.c delete mode 100644 net/ipv6/udplite_ipv6.c (limited to 'include') diff --git a/include/linux/udp.h b/include/linux/udp.h index 4144664d69d9..1e7b7cb5703b 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -70,10 +70,8 @@ struct udp_sock { #define UDPLITE_BIT 0x1 /* set by udplite proto init function */ #define UDPLITE_SEND_CC 0x2 /* set via udplite setsockopt */ #define UDPLITE_RECV_CC 0x4 /* set via udplite setsocktopt */ -#ifdef CONFIG_IP_UDPLITE __u8 pcflag; /* marks socket as UDP-Lite if > 0 */ __u8 unused[3]; -#endif /* * For encapsulation sockets. */ @@ -85,15 +83,7 @@ static inline struct udp_sock *udp_sk(const struct sock *sk) return (struct udp_sock *)sk; } -#ifdef CONFIG_IP_UDPLITE #define IS_UDPLITE(__sk) (udp_sk(__sk)->pcflag) -#define IS_PROTO_UDPLITE(__proto) ((__proto) == IPPROTO_UDPLITE) -#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP || (level) == SOL_UDPLITE) -#else -#define IS_UDPLITE(__sk) 0 -#define IS_PROTO_UDPLITE(__proto) 0 -#define IS_SOL_UDPFAMILY(level) ((level) == SOL_UDP) -#endif #endif diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 5f6df50a33a9..8db06af1efbb 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -599,13 +599,8 @@ extern int tcp6_proc_init(void); extern void tcp6_proc_exit(void); extern int udp6_proc_init(void); extern void udp6_proc_exit(void); -#ifdef CONFIG_IP_UDPLITE extern int udplite6_proc_init(void); extern void udplite6_proc_exit(void); -#else -static inline int udplite6_proc_init(void) { return 0; } -static inline void udplite6_proc_exit(void) { } -#endif extern int ipv6_misc_proc_init(void); extern void ipv6_misc_proc_exit(void); extern int snmp6_register_dev(struct inet6_dev *idev); diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h index 902e6c6bc793..27394e0447d8 100644 --- a/include/net/transp_v6.h +++ b/include/net/transp_v6.h @@ -27,13 +27,8 @@ extern int rawv6_init(void); extern void rawv6_exit(void); extern int udpv6_init(void); extern void udpv6_exit(void); -#ifdef CONFIG_IP_UDPLITE extern int udplitev6_init(void); extern void udplitev6_exit(void); -#else -static inline int udplitev6_init(void) { return 0; } -static inline void udplitev6_exit(void) { } -#endif extern int tcpv6_init(void); extern void tcpv6_exit(void); diff --git a/include/net/udplite.h b/include/net/udplite.h index 01ddb2c20264..b76b2e377af4 100644 --- a/include/net/udplite.h +++ b/include/net/udplite.h @@ -25,9 +25,7 @@ static __inline__ int udplite_getfrag(void *from, char *to, int offset, /* Designate sk as UDP-Lite socket */ static inline int udplite_sk_init(struct sock *sk) { -#ifdef CONFIG_IP_UDPLITE udp_sk(sk)->pcflag = UDPLITE_BIT; -#endif return 0; } @@ -71,7 +69,7 @@ static inline int udplite_checksum_init(struct sk_buff *skb, struct udphdr *uh) static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) { int cscov = up->len; -#ifdef CONFIG_IP_UDPLITE + /* * Sender has set `partial coverage' option on UDP-Lite socket */ @@ -95,15 +93,13 @@ static inline int udplite_sender_cscov(struct udp_sock *up, struct udphdr *uh) * illegal, we fall back to the defaults here. */ } -#endif return cscov; } static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) { - __wsum csum = 0; -#ifdef CONFIG_IP_UDPLITE int cscov = udplite_sender_cscov(udp_sk(sk), udp_hdr(skb)); + __wsum csum = 0; skb->ip_summed = CHECKSUM_NONE; /* no HW support for checksumming */ @@ -116,7 +112,6 @@ static inline __wsum udplite_csum_outgoing(struct sock *sk, struct sk_buff *skb) if ((cscov -= len) <= 0) break; } -#endif return csum; } diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 5098fd2ff4d0..9c7e5ffb223d 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -632,15 +632,5 @@ config TCP_MD5SIG If unsure, say N. -config IP_UDPLITE - bool "IP: UDP-Lite Protocol (RFC 3828)" - default n - ---help--- - UDP-Lite (RFC 3828) is a UDP-like protocol with variable-length - checksum. Read for - details. - - If unsure, say N. - source "net/ipv4/ipvs/Kconfig" diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index d5226241d5ed..ad40ef3f9ebc 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -8,7 +8,7 @@ obj-y := route.o inetpeer.o protocol.o \ inet_timewait_sock.o inet_connection_sock.o \ tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \ tcp_minisocks.o tcp_cong.o \ - datagram.o raw.o udp.o udp_ipv4.o \ + datagram.o raw.o udp.o udplite.o \ arp.o icmp.o devinet.o af_inet.o igmp.o \ fib_frontend.o fib_semantics.o \ inet_fragment.o @@ -49,7 +49,6 @@ obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o -obj-$(CONFIG_IP_UDPLITE) += udplite_ipv4.o obj-$(CONFIG_NETLABEL) += cipso_ipv4.o obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 25871c6c7444..4cb8a1385539 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1317,18 +1317,15 @@ static int __init init_ipv4_mibs(void) if (snmp_mib_init((void **)udp_statistics, sizeof(struct udp_mib)) < 0) goto err_udp_mib; -#ifdef CONFIG_IP_UDPLITE if (snmp_mib_init((void **)udplite_statistics, sizeof(struct udp_mib)) < 0) goto err_udplite_mib; -#endif + tcp_mib_init(); return 0; -#ifdef CONFIG_IP_UDPLITE err_udplite_mib: -#endif snmp_mib_free((void **)udp_statistics); err_udp_mib: snmp_mib_free((void **)tcp_statistics); @@ -1426,10 +1423,8 @@ static int __init inet_init(void) /* Setup UDP memory threshold */ udp_init(); -#ifdef CONFIG_IP_UDPLITE /* Add UDP-Lite (RFC 3828) */ udplite4_register(); -#endif /* * Set the ICMP layer up diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index d75ddb7fa4b8..d63474c6b400 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -59,9 +59,7 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) atomic_read(&tcp_memory_allocated)); seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot), atomic_read(&udp_memory_allocated)); -#ifdef CONFIG_IP_UDPLITE seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot)); -#endif seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot)); seq_printf(seq, "FRAG: inuse %d memory %d\n", ip_frag_nqueues(&init_net), ip_frag_mem(&init_net)); @@ -351,7 +349,6 @@ static int snmp_seq_show(struct seq_file *seq, void *v) snmp_fold_field((void **)udp_statistics, snmp4_udp_list[i].entry)); -#ifdef CONFIG_IP_UDPLITE /* the UDP and UDP-Lite MIBs are the same */ seq_puts(seq, "\nUdpLite:"); for (i = 0; snmp4_udp_list[i].name != NULL; i++) @@ -362,7 +359,7 @@ static int snmp_seq_show(struct seq_file *seq, void *v) seq_printf(seq, " %lu", snmp_fold_field((void **)udplite_statistics, snmp4_udp_list[i].entry)); -#endif + seq_putc(seq, '\n'); return 0; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index c53d7673b57d..7ea1b67b6de1 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -246,6 +246,553 @@ int udp_get_port(struct sock *sk, unsigned short snum, return __udp_lib_get_port(sk, snum, udp_hash, scmp); } +int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) +{ + struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2); + + return ( !ipv6_only_sock(sk2) && + (!inet1->rcv_saddr || !inet2->rcv_saddr || + inet1->rcv_saddr == inet2->rcv_saddr )); +} + +static inline int udp_v4_get_port(struct sock *sk, unsigned short snum) +{ + return udp_get_port(sk, snum, ipv4_rcv_saddr_equal); +} + +/* UDP is nearly always wildcards out the wazoo, it makes no sense to try + * harder than this. -DaveM + */ +static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, + __be16 sport, __be32 daddr, __be16 dport, + int dif, struct hlist_head udptable[]) +{ + struct sock *sk, *result = NULL; + struct hlist_node *node; + unsigned short hnum = ntohs(dport); + int badness = -1; + + read_lock(&udp_hash_lock); + sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { + struct inet_sock *inet = inet_sk(sk); + + if (sk->sk_net == net && sk->sk_hash == hnum && + !ipv6_only_sock(sk)) { + int score = (sk->sk_family == PF_INET ? 1 : 0); + if (inet->rcv_saddr) { + if (inet->rcv_saddr != daddr) + continue; + score+=2; + } + if (inet->daddr) { + if (inet->daddr != saddr) + continue; + score+=2; + } + if (inet->dport) { + if (inet->dport != sport) + continue; + score+=2; + } + if (sk->sk_bound_dev_if) { + if (sk->sk_bound_dev_if != dif) + continue; + score+=2; + } + if (score == 9) { + result = sk; + break; + } else if (score > badness) { + result = sk; + badness = score; + } + } + } + if (result) + sock_hold(result); + read_unlock(&udp_hash_lock); + return result; +} + +static inline struct sock *udp_v4_mcast_next(struct sock *sk, + __be16 loc_port, __be32 loc_addr, + __be16 rmt_port, __be32 rmt_addr, + int dif) +{ + struct hlist_node *node; + struct sock *s = sk; + unsigned short hnum = ntohs(loc_port); + + sk_for_each_from(s, node) { + struct inet_sock *inet = inet_sk(s); + + if (s->sk_hash != hnum || + (inet->daddr && inet->daddr != rmt_addr) || + (inet->dport != rmt_port && inet->dport) || + (inet->rcv_saddr && inet->rcv_saddr != loc_addr) || + ipv6_only_sock(s) || + (s->sk_bound_dev_if && s->sk_bound_dev_if != dif)) + continue; + if (!ip_mc_sf_allow(s, loc_addr, rmt_addr, dif)) + continue; + goto found; + } + s = NULL; +found: + return s; +} + +/* + * 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. + * Header points to the ip header of the error packet. We move + * on past this. Then (as it used to claim before adjustment) + * header points to the first 8 bytes of the udp header. We need + * to find the appropriate port. + */ + +void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[]) +{ + struct inet_sock *inet; + struct iphdr *iph = (struct iphdr*)skb->data; + struct udphdr *uh = (struct udphdr*)(skb->data+(iph->ihl<<2)); + const int type = icmp_hdr(skb)->type; + const int code = icmp_hdr(skb)->code; + struct sock *sk; + int harderr; + int err; + + sk = __udp4_lib_lookup(skb->dev->nd_net, iph->daddr, uh->dest, + iph->saddr, uh->source, skb->dev->ifindex, udptable); + if (sk == NULL) { + ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); + return; /* No socket for error */ + } + + err = 0; + harderr = 0; + inet = inet_sk(sk); + + switch (type) { + default: + case ICMP_TIME_EXCEEDED: + err = EHOSTUNREACH; + break; + case ICMP_SOURCE_QUENCH: + goto out; + case ICMP_PARAMETERPROB: + err = EPROTO; + harderr = 1; + break; + case ICMP_DEST_UNREACH: + if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */ + if (inet->pmtudisc != IP_PMTUDISC_DONT) { + err = EMSGSIZE; + harderr = 1; + break; + } + goto out; + } + err = EHOSTUNREACH; + if (code <= NR_ICMP_UNREACH) { + harderr = icmp_err_convert[code].fatal; + err = icmp_err_convert[code].errno; + } + break; + } + + /* + * RFC1122: OK. Passes ICMP errors back to application, as per + * 4.1.3.3. + */ + if (!inet->recverr) { + if (!harderr || sk->sk_state != TCP_ESTABLISHED) + goto out; + } else { + ip_icmp_error(sk, skb, err, uh->dest, info, (u8*)(uh+1)); + } + sk->sk_err = err; + sk->sk_error_report(sk); +out: + sock_put(sk); +} + +void udp_err(struct sk_buff *skb, u32 info) +{ + __udp4_lib_err(skb, info, udp_hash); +} + +/* + * Throw away all pending data and cancel the corking. Socket is locked. + */ +static void udp_flush_pending_frames(struct sock *sk) +{ + struct udp_sock *up = udp_sk(sk); + + if (up->pending) { + up->len = 0; + up->pending = 0; + ip_flush_pending_frames(sk); + } +} + +/** + * udp4_hwcsum_outgoing - handle outgoing HW checksumming + * @sk: socket we are sending on + * @skb: sk_buff containing the filled-in UDP header + * (checksum field must be zeroed out) + */ +static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, + __be32 src, __be32 dst, int len ) +{ + unsigned int offset; + struct udphdr *uh = udp_hdr(skb); + __wsum csum = 0; + + if (skb_queue_len(&sk->sk_write_queue) == 1) { + /* + * Only one fragment on the socket. + */ + skb->csum_start = skb_transport_header(skb) - skb->head; + skb->csum_offset = offsetof(struct udphdr, check); + uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); + } else { + /* + * HW-checksum won't work as there are two or more + * fragments on the socket so that all csums of sk_buffs + * should be together + */ + offset = skb_transport_offset(skb); + skb->csum = skb_checksum(skb, offset, skb->len - offset, 0); + + skb->ip_summed = CHECKSUM_NONE; + + skb_queue_walk(&sk->sk_write_queue, skb) { + csum = csum_add(csum, skb->csum); + } + + uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum); + if (uh->check == 0) + uh->check = CSUM_MANGLED_0; + } +} + +/* + * Push out all pending data as one UDP datagram. Socket is locked. + */ +static int udp_push_pending_frames(struct sock *sk) +{ + struct udp_sock *up = udp_sk(sk); + struct inet_sock *inet = inet_sk(sk); + struct flowi *fl = &inet->cork.fl; + struct sk_buff *skb; + struct udphdr *uh; + int err = 0; + int is_udplite = IS_UDPLITE(sk); + __wsum csum = 0; + + /* Grab the skbuff where UDP header space exists. */ + if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) + goto out; + + /* + * Create a UDP header + */ + uh = udp_hdr(skb); + uh->source = fl->fl_ip_sport; + uh->dest = fl->fl_ip_dport; + uh->len = htons(up->len); + uh->check = 0; + + if (is_udplite) /* UDP-Lite */ + csum = udplite_csum_outgoing(sk, skb); + + else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ + + skb->ip_summed = CHECKSUM_NONE; + goto send; + + } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ + + udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len); + goto send; + + } else /* `normal' UDP */ + csum = udp_csum_outgoing(sk, skb); + + /* add protocol-dependent pseudo-header */ + uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len, + sk->sk_protocol, csum ); + if (uh->check == 0) + uh->check = CSUM_MANGLED_0; + +send: + err = ip_push_pending_frames(sk); +out: + up->len = 0; + up->pending = 0; + if (!err) + UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); + return err; +} + +int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len) +{ + struct inet_sock *inet = inet_sk(sk); + struct udp_sock *up = udp_sk(sk); + int ulen = len; + struct ipcm_cookie ipc; + struct rtable *rt = NULL; + int free = 0; + int connected = 0; + __be32 daddr, faddr, saddr; + __be16 dport; + u8 tos; + int err, is_udplite = IS_UDPLITE(sk); + int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; + int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); + + if (len > 0xFFFF) + return -EMSGSIZE; + + /* + * Check the flags. + */ + + if (msg->msg_flags&MSG_OOB) /* Mirror BSD error message compatibility */ + return -EOPNOTSUPP; + + ipc.opt = NULL; + + if (up->pending) { + /* + * There are pending frames. + * The socket lock must be held while it's corked. + */ + lock_sock(sk); + if (likely(up->pending)) { + if (unlikely(up->pending != AF_INET)) { + release_sock(sk); + return -EINVAL; + } + goto do_append_data; + } + release_sock(sk); + } + ulen += sizeof(struct udphdr); + + /* + * Get and verify the address. + */ + if (msg->msg_name) { + struct sockaddr_in * usin = (struct sockaddr_in*)msg->msg_name; + if (msg->msg_namelen < sizeof(*usin)) + return -EINVAL; + if (usin->sin_family != AF_INET) { + if (usin->sin_family != AF_UNSPEC) + return -EAFNOSUPPORT; + } + + daddr = usin->sin_addr.s_addr; + dport = usin->sin_port; + if (dport == 0) + return -EINVAL; + } else { + if (sk->sk_state != TCP_ESTABLISHED) + return -EDESTADDRREQ; + daddr = inet->daddr; + dport = inet->dport; + /* Open fast path for connected socket. + Route will not be used, if at least one option is set. + */ + connected = 1; + } + ipc.addr = inet->saddr; + + ipc.oif = sk->sk_bound_dev_if; + if (msg->msg_controllen) { + err = ip_cmsg_send(msg, &ipc); + if (err) + return err; + if (ipc.opt) + free = 1; + connected = 0; + } + if (!ipc.opt) + ipc.opt = inet->opt; + + saddr = ipc.addr; + ipc.addr = faddr = daddr; + + if (ipc.opt && ipc.opt->srr) { + if (!daddr) + return -EINVAL; + faddr = ipc.opt->faddr; + connected = 0; + } + tos = RT_TOS(inet->tos); + if (sock_flag(sk, SOCK_LOCALROUTE) || + (msg->msg_flags & MSG_DONTROUTE) || + (ipc.opt && ipc.opt->is_strictroute)) { + tos |= RTO_ONLINK; + connected = 0; + } + + if (ipv4_is_multicast(daddr)) { + if (!ipc.oif) + ipc.oif = inet->mc_index; + if (!saddr) + saddr = inet->mc_addr; + connected = 0; + } + + if (connected) + rt = (struct rtable*)sk_dst_check(sk, 0); + + if (rt == NULL) { + struct flowi fl = { .oif = ipc.oif, + .nl_u = { .ip4_u = + { .daddr = faddr, + .saddr = saddr, + .tos = tos } }, + .proto = sk->sk_protocol, + .uli_u = { .ports = + { .sport = inet->sport, + .dport = dport } } }; + security_sk_classify_flow(sk, &fl); + err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1); + if (err) { + if (err == -ENETUNREACH) + IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); + goto out; + } + + err = -EACCES; + if ((rt->rt_flags & RTCF_BROADCAST) && + !sock_flag(sk, SOCK_BROADCAST)) + goto out; + if (connected) + sk_dst_set(sk, dst_clone(&rt->u.dst)); + } + + if (msg->msg_flags&MSG_CONFIRM) + goto do_confirm; +back_from_confirm: + + saddr = rt->rt_src; + if (!ipc.addr) + daddr = ipc.addr = rt->rt_dst; + + lock_sock(sk); + if (unlikely(up->pending)) { + /* The socket is already corked while preparing it. */ + /* ... which is an evident application bug. --ANK */ + release_sock(sk); + + LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n"); + err = -EINVAL; + goto out; + } + /* + * Now cork the socket to pend data. + */ + inet->cork.fl.fl4_dst = daddr; + inet->cork.fl.fl_ip_dport = dport; + inet->cork.fl.fl4_src = saddr; + inet->cork.fl.fl_ip_sport = inet->sport; + up->pending = AF_INET; + +do_append_data: + up->len += ulen; + getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; + err = ip_append_data(sk, getfrag, msg->msg_iov, ulen, + sizeof(struct udphdr), &ipc, rt, + corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); + if (err) + udp_flush_pending_frames(sk); + else if (!corkreq) + err = udp_push_pending_frames(sk); + else if (unlikely(skb_queue_empty(&sk->sk_write_queue))) + up->pending = 0; + release_sock(sk); + +out: + ip_rt_put(rt); + if (free) + kfree(ipc.opt); + if (!err) + return len; + /* + * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space. Reporting + * ENOBUFS might not be good (it's not tunable per se), but otherwise + * we don't have a good statistic (IpOutDiscards but it can be too many + * things). We could add another new stat but at least for now that + * seems like overkill. + */ + if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { + UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); + } + return err; + +do_confirm: + dst_confirm(&rt->u.dst); + if (!(msg->msg_flags&MSG_PROBE) || len) + goto back_from_confirm; + err = 0; + goto out; +} + +int udp_sendpage(struct sock *sk, struct page *page, int offset, + size_t size, int flags) +{ + struct udp_sock *up = udp_sk(sk); + int ret; + + if (!up->pending) { + struct msghdr msg = { .msg_flags = flags|MSG_MORE }; + + /* Call udp_sendmsg to specify destination address which + * sendpage interface can't pass. + * This will succeed only when the socket is connected. + */ + ret = udp_sendmsg(NULL, sk, &msg, 0); + if (ret < 0) + return ret; + } + + lock_sock(sk); + + if (unlikely(!up->pending)) { + release_sock(sk); + + LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 3\n"); + return -EINVAL; + } + + ret = ip_append_page(sk, page, offset, size, flags); + if (ret == -EOPNOTSUPP) { + release_sock(sk); + return sock_no_sendpage(sk->sk_socket, page, offset, + size, flags); + } + if (ret < 0) { + udp_flush_pending_frames(sk); + goto out; + } + + up->len += size; + if (!(up->corkflag || (flags&MSG_MORE))) + ret = udp_push_pending_frames(sk); + if (!ret) + ret = size; +out: + release_sock(sk); + return ret; +} + /* * IOCTL requests applicable to the UDP protocol */ @@ -286,6 +833,107 @@ int udp_ioctl(struct sock *sk, int cmd, unsigned long arg) return 0; } +/* + * This should be easy, if there is something there we + * return it, otherwise we block. + */ + +int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len) +{ + struct inet_sock *inet = inet_sk(sk); + struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name; + struct sk_buff *skb; + unsigned int ulen, copied; + int peeked; + int err; + int is_udplite = IS_UDPLITE(sk); + + /* + * Check any passed addresses + */ + if (addr_len) + *addr_len=sizeof(*sin); + + if (flags & MSG_ERRQUEUE) + return ip_recv_error(sk, msg, len); + +try_again: + skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0), + &peeked, &err); + if (!skb) + goto out; + + ulen = skb->len - sizeof(struct udphdr); + copied = len; + if (copied > ulen) + copied = ulen; + else if (copied < ulen) + msg->msg_flags |= MSG_TRUNC; + + /* + * If checksum is needed at all, try to do it while copying the + * data. If the data is truncated, or if we only want a partial + * coverage checksum (UDP-Lite), do it before the copy. + */ + + if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) { + if (udp_lib_checksum_complete(skb)) + goto csum_copy_err; + } + + if (skb_csum_unnecessary(skb)) + err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), + msg->msg_iov, copied ); + else { + err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); + + if (err == -EINVAL) + goto csum_copy_err; + } + + if (err) + goto out_free; + + if (!peeked) + UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite); + + sock_recv_timestamp(msg, sk, skb); + + /* Copy the address. */ + if (sin) + { + sin->sin_family = AF_INET; + sin->sin_port = udp_hdr(skb)->source; + sin->sin_addr.s_addr = ip_hdr(skb)->saddr; + memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); + } + if (inet->cmsg_flags) + ip_cmsg_recv(msg, skb); + + err = copied; + if (flags & MSG_TRUNC) + err = ulen; + +out_free: + lock_sock(sk); + skb_free_datagram(sk, skb); + release_sock(sk); +out: + return err; + +csum_copy_err: + lock_sock(sk); + if (!skb_kill_datagram(sk, skb, flags)) + UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); + release_sock(sk); + + if (noblock) + return -EAGAIN; + goto try_again; +} + + int udp_disconnect(struct sock *sk, int flags) { struct inet_sock *inet = inet_sk(sk); @@ -308,6 +956,319 @@ int udp_disconnect(struct sock *sk, int flags) return 0; } +/* returns: + * -1: error + * 0: success + * >0: "udp encap" protocol resubmission + * + * Note that in the success and error cases, the skb is assumed to + * have either been requeued or freed. + */ +int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) +{ + struct udp_sock *up = udp_sk(sk); + int rc; + int is_udplite = IS_UDPLITE(sk); + + /* + * Charge it to the socket, dropping if the queue is full. + */ + if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) + goto drop; + nf_reset(skb); + + if (up->encap_type) { + /* + * This is an encapsulation socket so pass the skb to + * the socket's udp_encap_rcv() hook. Otherwise, just + * fall through and pass this up the UDP socket. + * up->encap_rcv() returns the following value: + * =0 if skb was successfully passed to the encap + * handler or was discarded by it. + * >0 if skb should be passed on to UDP. + * <0 if skb should be resubmitted as proto -N + */ + + /* if we're overly short, let UDP handle it */ + if (skb->len > sizeof(struct udphdr) && + up->encap_rcv != NULL) { + int ret; + + ret = (*up->encap_rcv)(sk, skb); + if (ret <= 0) { + UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, + is_udplite); + return -ret; + } + } + + /* FALLTHROUGH -- it's a UDP Packet */ + } + + /* + * UDP-Lite specific tests, ignored on UDP sockets + */ + if ((is_udplite & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + + /* + * MIB statistics other than incrementing the error count are + * disabled for the following two types of errors: these depend + * on the application settings, not on the functioning of the + * protocol stack as such. + * + * RFC 3828 here recommends (sec 3.3): "There should also be a + * way ... to ... at least let the receiving application block + * delivery of packets with coverage values less than a value + * provided by the application." + */ + if (up->pcrlen == 0) { /* full coverage was set */ + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage " + "%d while full coverage %d requested\n", + UDP_SKB_CB(skb)->cscov, skb->len); + goto drop; + } + /* The next case involves violating the min. coverage requested + * by the receiver. This is subtle: if receiver wants x and x is + * greater than the buffersize/MTU then receiver will complain + * that it wants x while sender emits packets of smaller size y. + * Therefore the above ...()->partial_cov statement is essential. + */ + if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { + LIMIT_NETDEBUG(KERN_WARNING + "UDPLITE: coverage %d too small, need min %d\n", + UDP_SKB_CB(skb)->cscov, up->pcrlen); + goto drop; + } + } + + if (sk->sk_filter) { + if (udp_lib_checksum_complete(skb)) + goto drop; + } + + if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { + /* Note that an ENOMEM error is charged twice */ + if (rc == -ENOMEM) + UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite); + goto drop; + } + + return 0; + +drop: + UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); + kfree_skb(skb); + return -1; +} + +/* + * Multicasts and broadcasts go to each listener. + * + * Note: called only from the BH handler context, + * so we don't need to lock the hashes. + */ +static int __udp4_lib_mcast_deliver(struct sk_buff *skb, + struct udphdr *uh, + __be32 saddr, __be32 daddr, + struct hlist_head udptable[]) +{ + struct sock *sk; + int dif; + + read_lock(&udp_hash_lock); + sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); + dif = skb->dev->ifindex; + sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); + if (sk) { + struct sock *sknext = NULL; + + do { + struct sk_buff *skb1 = skb; + + sknext = udp_v4_mcast_next(sk_next(sk), uh->dest, daddr, + uh->source, saddr, dif); + if (sknext) + skb1 = skb_clone(skb, GFP_ATOMIC); + + if (skb1) { + int ret = 0; + + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + ret = udp_queue_rcv_skb(sk, skb1); + else + sk_add_backlog(sk, skb1); + bh_unlock_sock(sk); + + if (ret > 0) + /* we should probably re-process instead + * of dropping packets here. */ + kfree_skb(skb1); + } + sk = sknext; + } while (sknext); + } else + kfree_skb(skb); + read_unlock(&udp_hash_lock); + return 0; +} + +/* Initialize UDP checksum. If exited with zero value (success), + * CHECKSUM_UNNECESSARY means, that no more checks are required. + * Otherwise, csum completion requires chacksumming packet body, + * including udp header and folding it to skb->csum. + */ +static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh, + int proto) +{ + const struct iphdr *iph; + int err; + + UDP_SKB_CB(skb)->partial_cov = 0; + UDP_SKB_CB(skb)->cscov = skb->len; + + if (proto == IPPROTO_UDPLITE) { + err = udplite_checksum_init(skb, uh); + if (err) + return err; + } + + iph = ip_hdr(skb); + if (uh->check == 0) { + skb->ip_summed = CHECKSUM_UNNECESSARY; + } else if (skb->ip_summed == CHECKSUM_COMPLETE) { + if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len, + proto, skb->csum)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + if (!skb_csum_unnecessary(skb)) + skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, + skb->len, proto, 0); + /* Probably, we should checksum udp header (it should be in cache + * in any case) and data in tiny packets (< rx copybreak). + */ + + return 0; +} + +/* + * All we need to do is get the socket, and then do a checksum. + */ + +int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], + int proto) +{ + struct sock *sk; + struct udphdr *uh = udp_hdr(skb); + unsigned short ulen; + struct rtable *rt = (struct rtable*)skb->dst; + __be32 saddr = ip_hdr(skb)->saddr; + __be32 daddr = ip_hdr(skb)->daddr; + + /* + * Validate the packet. + */ + if (!pskb_may_pull(skb, sizeof(struct udphdr))) + goto drop; /* No space for header. */ + + ulen = ntohs(uh->len); + if (ulen > skb->len) + goto short_packet; + + if (proto == IPPROTO_UDP) { + /* UDP validates ulen. */ + if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen)) + goto short_packet; + uh = udp_hdr(skb); + } + + if (udp4_csum_init(skb, uh, proto)) + goto csum_error; + + if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST)) + return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable); + + sk = __udp4_lib_lookup(skb->dev->nd_net, saddr, uh->source, daddr, + uh->dest, inet_iif(skb), udptable); + + if (sk != NULL) { + int ret = 0; + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + ret = udp_queue_rcv_skb(sk, skb); + else + sk_add_backlog(sk, skb); + bh_unlock_sock(sk); + sock_put(sk); + + /* a return value > 0 means to resubmit the input, but + * it wants the return to be -protocol, or 0 + */ + if (ret > 0) + return -ret; + return 0; + } + + if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) + goto drop; + nf_reset(skb); + + /* No socket. Drop packet silently, if checksum is wrong */ + if (udp_lib_checksum_complete(skb)) + goto csum_error; + + UDP_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); + icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); + + /* + * Hmm. We got an UDP packet to a port to which we + * don't wanna listen. Ignore it. + */ + kfree_skb(skb); + return 0; + +short_packet: + LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", + proto == IPPROTO_UDPLITE ? "-Lite" : "", + NIPQUAD(saddr), + ntohs(uh->source), + ulen, + skb->len, + NIPQUAD(daddr), + ntohs(uh->dest)); + goto drop; + +csum_error: + /* + * RFC1122: OK. Discards the bad packet silently (as far as + * the network is concerned, anyway) as per 4.1.3.4 (MUST). + */ + LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", + proto == IPPROTO_UDPLITE ? "-Lite" : "", + NIPQUAD(saddr), + ntohs(uh->source), + NIPQUAD(daddr), + ntohs(uh->dest), + ulen); +drop: + UDP_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE); + kfree_skb(skb); + return 0; +} + +int udp_rcv(struct sk_buff *skb) +{ + return __udp4_lib_rcv(skb, udp_hash, IPPROTO_UDP); +} + +int udp_destroy_sock(struct sock *sk) +{ + lock_sock(sk); + udp_flush_pending_frames(sk); + release_sock(sk); + return 0; +} + /* * Socket option code for UDP */ @@ -318,9 +1279,7 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, struct udp_sock *up = udp_sk(sk); int val; int err = 0; -#ifdef CONFIG_IP_UDPLITE int is_udplite = IS_UDPLITE(sk); -#endif if (optlenpcrlen = val; up->pcflag |= UDPLITE_RECV_CC; break; -#endif default: err = -ENOPROTOOPT; @@ -392,6 +1349,26 @@ int udp_lib_setsockopt(struct sock *sk, int level, int optname, return err; } +int udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_push_pending_frames); + return ip_setsockopt(sk, level, optname, optval, optlen); +} + +#ifdef CONFIG_COMPAT +int compat_udp_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_push_pending_frames); + return compat_ip_setsockopt(sk, level, optname, optval, optlen); +} +#endif + int udp_lib_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) { @@ -436,6 +1413,23 @@ int udp_lib_getsockopt(struct sock *sk, int level, int optname, return 0; } +int udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_getsockopt(sk, level, optname, optval, optlen); + return ip_getsockopt(sk, level, optname, optval, optlen); +} + +#ifdef CONFIG_COMPAT +int compat_udp_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_getsockopt(sk, level, optname, optval, optlen); + return compat_ip_getsockopt(sk, level, optname, optval, optlen); +} +#endif /** * udp_poll - wait for a UDP event. * @file - file struct @@ -480,6 +1474,36 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) } +DEFINE_PROTO_INUSE(udp) + +struct proto udp_prot = { + .name = "UDP", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip4_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .destroy = udp_destroy_sock, + .setsockopt = udp_setsockopt, + .getsockopt = udp_getsockopt, + .sendmsg = udp_sendmsg, + .recvmsg = udp_recvmsg, + .sendpage = udp_sendpage, + .backlog_rcv = udp_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udp_v4_get_port, + .memory_allocated = &udp_memory_allocated, + .sysctl_mem = sysctl_udp_mem, + .sysctl_wmem = &sysctl_udp_wmem_min, + .sysctl_rmem = &sysctl_udp_rmem_min, + .obj_size = sizeof(struct udp_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udp_setsockopt, + .compat_getsockopt = compat_udp_getsockopt, +#endif + REF_PROTO_INUSE(udp) +}; /* ------------------------------------------------------------------------ */ #ifdef CONFIG_PROC_FS @@ -612,6 +1636,62 @@ void udp_proc_unregister(struct udp_seq_afinfo *afinfo) proc_net_remove(&init_net, afinfo->name); memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } + +/* ------------------------------------------------------------------------ */ +static void udp4_format_sock(struct sock *sp, char *tmpbuf, int bucket) +{ + struct inet_sock *inet = inet_sk(sp); + __be32 dest = inet->daddr; + __be32 src = inet->rcv_saddr; + __u16 destp = ntohs(inet->dport); + __u16 srcp = ntohs(inet->sport); + + sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X" + " %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p", + bucket, src, srcp, dest, destp, sp->sk_state, + atomic_read(&sp->sk_wmem_alloc), + atomic_read(&sp->sk_rmem_alloc), + 0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp), + atomic_read(&sp->sk_refcnt), sp); +} + +int udp4_seq_show(struct seq_file *seq, void *v) +{ + if (v == SEQ_START_TOKEN) + seq_printf(seq, "%-127s\n", + " sl local_address rem_address st tx_queue " + "rx_queue tr tm->when retrnsmt uid timeout " + "inode"); + else { + char tmpbuf[129]; + struct udp_iter_state *state = seq->private; + + udp4_format_sock(v, tmpbuf, state->bucket); + seq_printf(seq, "%-127s\n", tmpbuf); + } + return 0; +} + +/* ------------------------------------------------------------------------ */ +static struct file_operations udp4_seq_fops; +static struct udp_seq_afinfo udp4_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udp", + .family = AF_INET, + .hashtable = udp_hash, + .seq_show = udp4_seq_show, + .seq_fops = &udp4_seq_fops, +}; + +int __init udp4_proc_init(void) +{ + return udp_proc_register(&udp4_seq_afinfo); +} + +void udp4_proc_exit(void) +{ + udp_proc_unregister(&udp4_seq_afinfo); +} #endif /* CONFIG_PROC_FS */ void __init udp_init(void) @@ -638,6 +1718,8 @@ EXPORT_SYMBOL(udp_hash); EXPORT_SYMBOL(udp_hash_lock); EXPORT_SYMBOL(udp_ioctl); EXPORT_SYMBOL(udp_get_port); +EXPORT_SYMBOL(udp_prot); +EXPORT_SYMBOL(udp_sendmsg); EXPORT_SYMBOL(udp_lib_getsockopt); EXPORT_SYMBOL(udp_lib_setsockopt); EXPORT_SYMBOL(udp_poll); diff --git a/net/ipv4/udp_ipv4.c b/net/ipv4/udp_ipv4.c deleted file mode 100644 index fd14c2c50ed4..000000000000 --- a/net/ipv4/udp_ipv4.c +++ /dev/null @@ -1,1134 +0,0 @@ -/* - * INET An implementation of the TCP/IP protocol suite for the LINUX - * operating system. INET is implemented using the BSD Socket - * interface as the means of communication with the user level. - * - * UDP for IPv4. - * - * For full credits, see net/ipv4/udp.c. - * - * This program 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 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include "udp_impl.h" - -int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) -{ - struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2); - - return ( !ipv6_only_sock(sk2) && - (!inet1->rcv_saddr || !inet2->rcv_saddr || - inet1->rcv_saddr == inet2->rcv_saddr )); -} - -static inline int udp_v4_get_port(struct sock *sk, unsigned short snum) -{ - return udp_get_port(sk, snum, ipv4_rcv_saddr_equal); -} - -/* UDP is nearly always wildcards out the wazoo, it makes no sense to try - * harder than this. -DaveM - */ -static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, - __be16 sport, __be32 daddr, __be16 dport, - int dif, struct hlist_head udptable[]) -{ - struct sock *sk, *result = NULL; - struct hlist_node *node; - unsigned short hnum = ntohs(dport); - int badness = -1; - - read_lock(&udp_hash_lock); - sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { - struct inet_sock *inet = inet_sk(sk); - - if (sk->sk_net == net && sk->sk_hash == hnum && - !ipv6_only_sock(sk)) { - int score = (sk->sk_family == PF_INET ? 1 : 0); - if (inet->rcv_saddr) { - if (inet->rcv_saddr != daddr) - continue; - score+=2; - } - if (inet->daddr) { - if (inet->daddr != saddr) - continue; - score+=2; - } - if (inet->dport) { - if (inet->dport != sport) - continue; - score+=2; - } - if (sk->sk_bound_dev_if) { - if (sk->sk_bound_dev_if != dif) - continue; - score+=2; - } - if (score == 9) { - result = sk; - break; - } else if (score > badness) { - result = sk; - badness = score; - } - } - } - if (result) - sock_hold(result); - read_unlock(&udp_hash_lock); - return result; -} - -static inline struct sock *udp_v4_mcast_next(struct sock *sk, - __be16 loc_port, __be32 loc_addr, - __be16 rmt_port, __be32 rmt_addr, - int dif) -{ - struct hlist_node *node; - struct sock *s = sk; - unsigned short hnum = ntohs(loc_port); - - sk_for_each_from(s, node) { - struct inet_sock *inet = inet_sk(s); - - if (s->sk_hash != hnum || - (inet->daddr && inet->daddr != rmt_addr) || - (inet->dport != rmt_port && inet->dport) || - (inet->rcv_saddr && inet->rcv_saddr != loc_addr) || - ipv6_only_sock(s) || - (s->sk_bound_dev_if && s->sk_bound_dev_if != dif)) - continue; - if (!ip_mc_sf_allow(s, loc_addr, rmt_addr, dif)) - continue; - goto found; - } - s = NULL; -found: - return s; -} - -/* - * 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. - * Header points to the ip header of the error packet. We move - * on past this. Then (as it used to claim before adjustment) - * header points to the first 8 bytes of the udp header. We need - * to find the appropriate port. - */ - -void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[]) -{ - struct inet_sock *inet; - struct iphdr *iph = (struct iphdr*)skb->data; - struct udphdr *uh = (struct udphdr*)(skb->data+(iph->ihl<<2)); - const int type = icmp_hdr(skb)->type; - const int code = icmp_hdr(skb)->code; - struct sock *sk; - int harderr; - int err; - - sk = __udp4_lib_lookup(skb->dev->nd_net, iph->daddr, uh->dest, - iph->saddr, uh->source, skb->dev->ifindex, udptable); - if (sk == NULL) { - ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); - return; /* No socket for error */ - } - - err = 0; - harderr = 0; - inet = inet_sk(sk); - - switch (type) { - default: - case ICMP_TIME_EXCEEDED: - err = EHOSTUNREACH; - break; - case ICMP_SOURCE_QUENCH: - goto out; - case ICMP_PARAMETERPROB: - err = EPROTO; - harderr = 1; - break; - case ICMP_DEST_UNREACH: - if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */ - if (inet->pmtudisc != IP_PMTUDISC_DONT) { - err = EMSGSIZE; - harderr = 1; - break; - } - goto out; - } - err = EHOSTUNREACH; - if (code <= NR_ICMP_UNREACH) { - harderr = icmp_err_convert[code].fatal; - err = icmp_err_convert[code].errno; - } - break; - } - - /* - * RFC1122: OK. Passes ICMP errors back to application, as per - * 4.1.3.3. - */ - if (!inet->recverr) { - if (!harderr || sk->sk_state != TCP_ESTABLISHED) - goto out; - } else { - ip_icmp_error(sk, skb, err, uh->dest, info, (u8*)(uh+1)); - } - sk->sk_err = err; - sk->sk_error_report(sk); -out: - sock_put(sk); -} - -void udp_err(struct sk_buff *skb, u32 info) -{ - __udp4_lib_err(skb, info, udp_hash); -} - -/* - * Throw away all pending data and cancel the corking. Socket is locked. - */ -static void udp_flush_pending_frames(struct sock *sk) -{ - struct udp_sock *up = udp_sk(sk); - - if (up->pending) { - up->len = 0; - up->pending = 0; - ip_flush_pending_frames(sk); - } -} - -/** - * udp4_hwcsum_outgoing - handle outgoing HW checksumming - * @sk: socket we are sending on - * @skb: sk_buff containing the filled-in UDP header - * (checksum field must be zeroed out) - */ -static void udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb, - __be32 src, __be32 dst, int len ) -{ - unsigned int offset; - struct udphdr *uh = udp_hdr(skb); - __wsum csum = 0; - - if (skb_queue_len(&sk->sk_write_queue) == 1) { - /* - * Only one fragment on the socket. - */ - skb->csum_start = skb_transport_header(skb) - skb->head; - skb->csum_offset = offsetof(struct udphdr, check); - uh->check = ~csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, 0); - } else { - /* - * HW-checksum won't work as there are two or more - * fragments on the socket so that all csums of sk_buffs - * should be together - */ - offset = skb_transport_offset(skb); - skb->csum = skb_checksum(skb, offset, skb->len - offset, 0); - - skb->ip_summed = CHECKSUM_NONE; - - skb_queue_walk(&sk->sk_write_queue, skb) { - csum = csum_add(csum, skb->csum); - } - - uh->check = csum_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - } -} - -/* - * Push out all pending data as one UDP datagram. Socket is locked. - */ -static int udp_push_pending_frames(struct sock *sk) -{ - struct udp_sock *up = udp_sk(sk); - struct inet_sock *inet = inet_sk(sk); - struct flowi *fl = &inet->cork.fl; - struct sk_buff *skb; - struct udphdr *uh; - int err = 0; - int is_udplite = IS_UDPLITE(sk); - __wsum csum = 0; - - /* Grab the skbuff where UDP header space exists. */ - if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) - goto out; - - /* - * Create a UDP header - */ - uh = udp_hdr(skb); - uh->source = fl->fl_ip_sport; - uh->dest = fl->fl_ip_dport; - uh->len = htons(up->len); - uh->check = 0; - - if (is_udplite) /* UDP-Lite */ - csum = udplite_csum_outgoing(sk, skb); - - else if (sk->sk_no_check == UDP_CSUM_NOXMIT) { /* UDP csum disabled */ - - skb->ip_summed = CHECKSUM_NONE; - goto send; - - } else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */ - - udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len); - goto send; - - } else /* `normal' UDP */ - csum = udp_csum_outgoing(sk, skb); - - /* add protocol-dependent pseudo-header */ - uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len, - sk->sk_protocol, csum ); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - -send: - err = ip_push_pending_frames(sk); -out: - up->len = 0; - up->pending = 0; - if (!err) - UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); - return err; -} - -int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, - size_t len) -{ - struct inet_sock *inet = inet_sk(sk); - struct udp_sock *up = udp_sk(sk); - int ulen = len; - struct ipcm_cookie ipc; - struct rtable *rt = NULL; - int free = 0; - int connected = 0; - __be32 daddr, faddr, saddr; - __be16 dport; - u8 tos; - int err, is_udplite = IS_UDPLITE(sk); - int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; - int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); - - if (len > 0xFFFF) - return -EMSGSIZE; - - /* - * Check the flags. - */ - - if (msg->msg_flags&MSG_OOB) /* Mirror BSD error message compatibility */ - return -EOPNOTSUPP; - - ipc.opt = NULL; - - if (up->pending) { - /* - * There are pending frames. - * The socket lock must be held while it's corked. - */ - lock_sock(sk); - if (likely(up->pending)) { - if (unlikely(up->pending != AF_INET)) { - release_sock(sk); - return -EINVAL; - } - goto do_append_data; - } - release_sock(sk); - } - ulen += sizeof(struct udphdr); - - /* - * Get and verify the address. - */ - if (msg->msg_name) { - struct sockaddr_in * usin = (struct sockaddr_in*)msg->msg_name; - if (msg->msg_namelen < sizeof(*usin)) - return -EINVAL; - if (usin->sin_family != AF_INET) { - if (usin->sin_family != AF_UNSPEC) - return -EAFNOSUPPORT; - } - - daddr = usin->sin_addr.s_addr; - dport = usin->sin_port; - if (dport == 0) - return -EINVAL; - } else { - if (sk->sk_state != TCP_ESTABLISHED) - return -EDESTADDRREQ; - daddr = inet->daddr; - dport = inet->dport; - /* Open fast path for connected socket. - Route will not be used, if at least one option is set. - */ - connected = 1; - } - ipc.addr = inet->saddr; - - ipc.oif = sk->sk_bound_dev_if; - if (msg->msg_controllen) { - err = ip_cmsg_send(msg, &ipc); - if (err) - return err; - if (ipc.opt) - free = 1; - connected = 0; - } - if (!ipc.opt) - ipc.opt = inet->opt; - - saddr = ipc.addr; - ipc.addr = faddr = daddr; - - if (ipc.opt && ipc.opt->srr) { - if (!daddr) - return -EINVAL; - faddr = ipc.opt->faddr; - connected = 0; - } - tos = RT_TOS(inet->tos); - if (sock_flag(sk, SOCK_LOCALROUTE) || - (msg->msg_flags & MSG_DONTROUTE) || - (ipc.opt && ipc.opt->is_strictroute)) { - tos |= RTO_ONLINK; - connected = 0; - } - - if (ipv4_is_multicast(daddr)) { - if (!ipc.oif) - ipc.oif = inet->mc_index; - if (!saddr) - saddr = inet->mc_addr; - connected = 0; - } - - if (connected) - rt = (struct rtable*)sk_dst_check(sk, 0); - - if (rt == NULL) { - struct flowi fl = { .oif = ipc.oif, - .nl_u = { .ip4_u = - { .daddr = faddr, - .saddr = saddr, - .tos = tos } }, - .proto = sk->sk_protocol, - .uli_u = { .ports = - { .sport = inet->sport, - .dport = dport } } }; - security_sk_classify_flow(sk, &fl); - err = ip_route_output_flow(&init_net, &rt, &fl, sk, 1); - if (err) { - if (err == -ENETUNREACH) - IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); - goto out; - } - - err = -EACCES; - if ((rt->rt_flags & RTCF_BROADCAST) && - !sock_flag(sk, SOCK_BROADCAST)) - goto out; - if (connected) - sk_dst_set(sk, dst_clone(&rt->u.dst)); - } - - if (msg->msg_flags&MSG_CONFIRM) - goto do_confirm; -back_from_confirm: - - saddr = rt->rt_src; - if (!ipc.addr) - daddr = ipc.addr = rt->rt_dst; - - lock_sock(sk); - if (unlikely(up->pending)) { - /* The socket is already corked while preparing it. */ - /* ... which is an evident application bug. --ANK */ - release_sock(sk); - - LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n"); - err = -EINVAL; - goto out; - } - /* - * Now cork the socket to pend data. - */ - inet->cork.fl.fl4_dst = daddr; - inet->cork.fl.fl_ip_dport = dport; - inet->cork.fl.fl4_src = saddr; - inet->cork.fl.fl_ip_sport = inet->sport; - up->pending = AF_INET; - -do_append_data: - up->len += ulen; - getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; - err = ip_append_data(sk, getfrag, msg->msg_iov, ulen, - sizeof(struct udphdr), &ipc, rt, - corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); - if (err) - udp_flush_pending_frames(sk); - else if (!corkreq) - err = udp_push_pending_frames(sk); - else if (unlikely(skb_queue_empty(&sk->sk_write_queue))) - up->pending = 0; - release_sock(sk); - -out: - ip_rt_put(rt); - if (free) - kfree(ipc.opt); - if (!err) - return len; - /* - * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space. Reporting - * ENOBUFS might not be good (it's not tunable per se), but otherwise - * we don't have a good statistic (IpOutDiscards but it can be too many - * things). We could add another new stat but at least for now that - * seems like overkill. - */ - if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); - } - return err; - -do_confirm: - dst_confirm(&rt->u.dst); - if (!(msg->msg_flags&MSG_PROBE) || len) - goto back_from_confirm; - err = 0; - goto out; -} - -int udp_sendpage(struct sock *sk, struct page *page, int offset, - size_t size, int flags) -{ - struct udp_sock *up = udp_sk(sk); - int ret; - - if (!up->pending) { - struct msghdr msg = { .msg_flags = flags|MSG_MORE }; - - /* Call udp_sendmsg to specify destination address which - * sendpage interface can't pass. - * This will succeed only when the socket is connected. - */ - ret = udp_sendmsg(NULL, sk, &msg, 0); - if (ret < 0) - return ret; - } - - lock_sock(sk); - - if (unlikely(!up->pending)) { - release_sock(sk); - - LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 3\n"); - return -EINVAL; - } - - ret = ip_append_page(sk, page, offset, size, flags); - if (ret == -EOPNOTSUPP) { - release_sock(sk); - return sock_no_sendpage(sk->sk_socket, page, offset, - size, flags); - } - if (ret < 0) { - udp_flush_pending_frames(sk); - goto out; - } - - up->len += size; - if (!(up->corkflag || (flags&MSG_MORE))) - ret = udp_push_pending_frames(sk); - if (!ret) - ret = size; -out: - release_sock(sk); - return ret; -} - -/* - * This should be easy, if there is something there we - * return it, otherwise we block. - */ - -int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, - size_t len, int noblock, int flags, int *addr_len) -{ - struct inet_sock *inet = inet_sk(sk); - struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name; - struct sk_buff *skb; - unsigned int ulen, copied; - int peeked; - int err; - int is_udplite = IS_UDPLITE(sk); - - /* - * Check any passed addresses - */ - if (addr_len) - *addr_len=sizeof(*sin); - - if (flags & MSG_ERRQUEUE) - return ip_recv_error(sk, msg, len); - -try_again: - skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0), - &peeked, &err); - if (!skb) - goto out; - - ulen = skb->len - sizeof(struct udphdr); - copied = len; - if (copied > ulen) - copied = ulen; - else if (copied < ulen) - msg->msg_flags |= MSG_TRUNC; - - /* - * If checksum is needed at all, try to do it while copying the - * data. If the data is truncated, or if we only want a partial - * coverage checksum (UDP-Lite), do it before the copy. - */ - - if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) { - if (udp_lib_checksum_complete(skb)) - goto csum_copy_err; - } - - if (skb_csum_unnecessary(skb)) - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), - msg->msg_iov, copied ); - else { - err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); - - if (err == -EINVAL) - goto csum_copy_err; - } - - if (err) - goto out_free; - - if (!peeked) - UDP_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite); - - sock_recv_timestamp(msg, sk, skb); - - /* Copy the address. */ - if (sin) - { - sin->sin_family = AF_INET; - sin->sin_port = udp_hdr(skb)->source; - sin->sin_addr.s_addr = ip_hdr(skb)->saddr; - memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); - } - if (inet->cmsg_flags) - ip_cmsg_recv(msg, skb); - - err = copied; - if (flags & MSG_TRUNC) - err = ulen; - -out_free: - lock_sock(sk); - skb_free_datagram(sk, skb); - release_sock(sk); -out: - return err; - -csum_copy_err: - lock_sock(sk); - if (!skb_kill_datagram(sk, skb, flags)) - UDP_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); - release_sock(sk); - - if (noblock) - return -EAGAIN; - goto try_again; -} - - -/* returns: - * -1: error - * 0: success - * >0: "udp encap" protocol resubmission - * - * Note that in the success and error cases, the skb is assumed to - * have either been requeued or freed. - */ -int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) -{ - struct udp_sock *up = udp_sk(sk); - int rc; - int is_udplite = IS_UDPLITE(sk); - - /* - * Charge it to the socket, dropping if the queue is full. - */ - if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb)) - goto drop; - nf_reset(skb); - - if (up->encap_type) { - /* - * This is an encapsulation socket so pass the skb to - * the socket's udp_encap_rcv() hook. Otherwise, just - * fall through and pass this up the UDP socket. - * up->encap_rcv() returns the following value: - * =0 if skb was successfully passed to the encap - * handler or was discarded by it. - * >0 if skb should be passed on to UDP. - * <0 if skb should be resubmitted as proto -N - */ - - /* if we're overly short, let UDP handle it */ - if (skb->len > sizeof(struct udphdr) && - up->encap_rcv != NULL) { - int ret; - - ret = (*up->encap_rcv)(sk, skb); - if (ret <= 0) { - UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, - is_udplite); - return -ret; - } - } - - /* FALLTHROUGH -- it's a UDP Packet */ - } - - /* - * UDP-Lite specific tests, ignored on UDP sockets - */ - if ((is_udplite & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { - - /* - * MIB statistics other than incrementing the error count are - * disabled for the following two types of errors: these depend - * on the application settings, not on the functioning of the - * protocol stack as such. - * - * RFC 3828 here recommends (sec 3.3): "There should also be a - * way ... to ... at least let the receiving application block - * delivery of packets with coverage values less than a value - * provided by the application." - */ - if (up->pcrlen == 0) { /* full coverage was set */ - LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage " - "%d while full coverage %d requested\n", - UDP_SKB_CB(skb)->cscov, skb->len); - goto drop; - } - /* The next case involves violating the min. coverage requested - * by the receiver. This is subtle: if receiver wants x and x is - * greater than the buffersize/MTU then receiver will complain - * that it wants x while sender emits packets of smaller size y. - * Therefore the above ...()->partial_cov statement is essential. - */ - if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { - LIMIT_NETDEBUG(KERN_WARNING - "UDPLITE: coverage %d too small, need min %d\n", - UDP_SKB_CB(skb)->cscov, up->pcrlen); - goto drop; - } - } - - if (sk->sk_filter) { - if (udp_lib_checksum_complete(skb)) - goto drop; - } - - if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { - /* Note that an ENOMEM error is charged twice */ - if (rc == -ENOMEM) - UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite); - goto drop; - } - - return 0; - -drop: - UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); - kfree_skb(skb); - return -1; -} - -/* - * Multicasts and broadcasts go to each listener. - * - * Note: called only from the BH handler context, - * so we don't need to lock the hashes. - */ -static int __udp4_lib_mcast_deliver(struct sk_buff *skb, - struct udphdr *uh, - __be32 saddr, __be32 daddr, - struct hlist_head udptable[]) -{ - struct sock *sk; - int dif; - - read_lock(&udp_hash_lock); - sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); - dif = skb->dev->ifindex; - sk = udp_v4_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); - if (sk) { - struct sock *sknext = NULL; - - do { - struct sk_buff *skb1 = skb; - - sknext = udp_v4_mcast_next(sk_next(sk), uh->dest, daddr, - uh->source, saddr, dif); - if (sknext) - skb1 = skb_clone(skb, GFP_ATOMIC); - - if (skb1) { - int ret = 0; - - bh_lock_sock_nested(sk); - if (!sock_owned_by_user(sk)) - ret = udp_queue_rcv_skb(sk, skb1); - else - sk_add_backlog(sk, skb1); - bh_unlock_sock(sk); - - if (ret > 0) - /* we should probably re-process instead - * of dropping packets here. */ - kfree_skb(skb1); - } - sk = sknext; - } while (sknext); - } else - kfree_skb(skb); - read_unlock(&udp_hash_lock); - return 0; -} - -/* Initialize UDP checksum. If exited with zero value (success), - * CHECKSUM_UNNECESSARY means, that no more checks are required. - * Otherwise, csum completion requires chacksumming packet body, - * including udp header and folding it to skb->csum. - */ -static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh, - int proto) -{ - const struct iphdr *iph; - int err; - - UDP_SKB_CB(skb)->partial_cov = 0; - UDP_SKB_CB(skb)->cscov = skb->len; - - if (IS_PROTO_UDPLITE(proto)) { - err = udplite_checksum_init(skb, uh); - if (err) - return err; - } - - iph = ip_hdr(skb); - if (uh->check == 0) { - skb->ip_summed = CHECKSUM_UNNECESSARY; - } else if (skb->ip_summed == CHECKSUM_COMPLETE) { - if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len, - proto, skb->csum)) - skb->ip_summed = CHECKSUM_UNNECESSARY; - } - if (!skb_csum_unnecessary(skb)) - skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, - skb->len, proto, 0); - /* Probably, we should checksum udp header (it should be in cache - * in any case) and data in tiny packets (< rx copybreak). - */ - - return 0; -} - -/* - * All we need to do is get the socket, and then do a checksum. - */ - -int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], - int proto) -{ - struct sock *sk; - struct udphdr *uh = udp_hdr(skb); - unsigned short ulen; - struct rtable *rt = skb->rtable; - __be32 saddr = ip_hdr(skb)->saddr; - __be32 daddr = ip_hdr(skb)->daddr; - - /* - * Validate the packet. - */ - if (!pskb_may_pull(skb, sizeof(struct udphdr))) - goto drop; /* No space for header. */ - - ulen = ntohs(uh->len); - if (ulen > skb->len) - goto short_packet; - - if (IS_PROTO_UDPLITE(proto)) { - /* UDP validates ulen. */ - if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen)) - goto short_packet; - uh = udp_hdr(skb); - } - - if (udp4_csum_init(skb, uh, proto)) - goto csum_error; - - if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST)) - return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable); - - sk = __udp4_lib_lookup(skb->dev->nd_net, saddr, uh->source, daddr, - uh->dest, inet_iif(skb), udptable); - - if (sk != NULL) { - int ret = 0; - bh_lock_sock_nested(sk); - if (!sock_owned_by_user(sk)) - ret = udp_queue_rcv_skb(sk, skb); - else - sk_add_backlog(sk, skb); - bh_unlock_sock(sk); - sock_put(sk); - - /* a return value > 0 means to resubmit the input, but - * it wants the return to be -protocol, or 0 - */ - if (ret > 0) - return -ret; - return 0; - } - - if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb)) - goto drop; - nf_reset(skb); - - /* No socket. Drop packet silently, if checksum is wrong */ - if (udp_lib_checksum_complete(skb)) - goto csum_error; - - UDP_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto)); - icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0); - - /* - * Hmm. We got an UDP packet to a port to which we - * don't wanna listen. Ignore it. - */ - kfree_skb(skb); - return 0; - -short_packet: - LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n", - IS_PROTO_UDPLITE(proto) ? "-Lite" : "", - NIPQUAD(saddr), - ntohs(uh->source), - ulen, - skb->len, - NIPQUAD(daddr), - ntohs(uh->dest)); - goto drop; - -csum_error: - /* - * RFC1122: OK. Discards the bad packet silently (as far as - * the network is concerned, anyway) as per 4.1.3.4 (MUST). - */ - LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n", - IS_PROTO_UDPLITE(proto) ? "-Lite" : "", - NIPQUAD(saddr), - ntohs(uh->source), - NIPQUAD(daddr), - ntohs(uh->dest), - ulen); -drop: - UDP_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto)); - kfree_skb(skb); - return 0; -} - -int udp_rcv(struct sk_buff *skb) -{ - return __udp4_lib_rcv(skb, udp_hash, IPPROTO_UDP); -} - -int udp_destroy_sock(struct sock *sk) -{ - lock_sock(sk); - udp_flush_pending_frames(sk); - release_sock(sk); - return 0; -} - -int udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_setsockopt(sk, level, optname, optval, optlen, - udp_push_pending_frames); - return ip_setsockopt(sk, level, optname, optval, optlen); -} - -#ifdef CONFIG_COMPAT -int compat_udp_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_setsockopt(sk, level, optname, optval, optlen, - udp_push_pending_frames); - return compat_ip_setsockopt(sk, level, optname, optval, optlen); -} -#endif - -int udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_getsockopt(sk, level, optname, optval, optlen); - return ip_getsockopt(sk, level, optname, optval, optlen); -} - -#ifdef CONFIG_COMPAT -int compat_udp_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_getsockopt(sk, level, optname, optval, optlen); - return compat_ip_getsockopt(sk, level, optname, optval, optlen); -} -#endif - -/* ------------------------------------------------------------------------ */ -DEFINE_PROTO_INUSE(udp) - -struct proto udp_prot = { - .name = "UDP", - .owner = THIS_MODULE, - .close = udp_lib_close, - .connect = ip4_datagram_connect, - .disconnect = udp_disconnect, - .ioctl = udp_ioctl, - .destroy = udp_destroy_sock, - .setsockopt = udp_setsockopt, - .getsockopt = udp_getsockopt, - .sendmsg = udp_sendmsg, - .recvmsg = udp_recvmsg, - .sendpage = udp_sendpage, - .backlog_rcv = udp_queue_rcv_skb, - .hash = udp_lib_hash, - .unhash = udp_lib_unhash, - .get_port = udp_v4_get_port, - .memory_allocated = &udp_memory_allocated, - .sysctl_mem = sysctl_udp_mem, - .sysctl_wmem = &sysctl_udp_wmem_min, - .sysctl_rmem = &sysctl_udp_rmem_min, - .obj_size = sizeof(struct udp_sock), -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_udp_setsockopt, - .compat_getsockopt = compat_udp_getsockopt, -#endif - REF_PROTO_INUSE(udp) -}; - -/* ------------------------------------------------------------------------ */ -static void udp4_format_sock(struct sock *sp, char *tmpbuf, int bucket) -{ - struct inet_sock *inet = inet_sk(sp); - __be32 dest = inet->daddr; - __be32 src = inet->rcv_saddr; - __u16 destp = ntohs(inet->dport); - __u16 srcp = ntohs(inet->sport); - - sprintf(tmpbuf, "%4d: %08X:%04X %08X:%04X" - " %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p", - bucket, src, srcp, dest, destp, sp->sk_state, - atomic_read(&sp->sk_wmem_alloc), - atomic_read(&sp->sk_rmem_alloc), - 0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp), - atomic_read(&sp->sk_refcnt), sp); -} - -int udp4_seq_show(struct seq_file *seq, void *v) -{ - if (v == SEQ_START_TOKEN) - seq_printf(seq, "%-127s\n", - " sl local_address rem_address st tx_queue " - "rx_queue tr tm->when retrnsmt uid timeout " - "inode"); - else { - char tmpbuf[129]; - struct udp_iter_state *state = seq->private; - - udp4_format_sock(v, tmpbuf, state->bucket); - seq_printf(seq, "%-127s\n", tmpbuf); - } - return 0; -} - -/* ------------------------------------------------------------------------ */ -#ifdef CONFIG_PROC_FS -static struct file_operations udp4_seq_fops; -static struct udp_seq_afinfo udp4_seq_afinfo = { - .owner = THIS_MODULE, - .name = "udp", - .family = AF_INET, - .hashtable = udp_hash, - .seq_show = udp4_seq_show, - .seq_fops = &udp4_seq_fops, -}; - -int __init udp4_proc_init(void) -{ - return udp_proc_register(&udp4_seq_afinfo); -} - -void udp4_proc_exit(void) -{ - udp_proc_unregister(&udp4_seq_afinfo); -} -#endif /* CONFIG_PROC_FS */ - -EXPORT_SYMBOL(udp_prot); -EXPORT_SYMBOL(udp_sendmsg); - diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c new file mode 100644 index 000000000000..d49c6d68c8a9 --- /dev/null +++ b/net/ipv4/udplite.c @@ -0,0 +1,121 @@ +/* + * UDPLITE An implementation of the UDP-Lite protocol (RFC 3828). + * + * Version: $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $ + * + * Authors: Gerrit Renker + * + * Changes: + * Fixes: + * This program 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 of the License, or (at your option) any later version. + */ +#include "udp_impl.h" +DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics) __read_mostly; + +struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; + +int udplite_get_port(struct sock *sk, unsigned short p, + int (*c)(const struct sock *, const struct sock *)) +{ + return __udp_lib_get_port(sk, p, udplite_hash, c); +} + +static int udplite_v4_get_port(struct sock *sk, unsigned short snum) +{ + return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal); +} + +static int udplite_rcv(struct sk_buff *skb) +{ + return __udp4_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE); +} + +static void udplite_err(struct sk_buff *skb, u32 info) +{ + __udp4_lib_err(skb, info, udplite_hash); +} + +static struct net_protocol udplite_protocol = { + .handler = udplite_rcv, + .err_handler = udplite_err, + .no_policy = 1, +}; + +DEFINE_PROTO_INUSE(udplite) + +struct proto udplite_prot = { + .name = "UDP-Lite", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip4_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .init = udplite_sk_init, + .destroy = udp_destroy_sock, + .setsockopt = udp_setsockopt, + .getsockopt = udp_getsockopt, + .sendmsg = udp_sendmsg, + .recvmsg = udp_recvmsg, + .sendpage = udp_sendpage, + .backlog_rcv = udp_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udplite_v4_get_port, + .obj_size = sizeof(struct udp_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udp_setsockopt, + .compat_getsockopt = compat_udp_getsockopt, +#endif + REF_PROTO_INUSE(udplite) +}; + +static struct inet_protosw udplite4_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDPLITE, + .prot = &udplite_prot, + .ops = &inet_dgram_ops, + .capability = -1, + .no_check = 0, /* must checksum (RFC 3828) */ + .flags = INET_PROTOSW_PERMANENT, +}; + +#ifdef CONFIG_PROC_FS +static struct file_operations udplite4_seq_fops; +static struct udp_seq_afinfo udplite4_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udplite", + .family = AF_INET, + .hashtable = udplite_hash, + .seq_show = udp4_seq_show, + .seq_fops = &udplite4_seq_fops, +}; +#endif + +void __init udplite4_register(void) +{ + if (proto_register(&udplite_prot, 1)) + goto out_register_err; + + if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0) + goto out_unregister_proto; + + inet_register_protosw(&udplite4_protosw); + +#ifdef CONFIG_PROC_FS + if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */ + printk(KERN_ERR "%s: Cannot register /proc!\n", __func__); +#endif + return; + +out_unregister_proto: + proto_unregister(&udplite_prot); +out_register_err: + printk(KERN_CRIT "%s: Cannot add UDP-Lite protocol.\n", __func__); +} + +EXPORT_SYMBOL(udplite_hash); +EXPORT_SYMBOL(udplite_prot); +EXPORT_SYMBOL(udplite_get_port); diff --git a/net/ipv4/udplite_ipv4.c b/net/ipv4/udplite_ipv4.c deleted file mode 100644 index d49c6d68c8a9..000000000000 --- a/net/ipv4/udplite_ipv4.c +++ /dev/null @@ -1,121 +0,0 @@ -/* - * UDPLITE An implementation of the UDP-Lite protocol (RFC 3828). - * - * Version: $Id: udplite.c,v 1.25 2006/10/19 07:22:36 gerrit Exp $ - * - * Authors: Gerrit Renker - * - * Changes: - * Fixes: - * This program 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 of the License, or (at your option) any later version. - */ -#include "udp_impl.h" -DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics) __read_mostly; - -struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; - -int udplite_get_port(struct sock *sk, unsigned short p, - int (*c)(const struct sock *, const struct sock *)) -{ - return __udp_lib_get_port(sk, p, udplite_hash, c); -} - -static int udplite_v4_get_port(struct sock *sk, unsigned short snum) -{ - return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal); -} - -static int udplite_rcv(struct sk_buff *skb) -{ - return __udp4_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE); -} - -static void udplite_err(struct sk_buff *skb, u32 info) -{ - __udp4_lib_err(skb, info, udplite_hash); -} - -static struct net_protocol udplite_protocol = { - .handler = udplite_rcv, - .err_handler = udplite_err, - .no_policy = 1, -}; - -DEFINE_PROTO_INUSE(udplite) - -struct proto udplite_prot = { - .name = "UDP-Lite", - .owner = THIS_MODULE, - .close = udp_lib_close, - .connect = ip4_datagram_connect, - .disconnect = udp_disconnect, - .ioctl = udp_ioctl, - .init = udplite_sk_init, - .destroy = udp_destroy_sock, - .setsockopt = udp_setsockopt, - .getsockopt = udp_getsockopt, - .sendmsg = udp_sendmsg, - .recvmsg = udp_recvmsg, - .sendpage = udp_sendpage, - .backlog_rcv = udp_queue_rcv_skb, - .hash = udp_lib_hash, - .unhash = udp_lib_unhash, - .get_port = udplite_v4_get_port, - .obj_size = sizeof(struct udp_sock), -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_udp_setsockopt, - .compat_getsockopt = compat_udp_getsockopt, -#endif - REF_PROTO_INUSE(udplite) -}; - -static struct inet_protosw udplite4_protosw = { - .type = SOCK_DGRAM, - .protocol = IPPROTO_UDPLITE, - .prot = &udplite_prot, - .ops = &inet_dgram_ops, - .capability = -1, - .no_check = 0, /* must checksum (RFC 3828) */ - .flags = INET_PROTOSW_PERMANENT, -}; - -#ifdef CONFIG_PROC_FS -static struct file_operations udplite4_seq_fops; -static struct udp_seq_afinfo udplite4_seq_afinfo = { - .owner = THIS_MODULE, - .name = "udplite", - .family = AF_INET, - .hashtable = udplite_hash, - .seq_show = udp4_seq_show, - .seq_fops = &udplite4_seq_fops, -}; -#endif - -void __init udplite4_register(void) -{ - if (proto_register(&udplite_prot, 1)) - goto out_register_err; - - if (inet_add_protocol(&udplite_protocol, IPPROTO_UDPLITE) < 0) - goto out_unregister_proto; - - inet_register_protosw(&udplite4_protosw); - -#ifdef CONFIG_PROC_FS - if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */ - printk(KERN_ERR "%s: Cannot register /proc!\n", __func__); -#endif - return; - -out_unregister_proto: - proto_unregister(&udplite_prot); -out_register_err: - printk(KERN_CRIT "%s: Cannot add UDP-Lite protocol.\n", __func__); -} - -EXPORT_SYMBOL(udplite_hash); -EXPORT_SYMBOL(udplite_prot); -EXPORT_SYMBOL(udplite_get_port); diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 107051f7c227..ae14617e607f 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -6,7 +6,7 @@ obj-$(CONFIG_IPV6) += ipv6.o ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ addrlabel.o \ - route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp_ipv6.o \ + route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \ raw.o protocol.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o @@ -17,7 +17,6 @@ ipv6-$(CONFIG_NETFILTER) += netfilter.o ipv6-$(CONFIG_IPV6_MULTIPLE_TABLES) += fib6_rules.o ipv6-$(CONFIG_PROC_FS) += proc.o ipv6-$(CONFIG_SYN_COOKIES) += syncookies.o -ipv6-$(CONFIG_IP_UDPLITE) += udplite_ipv6.o ipv6-objs += $(ipv6-y) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index afe9276d0420..730a861b8f41 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -813,16 +813,12 @@ static int __init init_ipv6_mibs(void) goto err_icmpmsg_mib; if (snmp_mib_init((void **)udp_stats_in6, sizeof (struct udp_mib)) < 0) goto err_udp_mib; -#ifdef CONFIG_IP_UDPLITE if (snmp_mib_init((void **)udplite_stats_in6, sizeof (struct udp_mib)) < 0) goto err_udplite_mib; -#endif return 0; -#ifdef CONFIG_IP_UDPLITE err_udplite_mib: -#endif snmp_mib_free((void **)udp_stats_in6); err_udp_mib: snmp_mib_free((void **)icmpv6msg_statistics); @@ -841,9 +837,7 @@ static void cleanup_ipv6_mibs(void) snmp_mib_free((void **)icmpv6_statistics); snmp_mib_free((void **)icmpv6msg_statistics); snmp_mib_free((void **)udp_stats_in6); -#ifdef CONFIG_IP_UDPLITE snmp_mib_free((void **)udplite_stats_in6); -#endif } static int inet6_net_init(struct net *net) @@ -888,11 +882,9 @@ static int __init inet6_init(void) if (err) goto out_unregister_tcp_proto; -#ifdef CONFIG_IP_UDPLITE err = proto_register(&udplitev6_prot, 1); if (err) goto out_unregister_udp_proto; -#endif err = proto_register(&rawv6_prot, 1); if (err) @@ -1063,10 +1055,8 @@ out_sock_register_fail: out_unregister_raw_proto: proto_unregister(&rawv6_prot); out_unregister_udplite_proto: -#ifdef CONFIG_IP_UDPLITE proto_unregister(&udplitev6_prot); out_unregister_udp_proto: -#endif proto_unregister(&udpv6_prot); out_unregister_tcp_proto: proto_unregister(&tcpv6_prot); @@ -1085,9 +1075,7 @@ static void __exit inet6_exit(void) ipv6_sysctl_unregister(); #endif udpv6_exit(); -#ifdef CONFIG_IP_UDPLITE udplitev6_exit(); -#endif tcpv6_exit(); /* Cleanup code parts. */ @@ -1117,9 +1105,7 @@ static void __exit inet6_exit(void) unregister_pernet_subsys(&inet6_net_ops); cleanup_ipv6_mibs(); proto_unregister(&rawv6_prot); -#ifdef CONFIG_IP_UDPLITE proto_unregister(&udplitev6_prot); -#endif proto_unregister(&udpv6_prot); proto_unregister(&tcpv6_prot); } diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 3bbfdff698d2..5eea6fa506e5 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -127,9 +127,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, struct sk_buff *pktopt; if (sk->sk_protocol != IPPROTO_UDP && -#ifdef CONFIG_IP_UDPLITE sk->sk_protocol != IPPROTO_UDPLITE && -#endif sk->sk_protocol != IPPROTO_TCP) break; @@ -169,7 +167,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, } else { struct proto *prot = &udp_prot; - if (IS_PROTO_UDPLITE(sk->sk_protocol)) + if (sk->sk_protocol == IPPROTO_UDPLITE) prot = &udplite_prot; local_bh_disable(); sock_prot_inuse_add(sk->sk_prot, -1); @@ -734,9 +732,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, switch (optname) { case IPV6_ADDRFORM: if (sk->sk_protocol != IPPROTO_UDP && -#ifdef CONFIG_IP_UDPLITE sk->sk_protocol != IPPROTO_UDPLITE && -#endif sk->sk_protocol != IPPROTO_TCP) return -EINVAL; if (sk->sk_state != TCP_ESTABLISHED) diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 2453f2229ef7..8a5be290c710 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -39,10 +39,8 @@ static int sockstat6_seq_show(struct seq_file *seq, void *v) sock_prot_inuse_get(&tcpv6_prot)); seq_printf(seq, "UDP6: inuse %d\n", sock_prot_inuse_get(&udpv6_prot)); -#ifdef CONFIG_IP_UDPLITE seq_printf(seq, "UDPLITE6: inuse %d\n", sock_prot_inuse_get(&udplitev6_prot)); -#endif seq_printf(seq, "RAW6: inuse %d\n", sock_prot_inuse_get(&rawv6_prot)); seq_printf(seq, "FRAG6: inuse %d memory %d\n", @@ -113,7 +111,6 @@ static struct snmp_mib snmp6_udp6_list[] = { SNMP_MIB_SENTINEL }; -#ifdef CONFIG_IP_UDPLITE static struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_ITEM("UdpLite6InDatagrams", UDP_MIB_INDATAGRAMS), SNMP_MIB_ITEM("UdpLite6NoPorts", UDP_MIB_NOPORTS), @@ -121,7 +118,6 @@ static struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_ITEM("UdpLite6OutDatagrams", UDP_MIB_OUTDATAGRAMS), SNMP_MIB_SENTINEL }; -#endif static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) { @@ -180,9 +176,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) snmp6_seq_show_item(seq, (void **)icmpv6_statistics, snmp6_icmp6_list); snmp6_seq_show_icmpv6msg(seq, (void **)icmpv6msg_statistics); snmp6_seq_show_item(seq, (void **)udp_stats_in6, snmp6_udp6_list); -#ifdef CONFIG_IP_UDPLITE snmp6_seq_show_item(seq, (void **)udplite_stats_in6, snmp6_udplite6_list); -#endif } return 0; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c new file mode 100644 index 000000000000..53739de829db --- /dev/null +++ b/net/ipv6/udp.c @@ -0,0 +1,1065 @@ +/* + * UDP over IPv6 + * Linux INET6 implementation + * + * Authors: + * Pedro Roque + * + * Based on linux/ipv4/udp.c + * + * $Id: udp.c,v 1.65 2002/02/01 22:01:04 davem Exp $ + * + * Fixes: + * Hideaki YOSHIFUJI : sin6_scope_id support + * YOSHIFUJI Hideaki @USAGI and: Support IPV6_V6ONLY socket option, which + * Alexey Kuznetsov allow both IPv4 and IPv6 sockets to bind + * a single port at the same time. + * Kazunori MIYAZAWA @USAGI: change process style to use ip6_append_data + * YOSHIFUJI Hideaki @USAGI: convert /proc/net/udp6 to seq_file. + * + * This program 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 of the License, or (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include "udp_impl.h" + +static inline int udp_v6_get_port(struct sock *sk, unsigned short snum) +{ + return udp_get_port(sk, snum, ipv6_rcv_saddr_equal); +} + +static struct sock *__udp6_lib_lookup(struct net *net, + struct in6_addr *saddr, __be16 sport, + struct in6_addr *daddr, __be16 dport, + int dif, struct hlist_head udptable[]) +{ + struct sock *sk, *result = NULL; + struct hlist_node *node; + unsigned short hnum = ntohs(dport); + int badness = -1; + + read_lock(&udp_hash_lock); + sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { + struct inet_sock *inet = inet_sk(sk); + + if (sk->sk_net == net && sk->sk_hash == hnum && + sk->sk_family == PF_INET6) { + struct ipv6_pinfo *np = inet6_sk(sk); + int score = 0; + if (inet->dport) { + if (inet->dport != sport) + continue; + score++; + } + if (!ipv6_addr_any(&np->rcv_saddr)) { + if (!ipv6_addr_equal(&np->rcv_saddr, daddr)) + continue; + score++; + } + if (!ipv6_addr_any(&np->daddr)) { + if (!ipv6_addr_equal(&np->daddr, saddr)) + continue; + score++; + } + if (sk->sk_bound_dev_if) { + if (sk->sk_bound_dev_if != dif) + continue; + score++; + } + if (score == 4) { + result = sk; + break; + } else if (score > badness) { + result = sk; + badness = score; + } + } + } + if (result) + sock_hold(result); + read_unlock(&udp_hash_lock); + return result; +} + +/* + * This should be easy, if there is something there we + * return it, otherwise we block. + */ + +int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len, + int noblock, int flags, int *addr_len) +{ + struct ipv6_pinfo *np = inet6_sk(sk); + struct inet_sock *inet = inet_sk(sk); + struct sk_buff *skb; + unsigned int ulen, copied; + int peeked; + int err; + int is_udplite = IS_UDPLITE(sk); + + if (addr_len) + *addr_len=sizeof(struct sockaddr_in6); + + if (flags & MSG_ERRQUEUE) + return ipv6_recv_error(sk, msg, len); + +try_again: + skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0), + &peeked, &err); + if (!skb) + goto out; + + ulen = skb->len - sizeof(struct udphdr); + copied = len; + if (copied > ulen) + copied = ulen; + else if (copied < ulen) + msg->msg_flags |= MSG_TRUNC; + + /* + * If checksum is needed at all, try to do it while copying the + * data. If the data is truncated, or if we only want a partial + * coverage checksum (UDP-Lite), do it before the copy. + */ + + if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) { + if (udp_lib_checksum_complete(skb)) + goto csum_copy_err; + } + + if (skb_csum_unnecessary(skb)) + err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), + msg->msg_iov, copied ); + else { + err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); + if (err == -EINVAL) + goto csum_copy_err; + } + if (err) + goto out_free; + + if (!peeked) + UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite); + + sock_recv_timestamp(msg, sk, skb); + + /* Copy the address. */ + if (msg->msg_name) { + struct sockaddr_in6 *sin6; + + sin6 = (struct sockaddr_in6 *) msg->msg_name; + sin6->sin6_family = AF_INET6; + sin6->sin6_port = udp_hdr(skb)->source; + sin6->sin6_flowinfo = 0; + sin6->sin6_scope_id = 0; + + if (skb->protocol == htons(ETH_P_IP)) + ipv6_addr_set(&sin6->sin6_addr, 0, 0, + htonl(0xffff), ip_hdr(skb)->saddr); + else { + ipv6_addr_copy(&sin6->sin6_addr, + &ipv6_hdr(skb)->saddr); + if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL) + sin6->sin6_scope_id = IP6CB(skb)->iif; + } + + } + if (skb->protocol == htons(ETH_P_IP)) { + if (inet->cmsg_flags) + ip_cmsg_recv(msg, skb); + } else { + if (np->rxopt.all) + datagram_recv_ctl(sk, msg, skb); + } + + err = copied; + if (flags & MSG_TRUNC) + err = ulen; + +out_free: + lock_sock(sk); + skb_free_datagram(sk, skb); + release_sock(sk); +out: + return err; + +csum_copy_err: + lock_sock(sk); + if (!skb_kill_datagram(sk, skb, flags)) + UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); + release_sock(sk); + + if (flags & MSG_DONTWAIT) + return -EAGAIN; + goto try_again; +} + +void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, + int type, int code, int offset, __be32 info, + struct hlist_head udptable[] ) +{ + struct ipv6_pinfo *np; + struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; + struct in6_addr *saddr = &hdr->saddr; + struct in6_addr *daddr = &hdr->daddr; + struct udphdr *uh = (struct udphdr*)(skb->data+offset); + struct sock *sk; + int err; + + sk = __udp6_lib_lookup(skb->dev->nd_net, daddr, uh->dest, + saddr, uh->source, inet6_iif(skb), udptable); + if (sk == NULL) + return; + + np = inet6_sk(sk); + + if (!icmpv6_err_convert(type, code, &err) && !np->recverr) + goto out; + + if (sk->sk_state != TCP_ESTABLISHED && !np->recverr) + goto out; + + if (np->recverr) + ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1)); + + sk->sk_err = err; + sk->sk_error_report(sk); +out: + sock_put(sk); +} + +static __inline__ void udpv6_err(struct sk_buff *skb, + struct inet6_skb_parm *opt, int type, + int code, int offset, __be32 info ) +{ + __udp6_lib_err(skb, opt, type, code, offset, info, udp_hash); +} + +int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) +{ + struct udp_sock *up = udp_sk(sk); + int rc; + int is_udplite = IS_UDPLITE(sk); + + if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) + goto drop; + + /* + * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c). + */ + if ((is_udplite & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { + + if (up->pcrlen == 0) { /* full coverage was set */ + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage" + " %d while full coverage %d requested\n", + UDP_SKB_CB(skb)->cscov, skb->len); + goto drop; + } + if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { + LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d " + "too small, need min %d\n", + UDP_SKB_CB(skb)->cscov, up->pcrlen); + goto drop; + } + } + + if (sk->sk_filter) { + if (udp_lib_checksum_complete(skb)) + goto drop; + } + + if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { + /* Note that an ENOMEM error is charged twice */ + if (rc == -ENOMEM) + UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite); + goto drop; + } + + return 0; +drop: + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); + kfree_skb(skb); + return -1; +} + +static struct sock *udp_v6_mcast_next(struct sock *sk, + __be16 loc_port, struct in6_addr *loc_addr, + __be16 rmt_port, struct in6_addr *rmt_addr, + int dif) +{ + struct hlist_node *node; + struct sock *s = sk; + unsigned short num = ntohs(loc_port); + + sk_for_each_from(s, node) { + struct inet_sock *inet = inet_sk(s); + + if (s->sk_hash == num && s->sk_family == PF_INET6) { + struct ipv6_pinfo *np = inet6_sk(s); + if (inet->dport) { + if (inet->dport != rmt_port) + continue; + } + if (!ipv6_addr_any(&np->daddr) && + !ipv6_addr_equal(&np->daddr, rmt_addr)) + continue; + + if (s->sk_bound_dev_if && s->sk_bound_dev_if != dif) + continue; + + if (!ipv6_addr_any(&np->rcv_saddr)) { + if (!ipv6_addr_equal(&np->rcv_saddr, loc_addr)) + continue; + } + if (!inet6_mc_check(s, loc_addr, rmt_addr)) + continue; + return s; + } + } + return NULL; +} + +/* + * Note: called only from the BH handler context, + * so we don't need to lock the hashes. + */ +static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr, + struct in6_addr *daddr, struct hlist_head udptable[]) +{ + struct sock *sk, *sk2; + const struct udphdr *uh = udp_hdr(skb); + int dif; + + read_lock(&udp_hash_lock); + sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); + dif = inet6_iif(skb); + sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); + if (!sk) { + kfree_skb(skb); + goto out; + } + + sk2 = sk; + while ((sk2 = udp_v6_mcast_next(sk_next(sk2), uh->dest, daddr, + uh->source, saddr, dif))) { + struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC); + if (buff) { + bh_lock_sock_nested(sk2); + if (!sock_owned_by_user(sk2)) + udpv6_queue_rcv_skb(sk2, buff); + else + sk_add_backlog(sk2, buff); + bh_unlock_sock(sk2); + } + } + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + udpv6_queue_rcv_skb(sk, skb); + else + sk_add_backlog(sk, skb); + bh_unlock_sock(sk); +out: + read_unlock(&udp_hash_lock); + return 0; +} + +static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, + int proto) +{ + int err; + + UDP_SKB_CB(skb)->partial_cov = 0; + UDP_SKB_CB(skb)->cscov = skb->len; + + if (proto == IPPROTO_UDPLITE) { + err = udplite_checksum_init(skb, uh); + if (err) + return err; + } + + if (uh->check == 0) { + /* RFC 2460 section 8.1 says that we SHOULD log + this error. Well, it is reasonable. + */ + LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n"); + return 1; + } + if (skb->ip_summed == CHECKSUM_COMPLETE && + !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, + skb->len, proto, skb->csum)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + + if (!skb_csum_unnecessary(skb)) + skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr, + &ipv6_hdr(skb)->daddr, + skb->len, proto, 0)); + + return 0; +} + +int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], + int proto) +{ + struct sock *sk; + struct udphdr *uh; + struct net_device *dev = skb->dev; + struct in6_addr *saddr, *daddr; + u32 ulen = 0; + + if (!pskb_may_pull(skb, sizeof(struct udphdr))) + goto short_packet; + + saddr = &ipv6_hdr(skb)->saddr; + daddr = &ipv6_hdr(skb)->daddr; + uh = udp_hdr(skb); + + ulen = ntohs(uh->len); + if (ulen > skb->len) + goto short_packet; + + if (proto == IPPROTO_UDP) { + /* UDP validates ulen. */ + + /* Check for jumbo payload */ + if (ulen == 0) + ulen = skb->len; + + if (ulen < sizeof(*uh)) + goto short_packet; + + if (ulen < skb->len) { + if (pskb_trim_rcsum(skb, ulen)) + goto short_packet; + saddr = &ipv6_hdr(skb)->saddr; + daddr = &ipv6_hdr(skb)->daddr; + uh = udp_hdr(skb); + } + } + + if (udp6_csum_init(skb, uh, proto)) + goto discard; + + /* + * Multicast receive code + */ + if (ipv6_addr_is_multicast(daddr)) + return __udp6_lib_mcast_deliver(skb, saddr, daddr, udptable); + + /* Unicast */ + + /* + * check socket cache ... must talk to Alan about his plans + * for sock caches... i'll skip this for now. + */ + sk = __udp6_lib_lookup(skb->dev->nd_net, saddr, uh->source, + daddr, uh->dest, inet6_iif(skb), udptable); + + if (sk == NULL) { + if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) + goto discard; + + if (udp_lib_checksum_complete(skb)) + goto discard; + UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE); + + icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev); + + kfree_skb(skb); + return 0; + } + + /* deliver */ + + bh_lock_sock_nested(sk); + if (!sock_owned_by_user(sk)) + udpv6_queue_rcv_skb(sk, skb); + else + sk_add_backlog(sk, skb); + bh_unlock_sock(sk); + sock_put(sk); + return 0; + +short_packet: + LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n", + proto == IPPROTO_UDPLITE ? "-Lite" : "", + ulen, skb->len); + +discard: + UDP6_INC_STATS_BH(UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE); + kfree_skb(skb); + return 0; +} + +static __inline__ int udpv6_rcv(struct sk_buff *skb) +{ + return __udp6_lib_rcv(skb, udp_hash, IPPROTO_UDP); +} + +/* + * Throw away all pending data and cancel the corking. Socket is locked. + */ +static void udp_v6_flush_pending_frames(struct sock *sk) +{ + struct udp_sock *up = udp_sk(sk); + + if (up->pending) { + up->len = 0; + up->pending = 0; + ip6_flush_pending_frames(sk); + } +} + +/* + * Sending + */ + +static int udp_v6_push_pending_frames(struct sock *sk) +{ + struct sk_buff *skb; + struct udphdr *uh; + struct udp_sock *up = udp_sk(sk); + struct inet_sock *inet = inet_sk(sk); + struct flowi *fl = &inet->cork.fl; + int err = 0; + int is_udplite = IS_UDPLITE(sk); + __wsum csum = 0; + + /* Grab the skbuff where UDP header space exists. */ + if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) + goto out; + + /* + * Create a UDP header + */ + uh = udp_hdr(skb); + uh->source = fl->fl_ip_sport; + uh->dest = fl->fl_ip_dport; + uh->len = htons(up->len); + uh->check = 0; + + if (is_udplite) + csum = udplite_csum_outgoing(sk, skb); + else + csum = udp_csum_outgoing(sk, skb); + + /* add protocol-dependent pseudo-header */ + uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, + up->len, fl->proto, csum ); + if (uh->check == 0) + uh->check = CSUM_MANGLED_0; + + err = ip6_push_pending_frames(sk); +out: + up->len = 0; + up->pending = 0; + if (!err) + UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); + return err; +} + +int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, + struct msghdr *msg, size_t len) +{ + struct ipv6_txoptions opt_space; + struct udp_sock *up = udp_sk(sk); + struct inet_sock *inet = inet_sk(sk); + struct ipv6_pinfo *np = inet6_sk(sk); + struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) msg->msg_name; + struct in6_addr *daddr, *final_p = NULL, final; + struct ipv6_txoptions *opt = NULL; + struct ip6_flowlabel *flowlabel = NULL; + struct flowi fl; + struct dst_entry *dst; + int addr_len = msg->msg_namelen; + int ulen = len; + int hlimit = -1; + int tclass = -1; + int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; + int err; + int connected = 0; + int is_udplite = IS_UDPLITE(sk); + int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); + + /* destination address check */ + if (sin6) { + if (addr_len < offsetof(struct sockaddr, sa_data)) + return -EINVAL; + + switch (sin6->sin6_family) { + case AF_INET6: + if (addr_len < SIN6_LEN_RFC2133) + return -EINVAL; + daddr = &sin6->sin6_addr; + break; + case AF_INET: + goto do_udp_sendmsg; + case AF_UNSPEC: + msg->msg_name = sin6 = NULL; + msg->msg_namelen = addr_len = 0; + daddr = NULL; + break; + default: + return -EINVAL; + } + } else if (!up->pending) { + if (sk->sk_state != TCP_ESTABLISHED) + return -EDESTADDRREQ; + daddr = &np->daddr; + } else + daddr = NULL; + + if (daddr) { + if (ipv6_addr_v4mapped(daddr)) { + struct sockaddr_in sin; + sin.sin_family = AF_INET; + sin.sin_port = sin6 ? sin6->sin6_port : inet->dport; + sin.sin_addr.s_addr = daddr->s6_addr32[3]; + msg->msg_name = &sin; + msg->msg_namelen = sizeof(sin); +do_udp_sendmsg: + if (__ipv6_only_sock(sk)) + return -ENETUNREACH; + return udp_sendmsg(iocb, sk, msg, len); + } + } + + if (up->pending == AF_INET) + return udp_sendmsg(iocb, sk, msg, len); + + /* Rough check on arithmetic overflow, + better check is made in ip6_append_data(). + */ + if (len > INT_MAX - sizeof(struct udphdr)) + return -EMSGSIZE; + + if (up->pending) { + /* + * There are pending frames. + * The socket lock must be held while it's corked. + */ + lock_sock(sk); + if (likely(up->pending)) { + if (unlikely(up->pending != AF_INET6)) { + release_sock(sk); + return -EAFNOSUPPORT; + } + dst = NULL; + goto do_append_data; + } + release_sock(sk); + } + ulen += sizeof(struct udphdr); + + memset(&fl, 0, sizeof(fl)); + + if (sin6) { + if (sin6->sin6_port == 0) + return -EINVAL; + + fl.fl_ip_dport = sin6->sin6_port; + daddr = &sin6->sin6_addr; + + if (np->sndflow) { + fl.fl6_flowlabel = sin6->sin6_flowinfo&IPV6_FLOWINFO_MASK; + if (fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) { + flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel); + if (flowlabel == NULL) + return -EINVAL; + daddr = &flowlabel->dst; + } + } + + /* + * Otherwise it will be difficult to maintain + * sk->sk_dst_cache. + */ + if (sk->sk_state == TCP_ESTABLISHED && + ipv6_addr_equal(daddr, &np->daddr)) + daddr = &np->daddr; + + if (addr_len >= sizeof(struct sockaddr_in6) && + sin6->sin6_scope_id && + ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL) + fl.oif = sin6->sin6_scope_id; + } else { + if (sk->sk_state != TCP_ESTABLISHED) + return -EDESTADDRREQ; + + fl.fl_ip_dport = inet->dport; + daddr = &np->daddr; + fl.fl6_flowlabel = np->flow_label; + connected = 1; + } + + if (!fl.oif) + fl.oif = sk->sk_bound_dev_if; + + if (msg->msg_controllen) { + opt = &opt_space; + memset(opt, 0, sizeof(struct ipv6_txoptions)); + opt->tot_len = sizeof(*opt); + + err = datagram_send_ctl(msg, &fl, opt, &hlimit, &tclass); + if (err < 0) { + fl6_sock_release(flowlabel); + return err; + } + if ((fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) && !flowlabel) { + flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel); + if (flowlabel == NULL) + return -EINVAL; + } + if (!(opt->opt_nflen|opt->opt_flen)) + opt = NULL; + connected = 0; + } + if (opt == NULL) + opt = np->opt; + if (flowlabel) + opt = fl6_merge_options(&opt_space, flowlabel, opt); + opt = ipv6_fixup_options(&opt_space, opt); + + fl.proto = sk->sk_protocol; + ipv6_addr_copy(&fl.fl6_dst, daddr); + if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr)) + ipv6_addr_copy(&fl.fl6_src, &np->saddr); + fl.fl_ip_sport = inet->sport; + + /* merge ip6_build_xmit from ip6_output */ + if (opt && opt->srcrt) { + struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; + ipv6_addr_copy(&final, &fl.fl6_dst); + ipv6_addr_copy(&fl.fl6_dst, rt0->addr); + final_p = &final; + connected = 0; + } + + if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) { + fl.oif = np->mcast_oif; + connected = 0; + } + + security_sk_classify_flow(sk, &fl); + + err = ip6_sk_dst_lookup(sk, &dst, &fl); + if (err) + goto out; + if (final_p) + ipv6_addr_copy(&fl.fl6_dst, final_p); + + if ((err = __xfrm_lookup(&dst, &fl, sk, XFRM_LOOKUP_WAIT)) < 0) { + if (err == -EREMOTE) + err = ip6_dst_blackhole(sk, &dst, &fl); + if (err < 0) + goto out; + } + + if (hlimit < 0) { + if (ipv6_addr_is_multicast(&fl.fl6_dst)) + hlimit = np->mcast_hops; + else + hlimit = np->hop_limit; + if (hlimit < 0) + hlimit = dst_metric(dst, RTAX_HOPLIMIT); + if (hlimit < 0) + hlimit = ipv6_get_hoplimit(dst->dev); + } + + if (tclass < 0) { + tclass = np->tclass; + if (tclass < 0) + tclass = 0; + } + + if (msg->msg_flags&MSG_CONFIRM) + goto do_confirm; +back_from_confirm: + + lock_sock(sk); + if (unlikely(up->pending)) { + /* The socket is already corked while preparing it. */ + /* ... which is an evident application bug. --ANK */ + release_sock(sk); + + LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n"); + err = -EINVAL; + goto out; + } + + up->pending = AF_INET6; + +do_append_data: + up->len += ulen; + getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; + err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen, + sizeof(struct udphdr), hlimit, tclass, opt, &fl, + (struct rt6_info*)dst, + corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); + if (err) + udp_v6_flush_pending_frames(sk); + else if (!corkreq) + err = udp_v6_push_pending_frames(sk); + else if (unlikely(skb_queue_empty(&sk->sk_write_queue))) + up->pending = 0; + + if (dst) { + if (connected) { + ip6_dst_store(sk, dst, + ipv6_addr_equal(&fl.fl6_dst, &np->daddr) ? + &np->daddr : NULL, +#ifdef CONFIG_IPV6_SUBTREES + ipv6_addr_equal(&fl.fl6_src, &np->saddr) ? + &np->saddr : +#endif + NULL); + } else { + dst_release(dst); + } + } + + if (err > 0) + err = np->recverr ? net_xmit_errno(err) : 0; + release_sock(sk); +out: + fl6_sock_release(flowlabel); + if (!err) + return len; + /* + * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space. Reporting + * ENOBUFS might not be good (it's not tunable per se), but otherwise + * we don't have a good statistic (IpOutDiscards but it can be too many + * things). We could add another new stat but at least for now that + * seems like overkill. + */ + if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { + UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); + } + return err; + +do_confirm: + dst_confirm(dst); + if (!(msg->msg_flags&MSG_PROBE) || len) + goto back_from_confirm; + err = 0; + goto out; +} + +int udpv6_destroy_sock(struct sock *sk) +{ + lock_sock(sk); + udp_v6_flush_pending_frames(sk); + release_sock(sk); + + inet6_destroy_sock(sk); + + return 0; +} + +/* + * Socket option code for UDP + */ +int udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_v6_push_pending_frames); + return ipv6_setsockopt(sk, level, optname, optval, optlen); +} + +#ifdef CONFIG_COMPAT +int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, + char __user *optval, int optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_setsockopt(sk, level, optname, optval, optlen, + udp_v6_push_pending_frames); + return compat_ipv6_setsockopt(sk, level, optname, optval, optlen); +} +#endif + +int udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_getsockopt(sk, level, optname, optval, optlen); + return ipv6_getsockopt(sk, level, optname, optval, optlen); +} + +#ifdef CONFIG_COMPAT +int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, + char __user *optval, int __user *optlen) +{ + if (level == SOL_UDP || level == SOL_UDPLITE) + return udp_lib_getsockopt(sk, level, optname, optval, optlen); + return compat_ipv6_getsockopt(sk, level, optname, optval, optlen); +} +#endif + +static struct inet6_protocol udpv6_protocol = { + .handler = udpv6_rcv, + .err_handler = udpv6_err, + .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, +}; + +/* ------------------------------------------------------------------------ */ +#ifdef CONFIG_PROC_FS + +static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket) +{ + struct inet_sock *inet = inet_sk(sp); + struct ipv6_pinfo *np = inet6_sk(sp); + struct in6_addr *dest, *src; + __u16 destp, srcp; + + dest = &np->daddr; + src = &np->rcv_saddr; + destp = ntohs(inet->dport); + srcp = ntohs(inet->sport); + seq_printf(seq, + "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " + "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p\n", + bucket, + src->s6_addr32[0], src->s6_addr32[1], + src->s6_addr32[2], src->s6_addr32[3], srcp, + dest->s6_addr32[0], dest->s6_addr32[1], + dest->s6_addr32[2], dest->s6_addr32[3], destp, + sp->sk_state, + atomic_read(&sp->sk_wmem_alloc), + atomic_read(&sp->sk_rmem_alloc), + 0, 0L, 0, + sock_i_uid(sp), 0, + sock_i_ino(sp), + atomic_read(&sp->sk_refcnt), sp); +} + +int udp6_seq_show(struct seq_file *seq, void *v) +{ + if (v == SEQ_START_TOKEN) + seq_printf(seq, + " sl " + "local_address " + "remote_address " + "st tx_queue rx_queue tr tm->when retrnsmt" + " uid timeout inode\n"); + else + udp6_sock_seq_show(seq, v, ((struct udp_iter_state *)seq->private)->bucket); + return 0; +} + +static struct file_operations udp6_seq_fops; +static struct udp_seq_afinfo udp6_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udp6", + .family = AF_INET6, + .hashtable = udp_hash, + .seq_show = udp6_seq_show, + .seq_fops = &udp6_seq_fops, +}; + +int __init udp6_proc_init(void) +{ + return udp_proc_register(&udp6_seq_afinfo); +} + +void udp6_proc_exit(void) { + udp_proc_unregister(&udp6_seq_afinfo); +} +#endif /* CONFIG_PROC_FS */ + +/* ------------------------------------------------------------------------ */ + +DEFINE_PROTO_INUSE(udpv6) + +struct proto udpv6_prot = { + .name = "UDPv6", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip6_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .destroy = udpv6_destroy_sock, + .setsockopt = udpv6_setsockopt, + .getsockopt = udpv6_getsockopt, + .sendmsg = udpv6_sendmsg, + .recvmsg = udpv6_recvmsg, + .backlog_rcv = udpv6_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udp_v6_get_port, + .memory_allocated = &udp_memory_allocated, + .sysctl_mem = sysctl_udp_mem, + .sysctl_wmem = &sysctl_udp_wmem_min, + .sysctl_rmem = &sysctl_udp_rmem_min, + .obj_size = sizeof(struct udp6_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udpv6_setsockopt, + .compat_getsockopt = compat_udpv6_getsockopt, +#endif + REF_PROTO_INUSE(udpv6) +}; + +static struct inet_protosw udpv6_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDP, + .prot = &udpv6_prot, + .ops = &inet6_dgram_ops, + .capability =-1, + .no_check = UDP_CSUM_DEFAULT, + .flags = INET_PROTOSW_PERMANENT, +}; + + +int __init udpv6_init(void) +{ + int ret; + + ret = inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP); + if (ret) + goto out; + + ret = inet6_register_protosw(&udpv6_protosw); + if (ret) + goto out_udpv6_protocol; +out: + return ret; + +out_udpv6_protocol: + inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP); + goto out; +} + +void udpv6_exit(void) +{ + inet6_unregister_protosw(&udpv6_protosw); + inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP); +} diff --git a/net/ipv6/udp_ipv6.c b/net/ipv6/udp_ipv6.c deleted file mode 100644 index 55feac7ba717..000000000000 --- a/net/ipv6/udp_ipv6.c +++ /dev/null @@ -1,1065 +0,0 @@ -/* - * UDP over IPv6 - * Linux INET6 implementation - * - * Authors: - * Pedro Roque - * - * Based on linux/ipv4/udp.c - * - * $Id: udp.c,v 1.65 2002/02/01 22:01:04 davem Exp $ - * - * Fixes: - * Hideaki YOSHIFUJI : sin6_scope_id support - * YOSHIFUJI Hideaki @USAGI and: Support IPV6_V6ONLY socket option, which - * Alexey Kuznetsov allow both IPv4 and IPv6 sockets to bind - * a single port at the same time. - * Kazunori MIYAZAWA @USAGI: change process style to use ip6_append_data - * YOSHIFUJI Hideaki @USAGI: convert /proc/net/udp6 to seq_file. - * - * This program 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 of the License, or (at your option) any later version. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include "udp_impl.h" - -static inline int udp_v6_get_port(struct sock *sk, unsigned short snum) -{ - return udp_get_port(sk, snum, ipv6_rcv_saddr_equal); -} - -static struct sock *__udp6_lib_lookup(struct net *net, - struct in6_addr *saddr, __be16 sport, - struct in6_addr *daddr, __be16 dport, - int dif, struct hlist_head udptable[]) -{ - struct sock *sk, *result = NULL; - struct hlist_node *node; - unsigned short hnum = ntohs(dport); - int badness = -1; - - read_lock(&udp_hash_lock); - sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { - struct inet_sock *inet = inet_sk(sk); - - if (sk->sk_net == net && sk->sk_hash == hnum && - sk->sk_family == PF_INET6) { - struct ipv6_pinfo *np = inet6_sk(sk); - int score = 0; - if (inet->dport) { - if (inet->dport != sport) - continue; - score++; - } - if (!ipv6_addr_any(&np->rcv_saddr)) { - if (!ipv6_addr_equal(&np->rcv_saddr, daddr)) - continue; - score++; - } - if (!ipv6_addr_any(&np->daddr)) { - if (!ipv6_addr_equal(&np->daddr, saddr)) - continue; - score++; - } - if (sk->sk_bound_dev_if) { - if (sk->sk_bound_dev_if != dif) - continue; - score++; - } - if (score == 4) { - result = sk; - break; - } else if (score > badness) { - result = sk; - badness = score; - } - } - } - if (result) - sock_hold(result); - read_unlock(&udp_hash_lock); - return result; -} - -/* - * This should be easy, if there is something there we - * return it, otherwise we block. - */ - -int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, - struct msghdr *msg, size_t len, - int noblock, int flags, int *addr_len) -{ - struct ipv6_pinfo *np = inet6_sk(sk); - struct inet_sock *inet = inet_sk(sk); - struct sk_buff *skb; - unsigned int ulen, copied; - int peeked; - int err; - int is_udplite = IS_UDPLITE(sk); - - if (addr_len) - *addr_len=sizeof(struct sockaddr_in6); - - if (flags & MSG_ERRQUEUE) - return ipv6_recv_error(sk, msg, len); - -try_again: - skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0), - &peeked, &err); - if (!skb) - goto out; - - ulen = skb->len - sizeof(struct udphdr); - copied = len; - if (copied > ulen) - copied = ulen; - else if (copied < ulen) - msg->msg_flags |= MSG_TRUNC; - - /* - * If checksum is needed at all, try to do it while copying the - * data. If the data is truncated, or if we only want a partial - * coverage checksum (UDP-Lite), do it before the copy. - */ - - if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) { - if (udp_lib_checksum_complete(skb)) - goto csum_copy_err; - } - - if (skb_csum_unnecessary(skb)) - err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr), - msg->msg_iov, copied ); - else { - err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov); - if (err == -EINVAL) - goto csum_copy_err; - } - if (err) - goto out_free; - - if (!peeked) - UDP6_INC_STATS_USER(UDP_MIB_INDATAGRAMS, is_udplite); - - sock_recv_timestamp(msg, sk, skb); - - /* Copy the address. */ - if (msg->msg_name) { - struct sockaddr_in6 *sin6; - - sin6 = (struct sockaddr_in6 *) msg->msg_name; - sin6->sin6_family = AF_INET6; - sin6->sin6_port = udp_hdr(skb)->source; - sin6->sin6_flowinfo = 0; - sin6->sin6_scope_id = 0; - - if (skb->protocol == htons(ETH_P_IP)) - ipv6_addr_set(&sin6->sin6_addr, 0, 0, - htonl(0xffff), ip_hdr(skb)->saddr); - else { - ipv6_addr_copy(&sin6->sin6_addr, - &ipv6_hdr(skb)->saddr); - if (ipv6_addr_type(&sin6->sin6_addr) & IPV6_ADDR_LINKLOCAL) - sin6->sin6_scope_id = IP6CB(skb)->iif; - } - - } - if (skb->protocol == htons(ETH_P_IP)) { - if (inet->cmsg_flags) - ip_cmsg_recv(msg, skb); - } else { - if (np->rxopt.all) - datagram_recv_ctl(sk, msg, skb); - } - - err = copied; - if (flags & MSG_TRUNC) - err = ulen; - -out_free: - lock_sock(sk); - skb_free_datagram(sk, skb); - release_sock(sk); -out: - return err; - -csum_copy_err: - lock_sock(sk); - if (!skb_kill_datagram(sk, skb, flags)) - UDP6_INC_STATS_USER(UDP_MIB_INERRORS, is_udplite); - release_sock(sk); - - if (flags & MSG_DONTWAIT) - return -EAGAIN; - goto try_again; -} - -void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, - int type, int code, int offset, __be32 info, - struct hlist_head udptable[] ) -{ - struct ipv6_pinfo *np; - struct ipv6hdr *hdr = (struct ipv6hdr*)skb->data; - struct in6_addr *saddr = &hdr->saddr; - struct in6_addr *daddr = &hdr->daddr; - struct udphdr *uh = (struct udphdr*)(skb->data+offset); - struct sock *sk; - int err; - - sk = __udp6_lib_lookup(skb->dev->nd_net, daddr, uh->dest, - saddr, uh->source, inet6_iif(skb), udptable); - if (sk == NULL) - return; - - np = inet6_sk(sk); - - if (!icmpv6_err_convert(type, code, &err) && !np->recverr) - goto out; - - if (sk->sk_state != TCP_ESTABLISHED && !np->recverr) - goto out; - - if (np->recverr) - ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1)); - - sk->sk_err = err; - sk->sk_error_report(sk); -out: - sock_put(sk); -} - -static __inline__ void udpv6_err(struct sk_buff *skb, - struct inet6_skb_parm *opt, int type, - int code, int offset, __be32 info ) -{ - __udp6_lib_err(skb, opt, type, code, offset, info, udp_hash); -} - -int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) -{ - struct udp_sock *up = udp_sk(sk); - int rc; - int is_udplite = IS_UDPLITE(sk); - - if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb)) - goto drop; - - /* - * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c). - */ - if ((is_udplite & UDPLITE_RECV_CC) && UDP_SKB_CB(skb)->partial_cov) { - - if (up->pcrlen == 0) { /* full coverage was set */ - LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage" - " %d while full coverage %d requested\n", - UDP_SKB_CB(skb)->cscov, skb->len); - goto drop; - } - if (UDP_SKB_CB(skb)->cscov < up->pcrlen) { - LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d " - "too small, need min %d\n", - UDP_SKB_CB(skb)->cscov, up->pcrlen); - goto drop; - } - } - - if (sk->sk_filter) { - if (udp_lib_checksum_complete(skb)) - goto drop; - } - - if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) { - /* Note that an ENOMEM error is charged twice */ - if (rc == -ENOMEM) - UDP6_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, is_udplite); - goto drop; - } - - return 0; -drop: - UDP6_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite); - kfree_skb(skb); - return -1; -} - -static struct sock *udp_v6_mcast_next(struct sock *sk, - __be16 loc_port, struct in6_addr *loc_addr, - __be16 rmt_port, struct in6_addr *rmt_addr, - int dif) -{ - struct hlist_node *node; - struct sock *s = sk; - unsigned short num = ntohs(loc_port); - - sk_for_each_from(s, node) { - struct inet_sock *inet = inet_sk(s); - - if (s->sk_hash == num && s->sk_family == PF_INET6) { - struct ipv6_pinfo *np = inet6_sk(s); - if (inet->dport) { - if (inet->dport != rmt_port) - continue; - } - if (!ipv6_addr_any(&np->daddr) && - !ipv6_addr_equal(&np->daddr, rmt_addr)) - continue; - - if (s->sk_bound_dev_if && s->sk_bound_dev_if != dif) - continue; - - if (!ipv6_addr_any(&np->rcv_saddr)) { - if (!ipv6_addr_equal(&np->rcv_saddr, loc_addr)) - continue; - } - if (!inet6_mc_check(s, loc_addr, rmt_addr)) - continue; - return s; - } - } - return NULL; -} - -/* - * Note: called only from the BH handler context, - * so we don't need to lock the hashes. - */ -static int __udp6_lib_mcast_deliver(struct sk_buff *skb, struct in6_addr *saddr, - struct in6_addr *daddr, struct hlist_head udptable[]) -{ - struct sock *sk, *sk2; - const struct udphdr *uh = udp_hdr(skb); - int dif; - - read_lock(&udp_hash_lock); - sk = sk_head(&udptable[ntohs(uh->dest) & (UDP_HTABLE_SIZE - 1)]); - dif = inet6_iif(skb); - sk = udp_v6_mcast_next(sk, uh->dest, daddr, uh->source, saddr, dif); - if (!sk) { - kfree_skb(skb); - goto out; - } - - sk2 = sk; - while ((sk2 = udp_v6_mcast_next(sk_next(sk2), uh->dest, daddr, - uh->source, saddr, dif))) { - struct sk_buff *buff = skb_clone(skb, GFP_ATOMIC); - if (buff) { - bh_lock_sock_nested(sk2); - if (!sock_owned_by_user(sk2)) - udpv6_queue_rcv_skb(sk2, buff); - else - sk_add_backlog(sk2, buff); - bh_unlock_sock(sk2); - } - } - bh_lock_sock_nested(sk); - if (!sock_owned_by_user(sk)) - udpv6_queue_rcv_skb(sk, skb); - else - sk_add_backlog(sk, skb); - bh_unlock_sock(sk); -out: - read_unlock(&udp_hash_lock); - return 0; -} - -static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, - int proto) -{ - int err; - - UDP_SKB_CB(skb)->partial_cov = 0; - UDP_SKB_CB(skb)->cscov = skb->len; - - if (IS_PROTO_UDPLITE(proto)) { - err = udplite_checksum_init(skb, uh); - if (err) - return err; - } - - if (uh->check == 0) { - /* RFC 2460 section 8.1 says that we SHOULD log - this error. Well, it is reasonable. - */ - LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n"); - return 1; - } - if (skb->ip_summed == CHECKSUM_COMPLETE && - !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, - skb->len, proto, skb->csum)) - skb->ip_summed = CHECKSUM_UNNECESSARY; - - if (!skb_csum_unnecessary(skb)) - skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr, - &ipv6_hdr(skb)->daddr, - skb->len, proto, 0)); - - return 0; -} - -int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], - int proto) -{ - struct sock *sk; - struct udphdr *uh; - struct net_device *dev = skb->dev; - struct in6_addr *saddr, *daddr; - u32 ulen = 0; - - if (!pskb_may_pull(skb, sizeof(struct udphdr))) - goto short_packet; - - saddr = &ipv6_hdr(skb)->saddr; - daddr = &ipv6_hdr(skb)->daddr; - uh = udp_hdr(skb); - - ulen = ntohs(uh->len); - if (ulen > skb->len) - goto short_packet; - - if (proto == IPPROTO_UDP) { - /* UDP validates ulen. */ - - /* Check for jumbo payload */ - if (ulen == 0) - ulen = skb->len; - - if (ulen < sizeof(*uh)) - goto short_packet; - - if (ulen < skb->len) { - if (pskb_trim_rcsum(skb, ulen)) - goto short_packet; - saddr = &ipv6_hdr(skb)->saddr; - daddr = &ipv6_hdr(skb)->daddr; - uh = udp_hdr(skb); - } - } - - if (udp6_csum_init(skb, uh, proto)) - goto discard; - - /* - * Multicast receive code - */ - if (ipv6_addr_is_multicast(daddr)) - return __udp6_lib_mcast_deliver(skb, saddr, daddr, udptable); - - /* Unicast */ - - /* - * check socket cache ... must talk to Alan about his plans - * for sock caches... i'll skip this for now. - */ - sk = __udp6_lib_lookup(skb->dev->nd_net, saddr, uh->source, - daddr, uh->dest, inet6_iif(skb), udptable); - - if (sk == NULL) { - if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb)) - goto discard; - - if (udp_lib_checksum_complete(skb)) - goto discard; - UDP6_INC_STATS_BH(UDP_MIB_NOPORTS, IS_PROTO_UDPLITE(proto)); - - icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0, dev); - - kfree_skb(skb); - return 0; - } - - /* deliver */ - - bh_lock_sock_nested(sk); - if (!sock_owned_by_user(sk)) - udpv6_queue_rcv_skb(sk, skb); - else - sk_add_backlog(sk, skb); - bh_unlock_sock(sk); - sock_put(sk); - return 0; - -short_packet: - LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: %d/%u\n", - IS_PROTO_UDPLITE(proto) ? "-Lite" : "", - ulen, skb->len); - -discard: - UDP6_INC_STATS_BH(UDP_MIB_INERRORS, IS_PROTO_UDPLITE(proto)); - kfree_skb(skb); - return 0; -} - -static __inline__ int udpv6_rcv(struct sk_buff *skb) -{ - return __udp6_lib_rcv(skb, udp_hash, IPPROTO_UDP); -} - -/* - * Throw away all pending data and cancel the corking. Socket is locked. - */ -static void udp_v6_flush_pending_frames(struct sock *sk) -{ - struct udp_sock *up = udp_sk(sk); - - if (up->pending) { - up->len = 0; - up->pending = 0; - ip6_flush_pending_frames(sk); - } -} - -/* - * Sending - */ - -static int udp_v6_push_pending_frames(struct sock *sk) -{ - struct sk_buff *skb; - struct udphdr *uh; - struct udp_sock *up = udp_sk(sk); - struct inet_sock *inet = inet_sk(sk); - struct flowi *fl = &inet->cork.fl; - int err = 0; - int is_udplite = IS_UDPLITE(sk); - __wsum csum = 0; - - /* Grab the skbuff where UDP header space exists. */ - if ((skb = skb_peek(&sk->sk_write_queue)) == NULL) - goto out; - - /* - * Create a UDP header - */ - uh = udp_hdr(skb); - uh->source = fl->fl_ip_sport; - uh->dest = fl->fl_ip_dport; - uh->len = htons(up->len); - uh->check = 0; - - if (is_udplite) - csum = udplite_csum_outgoing(sk, skb); - else - csum = udp_csum_outgoing(sk, skb); - - /* add protocol-dependent pseudo-header */ - uh->check = csum_ipv6_magic(&fl->fl6_src, &fl->fl6_dst, - up->len, fl->proto, csum ); - if (uh->check == 0) - uh->check = CSUM_MANGLED_0; - - err = ip6_push_pending_frames(sk); -out: - up->len = 0; - up->pending = 0; - if (!err) - UDP6_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite); - return err; -} - -int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk, - struct msghdr *msg, size_t len) -{ - struct ipv6_txoptions opt_space; - struct udp_sock *up = udp_sk(sk); - struct inet_sock *inet = inet_sk(sk); - struct ipv6_pinfo *np = inet6_sk(sk); - struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) msg->msg_name; - struct in6_addr *daddr, *final_p = NULL, final; - struct ipv6_txoptions *opt = NULL; - struct ip6_flowlabel *flowlabel = NULL; - struct flowi fl; - struct dst_entry *dst; - int addr_len = msg->msg_namelen; - int ulen = len; - int hlimit = -1; - int tclass = -1; - int corkreq = up->corkflag || msg->msg_flags&MSG_MORE; - int err; - int connected = 0; - int is_udplite = IS_UDPLITE(sk); - int (*getfrag)(void *, char *, int, int, int, struct sk_buff *); - - /* destination address check */ - if (sin6) { - if (addr_len < offsetof(struct sockaddr, sa_data)) - return -EINVAL; - - switch (sin6->sin6_family) { - case AF_INET6: - if (addr_len < SIN6_LEN_RFC2133) - return -EINVAL; - daddr = &sin6->sin6_addr; - break; - case AF_INET: - goto do_udp_sendmsg; - case AF_UNSPEC: - msg->msg_name = sin6 = NULL; - msg->msg_namelen = addr_len = 0; - daddr = NULL; - break; - default: - return -EINVAL; - } - } else if (!up->pending) { - if (sk->sk_state != TCP_ESTABLISHED) - return -EDESTADDRREQ; - daddr = &np->daddr; - } else - daddr = NULL; - - if (daddr) { - if (ipv6_addr_v4mapped(daddr)) { - struct sockaddr_in sin; - sin.sin_family = AF_INET; - sin.sin_port = sin6 ? sin6->sin6_port : inet->dport; - sin.sin_addr.s_addr = daddr->s6_addr32[3]; - msg->msg_name = &sin; - msg->msg_namelen = sizeof(sin); -do_udp_sendmsg: - if (__ipv6_only_sock(sk)) - return -ENETUNREACH; - return udp_sendmsg(iocb, sk, msg, len); - } - } - - if (up->pending == AF_INET) - return udp_sendmsg(iocb, sk, msg, len); - - /* Rough check on arithmetic overflow, - better check is made in ip6_append_data(). - */ - if (len > INT_MAX - sizeof(struct udphdr)) - return -EMSGSIZE; - - if (up->pending) { - /* - * There are pending frames. - * The socket lock must be held while it's corked. - */ - lock_sock(sk); - if (likely(up->pending)) { - if (unlikely(up->pending != AF_INET6)) { - release_sock(sk); - return -EAFNOSUPPORT; - } - dst = NULL; - goto do_append_data; - } - release_sock(sk); - } - ulen += sizeof(struct udphdr); - - memset(&fl, 0, sizeof(fl)); - - if (sin6) { - if (sin6->sin6_port == 0) - return -EINVAL; - - fl.fl_ip_dport = sin6->sin6_port; - daddr = &sin6->sin6_addr; - - if (np->sndflow) { - fl.fl6_flowlabel = sin6->sin6_flowinfo&IPV6_FLOWINFO_MASK; - if (fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) { - flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel); - if (flowlabel == NULL) - return -EINVAL; - daddr = &flowlabel->dst; - } - } - - /* - * Otherwise it will be difficult to maintain - * sk->sk_dst_cache. - */ - if (sk->sk_state == TCP_ESTABLISHED && - ipv6_addr_equal(daddr, &np->daddr)) - daddr = &np->daddr; - - if (addr_len >= sizeof(struct sockaddr_in6) && - sin6->sin6_scope_id && - ipv6_addr_type(daddr)&IPV6_ADDR_LINKLOCAL) - fl.oif = sin6->sin6_scope_id; - } else { - if (sk->sk_state != TCP_ESTABLISHED) - return -EDESTADDRREQ; - - fl.fl_ip_dport = inet->dport; - daddr = &np->daddr; - fl.fl6_flowlabel = np->flow_label; - connected = 1; - } - - if (!fl.oif) - fl.oif = sk->sk_bound_dev_if; - - if (msg->msg_controllen) { - opt = &opt_space; - memset(opt, 0, sizeof(struct ipv6_txoptions)); - opt->tot_len = sizeof(*opt); - - err = datagram_send_ctl(msg, &fl, opt, &hlimit, &tclass); - if (err < 0) { - fl6_sock_release(flowlabel); - return err; - } - if ((fl.fl6_flowlabel&IPV6_FLOWLABEL_MASK) && !flowlabel) { - flowlabel = fl6_sock_lookup(sk, fl.fl6_flowlabel); - if (flowlabel == NULL) - return -EINVAL; - } - if (!(opt->opt_nflen|opt->opt_flen)) - opt = NULL; - connected = 0; - } - if (opt == NULL) - opt = np->opt; - if (flowlabel) - opt = fl6_merge_options(&opt_space, flowlabel, opt); - opt = ipv6_fixup_options(&opt_space, opt); - - fl.proto = sk->sk_protocol; - ipv6_addr_copy(&fl.fl6_dst, daddr); - if (ipv6_addr_any(&fl.fl6_src) && !ipv6_addr_any(&np->saddr)) - ipv6_addr_copy(&fl.fl6_src, &np->saddr); - fl.fl_ip_sport = inet->sport; - - /* merge ip6_build_xmit from ip6_output */ - if (opt && opt->srcrt) { - struct rt0_hdr *rt0 = (struct rt0_hdr *) opt->srcrt; - ipv6_addr_copy(&final, &fl.fl6_dst); - ipv6_addr_copy(&fl.fl6_dst, rt0->addr); - final_p = &final; - connected = 0; - } - - if (!fl.oif && ipv6_addr_is_multicast(&fl.fl6_dst)) { - fl.oif = np->mcast_oif; - connected = 0; - } - - security_sk_classify_flow(sk, &fl); - - err = ip6_sk_dst_lookup(sk, &dst, &fl); - if (err) - goto out; - if (final_p) - ipv6_addr_copy(&fl.fl6_dst, final_p); - - if ((err = __xfrm_lookup(&dst, &fl, sk, XFRM_LOOKUP_WAIT)) < 0) { - if (err == -EREMOTE) - err = ip6_dst_blackhole(sk, &dst, &fl); - if (err < 0) - goto out; - } - - if (hlimit < 0) { - if (ipv6_addr_is_multicast(&fl.fl6_dst)) - hlimit = np->mcast_hops; - else - hlimit = np->hop_limit; - if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); - } - - if (tclass < 0) { - tclass = np->tclass; - if (tclass < 0) - tclass = 0; - } - - if (msg->msg_flags&MSG_CONFIRM) - goto do_confirm; -back_from_confirm: - - lock_sock(sk); - if (unlikely(up->pending)) { - /* The socket is already corked while preparing it. */ - /* ... which is an evident application bug. --ANK */ - release_sock(sk); - - LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n"); - err = -EINVAL; - goto out; - } - - up->pending = AF_INET6; - -do_append_data: - up->len += ulen; - getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag; - err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen, - sizeof(struct udphdr), hlimit, tclass, opt, &fl, - (struct rt6_info*)dst, - corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags); - if (err) - udp_v6_flush_pending_frames(sk); - else if (!corkreq) - err = udp_v6_push_pending_frames(sk); - else if (unlikely(skb_queue_empty(&sk->sk_write_queue))) - up->pending = 0; - - if (dst) { - if (connected) { - ip6_dst_store(sk, dst, - ipv6_addr_equal(&fl.fl6_dst, &np->daddr) ? - &np->daddr : NULL, -#ifdef CONFIG_IPV6_SUBTREES - ipv6_addr_equal(&fl.fl6_src, &np->saddr) ? - &np->saddr : -#endif - NULL); - } else { - dst_release(dst); - } - } - - if (err > 0) - err = np->recverr ? net_xmit_errno(err) : 0; - release_sock(sk); -out: - fl6_sock_release(flowlabel); - if (!err) - return len; - /* - * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space. Reporting - * ENOBUFS might not be good (it's not tunable per se), but otherwise - * we don't have a good statistic (IpOutDiscards but it can be too many - * things). We could add another new stat but at least for now that - * seems like overkill. - */ - if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) { - UDP6_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite); - } - return err; - -do_confirm: - dst_confirm(dst); - if (!(msg->msg_flags&MSG_PROBE) || len) - goto back_from_confirm; - err = 0; - goto out; -} - -int udpv6_destroy_sock(struct sock *sk) -{ - lock_sock(sk); - udp_v6_flush_pending_frames(sk); - release_sock(sk); - - inet6_destroy_sock(sk); - - return 0; -} - -/* - * Socket option code for UDP - */ -int udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_setsockopt(sk, level, optname, optval, optlen, - udp_v6_push_pending_frames); - return ipv6_setsockopt(sk, level, optname, optval, optlen); -} - -#ifdef CONFIG_COMPAT -int compat_udpv6_setsockopt(struct sock *sk, int level, int optname, - char __user *optval, int optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_setsockopt(sk, level, optname, optval, optlen, - udp_v6_push_pending_frames); - return compat_ipv6_setsockopt(sk, level, optname, optval, optlen); -} -#endif - -int udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_getsockopt(sk, level, optname, optval, optlen); - return ipv6_getsockopt(sk, level, optname, optval, optlen); -} - -#ifdef CONFIG_COMPAT -int compat_udpv6_getsockopt(struct sock *sk, int level, int optname, - char __user *optval, int __user *optlen) -{ - if (IS_SOL_UDPFAMILY(level)) - return udp_lib_getsockopt(sk, level, optname, optval, optlen); - return compat_ipv6_getsockopt(sk, level, optname, optval, optlen); -} -#endif - -static struct inet6_protocol udpv6_protocol = { - .handler = udpv6_rcv, - .err_handler = udpv6_err, - .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, -}; - -/* ------------------------------------------------------------------------ */ -#ifdef CONFIG_PROC_FS - -static void udp6_sock_seq_show(struct seq_file *seq, struct sock *sp, int bucket) -{ - struct inet_sock *inet = inet_sk(sp); - struct ipv6_pinfo *np = inet6_sk(sp); - struct in6_addr *dest, *src; - __u16 destp, srcp; - - dest = &np->daddr; - src = &np->rcv_saddr; - destp = ntohs(inet->dport); - srcp = ntohs(inet->sport); - seq_printf(seq, - "%4d: %08X%08X%08X%08X:%04X %08X%08X%08X%08X:%04X " - "%02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p\n", - bucket, - src->s6_addr32[0], src->s6_addr32[1], - src->s6_addr32[2], src->s6_addr32[3], srcp, - dest->s6_addr32[0], dest->s6_addr32[1], - dest->s6_addr32[2], dest->s6_addr32[3], destp, - sp->sk_state, - atomic_read(&sp->sk_wmem_alloc), - atomic_read(&sp->sk_rmem_alloc), - 0, 0L, 0, - sock_i_uid(sp), 0, - sock_i_ino(sp), - atomic_read(&sp->sk_refcnt), sp); -} - -int udp6_seq_show(struct seq_file *seq, void *v) -{ - if (v == SEQ_START_TOKEN) - seq_printf(seq, - " sl " - "local_address " - "remote_address " - "st tx_queue rx_queue tr tm->when retrnsmt" - " uid timeout inode\n"); - else - udp6_sock_seq_show(seq, v, ((struct udp_iter_state *)seq->private)->bucket); - return 0; -} - -static struct file_operations udp6_seq_fops; -static struct udp_seq_afinfo udp6_seq_afinfo = { - .owner = THIS_MODULE, - .name = "udp6", - .family = AF_INET6, - .hashtable = udp_hash, - .seq_show = udp6_seq_show, - .seq_fops = &udp6_seq_fops, -}; - -int __init udp6_proc_init(void) -{ - return udp_proc_register(&udp6_seq_afinfo); -} - -void udp6_proc_exit(void) { - udp_proc_unregister(&udp6_seq_afinfo); -} -#endif /* CONFIG_PROC_FS */ - -/* ------------------------------------------------------------------------ */ - -DEFINE_PROTO_INUSE(udpv6) - -struct proto udpv6_prot = { - .name = "UDPv6", - .owner = THIS_MODULE, - .close = udp_lib_close, - .connect = ip6_datagram_connect, - .disconnect = udp_disconnect, - .ioctl = udp_ioctl, - .destroy = udpv6_destroy_sock, - .setsockopt = udpv6_setsockopt, - .getsockopt = udpv6_getsockopt, - .sendmsg = udpv6_sendmsg, - .recvmsg = udpv6_recvmsg, - .backlog_rcv = udpv6_queue_rcv_skb, - .hash = udp_lib_hash, - .unhash = udp_lib_unhash, - .get_port = udp_v6_get_port, - .memory_allocated = &udp_memory_allocated, - .sysctl_mem = sysctl_udp_mem, - .sysctl_wmem = &sysctl_udp_wmem_min, - .sysctl_rmem = &sysctl_udp_rmem_min, - .obj_size = sizeof(struct udp6_sock), -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_udpv6_setsockopt, - .compat_getsockopt = compat_udpv6_getsockopt, -#endif - REF_PROTO_INUSE(udpv6) -}; - -static struct inet_protosw udpv6_protosw = { - .type = SOCK_DGRAM, - .protocol = IPPROTO_UDP, - .prot = &udpv6_prot, - .ops = &inet6_dgram_ops, - .capability =-1, - .no_check = UDP_CSUM_DEFAULT, - .flags = INET_PROTOSW_PERMANENT, -}; - - -int __init udpv6_init(void) -{ - int ret; - - ret = inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP); - if (ret) - goto out; - - ret = inet6_register_protosw(&udpv6_protosw); - if (ret) - goto out_udpv6_protocol; -out: - return ret; - -out_udpv6_protocol: - inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP); - goto out; -} - -void udpv6_exit(void) -{ - inet6_unregister_protosw(&udpv6_protosw); - inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP); -} diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c new file mode 100644 index 000000000000..87d4202522ee --- /dev/null +++ b/net/ipv6/udplite.c @@ -0,0 +1,125 @@ +/* + * UDPLITEv6 An implementation of the UDP-Lite protocol over IPv6. + * See also net/ipv4/udplite.c + * + * Version: $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $ + * + * Authors: Gerrit Renker + * + * Changes: + * Fixes: + * This program 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 of the License, or (at your option) any later version. + */ +#include "udp_impl.h" + +DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6) __read_mostly; + +static int udplitev6_rcv(struct sk_buff *skb) +{ + return __udp6_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE); +} + +static void udplitev6_err(struct sk_buff *skb, + struct inet6_skb_parm *opt, + int type, int code, int offset, __be32 info) +{ + __udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash); +} + +static struct inet6_protocol udplitev6_protocol = { + .handler = udplitev6_rcv, + .err_handler = udplitev6_err, + .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, +}; + +static int udplite_v6_get_port(struct sock *sk, unsigned short snum) +{ + return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal); +} + +DEFINE_PROTO_INUSE(udplitev6) + +struct proto udplitev6_prot = { + .name = "UDPLITEv6", + .owner = THIS_MODULE, + .close = udp_lib_close, + .connect = ip6_datagram_connect, + .disconnect = udp_disconnect, + .ioctl = udp_ioctl, + .init = udplite_sk_init, + .destroy = udpv6_destroy_sock, + .setsockopt = udpv6_setsockopt, + .getsockopt = udpv6_getsockopt, + .sendmsg = udpv6_sendmsg, + .recvmsg = udpv6_recvmsg, + .backlog_rcv = udpv6_queue_rcv_skb, + .hash = udp_lib_hash, + .unhash = udp_lib_unhash, + .get_port = udplite_v6_get_port, + .obj_size = sizeof(struct udp6_sock), +#ifdef CONFIG_COMPAT + .compat_setsockopt = compat_udpv6_setsockopt, + .compat_getsockopt = compat_udpv6_getsockopt, +#endif + REF_PROTO_INUSE(udplitev6) +}; + +static struct inet_protosw udplite6_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_UDPLITE, + .prot = &udplitev6_prot, + .ops = &inet6_dgram_ops, + .capability = -1, + .no_check = 0, + .flags = INET_PROTOSW_PERMANENT, +}; + +int __init udplitev6_init(void) +{ + int ret; + + ret = inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE); + if (ret) + goto out; + + ret = inet6_register_protosw(&udplite6_protosw); + if (ret) + goto out_udplitev6_protocol; +out: + return ret; + +out_udplitev6_protocol: + inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE); + goto out; +} + +void udplitev6_exit(void) +{ + inet6_unregister_protosw(&udplite6_protosw); + inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE); +} + +#ifdef CONFIG_PROC_FS +static struct file_operations udplite6_seq_fops; +static struct udp_seq_afinfo udplite6_seq_afinfo = { + .owner = THIS_MODULE, + .name = "udplite6", + .family = AF_INET6, + .hashtable = udplite_hash, + .seq_show = udp6_seq_show, + .seq_fops = &udplite6_seq_fops, +}; + +int __init udplite6_proc_init(void) +{ + return udp_proc_register(&udplite6_seq_afinfo); +} + +void udplite6_proc_exit(void) +{ + udp_proc_unregister(&udplite6_seq_afinfo); +} +#endif diff --git a/net/ipv6/udplite_ipv6.c b/net/ipv6/udplite_ipv6.c deleted file mode 100644 index 87d4202522ee..000000000000 --- a/net/ipv6/udplite_ipv6.c +++ /dev/null @@ -1,125 +0,0 @@ -/* - * UDPLITEv6 An implementation of the UDP-Lite protocol over IPv6. - * See also net/ipv4/udplite.c - * - * Version: $Id: udplite.c,v 1.9 2006/10/19 08:28:10 gerrit Exp $ - * - * Authors: Gerrit Renker - * - * Changes: - * Fixes: - * This program 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 of the License, or (at your option) any later version. - */ -#include "udp_impl.h" - -DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6) __read_mostly; - -static int udplitev6_rcv(struct sk_buff *skb) -{ - return __udp6_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE); -} - -static void udplitev6_err(struct sk_buff *skb, - struct inet6_skb_parm *opt, - int type, int code, int offset, __be32 info) -{ - __udp6_lib_err(skb, opt, type, code, offset, info, udplite_hash); -} - -static struct inet6_protocol udplitev6_protocol = { - .handler = udplitev6_rcv, - .err_handler = udplitev6_err, - .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, -}; - -static int udplite_v6_get_port(struct sock *sk, unsigned short snum) -{ - return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal); -} - -DEFINE_PROTO_INUSE(udplitev6) - -struct proto udplitev6_prot = { - .name = "UDPLITEv6", - .owner = THIS_MODULE, - .close = udp_lib_close, - .connect = ip6_datagram_connect, - .disconnect = udp_disconnect, - .ioctl = udp_ioctl, - .init = udplite_sk_init, - .destroy = udpv6_destroy_sock, - .setsockopt = udpv6_setsockopt, - .getsockopt = udpv6_getsockopt, - .sendmsg = udpv6_sendmsg, - .recvmsg = udpv6_recvmsg, - .backlog_rcv = udpv6_queue_rcv_skb, - .hash = udp_lib_hash, - .unhash = udp_lib_unhash, - .get_port = udplite_v6_get_port, - .obj_size = sizeof(struct udp6_sock), -#ifdef CONFIG_COMPAT - .compat_setsockopt = compat_udpv6_setsockopt, - .compat_getsockopt = compat_udpv6_getsockopt, -#endif - REF_PROTO_INUSE(udplitev6) -}; - -static struct inet_protosw udplite6_protosw = { - .type = SOCK_DGRAM, - .protocol = IPPROTO_UDPLITE, - .prot = &udplitev6_prot, - .ops = &inet6_dgram_ops, - .capability = -1, - .no_check = 0, - .flags = INET_PROTOSW_PERMANENT, -}; - -int __init udplitev6_init(void) -{ - int ret; - - ret = inet6_add_protocol(&udplitev6_protocol, IPPROTO_UDPLITE); - if (ret) - goto out; - - ret = inet6_register_protosw(&udplite6_protosw); - if (ret) - goto out_udplitev6_protocol; -out: - return ret; - -out_udplitev6_protocol: - inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE); - goto out; -} - -void udplitev6_exit(void) -{ - inet6_unregister_protosw(&udplite6_protosw); - inet6_del_protocol(&udplitev6_protocol, IPPROTO_UDPLITE); -} - -#ifdef CONFIG_PROC_FS -static struct file_operations udplite6_seq_fops; -static struct udp_seq_afinfo udplite6_seq_afinfo = { - .owner = THIS_MODULE, - .name = "udplite6", - .family = AF_INET6, - .hashtable = udplite_hash, - .seq_show = udp6_seq_show, - .seq_fops = &udplite6_seq_fops, -}; - -int __init udplite6_proc_init(void) -{ - return udp_proc_register(&udplite6_seq_afinfo); -} - -void udplite6_proc_exit(void) -{ - udp_proc_unregister(&udplite6_seq_afinfo); -} -#endif -- cgit v1.2.3 From 1762f7e88eb34f653b4a915be99a102e347dd45e Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 7 Mar 2008 11:15:34 -0800 Subject: [NETNS][IPV6] ndisc - make socket control per namespace Make ndisc socket control per namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 1 + net/ipv6/ndisc.c | 77 +++++++++++++++++++++++++++++++++--------------- 2 files changed, 54 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 90e6e24df858..b773256d0d1d 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -51,5 +51,6 @@ struct netns_ipv6 { struct fib_rules_ops *fib6_rules_ops; #endif struct sock **icmp_sk; + struct sock *ndisc_sk; }; #endif diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index a539b9ea53fd..24e76ed98884 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -89,8 +89,6 @@ #include #include -static struct socket *ndisc_socket; - static u32 ndisc_hash(const void *pkey, const struct net_device *dev); static int ndisc_constructor(struct neighbour *neigh); static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb); @@ -449,7 +447,8 @@ static void __ndisc_send(struct net_device *dev, { struct flowi fl; struct dst_entry *dst; - struct sock *sk = ndisc_socket->sk; + struct net *net = dev->nd_net; + struct sock *sk = net->ipv6.ndisc_sk; struct sk_buff *skb; struct icmp6hdr *hdr; struct inet6_dev *idev; @@ -459,8 +458,7 @@ static void __ndisc_send(struct net_device *dev, type = icmp6h->icmp6_type; - icmpv6_flow_init(ndisc_socket->sk, &fl, type, - saddr, daddr, dev->ifindex); + icmpv6_flow_init(sk, &fl, type, saddr, daddr, dev->ifindex); dst = icmp6_dst_alloc(dev, neigh, daddr); if (!dst) @@ -1394,13 +1392,14 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, struct in6_addr *target) { - struct sock *sk = ndisc_socket->sk; + struct net_device *dev = skb->dev; + struct net *net = dev->nd_net; + struct sock *sk = net->ipv6.ndisc_sk; int len = sizeof(struct icmp6hdr) + 2 * sizeof(struct in6_addr); struct sk_buff *buff; struct icmp6hdr *icmph; struct in6_addr saddr_buf; struct in6_addr *addrp; - struct net_device *dev; struct rt6_info *rt; struct dst_entry *dst; struct inet6_dev *idev; @@ -1411,8 +1410,6 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, int hlen; u8 ha_buf[MAX_ADDR_LEN], *ha = NULL; - dev = skb->dev; - if (ipv6_get_lladdr(dev, &saddr_buf, IFA_F_TENTATIVE)) { ND_PRINTK2(KERN_WARNING "ICMPv6 Redirect: no link-local address on %s\n", @@ -1427,10 +1424,10 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, return; } - icmpv6_flow_init(ndisc_socket->sk, &fl, NDISC_REDIRECT, + icmpv6_flow_init(sk, &fl, NDISC_REDIRECT, &saddr_buf, &ipv6_hdr(skb)->saddr, dev->ifindex); - dst = ip6_route_output(&init_net, NULL, &fl); + dst = ip6_route_output(net, NULL, &fl); if (dst == NULL) return; @@ -1719,22 +1716,24 @@ static int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name, #endif -int __init ndisc_init(void) +static int ndisc_net_init(struct net *net) { + struct socket *sock; struct ipv6_pinfo *np; struct sock *sk; int err; - err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &ndisc_socket); + err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &sock); if (err < 0) { ND_PRINTK0(KERN_ERR "ICMPv6 NDISC: Failed to initialize the control socket (err %d).\n", err); - ndisc_socket = NULL; /* For safety. */ return err; } - sk = ndisc_socket->sk; + net->ipv6.ndisc_sk = sk = sock->sk; + sk_change_net(sk, net); + np = inet6_sk(sk); sk->sk_allocation = GFP_ATOMIC; np->hop_limit = 255; @@ -1742,21 +1741,52 @@ int __init ndisc_init(void) np->mc_loop = 0; sk->sk_prot->unhash(sk); + return 0; +} + +static void ndisc_net_exit(struct net *net) +{ + sk_release_kernel(net->ipv6.ndisc_sk); +} + +static struct pernet_operations ndisc_net_ops = { + .init = ndisc_net_init, + .exit = ndisc_net_exit, +}; + +int __init ndisc_init(void) +{ + int err; + + err = register_pernet_subsys(&ndisc_net_ops); + if (err) + return err; /* * Initialize the neighbour table */ - neigh_table_init(&nd_tbl); #ifdef CONFIG_SYSCTL - neigh_sysctl_register(NULL, &nd_tbl.parms, NET_IPV6, NET_IPV6_NEIGH, - "ipv6", - &ndisc_ifinfo_sysctl_change, - &ndisc_ifinfo_sysctl_strategy); + err = neigh_sysctl_register(NULL, &nd_tbl.parms, NET_IPV6, + NET_IPV6_NEIGH, "ipv6", + &ndisc_ifinfo_sysctl_change, + &ndisc_ifinfo_sysctl_strategy); + if (err) + goto out_unregister_pernet; #endif + err = register_netdevice_notifier(&ndisc_netdev_notifier); + if (err) + goto out_unregister_sysctl; +out: + return err; - register_netdevice_notifier(&ndisc_netdev_notifier); - return 0; +out_unregister_sysctl: +#ifdef CONFIG_SYSCTL + neigh_sysctl_unregister(&nd_tbl.parms); +out_unregister_pernet: +#endif + unregister_pernet_subsys(&ndisc_net_ops); + goto out; } void ndisc_cleanup(void) @@ -1766,6 +1796,5 @@ void ndisc_cleanup(void) neigh_sysctl_unregister(&nd_tbl.parms); #endif neigh_table_clear(&nd_tbl); - sock_release(ndisc_socket); - ndisc_socket = NULL; /* For safety. */ + unregister_pernet_subsys(&ndisc_net_ops); } -- cgit v1.2.3 From 93ec926b075d14bb7edcbc054e92199e9ad4ad1d Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 7 Mar 2008 11:16:02 -0800 Subject: [NETNS][IPV6] tcp6 - make socket control per namespace Instead of having a tcp6_socket global to all the namespace, there is tcp6 socket control per namespace. That is consistent with which namespace sent a RST and allows to pass the socket to the underlying function to retrieve the network namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 1 + net/ipv6/tcp_ipv6.c | 43 +++++++++++++++++++++++++++++++++---------- 2 files changed, 34 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index b773256d0d1d..ddb9ccddcc4d 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -52,5 +52,6 @@ struct netns_ipv6 { #endif struct sock **icmp_sk; struct sock *ndisc_sk; + struct sock *tcp_sk; }; #endif diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index b47cce8fea44..7954b4e99c4f 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -69,9 +69,6 @@ #include #include -/* Socket used for sending RSTs and ACKs */ -static struct socket *tcp6_socket; - static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb); static void tcp_v6_reqsk_send_ack(struct sk_buff *skb, struct request_sock *req); static void tcp_v6_send_check(struct sock *sk, int len, @@ -1075,10 +1072,11 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb) * Underlying function will use this to retrieve the network * namespace */ - if (!ip6_dst_lookup(tcp6_socket->sk, &buff->dst, &fl)) { + if (!ip6_dst_lookup(init_net.ipv6.tcp_sk, &buff->dst, &fl)) { if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) { - ip6_xmit(tcp6_socket->sk, buff, &fl, NULL, 0); + ip6_xmit(init_net.ipv6.tcp_sk, + buff, &fl, NULL, 0); TCP_INC_STATS_BH(TCP_MIB_OUTSEGS); TCP_INC_STATS_BH(TCP_MIB_OUTRSTS); return; @@ -1175,9 +1173,10 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw, fl.fl_ip_sport = t1->source; security_skb_classify_flow(skb, &fl); - if (!ip6_dst_lookup(tcp6_socket->sk, &buff->dst, &fl)) { + if (!ip6_dst_lookup(init_net.ipv6.tcp_sk, &buff->dst, &fl)) { if (xfrm_lookup(&buff->dst, &fl, NULL, 0) >= 0) { - ip6_xmit(tcp6_socket->sk, buff, &fl, NULL, 0); + ip6_xmit(init_net.ipv6.tcp_sk, + buff, &fl, NULL, 0); TCP_INC_STATS_BH(TCP_MIB_OUTSEGS); return; } @@ -2198,6 +2197,31 @@ static struct inet_protosw tcpv6_protosw = { INET_PROTOSW_ICSK, }; +static int tcpv6_net_init(struct net *net) +{ + int err; + struct socket *sock; + struct sock *sk; + + err = inet_csk_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP); + if (err) + return err; + + net->ipv6.tcp_sk = sk = sock->sk; + sk_change_net(sk, net); + return err; +} + +static void tcpv6_net_exit(struct net *net) +{ + sk_release_kernel(net->ipv6.tcp_sk); +} + +static struct pernet_operations tcpv6_net_ops = { + .init = tcpv6_net_init, + .exit = tcpv6_net_exit, +}; + int __init tcpv6_init(void) { int ret; @@ -2211,8 +2235,7 @@ int __init tcpv6_init(void) if (ret) goto out_tcpv6_protocol; - ret = inet_csk_ctl_sock_create(&tcp6_socket, PF_INET6, - SOCK_RAW, IPPROTO_TCP); + ret = register_pernet_subsys(&tcpv6_net_ops); if (ret) goto out_tcpv6_protosw; out: @@ -2227,7 +2250,7 @@ out_tcpv6_protosw: void tcpv6_exit(void) { - sock_release(tcp6_socket); + unregister_pernet_subsys(&tcpv6_net_ops); inet6_unregister_protosw(&tcpv6_protosw); inet6_del_protocol(&tcpv6_protocol, IPPROTO_TCP); } -- cgit v1.2.3 From b8ad0cbc58f703972e9e37c4e2a8081dd7e6a551 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 7 Mar 2008 11:16:55 -0800 Subject: [NETNS][IPV6] mcast - handle several network namespace This patch make use of the network namespace information at the right places to handle the multicast for several network namespaces. It makes the socket control to be per namespace too. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/netns/ipv6.h | 1 + net/ipv6/mcast.c | 116 +++++++++++++++++++++++++++++++---------------- net/ipv6/udp.c | 3 ++ 3 files changed, 82 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index ddb9ccddcc4d..ac053be6c256 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -53,5 +53,6 @@ struct netns_ipv6 { struct sock **icmp_sk; struct sock *ndisc_sk; struct sock *tcp_sk; + struct sock *igmp_sk; }; #endif diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 197ca390a15d..f2879056fab0 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -126,8 +126,6 @@ static struct in6_addr mld2_all_mcr = MLD2_ALL_MCR_INIT; /* Big mc list lock for all the sockets */ static DEFINE_RWLOCK(ipv6_sk_mc_lock); -static struct socket *igmp6_socket; - int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr); static void igmp6_join_group(struct ifmcaddr6 *ma); @@ -183,6 +181,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr) struct net_device *dev = NULL; struct ipv6_mc_socklist *mc_lst; struct ipv6_pinfo *np = inet6_sk(sk); + struct net *net = sk->sk_net; int err; if (!ipv6_addr_is_multicast(addr)) @@ -208,14 +207,14 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr) if (ifindex == 0) { struct rt6_info *rt; - rt = rt6_lookup(&init_net, addr, NULL, 0, 0); + rt = rt6_lookup(net, addr, NULL, 0, 0); if (rt) { dev = rt->rt6i_dev; dev_hold(dev); dst_release(&rt->u.dst); } } else - dev = dev_get_by_index(&init_net, ifindex); + dev = dev_get_by_index(net, ifindex); if (dev == NULL) { sock_kfree_s(sk, mc_lst, sizeof(*mc_lst)); @@ -256,6 +255,7 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr) { struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6_mc_socklist *mc_lst, **lnk; + struct net *net = sk->sk_net; write_lock_bh(&ipv6_sk_mc_lock); for (lnk = &np->ipv6_mc_list; (mc_lst = *lnk) !=NULL ; lnk = &mc_lst->next) { @@ -266,7 +266,8 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr) *lnk = mc_lst->next; write_unlock_bh(&ipv6_sk_mc_lock); - if ((dev = dev_get_by_index(&init_net, mc_lst->ifindex)) != NULL) { + dev = dev_get_by_index(net, mc_lst->ifindex); + if (dev != NULL) { struct inet6_dev *idev = in6_dev_get(dev); (void) ip6_mc_leave_src(sk, mc_lst, idev); @@ -286,7 +287,9 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr) return -EADDRNOTAVAIL; } -static struct inet6_dev *ip6_mc_find_dev(struct in6_addr *group, int ifindex) +static struct inet6_dev *ip6_mc_find_dev(struct net *net, + struct in6_addr *group, + int ifindex) { struct net_device *dev = NULL; struct inet6_dev *idev = NULL; @@ -294,14 +297,14 @@ static struct inet6_dev *ip6_mc_find_dev(struct in6_addr *group, int ifindex) if (ifindex == 0) { struct rt6_info *rt; - rt = rt6_lookup(&init_net, group, NULL, 0, 0); + rt = rt6_lookup(net, group, NULL, 0, 0); if (rt) { dev = rt->rt6i_dev; dev_hold(dev); dst_release(&rt->u.dst); } } else - dev = dev_get_by_index(&init_net, ifindex); + dev = dev_get_by_index(net, ifindex); if (!dev) return NULL; @@ -324,6 +327,7 @@ void ipv6_sock_mc_close(struct sock *sk) { struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6_mc_socklist *mc_lst; + struct net *net = sk->sk_net; write_lock_bh(&ipv6_sk_mc_lock); while ((mc_lst = np->ipv6_mc_list) != NULL) { @@ -332,7 +336,7 @@ void ipv6_sock_mc_close(struct sock *sk) np->ipv6_mc_list = mc_lst->next; write_unlock_bh(&ipv6_sk_mc_lock); - dev = dev_get_by_index(&init_net, mc_lst->ifindex); + dev = dev_get_by_index(net, mc_lst->ifindex); if (dev) { struct inet6_dev *idev = in6_dev_get(dev); @@ -361,6 +365,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk, struct inet6_dev *idev; struct ipv6_pinfo *inet6 = inet6_sk(sk); struct ip6_sf_socklist *psl; + struct net *net = sk->sk_net; int i, j, rv; int leavegroup = 0; int pmclocked = 0; @@ -376,7 +381,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk, if (!ipv6_addr_is_multicast(group)) return -EINVAL; - idev = ip6_mc_find_dev(group, pgsr->gsr_interface); + idev = ip6_mc_find_dev(net, group, pgsr->gsr_interface); if (!idev) return -ENODEV; dev = idev->dev; @@ -500,6 +505,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf) struct inet6_dev *idev; struct ipv6_pinfo *inet6 = inet6_sk(sk); struct ip6_sf_socklist *newpsl, *psl; + struct net *net = sk->sk_net; int leavegroup = 0; int i, err; @@ -511,7 +517,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf) gsf->gf_fmode != MCAST_EXCLUDE) return -EINVAL; - idev = ip6_mc_find_dev(group, gsf->gf_interface); + idev = ip6_mc_find_dev(net, group, gsf->gf_interface); if (!idev) return -ENODEV; @@ -592,13 +598,14 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf, struct net_device *dev; struct ipv6_pinfo *inet6 = inet6_sk(sk); struct ip6_sf_socklist *psl; + struct net *net = sk->sk_net; group = &((struct sockaddr_in6 *)&gsf->gf_group)->sin6_addr; if (!ipv6_addr_is_multicast(group)) return -EINVAL; - idev = ip6_mc_find_dev(group, gsf->gf_interface); + idev = ip6_mc_find_dev(net, group, gsf->gf_interface); if (!idev) return -ENODEV; @@ -1393,7 +1400,8 @@ mld_scount(struct ifmcaddr6 *pmc, int type, int gdeleted, int sdeleted) static struct sk_buff *mld_newpack(struct net_device *dev, int size) { - struct sock *sk = igmp6_socket->sk; + struct net *net = dev->nd_net; + struct sock *sk = net->ipv6.igmp_sk; struct sk_buff *skb; struct mld2_report *pmr; struct in6_addr addr_buf; @@ -1440,6 +1448,7 @@ static void mld_sendpack(struct sk_buff *skb) (struct mld2_report *)skb_transport_header(skb); int payload_len, mldlen; struct inet6_dev *idev = in6_dev_get(skb->dev); + struct net *net = skb->dev->nd_net; int err; struct flowi fl; @@ -1459,7 +1468,7 @@ static void mld_sendpack(struct sk_buff *skb) goto err_out; } - icmpv6_flow_init(igmp6_socket->sk, &fl, ICMPV6_MLD2_REPORT, + icmpv6_flow_init(net->ipv6.igmp_sk, &fl, ICMPV6_MLD2_REPORT, &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->dev->ifindex); @@ -1753,7 +1762,8 @@ static void mld_send_cr(struct inet6_dev *idev) static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) { - struct sock *sk = igmp6_socket->sk; + struct net *net = dev->nd_net; + struct sock *sk = net->ipv6.igmp_sk; struct inet6_dev *idev; struct sk_buff *skb; struct icmp6hdr *hdr; @@ -1824,7 +1834,7 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) goto err_out; } - icmpv6_flow_init(igmp6_socket->sk, &fl, type, + icmpv6_flow_init(sk, &fl, type, &ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr, skb->dev->ifindex); @@ -2334,6 +2344,7 @@ void ipv6_mc_destroy_dev(struct inet6_dev *idev) #ifdef CONFIG_PROC_FS struct igmp6_mc_iter_state { + struct seq_net_private p; struct net_device *dev; struct inet6_dev *idev; }; @@ -2344,9 +2355,10 @@ static inline struct ifmcaddr6 *igmp6_mc_get_first(struct seq_file *seq) { struct ifmcaddr6 *im = NULL; struct igmp6_mc_iter_state *state = igmp6_mc_seq_private(seq); + struct net *net = state->p.net; state->idev = NULL; - for_each_netdev(&init_net, state->dev) { + for_each_netdev(net, state->dev) { struct inet6_dev *idev; idev = in6_dev_get(state->dev); if (!idev) @@ -2448,8 +2460,8 @@ static const struct seq_operations igmp6_mc_seq_ops = { static int igmp6_mc_seq_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &igmp6_mc_seq_ops, - sizeof(struct igmp6_mc_iter_state)); + return seq_open_net(inode, file, &igmp6_mc_seq_ops, + sizeof(struct igmp6_mc_iter_state)); } static const struct file_operations igmp6_mc_seq_fops = { @@ -2457,10 +2469,11 @@ static const struct file_operations igmp6_mc_seq_fops = { .open = igmp6_mc_seq_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; struct igmp6_mcf_iter_state { + struct seq_net_private p; struct net_device *dev; struct inet6_dev *idev; struct ifmcaddr6 *im; @@ -2473,10 +2486,11 @@ static inline struct ip6_sf_list *igmp6_mcf_get_first(struct seq_file *seq) struct ip6_sf_list *psf = NULL; struct ifmcaddr6 *im = NULL; struct igmp6_mcf_iter_state *state = igmp6_mcf_seq_private(seq); + struct net *net = state->p.net; state->idev = NULL; state->im = NULL; - for_each_netdev(&init_net, state->dev) { + for_each_netdev(net, state->dev) { struct inet6_dev *idev; idev = in6_dev_get(state->dev); if (unlikely(idev == NULL)) @@ -2608,8 +2622,8 @@ static const struct seq_operations igmp6_mcf_seq_ops = { static int igmp6_mcf_seq_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &igmp6_mcf_seq_ops, - sizeof(struct igmp6_mcf_iter_state)); + return seq_open_net(inode, file, &igmp6_mcf_seq_ops, + sizeof(struct igmp6_mcf_iter_state)); } static const struct file_operations igmp6_mcf_seq_fops = { @@ -2617,26 +2631,27 @@ static const struct file_operations igmp6_mcf_seq_fops = { .open = igmp6_mcf_seq_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; #endif -int __init igmp6_init(void) +static int igmp6_net_init(struct net *net) { struct ipv6_pinfo *np; + struct socket *sock; struct sock *sk; int err; - err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &igmp6_socket); + err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6, &sock); if (err < 0) { printk(KERN_ERR "Failed to initialize the IGMP6 control socket (err %d).\n", err); - igmp6_socket = NULL; /* For safety. */ - return err; + goto out; } - sk = igmp6_socket->sk; + net->ipv6.igmp_sk = sk = sock->sk; + sk_change_net(sk, net); sk->sk_allocation = GFP_ATOMIC; sk->sk_prot->unhash(sk); @@ -2644,20 +2659,45 @@ int __init igmp6_init(void) np->hop_limit = 1; #ifdef CONFIG_PROC_FS - proc_net_fops_create(&init_net, "igmp6", S_IRUGO, &igmp6_mc_seq_fops); - proc_net_fops_create(&init_net, "mcfilter6", S_IRUGO, &igmp6_mcf_seq_fops); + err = -ENOMEM; + if (!proc_net_fops_create(net, "igmp6", S_IRUGO, &igmp6_mc_seq_fops)) + goto out_sock_create; + if (!proc_net_fops_create(net, "mcfilter6", S_IRUGO, + &igmp6_mcf_seq_fops)) { + proc_net_remove(net, "igmp6"); + goto out_sock_create; + } #endif - return 0; + err = 0; +out: + return err; + +out_sock_create: + sk_release_kernel(net->ipv6.igmp_sk); + goto out; } -void igmp6_cleanup(void) +static void igmp6_net_exit(struct net *net) { - sock_release(igmp6_socket); - igmp6_socket = NULL; /* for safety */ - + sk_release_kernel(net->ipv6.igmp_sk); #ifdef CONFIG_PROC_FS - proc_net_remove(&init_net, "mcfilter6"); - proc_net_remove(&init_net, "igmp6"); + proc_net_remove(net, "mcfilter6"); + proc_net_remove(net, "igmp6"); #endif } + +static struct pernet_operations igmp6_net_ops = { + .init = igmp6_net_init, + .exit = igmp6_net_exit, +}; + +int __init igmp6_init(void) +{ + return register_pernet_subsys(&igmp6_net_ops); +} + +void igmp6_cleanup(void) +{ + unregister_pernet_subsys(&igmp6_net_ops); +} diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 53739de829db..d6e311f6c8eb 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -323,6 +323,9 @@ static struct sock *udp_v6_mcast_next(struct sock *sk, sk_for_each_from(s, node) { struct inet_sock *inet = inet_sk(s); + if (s->sk_net != sk->sk_net) + continue; + if (s->sk_hash == num && s->sk_family == PF_INET6) { struct ipv6_pinfo *np = inet6_sk(s); if (inet->dport) { -- cgit v1.2.3 From 11f4b1cec98ad95abda80dc20bdc3cecac145d77 Mon Sep 17 00:00:00 2001 From: Ron Rindjunsky Date: Tue, 4 Mar 2008 18:09:26 -0800 Subject: mac80211: adding mac80211_tx_control_flags and HT flags This patch makes enum from the defines previously dwelled inside ieee80211_tx_control for better readability. The patch also addes HT flags, for 802.11n drivers: - IEEE80211_TXCTL_OFDM_HT: request low-level driver to use HT OFDM rates - IEEE80211_TXCTL_GREEN_FIELD: use green field protection - IEEE80211_TXCTL_DUP_DATA: duplicate data on both 20 Mhz channels - IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame in 40Mhz width - IEEE80211_TXCTL_SHORT_GI: send this frame with short guard interval Tx command can be a combination of any of these flags, along with bitrate represented by ieee80211_rate. this will allow legacy drivers to switch easily to any 11n rate representation. Signed-off-by: Ron Rindjunsky Signed-off-by: Tomas Winkler CC: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 86 ++++++++++++++++++++++++++++++++------------------ 1 file changed, 55 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index d002b1c6e78e..cde1b4953534 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -205,6 +205,58 @@ struct ieee80211_bss_conf { bool use_short_preamble; }; +/** + * enum mac80211_tx_control_flags - flags to describe Tx configuration for + * the Tx frame + * + * These flags are used with the @flags member of &ieee80211_tx_control + * + * @IEEE80211_TXCTL_REQ_TX_STATUS: request TX status callback for this frame. + * @IEEE80211_TXCTL_DO_NOT_ENCRYPT: send this frame without encryption; + * e.g., for EAPOL frame + * @IEEE80211_TXCTL_USE_RTS_CTS: use RTS-CTS before sending frame + * @IEEE80211_TXCTL_USE_CTS_PROTECT: use CTS protection for the frame (e.g., + * for combined 802.11g / 802.11b networks) + * @IEEE80211_TXCTL_NO_ACK: tell the low level not to wait for an ack + * @IEEE80211_TXCTL_RATE_CTRL_PROBE + * @EEE80211_TXCTL_CLEAR_PS_FILT: clear powersave filter + * for destination station + * @IEEE80211_TXCTL_REQUEUE: + * @IEEE80211_TXCTL_FIRST_FRAGMENT: this is a first fragment of the frame + * @IEEE80211_TXCTL_LONG_RETRY_LIMIT: this frame should be send using the + * through set_retry_limit configured long + * retry value + * @IEEE80211_TXCTL_EAPOL_FRAME: internal to mac80211 + * @IEEE80211_TXCTL_SEND_AFTER_DTIM: send this frame after DTIM beacon + * @IEEE80211_TXCTL_AMPDU: this frame should be sent as part of an A-MPDU + * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates + * @IEEE80211_TXCTL_GREEN_FIELD: use green field protection for this frame + * @IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width + * @IEEE80211_TXCTL_DUP_DATA: duplicate data frame on both 20 Mhz channels + * @IEEE80211_TXCTL_SHORT_GI: send this frame using short guard interval + */ +enum mac80211_tx_control_flags { + IEEE80211_TXCTL_REQ_TX_STATUS = (1<<0), + IEEE80211_TXCTL_DO_NOT_ENCRYPT = (1<<1), + IEEE80211_TXCTL_USE_RTS_CTS = (1<<2), + IEEE80211_TXCTL_USE_CTS_PROTECT = (1<<3), + IEEE80211_TXCTL_NO_ACK = (1<<4), + IEEE80211_TXCTL_RATE_CTRL_PROBE = (1<<5), + IEEE80211_TXCTL_CLEAR_PS_FILT = (1<<6), + IEEE80211_TXCTL_REQUEUE = (1<<7), + IEEE80211_TXCTL_FIRST_FRAGMENT = (1<<8), + IEEE80211_TXCTL_SHORT_PREAMBLE = (1<<9), + IEEE80211_TXCTL_LONG_RETRY_LIMIT = (1<<10), + IEEE80211_TXCTL_EAPOL_FRAME = (1<<11), + IEEE80211_TXCTL_SEND_AFTER_DTIM = (1<<12), + IEEE80211_TXCTL_AMPDU = (1<<13), + IEEE80211_TXCTL_OFDM_HT = (1<<14), + IEEE80211_TXCTL_GREEN_FIELD = (1<<15), + IEEE80211_TXCTL_40_MHZ_WIDTH = (1<<16), + IEEE80211_TXCTL_DUP_DATA = (1<<17), + IEEE80211_TXCTL_SHORT_GI = (1<<18), +}; + /* Transmit control fields. This data structure is passed to low-level driver * with each TX frame. The low-level driver is responsible for configuring * the hardware to use given values (depending on what is supported). */ @@ -219,42 +271,14 @@ struct ieee80211_tx_control { /* retry rate for the last retries */ struct ieee80211_rate *alt_retry_rate; -#define IEEE80211_TXCTL_REQ_TX_STATUS (1<<0)/* request TX status callback for - * this frame */ -#define IEEE80211_TXCTL_DO_NOT_ENCRYPT (1<<1) /* send this frame without - * encryption; e.g., for EAPOL - * frames */ -#define IEEE80211_TXCTL_USE_RTS_CTS (1<<2) /* use RTS-CTS before sending - * frame */ -#define IEEE80211_TXCTL_USE_CTS_PROTECT (1<<3) /* use CTS protection for the - * frame (e.g., for combined - * 802.11g / 802.11b networks) */ -#define IEEE80211_TXCTL_NO_ACK (1<<4) /* tell the low level not to - * wait for an ack */ -#define IEEE80211_TXCTL_RATE_CTRL_PROBE (1<<5) -#define IEEE80211_TXCTL_CLEAR_PS_FILT (1<<6) /* clear powersave filter - * for destination station */ -#define IEEE80211_TXCTL_REQUEUE (1<<7) -#define IEEE80211_TXCTL_FIRST_FRAGMENT (1<<8) /* this is a first fragment of - * the frame */ -#define IEEE80211_TXCTL_SHORT_PREAMBLE (1<<9) -#define IEEE80211_TXCTL_LONG_RETRY_LIMIT (1<<10) /* this frame should be send - * using the through - * set_retry_limit configured - * long retry value */ -#define IEEE80211_TXCTL_EAPOL_FRAME (1<<11) /* internal to mac80211 */ -#define IEEE80211_TXCTL_SEND_AFTER_DTIM (1<<12) /* send this frame after DTIM - * beacon */ -#define IEEE80211_TXCTL_AMPDU (1<<13) /* this frame should be sent - * as part of an A-MPDU */ - u32 flags; /* tx control flags defined - * above */ + u32 flags; /* tx control flags defined above */ u8 key_idx; /* keyidx from hw->set_key(), undefined if * IEEE80211_TXCTL_DO_NOT_ENCRYPT is set */ u8 retry_limit; /* 1 = only first attempt, 2 = one retry, .. * This could be used when set_retry_limit * is not implemented by the driver */ - u8 antenna_sel_tx; /* 0 = default/diversity, 1 = Ant0, 2 = Ant1 */ + u8 antenna_sel_tx; /* 0 = default/diversity, otherwise bit + * position represents antenna number used */ u8 icv_len; /* length of the ICV/MIC field in octets */ u8 iv_len; /* length of the IV field in octets */ u8 queue; /* hardware queue to use for this frame; -- cgit v1.2.3 From 6c5ef8a7059e4f7adc37b337face8b0a8cbd4f48 Mon Sep 17 00:00:00 2001 From: Ron Rindjunsky Date: Wed, 5 Mar 2008 17:38:59 +0200 Subject: mac80211: document IEEE80211_TXCTL_OFDM_HT This patch clarifies the use of IEEE80211_TXCTL_OFDM_HT flag. Can by united with patch "mac80211: adding mac80211_tx_control flags and HT flags" Signed-off-by: Ron Rindjunsky Signed-off-by: John W. Linville --- include/net/mac80211.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index cde1b4953534..5ab6a350ee6d 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -229,7 +229,11 @@ struct ieee80211_bss_conf { * @IEEE80211_TXCTL_EAPOL_FRAME: internal to mac80211 * @IEEE80211_TXCTL_SEND_AFTER_DTIM: send this frame after DTIM beacon * @IEEE80211_TXCTL_AMPDU: this frame should be sent as part of an A-MPDU - * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates + * @IEEE80211_TXCTL_OFDM_HT: this frame can be sent in HT OFDM rates. number + * of streams when this flag is on can be extracted + * from antenna_sel_tx, so if 1 antenna is marked + * use SISO, 2 antennas marked use MIMO, n antennas + * marked use MIMO_n. * @IEEE80211_TXCTL_GREEN_FIELD: use green field protection for this frame * @IEEE80211_TXCTL_40_MHZ_WIDTH: send this frame using 40 Mhz channel width * @IEEE80211_TXCTL_DUP_DATA: duplicate data frame on both 20 Mhz channels -- cgit v1.2.3 From 7524d7d6de5d5d3f081de8cf5479819fad339661 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 4 Mar 2008 15:26:14 -0800 Subject: the scheduled ieee80211 softmac removal Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: John W. Linville --- Documentation/feature-removal-schedule.txt | 8 - MAINTAINERS | 6 - include/net/ieee80211softmac.h | 373 ---------------- include/net/ieee80211softmac_wx.h | 99 ----- net/ieee80211/Kconfig | 1 - net/ieee80211/Makefile | 1 - net/ieee80211/softmac/Kconfig | 12 - net/ieee80211/softmac/Makefile | 9 - net/ieee80211/softmac/ieee80211softmac_assoc.c | 489 -------------------- net/ieee80211/softmac/ieee80211softmac_auth.c | 413 ----------------- net/ieee80211/softmac/ieee80211softmac_event.c | 189 -------- net/ieee80211/softmac/ieee80211softmac_io.c | 488 -------------------- net/ieee80211/softmac/ieee80211softmac_module.c | 568 ------------------------ net/ieee80211/softmac/ieee80211softmac_priv.h | 244 ---------- net/ieee80211/softmac/ieee80211softmac_scan.c | 254 ----------- net/ieee80211/softmac/ieee80211softmac_wx.c | 508 --------------------- 16 files changed, 3662 deletions(-) delete mode 100644 include/net/ieee80211softmac.h delete mode 100644 include/net/ieee80211softmac_wx.h delete mode 100644 net/ieee80211/softmac/Kconfig delete mode 100644 net/ieee80211/softmac/Makefile delete mode 100644 net/ieee80211/softmac/ieee80211softmac_assoc.c delete mode 100644 net/ieee80211/softmac/ieee80211softmac_auth.c delete mode 100644 net/ieee80211/softmac/ieee80211softmac_event.c delete mode 100644 net/ieee80211/softmac/ieee80211softmac_io.c delete mode 100644 net/ieee80211/softmac/ieee80211softmac_module.c delete mode 100644 net/ieee80211/softmac/ieee80211softmac_priv.h delete mode 100644 net/ieee80211/softmac/ieee80211softmac_scan.c delete mode 100644 net/ieee80211/softmac/ieee80211softmac_wx.c (limited to 'include') diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index dade1d14f5b2..2cf4d7ad7b0b 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt @@ -230,14 +230,6 @@ Who: Jean Delvare --------------------------- -What: ieee80211 softmac wireless networking component -When: 2.6.26 (or after removal of bcm43xx and port of zd1211rw to mac80211) -Files: net/ieee80211/softmac -Why: No in-kernel drivers will depend on it any longer. -Who: John W. Linville - ---------------------------- - What: rc80211-simple rate control algorithm for mac80211 When: 2.6.26 Files: net/mac80211/rc80211-simple.c diff --git a/MAINTAINERS b/MAINTAINERS index 38b767611bbf..b9fc1f620f29 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3548,12 +3548,6 @@ M: mhoffman@lightlink.com L: lm-sensors@lm-sensors.org S: Maintained -SOFTMAC LAYER (IEEE 802.11) -P: Daniel Drake -M: dsd@gentoo.org -L: linux-wireless@vger.kernel.org -S: Obsolete - SOFTWARE RAID (Multiple Disks) SUPPORT P: Ingo Molnar M: mingo@redhat.com diff --git a/include/net/ieee80211softmac.h b/include/net/ieee80211softmac.h deleted file mode 100644 index 1ef6282fdded..000000000000 --- a/include/net/ieee80211softmac.h +++ /dev/null @@ -1,373 +0,0 @@ -/* - * ieee80211softmac.h - public interface to the softmac - * - * Copyright (c) 2005 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#ifndef IEEE80211SOFTMAC_H_ -#define IEEE80211SOFTMAC_H_ - -#include -#include -#include -#include -#include - -/* Once the API is considered more or less stable, - * this should be incremented on API incompatible changes. - */ -#define IEEE80211SOFTMAC_API 0 - -#define IEEE80211SOFTMAC_MAX_RATES_LEN 8 -#define IEEE80211SOFTMAC_MAX_EX_RATES_LEN 255 - -struct ieee80211softmac_ratesinfo { - u8 count; - u8 rates[IEEE80211SOFTMAC_MAX_RATES_LEN + IEEE80211SOFTMAC_MAX_EX_RATES_LEN]; -}; - -/* internal structures */ -struct ieee80211softmac_network; -struct ieee80211softmac_scaninfo; - -struct ieee80211softmac_essid { - u8 len; - char data[IW_ESSID_MAX_SIZE+1]; -}; - -struct ieee80211softmac_wpa { - char *IE; - int IElen; - int IEbuflen; -}; - -/* - * Information about association - */ -struct ieee80211softmac_assoc_info { - - struct mutex mutex; - - /* - * This is the requested ESSID. It is written - * only by the WX handlers. - * - */ - struct ieee80211softmac_essid req_essid; - /* - * the ESSID of the network we're currently - * associated (or trying) to. This is - * updated to the network's actual ESSID - * even if the requested ESSID was 'ANY' - */ - struct ieee80211softmac_essid associate_essid; - - /* BSSID we're trying to associate to */ - char bssid[ETH_ALEN]; - - /* some flags. - * static_essid is valid if the essid is constant, - * this is for use by the wx handlers only. - * - * associating is true, if the network has been - * auth'ed on and we are in the process of associating. - * - * bssvalid is true if we found a matching network - * and saved it's BSSID into the bssid above. - * - * bssfixed is used for SIOCSIWAP. - */ - u8 static_essid; - u8 short_preamble_available; - u8 associating; - u8 associated; - u8 assoc_wait; - u8 bssvalid; - u8 bssfixed; - - /* Scan retries remaining */ - int scan_retry; - - struct delayed_work work; - struct delayed_work timeout; -}; - -struct ieee80211softmac_bss_info { - /* Rates supported by the network */ - struct ieee80211softmac_ratesinfo supported_rates; - - /* This indicates whether frames can currently be transmitted with - * short preamble (only use this variable during TX at CCK rates) */ - u8 short_preamble:1; - - /* This indicates whether protection (e.g. self-CTS) should be used - * when transmitting with OFDM modulation */ - u8 use_protection:1; -}; - -enum { - IEEE80211SOFTMAC_AUTH_OPEN_REQUEST = 1, - IEEE80211SOFTMAC_AUTH_OPEN_RESPONSE = 2, -}; - -enum { - IEEE80211SOFTMAC_AUTH_SHARED_REQUEST = 1, - IEEE80211SOFTMAC_AUTH_SHARED_CHALLENGE = 2, - IEEE80211SOFTMAC_AUTH_SHARED_RESPONSE = 3, - IEEE80211SOFTMAC_AUTH_SHARED_PASS = 4, -}; - -/* We should make these tunable - * AUTH_TIMEOUT seems really long, but that's what it is in BSD */ -#define IEEE80211SOFTMAC_AUTH_TIMEOUT (12 * HZ) -#define IEEE80211SOFTMAC_AUTH_RETRY_LIMIT 5 -#define IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT 3 - -struct ieee80211softmac_txrates { - /* The Bit-Rate to be used for multicast frames. */ - u8 mcast_rate; - - /* The Bit-Rate to be used for multicast management frames. */ - u8 mgt_mcast_rate; - - /* The Bit-Rate to be used for any other (normal) data packet. */ - u8 default_rate; - /* The Bit-Rate to be used for default fallback - * (If the device supports fallback and hardware-retry) - */ - u8 default_fallback; - - /* This is the rate that the user asked for */ - u8 user_rate; -}; - -/* Bits for txrates_change callback. */ -#define IEEE80211SOFTMAC_TXRATECHG_DEFAULT (1 << 0) /* default_rate */ -#define IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK (1 << 1) /* default_fallback */ -#define IEEE80211SOFTMAC_TXRATECHG_MCAST (1 << 2) /* mcast_rate */ -#define IEEE80211SOFTMAC_TXRATECHG_MGT_MCAST (1 << 3) /* mgt_mcast_rate */ - -#define IEEE80211SOFTMAC_BSSINFOCHG_RATES (1 << 0) /* supported_rates */ -#define IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE (1 << 1) /* short_preamble */ -#define IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION (1 << 2) /* use_protection */ - -struct ieee80211softmac_device { - /* 802.11 structure for data stuff */ - struct ieee80211_device *ieee; - struct net_device *dev; - - /* only valid if associated, then holds the Association ID */ - u16 association_id; - - /* the following methods are callbacks that the driver - * using this framework has to assign - */ - - /* always assign these */ - void (*set_bssid_filter)(struct net_device *dev, const u8 *bssid); - void (*set_channel)(struct net_device *dev, u8 channel); - - /* assign if you need it, informational only */ - void (*link_change)(struct net_device *dev); - - /* If the hardware can do scanning, assign _all_ three of these callbacks. - * When the scan finishes, call ieee80211softmac_scan_finished(). - */ - - /* when called, start_scan is guaranteed to not be called again - * until you call ieee80211softmac_scan_finished. - * Return 0 if scanning could start, error otherwise. - * SOFTMAC AUTHORS: don't call this, use ieee80211softmac_start_scan */ - int (*start_scan)(struct net_device *dev); - /* this should block until after ieee80211softmac_scan_finished was called - * SOFTMAC AUTHORS: don't call this, use ieee80211softmac_wait_for_scan */ - void (*wait_for_scan)(struct net_device *dev); - /* stop_scan aborts a scan, but is asynchronous. - * if you want to wait for it too, use wait_for_scan - * SOFTMAC AUTHORS: don't call this, use ieee80211softmac_stop_scan */ - void (*stop_scan)(struct net_device *dev); - - /* we'll need something about beacons here too, for AP or ad-hoc modes */ - - /* Transmission rates to be used by the driver. - * The SoftMAC figures out the best possible rates. - * The driver just needs to read them. - */ - struct ieee80211softmac_txrates txrates; - - /* If the driver needs to do stuff on TX rate changes, assign this - * callback. See IEEE80211SOFTMAC_TXRATECHG for change flags. */ - void (*txrates_change)(struct net_device *dev, - u32 changes); - - /* If the driver needs to do stuff when BSS properties change, assign - * this callback. see IEEE80211SOFTMAC_BSSINFOCHG for change flags. */ - void (*bssinfo_change)(struct net_device *dev, - u32 changes); - - /* private stuff follows */ - /* this lock protects this structure */ - spinlock_t lock; - - struct workqueue_struct *wq; - - u8 running; /* SoftMAC started? */ - u8 scanning; - - struct ieee80211softmac_scaninfo *scaninfo; - struct ieee80211softmac_assoc_info associnfo; - struct ieee80211softmac_bss_info bssinfo; - - struct list_head auth_queue; - struct list_head events; - - struct ieee80211softmac_ratesinfo ratesinfo; - int txrate_badness; - - /* WPA stuff */ - struct ieee80211softmac_wpa wpa; - - /* we need to keep a list of network structs we copied */ - struct list_head network_list; - - /* This must be the last item so that it points to the data - * allocated beyond this structure by alloc_ieee80211 */ - u8 priv[0]; -}; - -extern void ieee80211softmac_scan_finished(struct ieee80211softmac_device *sm); - -static inline void * ieee80211softmac_priv(struct net_device *dev) -{ - return ((struct ieee80211softmac_device *)ieee80211_priv(dev))->priv; -} - -extern struct net_device * alloc_ieee80211softmac(int sizeof_priv); -extern void free_ieee80211softmac(struct net_device *dev); - -/* Call this function if you detect a lost TX fragment. - * (If the device indicates failure of ACK RX, for example.) - * It is wise to call this function if you are able to detect lost packets, - * because it contributes to the TX Rates auto adjustment. - */ -extern void ieee80211softmac_fragment_lost(struct net_device *dev, - u16 wireless_sequence_number); -/* Call this function before _start to tell the softmac what rates - * the hw supports. The rates parameter is copied, so you can - * free it right after calling this function. - * Note that the rates need to be sorted. */ -extern void ieee80211softmac_set_rates(struct net_device *dev, u8 count, u8 *rates); - -/* Finds the highest rate which is: - * 1. Present in ri (optionally a basic rate) - * 2. Supported by the device - * 3. Less than or equal to the user-defined rate - */ -extern u8 ieee80211softmac_highest_supported_rate(struct ieee80211softmac_device *mac, - struct ieee80211softmac_ratesinfo *ri, int basic_only); - -/* Helper function which advises you the rate at which a frame should be - * transmitted at. */ -static inline u8 ieee80211softmac_suggest_txrate(struct ieee80211softmac_device *mac, - int is_multicast, - int is_mgt) -{ - struct ieee80211softmac_txrates *txrates = &mac->txrates; - - if (!mac->associnfo.associated) - return txrates->mgt_mcast_rate; - - /* We are associated, sending unicast frame */ - if (!is_multicast) - return txrates->default_rate; - - /* We are associated, sending multicast frame */ - if (is_mgt) - return txrates->mgt_mcast_rate; - else - return txrates->mcast_rate; -} - -/* Helper function which advises you when it is safe to transmit with short - * preamble. - * You should only call this function when transmitting at CCK rates. */ -static inline int ieee80211softmac_short_preamble_ok(struct ieee80211softmac_device *mac, - int is_multicast, - int is_mgt) -{ - return (is_multicast && is_mgt) ? 0 : mac->bssinfo.short_preamble; -} - -/* Helper function which advises you whether protection (e.g. self-CTS) is - * needed. 1 = protection needed, 0 = no protection needed - * Only use this function when transmitting with OFDM modulation. */ -static inline int ieee80211softmac_protection_needed(struct ieee80211softmac_device *mac) -{ - return mac->bssinfo.use_protection; -} - -/* Start the SoftMAC. Call this after you initialized the device - * and it is ready to run. - */ -extern void ieee80211softmac_start(struct net_device *dev); -/* Stop the SoftMAC. Call this before you shutdown the device. */ -extern void ieee80211softmac_stop(struct net_device *dev); - -/* - * Event system - */ - -/* valid event types */ -#define IEEE80211SOFTMAC_EVENT_ANY -1 /*private use only*/ -#define IEEE80211SOFTMAC_EVENT_SCAN_FINISHED 0 -#define IEEE80211SOFTMAC_EVENT_ASSOCIATED 1 -#define IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED 2 -#define IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT 3 -#define IEEE80211SOFTMAC_EVENT_AUTHENTICATED 4 -#define IEEE80211SOFTMAC_EVENT_AUTH_FAILED 5 -#define IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT 6 -#define IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND 7 -#define IEEE80211SOFTMAC_EVENT_DISASSOCIATED 8 -/* keep this updated! */ -#define IEEE80211SOFTMAC_EVENT_LAST 8 -/* - * If you want to be notified of certain events, you can call - * ieee80211softmac_notify[_atomic] with - * - event set to one of the constants below - * - fun set to a function pointer of the appropriate type - * - context set to the context data you want passed - * The return value is 0, or an error. - */ -typedef void (*notify_function_ptr)(struct net_device *dev, int event_type, void *context); - -#define ieee80211softmac_notify(dev, event, fun, context) ieee80211softmac_notify_gfp(dev, event, fun, context, GFP_KERNEL); -#define ieee80211softmac_notify_atomic(dev, event, fun, context) ieee80211softmac_notify_gfp(dev, event, fun, context, GFP_ATOMIC); - -extern int ieee80211softmac_notify_gfp(struct net_device *dev, - int event, notify_function_ptr fun, void *context, gfp_t gfp_mask); - -/* To clear pending work (for ifconfig down, etc.) */ -extern void -ieee80211softmac_clear_pending_work(struct ieee80211softmac_device *sm); - -#endif /* IEEE80211SOFTMAC_H_ */ diff --git a/include/net/ieee80211softmac_wx.h b/include/net/ieee80211softmac_wx.h deleted file mode 100644 index 4ee3ad57283f..000000000000 --- a/include/net/ieee80211softmac_wx.h +++ /dev/null @@ -1,99 +0,0 @@ -/* - * This file contains the prototypes for the wireless extension - * handlers that the softmac API provides. Include this file to - * use the wx handlers, you can assign these directly. - * - * Copyright (c) 2005 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#ifndef _IEEE80211SOFTMAC_WX_H -#define _IEEE80211SOFTMAC_WX_H - -#include -#include - -extern int -ieee80211softmac_wx_trigger_scan(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_get_scan_results(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_set_essid(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_get_essid(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_set_rate(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_get_rate(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_get_wap(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_set_wap(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra); - -extern int -ieee80211softmac_wx_set_genie(struct net_device *dev, - struct iw_request_info *info, - union iwreq_data *wrqu, - char *extra); - -extern int -ieee80211softmac_wx_get_genie(struct net_device *dev, - struct iw_request_info *info, - union iwreq_data *wrqu, - char *extra); -extern int -ieee80211softmac_wx_set_mlme(struct net_device *dev, - struct iw_request_info *info, - union iwreq_data *wrqu, - char *extra); -#endif /* _IEEE80211SOFTMAC_WX */ diff --git a/net/ieee80211/Kconfig b/net/ieee80211/Kconfig index bd501046c9c0..94ed7d3cd9da 100644 --- a/net/ieee80211/Kconfig +++ b/net/ieee80211/Kconfig @@ -71,4 +71,3 @@ config IEEE80211_CRYPT_TKIP This can be compiled as a module and it will be called "ieee80211_crypt_tkip". -source "net/ieee80211/softmac/Kconfig" diff --git a/net/ieee80211/Makefile b/net/ieee80211/Makefile index 796a7c76ee48..f988417121da 100644 --- a/net/ieee80211/Makefile +++ b/net/ieee80211/Makefile @@ -10,4 +10,3 @@ ieee80211-objs := \ ieee80211_wx.o \ ieee80211_geo.o -obj-$(CONFIG_IEEE80211_SOFTMAC) += softmac/ diff --git a/net/ieee80211/softmac/Kconfig b/net/ieee80211/softmac/Kconfig deleted file mode 100644 index 2811651cb134..000000000000 --- a/net/ieee80211/softmac/Kconfig +++ /dev/null @@ -1,12 +0,0 @@ -config IEEE80211_SOFTMAC - tristate "Software MAC add-on to the IEEE 802.11 networking stack" - depends on IEEE80211 && EXPERIMENTAL - select WIRELESS_EXT - select IEEE80211_CRYPT_WEP - ---help--- - This option enables the hardware independent software MAC addon - for the IEEE 802.11 networking stack. - -config IEEE80211_SOFTMAC_DEBUG - bool "Enable full debugging output" - depends on IEEE80211_SOFTMAC diff --git a/net/ieee80211/softmac/Makefile b/net/ieee80211/softmac/Makefile deleted file mode 100644 index bfcb391bb2c7..000000000000 --- a/net/ieee80211/softmac/Makefile +++ /dev/null @@ -1,9 +0,0 @@ -obj-$(CONFIG_IEEE80211_SOFTMAC) += ieee80211softmac.o -ieee80211softmac-objs := \ - ieee80211softmac_io.o \ - ieee80211softmac_auth.o \ - ieee80211softmac_module.o \ - ieee80211softmac_scan.o \ - ieee80211softmac_wx.o \ - ieee80211softmac_assoc.o \ - ieee80211softmac_event.o diff --git a/net/ieee80211/softmac/ieee80211softmac_assoc.c b/net/ieee80211/softmac/ieee80211softmac_assoc.c deleted file mode 100644 index c4d122ddd72c..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_assoc.c +++ /dev/null @@ -1,489 +0,0 @@ -/* - * This file contains the softmac's association logic. - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#include "ieee80211softmac_priv.h" - -/* - * Overview - * - * Before you can associate, you have to authenticate. - * - */ - -/* Sends out an association request to the desired AP */ -static void -ieee80211softmac_assoc(struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net) -{ - unsigned long flags; - - /* Switch to correct channel for this network */ - mac->set_channel(mac->dev, net->channel); - - /* Send association request */ - ieee80211softmac_send_mgt_frame(mac, net, IEEE80211_STYPE_ASSOC_REQ, 0); - - dprintk(KERN_INFO PFX "sent association request!\n"); - - spin_lock_irqsave(&mac->lock, flags); - mac->associnfo.associated = 0; /* just to make sure */ - - /* Set a timer for timeout */ - /* FIXME: make timeout configurable */ - if (likely(mac->running)) - queue_delayed_work(mac->wq, &mac->associnfo.timeout, 5 * HZ); - spin_unlock_irqrestore(&mac->lock, flags); -} - -void -ieee80211softmac_assoc_timeout(struct work_struct *work) -{ - struct ieee80211softmac_device *mac = - container_of(work, struct ieee80211softmac_device, - associnfo.timeout.work); - struct ieee80211softmac_network *n; - - mutex_lock(&mac->associnfo.mutex); - /* we might race against ieee80211softmac_handle_assoc_response, - * so make sure only one of us does something */ - if (!mac->associnfo.associating) - goto out; - mac->associnfo.associating = 0; - mac->associnfo.bssvalid = 0; - mac->associnfo.associated = 0; - - n = ieee80211softmac_get_network_by_bssid_locked(mac, mac->associnfo.bssid); - - dprintk(KERN_INFO PFX "assoc request timed out!\n"); - ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_TIMEOUT, n); -out: - mutex_unlock(&mac->associnfo.mutex); -} - -void -ieee80211softmac_disassoc(struct ieee80211softmac_device *mac) -{ - unsigned long flags; - - spin_lock_irqsave(&mac->lock, flags); - if (mac->associnfo.associating) - cancel_delayed_work(&mac->associnfo.timeout); - - netif_carrier_off(mac->dev); - - mac->associnfo.associated = 0; - mac->associnfo.bssvalid = 0; - mac->associnfo.associating = 0; - ieee80211softmac_init_bss(mac); - ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_DISASSOCIATED, NULL); - spin_unlock_irqrestore(&mac->lock, flags); -} - -/* Sends out a disassociation request to the desired AP */ -void -ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reason) -{ - struct ieee80211softmac_network *found; - - if (mac->associnfo.bssvalid && mac->associnfo.associated) { - found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid); - if (found) - ieee80211softmac_send_mgt_frame(mac, found, IEEE80211_STYPE_DISASSOC, reason); - } - - ieee80211softmac_disassoc(mac); -} - -static inline int -we_support_all_basic_rates(struct ieee80211softmac_device *mac, u8 *from, u8 from_len) -{ - int idx; - u8 rate; - - for (idx = 0; idx < (from_len); idx++) { - rate = (from)[idx]; - if (!(rate & IEEE80211_BASIC_RATE_MASK)) - continue; - rate &= ~IEEE80211_BASIC_RATE_MASK; - if (!ieee80211softmac_ratesinfo_rate_supported(&mac->ratesinfo, rate)) - return 0; - } - return 1; -} - -static int -network_matches_request(struct ieee80211softmac_device *mac, struct ieee80211_network *net) -{ - /* we cannot associate to networks whose name we don't know */ - if (ieee80211_is_empty_essid(net->ssid, net->ssid_len)) - return 0; - /* do not associate to a network whose BSSBasicRateSet we cannot support */ - if (!we_support_all_basic_rates(mac, net->rates, net->rates_len)) - return 0; - /* do we really need to check the ex rates? */ - if (!we_support_all_basic_rates(mac, net->rates_ex, net->rates_ex_len)) - return 0; - - /* assume that users know what they're doing ... - * (note we don't let them select a net we're incompatible with) */ - if (mac->associnfo.bssfixed) { - return !memcmp(mac->associnfo.bssid, net->bssid, ETH_ALEN); - } - - /* if 'ANY' network requested, take any that doesn't have privacy enabled */ - if (mac->associnfo.req_essid.len == 0 - && !(net->capability & WLAN_CAPABILITY_PRIVACY)) - return 1; - if (net->ssid_len != mac->associnfo.req_essid.len) - return 0; - if (!memcmp(net->ssid, mac->associnfo.req_essid.data, mac->associnfo.req_essid.len)) - return 1; - return 0; -} - -static void -ieee80211softmac_assoc_notify_scan(struct net_device *dev, int event_type, void *context) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - ieee80211softmac_assoc_work(&mac->associnfo.work.work); -} - -static void -ieee80211softmac_assoc_notify_auth(struct net_device *dev, int event_type, void *context) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - - switch (event_type) { - case IEEE80211SOFTMAC_EVENT_AUTHENTICATED: - ieee80211softmac_assoc_work(&mac->associnfo.work.work); - break; - case IEEE80211SOFTMAC_EVENT_AUTH_FAILED: - case IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT: - ieee80211softmac_disassoc(mac); - break; - } -} - -/* This function is called to handle userspace requests (asynchronously) */ -void -ieee80211softmac_assoc_work(struct work_struct *work) -{ - struct ieee80211softmac_device *mac = - container_of(work, struct ieee80211softmac_device, - associnfo.work.work); - struct ieee80211softmac_network *found = NULL; - struct ieee80211_network *net = NULL, *best = NULL; - int bssvalid; - unsigned long flags; - - mutex_lock(&mac->associnfo.mutex); - - if (!mac->associnfo.associating) - goto out; - - /* ieee80211_disassoc might clear this */ - bssvalid = mac->associnfo.bssvalid; - - /* meh */ - if (mac->associnfo.associated) - ieee80211softmac_send_disassoc_req(mac, WLAN_REASON_DISASSOC_STA_HAS_LEFT); - - /* try to find the requested network in our list, if we found one already */ - if (bssvalid || mac->associnfo.bssfixed) - found = ieee80211softmac_get_network_by_bssid(mac, mac->associnfo.bssid); - - /* Search the ieee80211 networks for this network if we didn't find it by bssid, - * but only if we've scanned at least once (to get a better list of networks to - * select from). If we have not scanned before, the !found logic below will be - * invoked and will scan. */ - if (!found && (mac->associnfo.scan_retry < IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT)) - { - s8 rssi = -128; /* if I don't initialise, gcc emits an invalid warning - because it cannot follow the best pointer logic. */ - spin_lock_irqsave(&mac->ieee->lock, flags); - list_for_each_entry(net, &mac->ieee->network_list, list) { - /* we're supposed to find the network with - * the best signal here, as we're asked to join - * any network with a specific ESSID, and many - * different ones could have that. - * - * I'll for now just go with the reported rssi. - * - * We also should take into account the rateset - * here to find the best BSSID to try. - */ - if (network_matches_request(mac, net)) { - if (!best) { - best = net; - rssi = best->stats.rssi; - continue; - } - /* we already had a matching network, so - * compare their properties to get the - * better of the two ... (see above) - */ - if (rssi < net->stats.rssi) { - best = net; - rssi = best->stats.rssi; - } - } - } - /* if we unlock here, we might get interrupted and the `best' - * pointer could go stale */ - if (best) { - found = ieee80211softmac_create_network(mac, best); - /* if found is still NULL, then we got -ENOMEM somewhere */ - if (found) - ieee80211softmac_add_network(mac, found); - } - spin_unlock_irqrestore(&mac->ieee->lock, flags); - } - - if (!found) { - if (mac->associnfo.scan_retry > 0) { - mac->associnfo.scan_retry--; - - /* We know of no such network. Let's scan. - * NB: this also happens if we had no memory to copy the network info... - * Maybe we can hope to have more memory after scanning finishes ;) - */ - dprintk(KERN_INFO PFX "Associate: Scanning for networks first.\n"); - ieee80211softmac_notify(mac->dev, IEEE80211SOFTMAC_EVENT_SCAN_FINISHED, ieee80211softmac_assoc_notify_scan, NULL); - if (ieee80211softmac_start_scan(mac)) { - dprintk(KERN_INFO PFX "Associate: failed to initiate scan. Is device up?\n"); - } - goto out; - } else { - mac->associnfo.associating = 0; - mac->associnfo.associated = 0; - - dprintk(KERN_INFO PFX "Unable to find matching network after scan!\n"); - /* reset the retry counter for the next user request since we - * break out and don't reschedule ourselves after this point. */ - mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT; - ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_NET_NOT_FOUND, NULL); - goto out; - } - } - - /* reset the retry counter for the next user request since we - * now found a net and will try to associate to it, but not - * schedule this function again. */ - mac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT; - mac->associnfo.bssvalid = 1; - memcpy(mac->associnfo.bssid, found->bssid, ETH_ALEN); - /* copy the ESSID for displaying it */ - mac->associnfo.associate_essid.len = found->essid.len; - memcpy(mac->associnfo.associate_essid.data, found->essid.data, IW_ESSID_MAX_SIZE + 1); - - /* we found a network! authenticate (if necessary) and associate to it. */ - if (found->authenticating) { - dprintk(KERN_INFO PFX "Already requested authentication, waiting...\n"); - if(!mac->associnfo.assoc_wait) { - mac->associnfo.assoc_wait = 1; - ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL); - } - goto out; - } - if (!found->authenticated && !found->authenticating) { - /* This relies on the fact that _auth_req only queues the work, - * otherwise adding the notification would be racy. */ - if (!ieee80211softmac_auth_req(mac, found)) { - if(!mac->associnfo.assoc_wait) { - dprintk(KERN_INFO PFX "Cannot associate without being authenticated, requested authentication\n"); - mac->associnfo.assoc_wait = 1; - ieee80211softmac_notify_internal(mac, IEEE80211SOFTMAC_EVENT_ANY, found, ieee80211softmac_assoc_notify_auth, NULL, GFP_KERNEL); - } - } else { - printkl(KERN_WARNING PFX "Not authenticated, but requesting authentication failed. Giving up to associate\n"); - mac->associnfo.assoc_wait = 0; - ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, found); - } - goto out; - } - /* finally! now we can start associating */ - mac->associnfo.assoc_wait = 0; - ieee80211softmac_assoc(mac, found); - -out: - mutex_unlock(&mac->associnfo.mutex); -} - -/* call this to do whatever is necessary when we're associated */ -static void -ieee80211softmac_associated(struct ieee80211softmac_device *mac, - struct ieee80211_assoc_response * resp, - struct ieee80211softmac_network *net) -{ - u16 cap = le16_to_cpu(resp->capability); - u8 erp_value = net->erp_value; - - mac->associnfo.associating = 0; - mac->bssinfo.supported_rates = net->supported_rates; - ieee80211softmac_recalc_txrates(mac); - - mac->associnfo.associated = 1; - - mac->associnfo.short_preamble_available = - (cap & WLAN_CAPABILITY_SHORT_PREAMBLE) != 0; - ieee80211softmac_process_erp(mac, erp_value); - - if (mac->set_bssid_filter) - mac->set_bssid_filter(mac->dev, net->bssid); - memcpy(mac->ieee->bssid, net->bssid, ETH_ALEN); - netif_carrier_on(mac->dev); - - mac->association_id = le16_to_cpup(&resp->aid); -} - -/* received frame handling functions */ -int -ieee80211softmac_handle_assoc_response(struct net_device * dev, - struct ieee80211_assoc_response * resp, - struct ieee80211_network * _ieee80211_network) -{ - /* NOTE: the network parameter has to be mostly ignored by - * this code because it is the ieee80211's pointer - * to the struct, not ours (we made a copy) - */ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - u16 status = le16_to_cpup(&resp->status); - struct ieee80211softmac_network *network = NULL; - unsigned long flags; - DECLARE_MAC_BUF(mac2); - - if (unlikely(!mac->running)) - return -ENODEV; - - spin_lock_irqsave(&mac->lock, flags); - - if (!mac->associnfo.associating) { - /* we race against the timeout function, so make sure - * only one of us can do work */ - spin_unlock_irqrestore(&mac->lock, flags); - return 0; - } - network = ieee80211softmac_get_network_by_bssid_locked(mac, resp->header.addr3); - - /* someone sending us things without us knowing him? Ignore. */ - if (!network) { - dprintk(KERN_INFO PFX "Received unrequested assocation response from %s\n", - print_mac(mac2, resp->header.addr3)); - spin_unlock_irqrestore(&mac->lock, flags); - return 0; - } - - /* now that we know it was for us, we can cancel the timeout */ - cancel_delayed_work(&mac->associnfo.timeout); - - /* if the association response included an ERP IE, update our saved - * copy */ - if (_ieee80211_network->flags & NETWORK_HAS_ERP_VALUE) - network->erp_value = _ieee80211_network->erp_value; - - switch (status) { - case 0: - dprintk(KERN_INFO PFX "associated!\n"); - ieee80211softmac_associated(mac, resp, network); - ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATED, network); - break; - case WLAN_REASON_STA_REQ_ASSOC_WITHOUT_AUTH: - if (!network->auth_desynced_once) { - /* there seem to be a few rare cases where our view of - * the world is obscured, or buggy APs that don't DEAUTH - * us properly. So we handle that, but allow it only once. - */ - printkl(KERN_INFO PFX "We were not authenticated during association, retrying...\n"); - network->authenticated = 0; - /* we don't want to do this more than once ... */ - network->auth_desynced_once = 1; - queue_delayed_work(mac->wq, &mac->associnfo.work, 0); - break; - } - default: - dprintk(KERN_INFO PFX "associating failed (reason: 0x%x)!\n", status); - mac->associnfo.associating = 0; - mac->associnfo.bssvalid = 0; - mac->associnfo.associated = 0; - ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_ASSOCIATE_FAILED, network); - } - - spin_unlock_irqrestore(&mac->lock, flags); - return 0; -} - -void -ieee80211softmac_try_reassoc(struct ieee80211softmac_device *mac) -{ - unsigned long flags; - - spin_lock_irqsave(&mac->lock, flags); - mac->associnfo.associating = 1; - queue_delayed_work(mac->wq, &mac->associnfo.work, 0); - spin_unlock_irqrestore(&mac->lock, flags); -} - -int -ieee80211softmac_handle_disassoc(struct net_device * dev, - struct ieee80211_disassoc *disassoc) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - - if (unlikely(!mac->running)) - return -ENODEV; - - if (memcmp(disassoc->header.addr2, mac->associnfo.bssid, ETH_ALEN)) - return 0; - - if (memcmp(disassoc->header.addr1, mac->dev->dev_addr, ETH_ALEN)) - return 0; - - dprintk(KERN_INFO PFX "got disassoc frame\n"); - ieee80211softmac_disassoc(mac); - - ieee80211softmac_try_reassoc(mac); - - return 0; -} - -int -ieee80211softmac_handle_reassoc_req(struct net_device * dev, - struct ieee80211_reassoc_request * resp) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - struct ieee80211softmac_network *network; - - if (unlikely(!mac->running)) - return -ENODEV; - - network = ieee80211softmac_get_network_by_bssid(mac, resp->header.addr3); - if (!network) { - dprintkl(KERN_INFO PFX "reassoc request from unknown network\n"); - return 0; - } - queue_delayed_work(mac->wq, &mac->associnfo.work, 0); - - return 0; -} diff --git a/net/ieee80211/softmac/ieee80211softmac_auth.c b/net/ieee80211/softmac/ieee80211softmac_auth.c deleted file mode 100644 index 1a96c2572578..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_auth.c +++ /dev/null @@ -1,413 +0,0 @@ -/* - * This file contains the softmac's authentication logic. - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#include "ieee80211softmac_priv.h" - -static void ieee80211softmac_auth_queue(struct work_struct *work); - -/* Queues an auth request to the desired AP */ -int -ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net) -{ - struct ieee80211softmac_auth_queue_item *auth; - unsigned long flags; - DECLARE_MAC_BUF(mac2); - - if (net->authenticating || net->authenticated) - return 0; - net->authenticating = 1; - - /* Add the network if it's not already added */ - ieee80211softmac_add_network(mac, net); - - dprintk(KERN_NOTICE PFX "Queueing Authentication Request to %s\n", print_mac(mac2, net->bssid)); - /* Queue the auth request */ - auth = (struct ieee80211softmac_auth_queue_item *) - kmalloc(sizeof(struct ieee80211softmac_auth_queue_item), GFP_KERNEL); - if(auth == NULL) - return -ENOMEM; - - auth->net = net; - auth->mac = mac; - auth->retry = IEEE80211SOFTMAC_AUTH_RETRY_LIMIT; - auth->state = IEEE80211SOFTMAC_AUTH_OPEN_REQUEST; - INIT_DELAYED_WORK(&auth->work, ieee80211softmac_auth_queue); - - /* Lock (for list) */ - spin_lock_irqsave(&mac->lock, flags); - - /* add to list */ - list_add_tail(&auth->list, &mac->auth_queue); - queue_delayed_work(mac->wq, &auth->work, 0); - spin_unlock_irqrestore(&mac->lock, flags); - - return 0; -} - - -/* Sends an auth request to the desired AP and handles timeouts */ -static void -ieee80211softmac_auth_queue(struct work_struct *work) -{ - struct ieee80211softmac_device *mac; - struct ieee80211softmac_auth_queue_item *auth; - struct ieee80211softmac_network *net; - unsigned long flags; - DECLARE_MAC_BUF(mac2); - - auth = container_of(work, struct ieee80211softmac_auth_queue_item, - work.work); - net = auth->net; - mac = auth->mac; - - if(auth->retry > 0) { - /* Switch to correct channel for this network */ - mac->set_channel(mac->dev, net->channel); - - /* Lock and set flags */ - spin_lock_irqsave(&mac->lock, flags); - if (unlikely(!mac->running)) { - /* Prevent reschedule on workqueue flush */ - spin_unlock_irqrestore(&mac->lock, flags); - return; - } - net->authenticated = 0; - /* add a timeout call so we eventually give up waiting for an auth reply */ - queue_delayed_work(mac->wq, &auth->work, IEEE80211SOFTMAC_AUTH_TIMEOUT); - auth->retry--; - spin_unlock_irqrestore(&mac->lock, flags); - if (ieee80211softmac_send_mgt_frame(mac, auth->net, IEEE80211_STYPE_AUTH, auth->state)) - dprintk(KERN_NOTICE PFX "Sending Authentication Request to %s failed (this shouldn't happen, wait for the timeout).\n", - print_mac(mac2, net->bssid)); - else - dprintk(KERN_NOTICE PFX "Sent Authentication Request to %s.\n", print_mac(mac2, net->bssid)); - return; - } - - printkl(KERN_WARNING PFX "Authentication timed out with %s\n", print_mac(mac2, net->bssid)); - /* Remove this item from the queue */ - spin_lock_irqsave(&mac->lock, flags); - net->authenticating = 0; - ieee80211softmac_call_events_locked(mac, IEEE80211SOFTMAC_EVENT_AUTH_TIMEOUT, net); - cancel_delayed_work(&auth->work); /* just to make sure... */ - list_del(&auth->list); - spin_unlock_irqrestore(&mac->lock, flags); - /* Free it */ - kfree(auth); -} - -/* Sends a response to an auth challenge (for shared key auth). */ -static void -ieee80211softmac_auth_challenge_response(struct work_struct *work) -{ - struct ieee80211softmac_auth_queue_item *aq = - container_of(work, struct ieee80211softmac_auth_queue_item, - work.work); - - /* Send our response */ - ieee80211softmac_send_mgt_frame(aq->mac, aq->net, IEEE80211_STYPE_AUTH, aq->state); -} - -/* Handle the auth response from the AP - * This should be registered with ieee80211 as handle_auth - */ -int -ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth) -{ - - struct list_head *list_ptr; - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - struct ieee80211softmac_auth_queue_item *aq = NULL; - struct ieee80211softmac_network *net = NULL; - unsigned long flags; - u8 * data; - DECLARE_MAC_BUF(mac2); - - if (unlikely(!mac->running)) - return -ENODEV; - - /* Find correct auth queue item */ - spin_lock_irqsave(&mac->lock, flags); - list_for_each(list_ptr, &mac->auth_queue) { - aq = list_entry(list_ptr, struct ieee80211softmac_auth_queue_item, list); - net = aq->net; - if (!memcmp(net->bssid, auth->header.addr2, ETH_ALEN)) - break; - else - aq = NULL; - } - spin_unlock_irqrestore(&mac->lock, flags); - - /* Make sure that we've got an auth queue item for this request */ - if(aq == NULL) - { - dprintkl(KERN_DEBUG PFX "Authentication response received from %s but no queue item exists.\n", print_mac(mac2, auth->header.addr2)); - /* Error #? */ - return -1; - } - - /* Check for out of order authentication */ - if(!net->authenticating) - { - dprintkl(KERN_DEBUG PFX "Authentication response received from %s but did not request authentication.\n",print_mac(mac2, auth->header.addr2)); - return -1; - } - - /* Parse the auth packet */ - switch(le16_to_cpu(auth->algorithm)) { - case WLAN_AUTH_OPEN: - /* Check the status code of the response */ - - switch(le16_to_cpu(auth->status)) { - case WLAN_STATUS_SUCCESS: - /* Update the status to Authenticated */ - spin_lock_irqsave(&mac->lock, flags); - net->authenticating = 0; - net->authenticated = 1; - spin_unlock_irqrestore(&mac->lock, flags); - - /* Send event */ - printkl(KERN_NOTICE PFX "Open Authentication completed with %s\n", print_mac(mac2, net->bssid)); - ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_AUTHENTICATED, net); - break; - default: - /* Lock and reset flags */ - spin_lock_irqsave(&mac->lock, flags); - net->authenticated = 0; - net->authenticating = 0; - spin_unlock_irqrestore(&mac->lock, flags); - - printkl(KERN_NOTICE PFX "Open Authentication with %s failed, error code: %i\n", - print_mac(mac2, net->bssid), le16_to_cpup(&auth->status)); - /* Count the error? */ - break; - } - goto free_aq; - break; - case WLAN_AUTH_SHARED_KEY: - /* Figure out where we are in the process */ - switch(le16_to_cpu(auth->transaction)) { - case IEEE80211SOFTMAC_AUTH_SHARED_CHALLENGE: - /* Check to make sure we have a challenge IE */ - data = (u8 *)auth->info_element; - if (*data++ != MFIE_TYPE_CHALLENGE) { - printkl(KERN_NOTICE PFX "Shared Key Authentication failed due to a missing challenge.\n"); - break; - } - /* Save the challenge */ - spin_lock_irqsave(&mac->lock, flags); - net->challenge_len = *data++; - if (net->challenge_len > WLAN_AUTH_CHALLENGE_LEN) - net->challenge_len = WLAN_AUTH_CHALLENGE_LEN; - kfree(net->challenge); - net->challenge = kmemdup(data, net->challenge_len, - GFP_ATOMIC); - if (net->challenge == NULL) { - printkl(KERN_NOTICE PFX "Shared Key " - "Authentication failed due to " - "memory shortage.\n"); - spin_unlock_irqrestore(&mac->lock, flags); - break; - } - aq->state = IEEE80211SOFTMAC_AUTH_SHARED_RESPONSE; - - /* We reuse the work struct from the auth request here. - * It is safe to do so as each one is per-request, and - * at this point (dealing with authentication response) - * we have obviously already sent the initial auth - * request. */ - cancel_delayed_work(&aq->work); - INIT_DELAYED_WORK(&aq->work, &ieee80211softmac_auth_challenge_response); - queue_delayed_work(mac->wq, &aq->work, 0); - spin_unlock_irqrestore(&mac->lock, flags); - return 0; - case IEEE80211SOFTMAC_AUTH_SHARED_PASS: - kfree(net->challenge); - net->challenge = NULL; - net->challenge_len = 0; - /* Check the status code of the response */ - switch(auth->status) { - case WLAN_STATUS_SUCCESS: - /* Update the status to Authenticated */ - spin_lock_irqsave(&mac->lock, flags); - net->authenticating = 0; - net->authenticated = 1; - spin_unlock_irqrestore(&mac->lock, flags); - printkl(KERN_NOTICE PFX "Shared Key Authentication completed with %s\n", - print_mac(mac2, net->bssid)); - ieee80211softmac_call_events(mac, IEEE80211SOFTMAC_EVENT_AUTHENTICATED, net); - break; - default: - printkl(KERN_NOTICE PFX "Shared Key Authentication with %s failed, error code: %i\n", - print_mac(mac2, net->bssid), le16_to_cpup(&auth->status)); - /* Lock and reset flags */ - spin_lock_irqsave(&mac->lock, flags); - net->authenticating = 0; - net->authenticated = 0; - spin_unlock_irqrestore(&mac->lock, flags); - /* Count the error? */ - break; - } - goto free_aq; - break; - default: - printkl(KERN_WARNING PFX "Unhandled Authentication Step: %i\n", auth->transaction); - break; - } - goto free_aq; - break; - default: - /* ERROR */ - goto free_aq; - break; - } - return 0; -free_aq: - /* Cancel the timeout */ - spin_lock_irqsave(&mac->lock, flags); - cancel_delayed_work(&aq->work); - /* Remove this item from the queue */ - list_del(&aq->list); - spin_unlock_irqrestore(&mac->lock, flags); - - /* Free it */ - kfree(aq); - return 0; -} - -/* - * Handle deauthorization - */ -static void -ieee80211softmac_deauth_from_net(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net) -{ - struct ieee80211softmac_auth_queue_item *aq = NULL; - struct list_head *list_ptr; - unsigned long flags; - - /* deauthentication implies disassociation */ - ieee80211softmac_disassoc(mac); - - /* Lock and reset status flags */ - spin_lock_irqsave(&mac->lock, flags); - net->authenticating = 0; - net->authenticated = 0; - - /* Find correct auth queue item, if it exists */ - list_for_each(list_ptr, &mac->auth_queue) { - aq = list_entry(list_ptr, struct ieee80211softmac_auth_queue_item, list); - if (!memcmp(net->bssid, aq->net->bssid, ETH_ALEN)) - break; - else - aq = NULL; - } - - /* Cancel pending work */ - if(aq != NULL) - /* Not entirely safe? What about running work? */ - cancel_delayed_work(&aq->work); - - /* Free our network ref */ - ieee80211softmac_del_network_locked(mac, net); - if(net->challenge != NULL) - kfree(net->challenge); - kfree(net); - - /* can't transmit data right now... */ - netif_carrier_off(mac->dev); - spin_unlock_irqrestore(&mac->lock, flags); - - ieee80211softmac_try_reassoc(mac); -} - -/* - * Sends a deauth request to the desired AP - */ -int -ieee80211softmac_deauth_req(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net, int reason) -{ - int ret; - - /* Make sure the network is authenticated */ - if (!net->authenticated) - { - dprintkl(KERN_DEBUG PFX "Can't send deauthentication packet, network is not authenticated.\n"); - /* Error okay? */ - return -EPERM; - } - - /* Send the de-auth packet */ - if((ret = ieee80211softmac_send_mgt_frame(mac, net, IEEE80211_STYPE_DEAUTH, reason))) - return ret; - - ieee80211softmac_deauth_from_net(mac, net); - return 0; -} - -/* - * This should be registered with ieee80211 as handle_deauth - */ -int -ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *deauth) -{ - - struct ieee80211softmac_network *net = NULL; - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - DECLARE_MAC_BUF(mac2); - - if (unlikely(!mac->running)) - return -ENODEV; - - if (!deauth) { - dprintk("deauth without deauth packet. eek!\n"); - return 0; - } - - net = ieee80211softmac_get_network_by_bssid(mac, deauth->header.addr2); - - if (net == NULL) { - dprintkl(KERN_DEBUG PFX "Received deauthentication packet from %s, but that network is unknown.\n", - print_mac(mac2, deauth->header.addr2)); - return 0; - } - - /* Make sure the network is authenticated */ - if(!net->authenticated) - { - dprintkl(KERN_DEBUG PFX "Can't perform deauthentication, network is not authenticated.\n"); - /* Error okay? */ - return -EPERM; - } - - ieee80211softmac_deauth_from_net(mac, net); - - /* let's try to re-associate */ - queue_delayed_work(mac->wq, &mac->associnfo.work, 0); - return 0; -} diff --git a/net/ieee80211/softmac/ieee80211softmac_event.c b/net/ieee80211/softmac/ieee80211softmac_event.c deleted file mode 100644 index 8cef05b60f16..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_event.c +++ /dev/null @@ -1,189 +0,0 @@ -/* - * Event system - * Also see comments in public header file and longer explanation below. - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#include "ieee80211softmac_priv.h" - -/* - * Each event has associated to it - * - an event type (see constants in public header) - * - an event context (see below) - * - the function to be called - * - a context (extra parameter to call the function with) - * - and the softmac struct - * - * The event context is private and can only be used from - * within this module. Its meaning varies with the event - * type: - * SCAN_FINISHED, - * DISASSOCIATED: NULL - * ASSOCIATED, - * ASSOCIATE_FAILED, - * ASSOCIATE_TIMEOUT, - * AUTHENTICATED, - * AUTH_FAILED, - * AUTH_TIMEOUT: a pointer to the network struct - * ... - * Code within this module can use the event context to be only - * called when the event is true for that specific context - * as per above table. - * If the event context is NULL, then the notification is always called, - * regardless of the event context. The event context is not passed to - * the callback, it is assumed that the context suffices. - * - * You can also use the event context only by setting the event type - * to -1 (private use only), in which case you'll be notified - * whenever the event context matches. - */ - -static char *event_descriptions[IEEE80211SOFTMAC_EVENT_LAST+1] = { - NULL, /* scan finished */ - NULL, /* associated */ - "associating failed", - "associating timed out", - "authenticated", - "authenticating failed", - "authenticating timed out", - "associating failed because no suitable network was found", - NULL, /* disassociated */ -}; - - -static void -ieee80211softmac_notify_callback(struct work_struct *work) -{ - struct ieee80211softmac_event *pevent = - container_of(work, struct ieee80211softmac_event, work.work); - struct ieee80211softmac_event event = *pevent; - kfree(pevent); - - event.fun(event.mac->dev, event.event_type, event.context); -} - -int -ieee80211softmac_notify_internal(struct ieee80211softmac_device *mac, - int event, void *event_context, notify_function_ptr fun, void *context, gfp_t gfp_mask) -{ - struct ieee80211softmac_event *eventptr; - unsigned long flags; - - if (event < -1 || event > IEEE80211SOFTMAC_EVENT_LAST) - return -ENOSYS; - - if (!fun) - return -EINVAL; - - eventptr = kmalloc(sizeof(struct ieee80211softmac_event), gfp_mask); - if (!eventptr) - return -ENOMEM; - - eventptr->event_type = event; - INIT_DELAYED_WORK(&eventptr->work, ieee80211softmac_notify_callback); - eventptr->fun = fun; - eventptr->context = context; - eventptr->mac = mac; - eventptr->event_context = event_context; - - spin_lock_irqsave(&mac->lock, flags); - list_add(&eventptr->list, &mac->events); - spin_unlock_irqrestore(&mac->lock, flags); - - return 0; -} - -int -ieee80211softmac_notify_gfp(struct net_device *dev, - int event, notify_function_ptr fun, void *context, gfp_t gfp_mask) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - - if (event < 0 || event > IEEE80211SOFTMAC_EVENT_LAST) - return -ENOSYS; - - return ieee80211softmac_notify_internal(mac, event, NULL, fun, context, gfp_mask); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_notify_gfp); - -/* private -- calling all callbacks that were specified */ -void -ieee80211softmac_call_events_locked(struct ieee80211softmac_device *mac, int event, void *event_ctx) -{ - struct ieee80211softmac_event *eventptr, *tmp; - struct ieee80211softmac_network *network; - - if (event >= 0) { - union iwreq_data wrqu; - int we_event; - char *msg = NULL; - - memset(&wrqu, '\0', sizeof (union iwreq_data)); - - switch(event) { - case IEEE80211SOFTMAC_EVENT_ASSOCIATED: - network = (struct ieee80211softmac_network *)event_ctx; - memcpy(wrqu.ap_addr.sa_data, &network->bssid[0], ETH_ALEN); - /* fall through */ - case IEEE80211SOFTMAC_EVENT_DISASSOCIATED: - wrqu.ap_addr.sa_family = ARPHRD_ETHER; - we_event = SIOCGIWAP; - break; - case IEEE80211SOFTMAC_EVENT_SCAN_FINISHED: - we_event = SIOCGIWSCAN; - break; - default: - msg = event_descriptions[event]; - if (!msg) - msg = "SOFTMAC EVENT BUG"; - wrqu.data.length = strlen(msg); - we_event = IWEVCUSTOM; - break; - } - wireless_send_event(mac->dev, we_event, &wrqu, msg); - } - - if (!list_empty(&mac->events)) - list_for_each_entry_safe(eventptr, tmp, &mac->events, list) { - if ((eventptr->event_type == event || eventptr->event_type == -1) - && (eventptr->event_context == NULL || eventptr->event_context == event_ctx)) { - list_del(&eventptr->list); - /* User may have subscribed to ANY event, so - * we tell them which event triggered it. */ - eventptr->event_type = event; - queue_delayed_work(mac->wq, &eventptr->work, 0); - } - } -} - -void -ieee80211softmac_call_events(struct ieee80211softmac_device *mac, int event, void *event_ctx) -{ - unsigned long flags; - - spin_lock_irqsave(&mac->lock, flags); - ieee80211softmac_call_events_locked(mac, event, event_ctx); - - spin_unlock_irqrestore(&mac->lock, flags); -} diff --git a/net/ieee80211/softmac/ieee80211softmac_io.c b/net/ieee80211/softmac/ieee80211softmac_io.c deleted file mode 100644 index 73b4b13fbd8f..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_io.c +++ /dev/null @@ -1,488 +0,0 @@ -/* - * Some parts based on code from net80211 - * Copyright (c) 2001 Atsushi Onoe - * Copyright (c) 2002-2005 Sam Leffler, Errno Consulting - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. The name of the author may not be used to endorse or promote products - * derived from this software without specific prior written permission. - * - * Alternatively, this software may be distributed under the terms of the - * GNU General Public License ("GPL") version 2 as published by the Free - * Software Foundation. - * - * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR - * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES - * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. - * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, - * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT - * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF - * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - * - */ - -#include "ieee80211softmac_priv.h" - -/* Helper functions for inserting data into the frames */ - -/* - * Adds an ESSID element to the frame - * - */ -static u8 * -ieee80211softmac_add_essid(u8 *dst, struct ieee80211softmac_essid *essid) -{ - if (essid) { - *dst++ = MFIE_TYPE_SSID; - *dst++ = essid->len; - memcpy(dst, essid->data, essid->len); - return dst+essid->len; - } else { - *dst++ = MFIE_TYPE_SSID; - *dst++ = 0; - return dst; - } -} - -/* Adds Supported Rates and if required Extended Rates Information Element - * to the frame, ASSUMES WE HAVE A SORTED LIST OF RATES */ -static u8 * -ieee80211softmac_frame_add_rates(u8 *dst, const struct ieee80211softmac_ratesinfo *r) -{ - int cck_len, ofdm_len; - *dst++ = MFIE_TYPE_RATES; - - for(cck_len=0; ieee80211_is_cck_rate(r->rates[cck_len]) && (cck_len < r->count);cck_len++); - - if(cck_len > IEEE80211SOFTMAC_MAX_RATES_LEN) - cck_len = IEEE80211SOFTMAC_MAX_RATES_LEN; - *dst++ = cck_len; - memcpy(dst, r->rates, cck_len); - dst += cck_len; - - if(cck_len < r->count){ - for (ofdm_len=0; ieee80211_is_ofdm_rate(r->rates[ofdm_len + cck_len]) && (ofdm_len + cck_len < r->count); ofdm_len++); - if (ofdm_len > 0) { - if (ofdm_len > IEEE80211SOFTMAC_MAX_EX_RATES_LEN) - ofdm_len = IEEE80211SOFTMAC_MAX_EX_RATES_LEN; - *dst++ = MFIE_TYPE_RATES_EX; - *dst++ = ofdm_len; - memcpy(dst, r->rates + cck_len, ofdm_len); - dst += ofdm_len; - } - } - return dst; -} - -/* Allocate a management frame */ -static u8 * -ieee80211softmac_alloc_mgt(u32 size) -{ - u8 * data; - - /* Add the header and FCS to the size */ - size = size + IEEE80211_3ADDR_LEN; - if(size > IEEE80211_DATA_LEN) - return NULL; - /* Allocate the frame */ - data = kzalloc(size, GFP_ATOMIC); - return data; -} - -/* - * Add a 2 Address Header - */ -static void -ieee80211softmac_hdr_2addr(struct ieee80211softmac_device *mac, - struct ieee80211_hdr_2addr *header, u32 type, u8 *dest) -{ - /* Fill in the frame control flags */ - header->frame_ctl = cpu_to_le16(type); - /* Control packets always have WEP turned off */ - if(type > IEEE80211_STYPE_CFENDACK && type < IEEE80211_STYPE_PSPOLL) - header->frame_ctl |= mac->ieee->sec.level ? cpu_to_le16(IEEE80211_FCTL_PROTECTED) : 0; - - /* Fill in the duration */ - header->duration_id = 0; - /* FIXME: How do I find this? - * calculate. But most drivers just fill in 0 (except if it's a station id of course) */ - - /* Fill in the Destination Address */ - if(dest == NULL) - memset(header->addr1, 0xFF, ETH_ALEN); - else - memcpy(header->addr1, dest, ETH_ALEN); - /* Fill in the Source Address */ - memcpy(header->addr2, mac->ieee->dev->dev_addr, ETH_ALEN); - -} - - -/* Add a 3 Address Header */ -static void -ieee80211softmac_hdr_3addr(struct ieee80211softmac_device *mac, - struct ieee80211_hdr_3addr *header, u32 type, u8 *dest, u8 *bssid) -{ - /* This is common with 2addr, so use that instead */ - ieee80211softmac_hdr_2addr(mac, (struct ieee80211_hdr_2addr *)header, type, dest); - - /* Fill in the BSS ID */ - if(bssid == NULL) - memset(header->addr3, 0xFF, ETH_ALEN); - else - memcpy(header->addr3, bssid, ETH_ALEN); - - /* Fill in the sequence # */ - /* FIXME: I need to add this to the softmac struct - * shouldn't the sequence number be in ieee80211? */ -} - -static __le16 -ieee80211softmac_capabilities(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net) -{ - __le16 capability = 0; - - /* ESS and IBSS bits are set according to the current mode */ - switch (mac->ieee->iw_mode) { - case IW_MODE_INFRA: - capability = cpu_to_le16(WLAN_CAPABILITY_ESS); - break; - case IW_MODE_ADHOC: - capability = cpu_to_le16(WLAN_CAPABILITY_IBSS); - break; - case IW_MODE_AUTO: - capability = cpu_to_le16(net->capabilities & - (WLAN_CAPABILITY_ESS|WLAN_CAPABILITY_IBSS)); - break; - default: - /* bleh. we don't ever go to these modes */ - printk(KERN_ERR PFX "invalid iw_mode!\n"); - break; - } - - /* CF Pollable / CF Poll Request */ - /* Needs to be implemented, for now, the 0's == not supported */ - - /* Privacy Bit */ - capability |= mac->ieee->sec.level ? - cpu_to_le16(WLAN_CAPABILITY_PRIVACY) : 0; - - /* Short Preamble */ - /* Always supported: we probably won't ever be powering devices which - * dont support this... */ - capability |= cpu_to_le16(WLAN_CAPABILITY_SHORT_PREAMBLE); - - /* PBCC */ - /* Not widely used */ - - /* Channel Agility */ - /* Not widely used */ - - /* Short Slot */ - /* Will be implemented later */ - - /* DSSS-OFDM */ - /* Not widely used */ - - return capability; -} - -/***************************************************************************** - * Create Management packets - *****************************************************************************/ - -/* Creates an association request packet */ -static u32 -ieee80211softmac_assoc_req(struct ieee80211_assoc_request **pkt, - struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net) -{ - u8 *data; - (*pkt) = (struct ieee80211_assoc_request *)ieee80211softmac_alloc_mgt( - 2 + /* Capability Info */ - 2 + /* Listen Interval */ - /* SSID IE */ - 1 + 1 + IW_ESSID_MAX_SIZE + - /* Rates IE */ - 1 + 1 + IEEE80211SOFTMAC_MAX_RATES_LEN + - /* Extended Rates IE */ - 1 + 1 + IEEE80211SOFTMAC_MAX_EX_RATES_LEN + - /* WPA IE if present */ - mac->wpa.IElen - /* Other IE's? Optional? - * Yeah, probably need an extra IE parameter -- lots of vendors like to - * fill in their own IEs */ - ); - if (unlikely((*pkt) == NULL)) - return 0; - ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_ASSOC_REQ, net->bssid, net->bssid); - - /* Fill in the capabilities */ - (*pkt)->capability = ieee80211softmac_capabilities(mac, net); - - /* Fill in Listen Interval (?) */ - (*pkt)->listen_interval = cpu_to_le16(10); - - data = (u8 *)(*pkt)->info_element; - /* Add SSID */ - data = ieee80211softmac_add_essid(data, &net->essid); - /* Add Rates */ - data = ieee80211softmac_frame_add_rates(data, &mac->ratesinfo); - /* Add WPA IE */ - if (mac->wpa.IElen && mac->wpa.IE) { - memcpy(data, mac->wpa.IE, mac->wpa.IElen); - data += mac->wpa.IElen; - } - /* Return the number of used bytes */ - return (data - (u8*)(*pkt)); -} - -/* Create a reassociation request packet */ -static u32 -ieee80211softmac_reassoc_req(struct ieee80211_reassoc_request **pkt, - struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net) -{ - u8 *data; - (*pkt) = (struct ieee80211_reassoc_request *)ieee80211softmac_alloc_mgt( - 2 + /* Capability Info */ - 2 + /* Listen Interval */ - ETH_ALEN + /* AP MAC */ - /* SSID IE */ - 1 + 1 + IW_ESSID_MAX_SIZE + - /* Rates IE */ - 1 + 1 + IEEE80211SOFTMAC_MAX_RATES_LEN + - /* Extended Rates IE */ - 1 + 1 + IEEE80211SOFTMAC_MAX_EX_RATES_LEN - /* Other IE's? */ - ); - if (unlikely((*pkt) == NULL)) - return 0; - ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_REASSOC_REQ, net->bssid, net->bssid); - - /* Fill in the capabilities */ - (*pkt)->capability = ieee80211softmac_capabilities(mac, net); - - /* Fill in Listen Interval (?) */ - (*pkt)->listen_interval = cpu_to_le16(10); - /* Fill in the current AP MAC */ - memcpy((*pkt)->current_ap, mac->ieee->bssid, ETH_ALEN); - - data = (u8 *)(*pkt)->info_element; - /* Add SSID */ - data = ieee80211softmac_add_essid(data, &net->essid); - /* Add Rates */ - data = ieee80211softmac_frame_add_rates(data, &mac->ratesinfo); - /* Return packet size */ - return (data - (u8 *)(*pkt)); -} - -/* Create an authentication packet */ -static u32 -ieee80211softmac_auth(struct ieee80211_auth **pkt, - struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net, - u16 transaction, u16 status, int *encrypt_mpdu) -{ - u8 *data; - int auth_mode = mac->ieee->sec.auth_mode; - int is_shared_response = (auth_mode == WLAN_AUTH_SHARED_KEY - && transaction == IEEE80211SOFTMAC_AUTH_SHARED_RESPONSE); - - /* Allocate Packet */ - (*pkt) = (struct ieee80211_auth *)ieee80211softmac_alloc_mgt( - 2 + /* Auth Algorithm */ - 2 + /* Auth Transaction Seq */ - 2 + /* Status Code */ - /* Challenge Text IE */ - (is_shared_response ? 1 + 1 + net->challenge_len : 0) - ); - if (unlikely((*pkt) == NULL)) - return 0; - ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_AUTH, net->bssid, net->bssid); - - /* Algorithm */ - (*pkt)->algorithm = cpu_to_le16(auth_mode); - /* Transaction */ - (*pkt)->transaction = cpu_to_le16(transaction); - /* Status */ - (*pkt)->status = cpu_to_le16(status); - - data = (u8 *)(*pkt)->info_element; - /* Challenge Text */ - if (is_shared_response) { - *data = MFIE_TYPE_CHALLENGE; - data++; - - /* Copy the challenge in */ - *data = net->challenge_len; - data++; - memcpy(data, net->challenge, net->challenge_len); - data += net->challenge_len; - - /* Make sure this frame gets encrypted with the shared key */ - *encrypt_mpdu = 1; - } else - *encrypt_mpdu = 0; - - /* Return the packet size */ - return (data - (u8 *)(*pkt)); -} - -/* Create a disassocation or deauthentication packet */ -static u32 -ieee80211softmac_disassoc_deauth(struct ieee80211_disassoc **pkt, - struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net, - u16 type, u16 reason) -{ - /* Allocate Packet */ - (*pkt) = (struct ieee80211_disassoc *)ieee80211softmac_alloc_mgt(2); - if (unlikely((*pkt) == NULL)) - return 0; - ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), type, net->bssid, net->bssid); - /* Reason */ - (*pkt)->reason = cpu_to_le16(reason); - /* Return the packet size */ - return (2 + IEEE80211_3ADDR_LEN); -} - -/* Create a probe request packet */ -static u32 -ieee80211softmac_probe_req(struct ieee80211_probe_request **pkt, - struct ieee80211softmac_device *mac, struct ieee80211softmac_essid *essid) -{ - u8 *data; - /* Allocate Packet */ - (*pkt) = (struct ieee80211_probe_request *)ieee80211softmac_alloc_mgt( - /* SSID of requested network */ - 1 + 1 + IW_ESSID_MAX_SIZE + - /* Rates IE */ - 1 + 1 + IEEE80211SOFTMAC_MAX_RATES_LEN + - /* Extended Rates IE */ - 1 + 1 + IEEE80211SOFTMAC_MAX_EX_RATES_LEN - ); - if (unlikely((*pkt) == NULL)) - return 0; - ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_PROBE_REQ, NULL, NULL); - - data = (u8 *)(*pkt)->info_element; - /* Add ESSID (can be NULL) */ - data = ieee80211softmac_add_essid(data, essid); - /* Add Rates */ - data = ieee80211softmac_frame_add_rates(data, &mac->ratesinfo); - /* Return packet size */ - return (data - (u8 *)(*pkt)); -} - -/* Create a probe response packet */ -/* FIXME: Not complete */ -static u32 -ieee80211softmac_probe_resp(struct ieee80211_probe_response **pkt, - struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net) -{ - u8 *data; - /* Allocate Packet */ - (*pkt) = (struct ieee80211_probe_response *)ieee80211softmac_alloc_mgt( - 8 + /* Timestamp */ - 2 + /* Beacon Interval */ - 2 + /* Capability Info */ - /* SSID IE */ - 1 + 1 + IW_ESSID_MAX_SIZE + - 7 + /* FH Parameter Set */ - 2 + /* DS Parameter Set */ - 8 + /* CF Parameter Set */ - 4 /* IBSS Parameter Set */ - ); - if (unlikely((*pkt) == NULL)) - return 0; - ieee80211softmac_hdr_3addr(mac, &((*pkt)->header), IEEE80211_STYPE_PROBE_RESP, net->bssid, net->bssid); - data = (u8 *)(*pkt)->info_element; - - /* Return the packet size */ - return (data - (u8 *)(*pkt)); -} - - -/* Sends a manangement packet - * FIXME: document the use of the arg parameter - * for _AUTH: (transaction #) | (status << 16) - */ -int -ieee80211softmac_send_mgt_frame(struct ieee80211softmac_device *mac, - void *ptrarg, u32 type, u32 arg) -{ - void *pkt = NULL; - u32 pkt_size = 0; - int encrypt_mpdu = 0; - - switch(type) { - case IEEE80211_STYPE_ASSOC_REQ: - pkt_size = ieee80211softmac_assoc_req((struct ieee80211_assoc_request **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg); - break; - case IEEE80211_STYPE_REASSOC_REQ: - pkt_size = ieee80211softmac_reassoc_req((struct ieee80211_reassoc_request **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg); - break; - case IEEE80211_STYPE_AUTH: - pkt_size = ieee80211softmac_auth((struct ieee80211_auth **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg, (u16)(arg & 0xFFFF), (u16) (arg >> 16), &encrypt_mpdu); - break; - case IEEE80211_STYPE_DISASSOC: - case IEEE80211_STYPE_DEAUTH: - pkt_size = ieee80211softmac_disassoc_deauth((struct ieee80211_disassoc **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg, type, (u16)(arg & 0xFFFF)); - break; - case IEEE80211_STYPE_PROBE_REQ: - pkt_size = ieee80211softmac_probe_req((struct ieee80211_probe_request **)(&pkt), mac, (struct ieee80211softmac_essid *)ptrarg); - break; - case IEEE80211_STYPE_PROBE_RESP: - pkt_size = ieee80211softmac_probe_resp((struct ieee80211_probe_response **)(&pkt), mac, (struct ieee80211softmac_network *)ptrarg); - break; - default: - printkl(KERN_DEBUG PFX "Unsupported Management Frame type: %i\n", type); - return -EINVAL; - }; - - if(pkt_size == 0 || pkt == NULL) { - printkl(KERN_DEBUG PFX "Error, packet is nonexistant or 0 length\n"); - return -ENOMEM; - } - - /* Send the packet to the ieee80211 layer for tx */ - /* we defined softmac->mgmt_xmit for this. Should we keep it - * as it is (that means we'd need to wrap this into a txb), - * modify the prototype (so it matches this function), - * or get rid of it alltogether? - * Does this work for you now? - */ - ieee80211_tx_frame(mac->ieee, (struct ieee80211_hdr *)pkt, - IEEE80211_3ADDR_LEN, pkt_size, encrypt_mpdu); - - kfree(pkt); - return 0; -} - -/* Beacon handling */ -int ieee80211softmac_handle_beacon(struct net_device *dev, - struct ieee80211_beacon *beacon, - struct ieee80211_network *network) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - - /* This might race, but we don't really care and it's not worth - * adding heavyweight locking in this fastpath. - */ - if (mac->associnfo.associated) { - if (memcmp(network->bssid, mac->associnfo.bssid, ETH_ALEN) == 0) - ieee80211softmac_process_erp(mac, network->erp_value); - } - - return 0; -} - diff --git a/net/ieee80211/softmac/ieee80211softmac_module.c b/net/ieee80211/softmac/ieee80211softmac_module.c deleted file mode 100644 index 07505ca859af..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_module.c +++ /dev/null @@ -1,568 +0,0 @@ -/* - * Contains some basic softmac functions along with module registration code etc. - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#include "ieee80211softmac_priv.h" -#include -#include - -struct net_device *alloc_ieee80211softmac(int sizeof_priv) -{ - struct ieee80211softmac_device *softmac; - struct net_device *dev; - - dev = alloc_ieee80211(sizeof(*softmac) + sizeof_priv); - if (!dev) - return NULL; - softmac = ieee80211_priv(dev); - softmac->wq = create_freezeable_workqueue("softmac"); - if (!softmac->wq) { - free_ieee80211(dev); - return NULL; - } - - softmac->dev = dev; - softmac->ieee = netdev_priv(dev); - spin_lock_init(&softmac->lock); - - softmac->ieee->handle_auth = ieee80211softmac_auth_resp; - softmac->ieee->handle_deauth = ieee80211softmac_deauth_resp; - softmac->ieee->handle_assoc_response = ieee80211softmac_handle_assoc_response; - softmac->ieee->handle_reassoc_request = ieee80211softmac_handle_reassoc_req; - softmac->ieee->handle_disassoc = ieee80211softmac_handle_disassoc; - softmac->ieee->handle_beacon = ieee80211softmac_handle_beacon; - softmac->scaninfo = NULL; - - softmac->associnfo.scan_retry = IEEE80211SOFTMAC_ASSOC_SCAN_RETRY_LIMIT; - - /* TODO: initialise all the other callbacks in the ieee struct - * (once they're written) - */ - - INIT_LIST_HEAD(&softmac->auth_queue); - INIT_LIST_HEAD(&softmac->network_list); - INIT_LIST_HEAD(&softmac->events); - - mutex_init(&softmac->associnfo.mutex); - INIT_DELAYED_WORK(&softmac->associnfo.work, ieee80211softmac_assoc_work); - INIT_DELAYED_WORK(&softmac->associnfo.timeout, ieee80211softmac_assoc_timeout); - softmac->start_scan = ieee80211softmac_start_scan_implementation; - softmac->wait_for_scan = ieee80211softmac_wait_for_scan_implementation; - softmac->stop_scan = ieee80211softmac_stop_scan_implementation; - - /* to start with, we can't send anything ... */ - netif_carrier_off(dev); - - return dev; -} -EXPORT_SYMBOL_GPL(alloc_ieee80211softmac); - -/* Clears the pending work queue items, stops all scans, etc. */ -void -ieee80211softmac_clear_pending_work(struct ieee80211softmac_device *sm) -{ - unsigned long flags; - struct ieee80211softmac_event *eventptr, *eventtmp; - struct ieee80211softmac_auth_queue_item *authptr, *authtmp; - struct ieee80211softmac_network *netptr, *nettmp; - - ieee80211softmac_stop_scan(sm); - ieee80211softmac_wait_for_scan(sm); - - spin_lock_irqsave(&sm->lock, flags); - sm->running = 0; - - /* Free all pending assoc work items */ - cancel_delayed_work(&sm->associnfo.work); - - /* Free all pending scan work items */ - if(sm->scaninfo != NULL) - cancel_delayed_work(&sm->scaninfo->softmac_scan); - - /* Free all pending auth work items */ - list_for_each_entry(authptr, &sm->auth_queue, list) - cancel_delayed_work(&authptr->work); - - /* delete all pending event calls and work items */ - list_for_each_entry_safe(eventptr, eventtmp, &sm->events, list) - cancel_delayed_work(&eventptr->work); - - spin_unlock_irqrestore(&sm->lock, flags); - flush_workqueue(sm->wq); - - /* now we should be save and no longer need locking... */ - spin_lock_irqsave(&sm->lock, flags); - /* Free all pending auth work items */ - list_for_each_entry_safe(authptr, authtmp, &sm->auth_queue, list) { - list_del(&authptr->list); - kfree(authptr); - } - - /* delete all pending event calls and work items */ - list_for_each_entry_safe(eventptr, eventtmp, &sm->events, list) { - list_del(&eventptr->list); - kfree(eventptr); - } - - /* Free all networks */ - list_for_each_entry_safe(netptr, nettmp, &sm->network_list, list) { - ieee80211softmac_del_network_locked(sm, netptr); - if(netptr->challenge != NULL) - kfree(netptr->challenge); - kfree(netptr); - } - - spin_unlock_irqrestore(&sm->lock, flags); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_clear_pending_work); - -void free_ieee80211softmac(struct net_device *dev) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(dev); - ieee80211softmac_clear_pending_work(sm); - kfree(sm->scaninfo); - kfree(sm->wpa.IE); - destroy_workqueue(sm->wq); - free_ieee80211(dev); -} -EXPORT_SYMBOL_GPL(free_ieee80211softmac); - -static void ieee80211softmac_start_check_rates(struct ieee80211softmac_device *mac) -{ - struct ieee80211softmac_ratesinfo *ri = &mac->ratesinfo; - /* I took out the sorting check, we're seperating by modulation now. */ - if (ri->count) - return; - /* otherwise assume we hav'em all! */ - if (mac->ieee->modulation & IEEE80211_CCK_MODULATION) { - ri->rates[ri->count++] = IEEE80211_CCK_RATE_1MB; - ri->rates[ri->count++] = IEEE80211_CCK_RATE_2MB; - ri->rates[ri->count++] = IEEE80211_CCK_RATE_5MB; - ri->rates[ri->count++] = IEEE80211_CCK_RATE_11MB; - } - if (mac->ieee->modulation & IEEE80211_OFDM_MODULATION) { - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_6MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_9MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_12MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_18MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_24MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_36MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_48MB; - ri->rates[ri->count++] = IEEE80211_OFDM_RATE_54MB; - } -} - -int ieee80211softmac_ratesinfo_rate_supported(struct ieee80211softmac_ratesinfo *ri, u8 rate) -{ - int search; - u8 search_rate; - - for (search = 0; search < ri->count; search++) { - search_rate = ri->rates[search]; - search_rate &= ~IEEE80211_BASIC_RATE_MASK; - if (rate == search_rate) - return 1; - } - - return 0; -} - -u8 ieee80211softmac_highest_supported_rate(struct ieee80211softmac_device *mac, - struct ieee80211softmac_ratesinfo *ri, int basic_only) -{ - u8 user_rate = mac->txrates.user_rate; - int i; - - if (ri->count == 0) - return IEEE80211_CCK_RATE_1MB; - - for (i = ri->count - 1; i >= 0; i--) { - u8 rate = ri->rates[i]; - if (basic_only && !(rate & IEEE80211_BASIC_RATE_MASK)) - continue; - rate &= ~IEEE80211_BASIC_RATE_MASK; - if (rate > user_rate) - continue; - if (ieee80211softmac_ratesinfo_rate_supported(&mac->ratesinfo, rate)) - return rate; - } - - /* If we haven't found a suitable rate by now, just trust the user */ - return user_rate; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_highest_supported_rate); - -void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac, - u8 erp_value) -{ - int use_protection; - int short_preamble; - u32 changes = 0; - - /* Barker preamble mode */ - short_preamble = ((erp_value & WLAN_ERP_BARKER_PREAMBLE) == 0 - && mac->associnfo.short_preamble_available) ? 1 : 0; - - /* Protection needed? */ - use_protection = (erp_value & WLAN_ERP_USE_PROTECTION) != 0; - - if (mac->bssinfo.short_preamble != short_preamble) { - changes |= IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE; - mac->bssinfo.short_preamble = short_preamble; - } - - if (mac->bssinfo.use_protection != use_protection) { - changes |= IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION; - mac->bssinfo.use_protection = use_protection; - } - - if (mac->bssinfo_change && changes) - mac->bssinfo_change(mac->dev, changes); -} - -void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac) -{ - struct ieee80211softmac_txrates *txrates = &mac->txrates; - u32 change = 0; - - change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT; - txrates->default_rate = ieee80211softmac_highest_supported_rate(mac, &mac->bssinfo.supported_rates, 0); - - change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK; - txrates->default_fallback = lower_rate(mac, txrates->default_rate); - - change |= IEEE80211SOFTMAC_TXRATECHG_MCAST; - txrates->mcast_rate = ieee80211softmac_highest_supported_rate(mac, &mac->bssinfo.supported_rates, 1); - - if (mac->txrates_change) - mac->txrates_change(mac->dev, change); - -} - -void ieee80211softmac_init_bss(struct ieee80211softmac_device *mac) -{ - struct ieee80211_device *ieee = mac->ieee; - u32 change = 0; - struct ieee80211softmac_txrates *txrates = &mac->txrates; - struct ieee80211softmac_bss_info *bssinfo = &mac->bssinfo; - - /* TODO: We need some kind of state machine to lower the default rates - * if we loose too many packets. - */ - /* Change the default txrate to the highest possible value. - * The txrate machine will lower it, if it is too high. - */ - if (ieee->modulation & IEEE80211_OFDM_MODULATION) - txrates->user_rate = IEEE80211_OFDM_RATE_24MB; - else - txrates->user_rate = IEEE80211_CCK_RATE_11MB; - - txrates->default_rate = IEEE80211_CCK_RATE_1MB; - change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT; - - txrates->default_fallback = IEEE80211_CCK_RATE_1MB; - change |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK; - - txrates->mcast_rate = IEEE80211_CCK_RATE_1MB; - change |= IEEE80211SOFTMAC_TXRATECHG_MCAST; - - txrates->mgt_mcast_rate = IEEE80211_CCK_RATE_1MB; - change |= IEEE80211SOFTMAC_TXRATECHG_MGT_MCAST; - - if (mac->txrates_change) - mac->txrates_change(mac->dev, change); - - change = 0; - - bssinfo->supported_rates.count = 0; - memset(bssinfo->supported_rates.rates, 0, - sizeof(bssinfo->supported_rates.rates)); - change |= IEEE80211SOFTMAC_BSSINFOCHG_RATES; - - bssinfo->short_preamble = 0; - change |= IEEE80211SOFTMAC_BSSINFOCHG_SHORT_PREAMBLE; - - bssinfo->use_protection = 0; - change |= IEEE80211SOFTMAC_BSSINFOCHG_PROTECTION; - - if (mac->bssinfo_change) - mac->bssinfo_change(mac->dev, change); - - mac->running = 1; -} - -void ieee80211softmac_start(struct net_device *dev) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - - ieee80211softmac_start_check_rates(mac); - ieee80211softmac_init_bss(mac); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_start); - -void ieee80211softmac_stop(struct net_device *dev) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - - ieee80211softmac_clear_pending_work(mac); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_stop); - -void ieee80211softmac_set_rates(struct net_device *dev, u8 count, u8 *rates) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - unsigned long flags; - - spin_lock_irqsave(&mac->lock, flags); - memcpy(mac->ratesinfo.rates, rates, count); - mac->ratesinfo.count = count; - spin_unlock_irqrestore(&mac->lock, flags); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_set_rates); - -static u8 raise_rate(struct ieee80211softmac_device *mac, u8 rate) -{ - int i; - struct ieee80211softmac_ratesinfo *ri = &mac->ratesinfo; - - for (i=0; icount-1; i++) { - if (ri->rates[i] == rate) - return ri->rates[i+1]; - } - /* I guess we can't go any higher... */ - return ri->rates[ri->count]; -} - -u8 ieee80211softmac_lower_rate_delta(struct ieee80211softmac_device *mac, u8 rate, int delta) -{ - int i; - struct ieee80211softmac_ratesinfo *ri = &mac->ratesinfo; - - for (i=delta; icount; i++) { - if (ri->rates[i] == rate) - return ri->rates[i-delta]; - } - /* I guess we can't go any lower... */ - return ri->rates[0]; -} - -static void ieee80211softmac_add_txrates_badness(struct ieee80211softmac_device *mac, - int amount) -{ - u8 default_rate = mac->txrates.default_rate; - u8 default_fallback = mac->txrates.default_fallback; - u32 changes = 0; - - //TODO: This is highly experimental code. - // Maybe the dynamic rate selection does not work - // and it has to be removed again. - -printk("badness %d\n", mac->txrate_badness); - mac->txrate_badness += amount; - if (mac->txrate_badness <= -1000) { - /* Very small badness. Try a faster bitrate. */ - default_rate = raise_rate(mac, default_rate); - changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT; - default_fallback = get_fallback_rate(mac, default_rate); - changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK; - mac->txrate_badness = 0; -printk("Bitrate raised to %u\n", default_rate); - } else if (mac->txrate_badness >= 10000) { - /* Very high badness. Try a slower bitrate. */ - default_rate = lower_rate(mac, default_rate); - changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT; - default_fallback = get_fallback_rate(mac, default_rate); - changes |= IEEE80211SOFTMAC_TXRATECHG_DEFAULT_FBACK; - mac->txrate_badness = 0; -printk("Bitrate lowered to %u\n", default_rate); - } - - mac->txrates.default_rate = default_rate; - mac->txrates.default_fallback = default_fallback; - - if (changes && mac->txrates_change) - mac->txrates_change(mac->dev, changes); -} - -void ieee80211softmac_fragment_lost(struct net_device *dev, - u16 wl_seq) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - unsigned long flags; - - spin_lock_irqsave(&mac->lock, flags); - ieee80211softmac_add_txrates_badness(mac, 1000); - //TODO - - spin_unlock_irqrestore(&mac->lock, flags); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_fragment_lost); - -static int rate_cmp(const void *a_, const void *b_) { - u8 *a, *b; - a = (u8*)a_; - b = (u8*)b_; - return ((*a & ~IEEE80211_BASIC_RATE_MASK) - (*b & ~IEEE80211_BASIC_RATE_MASK)); -} - -/* Allocate a softmac network struct and fill it from a network */ -struct ieee80211softmac_network * -ieee80211softmac_create_network(struct ieee80211softmac_device *mac, - struct ieee80211_network *net) -{ - struct ieee80211softmac_network *softnet; - softnet = kzalloc(sizeof(struct ieee80211softmac_network), GFP_ATOMIC); - if(softnet == NULL) - return NULL; - memcpy(softnet->bssid, net->bssid, ETH_ALEN); - softnet->channel = net->channel; - softnet->essid.len = net->ssid_len; - memcpy(softnet->essid.data, net->ssid, softnet->essid.len); - - /* copy rates over */ - softnet->supported_rates.count = net->rates_len; - memcpy(&softnet->supported_rates.rates[0], net->rates, net->rates_len); - memcpy(&softnet->supported_rates.rates[softnet->supported_rates.count], net->rates_ex, net->rates_ex_len); - softnet->supported_rates.count += net->rates_ex_len; - sort(softnet->supported_rates.rates, softnet->supported_rates.count, sizeof(softnet->supported_rates.rates[0]), rate_cmp, NULL); - - /* we save the ERP value because it is needed at association time, and - * many AP's do not include an ERP IE in the association response. */ - softnet->erp_value = net->erp_value; - - softnet->capabilities = net->capability; - return softnet; -} - - -/* Add a network to the list, while locked */ -void -ieee80211softmac_add_network_locked(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *add_net) -{ - struct ieee80211softmac_network *softmac_net; - - list_for_each_entry(softmac_net, &mac->network_list, list) { - if(!memcmp(softmac_net->bssid, add_net->bssid, ETH_ALEN)) - return; - } - list_add(&(add_net->list), &mac->network_list); -} - -/* Add a network to the list, with locking */ -void -ieee80211softmac_add_network(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *add_net) -{ - unsigned long flags; - spin_lock_irqsave(&mac->lock, flags); - ieee80211softmac_add_network_locked(mac, add_net); - spin_unlock_irqrestore(&mac->lock, flags); -} - - -/* Delete a network from the list, while locked*/ -void -ieee80211softmac_del_network_locked(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *del_net) -{ - list_del(&(del_net->list)); -} - -/* Delete a network from the list with locking */ -void -ieee80211softmac_del_network(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *del_net) -{ - unsigned long flags; - spin_lock_irqsave(&mac->lock, flags); - ieee80211softmac_del_network_locked(mac, del_net); - spin_unlock_irqrestore(&mac->lock, flags); -} - -/* Get a network from the list by MAC while locked */ -struct ieee80211softmac_network * -ieee80211softmac_get_network_by_bssid_locked(struct ieee80211softmac_device *mac, - u8 *bssid) -{ - struct ieee80211softmac_network *softmac_net; - - list_for_each_entry(softmac_net, &mac->network_list, list) { - if(!memcmp(softmac_net->bssid, bssid, ETH_ALEN)) - return softmac_net; - } - return NULL; -} - -/* Get a network from the list by BSSID with locking */ -struct ieee80211softmac_network * -ieee80211softmac_get_network_by_bssid(struct ieee80211softmac_device *mac, - u8 *bssid) -{ - unsigned long flags; - struct ieee80211softmac_network *softmac_net; - - spin_lock_irqsave(&mac->lock, flags); - softmac_net = ieee80211softmac_get_network_by_bssid_locked(mac, bssid); - spin_unlock_irqrestore(&mac->lock, flags); - return softmac_net; -} - -/* Get a network from the list by ESSID while locked */ -struct ieee80211softmac_network * -ieee80211softmac_get_network_by_essid_locked(struct ieee80211softmac_device *mac, - struct ieee80211softmac_essid *essid) -{ - struct ieee80211softmac_network *softmac_net; - - list_for_each_entry(softmac_net, &mac->network_list, list) { - if (softmac_net->essid.len == essid->len && - !memcmp(softmac_net->essid.data, essid->data, essid->len)) - return softmac_net; - } - return NULL; -} - -/* Get a network from the list by ESSID with locking */ -struct ieee80211softmac_network * -ieee80211softmac_get_network_by_essid(struct ieee80211softmac_device *mac, - struct ieee80211softmac_essid *essid) -{ - unsigned long flags; - struct ieee80211softmac_network *softmac_net = NULL; - - spin_lock_irqsave(&mac->lock, flags); - softmac_net = ieee80211softmac_get_network_by_essid_locked(mac, essid); - spin_unlock_irqrestore(&mac->lock, flags); - return softmac_net; -} - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Johannes Berg"); -MODULE_AUTHOR("Joseph Jezak"); -MODULE_AUTHOR("Larry Finger"); -MODULE_AUTHOR("Danny van Dyk"); -MODULE_AUTHOR("Michael Buesch"); -MODULE_DESCRIPTION("802.11 software MAC"); diff --git a/net/ieee80211/softmac/ieee80211softmac_priv.h b/net/ieee80211/softmac/ieee80211softmac_priv.h deleted file mode 100644 index f9232c8f6bdc..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_priv.h +++ /dev/null @@ -1,244 +0,0 @@ -/* - * Internal softmac API definitions. - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#ifndef IEEE80211SOFTMAC_PRIV_H_ -#define IEEE80211SOFTMAC_PRIV_H_ - -#include -#include -#include -#include - - -#define PFX "SoftMAC: " - -#ifdef assert -# undef assert -#endif -#ifdef CONFIG_IEEE80211_SOFTMAC_DEBUG -#define assert(expr) \ - do { \ - if (unlikely(!(expr))) { \ - printkl(KERN_ERR PFX "ASSERTION FAILED (%s) at: %s:%d:%s()\n", #expr, \ - __FILE__, __LINE__, __func__); \ - } \ - } while (0) -#else -#define assert(expr) do {} while (0) -#endif - -/* rate limited printk(). */ -#ifdef printkl -# undef printkl -#endif -#define printkl(f, x...) do { if (printk_ratelimit()) printk(f ,##x); } while (0) -/* rate limited printk() for debugging */ -#ifdef dprintkl -# undef dprintkl -#endif -#ifdef CONFIG_IEEE80211_SOFTMAC_DEBUG -# define dprintkl printkl -#else -# define dprintkl(f, x...) do { /* nothing */ } while (0) -#endif - -/* debugging printk() */ -#ifdef dprintk -# undef dprintk -#endif -#ifdef CONFIG_IEEE80211_SOFTMAC_DEBUG -# define dprintk(f, x...) do { printk(f ,##x); } while (0) -#else -# define dprintk(f, x...) do { /* nothing */ } while (0) -#endif - -/* private definitions and prototypes */ - -/*** prototypes from _scan.c */ -void ieee80211softmac_scan(struct work_struct *work); -/* for internal use if scanning is needed */ -int ieee80211softmac_start_scan(struct ieee80211softmac_device *mac); -void ieee80211softmac_stop_scan(struct ieee80211softmac_device *mac); -void ieee80211softmac_wait_for_scan(struct ieee80211softmac_device *mac); - -/* for use by _module.c to assign to the callbacks */ -int ieee80211softmac_start_scan_implementation(struct net_device *dev); -void ieee80211softmac_stop_scan_implementation(struct net_device *dev); -void ieee80211softmac_wait_for_scan_implementation(struct net_device *dev); - -/*** Network prototypes from _module.c */ -struct ieee80211softmac_network * ieee80211softmac_create_network( - struct ieee80211softmac_device *mac, struct ieee80211_network *net); -void ieee80211softmac_add_network_locked(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net); -void ieee80211softmac_add_network(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net); -void ieee80211softmac_del_network_locked(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net); -void ieee80211softmac_del_network(struct ieee80211softmac_device *mac, - struct ieee80211softmac_network *net); -struct ieee80211softmac_network * ieee80211softmac_get_network_by_bssid_locked( - struct ieee80211softmac_device *mac, u8 *ea); -struct ieee80211softmac_network * ieee80211softmac_get_network_by_bssid( - struct ieee80211softmac_device *mac, u8 *ea); -struct ieee80211softmac_network * ieee80211softmac_get_network_by_ssid_locked( - struct ieee80211softmac_device *mac, u8 *ssid, u8 ssid_len); -struct ieee80211softmac_network * ieee80211softmac_get_network_by_ssid( - struct ieee80211softmac_device *mac, u8 *ssid, u8 ssid_len); -struct ieee80211softmac_network * -ieee80211softmac_get_network_by_essid_locked(struct ieee80211softmac_device *mac, - struct ieee80211softmac_essid *essid); -struct ieee80211softmac_network * -ieee80211softmac_get_network_by_essid(struct ieee80211softmac_device *mac, - struct ieee80211softmac_essid *essid); - -/* Rates related */ -void ieee80211softmac_process_erp(struct ieee80211softmac_device *mac, - u8 erp_value); -int ieee80211softmac_ratesinfo_rate_supported(struct ieee80211softmac_ratesinfo *ri, u8 rate); -u8 ieee80211softmac_lower_rate_delta(struct ieee80211softmac_device *mac, u8 rate, int delta); -void ieee80211softmac_init_bss(struct ieee80211softmac_device *mac); -void ieee80211softmac_recalc_txrates(struct ieee80211softmac_device *mac); -static inline u8 lower_rate(struct ieee80211softmac_device *mac, u8 rate) { - return ieee80211softmac_lower_rate_delta(mac, rate, 1); -} - -static inline u8 get_fallback_rate(struct ieee80211softmac_device *mac, u8 rate) -{ - return ieee80211softmac_lower_rate_delta(mac, rate, 2); -} - - -/*** prototypes from _io.c */ -int ieee80211softmac_send_mgt_frame(struct ieee80211softmac_device *mac, - void* ptrarg, u32 type, u32 arg); -int ieee80211softmac_handle_beacon(struct net_device *dev, - struct ieee80211_beacon *beacon, - struct ieee80211_network *network); - -/*** prototypes from _auth.c */ -/* do these have to go into the public header? */ -int ieee80211softmac_auth_req(struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net); -int ieee80211softmac_deauth_req(struct ieee80211softmac_device *mac, struct ieee80211softmac_network *net, int reason); - -/* for use by _module.c to assign to the callbacks */ -int ieee80211softmac_auth_resp(struct net_device *dev, struct ieee80211_auth *auth); -int ieee80211softmac_deauth_resp(struct net_device *dev, struct ieee80211_deauth *deauth); - -/*** prototypes from _assoc.c */ -void ieee80211softmac_assoc_work(struct work_struct *work); -int ieee80211softmac_handle_assoc_response(struct net_device * dev, - struct ieee80211_assoc_response * resp, - struct ieee80211_network * network); -int ieee80211softmac_handle_disassoc(struct net_device * dev, - struct ieee80211_disassoc * disassoc); -int ieee80211softmac_handle_reassoc_req(struct net_device * dev, - struct ieee80211_reassoc_request * reassoc); -void ieee80211softmac_assoc_timeout(struct work_struct *work); -void ieee80211softmac_send_disassoc_req(struct ieee80211softmac_device *mac, u16 reason); -void ieee80211softmac_disassoc(struct ieee80211softmac_device *mac); - -/* some helper functions */ -static inline int ieee80211softmac_scan_handlers_check_self(struct ieee80211softmac_device *sm) -{ - return (sm->start_scan == ieee80211softmac_start_scan_implementation) && - (sm->stop_scan == ieee80211softmac_stop_scan_implementation) && - (sm->wait_for_scan == ieee80211softmac_wait_for_scan_implementation); -} - -static inline int ieee80211softmac_scan_sanity_check(struct ieee80211softmac_device *sm) -{ - return ((sm->start_scan != ieee80211softmac_start_scan_implementation) && - (sm->stop_scan != ieee80211softmac_stop_scan_implementation) && - (sm->wait_for_scan != ieee80211softmac_wait_for_scan_implementation) - ) || ieee80211softmac_scan_handlers_check_self(sm); -} - -#define IEEE80211SOFTMAC_PROBE_DELAY HZ/50 -#define IEEE80211SOFTMAC_WORKQUEUE_NAME_LEN (17 + IFNAMSIZ) - -struct ieee80211softmac_network { - struct list_head list; /* List */ - /* Network information copied from ieee80211_network */ - u8 bssid[ETH_ALEN]; - u8 channel; - struct ieee80211softmac_essid essid; - - struct ieee80211softmac_ratesinfo supported_rates; - - /* SoftMAC specific */ - u16 authenticating:1, /* Status Flags */ - authenticated:1, - auth_desynced_once:1; - - u8 erp_value; /* Saved ERP value */ - u16 capabilities; /* Capabilities bitfield */ - u8 challenge_len; /* Auth Challenge length */ - char *challenge; /* Challenge Text */ -}; - -/* structure used to keep track of networks we're auth'ing to */ -struct ieee80211softmac_auth_queue_item { - struct list_head list; /* List head */ - struct ieee80211softmac_network *net; /* Network to auth */ - struct ieee80211softmac_device *mac; /* SoftMAC device */ - u8 retry; /* Retry limit */ - u8 state; /* Auth State */ - struct delayed_work work; /* Work queue */ -}; - -/* scanning information */ -struct ieee80211softmac_scaninfo { - u8 current_channel_idx, - number_channels; - struct ieee80211_channel *channels; - u8 started:1, - stop:1; - u8 skip_flags; - struct completion finished; - struct delayed_work softmac_scan; - struct ieee80211softmac_device *mac; -}; - -/* private event struct */ -struct ieee80211softmac_event { - struct list_head list; - int event_type; - void *event_context; - struct delayed_work work; - notify_function_ptr fun; - void *context; - struct ieee80211softmac_device *mac; -}; - -void ieee80211softmac_call_events(struct ieee80211softmac_device *mac, int event, void *event_context); -void ieee80211softmac_call_events_locked(struct ieee80211softmac_device *mac, int event, void *event_context); -int ieee80211softmac_notify_internal(struct ieee80211softmac_device *mac, - int event, void *event_context, notify_function_ptr fun, void *context, gfp_t gfp_mask); - -void ieee80211softmac_try_reassoc(struct ieee80211softmac_device *mac); - -#endif /* IEEE80211SOFTMAC_PRIV_H_ */ diff --git a/net/ieee80211/softmac/ieee80211softmac_scan.c b/net/ieee80211/softmac/ieee80211softmac_scan.c deleted file mode 100644 index bfab8d7db88f..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_scan.c +++ /dev/null @@ -1,254 +0,0 @@ -/* - * Scanning routines. - * - * These are not exported because they're assigned to the function pointers. - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#include -#include "ieee80211softmac_priv.h" - -/* internal, use to trigger scanning if needed. - * Returns -EBUSY if already scanning, - * result of start_scan otherwise */ -int -ieee80211softmac_start_scan(struct ieee80211softmac_device *sm) -{ - unsigned long flags; - int ret; - - spin_lock_irqsave(&sm->lock, flags); - if (sm->scanning) - { - spin_unlock_irqrestore(&sm->lock, flags); - return -EINPROGRESS; - } - sm->scanning = 1; - spin_unlock_irqrestore(&sm->lock, flags); - - ret = sm->start_scan(sm->dev); - if (ret) { - spin_lock_irqsave(&sm->lock, flags); - sm->scanning = 0; - spin_unlock_irqrestore(&sm->lock, flags); - } - return ret; -} - -void -ieee80211softmac_stop_scan(struct ieee80211softmac_device *sm) -{ - unsigned long flags; - - spin_lock_irqsave(&sm->lock, flags); - - if (!sm->scanning) { - spin_unlock_irqrestore(&sm->lock, flags); - return; - } - - spin_unlock_irqrestore(&sm->lock, flags); - sm->stop_scan(sm->dev); -} - -void -ieee80211softmac_wait_for_scan(struct ieee80211softmac_device *sm) -{ - unsigned long flags; - - spin_lock_irqsave(&sm->lock, flags); - - if (!sm->scanning) { - spin_unlock_irqrestore(&sm->lock, flags); - return; - } - - spin_unlock_irqrestore(&sm->lock, flags); - sm->wait_for_scan(sm->dev); -} - - -/* internal scanning implementation follows */ -void ieee80211softmac_scan(struct work_struct *work) -{ - int invalid_channel; - u8 current_channel_idx; - struct ieee80211softmac_scaninfo *si = - container_of(work, struct ieee80211softmac_scaninfo, - softmac_scan.work); - struct ieee80211softmac_device *sm = si->mac; - unsigned long flags; - - while (!(si->stop) && (si->current_channel_idx < si->number_channels)) { - current_channel_idx = si->current_channel_idx; - si->current_channel_idx++; /* go to the next channel */ - - invalid_channel = (si->skip_flags & si->channels[current_channel_idx].flags); - - if (!invalid_channel) { - sm->set_channel(sm->dev, si->channels[current_channel_idx].channel); - // FIXME make this user configurable (active/passive) - if(ieee80211softmac_send_mgt_frame(sm, NULL, IEEE80211_STYPE_PROBE_REQ, 0)) - printkl(KERN_DEBUG PFX "Sending Probe Request Failed\n"); - - /* also send directed management frame for the network we're looking for */ - // TODO: is this if correct, or should we do this only if scanning from assoc request? - if (sm->associnfo.req_essid.len) - ieee80211softmac_send_mgt_frame(sm, &sm->associnfo.req_essid, IEEE80211_STYPE_PROBE_REQ, 0); - - spin_lock_irqsave(&sm->lock, flags); - if (unlikely(!sm->running)) { - /* Prevent reschedule on workqueue flush */ - spin_unlock_irqrestore(&sm->lock, flags); - break; - } - queue_delayed_work(sm->wq, &si->softmac_scan, IEEE80211SOFTMAC_PROBE_DELAY); - spin_unlock_irqrestore(&sm->lock, flags); - return; - } else { - dprintk(PFX "Not probing Channel %d (not allowed here)\n", si->channels[current_channel_idx].channel); - } - } - - spin_lock_irqsave(&sm->lock, flags); - cancel_delayed_work(&si->softmac_scan); - si->started = 0; - spin_unlock_irqrestore(&sm->lock, flags); - - dprintk(PFX "Scanning finished: scanned %d channels starting with channel %d\n", - sm->scaninfo->number_channels, sm->scaninfo->channels[0].channel); - ieee80211softmac_scan_finished(sm); - complete_all(&sm->scaninfo->finished); -} - -static inline struct ieee80211softmac_scaninfo *allocate_scaninfo(struct ieee80211softmac_device *mac) -{ - /* ugh. can we call this without having the spinlock held? */ - struct ieee80211softmac_scaninfo *info = kmalloc(sizeof(struct ieee80211softmac_scaninfo), GFP_ATOMIC); - if (unlikely(!info)) - return NULL; - INIT_DELAYED_WORK(&info->softmac_scan, ieee80211softmac_scan); - info->mac = mac; - init_completion(&info->finished); - return info; -} - -int ieee80211softmac_start_scan_implementation(struct net_device *dev) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(dev); - unsigned long flags; - - if (!(dev->flags & IFF_UP)) - return -ENODEV; - - assert(ieee80211softmac_scan_handlers_check_self(sm)); - if (!ieee80211softmac_scan_handlers_check_self(sm)) - return -EINVAL; - - spin_lock_irqsave(&sm->lock, flags); - /* it looks like we need to hold the lock here - * to make sure we don't allocate two of these... */ - if (unlikely(!sm->scaninfo)) - sm->scaninfo = allocate_scaninfo(sm); - if (unlikely(!sm->scaninfo)) { - spin_unlock_irqrestore(&sm->lock, flags); - return -ENOMEM; - } - - sm->scaninfo->skip_flags = IEEE80211_CH_INVALID; - if (0 /* not scanning in IEEE802.11b */)//TODO - sm->scaninfo->skip_flags |= IEEE80211_CH_B_ONLY; - if (0 /* IEEE802.11a */) {//TODO - sm->scaninfo->channels = sm->ieee->geo.a; - sm->scaninfo->number_channels = sm->ieee->geo.a_channels; - } else { - sm->scaninfo->channels = sm->ieee->geo.bg; - sm->scaninfo->number_channels = sm->ieee->geo.bg_channels; - } - sm->scaninfo->current_channel_idx = 0; - sm->scaninfo->started = 1; - sm->scaninfo->stop = 0; - INIT_COMPLETION(sm->scaninfo->finished); - queue_delayed_work(sm->wq, &sm->scaninfo->softmac_scan, 0); - spin_unlock_irqrestore(&sm->lock, flags); - return 0; -} - -void ieee80211softmac_stop_scan_implementation(struct net_device *dev) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(dev); - unsigned long flags; - - assert(ieee80211softmac_scan_handlers_check_self(sm)); - if (!ieee80211softmac_scan_handlers_check_self(sm)) - return; - - spin_lock_irqsave(&sm->lock, flags); - assert(sm->scaninfo != NULL); - if (sm->scaninfo) { - if (sm->scaninfo->started) - sm->scaninfo->stop = 1; - else - complete_all(&sm->scaninfo->finished); - } - spin_unlock_irqrestore(&sm->lock, flags); -} - -void ieee80211softmac_wait_for_scan_implementation(struct net_device *dev) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(dev); - unsigned long flags; - - assert(ieee80211softmac_scan_handlers_check_self(sm)); - if (!ieee80211softmac_scan_handlers_check_self(sm)) - return; - - spin_lock_irqsave(&sm->lock, flags); - if (!sm->scaninfo->started) { - spin_unlock_irqrestore(&sm->lock, flags); - return; - } - spin_unlock_irqrestore(&sm->lock, flags); - wait_for_completion(&sm->scaninfo->finished); -} - -/* this is what drivers (that do scanning) call when they're done */ -void ieee80211softmac_scan_finished(struct ieee80211softmac_device *sm) -{ - unsigned long flags; - - spin_lock_irqsave(&sm->lock, flags); - sm->scanning = 0; - spin_unlock_irqrestore(&sm->lock, flags); - - if (sm->associnfo.bssvalid) { - struct ieee80211softmac_network *net; - - net = ieee80211softmac_get_network_by_bssid(sm, sm->associnfo.bssid); - if (net) - sm->set_channel(sm->dev, net->channel); - } - ieee80211softmac_call_events(sm, IEEE80211SOFTMAC_EVENT_SCAN_FINISHED, NULL); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_scan_finished); diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c deleted file mode 100644 index e01b59aedc54..000000000000 --- a/net/ieee80211/softmac/ieee80211softmac_wx.c +++ /dev/null @@ -1,508 +0,0 @@ -/* - * This file contains our _wx handlers. Make sure you EXPORT_SYMBOL_GPL them - * - * Copyright (c) 2005, 2006 Johannes Berg - * Joseph Jezak - * Larry Finger - * Danny van Dyk - * Michael Buesch - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program 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 this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - * The full GNU General Public License is included in this distribution in the - * file called COPYING. - */ - -#include "ieee80211softmac_priv.h" - -#include -/* for is_broadcast_ether_addr and is_zero_ether_addr */ -#include - -int -ieee80211softmac_wx_trigger_scan(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(net_dev); - return ieee80211softmac_start_scan(sm); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_trigger_scan); - - -/* if we're still scanning, return -EAGAIN so that userspace tools - * can get the complete scan results, otherwise return 0. */ -int -ieee80211softmac_wx_get_scan_results(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - unsigned long flags; - struct ieee80211softmac_device *sm = ieee80211_priv(net_dev); - - spin_lock_irqsave(&sm->lock, flags); - if (sm->scanning) { - spin_unlock_irqrestore(&sm->lock, flags); - return -EAGAIN; - } - spin_unlock_irqrestore(&sm->lock, flags); - return ieee80211_wx_get_scan(sm->ieee, info, data, extra); -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_scan_results); - -int -ieee80211softmac_wx_set_essid(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(net_dev); - struct ieee80211softmac_auth_queue_item *authptr; - int length = 0; - DECLARE_MAC_BUF(mac); - -check_assoc_again: - mutex_lock(&sm->associnfo.mutex); - if((sm->associnfo.associating || sm->associnfo.associated) && - (data->essid.flags && data->essid.length)) { - dprintk(KERN_INFO PFX "Canceling existing associate request!\n"); - /* Cancel assoc work */ - cancel_delayed_work(&sm->associnfo.work); - /* We don't have to do this, but it's a little cleaner */ - list_for_each_entry(authptr, &sm->auth_queue, list) - cancel_delayed_work(&authptr->work); - sm->associnfo.bssvalid = 0; - sm->associnfo.bssfixed = 0; - sm->associnfo.associating = 0; - sm->associnfo.associated = 0; - /* We must unlock to avoid deadlocks with the assoc workqueue - * on the associnfo.mutex */ - mutex_unlock(&sm->associnfo.mutex); - flush_workqueue(sm->wq); - /* Avoid race! Check assoc status again. Maybe someone started an - * association while we flushed. */ - goto check_assoc_again; - } - - sm->associnfo.static_essid = 0; - sm->associnfo.assoc_wait = 0; - - if (data->essid.flags && data->essid.length) { - length = min((int)data->essid.length, IW_ESSID_MAX_SIZE); - if (length) { - memcpy(sm->associnfo.req_essid.data, extra, length); - sm->associnfo.static_essid = 1; - } - } - - /* set our requested ESSID length. - * If applicable, we have already copied the data in */ - sm->associnfo.req_essid.len = length; - - sm->associnfo.associating = 1; - /* queue lower level code to do work (if necessary) */ - queue_delayed_work(sm->wq, &sm->associnfo.work, 0); - - mutex_unlock(&sm->associnfo.mutex); - - return 0; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_essid); - -int -ieee80211softmac_wx_get_essid(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *sm = ieee80211_priv(net_dev); - - mutex_lock(&sm->associnfo.mutex); - /* If all fails, return ANY (empty) */ - data->essid.length = 0; - data->essid.flags = 0; /* active */ - - /* If we have a statically configured ESSID then return it */ - if (sm->associnfo.static_essid) { - data->essid.length = sm->associnfo.req_essid.len; - data->essid.flags = 1; /* active */ - memcpy(extra, sm->associnfo.req_essid.data, sm->associnfo.req_essid.len); - dprintk(KERN_INFO PFX "Getting essid from req_essid\n"); - } else if (sm->associnfo.associated || sm->associnfo.associating) { - /* If we're associating/associated, return that */ - data->essid.length = sm->associnfo.associate_essid.len; - data->essid.flags = 1; /* active */ - memcpy(extra, sm->associnfo.associate_essid.data, sm->associnfo.associate_essid.len); - dprintk(KERN_INFO PFX "Getting essid from associate_essid\n"); - } - mutex_unlock(&sm->associnfo.mutex); - - return 0; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_essid); - -int -ieee80211softmac_wx_set_rate(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(net_dev); - struct ieee80211_device *ieee = mac->ieee; - unsigned long flags; - s32 in_rate = data->bitrate.value; - u8 rate; - int is_ofdm = 0; - int err = -EINVAL; - - if (in_rate == -1) { - if (ieee->modulation & IEEE80211_OFDM_MODULATION) - in_rate = 24000000; - else - in_rate = 11000000; - } - - switch (in_rate) { - case 1000000: - rate = IEEE80211_CCK_RATE_1MB; - break; - case 2000000: - rate = IEEE80211_CCK_RATE_2MB; - break; - case 5500000: - rate = IEEE80211_CCK_RATE_5MB; - break; - case 11000000: - rate = IEEE80211_CCK_RATE_11MB; - break; - case 6000000: - rate = IEEE80211_OFDM_RATE_6MB; - is_ofdm = 1; - break; - case 9000000: - rate = IEEE80211_OFDM_RATE_9MB; - is_ofdm = 1; - break; - case 12000000: - rate = IEEE80211_OFDM_RATE_12MB; - is_ofdm = 1; - break; - case 18000000: - rate = IEEE80211_OFDM_RATE_18MB; - is_ofdm = 1; - break; - case 24000000: - rate = IEEE80211_OFDM_RATE_24MB; - is_ofdm = 1; - break; - case 36000000: - rate = IEEE80211_OFDM_RATE_36MB; - is_ofdm = 1; - break; - case 48000000: - rate = IEEE80211_OFDM_RATE_48MB; - is_ofdm = 1; - break; - case 54000000: - rate = IEEE80211_OFDM_RATE_54MB; - is_ofdm = 1; - break; - default: - goto out; - } - - spin_lock_irqsave(&mac->lock, flags); - - /* Check if correct modulation for this PHY. */ - if (is_ofdm && !(ieee->modulation & IEEE80211_OFDM_MODULATION)) - goto out_unlock; - - mac->txrates.user_rate = rate; - ieee80211softmac_recalc_txrates(mac); - err = 0; - -out_unlock: - spin_unlock_irqrestore(&mac->lock, flags); -out: - return err; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_rate); - -int -ieee80211softmac_wx_get_rate(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(net_dev); - unsigned long flags; - int err = -EINVAL; - - spin_lock_irqsave(&mac->lock, flags); - - if (unlikely(!mac->running)) { - err = -ENODEV; - goto out_unlock; - } - - switch (mac->txrates.default_rate) { - case IEEE80211_CCK_RATE_1MB: - data->bitrate.value = 1000000; - break; - case IEEE80211_CCK_RATE_2MB: - data->bitrate.value = 2000000; - break; - case IEEE80211_CCK_RATE_5MB: - data->bitrate.value = 5500000; - break; - case IEEE80211_CCK_RATE_11MB: - data->bitrate.value = 11000000; - break; - case IEEE80211_OFDM_RATE_6MB: - data->bitrate.value = 6000000; - break; - case IEEE80211_OFDM_RATE_9MB: - data->bitrate.value = 9000000; - break; - case IEEE80211_OFDM_RATE_12MB: - data->bitrate.value = 12000000; - break; - case IEEE80211_OFDM_RATE_18MB: - data->bitrate.value = 18000000; - break; - case IEEE80211_OFDM_RATE_24MB: - data->bitrate.value = 24000000; - break; - case IEEE80211_OFDM_RATE_36MB: - data->bitrate.value = 36000000; - break; - case IEEE80211_OFDM_RATE_48MB: - data->bitrate.value = 48000000; - break; - case IEEE80211_OFDM_RATE_54MB: - data->bitrate.value = 54000000; - break; - default: - assert(0); - goto out_unlock; - } - err = 0; -out_unlock: - spin_unlock_irqrestore(&mac->lock, flags); - - return err; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_rate); - -int -ieee80211softmac_wx_get_wap(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(net_dev); - int err = 0; - - mutex_lock(&mac->associnfo.mutex); - if (mac->associnfo.bssvalid) - memcpy(data->ap_addr.sa_data, mac->associnfo.bssid, ETH_ALEN); - else - memset(data->ap_addr.sa_data, 0xff, ETH_ALEN); - data->ap_addr.sa_family = ARPHRD_ETHER; - mutex_unlock(&mac->associnfo.mutex); - - return err; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_wap); - -int -ieee80211softmac_wx_set_wap(struct net_device *net_dev, - struct iw_request_info *info, - union iwreq_data *data, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(net_dev); - - /* sanity check */ - if (data->ap_addr.sa_family != ARPHRD_ETHER) { - return -EINVAL; - } - - mutex_lock(&mac->associnfo.mutex); - if (is_broadcast_ether_addr(data->ap_addr.sa_data)) { - /* the bssid we have is not to be fixed any longer, - * and we should reassociate to the best AP. */ - mac->associnfo.bssfixed = 0; - /* force reassociation */ - mac->associnfo.bssvalid = 0; - if (mac->associnfo.associated) - queue_delayed_work(mac->wq, &mac->associnfo.work, 0); - } else if (is_zero_ether_addr(data->ap_addr.sa_data)) { - /* the bssid we have is no longer fixed */ - mac->associnfo.bssfixed = 0; - } else { - if (!memcmp(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN)) { - if (mac->associnfo.associating || mac->associnfo.associated) { - /* bssid unchanged and associated or associating - just return */ - goto out; - } - } else { - /* copy new value in data->ap_addr.sa_data to bssid */ - memcpy(mac->associnfo.bssid, data->ap_addr.sa_data, ETH_ALEN); - } - /* tell the other code that this bssid should be used no matter what */ - mac->associnfo.bssfixed = 1; - /* queue associate if new bssid or (old one again and not associated) */ - queue_delayed_work(mac->wq, &mac->associnfo.work, 0); - } - - out: - mutex_unlock(&mac->associnfo.mutex); - - return 0; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_wap); - -int -ieee80211softmac_wx_set_genie(struct net_device *dev, - struct iw_request_info *info, - union iwreq_data *wrqu, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - unsigned long flags; - int err = 0; - char *buf; - int i; - - mutex_lock(&mac->associnfo.mutex); - spin_lock_irqsave(&mac->lock, flags); - /* bleh. shouldn't be locked for that kmalloc... */ - - if (wrqu->data.length) { - if ((wrqu->data.length < 2) || (extra[1]+2 != wrqu->data.length)) { - /* this is an IE, so the length must be - * correct. Is it possible though that - * more than one IE is passed in? - */ - err = -EINVAL; - goto out; - } - if (mac->wpa.IEbuflen <= wrqu->data.length) { - buf = kmalloc(wrqu->data.length, GFP_ATOMIC); - if (!buf) { - err = -ENOMEM; - goto out; - } - kfree(mac->wpa.IE); - mac->wpa.IE = buf; - mac->wpa.IEbuflen = wrqu->data.length; - } - memcpy(mac->wpa.IE, extra, wrqu->data.length); - dprintk(KERN_INFO PFX "generic IE set to "); - for (i=0;idata.length;i++) - dprintk("%.2x", (u8)mac->wpa.IE[i]); - dprintk("\n"); - mac->wpa.IElen = wrqu->data.length; - } else { - kfree(mac->wpa.IE); - mac->wpa.IE = NULL; - mac->wpa.IElen = 0; - mac->wpa.IEbuflen = 0; - } - - out: - spin_unlock_irqrestore(&mac->lock, flags); - mutex_unlock(&mac->associnfo.mutex); - - return err; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_genie); - -int -ieee80211softmac_wx_get_genie(struct net_device *dev, - struct iw_request_info *info, - union iwreq_data *wrqu, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - unsigned long flags; - int err = 0; - int space = wrqu->data.length; - - mutex_lock(&mac->associnfo.mutex); - spin_lock_irqsave(&mac->lock, flags); - - wrqu->data.length = 0; - - if (mac->wpa.IE && mac->wpa.IElen) { - wrqu->data.length = mac->wpa.IElen; - if (mac->wpa.IElen <= space) - memcpy(extra, mac->wpa.IE, mac->wpa.IElen); - else - err = -E2BIG; - } - spin_unlock_irqrestore(&mac->lock, flags); - mutex_unlock(&mac->associnfo.mutex); - - return err; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_get_genie); - -int -ieee80211softmac_wx_set_mlme(struct net_device *dev, - struct iw_request_info *info, - union iwreq_data *wrqu, - char *extra) -{ - struct ieee80211softmac_device *mac = ieee80211_priv(dev); - struct iw_mlme *mlme = (struct iw_mlme *)extra; - u16 reason = mlme->reason_code; - struct ieee80211softmac_network *net; - int err = -EINVAL; - - mutex_lock(&mac->associnfo.mutex); - - if (memcmp(mac->associnfo.bssid, mlme->addr.sa_data, ETH_ALEN)) { - printk(KERN_DEBUG PFX "wx_set_mlme: requested operation on net we don't use\n"); - goto out; - } - - switch (mlme->cmd) { - case IW_MLME_DEAUTH: - net = ieee80211softmac_get_network_by_bssid_locked(mac, mlme->addr.sa_data); - if (!net) { - printk(KERN_DEBUG PFX "wx_set_mlme: we should know the net here...\n"); - goto out; - } - err = ieee80211softmac_deauth_req(mac, net, reason); - goto out; - case IW_MLME_DISASSOC: - ieee80211softmac_send_disassoc_req(mac, reason); - mac->associnfo.associated = 0; - mac->associnfo.associating = 0; - err = 0; - goto out; - default: - err = -EOPNOTSUPP; - } - -out: - mutex_unlock(&mac->associnfo.mutex); - - return err; -} -EXPORT_SYMBOL_GPL(ieee80211softmac_wx_set_mlme); -- cgit v1.2.3 From 068edceb7e73c05f77e204442ea8f86e238575da Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 9 Mar 2008 16:55:10 -0700 Subject: include/net/ieee80211.h - remove duplicate include Signed-off-by: Joe Perches Signed-off-by: John W. Linville --- include/net/ieee80211.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/net/ieee80211.h b/include/net/ieee80211.h index 285b2adfa648..529816bfbc52 100644 --- a/include/net/ieee80211.h +++ b/include/net/ieee80211.h @@ -183,7 +183,6 @@ const char *escape_essid(const char *essid, u8 essid_len); #define IEEE80211_DEBUG_RX(f, a...) IEEE80211_DEBUG(IEEE80211_DL_RX, f, ## a) #define IEEE80211_DEBUG_QOS(f, a...) IEEE80211_DEBUG(IEEE80211_DL_QOS, f, ## a) #include -#include #include /* ARPHRD_ETHER */ #ifndef WIRELESS_SPY -- cgit v1.2.3 From e7ec2e3230633a858af1b0b359f6c4670dbeb997 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Mon, 10 Mar 2008 17:26:32 +0100 Subject: ssb: Add SPROM/invariants support for PCMCIA devices This adds support for reading/writing the SPROM invariants for PCMCIA based devices. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/ssb/Kconfig | 6 + drivers/ssb/Makefile | 1 + drivers/ssb/main.c | 23 +- drivers/ssb/pci.c | 113 +--------- drivers/ssb/pcmcia.c | 518 ++++++++++++++++++++++++++++++++++++++++------ drivers/ssb/sprom.c | 133 ++++++++++++ drivers/ssb/ssb_private.h | 17 ++ include/linux/ssb/ssb.h | 4 +- 8 files changed, 648 insertions(+), 167 deletions(-) create mode 100644 drivers/ssb/sprom.c (limited to 'include') diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig index f69ef0ba2613..0f7cce2560d1 100644 --- a/drivers/ssb/Kconfig +++ b/drivers/ssb/Kconfig @@ -20,6 +20,10 @@ config SSB If unsure, say N. +# Common SPROM support routines +config SSB_SPROM + bool + config SSB_PCIHOST_POSSIBLE bool depends on SSB && (PCI = y || PCI = SSB) @@ -28,6 +32,7 @@ config SSB_PCIHOST_POSSIBLE config SSB_PCIHOST bool "Support for SSB on PCI-bus host" depends on SSB_PCIHOST_POSSIBLE + select SSB_SPROM default y help Support for a Sonics Silicon Backplane on top @@ -48,6 +53,7 @@ config SSB_PCMCIAHOST_POSSIBLE config SSB_PCMCIAHOST bool "Support for SSB on PCMCIA-bus host (EXPERIMENTAL)" depends on SSB_PCMCIAHOST_POSSIBLE + select SSB_SPROM help Support for a Sonics Silicon Backplane on top of a PCMCIA device. diff --git a/drivers/ssb/Makefile b/drivers/ssb/Makefile index 910f35e32fc9..6f255e9c5af9 100644 --- a/drivers/ssb/Makefile +++ b/drivers/ssb/Makefile @@ -1,6 +1,7 @@ # core ssb-y += main.o scan.o ssb-$(CONFIG_SSB_EMBEDDED) += embedded.o +ssb-$(CONFIG_SSB_SPROM) += sprom.o # host support ssb-$(CONFIG_SSB_PCIHOST) += pci.o pcihost_wrapper.o diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c index 49d7bbb9bea7..e12371916444 100644 --- a/drivers/ssb/main.c +++ b/drivers/ssb/main.c @@ -69,6 +69,25 @@ found: } #endif /* CONFIG_SSB_PCIHOST */ +#ifdef CONFIG_SSB_PCMCIAHOST +struct ssb_bus *ssb_pcmcia_dev_to_bus(struct pcmcia_device *pdev) +{ + struct ssb_bus *bus; + + ssb_buses_lock(); + list_for_each_entry(bus, &buses, list) { + if (bus->bustype == SSB_BUSTYPE_PCMCIA && + bus->host_pcmcia == pdev) + goto found; + } + bus = NULL; +found: + ssb_buses_unlock(); + + return bus; +} +#endif /* CONFIG_SSB_PCMCIAHOST */ + int ssb_for_each_bus_call(unsigned long data, int (*func)(struct ssb_bus *bus, unsigned long data)) { @@ -398,7 +417,7 @@ void ssb_bus_unregister(struct ssb_bus *bus) list_del(&bus->list); ssb_buses_unlock(); - /* ssb_pcmcia_exit(bus); */ + ssb_pcmcia_exit(bus); ssb_pci_exit(bus); ssb_iounmap(bus); } @@ -663,7 +682,7 @@ out: err_dequeue: list_del(&bus->list); err_pcmcia_exit: -/* ssb_pcmcia_exit(bus); */ + ssb_pcmcia_exit(bus); err_pci_exit: ssb_pci_exit(bus); err_unmap: diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c index 1facc7620fc8..f1514b33cfae 100644 --- a/drivers/ssb/pci.c +++ b/drivers/ssb/pci.c @@ -227,7 +227,7 @@ static u8 ssb_sprom_crc(const u16 *sprom, u16 size) return crc; } -static int sprom_check_crc(const u16 *sprom, u16 size) +static int sprom_check_crc(const u16 *sprom, size_t size) { u8 crc; u8 expected_crc; @@ -242,12 +242,14 @@ static int sprom_check_crc(const u16 *sprom, u16 size) return 0; } -static void sprom_do_read(struct ssb_bus *bus, u16 *sprom) +static int sprom_do_read(struct ssb_bus *bus, u16 *sprom) { int i; for (i = 0; i < bus->sprom_size; i++) sprom[i] = ioread16(bus->mmio + SSB_SPROM_BASE + (i * 2)); + + return 0; } static int sprom_do_write(struct ssb_bus *bus, const u16 *sprom) @@ -660,71 +662,18 @@ const struct ssb_bus_ops ssb_pci_ops = { .write32 = ssb_pci_write32, }; -static int sprom2hex(const u16 *sprom, char *buf, size_t buf_len, u16 size) -{ - int i, pos = 0; - - for (i = 0; i < size; i++) - pos += snprintf(buf + pos, buf_len - pos - 1, - "%04X", swab16(sprom[i]) & 0xFFFF); - pos += snprintf(buf + pos, buf_len - pos - 1, "\n"); - - return pos + 1; -} - -static int hex2sprom(u16 *sprom, const char *dump, size_t len, u16 size) -{ - char tmp[5] = { 0 }; - int cnt = 0; - unsigned long parsed; - - if (len < size * 2) - return -EINVAL; - - while (cnt < size) { - memcpy(tmp, dump, 4); - dump += 4; - parsed = simple_strtoul(tmp, NULL, 16); - sprom[cnt++] = swab16((u16)parsed); - } - - return 0; -} - static ssize_t ssb_pci_attr_sprom_show(struct device *pcidev, struct device_attribute *attr, char *buf) { struct pci_dev *pdev = container_of(pcidev, struct pci_dev, dev); struct ssb_bus *bus; - u16 *sprom; - int err = -ENODEV; - ssize_t count = 0; bus = ssb_pci_dev_to_bus(pdev); if (!bus) - goto out; - err = -ENOMEM; - sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL); - if (!sprom) - goto out; + return -ENODEV; - /* Use interruptible locking, as the SPROM write might - * be holding the lock for several seconds. So allow userspace - * to cancel operation. */ - err = -ERESTARTSYS; - if (mutex_lock_interruptible(&bus->pci_sprom_mutex)) - goto out_kfree; - sprom_do_read(bus, sprom); - mutex_unlock(&bus->pci_sprom_mutex); - - count = sprom2hex(sprom, buf, PAGE_SIZE, bus->sprom_size); - err = 0; - -out_kfree: - kfree(sprom); -out: - return err ? err : count; + return ssb_attr_sprom_show(bus, buf, sprom_do_read); } static ssize_t ssb_pci_attr_sprom_store(struct device *pcidev, @@ -733,55 +682,13 @@ static ssize_t ssb_pci_attr_sprom_store(struct device *pcidev, { struct pci_dev *pdev = container_of(pcidev, struct pci_dev, dev); struct ssb_bus *bus; - u16 *sprom; - int res = 0, err = -ENODEV; bus = ssb_pci_dev_to_bus(pdev); if (!bus) - goto out; - err = -ENOMEM; - sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL); - if (!sprom) - goto out; - err = hex2sprom(sprom, buf, count, bus->sprom_size); - if (err) { - err = -EINVAL; - goto out_kfree; - } - err = sprom_check_crc(sprom, bus->sprom_size); - if (err) { - err = -EINVAL; - goto out_kfree; - } + return -ENODEV; - /* Use interruptible locking, as the SPROM write might - * be holding the lock for several seconds. So allow userspace - * to cancel operation. */ - err = -ERESTARTSYS; - if (mutex_lock_interruptible(&bus->pci_sprom_mutex)) - goto out_kfree; - err = ssb_devices_freeze(bus); - if (err == -EOPNOTSUPP) { - ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze devices. " - "No suspend support. Is CONFIG_PM enabled?\n"); - goto out_unlock; - } - if (err) { - ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze all devices\n"); - goto out_unlock; - } - res = sprom_do_write(bus, sprom); - err = ssb_devices_thaw(bus); - if (err) - ssb_printk(KERN_ERR PFX "SPROM write: Could not thaw all devices\n"); -out_unlock: - mutex_unlock(&bus->pci_sprom_mutex); -out_kfree: - kfree(sprom); -out: - if (res) - return res; - return err ? err : count; + return ssb_attr_sprom_store(bus, buf, count, + sprom_check_crc, sprom_do_write); } static DEVICE_ATTR(ssb_sprom, 0600, @@ -808,7 +715,7 @@ int ssb_pci_init(struct ssb_bus *bus) return 0; pdev = bus->host_pci; - mutex_init(&bus->pci_sprom_mutex); + mutex_init(&bus->sprom_mutex); err = device_create_file(&pdev->dev, &dev_attr_ssb_sprom); if (err) goto out; diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c index 84b3a845a8a8..cd49f7c65531 100644 --- a/drivers/ssb/pcmcia.c +++ b/drivers/ssb/pcmcia.c @@ -3,7 +3,7 @@ * PCMCIA-Hostbus related functions * * Copyright 2006 Johannes Berg - * Copyright 2007 Michael Buesch + * Copyright 2007-2008 Michael Buesch * * Licensed under the GNU/GPL. See COPYING for details. */ @@ -11,6 +11,7 @@ #include #include #include +#include #include #include @@ -26,59 +27,132 @@ #define SSB_VERBOSE_PCMCIACORESWITCH_DEBUG 0 +/* PCMCIA configuration registers */ +#define SSB_PCMCIA_CORECTL 0x00 +#define SSB_PCMCIA_CORECTL_RESET 0x80 /* Core reset */ +#define SSB_PCMCIA_CORECTL_IRQEN 0x04 /* IRQ enable */ +#define SSB_PCMCIA_CORECTL_FUNCEN 0x01 /* Function enable */ +#define SSB_PCMCIA_CORECTL2 0x80 +#define SSB_PCMCIA_ADDRESS0 0x2E +#define SSB_PCMCIA_ADDRESS1 0x30 +#define SSB_PCMCIA_ADDRESS2 0x32 +#define SSB_PCMCIA_MEMSEG 0x34 +#define SSB_PCMCIA_SPROMCTL 0x36 +#define SSB_PCMCIA_SPROMCTL_IDLE 0 +#define SSB_PCMCIA_SPROMCTL_WRITE 1 +#define SSB_PCMCIA_SPROMCTL_READ 2 +#define SSB_PCMCIA_SPROMCTL_WRITEEN 4 +#define SSB_PCMCIA_SPROMCTL_WRITEDIS 7 +#define SSB_PCMCIA_SPROMCTL_DONE 8 +#define SSB_PCMCIA_SPROM_DATALO 0x38 +#define SSB_PCMCIA_SPROM_DATAHI 0x3A +#define SSB_PCMCIA_SPROM_ADDRLO 0x3C +#define SSB_PCMCIA_SPROM_ADDRHI 0x3E + +/* Hardware invariants CIS tuples */ +#define SSB_PCMCIA_CIS 0x80 +#define SSB_PCMCIA_CIS_ID 0x01 +#define SSB_PCMCIA_CIS_BOARDREV 0x02 +#define SSB_PCMCIA_CIS_PA 0x03 +#define SSB_PCMCIA_CIS_PA_PA0B0_LO 0 +#define SSB_PCMCIA_CIS_PA_PA0B0_HI 1 +#define SSB_PCMCIA_CIS_PA_PA0B1_LO 2 +#define SSB_PCMCIA_CIS_PA_PA0B1_HI 3 +#define SSB_PCMCIA_CIS_PA_PA0B2_LO 4 +#define SSB_PCMCIA_CIS_PA_PA0B2_HI 5 +#define SSB_PCMCIA_CIS_PA_ITSSI 6 +#define SSB_PCMCIA_CIS_PA_MAXPOW 7 +#define SSB_PCMCIA_CIS_OEMNAME 0x04 +#define SSB_PCMCIA_CIS_CCODE 0x05 +#define SSB_PCMCIA_CIS_ANTENNA 0x06 +#define SSB_PCMCIA_CIS_ANTGAIN 0x07 +#define SSB_PCMCIA_CIS_BFLAGS 0x08 +#define SSB_PCMCIA_CIS_LEDS 0x09 + +/* PCMCIA SPROM size. */ +#define SSB_PCMCIA_SPROM_SIZE 256 +#define SSB_PCMCIA_SPROM_SIZE_BYTES (SSB_PCMCIA_SPROM_SIZE * sizeof(u16)) + + +/* Write to a PCMCIA configuration register. */ +static int ssb_pcmcia_cfg_write(struct ssb_bus *bus, u8 offset, u8 value) +{ + conf_reg_t reg; + int res; + + memset(®, 0, sizeof(reg)); + reg.Offset = offset; + reg.Action = CS_WRITE; + reg.Value = value; + res = pcmcia_access_configuration_register(bus->host_pcmcia, ®); + if (unlikely(res != CS_SUCCESS)) + return -EBUSY; + + return 0; +} + +/* Read from a PCMCIA configuration register. */ +static int ssb_pcmcia_cfg_read(struct ssb_bus *bus, u8 offset, u8 *value) +{ + conf_reg_t reg; + int res; + + memset(®, 0, sizeof(reg)); + reg.Offset = offset; + reg.Action = CS_READ; + res = pcmcia_access_configuration_register(bus->host_pcmcia, ®); + if (unlikely(res != CS_SUCCESS)) + return -EBUSY; + *value = reg.Value; + + return 0; +} + int ssb_pcmcia_switch_coreidx(struct ssb_bus *bus, u8 coreidx) { - struct pcmcia_device *pdev = bus->host_pcmcia; int err; int attempts = 0; u32 cur_core; - conf_reg_t reg; u32 addr; u32 read_addr; + u8 val; addr = (coreidx * SSB_CORE_SIZE) + SSB_ENUM_BASE; while (1) { - reg.Action = CS_WRITE; - reg.Offset = 0x2E; - reg.Value = (addr & 0x0000F000) >> 12; - err = pcmcia_access_configuration_register(pdev, ®); - if (err != CS_SUCCESS) + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_ADDRESS0, + (addr & 0x0000F000) >> 12); + if (err) goto error; - reg.Offset = 0x30; - reg.Value = (addr & 0x00FF0000) >> 16; - err = pcmcia_access_configuration_register(pdev, ®); - if (err != CS_SUCCESS) + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_ADDRESS1, + (addr & 0x00FF0000) >> 16); + if (err) goto error; - reg.Offset = 0x32; - reg.Value = (addr & 0xFF000000) >> 24; - err = pcmcia_access_configuration_register(pdev, ®); - if (err != CS_SUCCESS) + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_ADDRESS2, + (addr & 0xFF000000) >> 24); + if (err) goto error; read_addr = 0; - reg.Action = CS_READ; - reg.Offset = 0x2E; - err = pcmcia_access_configuration_register(pdev, ®); - if (err != CS_SUCCESS) + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_ADDRESS0, &val); + if (err) goto error; - read_addr |= ((u32)(reg.Value & 0x0F)) << 12; - reg.Offset = 0x30; - err = pcmcia_access_configuration_register(pdev, ®); - if (err != CS_SUCCESS) + read_addr |= ((u32)(val & 0x0F)) << 12; + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_ADDRESS1, &val); + if (err) goto error; - read_addr |= ((u32)reg.Value) << 16; - reg.Offset = 0x32; - err = pcmcia_access_configuration_register(pdev, ®); - if (err != CS_SUCCESS) + read_addr |= ((u32)val) << 16; + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_ADDRESS2, &val); + if (err) goto error; - read_addr |= ((u32)reg.Value) << 24; + read_addr |= ((u32)val) << 24; cur_core = (read_addr - SSB_ENUM_BASE) / SSB_CORE_SIZE; if (cur_core == coreidx) break; + err = -ETIMEDOUT; if (attempts++ > SSB_BAR0_MAX_RETRIES) goto error; udelay(10); @@ -87,7 +161,7 @@ int ssb_pcmcia_switch_coreidx(struct ssb_bus *bus, return 0; error: ssb_printk(KERN_ERR PFX "Failed to switch to core %u\n", coreidx); - return -ENODEV; + return err; } int ssb_pcmcia_switch_core(struct ssb_bus *bus, @@ -112,27 +186,21 @@ int ssb_pcmcia_switch_core(struct ssb_bus *bus, int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg) { int attempts = 0; - conf_reg_t reg; - int res; + int err; + u8 val; SSB_WARN_ON((seg != 0) && (seg != 1)); - reg.Offset = 0x34; - reg.Function = 0; while (1) { - reg.Action = CS_WRITE; - reg.Value = seg; - res = pcmcia_access_configuration_register(bus->host_pcmcia, ®); - if (unlikely(res != CS_SUCCESS)) + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_MEMSEG, seg); + if (err) goto error; - reg.Value = 0xFF; - reg.Action = CS_READ; - res = pcmcia_access_configuration_register(bus->host_pcmcia, ®); - if (unlikely(res != CS_SUCCESS)) + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_MEMSEG, &val); + if (err) goto error; - - if (reg.Value == seg) + if (val == seg) break; + err = -ETIMEDOUT; if (unlikely(attempts++ > SSB_BAR0_MAX_RETRIES)) goto error; udelay(10); @@ -142,7 +210,7 @@ int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg) return 0; error: ssb_printk(KERN_ERR PFX "Failed to switch pcmcia segment\n"); - return -ENODEV; + return err; } static int select_core_and_segment(struct ssb_device *dev, @@ -276,18 +344,344 @@ const struct ssb_bus_ops ssb_pcmcia_ops = { .write32 = ssb_pcmcia_write32, }; -#include +static int ssb_pcmcia_sprom_command(struct ssb_bus *bus, u8 command) +{ + unsigned int i; + int err; + u8 value; + + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROMCTL, command); + if (err) + return err; + for (i = 0; i < 1000; i++) { + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_SPROMCTL, &value); + if (err) + return err; + if (value & SSB_PCMCIA_SPROMCTL_DONE) + return 0; + udelay(10); + } + + return -ETIMEDOUT; +} + +/* offset is the 16bit word offset */ +static int ssb_pcmcia_sprom_read(struct ssb_bus *bus, u16 offset, u16 *value) +{ + int err; + u8 lo, hi; + + offset *= 2; /* Make byte offset */ + + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRLO, + (offset & 0x00FF)); + if (err) + return err; + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRHI, + (offset & 0xFF00) >> 8); + if (err) + return err; + err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_READ); + if (err) + return err; + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_SPROM_DATALO, &lo); + if (err) + return err; + err = ssb_pcmcia_cfg_read(bus, SSB_PCMCIA_SPROM_DATAHI, &hi); + if (err) + return err; + *value = (lo | (((u16)hi) << 8)); + + return 0; +} + +/* offset is the 16bit word offset */ +static int ssb_pcmcia_sprom_write(struct ssb_bus *bus, u16 offset, u16 value) +{ + int err; + + offset *= 2; /* Make byte offset */ + + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRLO, + (offset & 0x00FF)); + if (err) + return err; + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_ADDRHI, + (offset & 0xFF00) >> 8); + if (err) + return err; + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_DATALO, + (value & 0x00FF)); + if (err) + return err; + err = ssb_pcmcia_cfg_write(bus, SSB_PCMCIA_SPROM_DATAHI, + (value & 0xFF00) >> 8); + if (err) + return err; + err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_WRITE); + if (err) + return err; + msleep(20); + + return 0; +} + +/* Read the SPROM image. bufsize is in 16bit words. */ +static int ssb_pcmcia_sprom_read_all(struct ssb_bus *bus, u16 *sprom) +{ + int err, i; + + for (i = 0; i < SSB_PCMCIA_SPROM_SIZE; i++) { + err = ssb_pcmcia_sprom_read(bus, i, &sprom[i]); + if (err) + return err; + } + + return 0; +} + +/* Write the SPROM image. size is in 16bit words. */ +static int ssb_pcmcia_sprom_write_all(struct ssb_bus *bus, const u16 *sprom) +{ + int i, err; + bool failed = 0; + size_t size = SSB_PCMCIA_SPROM_SIZE; + + ssb_printk(KERN_NOTICE PFX + "Writing SPROM. Do NOT turn off the power! " + "Please stand by...\n"); + err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_WRITEEN); + if (err) { + ssb_printk(KERN_NOTICE PFX + "Could not enable SPROM write access.\n"); + return -EBUSY; + } + ssb_printk(KERN_NOTICE PFX "[ 0%%"); + msleep(500); + for (i = 0; i < size; i++) { + if (i == size / 4) + ssb_printk("25%%"); + else if (i == size / 2) + ssb_printk("50%%"); + else if (i == (size * 3) / 4) + ssb_printk("75%%"); + else if (i % 2) + ssb_printk("."); + err = ssb_pcmcia_sprom_write(bus, i, sprom[i]); + if (err) { + ssb_printk("\n" KERN_NOTICE PFX + "Failed to write to SPROM.\n"); + failed = 1; + break; + } + } + err = ssb_pcmcia_sprom_command(bus, SSB_PCMCIA_SPROMCTL_WRITEDIS); + if (err) { + ssb_printk("\n" KERN_NOTICE PFX + "Could not disable SPROM write access.\n"); + failed = 1; + } + msleep(500); + if (!failed) { + ssb_printk("100%% ]\n"); + ssb_printk(KERN_NOTICE PFX "SPROM written.\n"); + } + + return failed ? -EBUSY : 0; +} + +static int ssb_pcmcia_sprom_check_crc(const u16 *sprom, size_t size) +{ + //TODO + return 0; +} + +#define GOTO_ERROR_ON(condition, description) do { \ + if (unlikely(condition)) { \ + error_description = description; \ + goto error; \ + } \ + } while (0) + int ssb_pcmcia_get_invariants(struct ssb_bus *bus, struct ssb_init_invariants *iv) { - //TODO - random_ether_addr(iv->sprom.il0mac); + tuple_t tuple; + int res; + unsigned char buf[32]; + struct ssb_sprom *sprom = &iv->sprom; + struct ssb_boardinfo *bi = &iv->boardinfo; + const char *error_description; + + memset(sprom, 0xFF, sizeof(*sprom)); + sprom->revision = 1; + sprom->boardflags_lo = 0; + sprom->boardflags_hi = 0; + + /* First fetch the MAC address. */ + memset(&tuple, 0, sizeof(tuple)); + tuple.DesiredTuple = CISTPL_FUNCE; + tuple.TupleData = buf; + tuple.TupleDataMax = sizeof(buf); + res = pcmcia_get_first_tuple(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "MAC first tpl"); + res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "MAC first tpl data"); + while (1) { + GOTO_ERROR_ON(tuple.TupleDataLen < 1, "MAC tpl < 1"); + if (tuple.TupleData[0] == CISTPL_FUNCE_LAN_NODE_ID) + break; + res = pcmcia_get_next_tuple(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "MAC next tpl"); + res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "MAC next tpl data"); + } + GOTO_ERROR_ON(tuple.TupleDataLen != ETH_ALEN + 2, "MAC tpl size"); + memcpy(sprom->il0mac, &tuple.TupleData[2], ETH_ALEN); + + /* Fetch the vendor specific tuples. */ + memset(&tuple, 0, sizeof(tuple)); + tuple.DesiredTuple = SSB_PCMCIA_CIS; + tuple.TupleData = buf; + tuple.TupleDataMax = sizeof(buf); + res = pcmcia_get_first_tuple(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "VEN first tpl"); + res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "VEN first tpl data"); + while (1) { + GOTO_ERROR_ON(tuple.TupleDataLen < 1, "VEN tpl < 1"); + switch (tuple.TupleData[0]) { + case SSB_PCMCIA_CIS_ID: + GOTO_ERROR_ON((tuple.TupleDataLen != 5) && + (tuple.TupleDataLen != 7), + "id tpl size"); + bi->vendor = tuple.TupleData[1] | + ((u16)tuple.TupleData[2] << 8); + break; + case SSB_PCMCIA_CIS_BOARDREV: + GOTO_ERROR_ON(tuple.TupleDataLen != 2, + "boardrev tpl size"); + sprom->board_rev = tuple.TupleData[1]; + break; + case SSB_PCMCIA_CIS_PA: + GOTO_ERROR_ON(tuple.TupleDataLen != 9, + "pa tpl size"); + sprom->pa0b0 = tuple.TupleData[1] | + ((u16)tuple.TupleData[2] << 8); + sprom->pa0b1 = tuple.TupleData[3] | + ((u16)tuple.TupleData[4] << 8); + sprom->pa0b2 = tuple.TupleData[5] | + ((u16)tuple.TupleData[6] << 8); + sprom->itssi_a = tuple.TupleData[7]; + sprom->itssi_bg = tuple.TupleData[7]; + sprom->maxpwr_a = tuple.TupleData[8]; + sprom->maxpwr_bg = tuple.TupleData[8]; + break; + case SSB_PCMCIA_CIS_OEMNAME: + /* We ignore this. */ + break; + case SSB_PCMCIA_CIS_CCODE: + GOTO_ERROR_ON(tuple.TupleDataLen != 2, + "ccode tpl size"); + sprom->country_code = tuple.TupleData[1]; + break; + case SSB_PCMCIA_CIS_ANTENNA: + GOTO_ERROR_ON(tuple.TupleDataLen != 2, + "ant tpl size"); + sprom->ant_available_a = tuple.TupleData[1]; + sprom->ant_available_bg = tuple.TupleData[1]; + break; + case SSB_PCMCIA_CIS_ANTGAIN: + GOTO_ERROR_ON(tuple.TupleDataLen != 2, + "antg tpl size"); + sprom->antenna_gain.ghz24.a0 = tuple.TupleData[1]; + sprom->antenna_gain.ghz24.a1 = tuple.TupleData[1]; + sprom->antenna_gain.ghz24.a2 = tuple.TupleData[1]; + sprom->antenna_gain.ghz24.a3 = tuple.TupleData[1]; + sprom->antenna_gain.ghz5.a0 = tuple.TupleData[1]; + sprom->antenna_gain.ghz5.a1 = tuple.TupleData[1]; + sprom->antenna_gain.ghz5.a2 = tuple.TupleData[1]; + sprom->antenna_gain.ghz5.a3 = tuple.TupleData[1]; + break; + case SSB_PCMCIA_CIS_BFLAGS: + GOTO_ERROR_ON(tuple.TupleDataLen != 3, + "bfl tpl size"); + sprom->boardflags_lo = tuple.TupleData[1] | + ((u16)tuple.TupleData[2] << 8); + break; + case SSB_PCMCIA_CIS_LEDS: + GOTO_ERROR_ON(tuple.TupleDataLen != 5, + "leds tpl size"); + sprom->gpio0 = tuple.TupleData[1]; + sprom->gpio1 = tuple.TupleData[2]; + sprom->gpio2 = tuple.TupleData[3]; + sprom->gpio3 = tuple.TupleData[4]; + break; + } + res = pcmcia_get_next_tuple(bus->host_pcmcia, &tuple); + if (res == CS_NO_MORE_ITEMS) + break; + GOTO_ERROR_ON(res != CS_SUCCESS, "VEN next tpl"); + res = pcmcia_get_tuple_data(bus->host_pcmcia, &tuple); + GOTO_ERROR_ON(res != CS_SUCCESS, "VEN next tpl data"); + } + return 0; +error: + ssb_printk(KERN_ERR PFX + "PCMCIA: Failed to fetch device invariants: %s\n", + error_description); + return -ENODEV; +} + +static ssize_t ssb_pcmcia_attr_sprom_show(struct device *pcmciadev, + struct device_attribute *attr, + char *buf) +{ + struct pcmcia_device *pdev = + container_of(pcmciadev, struct pcmcia_device, dev); + struct ssb_bus *bus; + + bus = ssb_pcmcia_dev_to_bus(pdev); + if (!bus) + return -ENODEV; + + return ssb_attr_sprom_show(bus, buf, + ssb_pcmcia_sprom_read_all); +} + +static ssize_t ssb_pcmcia_attr_sprom_store(struct device *pcmciadev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct pcmcia_device *pdev = + container_of(pcmciadev, struct pcmcia_device, dev); + struct ssb_bus *bus; + + bus = ssb_pcmcia_dev_to_bus(pdev); + if (!bus) + return -ENODEV; + + return ssb_attr_sprom_store(bus, buf, count, + ssb_pcmcia_sprom_check_crc, + ssb_pcmcia_sprom_write_all); +} + +static DEVICE_ATTR(ssb_sprom, 0600, + ssb_pcmcia_attr_sprom_show, + ssb_pcmcia_attr_sprom_store); + +void ssb_pcmcia_exit(struct ssb_bus *bus) +{ + if (bus->bustype != SSB_BUSTYPE_PCMCIA) + return; + + device_remove_file(&bus->host_pcmcia->dev, &dev_attr_ssb_sprom); } int ssb_pcmcia_init(struct ssb_bus *bus) { - conf_reg_t reg; + u8 val, offset; int err; if (bus->bustype != SSB_BUSTYPE_PCMCIA) @@ -298,22 +692,26 @@ int ssb_pcmcia_init(struct ssb_bus *bus) ssb_pcmcia_switch_segment(bus, 0); /* Init IRQ routing */ - reg.Action = CS_READ; - reg.Function = 0; if (bus->chip_id == 0x4306) - reg.Offset = 0x00; + offset = SSB_PCMCIA_CORECTL; else - reg.Offset = 0x80; - err = pcmcia_access_configuration_register(bus->host_pcmcia, ®); - if (err != CS_SUCCESS) + offset = SSB_PCMCIA_CORECTL2; + err = ssb_pcmcia_cfg_read(bus, offset, &val); + if (err) goto error; - reg.Action = CS_WRITE; - reg.Value |= 0x04 | 0x01; - err = pcmcia_access_configuration_register(bus->host_pcmcia, ®); - if (err != CS_SUCCESS) + val |= SSB_PCMCIA_CORECTL_IRQEN | SSB_PCMCIA_CORECTL_FUNCEN; + err = ssb_pcmcia_cfg_write(bus, offset, val); + if (err) + goto error; + + bus->sprom_size = SSB_PCMCIA_SPROM_SIZE; + mutex_init(&bus->sprom_mutex); + err = device_create_file(&bus->host_pcmcia->dev, &dev_attr_ssb_sprom); + if (err) goto error; return 0; error: - return -ENODEV; + ssb_printk(KERN_ERR PFX "Failed to initialize PCMCIA host device\n"); + return err; } diff --git a/drivers/ssb/sprom.c b/drivers/ssb/sprom.c new file mode 100644 index 000000000000..3668edb39315 --- /dev/null +++ b/drivers/ssb/sprom.c @@ -0,0 +1,133 @@ +/* + * Sonics Silicon Backplane + * Common SPROM support routines + * + * Copyright (C) 2005-2008 Michael Buesch + * Copyright (C) 2005 Martin Langer + * Copyright (C) 2005 Stefano Brivio + * Copyright (C) 2005 Danny van Dyk + * Copyright (C) 2005 Andreas Jaggi + * + * Licensed under the GNU/GPL. See COPYING for details. + */ + +#include "ssb_private.h" + + +static int sprom2hex(const u16 *sprom, char *buf, size_t buf_len, + size_t sprom_size_words) +{ + int i, pos = 0; + + for (i = 0; i < sprom_size_words; i++) + pos += snprintf(buf + pos, buf_len - pos - 1, + "%04X", swab16(sprom[i]) & 0xFFFF); + pos += snprintf(buf + pos, buf_len - pos - 1, "\n"); + + return pos + 1; +} + +static int hex2sprom(u16 *sprom, const char *dump, size_t len, + size_t sprom_size_words) +{ + char tmp[5] = { 0 }; + int cnt = 0; + unsigned long parsed; + + if (len < sprom_size_words * 2) + return -EINVAL; + + while (cnt < sprom_size_words) { + memcpy(tmp, dump, 4); + dump += 4; + parsed = simple_strtoul(tmp, NULL, 16); + sprom[cnt++] = swab16((u16)parsed); + } + + return 0; +} + +/* Common sprom device-attribute show-handler */ +ssize_t ssb_attr_sprom_show(struct ssb_bus *bus, char *buf, + int (*sprom_read)(struct ssb_bus *bus, u16 *sprom)) +{ + u16 *sprom; + int err = -ENOMEM; + ssize_t count = 0; + size_t sprom_size_words = bus->sprom_size; + + sprom = kcalloc(sprom_size_words, sizeof(u16), GFP_KERNEL); + if (!sprom) + goto out; + + /* Use interruptible locking, as the SPROM write might + * be holding the lock for several seconds. So allow userspace + * to cancel operation. */ + err = -ERESTARTSYS; + if (mutex_lock_interruptible(&bus->sprom_mutex)) + goto out_kfree; + err = sprom_read(bus, sprom); + mutex_unlock(&bus->sprom_mutex); + + if (!err) + count = sprom2hex(sprom, buf, PAGE_SIZE, sprom_size_words); + +out_kfree: + kfree(sprom); +out: + return err ? err : count; +} + +/* Common sprom device-attribute store-handler */ +ssize_t ssb_attr_sprom_store(struct ssb_bus *bus, + const char *buf, size_t count, + int (*sprom_check_crc)(const u16 *sprom, size_t size), + int (*sprom_write)(struct ssb_bus *bus, const u16 *sprom)) +{ + u16 *sprom; + int res = 0, err = -ENOMEM; + size_t sprom_size_words = bus->sprom_size; + + sprom = kcalloc(bus->sprom_size, sizeof(u16), GFP_KERNEL); + if (!sprom) + goto out; + err = hex2sprom(sprom, buf, count, sprom_size_words); + if (err) { + err = -EINVAL; + goto out_kfree; + } + err = sprom_check_crc(sprom, sprom_size_words); + if (err) { + err = -EINVAL; + goto out_kfree; + } + + /* Use interruptible locking, as the SPROM write might + * be holding the lock for several seconds. So allow userspace + * to cancel operation. */ + err = -ERESTARTSYS; + if (mutex_lock_interruptible(&bus->sprom_mutex)) + goto out_kfree; + err = ssb_devices_freeze(bus); + if (err == -EOPNOTSUPP) { + ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze devices. " + "No suspend support. Is CONFIG_PM enabled?\n"); + goto out_unlock; + } + if (err) { + ssb_printk(KERN_ERR PFX "SPROM write: Could not freeze all devices\n"); + goto out_unlock; + } + res = sprom_write(bus, sprom); + err = ssb_devices_thaw(bus); + if (err) + ssb_printk(KERN_ERR PFX "SPROM write: Could not thaw all devices\n"); +out_unlock: + mutex_unlock(&bus->sprom_mutex); +out_kfree: + kfree(sprom); +out: + if (res) + return res; + return err ? err : count; +} diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h index d03b20983b1e..a83bf7a4d80b 100644 --- a/drivers/ssb/ssb_private.h +++ b/drivers/ssb/ssb_private.h @@ -81,6 +81,7 @@ extern int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg); extern int ssb_pcmcia_get_invariants(struct ssb_bus *bus, struct ssb_init_invariants *iv); +extern void ssb_pcmcia_exit(struct ssb_bus *bus); extern int ssb_pcmcia_init(struct ssb_bus *bus); extern const struct ssb_bus_ops ssb_pcmcia_ops; #else /* CONFIG_SSB_PCMCIAHOST */ @@ -99,6 +100,9 @@ static inline int ssb_pcmcia_switch_segment(struct ssb_bus *bus, { return 0; } +static inline void ssb_pcmcia_exit(struct ssb_bus *bus) +{ +} static inline int ssb_pcmcia_init(struct ssb_bus *bus) { return 0; @@ -113,6 +117,17 @@ extern int ssb_bus_scan(struct ssb_bus *bus, extern void ssb_iounmap(struct ssb_bus *ssb); +/* sprom.c */ +extern +ssize_t ssb_attr_sprom_show(struct ssb_bus *bus, char *buf, + int (*sprom_read)(struct ssb_bus *bus, u16 *sprom)); +extern +ssize_t ssb_attr_sprom_store(struct ssb_bus *bus, + const char *buf, size_t count, + int (*sprom_check_crc)(const u16 *sprom, size_t size), + int (*sprom_write)(struct ssb_bus *bus, const u16 *sprom)); + + /* core.c */ extern u32 ssb_calc_clock_rate(u32 plltype, u32 n, u32 m); extern int ssb_devices_freeze(struct ssb_bus *bus); @@ -120,6 +135,8 @@ extern int ssb_devices_thaw(struct ssb_bus *bus); extern struct ssb_bus *ssb_pci_dev_to_bus(struct pci_dev *pdev); int ssb_for_each_bus_call(unsigned long data, int (*func)(struct ssb_bus *bus, unsigned long data)); +extern struct ssb_bus *ssb_pcmcia_dev_to_bus(struct pcmcia_device *pdev); + /* b43_pci_bridge.c */ #ifdef CONFIG_SSB_B43_PCI_BRIDGE diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h index b7c388972fcf..8644e03cf588 100644 --- a/include/linux/ssb/ssb.h +++ b/include/linux/ssb/ssb.h @@ -245,9 +245,9 @@ struct ssb_bus { /* Pointer to the PCMCIA device (only if bustype == SSB_BUSTYPE_PCMCIA). */ struct pcmcia_device *host_pcmcia; -#ifdef CONFIG_SSB_PCIHOST +#ifdef CONFIG_SSB_SPROM /* Mutex to protect the SPROM writing. */ - struct mutex pci_sprom_mutex; + struct mutex sprom_mutex; #endif /* ID information about the Chip. */ -- cgit v1.2.3 From 3e94794355724f77dc6cbb5ad956f7c72d8313a4 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Fri, 22 Feb 2008 19:55:15 +0900 Subject: smc91x: introduce platform data flags V2 This patch introduces struct smc91x_platdata and modifies the driver so bus width is checked during run time using SMC_nBIT() instead of SMC_CAN_USE_nBIT. V2 keeps static configuration lean using SMC_DYNAMIC_BUS_CONFIG. Signed-off-by: Magnus Damm Acked-by: Nicolas Pitre Signed-off-by: Jeff Garzik --- drivers/net/smc91x.c | 34 +++++++++++++++++++++++++----- drivers/net/smc91x.h | 57 +++++++++++++++++++++++++++++++------------------- include/linux/smc91x.h | 13 ++++++++++++ 3 files changed, 77 insertions(+), 27 deletions(-) create mode 100644 include/linux/smc91x.h (limited to 'include') diff --git a/drivers/net/smc91x.c b/drivers/net/smc91x.c index d0ef80ae018a..97bdb2a43bc8 100644 --- a/drivers/net/smc91x.c +++ b/drivers/net/smc91x.c @@ -1997,6 +1997,8 @@ err_out: static int smc_enable_device(struct platform_device *pdev) { + struct net_device *ndev = platform_get_drvdata(pdev); + struct smc_local *lp = netdev_priv(ndev); unsigned long flags; unsigned char ecor, ecsr; void __iomem *addr; @@ -2039,7 +2041,7 @@ static int smc_enable_device(struct platform_device *pdev) * Set the appropriate byte/word mode. */ ecsr = readb(addr + (ECSR << SMC_IO_SHIFT)) & ~ECSR_IOIS8; - if (!SMC_CAN_USE_16BIT) + if (!SMC_16BIT(lp)) ecsr |= ECSR_IOIS8; writeb(ecsr, addr + (ECSR << SMC_IO_SHIFT)); local_irq_restore(flags); @@ -2124,10 +2126,11 @@ static void smc_release_datacs(struct platform_device *pdev, struct net_device * */ static int smc_drv_probe(struct platform_device *pdev) { + struct smc91x_platdata *pd = pdev->dev.platform_data; + struct smc_local *lp; struct net_device *ndev; struct resource *res, *ires; unsigned int __iomem *addr; - unsigned long irq_flags = SMC_IRQ_FLAGS; int ret; res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "smc91x-regs"); @@ -2152,6 +2155,27 @@ static int smc_drv_probe(struct platform_device *pdev) } SET_NETDEV_DEV(ndev, &pdev->dev); + /* get configuration from platform data, only allow use of + * bus width if both SMC_CAN_USE_xxx and SMC91X_USE_xxx are set. + */ + + lp = netdev_priv(ndev); + lp->cfg.irq_flags = SMC_IRQ_FLAGS; + +#ifdef SMC_DYNAMIC_BUS_CONFIG + if (pd) + memcpy(&lp->cfg, pd, sizeof(lp->cfg)); + else { + lp->cfg.flags = SMC91X_USE_8BIT; + lp->cfg.flags |= SMC91X_USE_16BIT; + lp->cfg.flags |= SMC91X_USE_32BIT; + } + + lp->cfg.flags &= ~(SMC_CAN_USE_8BIT ? 0 : SMC91X_USE_8BIT); + lp->cfg.flags &= ~(SMC_CAN_USE_16BIT ? 0 : SMC91X_USE_16BIT); + lp->cfg.flags &= ~(SMC_CAN_USE_32BIT ? 0 : SMC91X_USE_32BIT); +#endif + ndev->dma = (unsigned char)-1; ires = platform_get_resource(pdev, IORESOURCE_IRQ, 0); @@ -2162,7 +2186,7 @@ static int smc_drv_probe(struct platform_device *pdev) ndev->irq = ires->start; if (SMC_IRQ_FLAGS == -1) - irq_flags = ires->flags & IRQF_TRIGGER_MASK; + lp->cfg.irq_flags = ires->flags & IRQF_TRIGGER_MASK; ret = smc_request_attrib(pdev); if (ret) @@ -2170,6 +2194,7 @@ static int smc_drv_probe(struct platform_device *pdev) #if defined(CONFIG_SA1100_ASSABET) NCR_0 |= NCR_ENET_OSC_EN; #endif + platform_set_drvdata(pdev, ndev); ret = smc_enable_device(pdev); if (ret) goto out_release_attrib; @@ -2188,8 +2213,7 @@ static int smc_drv_probe(struct platform_device *pdev) } #endif - platform_set_drvdata(pdev, ndev); - ret = smc_probe(ndev, addr, irq_flags); + ret = smc_probe(ndev, addr, lp->cfg.irq_flags); if (ret != 0) goto out_iounmap; diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h index 92ff9c42367e..e044b4de1397 100644 --- a/drivers/net/smc91x.h +++ b/drivers/net/smc91x.h @@ -34,6 +34,7 @@ #ifndef _SMC91X_H_ #define _SMC91X_H_ +#include /* * Define your architecture specific bus configuration parameters here. @@ -481,6 +482,7 @@ static inline void LPD7_SMC_outsw (unsigned char* a, int r, #define RPC_LSA_DEFAULT RPC_LED_100_10 #define RPC_LSB_DEFAULT RPC_LED_TX_RX +#define SMC_DYNAMIC_BUS_CONFIG #endif @@ -526,8 +528,19 @@ struct smc_local { #endif void __iomem *base; void __iomem *datacs; + + struct smc91x_platdata cfg; }; +#ifdef SMC_DYNAMIC_BUS_CONFIG +#define SMC_8BIT(p) (((p)->cfg.flags & SMC91X_USE_8BIT) && SMC_CAN_USE_8BIT) +#define SMC_16BIT(p) (((p)->cfg.flags & SMC91X_USE_16BIT) && SMC_CAN_USE_16BIT) +#define SMC_32BIT(p) (((p)->cfg.flags & SMC91X_USE_32BIT) && SMC_CAN_USE_32BIT) +#else +#define SMC_8BIT(p) SMC_CAN_USE_8BIT +#define SMC_16BIT(p) SMC_CAN_USE_16BIT +#define SMC_32BIT(p) SMC_CAN_USE_32BIT +#endif #ifdef SMC_USE_PXA_DMA /* @@ -1108,41 +1121,41 @@ static const char * chip_ids[ 16 ] = { * * Enforce it on any 32-bit capable setup for now. */ -#define SMC_MUST_ALIGN_WRITE SMC_CAN_USE_32BIT +#define SMC_MUST_ALIGN_WRITE(lp) SMC_32BIT(lp) #define SMC_GET_PN(lp) \ - (SMC_CAN_USE_8BIT ? (SMC_inb(ioaddr, PN_REG(lp))) \ + (SMC_8BIT(lp) ? (SMC_inb(ioaddr, PN_REG(lp))) \ : (SMC_inw(ioaddr, PN_REG(lp)) & 0xFF)) #define SMC_SET_PN(lp, x) \ do { \ - if (SMC_MUST_ALIGN_WRITE) \ + if (SMC_MUST_ALIGN_WRITE(lp)) \ SMC_outl((x)<<16, ioaddr, SMC_REG(lp, 0, 2)); \ - else if (SMC_CAN_USE_8BIT) \ + else if (SMC_8BIT(lp)) \ SMC_outb(x, ioaddr, PN_REG(lp)); \ else \ SMC_outw(x, ioaddr, PN_REG(lp)); \ } while (0) #define SMC_GET_AR(lp) \ - (SMC_CAN_USE_8BIT ? (SMC_inb(ioaddr, AR_REG(lp))) \ + (SMC_8BIT(lp) ? (SMC_inb(ioaddr, AR_REG(lp))) \ : (SMC_inw(ioaddr, PN_REG(lp)) >> 8)) #define SMC_GET_TXFIFO(lp) \ - (SMC_CAN_USE_8BIT ? (SMC_inb(ioaddr, TXFIFO_REG(lp))) \ + (SMC_8BIT(lp) ? (SMC_inb(ioaddr, TXFIFO_REG(lp))) \ : (SMC_inw(ioaddr, TXFIFO_REG(lp)) & 0xFF)) #define SMC_GET_RXFIFO(lp) \ - (SMC_CAN_USE_8BIT ? (SMC_inb(ioaddr, RXFIFO_REG(lp))) \ + (SMC_8BIT(lp) ? (SMC_inb(ioaddr, RXFIFO_REG(lp))) \ : (SMC_inw(ioaddr, TXFIFO_REG(lp)) >> 8)) #define SMC_GET_INT(lp) \ - (SMC_CAN_USE_8BIT ? (SMC_inb(ioaddr, INT_REG(lp))) \ + (SMC_8BIT(lp) ? (SMC_inb(ioaddr, INT_REG(lp))) \ : (SMC_inw(ioaddr, INT_REG(lp)) & 0xFF)) #define SMC_ACK_INT(lp, x) \ do { \ - if (SMC_CAN_USE_8BIT) \ + if (SMC_8BIT(lp)) \ SMC_outb(x, ioaddr, INT_REG(lp)); \ else { \ unsigned long __flags; \ @@ -1155,12 +1168,12 @@ static const char * chip_ids[ 16 ] = { } while (0) #define SMC_GET_INT_MASK(lp) \ - (SMC_CAN_USE_8BIT ? (SMC_inb(ioaddr, IM_REG(lp))) \ + (SMC_8BIT(lp) ? (SMC_inb(ioaddr, IM_REG(lp))) \ : (SMC_inw(ioaddr, INT_REG(lp)) >> 8)) #define SMC_SET_INT_MASK(lp, x) \ do { \ - if (SMC_CAN_USE_8BIT) \ + if (SMC_8BIT(lp)) \ SMC_outb(x, ioaddr, IM_REG(lp)); \ else \ SMC_outw((x) << 8, ioaddr, INT_REG(lp)); \ @@ -1170,7 +1183,7 @@ static const char * chip_ids[ 16 ] = { #define SMC_SELECT_BANK(lp, x) \ do { \ - if (SMC_MUST_ALIGN_WRITE) \ + if (SMC_MUST_ALIGN_WRITE(lp)) \ SMC_outl((x)<<16, ioaddr, 12<> 16; \ @@ -1290,7 +1303,7 @@ static const char * chip_ids[ 16 ] = { #define SMC_PUSH_DATA(lp, p, l) \ do { \ - if (SMC_CAN_USE_32BIT) { \ + if (SMC_32BIT(lp)) { \ void *__ptr = (p); \ int __len = (l); \ void __iomem *__ioaddr = ioaddr; \ @@ -1308,15 +1321,15 @@ static const char * chip_ids[ 16 ] = { SMC_outw(*((u16 *)__ptr), ioaddr, \ DATA_REG(lp)); \ } \ - } else if (SMC_CAN_USE_16BIT) \ + } else if (SMC_16BIT(lp)) \ SMC_outsw(ioaddr, DATA_REG(lp), p, (l) >> 1); \ - else if (SMC_CAN_USE_8BIT) \ + else if (SMC_8BIT(lp)) \ SMC_outsb(ioaddr, DATA_REG(lp), p, l); \ } while (0) #define SMC_PULL_DATA(lp, p, l) \ do { \ - if (SMC_CAN_USE_32BIT) { \ + if (SMC_32BIT(lp)) { \ void *__ptr = (p); \ int __len = (l); \ void __iomem *__ioaddr = ioaddr; \ @@ -1343,9 +1356,9 @@ static const char * chip_ids[ 16 ] = { __ioaddr = lp->datacs; \ __len += 2; \ SMC_insl(__ioaddr, DATA_REG(lp), __ptr, __len>>2); \ - } else if (SMC_CAN_USE_16BIT) \ + } else if (SMC_16BIT(lp)) \ SMC_insw(ioaddr, DATA_REG(lp), p, (l) >> 1); \ - else if (SMC_CAN_USE_8BIT) \ + else if (SMC_8BIT(lp)) \ SMC_insb(ioaddr, DATA_REG(lp), p, l); \ } while (0) diff --git a/include/linux/smc91x.h b/include/linux/smc91x.h new file mode 100644 index 000000000000..8e0556b8781c --- /dev/null +++ b/include/linux/smc91x.h @@ -0,0 +1,13 @@ +#ifndef __SMC91X_H__ +#define __SMC91X_H__ + +#define SMC91X_USE_8BIT (1 << 0) +#define SMC91X_USE_16BIT (1 << 1) +#define SMC91X_USE_32BIT (1 << 2) + +struct smc91x_platdata { + unsigned long flags; + unsigned long irq_flags; /* IRQF_... */ +}; + +#endif /* __SMC91X_H__ */ -- cgit v1.2.3 From 82cc1a7a56872056af0ead6c7d695aa223f36695 Mon Sep 17 00:00:00 2001 From: Peter P Waskiewicz Jr Date: Fri, 21 Mar 2008 03:43:19 -0700 Subject: [NET]: Add per-connection option to set max TSO frame size Update: My mailer ate one of Jarek's feedback mails... Fixed the parameter in netif_set_gso_max_size() to be u32, not u16. Fixed the whitespace issue due to a patch import botch. Changed the types from u32 to unsigned int to be more consistent with other variables in the area. Also brought the patch up to the latest net-2.6.26 tree. Update: Made gso_max_size container 32 bits, not 16. Moved the location of gso_max_size within netdev to be less hotpath. Made more consistent names between the sock and netdev layers, and added a define for the max GSO size. Update: Respun for net-2.6.26 tree. Update: changed max_gso_frame_size and sk_gso_max_size from signed to unsigned - thanks Stephen! This patch adds the ability for device drivers to control the size of the TSO frames being sent to them, per TCP connection. By setting the netdevice's gso_max_size value, the socket layer will set the GSO frame size based on that value. This will propogate into the TCP layer, and send TSO's of that size to the hardware. This can be desirable to help tune the bursty nature of TSO on a per-adapter basis, where one may have 1 GbE and 10 GbE devices coexisting in a system, one running multiqueue and the other not, etc. This can also be desirable for devices that cannot support full 64 KB TSO's, but still want to benefit from some level of segmentation offloading. Signed-off-by: Peter P Waskiewicz Jr Signed-off-by: David S. Miller --- include/linux/netdevice.h | 10 ++++++++++ include/net/sock.h | 2 ++ net/core/dev.c | 1 + net/core/sock.c | 6 ++++-- net/ipv4/tcp_output.c | 4 ++-- 5 files changed, 19 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index a2f003239c85..ced61f87660e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -724,6 +724,10 @@ struct net_device /* rtnetlink link ops */ const struct rtnl_link_ops *rtnl_link_ops; + /* for setting kernel sock attribute on TCP connection setup */ +#define GSO_MAX_SIZE 65536 + unsigned int gso_max_size; + /* The TX queue control structures */ unsigned int egress_subqueue_count; struct net_device_subqueue egress_subqueue[1]; @@ -1475,6 +1479,12 @@ static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb) unlikely(skb->ip_summed != CHECKSUM_PARTIAL)); } +static inline void netif_set_gso_max_size(struct net_device *dev, + unsigned int size) +{ + dev->gso_max_size = size; +} + /* On bonding slaves other than the currently active slave, suppress * duplicates except for 802.3ad ETH_P_SLOW, alb non-mcast/bcast, and * ARP on active-backup slaves with arp_validate enabled. diff --git a/include/net/sock.h b/include/net/sock.h index 39112e75411c..8358fff002eb 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -151,6 +151,7 @@ struct sock_common { * @sk_no_check: %SO_NO_CHECK setting, wether or not checkup packets * @sk_route_caps: route capabilities (e.g. %NETIF_F_TSO) * @sk_gso_type: GSO type (e.g. %SKB_GSO_TCPV4) + * @sk_gso_max_size: Maximum GSO segment size to build * @sk_lingertime: %SO_LINGER l_linger setting * @sk_backlog: always used with the per-socket spinlock held * @sk_callback_lock: used with the callbacks in the end of this struct @@ -237,6 +238,7 @@ struct sock { gfp_t sk_allocation; int sk_route_caps; int sk_gso_type; + unsigned int sk_gso_max_size; int sk_rcvlowat; unsigned long sk_flags; unsigned long sk_lingertime; diff --git a/net/core/dev.c b/net/core/dev.c index fcdf03cf3b3f..f973e38b81af 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4021,6 +4021,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, } dev->egress_subqueue_count = queue_count; + dev->gso_max_size = GSO_MAX_SIZE; dev->get_stats = internal_stats; netpoll_netdev_init(dev); diff --git a/net/core/sock.c b/net/core/sock.c index bb5236aee643..b1a6ed4d33c1 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1095,10 +1095,12 @@ void sk_setup_caps(struct sock *sk, struct dst_entry *dst) if (sk->sk_route_caps & NETIF_F_GSO) sk->sk_route_caps |= NETIF_F_GSO_SOFTWARE; if (sk_can_gso(sk)) { - if (dst->header_len) + if (dst->header_len) { sk->sk_route_caps &= ~NETIF_F_GSO_MASK; - else + } else { sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM; + sk->sk_gso_max_size = dst->dev->gso_max_size; + } } } EXPORT_SYMBOL_GPL(sk_setup_caps); diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index b4e11d834c9f..a627616314ba 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -998,7 +998,7 @@ unsigned int tcp_current_mss(struct sock *sk, int large_allowed) xmit_size_goal = mss_now; if (doing_tso) { - xmit_size_goal = (65535 - + xmit_size_goal = ((sk->sk_gso_max_size - 1) - inet_csk(sk)->icsk_af_ops->net_header_len - inet_csk(sk)->icsk_ext_hdr_len - tp->tcp_header_len); @@ -1282,7 +1282,7 @@ static int tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb) limit = min(send_win, cong_win); /* If a full-sized TSO skb can be sent, do it. */ - if (limit >= 65536) + if (limit >= sk->sk_gso_max_size) goto send_now; if (sysctl_tcp_tso_win_divisor) { -- cgit v1.2.3 From a91275eff43a527e1a25d6d034cbcd19ee323e64 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 21 Mar 2008 04:11:58 -0700 Subject: [NETNS][IPV6] udp - make proc handle the network namespace This patch makes the common udp proc functions to take care of which socket they should show taking into account the namespace it belongs. Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/udp.h | 1 + net/ipv4/udp.c | 32 ++++++++++++++++++++++++++++---- 2 files changed, 29 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index c6669c0a74c7..a1b33d667199 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -192,6 +192,7 @@ struct udp_seq_afinfo { }; struct udp_iter_state { + struct net *net; sa_family_t family; struct hlist_head *hashtable; int bucket; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 7ea1b67b6de1..049e92519616 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1512,10 +1512,13 @@ static struct sock *udp_get_first(struct seq_file *seq) { struct sock *sk; struct udp_iter_state *state = seq->private; + struct net *net = state->net; for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) { struct hlist_node *node; sk_for_each(sk, node, state->hashtable + state->bucket) { + if (sk->sk_net != net) + continue; if (sk->sk_family == state->family) goto found; } @@ -1528,12 +1531,13 @@ found: static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk) { struct udp_iter_state *state = seq->private; + struct net *net = state->net; do { sk = sk_next(sk); try_again: ; - } while (sk && sk->sk_family != state->family); + } while (sk && sk->sk_net != net && sk->sk_family != state->family); if (!sk && ++state->bucket < UDP_HTABLE_SIZE) { sk = sk_head(state->hashtable + state->bucket); @@ -1582,31 +1586,51 @@ static int udp_seq_open(struct inode *inode, struct file *file) { struct udp_seq_afinfo *afinfo = PDE(inode)->data; struct seq_file *seq; + struct net *net; int rc = -ENOMEM; struct udp_iter_state *s = kzalloc(sizeof(*s), GFP_KERNEL); if (!s) goto out; + + rc = -ENXIO; + net = get_proc_net(inode); + if (!net) + goto out_kfree; + s->family = afinfo->family; s->hashtable = afinfo->hashtable; s->seq_ops.start = udp_seq_start; s->seq_ops.next = udp_seq_next; s->seq_ops.show = afinfo->seq_show; s->seq_ops.stop = udp_seq_stop; + s->net = net; rc = seq_open(file, &s->seq_ops); if (rc) - goto out_kfree; + goto out_put_net; - seq = file->private_data; + seq = file->private_data; seq->private = s; out: return rc; +out_put_net: + put_net(net); out_kfree: kfree(s); goto out; } +static int udp_seq_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct udp_iter_state *s = seq->private; + + put_net(s->net); + seq_release_private(inode, file); + return 0; +} + /* ------------------------------------------------------------------------ */ int udp_proc_register(struct udp_seq_afinfo *afinfo) { @@ -1619,7 +1643,7 @@ int udp_proc_register(struct udp_seq_afinfo *afinfo) afinfo->seq_fops->open = udp_seq_open; afinfo->seq_fops->read = seq_read; afinfo->seq_fops->llseek = seq_lseek; - afinfo->seq_fops->release = seq_release_private; + afinfo->seq_fops->release = udp_seq_release; p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops); if (p) -- cgit v1.2.3 From f40c8174d3c21bf178283f3ef3aa8c7bf238fdec Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 21 Mar 2008 04:13:54 -0700 Subject: [NETNS][IPV4] tcp - make proc handle the network namespaces This patch, like udp proc, makes the proc functions to take care of which namespace the socket belongs. Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/tcp.h | 1 + net/ipv4/tcp_ipv4.c | 44 +++++++++++++++++++++++++++++++++++--------- 2 files changed, 36 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 11119e33acfe..6b08dab1b1fa 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1328,6 +1328,7 @@ struct tcp_seq_afinfo { }; struct tcp_iter_state { + struct net *net; sa_family_t family; enum tcp_seq_states state; struct sock *syn_wait_sk; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index a79e324638eb..f9b30dc3bd6c 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1948,6 +1948,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur) struct hlist_node *node; struct sock *sk = cur; struct tcp_iter_state* st = seq->private; + struct net *net = st->net; if (!sk) { st->bucket = 0; @@ -1964,7 +1965,8 @@ static void *listening_get_next(struct seq_file *seq, void *cur) req = req->dl_next; while (1) { while (req) { - if (req->rsk_ops->family == st->family) { + if (req->rsk_ops->family == st->family && + req->sk->sk_net == net) { cur = req; goto out; } @@ -1988,7 +1990,7 @@ get_req: } get_sk: sk_for_each_from(sk, node) { - if (sk->sk_family == st->family) { + if (sk->sk_family == st->family && sk->sk_net == net) { cur = sk; goto out; } @@ -2027,6 +2029,7 @@ static void *listening_get_idx(struct seq_file *seq, loff_t *pos) static void *established_get_first(struct seq_file *seq) { struct tcp_iter_state* st = seq->private; + struct net *net = st->net; void *rc = NULL; for (st->bucket = 0; st->bucket < tcp_hashinfo.ehash_size; ++st->bucket) { @@ -2037,7 +2040,8 @@ static void *established_get_first(struct seq_file *seq) read_lock_bh(lock); sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) { - if (sk->sk_family != st->family) { + if (sk->sk_family != st->family || + sk->sk_net != net) { continue; } rc = sk; @@ -2046,7 +2050,8 @@ static void *established_get_first(struct seq_file *seq) st->state = TCP_SEQ_STATE_TIME_WAIT; inet_twsk_for_each(tw, node, &tcp_hashinfo.ehash[st->bucket].twchain) { - if (tw->tw_family != st->family) { + if (tw->tw_family != st->family && + tw->tw_net != net) { continue; } rc = tw; @@ -2065,6 +2070,7 @@ static void *established_get_next(struct seq_file *seq, void *cur) struct inet_timewait_sock *tw; struct hlist_node *node; struct tcp_iter_state* st = seq->private; + struct net *net = st->net; ++st->num; @@ -2072,7 +2078,7 @@ static void *established_get_next(struct seq_file *seq, void *cur) tw = cur; tw = tw_next(tw); get_tw: - while (tw && tw->tw_family != st->family) { + while (tw && tw->tw_family != st->family && tw->tw_net != net) { tw = tw_next(tw); } if (tw) { @@ -2093,7 +2099,7 @@ get_tw: sk = sk_next(sk); sk_for_each_from(sk, node) { - if (sk->sk_family == st->family) + if (sk->sk_family == st->family && sk->sk_net == net) goto found; } @@ -2201,6 +2207,7 @@ static int tcp_seq_open(struct inode *inode, struct file *file) struct tcp_seq_afinfo *afinfo = PDE(inode)->data; struct seq_file *seq; struct tcp_iter_state *s; + struct net *net; int rc; if (unlikely(afinfo == NULL)) @@ -2209,24 +2216,43 @@ static int tcp_seq_open(struct inode *inode, struct file *file) s = kzalloc(sizeof(*s), GFP_KERNEL); if (!s) return -ENOMEM; + + rc = -ENXIO; + net = get_proc_net(inode); + if (!net) + goto out_kfree; + s->family = afinfo->family; s->seq_ops.start = tcp_seq_start; s->seq_ops.next = tcp_seq_next; s->seq_ops.show = afinfo->seq_show; s->seq_ops.stop = tcp_seq_stop; + s->net = net; rc = seq_open(file, &s->seq_ops); if (rc) - goto out_kfree; - seq = file->private_data; + goto out_put_net; + seq = file->private_data; seq->private = s; out: return rc; +out_put_net: + put_net(net); out_kfree: kfree(s); goto out; } +static int tcp_seq_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + struct tcp_iter_state *s = seq->private; + + put_net(s->net); + seq_release_private(inode, file); + return 0; +} + int tcp_proc_register(struct tcp_seq_afinfo *afinfo) { int rc = 0; @@ -2238,7 +2264,7 @@ int tcp_proc_register(struct tcp_seq_afinfo *afinfo) afinfo->seq_fops->open = tcp_seq_open; afinfo->seq_fops->read = seq_read; afinfo->seq_fops->llseek = seq_lseek; - afinfo->seq_fops->release = seq_release_private; + afinfo->seq_fops->release = tcp_seq_release; p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops); if (p) -- cgit v1.2.3 From 0c96d8c50bffb7f02690dd8a8cf1adb8e07e100f Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 21 Mar 2008 04:14:17 -0700 Subject: [NETNS][IPV6] udp6 - make proc per namespace The proc init/exit functions take a new network namespace parameter in order to register/unregister /proc/net/udp6 for a namespace. Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/ipv6.h | 4 ++-- include/net/udp.h | 4 ++-- net/ipv4/udp.c | 12 ++++++------ net/ipv4/udplite.c | 2 +- net/ipv6/af_inet6.c | 19 ++++++++++++------- net/ipv6/udp.c | 8 ++++---- net/ipv6/udplite.c | 4 ++-- 7 files changed, 29 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 8db06af1efbb..e01a563132e8 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -597,8 +597,8 @@ extern int raw6_proc_init(void); extern void raw6_proc_exit(void); extern int tcp6_proc_init(void); extern void tcp6_proc_exit(void); -extern int udp6_proc_init(void); -extern void udp6_proc_exit(void); +extern int udp6_proc_init(struct net *net); +extern void udp6_proc_exit(struct net *net); extern int udplite6_proc_init(void); extern void udplite6_proc_exit(void); extern int ipv6_misc_proc_init(void); diff --git a/include/net/udp.h b/include/net/udp.h index a1b33d667199..b4cbdce883d1 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -200,8 +200,8 @@ struct udp_iter_state { }; #ifdef CONFIG_PROC_FS -extern int udp_proc_register(struct udp_seq_afinfo *afinfo); -extern void udp_proc_unregister(struct udp_seq_afinfo *afinfo); +extern int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo); +extern void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo); extern int udp4_proc_init(void); extern void udp4_proc_exit(void); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 049e92519616..a98c43c0a89c 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1632,7 +1632,7 @@ static int udp_seq_release(struct inode *inode, struct file *file) } /* ------------------------------------------------------------------------ */ -int udp_proc_register(struct udp_seq_afinfo *afinfo) +int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) { struct proc_dir_entry *p; int rc = 0; @@ -1645,7 +1645,7 @@ int udp_proc_register(struct udp_seq_afinfo *afinfo) afinfo->seq_fops->llseek = seq_lseek; afinfo->seq_fops->release = udp_seq_release; - p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops); + p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops); if (p) p->data = afinfo; else @@ -1653,11 +1653,11 @@ int udp_proc_register(struct udp_seq_afinfo *afinfo) return rc; } -void udp_proc_unregister(struct udp_seq_afinfo *afinfo) +void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo) { if (!afinfo) return; - proc_net_remove(&init_net, afinfo->name); + proc_net_remove(net, afinfo->name); memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } @@ -1709,12 +1709,12 @@ static struct udp_seq_afinfo udp4_seq_afinfo = { int __init udp4_proc_init(void) { - return udp_proc_register(&udp4_seq_afinfo); + return udp_proc_register(&init_net, &udp4_seq_afinfo); } void udp4_proc_exit(void) { - udp_proc_unregister(&udp4_seq_afinfo); + udp_proc_unregister(&init_net, &udp4_seq_afinfo); } #endif /* CONFIG_PROC_FS */ diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index d49c6d68c8a9..2469b5104587 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -105,7 +105,7 @@ void __init udplite4_register(void) inet_register_protosw(&udplite4_protosw); #ifdef CONFIG_PROC_FS - if (udp_proc_register(&udplite4_seq_afinfo)) /* udplite4_proc_init() */ + if (udp_proc_register(&init_net, &udplite4_seq_afinfo)) printk(KERN_ERR "%s: Cannot register /proc!\n", __func__); #endif return; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 730a861b8f41..e3e09147d134 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -842,6 +842,8 @@ static void cleanup_ipv6_mibs(void) static int inet6_net_init(struct net *net) { + int err = 0; + net->ipv6.sysctl.bindv6only = 0; net->ipv6.sysctl.flush_delay = 0; net->ipv6.sysctl.ip6_rt_max_size = 4096; @@ -853,12 +855,20 @@ static int inet6_net_init(struct net *net) net->ipv6.sysctl.ip6_rt_min_advmss = IPV6_MIN_MTU - 20 - 40; net->ipv6.sysctl.icmpv6_time = 1*HZ; - return 0; +#ifdef CONFIG_PROC_FS + err = udp6_proc_init(net); + if (err) + goto out; +out: +#endif + return err; } static void inet6_net_exit(struct net *net) { - return; +#ifdef CONFIG_PROC_FS + udp6_proc_exit(net); +#endif } static struct pernet_operations inet6_net_ops = { @@ -943,8 +953,6 @@ static int __init inet6_init(void) goto proc_raw6_fail; if (tcp6_proc_init()) goto proc_tcp6_fail; - if (udp6_proc_init()) - goto proc_udp6_fail; if (udplite6_proc_init()) goto proc_udplite6_fail; if (ipv6_misc_proc_init()) @@ -1029,8 +1037,6 @@ proc_anycast6_fail: proc_misc6_fail: udplite6_proc_exit(); proc_udplite6_fail: - udp6_proc_exit(); -proc_udp6_fail: tcp6_proc_exit(); proc_tcp6_fail: raw6_proc_exit(); @@ -1092,7 +1098,6 @@ static void __exit inet6_exit(void) ac6_proc_exit(); ipv6_misc_proc_exit(); udplite6_proc_exit(); - udp6_proc_exit(); tcp6_proc_exit(); raw6_proc_exit(); #endif diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index d6e311f6c8eb..af619d48ba80 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -989,13 +989,13 @@ static struct udp_seq_afinfo udp6_seq_afinfo = { .seq_fops = &udp6_seq_fops, }; -int __init udp6_proc_init(void) +int udp6_proc_init(struct net *net) { - return udp_proc_register(&udp6_seq_afinfo); + return udp_proc_register(net, &udp6_seq_afinfo); } -void udp6_proc_exit(void) { - udp_proc_unregister(&udp6_seq_afinfo); +void udp6_proc_exit(struct net *net) { + udp_proc_unregister(net, &udp6_seq_afinfo); } #endif /* CONFIG_PROC_FS */ diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 87d4202522ee..815190be528c 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -115,11 +115,11 @@ static struct udp_seq_afinfo udplite6_seq_afinfo = { int __init udplite6_proc_init(void) { - return udp_proc_register(&udplite6_seq_afinfo); + return udp_proc_register(&init_net, &udplite6_seq_afinfo); } void udplite6_proc_exit(void) { - udp_proc_unregister(&udplite6_seq_afinfo); + udp_proc_unregister(&init_net, &udplite6_seq_afinfo); } #endif -- cgit v1.2.3 From 6f8b13bcb3369a5df2e63acc422bed6098f5b8c4 Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Fri, 21 Mar 2008 04:14:45 -0700 Subject: [NETNS][IPV6] tcp6 - make proc per namespace Make the proc for tcp6 to be per namespace. Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/ipv6.h | 4 ++-- include/net/tcp.h | 4 ++-- net/ipv4/tcp_ipv4.c | 12 ++++++------ net/ipv6/af_inet6.c | 15 ++++++++++----- net/ipv6/tcp_ipv6.c | 8 ++++---- 5 files changed, 24 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index e01a563132e8..e82f1814d96b 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -595,8 +595,8 @@ extern int ac6_proc_init(void); extern void ac6_proc_exit(void); extern int raw6_proc_init(void); extern void raw6_proc_exit(void); -extern int tcp6_proc_init(void); -extern void tcp6_proc_exit(void); +extern int tcp6_proc_init(struct net *net); +extern void tcp6_proc_exit(struct net *net); extern int udp6_proc_init(struct net *net); extern void udp6_proc_exit(struct net *net); extern int udplite6_proc_init(void); diff --git a/include/net/tcp.h b/include/net/tcp.h index 6b08dab1b1fa..847e1634e1f4 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1336,8 +1336,8 @@ struct tcp_iter_state { struct seq_operations seq_ops; }; -extern int tcp_proc_register(struct tcp_seq_afinfo *afinfo); -extern void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo); +extern int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo); +extern void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo); extern struct request_sock_ops tcp_request_sock_ops; extern struct request_sock_ops tcp6_request_sock_ops; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index f9b30dc3bd6c..744bc9d6cebc 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2253,7 +2253,7 @@ static int tcp_seq_release(struct inode *inode, struct file *file) return 0; } -int tcp_proc_register(struct tcp_seq_afinfo *afinfo) +int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) { int rc = 0; struct proc_dir_entry *p; @@ -2266,7 +2266,7 @@ int tcp_proc_register(struct tcp_seq_afinfo *afinfo) afinfo->seq_fops->llseek = seq_lseek; afinfo->seq_fops->release = tcp_seq_release; - p = proc_net_fops_create(&init_net, afinfo->name, S_IRUGO, afinfo->seq_fops); + p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops); if (p) p->data = afinfo; else @@ -2274,11 +2274,11 @@ int tcp_proc_register(struct tcp_seq_afinfo *afinfo) return rc; } -void tcp_proc_unregister(struct tcp_seq_afinfo *afinfo) +void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo) { if (!afinfo) return; - proc_net_remove(&init_net, afinfo->name); + proc_net_remove(net, afinfo->name); memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } @@ -2419,12 +2419,12 @@ static struct tcp_seq_afinfo tcp4_seq_afinfo = { int __init tcp4_proc_init(void) { - return tcp_proc_register(&tcp4_seq_afinfo); + return tcp_proc_register(&init_net, &tcp4_seq_afinfo); } void tcp4_proc_exit(void) { - tcp_proc_unregister(&tcp4_seq_afinfo); + tcp_proc_unregister(&init_net, &tcp4_seq_afinfo); } #endif /* CONFIG_PROC_FS */ diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index e3e09147d134..f52bdaed8a1b 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -859,15 +859,25 @@ static int inet6_net_init(struct net *net) err = udp6_proc_init(net); if (err) goto out; + err = tcp6_proc_init(net); + if (err) + goto proc_tcp6_fail; out: #endif return err; + +#ifdef CONFIG_PROC_FS +proc_tcp6_fail: + udp6_proc_exit(net); + goto out; +#endif } static void inet6_net_exit(struct net *net) { #ifdef CONFIG_PROC_FS udp6_proc_exit(net); + tcp6_proc_exit(net); #endif } @@ -951,8 +961,6 @@ static int __init inet6_init(void) err = -ENOMEM; if (raw6_proc_init()) goto proc_raw6_fail; - if (tcp6_proc_init()) - goto proc_tcp6_fail; if (udplite6_proc_init()) goto proc_udplite6_fail; if (ipv6_misc_proc_init()) @@ -1037,8 +1045,6 @@ proc_anycast6_fail: proc_misc6_fail: udplite6_proc_exit(); proc_udplite6_fail: - tcp6_proc_exit(); -proc_tcp6_fail: raw6_proc_exit(); proc_raw6_fail: #endif @@ -1098,7 +1104,6 @@ static void __exit inet6_exit(void) ac6_proc_exit(); ipv6_misc_proc_exit(); udplite6_proc_exit(); - tcp6_proc_exit(); raw6_proc_exit(); #endif ipv6_netfilter_fini(); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index caf0cc1c00e1..56d0cea7d578 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2129,14 +2129,14 @@ static struct tcp_seq_afinfo tcp6_seq_afinfo = { .seq_fops = &tcp6_seq_fops, }; -int __init tcp6_proc_init(void) +int tcp6_proc_init(struct net *net) { - return tcp_proc_register(&tcp6_seq_afinfo); + return tcp_proc_register(net, &tcp6_seq_afinfo); } -void tcp6_proc_exit(void) +void tcp6_proc_exit(struct net *net) { - tcp_proc_unregister(&tcp6_seq_afinfo); + tcp_proc_unregister(net, &tcp6_seq_afinfo); } #endif -- cgit v1.2.3 From 4cd9029d25f6612302f82634620f107c65f790b1 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 21 Mar 2008 15:54:53 -0700 Subject: socket: SOCK_DEBUG type checking Use the inline trick (same as pr_debug) to get checking of debug statements even if no code is generated. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/net/sock.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 8358fff002eb..b89680d2693b 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -70,7 +70,11 @@ #define SOCK_DEBUG(sk, msg...) do { if ((sk) && sock_flag((sk), SOCK_DBG)) \ printk(KERN_DEBUG msg); } while (0) #else -#define SOCK_DEBUG(sk, msg...) do { } while (0) +/* Validate arguments and do nothing */ +static void inline int __attribute__ ((format (printf, 2, 3))) +SOCK_DEBUG(struct sock *sk, const char *msg, ...) +{ +} #endif /* This is the per-socket lock. The spinlock provides a synchronization -- cgit v1.2.3 From ec3c0982a2dd1e671bad8e9d26c28dcba0039d87 Mon Sep 17 00:00:00 2001 From: Patrick McManus Date: Fri, 21 Mar 2008 16:33:01 -0700 Subject: [TCP]: TCP_DEFER_ACCEPT updates - process as established Change TCP_DEFER_ACCEPT implementation so that it transitions a connection to ESTABLISHED after handshake is complete instead of leaving it in SYN-RECV until some data arrvies. Place connection in accept queue when first data packet arrives from slow path. Benefits: - established connection is now reset if it never makes it to the accept queue - diagnostic state of established matches with the packet traces showing completed handshake - TCP_DEFER_ACCEPT timeouts are expressed in seconds and can now be enforced with reasonable accuracy instead of rounding up to next exponential back-off of syn-ack retry. Signed-off-by: Patrick McManus Signed-off-by: David S. Miller --- include/linux/tcp.h | 7 +++++++ include/net/request_sock.h | 4 ++-- include/net/tcp.h | 1 + net/ipv4/inet_connection_sock.c | 11 +++------- net/ipv4/tcp.c | 18 +++++++--------- net/ipv4/tcp_input.c | 46 +++++++++++++++++++++++++++++++++++++++++ net/ipv4/tcp_ipv4.c | 8 +++++++ net/ipv4/tcp_minisocks.c | 32 +++++++++++++++++----------- net/ipv4/tcp_timer.c | 5 +++++ 9 files changed, 99 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 08027f1d7f31..d96d9b122304 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -239,6 +239,11 @@ static inline struct tcp_request_sock *tcp_rsk(const struct request_sock *req) return (struct tcp_request_sock *)req; } +struct tcp_deferred_accept_info { + struct sock *listen_sk; + struct request_sock *request; +}; + struct tcp_sock { /* inet_connection_sock has to be the first member of tcp_sock */ struct inet_connection_sock inet_conn; @@ -374,6 +379,8 @@ struct tcp_sock { unsigned int keepalive_intvl; /* time interval between keep alive probes */ int linger2; + struct tcp_deferred_accept_info defer_tcp_accept; + unsigned long last_synq_overflow; u32 tso_deferred; diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 040780add355..0369f98e9f3a 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -115,8 +115,8 @@ struct request_sock_queue { struct request_sock *rskq_accept_head; struct request_sock *rskq_accept_tail; rwlock_t syn_wait_lock; - u8 rskq_defer_accept; - /* 3 bytes hole, try to pack */ + u16 rskq_defer_accept; + /* 2 bytes hole, try to pack */ struct listen_sock *listen_opt; }; diff --git a/include/net/tcp.h b/include/net/tcp.h index 847e1634e1f4..67cc3956d29c 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -139,6 +139,7 @@ extern void tcp_time_wait(struct sock *sk, int state, int timeo); #define MAX_TCP_KEEPINTVL 32767 #define MAX_TCP_KEEPCNT 127 #define MAX_TCP_SYNCNT 127 +#define MAX_TCP_ACCEPT_DEFERRED 65535 #define TCP_SYNQ_INTERVAL (HZ/5) /* Period of SYNACK timer */ diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 8a45be988709..cc1a1859a61b 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -414,8 +414,7 @@ void inet_csk_reqsk_queue_prune(struct sock *parent, struct inet_connection_sock *icsk = inet_csk(parent); struct request_sock_queue *queue = &icsk->icsk_accept_queue; struct listen_sock *lopt = queue->listen_opt; - int max_retries = icsk->icsk_syn_retries ? : sysctl_tcp_synack_retries; - int thresh = max_retries; + int thresh = icsk->icsk_syn_retries ? : sysctl_tcp_synack_retries; unsigned long now = jiffies; struct request_sock **reqp, *req; int i, budget; @@ -451,9 +450,6 @@ void inet_csk_reqsk_queue_prune(struct sock *parent, } } - if (queue->rskq_defer_accept) - max_retries = queue->rskq_defer_accept; - budget = 2 * (lopt->nr_table_entries / (timeout / interval)); i = lopt->clock_hand; @@ -461,9 +457,8 @@ void inet_csk_reqsk_queue_prune(struct sock *parent, reqp=&lopt->syn_table[i]; while ((req = *reqp) != NULL) { if (time_after_eq(now, req->expires)) { - if ((req->retrans < (inet_rsk(req)->acked ? max_retries : thresh)) && - (inet_rsk(req)->acked || - !req->rsk_ops->rtx_syn_ack(parent, req))) { + if (req->retrans < thresh && + !req->rsk_ops->rtx_syn_ack(parent, req)) { unsigned long timeo; if (req->retrans++ == 0) diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 071e83a894ad..e0fbc25ca816 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -2105,15 +2105,12 @@ static int do_tcp_setsockopt(struct sock *sk, int level, break; case TCP_DEFER_ACCEPT: - icsk->icsk_accept_queue.rskq_defer_accept = 0; - if (val > 0) { - /* Translate value in seconds to number of - * retransmits */ - while (icsk->icsk_accept_queue.rskq_defer_accept < 32 && - val > ((TCP_TIMEOUT_INIT / HZ) << - icsk->icsk_accept_queue.rskq_defer_accept)) - icsk->icsk_accept_queue.rskq_defer_accept++; - icsk->icsk_accept_queue.rskq_defer_accept++; + if (val < 0) { + err = -EINVAL; + } else { + if (val > MAX_TCP_ACCEPT_DEFERRED) + val = MAX_TCP_ACCEPT_DEFERRED; + icsk->icsk_accept_queue.rskq_defer_accept = val; } break; @@ -2295,8 +2292,7 @@ static int do_tcp_getsockopt(struct sock *sk, int level, val = (val ? : sysctl_tcp_fin_timeout) / HZ; break; case TCP_DEFER_ACCEPT: - val = !icsk->icsk_accept_queue.rskq_defer_accept ? 0 : - ((TCP_TIMEOUT_INIT / HZ) << (icsk->icsk_accept_queue.rskq_defer_accept - 1)); + val = icsk->icsk_accept_queue.rskq_defer_accept; break; case TCP_WINDOW_CLAMP: val = tp->window_clamp; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 9cf446427cc2..6e46b4c0f28c 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4451,6 +4451,49 @@ static void tcp_urg(struct sock *sk, struct sk_buff *skb, struct tcphdr *th) } } +static int tcp_defer_accept_check(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (tp->defer_tcp_accept.request) { + int queued_data = tp->rcv_nxt - tp->copied_seq; + int hasfin = !skb_queue_empty(&sk->sk_receive_queue) ? + tcp_hdr((struct sk_buff *) + sk->sk_receive_queue.prev)->fin : 0; + + if (queued_data && hasfin) + queued_data--; + + if (queued_data && + tp->defer_tcp_accept.listen_sk->sk_state == TCP_LISTEN) { + if (sock_flag(sk, SOCK_KEEPOPEN)) { + inet_csk_reset_keepalive_timer(sk, + keepalive_time_when(tp)); + } else { + inet_csk_delete_keepalive_timer(sk); + } + + inet_csk_reqsk_queue_add( + tp->defer_tcp_accept.listen_sk, + tp->defer_tcp_accept.request, + sk); + + tp->defer_tcp_accept.listen_sk->sk_data_ready( + tp->defer_tcp_accept.listen_sk, 0); + + sock_put(tp->defer_tcp_accept.listen_sk); + sock_put(sk); + tp->defer_tcp_accept.listen_sk = NULL; + tp->defer_tcp_accept.request = NULL; + } else if (hasfin || + tp->defer_tcp_accept.listen_sk->sk_state != TCP_LISTEN) { + tcp_reset(sk); + return -1; + } + } + return 0; +} + static int tcp_copy_to_iovec(struct sock *sk, struct sk_buff *skb, int hlen) { struct tcp_sock *tp = tcp_sk(sk); @@ -4811,6 +4854,9 @@ step5: tcp_data_snd_check(sk); tcp_ack_snd_check(sk); + + if (tcp_defer_accept_check(sk)) + return -1; return 0; csum_error: diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 0ba6e911c979..167a0f557531 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1920,6 +1920,14 @@ int tcp_v4_destroy_sock(struct sock *sk) sk->sk_sndmsg_page = NULL; } + if (tp->defer_tcp_accept.request) { + reqsk_free(tp->defer_tcp_accept.request); + sock_put(tp->defer_tcp_accept.listen_sk); + sock_put(sk); + tp->defer_tcp_accept.listen_sk = NULL; + tp->defer_tcp_accept.request = NULL; + } + atomic_dec(&tcp_sockets_allocated); return 0; diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 8245247a6ceb..019c8c16e5cc 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -571,10 +571,8 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, does sequence test, SYN is truncated, and thus we consider it a bare ACK. - If icsk->icsk_accept_queue.rskq_defer_accept, we silently drop this - bare ACK. Otherwise, we create an established connection. Both - ends (listening sockets) accept the new incoming connection and try - to talk to each other. 8-) + Both ends (listening sockets) accept the new incoming + connection and try to talk to each other. 8-) Note: This case is both harmless, and rare. Possibility is about the same as us discovering intelligent life on another plant tomorrow. @@ -642,13 +640,6 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, if (!(flg & TCP_FLAG_ACK)) return NULL; - /* If TCP_DEFER_ACCEPT is set, drop bare ACK. */ - if (inet_csk(sk)->icsk_accept_queue.rskq_defer_accept && - TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) { - inet_rsk(req)->acked = 1; - return NULL; - } - /* OK, ACK is valid, create big socket and * feed this segment to it. It will repeat all * the tests. THIS SEGMENT MUST MOVE SOCKET TO @@ -687,7 +678,24 @@ struct sock *tcp_check_req(struct sock *sk,struct sk_buff *skb, inet_csk_reqsk_queue_unlink(sk, req, prev); inet_csk_reqsk_queue_removed(sk, req); - inet_csk_reqsk_queue_add(sk, req, child); + if (inet_csk(sk)->icsk_accept_queue.rskq_defer_accept && + TCP_SKB_CB(skb)->end_seq == tcp_rsk(req)->rcv_isn + 1) { + + /* the accept queue handling is done is est recv slow + * path so lets make sure to start there + */ + tcp_sk(child)->pred_flags = 0; + sock_hold(sk); + sock_hold(child); + tcp_sk(child)->defer_tcp_accept.listen_sk = sk; + tcp_sk(child)->defer_tcp_accept.request = req; + + inet_csk_reset_keepalive_timer(child, + inet_csk(sk)->icsk_accept_queue.rskq_defer_accept * HZ); + } else { + inet_csk_reqsk_queue_add(sk, req, child); + } + return child; listen_overflow: diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c index 803d758a2b12..160d16f9f4fc 100644 --- a/net/ipv4/tcp_timer.c +++ b/net/ipv4/tcp_timer.c @@ -481,6 +481,11 @@ static void tcp_keepalive_timer (unsigned long data) goto death; } + if (tp->defer_tcp_accept.request && sk->sk_state == TCP_ESTABLISHED) { + tcp_send_active_reset(sk, GFP_ATOMIC); + goto death; + } + if (!sock_flag(sk, SOCK_KEEPOPEN) || sk->sk_state == TCP_CLOSE) goto out; -- cgit v1.2.3 From ef722495c8867aacc1db0675a6737e5cf1e72e07 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Sat, 22 Mar 2008 16:35:29 -0700 Subject: [IPV4]: Remove unused ip_options->is_data. ip_options->is_data is assigned only and never checked. The structure is not a part of kernel interface to the userspace. So, it is safe to remove this field. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/inet_sock.h | 3 +-- net/ipv4/cipso_ipv4.c | 1 - net/ipv4/ip_options.c | 5 ----- 3 files changed, 1 insertion(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index 8660cb0fa0dd..b6db16d2766a 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -43,8 +43,7 @@ struct ip_options { unsigned char srr; unsigned char rr; unsigned char ts; - unsigned char is_data:1, - is_strictroute:1, + unsigned char is_strictroute:1, srr_is_hit:1, is_changed:1, rr_needaddr:1, diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 8cd357f41283..4637ded3dba8 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -1800,7 +1800,6 @@ int cipso_v4_sock_setattr(struct sock *sk, } memcpy(opt->__data, buf, buf_len); opt->optlen = opt_len; - opt->is_data = 1; opt->cipso = sizeof(struct iphdr); kfree(buf); buf = NULL; diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 7e94bf850a07..2fa411381b11 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -45,7 +45,6 @@ void ip_options_build(struct sk_buff * skb, struct ip_options * opt, memcpy(&(IPCB(skb)->opt), opt, sizeof(struct ip_options)); memcpy(iph+sizeof(struct iphdr), opt->__data, opt->optlen); opt = &(IPCB(skb)->opt); - opt->is_data = 0; if (opt->srr) memcpy(iph+opt->srr+iph[opt->srr+1]-4, &daddr, 4); @@ -95,8 +94,6 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb) memset(dopt, 0, sizeof(struct ip_options)); - dopt->is_data = 1; - sopt = &(IPCB(skb)->opt); if (sopt->optlen == 0) { @@ -265,7 +262,6 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb) iph = skb_network_header(skb); opt->optlen = ((struct iphdr *)iph)->ihl*4 - sizeof(struct iphdr); optptr = iph + sizeof(struct iphdr); - opt->is_data = 0; } else { optptr = opt->__data; iph = optptr - sizeof(struct iphdr); @@ -519,7 +515,6 @@ static int ip_options_get_finish(struct ip_options **optp, while (optlen & 3) opt->__data[optlen++] = IPOPT_END; opt->optlen = optlen; - opt->is_data = 1; if (optlen && ip_options_compile(opt, NULL)) { kfree(opt); return -EINVAL; -- cgit v1.2.3 From 39d8cda76cfb1178455f9d196b39e773878e6c05 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Sat, 22 Mar 2008 16:50:58 -0700 Subject: [SOCK]: Add udp_hash member to struct proto. Inspired by the commit ab1e0a13 ([SOCK] proto: Add hashinfo member to struct proto) from Arnaldo, I made similar thing for UDP/-Lite IPv4 and -v6 protocols. The result is not that exciting, but it removes some levels of indirection in udpxxx_get_port and saves some space in code and text. The first step is to union existing hashinfo and new udp_hash on the struct proto and give a name to this union, since future initialization of tcpxxx_prot, dccp_vx_protinfo and udpxxx_protinfo will cause gcc warning about inability to initialize anonymous member this way. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/inet_hashtables.h | 2 +- include/net/sock.h | 5 ++++- net/dccp/ipv4.c | 2 +- net/dccp/ipv6.c | 2 +- net/ipv4/inet_connection_sock.c | 2 +- net/ipv4/inet_hashtables.c | 8 ++++---- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/inet6_hashtables.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 9 files changed, 15 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 97dc35ad09be..d99c1ba2ece0 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -223,7 +223,7 @@ static inline int inet_sk_listen_hashfn(const struct sock *sk) /* Caller must disable local BH processing. */ static inline void __inet_inherit_port(struct sock *sk, struct sock *child) { - struct inet_hashinfo *table = sk->sk_prot->hashinfo; + struct inet_hashinfo *table = sk->sk_prot->h.hashinfo; const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size); struct inet_bind_hashbucket *head = &table->bhash[bhash]; struct inet_bind_bucket *tb; diff --git a/include/net/sock.h b/include/net/sock.h index b89680d2693b..c3175c400b79 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -586,7 +586,10 @@ struct proto { struct request_sock_ops *rsk_prot; struct timewait_sock_ops *twsk_prot; - struct inet_hashinfo *hashinfo; + union { + struct inet_hashinfo *hashinfo; + struct hlist_head *udp_hash; + } h; struct module *owner; diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 17ad69e90e48..4ca8b0c93c80 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -941,7 +941,7 @@ static struct proto dccp_v4_prot = { .obj_size = sizeof(struct dccp_sock), .rsk_prot = &dccp_request_sock_ops, .twsk_prot = &dccp_timewait_sock_ops, - .hashinfo = &dccp_hashinfo, + .h.hashinfo = &dccp_hashinfo, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_dccp_setsockopt, .compat_getsockopt = compat_dccp_getsockopt, diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 1a5e50b90677..2fec1af1a8c3 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -1126,7 +1126,7 @@ static struct proto dccp_v6_prot = { .obj_size = sizeof(struct dccp6_sock), .rsk_prot = &dccp6_request_sock_ops, .twsk_prot = &dccp6_timewait_sock_ops, - .hashinfo = &dccp_hashinfo, + .h.hashinfo = &dccp_hashinfo, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_dccp_setsockopt, .compat_getsockopt = compat_dccp_getsockopt, diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index cc1a1859a61b..f9c5c4def1ba 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -80,7 +80,7 @@ EXPORT_SYMBOL_GPL(inet_csk_bind_conflict); */ int inet_csk_get_port(struct sock *sk, unsigned short snum) { - struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo; + struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; struct inet_bind_hashbucket *head; struct hlist_node *node; struct inet_bind_bucket *tb; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 1aba606f6bbb..8cd1ad9b9111 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -68,7 +68,7 @@ void inet_bind_hash(struct sock *sk, struct inet_bind_bucket *tb, */ static void __inet_put_port(struct sock *sk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo; + struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; const int bhash = inet_bhashfn(inet_sk(sk)->num, hashinfo->bhash_size); struct inet_bind_hashbucket *head = &hashinfo->bhash[bhash]; struct inet_bind_bucket *tb; @@ -318,7 +318,7 @@ static inline u32 inet_sk_port_offset(const struct sock *sk) void __inet_hash_nolisten(struct sock *sk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo; + struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; struct hlist_head *list; rwlock_t *lock; struct inet_ehash_bucket *head; @@ -339,7 +339,7 @@ EXPORT_SYMBOL_GPL(__inet_hash_nolisten); static void __inet_hash(struct sock *sk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo; + struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; struct hlist_head *list; rwlock_t *lock; @@ -372,7 +372,7 @@ EXPORT_SYMBOL_GPL(inet_hash); void inet_unhash(struct sock *sk) { rwlock_t *lock; - struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo; + struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; if (sk_unhashed(sk)) goto out; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 167a0f557531..1a47719a55ad 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2468,7 +2468,7 @@ struct proto tcp_prot = { .obj_size = sizeof(struct tcp_sock), .twsk_prot = &tcp_timewait_sock_ops, .rsk_prot = &tcp_request_sock_ops, - .hashinfo = &tcp_hashinfo, + .h.hashinfo = &tcp_hashinfo, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_tcp_setsockopt, .compat_getsockopt = compat_tcp_getsockopt, diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 99fd25f7f005..c0c8d2d17682 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -24,7 +24,7 @@ void __inet6_hash(struct sock *sk) { - struct inet_hashinfo *hashinfo = sk->sk_prot->hashinfo; + struct inet_hashinfo *hashinfo = sk->sk_prot->h.hashinfo; struct hlist_head *list; rwlock_t *lock; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 56d0cea7d578..8dd72966ff78 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2172,7 +2172,7 @@ struct proto tcpv6_prot = { .obj_size = sizeof(struct tcp6_sock), .twsk_prot = &tcp6_timewait_sock_ops, .rsk_prot = &tcp6_request_sock_ops, - .hashinfo = &tcp_hashinfo, + .h.hashinfo = &tcp_hashinfo, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_tcp_setsockopt, .compat_getsockopt = compat_tcp_getsockopt, -- cgit v1.2.3 From 6ba5a3c52da00015e739469e3b00cd6d0d4c5c67 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Sat, 22 Mar 2008 16:51:21 -0700 Subject: [UDP]: Make full use of proto.h.udp_hash innovation. After this we have only udp_lib_get_port to get the port and two stubs for ipv4 and ipv6. No difference in udp and udplite except for initialized h.udp_hash member. I tried to find a graceful way to drop the only difference between udp_v4_get_port and udp_v6_get_port (i.e. the rcv_saddr comparison routine), but adding one more callback on the struct proto didn't appear such :( Maybe later. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/udp.h | 2 ++ net/ipv4/udp.c | 22 ++++++++-------------- net/ipv4/udp_impl.h | 6 +----- net/ipv4/udplite.c | 15 ++------------- net/ipv6/udp.c | 5 +++-- net/ipv6/udp_impl.h | 2 ++ net/ipv6/udplite.c | 8 ++------ 7 files changed, 20 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index b4cbdce883d1..635940d374ab 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -125,6 +125,8 @@ static inline void udp_lib_close(struct sock *sk, long timeout) sk_common_release(sk); } +extern int udp_lib_get_port(struct sock *sk, unsigned short snum, + int (*)(const struct sock*,const struct sock*)); /* net/ipv4/udp.c */ extern int udp_get_port(struct sock *sk, unsigned short snum, diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index fa946829d1e8..8c1f5eaafcd2 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -143,18 +143,17 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num, } /** - * __udp_lib_get_port - UDP/-Lite port lookup for IPv4 and IPv6 + * udp_lib_get_port - UDP/-Lite port lookup for IPv4 and IPv6 * * @sk: socket struct in question * @snum: port number to look up - * @udptable: hash list table, must be of UDP_HTABLE_SIZE * @saddr_comp: AF-dependent comparison of bound local IP addresses */ -int __udp_lib_get_port(struct sock *sk, unsigned short snum, - struct hlist_head udptable[], +int udp_lib_get_port(struct sock *sk, unsigned short snum, int (*saddr_comp)(const struct sock *sk1, const struct sock *sk2 ) ) { + struct hlist_head *udptable = sk->sk_prot->h.udp_hash; struct hlist_node *node; struct hlist_head *head; struct sock *sk2; @@ -240,13 +239,7 @@ fail: return error; } -int udp_get_port(struct sock *sk, unsigned short snum, - int (*scmp)(const struct sock *, const struct sock *)) -{ - return __udp_lib_get_port(sk, snum, udp_hash, scmp); -} - -int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) +static int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) { struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2); @@ -255,9 +248,9 @@ int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2) inet1->rcv_saddr == inet2->rcv_saddr )); } -static inline int udp_v4_get_port(struct sock *sk, unsigned short snum) +int udp_v4_get_port(struct sock *sk, unsigned short snum) { - return udp_get_port(sk, snum, ipv4_rcv_saddr_equal); + return udp_lib_get_port(sk, snum, ipv4_rcv_saddr_equal); } /* UDP is nearly always wildcards out the wazoo, it makes no sense to try @@ -1498,6 +1491,7 @@ struct proto udp_prot = { .sysctl_wmem = &sysctl_udp_wmem_min, .sysctl_rmem = &sysctl_udp_rmem_min, .obj_size = sizeof(struct udp_sock), + .h.udp_hash = udp_hash, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_udp_setsockopt, .compat_getsockopt = compat_udp_getsockopt, @@ -1741,12 +1735,12 @@ EXPORT_SYMBOL(udp_disconnect); EXPORT_SYMBOL(udp_hash); EXPORT_SYMBOL(udp_hash_lock); EXPORT_SYMBOL(udp_ioctl); -EXPORT_SYMBOL(udp_get_port); EXPORT_SYMBOL(udp_prot); EXPORT_SYMBOL(udp_sendmsg); EXPORT_SYMBOL(udp_lib_getsockopt); EXPORT_SYMBOL(udp_lib_setsockopt); EXPORT_SYMBOL(udp_poll); +EXPORT_SYMBOL(udp_lib_get_port); #ifdef CONFIG_PROC_FS EXPORT_SYMBOL(udp_proc_register); diff --git a/net/ipv4/udp_impl.h b/net/ipv4/udp_impl.h index 6c55828e41ba..7288bf7977fb 100644 --- a/net/ipv4/udp_impl.h +++ b/net/ipv4/udp_impl.h @@ -8,11 +8,7 @@ extern int __udp4_lib_rcv(struct sk_buff *, struct hlist_head [], int ); extern void __udp4_lib_err(struct sk_buff *, u32, struct hlist_head []); -extern int __udp_lib_get_port(struct sock *sk, unsigned short snum, - struct hlist_head udptable[], - int (*)(const struct sock*,const struct sock*)); -extern int ipv4_rcv_saddr_equal(const struct sock *, const struct sock *); - +extern int udp_v4_get_port(struct sock *sk, unsigned short snum); extern int udp_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen); diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 2469b5104587..8d42e344b043 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -17,17 +17,6 @@ DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics) __read_mostly; struct hlist_head udplite_hash[UDP_HTABLE_SIZE]; -int udplite_get_port(struct sock *sk, unsigned short p, - int (*c)(const struct sock *, const struct sock *)) -{ - return __udp_lib_get_port(sk, p, udplite_hash, c); -} - -static int udplite_v4_get_port(struct sock *sk, unsigned short snum) -{ - return udplite_get_port(sk, snum, ipv4_rcv_saddr_equal); -} - static int udplite_rcv(struct sk_buff *skb) { return __udp4_lib_rcv(skb, udplite_hash, IPPROTO_UDPLITE); @@ -63,8 +52,9 @@ struct proto udplite_prot = { .backlog_rcv = udp_queue_rcv_skb, .hash = udp_lib_hash, .unhash = udp_lib_unhash, - .get_port = udplite_v4_get_port, + .get_port = udp_v4_get_port, .obj_size = sizeof(struct udp_sock), + .h.udp_hash = udplite_hash, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_udp_setsockopt, .compat_getsockopt = compat_udp_getsockopt, @@ -118,4 +108,3 @@ out_register_err: EXPORT_SYMBOL(udplite_hash); EXPORT_SYMBOL(udplite_prot); -EXPORT_SYMBOL(udplite_get_port); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index af619d48ba80..5f5d1218c34e 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -51,9 +51,9 @@ #include #include "udp_impl.h" -static inline int udp_v6_get_port(struct sock *sk, unsigned short snum) +int udp_v6_get_port(struct sock *sk, unsigned short snum) { - return udp_get_port(sk, snum, ipv6_rcv_saddr_equal); + return udp_lib_get_port(sk, snum, ipv6_rcv_saddr_equal); } static struct sock *__udp6_lib_lookup(struct net *net, @@ -1024,6 +1024,7 @@ struct proto udpv6_prot = { .sysctl_wmem = &sysctl_udp_wmem_min, .sysctl_rmem = &sysctl_udp_rmem_min, .obj_size = sizeof(struct udp6_sock), + .h.udp_hash = udp_hash, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_udpv6_setsockopt, .compat_getsockopt = compat_udpv6_getsockopt, diff --git a/net/ipv6/udp_impl.h b/net/ipv6/udp_impl.h index 21be3a83e7bc..321b81a4d418 100644 --- a/net/ipv6/udp_impl.h +++ b/net/ipv6/udp_impl.h @@ -11,6 +11,8 @@ extern int __udp6_lib_rcv(struct sk_buff *, struct hlist_head [], int ); extern void __udp6_lib_err(struct sk_buff *, struct inet6_skb_parm *, int , int , int , __be32 , struct hlist_head []); +extern int udp_v6_get_port(struct sock *sk, unsigned short snum); + extern int udpv6_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen); extern int udpv6_setsockopt(struct sock *sk, int level, int optname, diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 815190be528c..93e52e0d57f2 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -35,11 +35,6 @@ static struct inet6_protocol udplitev6_protocol = { .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -static int udplite_v6_get_port(struct sock *sk, unsigned short snum) -{ - return udplite_get_port(sk, snum, ipv6_rcv_saddr_equal); -} - DEFINE_PROTO_INUSE(udplitev6) struct proto udplitev6_prot = { @@ -58,8 +53,9 @@ struct proto udplitev6_prot = { .backlog_rcv = udpv6_queue_rcv_skb, .hash = udp_lib_hash, .unhash = udp_lib_unhash, - .get_port = udplite_v6_get_port, + .get_port = udp_v6_get_port, .obj_size = sizeof(struct udp6_sock), + .h.udp_hash = udplite_hash, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_udpv6_setsockopt, .compat_getsockopt = compat_udpv6_getsockopt, -- cgit v1.2.3 From fc8717baa8f52dd8d1b90df9008300ef3ec794ed Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Sat, 22 Mar 2008 16:56:51 -0700 Subject: [RAW]: Add raw_hashinfo member on struct proto. Sorry for the patch sequence confusion :| but I found that the similar thing can be done for raw sockets easily too late. Expand the proto.h union with the raw_hashinfo member and use it in raw_prot and rawv6_prot. This allows to drop the protocol specific versions of hash and unhash callbacks. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/raw.h | 4 ++-- include/net/sock.h | 2 ++ net/ipv4/raw.c | 22 ++++++++-------------- net/ipv6/raw.c | 16 +++------------- 4 files changed, 15 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/include/net/raw.h b/include/net/raw.h index 1828f81fe374..6c14a656357a 100644 --- a/include/net/raw.h +++ b/include/net/raw.h @@ -53,7 +53,7 @@ int raw_seq_open(struct inode *ino, struct file *file, #endif -void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h); -void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h); +void raw_hash_sk(struct sock *sk); +void raw_unhash_sk(struct sock *sk); #endif /* _RAW_H */ diff --git a/include/net/sock.h b/include/net/sock.h index c3175c400b79..b433b1ed203d 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -504,6 +504,7 @@ extern int sk_wait_data(struct sock *sk, long *timeo); struct request_sock_ops; struct timewait_sock_ops; struct inet_hashinfo; +struct raw_hashinfo; /* Networking protocol blocks we attach to sockets. * socket layer -> transport layer interface @@ -589,6 +590,7 @@ struct proto { union { struct inet_hashinfo *hashinfo; struct hlist_head *udp_hash; + struct raw_hashinfo *raw_hash; } h; struct module *owner; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index e10ef6c2202a..b433b485a883 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -84,8 +84,9 @@ static struct raw_hashinfo raw_v4_hashinfo = { .lock = __RW_LOCK_UNLOCKED(raw_v4_hashinfo.lock), }; -void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h) +void raw_hash_sk(struct sock *sk) { + struct raw_hashinfo *h = sk->sk_prot->h.raw_hash; struct hlist_head *head; head = &h->ht[inet_sk(sk)->num & (RAW_HTABLE_SIZE - 1)]; @@ -97,8 +98,10 @@ void raw_hash_sk(struct sock *sk, struct raw_hashinfo *h) } EXPORT_SYMBOL_GPL(raw_hash_sk); -void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h) +void raw_unhash_sk(struct sock *sk) { + struct raw_hashinfo *h = sk->sk_prot->h.raw_hash; + write_lock_bh(&h->lock); if (sk_del_node_init(sk)) sock_prot_inuse_add(sk->sk_prot, -1); @@ -106,16 +109,6 @@ void raw_unhash_sk(struct sock *sk, struct raw_hashinfo *h) } EXPORT_SYMBOL_GPL(raw_unhash_sk); -static void raw_v4_hash(struct sock *sk) -{ - raw_hash_sk(sk, &raw_v4_hashinfo); -} - -static void raw_v4_unhash(struct sock *sk) -{ - raw_unhash_sk(sk, &raw_v4_hashinfo); -} - static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk, unsigned short num, __be32 raddr, __be32 laddr, int dif) { @@ -841,9 +834,10 @@ struct proto raw_prot = { .recvmsg = raw_recvmsg, .bind = raw_bind, .backlog_rcv = raw_rcv_skb, - .hash = raw_v4_hash, - .unhash = raw_v4_unhash, + .hash = raw_hash_sk, + .unhash = raw_unhash_sk, .obj_size = sizeof(struct raw_sock), + .h.raw_hash = &raw_v4_hashinfo, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_raw_setsockopt, .compat_getsockopt = compat_raw_getsockopt, diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index d6afa0216759..a9e4235157a2 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -65,17 +65,6 @@ static struct raw_hashinfo raw_v6_hashinfo = { .lock = __RW_LOCK_UNLOCKED(raw_v6_hashinfo.lock), }; -static void raw_v6_hash(struct sock *sk) -{ - raw_hash_sk(sk, &raw_v6_hashinfo); -} - -static void raw_v6_unhash(struct sock *sk) -{ - raw_unhash_sk(sk, &raw_v6_hashinfo); -} - - static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, unsigned short num, struct in6_addr *loc_addr, struct in6_addr *rmt_addr, int dif) @@ -1201,9 +1190,10 @@ struct proto rawv6_prot = { .recvmsg = rawv6_recvmsg, .bind = rawv6_bind, .backlog_rcv = rawv6_rcv_skb, - .hash = raw_v6_hash, - .unhash = raw_v6_unhash, + .hash = raw_hash_sk, + .unhash = raw_unhash_sk, .obj_size = sizeof(struct raw6_sock), + .h.raw_hash = &raw_v6_hashinfo, #ifdef CONFIG_COMPAT .compat_setsockopt = compat_rawv6_setsockopt, .compat_getsockopt = compat_rawv6_getsockopt, -- cgit v1.2.3 From 0098b7273e968fb9989a6e1e4e4c024cd081fe0d Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Sat, 22 Mar 2008 17:18:47 -0700 Subject: [NET]: NPROTO is redundant; it's equal to AF_MAX/PF_MAX. DaveM pointed out NPROTO exposed to userspace, so keep it around, just make sure it stays in sync. Signed-off-by: Rusty Russell Signed-off-by: David S. Miller --- include/linux/net.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/net.h b/include/linux/net.h index c414d90e647b..71f7dd559285 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -19,6 +19,7 @@ #define _LINUX_NET_H #include +#include #include struct poll_table_struct; @@ -26,7 +27,7 @@ struct pipe_inode_info; struct inode; struct net; -#define NPROTO 34 /* should be enough for now.. */ +#define NPROTO AF_MAX #define SYS_SOCKET 1 /* sys_socket(2) */ #define SYS_BIND 2 /* sys_bind(2) */ -- cgit v1.2.3 From 7d164be8aa4392fe55474f4608547f2097e07c41 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 23 Mar 2008 22:03:56 -0700 Subject: [NET]: include/net/route.h - remove duplicate include Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/net/route.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/net/route.h b/include/net/route.h index eadad5901429..28dba925663c 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -34,7 +34,6 @@ #include #include #include -#include #ifndef __KERNEL__ #warning This file is not supposed to be used outside of kernel. -- cgit v1.2.3 From 414f69d8a6ff0b30e7ea5ce10534b19f851e172e Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 23 Mar 2008 22:04:31 -0700 Subject: [NET]: include/linux/atalk.h - remove duplicate include Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/linux/atalk.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/atalk.h b/include/linux/atalk.h index ced8a1ed080c..e9ebac2e2ecc 100644 --- a/include/linux/atalk.h +++ b/include/linux/atalk.h @@ -85,8 +85,6 @@ static inline struct atalk_sock *at_sk(struct sock *sk) return (struct atalk_sock *)sk; } -#include - struct ddpehdr { __be16 deh_len_hops; /* lower 10 bits are length, next 4 - hops */ __be16 deh_sum; -- cgit v1.2.3 From cc32e05416b4023a5466a2f66e3c02236a771c5b Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 23 Mar 2008 22:05:44 -0700 Subject: [NET]: include/linux/igmp.h - remove duplicate include Removed duplicate #include Combined #ifdef __KERNEL__ blocks Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/linux/igmp.h | 38 +++++++++++++++++--------------------- 1 file changed, 17 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/linux/igmp.h b/include/linux/igmp.h index f510e7e382a8..f5a1a0db2e8e 100644 --- a/include/linux/igmp.h +++ b/include/linux/igmp.h @@ -80,27 +80,6 @@ struct igmpv3_query { __be32 srcs[0]; }; -#ifdef __KERNEL__ -#include - -static inline struct igmphdr *igmp_hdr(const struct sk_buff *skb) -{ - return (struct igmphdr *)skb_transport_header(skb); -} - -static inline struct igmpv3_report * - igmpv3_report_hdr(const struct sk_buff *skb) -{ - return (struct igmpv3_report *)skb_transport_header(skb); -} - -static inline struct igmpv3_query * - igmpv3_query_hdr(const struct sk_buff *skb) -{ - return (struct igmpv3_query *)skb_transport_header(skb); -} -#endif - #define IGMP_HOST_MEMBERSHIP_QUERY 0x11 /* From RFC1112 */ #define IGMP_HOST_MEMBERSHIP_REPORT 0x12 /* Ditto */ #define IGMP_DVMRP 0x13 /* DVMRP routing */ @@ -151,6 +130,23 @@ static inline struct igmpv3_query * #include #include +static inline struct igmphdr *igmp_hdr(const struct sk_buff *skb) +{ + return (struct igmphdr *)skb_transport_header(skb); +} + +static inline struct igmpv3_report * + igmpv3_report_hdr(const struct sk_buff *skb) +{ + return (struct igmpv3_report *)skb_transport_header(skb); +} + +static inline struct igmpv3_query * + igmpv3_query_hdr(const struct sk_buff *skb) +{ + return (struct igmpv3_query *)skb_transport_header(skb); +} + extern int sysctl_igmp_max_memberships; extern int sysctl_igmp_max_msf; -- cgit v1.2.3 From 310afe86af8ddd96a06b75aa61ef1af233f80e89 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sun, 23 Mar 2008 22:06:51 -0700 Subject: [NET]: include/linux/udp.h - remove duplicate include Remove duplicate #include Combine #ifdef __KERNEL__ blocks Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/linux/udp.h | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/udp.h b/include/linux/udp.h index 1e7b7cb5703b..581ca2c14c52 100644 --- a/include/linux/udp.h +++ b/include/linux/udp.h @@ -26,15 +26,6 @@ struct udphdr { __sum16 check; }; -#ifdef __KERNEL__ -#include - -static inline struct udphdr *udp_hdr(const struct sk_buff *skb) -{ - return (struct udphdr *)skb_transport_header(skb); -} -#endif - /* UDP socket options */ #define UDP_CORK 1 /* Never send partially complete segments */ #define UDP_ENCAP 100 /* Set the socket to accept encapsulated packets */ @@ -45,9 +36,14 @@ static inline struct udphdr *udp_hdr(const struct sk_buff *skb) #define UDP_ENCAP_L2TPINUDP 3 /* rfc2661 */ #ifdef __KERNEL__ -#include - #include +#include + +static inline struct udphdr *udp_hdr(const struct sk_buff *skb) +{ + return (struct udphdr *)skb_transport_header(skb); +} + #define UDP_HTABLE_SIZE 128 struct udp_sock { -- cgit v1.2.3 From 2051f11fb86b0056fec440fe7e9fa8370d60a5c6 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 23 Mar 2008 22:21:28 -0700 Subject: [TCP]: Shrink syncookie_secret by 8 byte. the first u32 copied from syncookie_secret is overwritten by the minute-counter four lines below. After adjusting the destination address, the size of syncookie_secret can be reduced accordingly. AFAICS, the only other user of syncookie_secret[] is the ipv6 syncookie support. Because ipv6 syncookies only grab 44 bytes from syncookie_secret[], this shouldn't affect them in any way. With fixes from Glenn Griffin. Signed-off-by: Florian Westphal Acked-by: Glenn Griffin Signed-off-by: David S. Miller --- include/net/tcp.h | 2 +- net/ipv4/syncookies.c | 4 ++-- net/ipv6/syncookies.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 67cc3956d29c..723b36851dde 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -436,7 +436,7 @@ extern int tcp_disconnect(struct sock *sk, int flags); extern void tcp_unhash(struct sock *sk); /* From syncookies.c */ -extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; +extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS]; extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt); extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index 4704f27f6c0b..abc752d45cf7 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -21,7 +21,7 @@ extern int sysctl_tcp_syncookies; -__u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; +__u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS]; EXPORT_SYMBOL(syncookie_secret); static __init int init_syncookies(void) @@ -41,7 +41,7 @@ static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport, { __u32 *tmp = __get_cpu_var(cookie_scratch); - memcpy(tmp + 3, syncookie_secret[c], sizeof(syncookie_secret[c])); + memcpy(tmp + 4, syncookie_secret[c], sizeof(syncookie_secret[c])); tmp[0] = (__force u32)saddr; tmp[1] = (__force u32)daddr; tmp[2] = ((__force u32)sport << 16) + (__force u32)dport; diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index 827c5aa7524c..3a622e7abc02 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -22,7 +22,7 @@ #include extern int sysctl_tcp_syncookies; -extern __u32 syncookie_secret[2][16-3+SHA_DIGEST_WORDS]; +extern __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS]; #define COOKIEBITS 24 /* Upper bits store count */ #define COOKIEMASK (((__u32)1 << COOKIEBITS) - 1) -- cgit v1.2.3 From 80445cfb28a6b093540582b68d9ae928bf34cfe7 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sun, 23 Mar 2008 22:47:08 -0700 Subject: [SCTP]: Remove redundant wrapper functions. sctp_datamsg_free and sctp_datamsg_track are just aliases for sctp_datamsg_put and sctp_chunk_hold, respectively. Saves 32 Bytes on x86. Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 2 -- net/sctp/chunk.c | 16 +--------------- net/sctp/socket.c | 4 ++-- 3 files changed, 3 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 8966599ddb9f..0ce0443c5b79 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -637,8 +637,6 @@ struct sctp_datamsg *sctp_datamsg_from_user(struct sctp_association *, struct sctp_sndrcvinfo *, struct msghdr *, int len); void sctp_datamsg_put(struct sctp_datamsg *); -void sctp_datamsg_free(struct sctp_datamsg *); -void sctp_datamsg_track(struct sctp_chunk *); void sctp_chunk_fail(struct sctp_chunk *, int error); int sctp_chunk_abandoned(struct sctp_chunk *); diff --git a/net/sctp/chunk.c b/net/sctp/chunk.c index e1f355080026..ed857643e3ff 100644 --- a/net/sctp/chunk.c +++ b/net/sctp/chunk.c @@ -136,20 +136,6 @@ void sctp_datamsg_put(struct sctp_datamsg *msg) 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) { @@ -295,7 +281,7 @@ errout: chunk = list_entry(pos, struct sctp_chunk, frag_list); sctp_chunk_free(chunk); } - sctp_datamsg_free(msg); + sctp_datamsg_put(msg); return NULL; } diff --git a/net/sctp/socket.c b/net/sctp/socket.c index a3138a0fe2c5..76c747056dd7 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1729,7 +1729,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk, /* 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); + sctp_chunk_hold(chunk); /* Do accounting for the write space. */ sctp_set_owner_w(chunk); @@ -1748,7 +1748,7 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk, SCTP_DEBUG_PRINTK("We sent primitively.\n"); } - sctp_datamsg_free(datamsg); + sctp_datamsg_put(datamsg); if (err) goto out_free; else -- cgit v1.2.3 From 0e6bd4a1c6c3881c9ed82985ecb9824d4450c4ba Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Mon, 24 Mar 2008 15:29:23 -0700 Subject: [NETNS]: Add namespace parameter to ip_options_compile. ip_options_compile uses inet_addr_type which requires a namespace. The packet argument is optional, so parameter is the only way to obtain it. Pass the init_net there for now. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/ip.h | 3 ++- net/ipv4/ip_input.c | 2 +- net/ipv4/ip_options.c | 7 ++++--- 3 files changed, 7 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index 9f50d4f1f157..bcc3afa3df36 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -347,7 +347,8 @@ extern int ip_forward(struct sk_buff *skb); extern void ip_options_build(struct sk_buff *skb, struct ip_options *opt, __be32 daddr, struct rtable *rt, int is_frag); extern int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb); extern void ip_options_fragment(struct sk_buff *skb); -extern int ip_options_compile(struct ip_options *opt, struct sk_buff *skb); +extern int ip_options_compile(struct net *net, + struct ip_options *opt, struct sk_buff *skb); extern int ip_options_get(struct ip_options **optp, unsigned char *data, int optlen); extern int ip_options_get_from_user(struct ip_options **optp, diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index e3a0c78fa7bf..f3a7a08a463f 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -286,7 +286,7 @@ static inline int ip_rcv_options(struct sk_buff *skb) opt = &(IPCB(skb)->opt); opt->optlen = iph->ihl*4 - sizeof(struct iphdr); - if (ip_options_compile(opt, skb)) { + if (ip_options_compile(&init_net, opt, skb)) { IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); goto drop; } diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index aeed4e5858ec..f0949b42e79e 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -248,7 +248,8 @@ void ip_options_fragment(struct sk_buff * skb) * If opt == NULL, then skb->data should point to IP header. */ -int ip_options_compile(struct ip_options * opt, struct sk_buff * skb) +int ip_options_compile(struct net *net, + struct ip_options * opt, struct sk_buff * skb) { int l; unsigned char * iph; @@ -389,7 +390,7 @@ int ip_options_compile(struct ip_options * opt, struct sk_buff * skb) { __be32 addr; memcpy(&addr, &optptr[optptr[2]-1], 4); - if (inet_addr_type(&init_net, addr) == RTN_UNICAST) + if (inet_addr_type(net, addr) == RTN_UNICAST) break; if (skb) timeptr = (__be32*)&optptr[optptr[2]+3]; @@ -512,7 +513,7 @@ static int ip_options_get_finish(struct ip_options **optp, while (optlen & 3) opt->__data[optlen++] = IPOPT_END; opt->optlen = optlen; - if (optlen && ip_options_compile(opt, NULL)) { + if (optlen && ip_options_compile(&init_net, opt, NULL)) { kfree(opt); return -EINVAL; } -- cgit v1.2.3 From f2c4802b3fdfb0d9596d932ca2af0ef6f8d60491 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Mon, 24 Mar 2008 15:29:55 -0700 Subject: [NETNS]: Add namespace parameter to ip_options_get(...). Pass the init_net there for now. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/ip.h | 4 ++-- net/ipv4/ip_options.c | 14 ++++++++------ net/ipv4/ip_sockglue.c | 4 ++-- 3 files changed, 12 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index bcc3afa3df36..531270dc48a6 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -349,9 +349,9 @@ extern int ip_options_echo(struct ip_options *dopt, struct sk_buff *skb); extern void ip_options_fragment(struct sk_buff *skb); extern int ip_options_compile(struct net *net, struct ip_options *opt, struct sk_buff *skb); -extern int ip_options_get(struct ip_options **optp, +extern int ip_options_get(struct net *net, struct ip_options **optp, unsigned char *data, int optlen); -extern int ip_options_get_from_user(struct ip_options **optp, +extern int ip_options_get_from_user(struct net *net, struct ip_options **optp, unsigned char __user *data, int optlen); extern void ip_options_undo(struct ip_options * opt); extern void ip_forward_options(struct sk_buff *skb); diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index f0949b42e79e..59f7ddfb29bf 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -507,13 +507,13 @@ static struct ip_options *ip_options_get_alloc(const int optlen) GFP_KERNEL); } -static int ip_options_get_finish(struct ip_options **optp, +static int ip_options_get_finish(struct net *net, struct ip_options **optp, struct ip_options *opt, int optlen) { while (optlen & 3) opt->__data[optlen++] = IPOPT_END; opt->optlen = optlen; - if (optlen && ip_options_compile(&init_net, opt, NULL)) { + if (optlen && ip_options_compile(net, opt, NULL)) { kfree(opt); return -EINVAL; } @@ -522,7 +522,8 @@ static int ip_options_get_finish(struct ip_options **optp, return 0; } -int ip_options_get_from_user(struct ip_options **optp, unsigned char __user *data, int optlen) +int ip_options_get_from_user(struct net *net, struct ip_options **optp, + unsigned char __user *data, int optlen) { struct ip_options *opt = ip_options_get_alloc(optlen); @@ -532,10 +533,11 @@ int ip_options_get_from_user(struct ip_options **optp, unsigned char __user *dat kfree(opt); return -EFAULT; } - return ip_options_get_finish(optp, opt, optlen); + return ip_options_get_finish(net, optp, opt, optlen); } -int ip_options_get(struct ip_options **optp, unsigned char *data, int optlen) +int ip_options_get(struct net *net, struct ip_options **optp, + unsigned char *data, int optlen) { struct ip_options *opt = ip_options_get_alloc(optlen); @@ -543,7 +545,7 @@ int ip_options_get(struct ip_options **optp, unsigned char *data, int optlen) return -ENOMEM; if (optlen) memcpy(opt->__data, data, optlen); - return ip_options_get_finish(optp, opt, optlen); + return ip_options_get_finish(net, optp, opt, optlen); } void ip_forward_options(struct sk_buff *skb) diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index bb3cbe5ec36d..1b86a50269bc 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -176,7 +176,7 @@ int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc) switch (cmsg->cmsg_type) { case IP_RETOPTS: err = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr)); - err = ip_options_get(&ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40); + err = ip_options_get(&init_net, &ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40); if (err) return err; break; @@ -449,7 +449,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, struct ip_options * opt = NULL; if (optlen > 40 || optlen < 0) goto e_inval; - err = ip_options_get_from_user(&opt, optval, optlen); + err = ip_options_get_from_user(&init_net, &opt, optval, optlen); if (err) break; if (inet->is_icsk) { -- cgit v1.2.3 From 7a6adb92fe301c10ca4dbd0d9f2422f5880595e7 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Mon, 24 Mar 2008 15:30:27 -0700 Subject: [NETNS]: Add namespace parameter to ip_cmsg_send. Pass the init_net there for now. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/ip.h | 3 ++- net/ipv4/ip_sockglue.c | 4 ++-- net/ipv4/raw.c | 2 +- net/ipv4/udp.c | 2 +- 4 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/ip.h b/include/net/ip.h index 531270dc48a6..6d7bcd5e62d4 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -362,7 +362,8 @@ extern int ip_options_rcv_srr(struct sk_buff *skb); */ extern void ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb); -extern int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc); +extern int ip_cmsg_send(struct net *net, + struct msghdr *msg, struct ipcm_cookie *ipc); extern int ip_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen); extern int ip_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen); extern int compat_ip_setsockopt(struct sock *sk, int level, diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 1b86a50269bc..0857f2d042cd 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -163,7 +163,7 @@ void ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb) ip_cmsg_recv_security(msg, skb); } -int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc) +int ip_cmsg_send(struct net *net, struct msghdr *msg, struct ipcm_cookie *ipc) { int err; struct cmsghdr *cmsg; @@ -176,7 +176,7 @@ int ip_cmsg_send(struct msghdr *msg, struct ipcm_cookie *ipc) switch (cmsg->cmsg_type) { case IP_RETOPTS: err = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr)); - err = ip_options_get(&init_net, &ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40); + err = ip_options_get(net, &ipc->opt, CMSG_DATA(cmsg), err < 40 ? err : 40); if (err) return err; break; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index b433b485a883..7d29a05bf887 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -499,7 +499,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.oif = sk->sk_bound_dev_if; if (msg->msg_controllen) { - err = ip_cmsg_send(msg, &ipc); + err = ip_cmsg_send(&init_net, msg, &ipc); if (err) goto out; if (ipc.opt) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 9cd48b0bb7d4..d142b87e9eb8 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -607,7 +607,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.oif = sk->sk_bound_dev_if; if (msg->msg_controllen) { - err = ip_cmsg_send(msg, &ipc); + err = ip_cmsg_send(&init_net, msg, &ipc); if (err) return err; if (ipc.opt) -- cgit v1.2.3 From f145049a06f470d0489f47cb83ff3ccb2a0de622 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Mon, 24 Mar 2008 15:33:00 -0700 Subject: [NETNS]: Drop packets in the non-initial namespace on the per/protocol basis. IP layer now can handle multiple namespaces normally. So, process such packets normally and drop them only if the transport layer is not aware about namespaces. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/protocol.h | 3 ++- net/ipv4/ip_input.c | 8 ++++---- 2 files changed, 6 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/protocol.h b/include/net/protocol.h index ad8c584233a6..8d024d7cb741 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -39,7 +39,8 @@ struct net_protocol { int (*gso_send_check)(struct sk_buff *skb); struct sk_buff *(*gso_segment)(struct sk_buff *skb, int features); - int no_policy; + unsigned int no_policy:1, + netns_ok:1; }; #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE) diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index eb1fa27dc0c4..2aeea5d15425 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -199,6 +199,8 @@ int ip_call_ra_chain(struct sk_buff *skb) static int ip_local_deliver_finish(struct sk_buff *skb) { + struct net *net = skb->dev->nd_net; + __skb_pull(skb, ip_hdrlen(skb)); /* Point into the IP datagram, just past the header. */ @@ -214,7 +216,8 @@ static int ip_local_deliver_finish(struct sk_buff *skb) raw = raw_local_deliver(skb, protocol); hash = protocol & (MAX_INET_PROTOS - 1); - if ((ipprot = rcu_dereference(inet_protos[hash])) != NULL) { + ipprot = rcu_dereference(inet_protos[hash]); + if (ipprot != NULL && (net == &init_net || ipprot->netns_ok)) { int ret; if (!ipprot->no_policy) { @@ -375,9 +378,6 @@ int ip_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct iphdr *iph; u32 len; - if (dev->nd_net != &init_net) - goto drop; - /* When the interface is in promisc. mode, drop all the crap * that it receives, do not try to analyse it. */ -- cgit v1.2.3 From 9bb182a7007515239091b237fe7169b1328a61d3 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Fri, 22 Feb 2008 14:48:22 +0900 Subject: [XFRM] MIP6: Fix address keys for routing search. Each MIPv6 XFRM state (DSTOPT/RH2) holds either destination or source address to be mangled in the IPv6 header (that is "CoA"). On Inter-MN communication after both nodes binds each other, they use route optimized traffic two MIPv6 states applied, and both source and destination address in the IPv6 header are replaced by the states respectively. The packet format is correct, however, next-hop routing search are not. This patch fixes it by remembering address pairs for later states. Based on patch from Masahide NAKAMURA . Signed-off-by: Masahide NAKAMURA Signed-off-by: YOSHIFUJI Hideaki --- include/net/xfrm.h | 17 +++++++++++++++++ net/xfrm/xfrm_policy.c | 49 ++++++++++++++++++++++++++++++++++++++++--------- 2 files changed, 57 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/xfrm.h b/include/net/xfrm.h index c435620dbb37..bed7d43932f6 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -1045,6 +1045,23 @@ xfrm_address_t *xfrm_flowi_saddr(struct flowi *fl, unsigned short family) return NULL; } +static __inline__ +void xfrm_flowi_addr_get(struct flowi *fl, + xfrm_address_t *saddr, xfrm_address_t *daddr, + unsigned short family) +{ + switch(family) { + case AF_INET: + memcpy(&saddr->a4, &fl->fl4_src, sizeof(saddr->a4)); + memcpy(&daddr->a4, &fl->fl4_dst, sizeof(daddr->a4)); + break; + case AF_INET6: + ipv6_addr_copy((struct in6_addr *)&saddr->a6, &fl->fl6_src); + ipv6_addr_copy((struct in6_addr *)&daddr->a6, &fl->fl6_dst); + break; + } +} + static __inline__ int __xfrm4_state_addr_check(struct xfrm_state *x, xfrm_address_t *daddr, xfrm_address_t *saddr) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index bae94a8031a2..8e588f20c60c 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -97,25 +97,52 @@ int xfrm_selector_match(struct xfrm_selector *sel, struct flowi *fl, return 0; } +static inline struct dst_entry *__xfrm_dst_lookup(int tos, + xfrm_address_t *saddr, + xfrm_address_t *daddr, + int family) +{ + struct xfrm_policy_afinfo *afinfo; + struct dst_entry *dst; + + afinfo = xfrm_policy_get_afinfo(family); + if (unlikely(afinfo == NULL)) + return ERR_PTR(-EAFNOSUPPORT); + + dst = afinfo->dst_lookup(tos, saddr, daddr); + + xfrm_policy_put_afinfo(afinfo); + + return dst; +} + static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x, int tos, + xfrm_address_t *prev_saddr, + xfrm_address_t *prev_daddr, int family) { xfrm_address_t *saddr = &x->props.saddr; xfrm_address_t *daddr = &x->id.daddr; - struct xfrm_policy_afinfo *afinfo; struct dst_entry *dst; - if (x->type->flags & XFRM_TYPE_LOCAL_COADDR) + if (x->type->flags & XFRM_TYPE_LOCAL_COADDR) { saddr = x->coaddr; - if (x->type->flags & XFRM_TYPE_REMOTE_COADDR) + daddr = prev_daddr; + } + if (x->type->flags & XFRM_TYPE_REMOTE_COADDR) { + saddr = prev_saddr; daddr = x->coaddr; + } - afinfo = xfrm_policy_get_afinfo(family); - if (unlikely(afinfo == NULL)) - return ERR_PTR(-EAFNOSUPPORT); + dst = __xfrm_dst_lookup(tos, saddr, daddr, family); + + if (!IS_ERR(dst)) { + if (prev_saddr != saddr) + memcpy(prev_saddr, saddr, sizeof(*prev_saddr)); + if (prev_daddr != daddr) + memcpy(prev_daddr, daddr, sizeof(*prev_daddr)); + } - dst = afinfo->dst_lookup(tos, saddr, daddr); - xfrm_policy_put_afinfo(afinfo); return dst; } @@ -1354,6 +1381,9 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy, int trailer_len = 0; int tos; int family = policy->selector.family; + xfrm_address_t saddr, daddr; + + xfrm_flowi_addr_get(fl, &saddr, &daddr, family); tos = xfrm_get_tos(fl, family); err = tos; @@ -1384,7 +1414,8 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy, if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) { family = xfrm[i]->props.family; - dst = xfrm_dst_lookup(xfrm[i], tos, family); + dst = xfrm_dst_lookup(xfrm[i], tos, &saddr, &daddr, + family); err = PTR_ERR(dst); if (IS_ERR(dst)) goto put_states; -- cgit v1.2.3 From c8cdaf998df221b01134a051aba38c570105061b Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 10 Mar 2008 04:30:37 -0400 Subject: [IPV4,IPV6]: Share cork.rt between IPv4 and IPv6. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/ipv6.h | 1 - include/net/inet_sock.h | 2 +- net/ipv4/ip_output.c | 14 ++++++-------- net/ipv6/ip6_output.c | 12 ++++++------ 4 files changed, 13 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 4aaefc349a4b..2102d8b67c01 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -315,7 +315,6 @@ struct ipv6_pinfo { struct sk_buff *pktoptions; struct { struct ipv6_txoptions *opt; - struct rt6_info *rt; int hop_limit; int tclass; } cork; diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h index b6db16d2766a..a42cd63d241a 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -136,7 +136,7 @@ struct inet_sock { unsigned int flags; unsigned int fragsize; struct ip_options *opt; - struct rtable *rt; + struct dst_entry *dst; int length; /* Total length of all frames */ __be32 addr; struct flowi fl; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 349fae58c1a3..913266cd9902 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -825,7 +825,7 @@ int ip_append_data(struct sock *sk, inet->cork.fragsize = mtu = inet->pmtudisc == IP_PMTUDISC_PROBE ? rt->u.dst.dev->mtu : dst_mtu(rt->u.dst.path); - inet->cork.rt = rt; + inet->cork.dst = &rt->u.dst; inet->cork.length = 0; sk->sk_sndmsg_page = NULL; sk->sk_sndmsg_off = 0; @@ -834,7 +834,7 @@ int ip_append_data(struct sock *sk, transhdrlen += exthdrlen; } } else { - rt = inet->cork.rt; + rt = (struct rtable *)inet->cork.dst; if (inet->cork.flags & IPCORK_OPT) opt = inet->cork.opt; @@ -1083,7 +1083,7 @@ ssize_t ip_append_page(struct sock *sk, struct page *page, if (skb_queue_empty(&sk->sk_write_queue)) return -EINVAL; - rt = inet->cork.rt; + rt = (struct rtable *)inet->cork.dst; if (inet->cork.flags & IPCORK_OPT) opt = inet->cork.opt; @@ -1208,10 +1208,8 @@ static void ip_cork_release(struct inet_sock *inet) inet->cork.flags &= ~IPCORK_OPT; kfree(inet->cork.opt); inet->cork.opt = NULL; - if (inet->cork.rt) { - ip_rt_put(inet->cork.rt); - inet->cork.rt = NULL; - } + dst_release(inet->cork.dst); + inet->cork.dst = NULL; } /* @@ -1224,7 +1222,7 @@ int ip_push_pending_frames(struct sock *sk) struct sk_buff **tail_skb; struct inet_sock *inet = inet_sk(sk); struct ip_options *opt = NULL; - struct rtable *rt = inet->cork.rt; + struct rtable *rt = (struct rtable *)inet->cork.dst; struct iphdr *iph; __be16 df = 0; __u8 ttl; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 98762fde2b65..ed6482667a25 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1115,7 +1115,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, /* need source address above miyazawa*/ } dst_hold(&rt->u.dst); - np->cork.rt = rt; + inet->cork.dst = &rt->u.dst; inet->cork.fl = *fl; np->cork.hop_limit = hlimit; np->cork.tclass = tclass; @@ -1136,7 +1136,7 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, length += exthdrlen; transhdrlen += exthdrlen; } else { - rt = np->cork.rt; + rt = (struct rt6_info *)inet->cork.dst; fl = &inet->cork.fl; if (inet->cork.flags & IPCORK_OPT) opt = np->cork.opt; @@ -1381,9 +1381,9 @@ static void ip6_cork_release(struct inet_sock *inet, struct ipv6_pinfo *np) inet->cork.flags &= ~IPCORK_OPT; kfree(np->cork.opt); np->cork.opt = NULL; - if (np->cork.rt) { - dst_release(&np->cork.rt->u.dst); - np->cork.rt = NULL; + if (inet->cork.dst) { + dst_release(inet->cork.dst); + inet->cork.dst = NULL; inet->cork.flags &= ~IPCORK_ALLFRAG; } memset(&inet->cork.fl, 0, sizeof(inet->cork.fl)); @@ -1398,7 +1398,7 @@ int ip6_push_pending_frames(struct sock *sk) struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6hdr *hdr; struct ipv6_txoptions *opt = np->cork.opt; - struct rt6_info *rt = np->cork.rt; + struct rt6_info *rt = (struct rt6_info *)inet->cork.dst; struct flowi *fl = &inet->cork.fl; unsigned char proto = fl->proto; int err = 0; -- cgit v1.2.3 From 4725474584d6aa2f07b3d47442dfbc4f6544f65e Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 10 Mar 2008 04:41:33 -0400 Subject: [IPV6]: Convert cork.hop_limit and cork.tclass into u8 instead of int. Values of those fields are always between 0 and 255 (inclusive), so use u8 and save some memory on 32bit systems. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/ipv6.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 2102d8b67c01..9b59e37afad9 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -315,8 +315,8 @@ struct ipv6_pinfo { struct sk_buff *pktoptions; struct { struct ipv6_txoptions *opt; - int hop_limit; - int tclass; + u8 hop_limit; + u8 tclass; } cork; }; -- cgit v1.2.3 From 6b75d0908185bf853b188afa6f269426f6554c5b Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 10 Mar 2008 06:00:30 -0400 Subject: [IPV6]: Optimize hop-limit determination. Last part of hop-limit determination is always: hoplimit = dst_metric(dst, RTAX_HOPLIMIT); if (hoplimit < 0) hoplimit = ipv6_get_hoplimit(dst->dev). Let's consolidate it as ip6_dst_hoplimit(dst). Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 2 -- include/net/ip6_route.h | 2 ++ net/ipv6/icmp.c | 8 ++------ net/ipv6/ip6_output.c | 4 +--- net/ipv6/ipv6_sockglue.c | 4 +--- net/ipv6/raw.c | 4 +--- net/ipv6/route.c | 20 +++++++++++--------- net/ipv6/udp.c | 4 +--- 8 files changed, 19 insertions(+), 29 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 232da20e7171..edcb4bbaab7d 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -123,8 +123,6 @@ extern int ipv6_is_mld(struct sk_buff *skb, int nexthdr); extern void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len); -extern int ipv6_get_hoplimit(struct net_device *dev); - /* * anycast prototypes (anycast.c) */ diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 0e2895c8b270..5c3b67c86aef 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -88,6 +88,8 @@ extern struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, const struct in6_addr *addr, int anycast); +extern int ip6_dst_hoplimit(struct dst_entry *dst); + /* * support functions for ND * diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 6b5391ab8346..86332417b402 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -464,9 +464,7 @@ route_done: else hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); tclass = np->tclass; if (tclass < 0) @@ -560,9 +558,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) else hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); tclass = np->tclass; if (tclass < 0) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index ed6482667a25..2a4f08c8a02d 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -237,9 +237,7 @@ int ip6_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, if (np) hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); tclass = -1; if (np) diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index c11c76cab371..8e29fb1d1df6 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -904,9 +904,7 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, dst = sk_dst_get(sk); if (dst) { if (val < 0) - val = dst_metric(dst, RTAX_HOPLIMIT); - if (val < 0) - val = ipv6_get_hoplimit(dst->dev); + val = ip6_dst_hoplimit(dst); dst_release(dst); } if (val < 0) diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index a9e4235157a2..548d0763f4d3 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -885,9 +885,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, else hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); } if (tclass < 0) { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index a4b5aee0f68a..aa3f08718e44 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1034,15 +1034,17 @@ static int ipv6_get_mtu(struct net_device *dev) return mtu; } -int ipv6_get_hoplimit(struct net_device *dev) -{ - int hoplimit = ipv6_devconf.hop_limit; - struct inet6_dev *idev; - - idev = in6_dev_get(dev); - if (idev) { - hoplimit = idev->cnf.hop_limit; - in6_dev_put(idev); +int ip6_dst_hoplimit(struct dst_entry *dst) +{ + int hoplimit = dst_metric(dst, RTAX_HOPLIMIT); + if (hoplimit < 0) { + struct net_device *dev = dst->dev; + struct inet6_dev *idev = in6_dev_get(dev); + if (idev) { + hoplimit = idev->cnf.hop_limit; + in6_dev_put(idev); + } else + hoplimit = ipv6_devconf.hop_limit; } return hoplimit; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 5f5d1218c34e..593d3efadaf9 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -792,9 +792,7 @@ do_udp_sendmsg: else hlimit = np->hop_limit; if (hlimit < 0) - hlimit = dst_metric(dst, RTAX_HOPLIMIT); - if (hlimit < 0) - hlimit = ipv6_get_hoplimit(dst->dev); + hlimit = ip6_dst_hoplimit(dst); } if (tclass < 0) { -- cgit v1.2.3 From 1d5d236d309ab90fa6aedf712f586b3595721373 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 10 Mar 2008 10:56:55 -0400 Subject: [IPV6]: Use bitfields for hop_limit and mcast_hops. Save some bits for future extensions. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/ipv6.h | 28 ++++++++++++++++++++++++---- 1 file changed, 24 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 9b59e37afad9..87ae4e389ce1 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -274,8 +274,29 @@ struct ipv6_pinfo { __be32 flow_label; __u32 frag_size; - __s16 hop_limit; - __s16 mcast_hops; + + /* + * Packed in 16bits. + * Omit one shift by by putting the signed field at MSB. + */ +#if defined(__BIG_ENDIAN_BITFIELD) + __s16 hop_limit:9; + __u16 __unused_1:7; +#else + __u16 __unused_1:7; + __s16 hop_limit:9; +#endif + +#if defined(__BIG_ENDIAN_BITFIELD) + /* Packed in 16bits. */ + __s16 mcast_hops:9; + __u16 __unused_2:6, + mc_loop:1; +#else + __u16 mc_loop:1, + __unused_2:6; + __s16 mcast_hops:9; +#endif int mcast_oif; /* pktoption flags */ @@ -298,8 +319,7 @@ struct ipv6_pinfo { } rxopt; /* sockopt flags */ - __u8 mc_loop:1, - recverr:1, + __u8 recverr:1, sndflow:1, pmtudisc:2, ipv6only:1; -- cgit v1.2.3 From 7cbca67c073263c179f605bdbbdc565ab29d801d Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 25 Mar 2008 09:37:42 +0900 Subject: [IPV6]: Support Source Address Selection API (RFC5014). Signed-off-by: YOSHIFUJI Hideaki --- include/linux/in6.h | 11 +++++++ include/linux/ipv6.h | 6 +++- include/net/addrconf.h | 1 + include/net/ip6_route.h | 9 ++++-- net/ipv6/addrconf.c | 17 +++++++++-- net/ipv6/fib6_rules.c | 12 +++++++- net/ipv6/ip6_output.c | 4 ++- net/ipv6/ipv6_sockglue.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/ndisc.c | 4 ++- net/ipv6/route.c | 11 ++++++- net/ipv6/xfrm6_policy.c | 2 +- net/sctp/ipv6.c | 4 ++- 12 files changed, 146 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/in6.h b/include/linux/in6.h index 2a61c82af115..f674000c6c99 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -249,4 +249,15 @@ struct in6_flowlabel_req * IP6T_SO_GET_REVISION_TARGET 69 */ +/* RFC5014: Source address selection */ +#define IPV6_ADDR_PREFERENCES 72 + +#define IPV6_PREFER_SRC_TMP 0x0001 +#define IPV6_PREFER_SRC_PUBLIC 0x0002 +#define IPV6_PREFER_SRC_PUBTMP_DEFAULT 0x0100 +#define IPV6_PREFER_SRC_COA 0x0004 +#define IPV6_PREFER_SRC_HOME 0x0400 +#define IPV6_PREFER_SRC_CGA 0x0008 +#define IPV6_PREFER_SRC_NONCGA 0x0800 + #endif diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 87ae4e389ce1..c9ba0da16ce9 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -322,7 +322,11 @@ struct ipv6_pinfo { __u8 recverr:1, sndflow:1, pmtudisc:2, - ipv6only:1; + ipv6only:1, + srcprefs:3; /* 001: prefer temporary address + * 010: prefer public address + * 100: prefer care-of address + */ __u8 tclass; __u32 dst_cookie; diff --git a/include/net/addrconf.h b/include/net/addrconf.h index edcb4bbaab7d..c9276c72764d 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -78,6 +78,7 @@ extern struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, extern int ipv6_dev_get_saddr(struct net_device *dev, struct in6_addr *daddr, + unsigned int srcprefs, struct in6_addr *saddr); extern int ipv6_get_lladdr(struct net_device *dev, struct in6_addr *addr, diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 5c3b67c86aef..3ae6799c2b14 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -30,9 +30,12 @@ struct route_info { #include #include -#define RT6_LOOKUP_F_IFACE 0x1 -#define RT6_LOOKUP_F_REACHABLE 0x2 -#define RT6_LOOKUP_F_HAS_SADDR 0x4 +#define RT6_LOOKUP_F_IFACE 0x00000001 +#define RT6_LOOKUP_F_REACHABLE 0x00000002 +#define RT6_LOOKUP_F_HAS_SADDR 0x00000004 +#define RT6_LOOKUP_F_SRCPREF_TMP 0x00000008 +#define RT6_LOOKUP_F_SRCPREF_PUBLIC 0x00000010 +#define RT6_LOOKUP_F_SRCPREF_COA 0x00000020 extern struct rt6_info *ip6_null_entry; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 787e90af166c..89954885dee1 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -909,6 +909,7 @@ struct ipv6_saddr_dst { int ifindex; int scope; int label; + unsigned int prefs; }; static inline int ipv6_saddr_preferred(int type) @@ -984,9 +985,12 @@ static int ipv6_get_saddr_eval(struct ipv6_saddr_score *score, break; #ifdef CONFIG_IPV6_MIP6 case IPV6_SADDR_RULE_HOA: + { /* Rule 4: Prefer home address */ - ret = !!(score->ifa->flags & IFA_F_HOMEADDRESS); + int prefhome = !(dst->prefs & IPV6_PREFER_SRC_COA); + ret = !(score->ifa->flags & IFA_F_HOMEADDRESS) ^ prefhome; break; + } #endif case IPV6_SADDR_RULE_OIF: /* Rule 5: Prefer outgoing interface */ @@ -1000,11 +1004,16 @@ static int ipv6_get_saddr_eval(struct ipv6_saddr_score *score, break; #ifdef CONFIG_IPV6_PRIVACY case IPV6_SADDR_RULE_PRIVACY: + { /* Rule 7: Prefer public address * Note: prefer temprary address if use_tempaddr >= 2 */ - ret = (!(score->ifa->flags & IFA_F_TEMPORARY)) ^ (score->ifa->idev->cnf.use_tempaddr >= 2); + int preftmp = dst->prefs & (IPV6_PREFER_SRC_PUBLIC|IPV6_PREFER_SRC_TMP) ? + !!(dst->prefs & IPV6_PREFER_SRC_TMP) : + score->ifa->idev->cnf.use_tempaddr >= 2; + ret = (!(score->ifa->flags & IFA_F_TEMPORARY)) ^ preftmp; break; + } #endif case IPV6_SADDR_RULE_ORCHID: /* Rule 8-: Prefer ORCHID vs ORCHID or @@ -1030,7 +1039,8 @@ out: } int ipv6_dev_get_saddr(struct net_device *dst_dev, - struct in6_addr *daddr, struct in6_addr *saddr) + struct in6_addr *daddr, unsigned int prefs, + struct in6_addr *saddr) { struct ipv6_saddr_score scores[2], *score = &scores[0], *hiscore = &scores[1]; @@ -1044,6 +1054,7 @@ int ipv6_dev_get_saddr(struct net_device *dst_dev, dst.ifindex = dst_dev ? dst_dev->ifindex : 0; dst.scope = __ipv6_addr_src_scope(dst_type); dst.label = ipv6_addr_label(daddr, dst_type, dst.ifindex); + dst.prefs = prefs; hiscore->rule = -1; hiscore->ifa = NULL; diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index 55137408f054..e7a7fe26cebf 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -84,8 +84,18 @@ static int fib6_rule_action(struct fib_rule *rule, struct flowi *flp, if ((rule->flags & FIB_RULE_FIND_SADDR) && r->src.plen && !(flags & RT6_LOOKUP_F_HAS_SADDR)) { struct in6_addr saddr; + unsigned int srcprefs = 0; + + if (flags & RT6_LOOKUP_F_SRCPREF_TMP) + srcprefs |= IPV6_PREFER_SRC_TMP; + if (flags & RT6_LOOKUP_F_SRCPREF_PUBLIC) + srcprefs |= IPV6_PREFER_SRC_PUBLIC; + if (flags & RT6_LOOKUP_F_SRCPREF_COA) + srcprefs |= IPV6_PREFER_SRC_COA; + if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, - &flp->fl6_dst, &saddr)) + &flp->fl6_dst, srcprefs, + &saddr)) goto again; if (!ipv6_prefix_equal(&saddr, &r->src.addr, r->src.plen)) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 2a4f08c8a02d..d34aa61353bb 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -920,7 +920,9 @@ static int ip6_dst_lookup_tail(struct sock *sk, if (ipv6_addr_any(&fl->fl6_src)) { err = ipv6_dev_get_saddr(ip6_dst_idev(*dst)->dev, - &fl->fl6_dst, &fl->fl6_src); + &fl->fl6_dst, + sk ? inet6_sk(sk)->srcprefs : 0, + &fl->fl6_src); if (err) goto out_err_release; } diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 8e29fb1d1df6..dc6695cc5767 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -617,7 +617,67 @@ done: retv = xfrm_user_policy(sk, optname, optval, optlen); break; + case IPV6_ADDR_PREFERENCES: + { + unsigned int pref = 0; + unsigned int prefmask = ~0; + + retv = -EINVAL; + + /* check PUBLIC/TMP/PUBTMP_DEFAULT conflicts */ + switch (val & (IPV6_PREFER_SRC_PUBLIC| + IPV6_PREFER_SRC_TMP| + IPV6_PREFER_SRC_PUBTMP_DEFAULT)) { + case IPV6_PREFER_SRC_PUBLIC: + pref |= IPV6_PREFER_SRC_PUBLIC; + break; + case IPV6_PREFER_SRC_TMP: + pref |= IPV6_PREFER_SRC_TMP; + break; + case IPV6_PREFER_SRC_PUBTMP_DEFAULT: + break; + case 0: + goto pref_skip_pubtmp; + default: + goto e_inval; + } + + prefmask &= ~(IPV6_PREFER_SRC_PUBLIC| + IPV6_PREFER_SRC_TMP); +pref_skip_pubtmp: + + /* check HOME/COA conflicts */ + switch (val & (IPV6_PREFER_SRC_HOME|IPV6_PREFER_SRC_COA)) { + case IPV6_PREFER_SRC_HOME: + break; + case IPV6_PREFER_SRC_COA: + pref |= IPV6_PREFER_SRC_COA; + case 0: + goto pref_skip_coa; + default: + goto e_inval; + } + + prefmask &= ~IPV6_PREFER_SRC_COA; +pref_skip_coa: + + /* check CGA/NONCGA conflicts */ + switch (val & (IPV6_PREFER_SRC_CGA|IPV6_PREFER_SRC_NONCGA)) { + case IPV6_PREFER_SRC_CGA: + case IPV6_PREFER_SRC_NONCGA: + case 0: + break; + default: + goto e_inval; + } + + np->srcprefs = (np->srcprefs & prefmask) | pref; + retv = 0; + + break; + } } + release_sock(sk); return retv; @@ -932,6 +992,24 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, val = np->sndflow; break; + case IPV6_ADDR_PREFERENCES: + val = 0; + + if (np->srcprefs & IPV6_PREFER_SRC_TMP) + val |= IPV6_PREFER_SRC_TMP; + else if (np->srcprefs & IPV6_PREFER_SRC_PUBLIC) + val |= IPV6_PREFER_SRC_PUBLIC; + else { + /* XXX: should we return system default? */ + val |= IPV6_PREFER_SRC_PUBTMP_DEFAULT; + } + + if (np->srcprefs & IPV6_PREFER_SRC_COA) + val |= IPV6_PREFER_SRC_COA; + else + val |= IPV6_PREFER_SRC_HOME; + break; + default: return -ENOPROTOOPT; } diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index e7d8e74704c1..3f68a6eae7b2 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -546,7 +546,9 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, override = 0; in6_ifa_put(ifp); } else { - if (ipv6_dev_get_saddr(dev, daddr, &tmpaddr)) + if (ipv6_dev_get_saddr(dev, daddr, + inet6_sk(dev->nd_net->ipv6.ndisc_sk)->srcprefs, + &tmpaddr)) return; src_addr = &tmpaddr; } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index aa3f08718e44..06faa46920e1 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -782,6 +782,15 @@ struct dst_entry * ip6_route_output(struct net *net, struct sock *sk, if (!ipv6_addr_any(&fl->fl6_src)) flags |= RT6_LOOKUP_F_HAS_SADDR; + else if (sk) { + unsigned int prefs = inet6_sk(sk)->srcprefs; + if (prefs & IPV6_PREFER_SRC_TMP) + flags |= RT6_LOOKUP_F_SRCPREF_TMP; + if (prefs & IPV6_PREFER_SRC_PUBLIC) + flags |= RT6_LOOKUP_F_SRCPREF_PUBLIC; + if (prefs & IPV6_PREFER_SRC_COA) + flags |= RT6_LOOKUP_F_SRCPREF_COA; + } return fib6_rule_lookup(net, fl, flags, ip6_pol_route_output); } @@ -2162,7 +2171,7 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, else if (dst) { struct in6_addr saddr_buf; if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, - dst, &saddr_buf) == 0) + dst, 0, &saddr_buf) == 0) NLA_PUT(skb, RTA_PREFSRC, 16, &saddr_buf); } diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index e96dafdc7032..d92d1fceb8cf 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -58,7 +58,7 @@ static int xfrm6_get_saddr(xfrm_address_t *saddr, xfrm_address_t *daddr) return -EHOSTUNREACH; ipv6_dev_get_saddr(ip6_dst_idev(dst)->dev, - (struct in6_addr *)&daddr->a6, + (struct in6_addr *)&daddr->a6, 0, (struct in6_addr *)&saddr->a6); dst_release(dst); return 0; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index 46c5b3c5cb99..dc71d0d83753 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -316,7 +316,9 @@ static void sctp_v6_get_saddr(struct sctp_association *asoc, if (!asoc) { ipv6_dev_get_saddr(dst ? ip6_dst_idev(dst)->dev : NULL, - &daddr->v6.sin6_addr, &saddr->v6.sin6_addr); + &daddr->v6.sin6_addr, + inet6_sk(asoc->base.sk)->srcprefs, + &saddr->v6.sin6_addr); SCTP_DEBUG_PRINTK("saddr from ipv6_get_saddr: " NIP6_FMT "\n", NIP6(saddr->v6.sin6_addr)); return; -- cgit v1.2.3 From c346dca10840a874240c78efe3f39acf4312a1f2 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Tue, 25 Mar 2008 21:47:49 +0900 Subject: [NET] NETNS: Omit net_device->nd_net without CONFIG_NET_NS. Introduce per-net_device inlines: dev_net(), dev_net_set(). Without CONFIG_NET_NS, no namespace other than &init_net exists. Let's explicitly define them to help compiler optimizations. Signed-off-by: YOSHIFUJI Hideaki --- arch/ia64/hp/sim/simeth.c | 2 +- drivers/block/aoe/aoenet.c | 2 +- drivers/net/bonding/bond_3ad.c | 2 +- drivers/net/bonding/bond_alb.c | 2 +- drivers/net/bonding/bond_main.c | 6 ++--- drivers/net/hamradio/bpqether.c | 4 ++-- drivers/net/loopback.c | 2 +- drivers/net/macvlan.c | 2 +- drivers/net/pppoe.c | 6 ++--- drivers/net/veth.c | 2 +- drivers/net/via-velocity.c | 2 +- drivers/net/wan/dlci.c | 2 +- drivers/net/wan/hdlc.c | 4 ++-- drivers/net/wan/lapbether.c | 4 ++-- drivers/net/wan/syncppp.c | 2 +- drivers/s390/net/qeth_l3_main.c | 2 +- include/linux/inetdevice.h | 6 ++--- include/linux/netdevice.h | 25 ++++++++++++++++++++- net/8021q/vlan.c | 2 +- net/8021q/vlan_dev.c | 2 +- net/appletalk/aarp.c | 4 ++-- net/appletalk/ddp.c | 6 ++--- net/atm/clip.c | 2 +- net/atm/mpc.c | 2 +- net/ax25/af_ax25.c | 2 +- net/ax25/ax25_in.c | 2 +- net/bridge/br_notify.c | 2 +- net/bridge/br_stp_bpdu.c | 2 +- net/can/af_can.c | 4 ++-- net/can/bcm.c | 2 +- net/can/raw.c | 2 +- net/core/dev.c | 22 +++++++++---------- net/core/dst.c | 2 +- net/core/fib_rules.c | 2 +- net/core/neighbour.c | 12 +++++----- net/core/pktgen.c | 2 +- net/core/rtnetlink.c | 4 ++-- net/decnet/af_decnet.c | 2 +- net/decnet/dn_route.c | 2 +- net/econet/af_econet.c | 4 ++-- net/ipv4/arp.c | 14 ++++++------ net/ipv4/devinet.c | 10 ++++----- net/ipv4/fib_frontend.c | 12 +++++----- net/ipv4/icmp.c | 8 +++---- net/ipv4/igmp.c | 16 +++++++------- net/ipv4/ip_fragment.c | 2 +- net/ipv4/ip_gre.c | 2 +- net/ipv4/ip_input.c | 6 ++--- net/ipv4/ip_options.c | 2 +- net/ipv4/ipconfig.c | 4 ++-- net/ipv4/ipmr.c | 2 +- net/ipv4/netfilter/ip_queue.c | 2 +- net/ipv4/netfilter/ipt_MASQUERADE.c | 2 +- net/ipv4/raw.c | 4 ++-- net/ipv4/route.c | 28 +++++++++++------------ net/ipv4/tcp_ipv4.c | 6 ++--- net/ipv4/udp.c | 4 ++-- net/ipv4/xfrm4_policy.c | 2 +- net/ipv6/addrconf.c | 44 ++++++++++++++++++------------------- net/ipv6/icmp.c | 4 ++-- net/ipv6/ip6_output.c | 2 +- net/ipv6/mcast.c | 6 ++--- net/ipv6/ndisc.c | 24 ++++++++++---------- net/ipv6/netfilter/ip6_queue.c | 2 +- net/ipv6/proc.c | 2 +- net/ipv6/raw.c | 4 ++-- net/ipv6/reassembly.c | 2 +- net/ipv6/route.c | 40 ++++++++++++++++----------------- net/ipv6/tcp_ipv6.c | 10 ++++----- net/ipv6/udp.c | 4 ++-- net/ipv6/xfrm6_policy.c | 2 +- net/ipx/af_ipx.c | 4 ++-- net/irda/irlap_frame.c | 2 +- net/llc/llc_input.c | 2 +- net/netfilter/core.c | 2 +- net/netfilter/nfnetlink_queue.c | 2 +- net/netlabel/netlabel_unlabeled.c | 2 +- net/netrom/af_netrom.c | 2 +- net/packet/af_packet.c | 8 +++---- net/rose/af_rose.c | 2 +- net/sctp/protocol.c | 2 +- net/tipc/eth_media.c | 4 ++-- net/wireless/wext.c | 2 +- net/x25/af_x25.c | 2 +- net/x25/x25_dev.c | 2 +- net/xfrm/xfrm_policy.c | 4 ++-- security/selinux/netif.c | 2 +- 87 files changed, 251 insertions(+), 228 deletions(-) (limited to 'include') diff --git a/arch/ia64/hp/sim/simeth.c b/arch/ia64/hp/sim/simeth.c index 969fe9f443c4..3d47839a0c48 100644 --- a/arch/ia64/hp/sim/simeth.c +++ b/arch/ia64/hp/sim/simeth.c @@ -294,7 +294,7 @@ simeth_device_event(struct notifier_block *this,unsigned long event, void *ptr) return NOTIFY_DONE; } - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if ( event != NETDEV_UP && event != NETDEV_DOWN ) return NOTIFY_DONE; diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index 8460ef736d56..18d243c73eee 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -115,7 +115,7 @@ aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, struct aoe_hdr *h; u32 n; - if (ifp->nd_net != &init_net) + if (dev_net(ifp) != &init_net) goto exit; skb = skb_share_check(skb, GFP_ATOMIC); diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index cb3c6faa7888..457d81f73e39 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -2429,7 +2429,7 @@ int bond_3ad_lacpdu_recv(struct sk_buff *skb, struct net_device *dev, struct pac struct slave *slave = NULL; int ret = NET_RX_DROP; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto out; if (!(dev->flags & IFF_MASTER)) diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c index b57bc9467dbe..b986dacf5d33 100644 --- a/drivers/net/bonding/bond_alb.c +++ b/drivers/net/bonding/bond_alb.c @@ -345,7 +345,7 @@ static int rlb_arp_recv(struct sk_buff *skb, struct net_device *bond_dev, struct struct arp_pkt *arp = (struct arp_pkt *)skb->data; int res = NET_RX_DROP; - if (bond_dev->nd_net != &init_net) + if (dev_net(bond_dev) != &init_net) goto out; if (!(bond_dev->flags & IFF_MASTER)) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 5fc9d8d58ece..ac688fcb27d7 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2629,7 +2629,7 @@ static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct pack unsigned char *arp_ptr; __be32 sip, tip; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto out; if (!(dev->priv_flags & IFF_BONDING) || !(dev->flags & IFF_MASTER)) @@ -3470,7 +3470,7 @@ static int bond_netdev_event(struct notifier_block *this, unsigned long event, v { struct net_device *event_dev = (struct net_device *)ptr; - if (event_dev->nd_net != &init_net) + if (dev_net(event_dev) != &init_net) return NOTIFY_DONE; dprintk("event_dev: %s, event: %lx\n", @@ -3508,7 +3508,7 @@ static int bond_inetaddr_event(struct notifier_block *this, unsigned long event, struct bonding *bond, *bond_next; struct vlan_entry *vlan, *vlan_next; - if (ifa->ifa_dev->dev->nd_net != &init_net) + if (dev_net(ifa->ifa_dev->dev) != &init_net) return NOTIFY_DONE; list_for_each_entry_safe(bond, bond_next, &bond_dev_list, bond_list) { diff --git a/drivers/net/hamradio/bpqether.c b/drivers/net/hamradio/bpqether.c index 5ddf8b0c34f9..5f4b4c6c9f76 100644 --- a/drivers/net/hamradio/bpqether.c +++ b/drivers/net/hamradio/bpqether.c @@ -172,7 +172,7 @@ static int bpq_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty struct ethhdr *eth; struct bpqdev *bpq; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) @@ -553,7 +553,7 @@ static int bpq_device_event(struct notifier_block *this,unsigned long event, voi { struct net_device *dev = (struct net_device *)ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (!dev_is_ethdev(dev)) diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index f2a6e7132241..41b774baac4d 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -258,7 +258,7 @@ static __net_init int loopback_net_init(struct net *net) if (!dev) goto out; - dev->nd_net = net; + dev_net_set(dev, net); err = register_netdev(dev); if (err) goto out_free_netdev; diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index f651a816b280..2056cfc624dc 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -402,7 +402,7 @@ static int macvlan_newlink(struct net_device *dev, if (!tb[IFLA_LINK]) return -EINVAL; - lowerdev = __dev_get_by_index(dev->nd_net, nla_get_u32(tb[IFLA_LINK])); + lowerdev = __dev_get_by_index(dev_net(dev), nla_get_u32(tb[IFLA_LINK])); if (lowerdev == NULL) return -ENODEV; diff --git a/drivers/net/pppoe.c b/drivers/net/pppoe.c index ac0ac98b19cd..4fad4ddb3504 100644 --- a/drivers/net/pppoe.c +++ b/drivers/net/pppoe.c @@ -301,7 +301,7 @@ static int pppoe_device_event(struct notifier_block *this, { struct net_device *dev = (struct net_device *) ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* Only look at sockets that are using this specific device. */ @@ -392,7 +392,7 @@ static int pppoe_rcv(struct sk_buff *skb, if (!(skb = skb_share_check(skb, GFP_ATOMIC))) goto out; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; if (!pskb_may_pull(skb, sizeof(struct pppoe_hdr))) @@ -424,7 +424,7 @@ static int pppoe_disc_rcv(struct sk_buff *skb, struct pppoe_hdr *ph; struct pppox_sock *po; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto abort; if (!pskb_may_pull(skb, sizeof(struct pppoe_hdr))) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index e2ad98bee6e7..31cd817f33f9 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -375,7 +375,7 @@ static int veth_newlink(struct net_device *dev, else snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d"); - peer = rtnl_create_link(dev->nd_net, ifname, &veth_link_ops, tbp); + peer = rtnl_create_link(dev_net(dev), ifname, &veth_link_ops, tbp); if (IS_ERR(peer)) return PTR_ERR(peer); diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c index 1525e8a89844..ed1afaf683a4 100644 --- a/drivers/net/via-velocity.c +++ b/drivers/net/via-velocity.c @@ -3464,7 +3464,7 @@ static int velocity_netdev_event(struct notifier_block *nb, unsigned long notifi struct velocity_info *vptr; unsigned long flags; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; spin_lock_irqsave(&velocity_dev_list_lock, flags); diff --git a/drivers/net/wan/dlci.c b/drivers/net/wan/dlci.c index 96b232446c0b..b14242768fad 100644 --- a/drivers/net/wan/dlci.c +++ b/drivers/net/wan/dlci.c @@ -517,7 +517,7 @@ static int dlci_dev_event(struct notifier_block *unused, { struct net_device *dev = (struct net_device *) ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event == NETDEV_UNREGISTER) { diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c index 39951d0c34d6..9a83c9d5b8cf 100644 --- a/drivers/net/wan/hdlc.c +++ b/drivers/net/wan/hdlc.c @@ -68,7 +68,7 @@ static int hdlc_rcv(struct sk_buff *skb, struct net_device *dev, { struct hdlc_device *hdlc = dev_to_hdlc(dev); - if (dev->nd_net != &init_net) { + if (dev_net(dev) != &init_net) { kfree_skb(skb); return 0; } @@ -105,7 +105,7 @@ static int hdlc_device_event(struct notifier_block *this, unsigned long event, unsigned long flags; int on; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (dev->get_stats != hdlc_get_stats) diff --git a/drivers/net/wan/lapbether.c b/drivers/net/wan/lapbether.c index fb37b8095231..629c909e05f9 100644 --- a/drivers/net/wan/lapbether.c +++ b/drivers/net/wan/lapbether.c @@ -91,7 +91,7 @@ static int lapbeth_rcv(struct sk_buff *skb, struct net_device *dev, struct packe int len, err; struct lapbethdev *lapbeth; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) @@ -393,7 +393,7 @@ static int lapbeth_device_event(struct notifier_block *this, struct lapbethdev *lapbeth; struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (!dev_is_ethdev(dev)) diff --git a/drivers/net/wan/syncppp.c b/drivers/net/wan/syncppp.c index 61e24b7a45a3..29b4b94e4947 100644 --- a/drivers/net/wan/syncppp.c +++ b/drivers/net/wan/syncppp.c @@ -1444,7 +1444,7 @@ static void sppp_print_bytes (u_char *p, u16 len) static int sppp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *p, struct net_device *orig_dev) { - if (dev->nd_net != &init_net) { + if (dev_net(dev) != &init_net) { kfree_skb(skb); return 0; } diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c index a856cb47fc78..21c439046b3c 100644 --- a/drivers/s390/net/qeth_l3_main.c +++ b/drivers/s390/net/qeth_l3_main.c @@ -3250,7 +3250,7 @@ static int qeth_l3_ip_event(struct notifier_block *this, struct qeth_ipaddr *addr; struct qeth_card *card; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; QETH_DBF_TEXT(trace, 3, "ipevent"); diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h index da05ab47ff2f..7009b0cdd06f 100644 --- a/include/linux/inetdevice.h +++ b/include/linux/inetdevice.h @@ -70,13 +70,13 @@ static inline void ipv4_devconf_setall(struct in_device *in_dev) ipv4_devconf_set((in_dev), NET_IPV4_CONF_ ## attr, (val)) #define IN_DEV_ANDCONF(in_dev, attr) \ - (IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr) && \ + (IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr) && \ IN_DEV_CONF_GET((in_dev), attr)) #define IN_DEV_ORCONF(in_dev, attr) \ - (IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr) || \ + (IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr) || \ IN_DEV_CONF_GET((in_dev), attr)) #define IN_DEV_MAXCONF(in_dev, attr) \ - (max(IPV4_DEVCONF_ALL(in_dev->dev->nd_net, attr), \ + (max(IPV4_DEVCONF_ALL(dev_net(in_dev->dev), attr), \ IN_DEV_CONF_GET((in_dev), attr))) #define IN_DEV_FORWARD(in_dev) IN_DEV_CONF_GET((in_dev), FORWARDING) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index ced61f87660e..d146be40f46c 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -708,8 +708,10 @@ struct net_device void (*poll_controller)(struct net_device *dev); #endif +#ifdef CONFIG_NET_NS /* Network namespace this network device is inside */ struct net *nd_net; +#endif /* bridge stuff */ struct net_bridge_port *br_port; @@ -737,6 +739,27 @@ struct net_device #define NETDEV_ALIGN 32 #define NETDEV_ALIGN_CONST (NETDEV_ALIGN - 1) +/* + * Net namespace inlines + */ +static inline +struct net *dev_net(const struct net_device *dev) +{ +#ifdef CONFIG_NET_NS + return dev->nd_net; +#else + return &init_net; +#endif +} + +static inline +void dev_net_set(struct net_device *dev, const struct net *net) +{ +#ifdef CONFIG_NET_NS + dev->nd_dev = net; +#endif +} + /** * netdev_priv - access network device private data * @dev: network device @@ -813,7 +836,7 @@ static inline struct net_device *next_net_device(struct net_device *dev) struct list_head *lh; struct net *net; - net = dev->nd_net; + net = dev_net(dev); lh = dev->dev_list.next; return lh == &net->dev_base_head ? NULL : net_device_entry(lh); } diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index dbc81b965096..c35dc230365c 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -382,7 +382,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, int i, flgs; struct net_device *vlandev; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (!grp) diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 1e5c9904571d..e536162b1ebc 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -153,7 +153,7 @@ int vlan_skb_recv(struct sk_buff *skb, struct net_device *dev, struct net_device_stats *stats; unsigned short vlan_TCI; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto err_free; skb = skb_share_check(skb, GFP_ATOMIC); diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index 61166f66479f..25aa37ce9430 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -333,7 +333,7 @@ static int aarp_device_event(struct notifier_block *this, unsigned long event, struct net_device *dev = ptr; int ct; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event == NETDEV_DOWN) { @@ -716,7 +716,7 @@ static int aarp_rcv(struct sk_buff *skb, struct net_device *dev, struct atalk_addr sa, *ma, da; struct atalk_iface *ifa; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto out0; /* We only do Ethernet SNAP AARP. */ diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 3be55c8ca4ef..44cd42f7786b 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -648,7 +648,7 @@ static int ddp_device_event(struct notifier_block *this, unsigned long event, { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event == NETDEV_DOWN) @@ -1405,7 +1405,7 @@ static int atalk_rcv(struct sk_buff *skb, struct net_device *dev, int origlen; __u16 len_hops; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto freeit; /* Don't mangle buffer if shared */ @@ -1493,7 +1493,7 @@ freeit: static int ltalk_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev) { - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto freeit; /* Expand any short form frames */ diff --git a/net/atm/clip.c b/net/atm/clip.c index e82da6746723..6f8223ebf551 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -612,7 +612,7 @@ static int clip_device_event(struct notifier_block *this, unsigned long event, { struct net_device *dev = arg; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event == NETDEV_UNREGISTER) { diff --git a/net/atm/mpc.c b/net/atm/mpc.c index 9c7f712fc7e9..9db332e7a6c0 100644 --- a/net/atm/mpc.c +++ b/net/atm/mpc.c @@ -964,7 +964,7 @@ static int mpoa_event_listener(struct notifier_block *mpoa_notifier, unsigned lo dev = (struct net_device *)dev_ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (dev->name == NULL || strncmp(dev->name, "lec", 3)) diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index 48bfcc741f25..ee9dd83e7561 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -116,7 +116,7 @@ static int ax25_device_event(struct notifier_block *this, unsigned long event, { struct net_device *dev = (struct net_device *)ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* Reject non AX.25 devices */ diff --git a/net/ax25/ax25_in.c b/net/ax25/ax25_in.c index d1be080dcb25..33790a8efbc8 100644 --- a/net/ax25/ax25_in.c +++ b/net/ax25/ax25_in.c @@ -451,7 +451,7 @@ int ax25_kiss_rcv(struct sk_buff *skb, struct net_device *dev, skb->sk = NULL; /* Initially we don't know who it's for */ skb->destructor = NULL; /* Who initializes this, dammit?! */ - if (dev->nd_net != &init_net) { + if (dev_net(dev) != &init_net) { kfree_skb(skb); return 0; } diff --git a/net/bridge/br_notify.c b/net/bridge/br_notify.c index 07ac3ae68d8f..00644a544e3c 100644 --- a/net/bridge/br_notify.c +++ b/net/bridge/br_notify.c @@ -37,7 +37,7 @@ static int br_device_event(struct notifier_block *unused, unsigned long event, v struct net_bridge_port *p = dev->br_port; struct net_bridge *br; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* not a port of a bridge */ diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c index 0edbd2a1c3f3..8deab645ef75 100644 --- a/net/bridge/br_stp_bpdu.c +++ b/net/bridge/br_stp_bpdu.c @@ -142,7 +142,7 @@ int br_stp_rcv(struct sk_buff *skb, struct net_device *dev, struct net_bridge *br; const unsigned char *buf; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto err; if (!p) diff --git a/net/can/af_can.c b/net/can/af_can.c index 36b9f22ed83a..2759b76f731c 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -599,7 +599,7 @@ static int can_rcv(struct sk_buff *skb, struct net_device *dev, struct dev_rcv_lists *d; int matches; - if (dev->type != ARPHRD_CAN || dev->nd_net != &init_net) { + if (dev->type != ARPHRD_CAN || dev_net(dev) != &init_net) { kfree_skb(skb); return 0; } @@ -710,7 +710,7 @@ static int can_notifier(struct notifier_block *nb, unsigned long msg, struct net_device *dev = (struct net_device *)data; struct dev_rcv_lists *d; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (dev->type != ARPHRD_CAN) diff --git a/net/can/bcm.c b/net/can/bcm.c index bd4282dae754..e9f99b2c6bc9 100644 --- a/net/can/bcm.c +++ b/net/can/bcm.c @@ -1285,7 +1285,7 @@ static int bcm_notifier(struct notifier_block *nb, unsigned long msg, struct bcm_op *op; int notify_enodev = 0; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (dev->type != ARPHRD_CAN) diff --git a/net/can/raw.c b/net/can/raw.c index 94cd7f27c444..ead50c7c0d40 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -210,7 +210,7 @@ static int raw_notifier(struct notifier_block *nb, struct raw_sock *ro = container_of(nb, struct raw_sock, notifier); struct sock *sk = &ro->sk; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (dev->type != ARPHRD_CAN) diff --git a/net/core/dev.c b/net/core/dev.c index aebd08606040..812534828914 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -216,7 +216,7 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex) /* Device list insertion */ static int list_netdevice(struct net_device *dev) { - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); ASSERT_RTNL(); @@ -852,8 +852,8 @@ int dev_alloc_name(struct net_device *dev, const char *name) struct net *net; int ret; - BUG_ON(!dev->nd_net); - net = dev->nd_net; + BUG_ON(!dev_net(dev)); + net = dev_net(dev); ret = __dev_alloc_name(net, name, buf); if (ret >= 0) strlcpy(dev->name, buf, IFNAMSIZ); @@ -877,9 +877,9 @@ int dev_change_name(struct net_device *dev, char *newname) struct net *net; ASSERT_RTNL(); - BUG_ON(!dev->nd_net); + BUG_ON(!dev_net(dev)); - net = dev->nd_net; + net = dev_net(dev); if (dev->flags & IFF_UP) return -EBUSY; @@ -2615,7 +2615,7 @@ static int ptype_seq_show(struct seq_file *seq, void *v) if (v == SEQ_START_TOKEN) seq_puts(seq, "Type Device Function\n"); - else if (pt->dev == NULL || pt->dev->nd_net == seq_file_net(seq)) { + else if (pt->dev == NULL || dev_net(pt->dev) == seq_file_net(seq)) { if (pt->type == htons(ETH_P_ALL)) seq_puts(seq, "ALL "); else @@ -3689,8 +3689,8 @@ int register_netdevice(struct net_device *dev) /* When net_device's are persistent, this will be fatal. */ BUG_ON(dev->reg_state != NETREG_UNINITIALIZED); - BUG_ON(!dev->nd_net); - net = dev->nd_net; + BUG_ON(!dev_net(dev)); + net = dev_net(dev); spin_lock_init(&dev->queue_lock); spin_lock_init(&dev->_xmit_lock); @@ -4011,7 +4011,7 @@ struct net_device *alloc_netdev_mq(int sizeof_priv, const char *name, dev = (struct net_device *) (((long)p + NETDEV_ALIGN_CONST) & ~NETDEV_ALIGN_CONST); dev->padded = (char *)dev - (char *)p; - dev->nd_net = &init_net; + dev_net_set(dev, &init_net); if (sizeof_priv) { dev->priv = ((char *)dev + @@ -4136,7 +4136,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char /* Get out if there is nothing todo */ err = 0; - if (dev->nd_net == net) + if (dev_net(dev) == net) goto out; /* Pick the destination device name, and ensure @@ -4187,7 +4187,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char dev_addr_discard(dev); /* Actually switch the network namespace */ - dev->nd_net = net; + dev_net_set(dev, net); /* Assign the new device name */ if (destname != dev->name) diff --git a/net/core/dst.c b/net/core/dst.c index 3a01a819ba47..694cd2a3f6d2 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -279,7 +279,7 @@ static inline void dst_ifdown(struct dst_entry *dst, struct net_device *dev, if (!unregister) { dst->input = dst->output = dst_discard; } else { - dst->dev = dst->dev->nd_net->loopback_dev; + dst->dev = dev_net(dst->dev)->loopback_dev; dev_hold(dst->dev); dev_put(dev); if (dst->neighbour && dst->neighbour->dev == dev) { diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 42ccaf5b8509..942be93a2eb0 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -618,7 +618,7 @@ static int fib_rules_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct fib_rules_ops *ops; ASSERT_RTNL(); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 23c0a10c0c37..c978bd1cd659 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -388,7 +388,7 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, hash_val = tbl->hash(pkey, NULL); for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) { if (!memcmp(n->primary_key, pkey, key_len) && - (net == n->dev->nd_net)) { + dev_net(n->dev) == net) { neigh_hold(n); NEIGH_CACHE_STAT_INC(tbl, hits); break; @@ -1298,7 +1298,7 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, struct neigh_parms *p, *ref; struct net *net; - net = dev->nd_net; + net = dev_net(dev); ref = lookup_neigh_params(tbl, net, 0); if (!ref) return NULL; @@ -2050,7 +2050,7 @@ static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, s_idx = 0; for (n = tbl->hash_buckets[h], idx = 0; n; n = n->next) { int lidx; - if (n->dev->nd_net != net) + if (dev_net(n->dev) != net) continue; lidx = idx++; if (lidx < s_idx) @@ -2155,7 +2155,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) n = tbl->hash_buckets[bucket]; while (n) { - if (n->dev->nd_net != net) + if (dev_net(n->dev) != net) goto next; if (state->neigh_sub_iter) { loff_t fakep = 0; @@ -2198,7 +2198,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, while (1) { while (n) { - if (n->dev->nd_net != net) + if (dev_net(n->dev) != net) goto next; if (state->neigh_sub_iter) { void *v = state->neigh_sub_iter(state, n, pos); @@ -2482,7 +2482,7 @@ static inline size_t neigh_nlmsg_size(void) static void __neigh_notify(struct neighbour *n, int type, int flags) { - struct net *net = n->dev->nd_net; + struct net *net = dev_net(n->dev); struct sk_buff *skb; int err = -ENOBUFS; diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 20e63b302ba6..a803b442234c 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -1874,7 +1874,7 @@ static int pktgen_device_event(struct notifier_block *unused, { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* It is OK that we do not hold the group lock right now, diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 2bd9c5f7627d..09250a0800f6 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -972,7 +972,7 @@ struct net_device *rtnl_create_link(struct net *net, char *ifname, goto err_free; } - dev->nd_net = net; + dev_net_set(dev, net); dev->rtnl_link_ops = ops; if (tb[IFLA_MTU]) @@ -1198,7 +1198,7 @@ static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb) void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) { - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct sk_buff *skb; int err = -ENOBUFS; diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 23fd95a7ad15..3554fb3d251c 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -2089,7 +2089,7 @@ static int dn_device_event(struct notifier_block *this, unsigned long event, { struct net_device *dev = (struct net_device *)ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; switch(event) { diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 9dc0abb50eaf..0a46b6c10e51 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -580,7 +580,7 @@ int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type struct dn_dev *dn = (struct dn_dev *)dev->dn_ptr; unsigned char padlen = 0; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto dump_it; if (dn == NULL) diff --git a/net/econet/af_econet.c b/net/econet/af_econet.c index bc0f6252613f..68d154480043 100644 --- a/net/econet/af_econet.c +++ b/net/econet/af_econet.c @@ -1064,7 +1064,7 @@ static int econet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet struct sock *sk; struct ec_device *edev = dev->ec_ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; if (skb->pkt_type == PACKET_OTHERHOST) @@ -1121,7 +1121,7 @@ static int econet_notifier(struct notifier_block *this, unsigned long msg, void struct net_device *dev = (struct net_device *)data; struct ec_device *edev; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; switch (msg) { diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 832473e30b36..3ce2e137e7bc 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -242,7 +242,7 @@ static int arp_constructor(struct neighbour *neigh) return -EINVAL; } - neigh->type = inet_addr_type(dev->nd_net, addr); + neigh->type = inet_addr_type(dev_net(dev), addr); parms = in_dev->arp_parms; __neigh_parms_put(neigh->parms); @@ -341,14 +341,14 @@ static void arp_solicit(struct neighbour *neigh, struct sk_buff *skb) switch (IN_DEV_ARP_ANNOUNCE(in_dev)) { default: case 0: /* By default announce any local IP */ - if (skb && inet_addr_type(dev->nd_net, ip_hdr(skb)->saddr) == RTN_LOCAL) + if (skb && inet_addr_type(dev_net(dev), ip_hdr(skb)->saddr) == RTN_LOCAL) saddr = ip_hdr(skb)->saddr; break; case 1: /* Restrict announcements of saddr in same subnet */ if (!skb) break; saddr = ip_hdr(skb)->saddr; - if (inet_addr_type(dev->nd_net, saddr) == RTN_LOCAL) { + if (inet_addr_type(dev_net(dev), saddr) == RTN_LOCAL) { /* saddr should be known to target */ if (inet_addr_onlink(in_dev, target, saddr)) break; @@ -424,7 +424,7 @@ static int arp_filter(__be32 sip, __be32 tip, struct net_device *dev) int flag = 0; /*unsigned long now; */ - if (ip_route_output_key(dev->nd_net, &rt, &fl) < 0) + if (ip_route_output_key(dev_net(dev), &rt, &fl) < 0) return 1; if (rt->u.dst.dev != dev) { NET_INC_STATS_BH(LINUX_MIB_ARPFILTER); @@ -477,7 +477,7 @@ int arp_find(unsigned char *haddr, struct sk_buff *skb) paddr = skb->rtable->rt_gateway; - if (arp_set_predefined(inet_addr_type(dev->nd_net, paddr), haddr, paddr, dev)) + if (arp_set_predefined(inet_addr_type(dev_net(dev), paddr), haddr, paddr, dev)) return 0; n = __neigh_lookup(&arp_tbl, &paddr, dev, 1); @@ -709,7 +709,7 @@ static int arp_process(struct sk_buff *skb) u16 dev_type = dev->type; int addr_type; struct neighbour *n; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); /* arp_rcv below verifies the ARP header and verifies the device * is ARP'able. @@ -858,7 +858,7 @@ static int arp_process(struct sk_buff *skb) n = __neigh_lookup(&arp_tbl, &sip, dev, 0); - if (IPV4_DEVCONF_ALL(dev->nd_net, ARP_ACCEPT)) { + if (IPV4_DEVCONF_ALL(dev_net(dev), ARP_ACCEPT)) { /* Unsolicited ARP is not accepted by default. It is possible, that this option should be enabled for some devices (strip is candidate) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 4a10dbbbe0a1..823c724a8593 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -165,7 +165,7 @@ static struct in_device *inetdev_init(struct net_device *dev) if (!in_dev) goto out; INIT_RCU_HEAD(&in_dev->rcu_head); - memcpy(&in_dev->cnf, dev->nd_net->ipv4.devconf_dflt, + memcpy(&in_dev->cnf, dev_net(dev)->ipv4.devconf_dflt, sizeof(in_dev->cnf)); in_dev->cnf.sysctl = NULL; in_dev->dev = dev; @@ -872,7 +872,7 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope) { __be32 addr = 0; struct in_device *in_dev; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); rcu_read_lock(); in_dev = __in_dev_get_rcu(dev); @@ -974,7 +974,7 @@ __be32 inet_confirm_addr(struct in_device *in_dev, if (scope != RT_SCOPE_LINK) return confirm_addr_indev(in_dev, dst, local, scope); - net = in_dev->dev->nd_net; + net = dev_net(in_dev->dev); read_lock(&dev_base_lock); rcu_read_lock(); for_each_netdev(net, dev) { @@ -1203,7 +1203,7 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa, struct nlmsghdr *nlh, int err = -ENOBUFS; struct net *net; - net = ifa->ifa_dev->dev->nd_net; + net = dev_net(ifa->ifa_dev->dev); skb = nlmsg_new(inet_nlmsg_size(), GFP_KERNEL); if (skb == NULL) goto errout; @@ -1517,7 +1517,7 @@ static void devinet_sysctl_register(struct in_device *idev) { neigh_sysctl_register(idev->dev, idev->arp_parms, NET_IPV4, NET_IPV4_NEIGH, "ipv4", NULL, NULL); - __devinet_sysctl_register(idev->dev->nd_net, idev->dev->name, + __devinet_sysctl_register(dev_net(idev->dev), idev->dev->name, idev->dev->ifindex, &idev->cnf); } diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 86ff2711fc95..0e4b34b07cb5 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -257,7 +257,7 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, if (in_dev == NULL) goto e_inval; - net = dev->nd_net; + net = dev_net(dev); if (fib_lookup(net, &fl, &res)) goto last_resort; if (res.type != RTN_UNICAST) @@ -674,7 +674,7 @@ out: static void fib_magic(int cmd, int type, __be32 dst, int dst_len, struct in_ifaddr *ifa) { - struct net *net = ifa->ifa_dev->dev->nd_net; + struct net *net = dev_net(ifa->ifa_dev->dev); struct fib_table *tb; struct fib_config cfg = { .fc_protocol = RTPROT_KERNEL, @@ -801,15 +801,15 @@ static void fib_del_ifaddr(struct in_ifaddr *ifa) fib_magic(RTM_DELROUTE, RTN_LOCAL, ifa->ifa_local, 32, prim); /* Check, that this local address finally disappeared. */ - if (inet_addr_type(dev->nd_net, ifa->ifa_local) != RTN_LOCAL) { + if (inet_addr_type(dev_net(dev), ifa->ifa_local) != RTN_LOCAL) { /* And the last, but not the least thing. We must flush stray FIB entries. First of all, we scan fib_info list searching for stray nexthop entries, then ignite fib_flush. */ - if (fib_sync_down_addr(dev->nd_net, ifa->ifa_local)) - fib_flush(dev->nd_net); + if (fib_sync_down_addr(dev_net(dev), ifa->ifa_local)) + fib_flush(dev_net(dev)); } } #undef LOCAL_OK @@ -899,7 +899,7 @@ static void nl_fib_lookup_exit(struct net *net) static void fib_disable_ip(struct net_device *dev, int force) { if (fib_sync_down_dev(dev, force)) - fib_flush(dev->nd_net); + fib_flush(dev_net(dev)); rt_cache_flush(0); arp_ifdown(dev); } diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index ff9a8e643fcc..f38f093ef751 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -351,7 +351,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param, struct sock *sk; struct sk_buff *skb; - sk = icmp_sk(rt->u.dst.dev->nd_net); + sk = icmp_sk(dev_net(rt->u.dst.dev)); if (ip_append_data(sk, icmp_glue_bits, icmp_param, icmp_param->data_len+icmp_param->head_len, icmp_param->head_len, @@ -382,7 +382,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) { struct ipcm_cookie ipc; struct rtable *rt = skb->rtable; - struct net *net = rt->u.dst.dev->nd_net; + struct net *net = dev_net(rt->u.dst.dev); struct sock *sk = icmp_sk(net); struct inet_sock *inet = inet_sk(sk); __be32 daddr; @@ -447,7 +447,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) if (!rt) goto out; - net = rt->u.dst.dev->nd_net; + net = dev_net(rt->u.dst.dev); sk = icmp_sk(net); /* @@ -677,7 +677,7 @@ static void icmp_unreach(struct sk_buff *skb) u32 info = 0; struct net *net; - net = skb->dst->dev->nd_net; + net = dev_net(skb->dst->dev); /* * Incomplete header ? diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 6a4ee8da6994..682f632bfb77 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -130,12 +130,12 @@ */ #define IGMP_V1_SEEN(in_dev) \ - (IPV4_DEVCONF_ALL(in_dev->dev->nd_net, FORCE_IGMP_VERSION) == 1 || \ + (IPV4_DEVCONF_ALL(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 1 || \ IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 1 || \ ((in_dev)->mr_v1_seen && \ time_before(jiffies, (in_dev)->mr_v1_seen))) #define IGMP_V2_SEEN(in_dev) \ - (IPV4_DEVCONF_ALL(in_dev->dev->nd_net, FORCE_IGMP_VERSION) == 2 || \ + (IPV4_DEVCONF_ALL(dev_net(in_dev->dev), FORCE_IGMP_VERSION) == 2 || \ IN_DEV_CONF_GET((in_dev), FORCE_IGMP_VERSION) == 2 || \ ((in_dev)->mr_v2_seen && \ time_before(jiffies, (in_dev)->mr_v2_seen))) @@ -1198,7 +1198,7 @@ void ip_mc_inc_group(struct in_device *in_dev, __be32 addr) ASSERT_RTNL(); - if (in_dev->dev->nd_net != &init_net) + if (dev_net(in_dev->dev) != &init_net) return; for (im=in_dev->mc_list; im; im=im->next) { @@ -1280,7 +1280,7 @@ void ip_mc_dec_group(struct in_device *in_dev, __be32 addr) ASSERT_RTNL(); - if (in_dev->dev->nd_net != &init_net) + if (dev_net(in_dev->dev) != &init_net) return; for (ip=&in_dev->mc_list; (i=*ip)!=NULL; ip=&i->next) { @@ -1310,7 +1310,7 @@ void ip_mc_down(struct in_device *in_dev) ASSERT_RTNL(); - if (in_dev->dev->nd_net != &init_net) + if (dev_net(in_dev->dev) != &init_net) return; for (i=in_dev->mc_list; i; i=i->next) @@ -1333,7 +1333,7 @@ void ip_mc_init_dev(struct in_device *in_dev) { ASSERT_RTNL(); - if (in_dev->dev->nd_net != &init_net) + if (dev_net(in_dev->dev) != &init_net) return; in_dev->mc_tomb = NULL; @@ -1359,7 +1359,7 @@ void ip_mc_up(struct in_device *in_dev) ASSERT_RTNL(); - if (in_dev->dev->nd_net != &init_net) + if (dev_net(in_dev->dev) != &init_net) return; ip_mc_inc_group(in_dev, IGMP_ALL_HOSTS); @@ -1378,7 +1378,7 @@ void ip_mc_destroy_dev(struct in_device *in_dev) ASSERT_RTNL(); - if (in_dev->dev->nd_net != &init_net) + if (dev_net(in_dev->dev) != &init_net) return; /* Deactivate timers */ diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 8b448c4b9080..fcb60e76b234 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -571,7 +571,7 @@ int ip_defrag(struct sk_buff *skb, u32 user) IP_INC_STATS_BH(IPSTATS_MIB_REASMREQDS); - net = skb->dev ? skb->dev->nd_net : skb->dst->dev->nd_net; + net = skb->dev ? dev_net(skb->dev) : dev_net(skb->dst->dev); /* Start by cleaning up the memory. */ if (atomic_read(&net->ipv4.frags.mem) > net->ipv4.frags.high_thresh) ip_evictor(net); diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index f9ee84420cb3..50972b397a9a 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -1190,7 +1190,7 @@ static int ipgre_close(struct net_device *dev) struct ip_tunnel *t = netdev_priv(dev); if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) { struct in_device *in_dev; - in_dev = inetdev_by_index(dev->nd_net, t->mlink); + in_dev = inetdev_by_index(dev_net(dev), t->mlink); if (in_dev) { ip_mc_dec_group(in_dev, t->parms.iph.daddr); in_dev_put(in_dev); diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 2aeea5d15425..26685c83a146 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -172,7 +172,7 @@ int ip_call_ra_chain(struct sk_buff *skb) if (sk && inet_sk(sk)->num == protocol && (!sk->sk_bound_dev_if || sk->sk_bound_dev_if == dev->ifindex) && - sk->sk_net == dev->nd_net) { + sk->sk_net == dev_net(dev)) { if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { if (ip_defrag(skb, IP_DEFRAG_CALL_RA_CHAIN)) { read_unlock(&ip_ra_lock); @@ -199,7 +199,7 @@ int ip_call_ra_chain(struct sk_buff *skb) static int ip_local_deliver_finish(struct sk_buff *skb) { - struct net *net = skb->dev->nd_net; + struct net *net = dev_net(skb->dev); __skb_pull(skb, ip_hdrlen(skb)); @@ -291,7 +291,7 @@ static inline int ip_rcv_options(struct sk_buff *skb) opt = &(IPCB(skb)->opt); opt->optlen = iph->ihl*4 - sizeof(struct iphdr); - if (ip_options_compile(dev->nd_net, opt, skb)) { + if (ip_options_compile(dev_net(dev), opt, skb)) { IP_INC_STATS_BH(IPSTATS_MIB_INHDRERRORS); goto drop; } diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index 87cc1222c600..d107543d3f81 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -145,7 +145,7 @@ int ip_options_echo(struct ip_options * dopt, struct sk_buff * skb) __be32 addr; memcpy(&addr, sptr+soffset-1, 4); - if (inet_addr_type(skb->dst->dev->nd_net, addr) != RTN_LOCAL) { + if (inet_addr_type(dev_net(skb->dst->dev), addr) != RTN_LOCAL) { dopt->ts_needtime = 1; soffset += 8; } diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 96138b128de8..08e8fb60d315 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -434,7 +434,7 @@ ic_rarp_recv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt unsigned char *sha, *tha; /* s for "source", t for "target" */ struct ic_device *d; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) @@ -854,7 +854,7 @@ static int __init ic_bootp_recv(struct sk_buff *skb, struct net_device *dev, str struct ic_device *d; int len, ext_len; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; /* Perform verifications before taking the lock. */ diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 7d63d74ef62a..e54bc1364473 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1089,7 +1089,7 @@ static int ipmr_device_event(struct notifier_block *this, unsigned long event, v struct vif_device *v; int ct; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event != NETDEV_UNREGISTER) diff --git a/net/ipv4/netfilter/ip_queue.c b/net/ipv4/netfilter/ip_queue.c index fe05da41d6ba..500998a2dec1 100644 --- a/net/ipv4/netfilter/ip_queue.c +++ b/net/ipv4/netfilter/ip_queue.c @@ -481,7 +481,7 @@ ipq_rcv_dev_event(struct notifier_block *this, { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* Drop any packets associated with the downed device */ diff --git a/net/ipv4/netfilter/ipt_MASQUERADE.c b/net/ipv4/netfilter/ipt_MASQUERADE.c index c6817b18366a..84c26dd27d81 100644 --- a/net/ipv4/netfilter/ipt_MASQUERADE.c +++ b/net/ipv4/netfilter/ipt_MASQUERADE.c @@ -120,7 +120,7 @@ static int masq_device_event(struct notifier_block *this, { const struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event == NETDEV_DOWN) { diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 3f68a937b602..8756d502a47f 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -168,7 +168,7 @@ static int raw_v4_input(struct sk_buff *skb, struct iphdr *iph, int hash) if (hlist_empty(head)) goto out; - net = skb->dev->nd_net; + net = dev_net(skb->dev); sk = __raw_v4_lookup(net, __sk_head(head), iph->protocol, iph->saddr, iph->daddr, skb->dev->ifindex); @@ -276,7 +276,7 @@ void raw_icmp_error(struct sk_buff *skb, int protocol, u32 info) raw_sk = sk_head(&raw_v4_hashinfo.ht[hash]); if (raw_sk != NULL) { iph = (struct iphdr *)skb->data; - net = skb->dev->nd_net; + net = dev_net(skb->dev); while ((raw_sk = __raw_v4_lookup(net, raw_sk, protocol, iph->daddr, iph->saddr, diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 2941ef21f203..7768d718e199 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -284,7 +284,7 @@ static struct rtable *rt_cache_get_first(struct rt_cache_iter_state *st) rcu_read_lock_bh(); r = rcu_dereference(rt_hash_table[st->bucket].chain); while (r) { - if (r->u.dst.dev->nd_net == st->p.net && + if (dev_net(r->u.dst.dev) == st->p.net && r->rt_genid == st->genid) return r; r = rcu_dereference(r->u.dst.rt_next); @@ -312,7 +312,7 @@ static struct rtable *rt_cache_get_next(struct rt_cache_iter_state *st, struct rtable *r) { while ((r = __rt_cache_get_next(st, r)) != NULL) { - if (r->u.dst.dev->nd_net != st->p.net) + if (dev_net(r->u.dst.dev) != st->p.net) continue; if (r->rt_genid == st->genid) break; @@ -680,7 +680,7 @@ static inline int compare_keys(struct flowi *fl1, struct flowi *fl2) static inline int compare_netns(struct rtable *rt1, struct rtable *rt2) { - return rt1->u.dst.dev->nd_net == rt2->u.dst.dev->nd_net; + return dev_net(rt1->u.dst.dev) == dev_net(rt2->u.dst.dev); } /* @@ -1164,7 +1164,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw, if (!in_dev) return; - net = dev->nd_net; + net = dev_net(dev); if (new_gw == old_gw || !IN_DEV_RX_REDIRECTS(in_dev) || ipv4_is_multicast(new_gw) || ipv4_is_lbcast(new_gw) || ipv4_is_zeronet(new_gw)) @@ -1195,7 +1195,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw, rth->fl.oif != ikeys[k] || rth->fl.iif != 0 || rth->rt_genid != atomic_read(&rt_genid) || - rth->u.dst.dev->nd_net != net) { + dev_net(rth->u.dst.dev) != net) { rthp = &rth->u.dst.rt_next; continue; } @@ -1454,7 +1454,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph, rth->rt_src == iph->saddr && rth->fl.iif == 0 && !(dst_metric_locked(&rth->u.dst, RTAX_MTU)) && - rth->u.dst.dev->nd_net == net && + dev_net(rth->u.dst.dev) == net && rth->rt_genid == atomic_read(&rt_genid)) { unsigned short mtu = new_mtu; @@ -1530,9 +1530,9 @@ static void ipv4_dst_ifdown(struct dst_entry *dst, struct net_device *dev, { struct rtable *rt = (struct rtable *) dst; struct in_device *idev = rt->idev; - if (dev != dev->nd_net->loopback_dev && idev && idev->dev == dev) { + if (dev != dev_net(dev)->loopback_dev && idev && idev->dev == dev) { struct in_device *loopback_idev = - in_dev_get(dev->nd_net->loopback_dev); + in_dev_get(dev_net(dev)->loopback_dev); if (loopback_idev) { rt->idev = loopback_idev; in_dev_put(idev); @@ -1576,7 +1576,7 @@ void ip_rt_get_source(u8 *addr, struct rtable *rt) if (rt->fl.iif == 0) src = rt->rt_src; - else if (fib_lookup(rt->u.dst.dev->nd_net, &rt->fl, &res) == 0) { + else if (fib_lookup(dev_net(rt->u.dst.dev), &rt->fl, &res) == 0) { src = FIB_RES_PREFSRC(res); fib_res_put(&res); } else @@ -1900,7 +1900,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, __be32 spec_dst; int err = -EINVAL; int free_res = 0; - struct net * net = dev->nd_net; + struct net * net = dev_net(dev); /* IP on this device is disabled. */ @@ -2071,7 +2071,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, int iif = dev->ifindex; struct net *net; - net = dev->nd_net; + net = dev_net(dev); tos &= IPTOS_RT_MASK; hash = rt_hash(daddr, saddr, iif); @@ -2084,7 +2084,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->fl.oif == 0 && rth->fl.mark == skb->mark && rth->fl.fl4_tos == tos && - rth->u.dst.dev->nd_net == net && + dev_net(rth->u.dst.dev) == net && rth->rt_genid == atomic_read(&rt_genid)) { dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(in_hit); @@ -2486,7 +2486,7 @@ int __ip_route_output_key(struct net *net, struct rtable **rp, rth->fl.mark == flp->mark && !((rth->fl.fl4_tos ^ flp->fl4_tos) & (IPTOS_RT_MASK | RTO_ONLINK)) && - rth->u.dst.dev->nd_net == net && + dev_net(rth->u.dst.dev) == net && rth->rt_genid == atomic_read(&rt_genid)) { dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(out_hit); @@ -2795,7 +2795,7 @@ int ip_rt_dump(struct sk_buff *skb, struct netlink_callback *cb) rcu_read_lock_bh(); for (rt = rcu_dereference(rt_hash_table[h].chain), idx = 0; rt; rt = rcu_dereference(rt->u.dst.rt_next), idx++) { - if (rt->u.dst.dev->nd_net != net || idx < s_idx) + if (dev_net(rt->u.dst.dev) != net || idx < s_idx) continue; if (rt->rt_genid != atomic_read(&rt_genid)) continue; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 649d00a50cb1..28bece6f281b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -353,7 +353,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info) return; } - sk = inet_lookup(skb->dev->nd_net, &tcp_hashinfo, iph->daddr, th->dest, + sk = inet_lookup(dev_net(skb->dev), &tcp_hashinfo, iph->daddr, th->dest, iph->saddr, th->source, inet_iif(skb)); if (!sk) { ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); @@ -1644,7 +1644,7 @@ int tcp_v4_rcv(struct sk_buff *skb) TCP_SKB_CB(skb)->flags = iph->tos; TCP_SKB_CB(skb)->sacked = 0; - sk = __inet_lookup(skb->dev->nd_net, &tcp_hashinfo, iph->saddr, + sk = __inet_lookup(dev_net(skb->dev), &tcp_hashinfo, iph->saddr, th->source, iph->daddr, th->dest, inet_iif(skb)); if (!sk) goto no_tcp_socket; @@ -1718,7 +1718,7 @@ do_time_wait: } switch (tcp_timewait_state_process(inet_twsk(sk), skb, th)) { case TCP_TW_SYN: { - struct sock *sk2 = inet_lookup_listener(skb->dev->nd_net, + struct sock *sk2 = inet_lookup_listener(dev_net(skb->dev), &tcp_hashinfo, iph->daddr, th->dest, inet_iif(skb)); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index b37581dfd029..e2cd93481359 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -357,7 +357,7 @@ void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[]) int harderr; int err; - sk = __udp4_lib_lookup(skb->dev->nd_net, iph->daddr, uh->dest, + sk = __udp4_lib_lookup(dev_net(skb->dev), iph->daddr, uh->dest, iph->saddr, uh->source, skb->dev->ifindex, udptable); if (sk == NULL) { ICMP_INC_STATS_BH(ICMP_MIB_INERRORS); @@ -1181,7 +1181,7 @@ int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST)) return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable); - sk = __udp4_lib_lookup(skb->dev->nd_net, saddr, uh->source, daddr, + sk = __udp4_lib_lookup(dev_net(skb->dev), saddr, uh->source, daddr, uh->dest, inet_iif(skb), udptable); if (sk != NULL) { diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 10ed70491434..c63de0a72aba 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -221,7 +221,7 @@ static void xfrm4_dst_ifdown(struct dst_entry *dst, struct net_device *dev, xdst = (struct xfrm_dst *)dst; if (xdst->u.rt.idev->dev == dev) { struct in_device *loopback_idev = - in_dev_get(dev->nd_net->loopback_dev); + in_dev_get(dev_net(dev)->loopback_dev); BUG_ON(!loopback_idev); do { diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 89954885dee1..d1de9ec74261 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -335,7 +335,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) rwlock_init(&ndev->lock); ndev->dev = dev; - memcpy(&ndev->cnf, dev->nd_net->ipv6.devconf_dflt, sizeof(ndev->cnf)); + memcpy(&ndev->cnf, dev_net(dev)->ipv6.devconf_dflt, sizeof(ndev->cnf)); ndev->cnf.mtu6 = dev->mtu; ndev->cnf.sysctl = NULL; ndev->nd_parms = neigh_parms_alloc(dev, &nd_tbl); @@ -561,7 +561,7 @@ ipv6_add_addr(struct inet6_dev *idev, const struct in6_addr *addr, int pfxlen, write_lock(&addrconf_hash_lock); /* Ignore adding duplicate addresses on an interface */ - if (ipv6_chk_same_addr(idev->dev->nd_net, addr, idev->dev)) { + if (ipv6_chk_same_addr(dev_net(idev->dev), addr, idev->dev)) { ADBG(("ipv6_add_addr: already assigned\n")); err = -EEXIST; goto out; @@ -751,7 +751,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) if ((ifp->flags & IFA_F_PERMANENT) && onlink < 1) { struct in6_addr prefix; struct rt6_info *rt; - struct net *net = ifp->idev->dev->nd_net; + struct net *net = dev_net(ifp->idev->dev); ipv6_addr_prefix(&prefix, &ifp->addr, ifp->prefix_len); rt = rt6_lookup(net, &prefix, NULL, ifp->idev->dev->ifindex, 1); @@ -1044,7 +1044,7 @@ int ipv6_dev_get_saddr(struct net_device *dst_dev, { struct ipv6_saddr_score scores[2], *score = &scores[0], *hiscore = &scores[1]; - struct net *net = dst_dev->nd_net; + struct net *net = dev_net(dst_dev); struct ipv6_saddr_dst dst; struct net_device *dev; int dst_type; @@ -1217,7 +1217,7 @@ int ipv6_chk_addr(struct net *net, struct in6_addr *addr, read_lock_bh(&addrconf_hash_lock); for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) { - if (ifp->idev->dev->nd_net != net) + if (dev_net(ifp->idev->dev) != net) continue; if (ipv6_addr_equal(&ifp->addr, addr) && !(ifp->flags&IFA_F_TENTATIVE)) { @@ -1239,7 +1239,7 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr, u8 hash = ipv6_addr_hash(addr); for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) { - if (ifp->idev->dev->nd_net != net) + if (dev_net(ifp->idev->dev) != net) continue; if (ipv6_addr_equal(&ifp->addr, addr)) { if (dev == NULL || ifp->idev->dev == dev) @@ -1257,7 +1257,7 @@ struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr, read_lock_bh(&addrconf_hash_lock); for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) { - if (ifp->idev->dev->nd_net != net) + if (dev_net(ifp->idev->dev) != net) continue; if (ipv6_addr_equal(&ifp->addr, addr)) { if (dev == NULL || ifp->idev->dev == dev || @@ -1559,7 +1559,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev, .fc_expires = expires, .fc_dst_len = plen, .fc_flags = RTF_UP | flags, - .fc_nlinfo.nl_net = dev->nd_net, + .fc_nlinfo.nl_net = dev_net(dev), }; ipv6_addr_copy(&cfg.fc_dst, pfx); @@ -1586,7 +1586,7 @@ static void addrconf_add_mroute(struct net_device *dev) .fc_ifindex = dev->ifindex, .fc_dst_len = 8, .fc_flags = RTF_UP, - .fc_nlinfo.nl_net = dev->nd_net, + .fc_nlinfo.nl_net = dev_net(dev), }; ipv6_addr_set(&cfg.fc_dst, htonl(0xFF000000), 0, 0, 0); @@ -1603,7 +1603,7 @@ static void sit_route_add(struct net_device *dev) .fc_ifindex = dev->ifindex, .fc_dst_len = 96, .fc_flags = RTF_UP | RTF_NONEXTHOP, - .fc_nlinfo.nl_net = dev->nd_net, + .fc_nlinfo.nl_net = dev_net(dev), }; /* prefix length - 96 bits "::d.d.d.d" */ @@ -1704,7 +1704,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len) if (pinfo->onlink) { struct rt6_info *rt; - rt = rt6_lookup(dev->nd_net, &pinfo->prefix, NULL, + rt = rt6_lookup(dev_net(dev), &pinfo->prefix, NULL, dev->ifindex, 1); if (rt && ((rt->rt6i_flags & (RTF_GATEWAY | RTF_DEFAULT)) == 0)) { @@ -1748,7 +1748,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len) ok: - ifp = ipv6_get_ifaddr(dev->nd_net, &addr, dev, 1); + ifp = ipv6_get_ifaddr(dev_net(dev), &addr, dev, 1); if (ifp == NULL && valid_lft) { int max_addresses = in6_dev->cnf.max_addresses; @@ -2071,7 +2071,7 @@ static void sit_add_v4_addrs(struct inet6_dev *idev) struct inet6_ifaddr * ifp; struct in6_addr addr; struct net_device *dev; - struct net *net = idev->dev->nd_net; + struct net *net = dev_net(idev->dev); int scope; ASSERT_RTNL(); @@ -2261,7 +2261,7 @@ ipv6_inherit_linklocal(struct inet6_dev *idev, struct net_device *link_dev) static void ip6_tnl_add_linklocal(struct inet6_dev *idev) { struct net_device *link_dev; - struct net *net = idev->dev->nd_net; + struct net *net = dev_net(idev->dev); /* first try to inherit the link-local address from the link device */ if (idev->dev->iflink && @@ -2442,7 +2442,7 @@ static int addrconf_ifdown(struct net_device *dev, int how) { struct inet6_dev *idev; struct inet6_ifaddr *ifa, **bifa; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); int i; ASSERT_RTNL(); @@ -2771,7 +2771,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq) for (state->bucket = 0; state->bucket < IN6_ADDR_HSIZE; ++state->bucket) { ifa = inet6_addr_lst[state->bucket]; - while (ifa && ifa->idev->dev->nd_net != net) + while (ifa && dev_net(ifa->idev->dev) != net) ifa = ifa->lst_next; if (ifa) break; @@ -2787,7 +2787,7 @@ static struct inet6_ifaddr *if6_get_next(struct seq_file *seq, struct inet6_ifad ifa = ifa->lst_next; try_again: if (ifa) { - if (ifa->idev->dev->nd_net != net) { + if (dev_net(ifa->idev->dev) != net) { ifa = ifa->lst_next; goto try_again; } @@ -2905,7 +2905,7 @@ int ipv6_chk_home_addr(struct net *net, struct in6_addr *addr) u8 hash = ipv6_addr_hash(addr); read_lock_bh(&addrconf_hash_lock); for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) { - if (ifp->idev->dev->nd_net != net) + if (dev_net(ifp->idev->dev) != net) continue; if (ipv6_addr_cmp(&ifp->addr, addr) == 0 && (ifp->flags & IFA_F_HOMEADDRESS)) { @@ -3469,7 +3469,7 @@ errout: static void inet6_ifa_notify(int event, struct inet6_ifaddr *ifa) { struct sk_buff *skb; - struct net *net = ifa->idev->dev->nd_net; + struct net *net = dev_net(ifa->idev->dev); int err = -ENOBUFS; skb = nlmsg_new(inet6_ifaddr_msgsize(), GFP_ATOMIC); @@ -3675,7 +3675,7 @@ cont: void inet6_ifinfo_notify(int event, struct inet6_dev *idev) { struct sk_buff *skb; - struct net *net = idev->dev->nd_net; + struct net *net = dev_net(idev->dev); int err = -ENOBUFS; skb = nlmsg_new(inet6_if_nlmsg_size(), GFP_ATOMIC); @@ -3745,7 +3745,7 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, struct prefix_info *pinfo) { struct sk_buff *skb; - struct net *net = idev->dev->nd_net; + struct net *net = dev_net(idev->dev); int err = -ENOBUFS; skb = nlmsg_new(inet6_prefix_nlmsg_size(), GFP_ATOMIC); @@ -4157,7 +4157,7 @@ static void addrconf_sysctl_register(struct inet6_dev *idev) NET_IPV6_NEIGH, "ipv6", &ndisc_ifinfo_sysctl_change, NULL); - __addrconf_sysctl_register(idev->dev->nd_net, idev->dev->name, + __addrconf_sysctl_register(dev_net(idev->dev), idev->dev->name, idev->dev->ifindex, idev, &idev->cnf); } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 86332417b402..50857662e6b7 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -306,7 +306,7 @@ static inline void mip6_addr_swap(struct sk_buff *skb) {} void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info, struct net_device *dev) { - struct net *net = skb->dev->nd_net; + struct net *net = dev_net(skb->dev); struct inet6_dev *idev = NULL; struct ipv6hdr *hdr = ipv6_hdr(skb); struct sock *sk; @@ -507,7 +507,7 @@ EXPORT_SYMBOL(icmpv6_send); static void icmpv6_echo_reply(struct sk_buff *skb) { - struct net *net = skb->dev->nd_net; + struct net *net = dev_net(skb->dev); struct sock *sk; struct inet6_dev *idev; struct ipv6_pinfo *np; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index d34aa61353bb..556300f0eba5 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -402,7 +402,7 @@ int ip6_forward(struct sk_buff *skb) struct dst_entry *dst = skb->dst; struct ipv6hdr *hdr = ipv6_hdr(skb); struct inet6_skb_parm *opt = IP6CB(skb); - struct net *net = dst->dev->nd_net; + struct net *net = dev_net(dst->dev); if (ipv6_devconf.forwarding == 0) goto error; diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 957ac7e9e929..0357de8e78c8 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1400,7 +1400,7 @@ mld_scount(struct ifmcaddr6 *pmc, int type, int gdeleted, int sdeleted) static struct sk_buff *mld_newpack(struct net_device *dev, int size) { - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct sock *sk = net->ipv6.igmp_sk; struct sk_buff *skb; struct mld2_report *pmr; @@ -1448,7 +1448,7 @@ static void mld_sendpack(struct sk_buff *skb) (struct mld2_report *)skb_transport_header(skb); int payload_len, mldlen; struct inet6_dev *idev = in6_dev_get(skb->dev); - struct net *net = skb->dev->nd_net; + struct net *net = dev_net(skb->dev); int err; struct flowi fl; @@ -1762,7 +1762,7 @@ static void mld_send_cr(struct inet6_dev *idev) static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) { - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct sock *sk = net->ipv6.igmp_sk; struct inet6_dev *idev; struct sk_buff *skb; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 3f68a6eae7b2..79af57f586e8 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -447,7 +447,7 @@ static void __ndisc_send(struct net_device *dev, { struct flowi fl; struct dst_entry *dst; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct sock *sk = net->ipv6.ndisc_sk; struct sk_buff *skb; struct icmp6hdr *hdr; @@ -539,7 +539,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, }; /* for anycast or proxy, solicited_addr != src_addr */ - ifp = ipv6_get_ifaddr(dev->nd_net, solicited_addr, dev, 1); + ifp = ipv6_get_ifaddr(dev_net(dev), solicited_addr, dev, 1); if (ifp) { src_addr = solicited_addr; if (ifp->flags & IFA_F_OPTIMISTIC) @@ -547,7 +547,7 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, in6_ifa_put(ifp); } else { if (ipv6_dev_get_saddr(dev, daddr, - inet6_sk(dev->nd_net->ipv6.ndisc_sk)->srcprefs, + inet6_sk(dev_net(dev)->ipv6.ndisc_sk)->srcprefs, &tmpaddr)) return; src_addr = &tmpaddr; @@ -601,7 +601,7 @@ void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr, * suppress the inclusion of the sllao. */ if (send_sllao) { - struct inet6_ifaddr *ifp = ipv6_get_ifaddr(dev->nd_net, saddr, + struct inet6_ifaddr *ifp = ipv6_get_ifaddr(dev_net(dev), saddr, dev, 1); if (ifp) { if (ifp->flags & IFA_F_OPTIMISTIC) { @@ -639,7 +639,7 @@ static void ndisc_solicit(struct neighbour *neigh, struct sk_buff *skb) struct in6_addr *target = (struct in6_addr *)&neigh->primary_key; int probes = atomic_read(&neigh->probes); - if (skb && ipv6_chk_addr(dev->nd_net, &ipv6_hdr(skb)->saddr, dev, 1)) + if (skb && ipv6_chk_addr(dev_net(dev), &ipv6_hdr(skb)->saddr, dev, 1)) saddr = &ipv6_hdr(skb)->saddr; if ((probes -= neigh->parms->ucast_probes) < 0) { @@ -727,7 +727,7 @@ static void ndisc_recv_ns(struct sk_buff *skb) inc = ipv6_addr_is_multicast(daddr); - ifp = ipv6_get_ifaddr(dev->nd_net, &msg->target, dev, 1); + ifp = ipv6_get_ifaddr(dev_net(dev), &msg->target, dev, 1); if (ifp) { if (ifp->flags & (IFA_F_TENTATIVE|IFA_F_OPTIMISTIC)) { @@ -776,7 +776,7 @@ static void ndisc_recv_ns(struct sk_buff *skb) if (ipv6_chk_acast_addr(dev, &msg->target) || (idev->cnf.forwarding && (ipv6_devconf.proxy_ndp || idev->cnf.proxy_ndp) && - (pneigh = pneigh_lookup(&nd_tbl, dev->nd_net, + (pneigh = pneigh_lookup(&nd_tbl, dev_net(dev), &msg->target, dev, 0)) != NULL)) { if (!(NEIGH_CB(skb)->flags & LOCALLY_ENQUEUED) && skb->pkt_type != PACKET_HOST && @@ -886,7 +886,7 @@ static void ndisc_recv_na(struct sk_buff *skb) return; } } - ifp = ipv6_get_ifaddr(dev->nd_net, &msg->target, dev, 1); + ifp = ipv6_get_ifaddr(dev_net(dev), &msg->target, dev, 1); if (ifp) { if (ifp->flags & IFA_F_TENTATIVE) { addrconf_dad_failure(ifp); @@ -918,7 +918,7 @@ static void ndisc_recv_na(struct sk_buff *skb) */ if (lladdr && !memcmp(lladdr, dev->dev_addr, dev->addr_len) && ipv6_devconf.forwarding && ipv6_devconf.proxy_ndp && - pneigh_lookup(&nd_tbl, dev->nd_net, &msg->target, dev, 0)) { + pneigh_lookup(&nd_tbl, dev_net(dev), &msg->target, dev, 0)) { /* XXX: idev->cnf.prixy_ndp */ goto out; } @@ -1008,7 +1008,7 @@ static void ndisc_ra_useropt(struct sk_buff *ra, struct nd_opt_hdr *opt) struct sk_buff *skb; struct nlmsghdr *nlh; struct nduseroptmsg *ndmsg; - struct net *net = ra->dev->nd_net; + struct net *net = dev_net(ra->dev); int err; int base_size = NLMSG_ALIGN(sizeof(struct nduseroptmsg) + (opt->nd_opt_len << 3)); @@ -1395,7 +1395,7 @@ void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, struct in6_addr *target) { struct net_device *dev = skb->dev; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct sock *sk = net->ipv6.ndisc_sk; int len = sizeof(struct icmp6hdr) + 2 * sizeof(struct in6_addr); struct sk_buff *buff; @@ -1597,7 +1597,7 @@ int ndisc_rcv(struct sk_buff *skb) static int ndisc_netdev_event(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); switch (event) { case NETDEV_CHANGEADDR: diff --git a/net/ipv6/netfilter/ip6_queue.c b/net/ipv6/netfilter/ip6_queue.c index cc2f9afcf808..a6d30626b47c 100644 --- a/net/ipv6/netfilter/ip6_queue.c +++ b/net/ipv6/netfilter/ip6_queue.c @@ -484,7 +484,7 @@ ipq_rcv_dev_event(struct notifier_block *this, { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* Drop any packets associated with the downed device */ diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 8a5be290c710..364dc332532c 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -214,7 +214,7 @@ int snmp6_register_dev(struct inet6_dev *idev) if (!idev || !idev->dev) return -EINVAL; - if (idev->dev->nd_net != &init_net) + if (dev_net(idev->dev) != &init_net) return 0; if (!proc_net_devsnmp6) diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 548d0763f4d3..efb0047f6880 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -176,7 +176,7 @@ static int ipv6_raw_deliver(struct sk_buff *skb, int nexthdr) if (sk == NULL) goto out; - net = skb->dev->nd_net; + net = dev_net(skb->dev); sk = __raw_v6_lookup(net, sk, nexthdr, daddr, saddr, IP6CB(skb)->iif); while (sk) { @@ -363,7 +363,7 @@ void raw6_icmp_error(struct sk_buff *skb, int nexthdr, if (sk != NULL) { saddr = &ipv6_hdr(skb)->saddr; daddr = &ipv6_hdr(skb)->daddr; - net = skb->dev->nd_net; + net = dev_net(skb->dev); while ((sk = __raw_v6_lookup(net, sk, nexthdr, saddr, daddr, IP6CB(skb)->iif))) { diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index f936d045a39d..4e1447634f36 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -600,7 +600,7 @@ static int ipv6_frag_rcv(struct sk_buff *skb) return 1; } - net = skb->dev->nd_net; + net = dev_net(skb->dev); if (atomic_read(&net->ipv6.frags.mem) > net->ipv6.frags.high_thresh) ip6_evictor(net, ip6_dst_idev(skb->dst)); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 06faa46920e1..65053fba8c1a 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -208,7 +208,7 @@ static void ip6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, struct rt6_info *rt = (struct rt6_info *)dst; struct inet6_dev *idev = rt->rt6i_idev; struct net_device *loopback_dev = - dev->nd_net->loopback_dev; + dev_net(dev)->loopback_dev; if (dev != loopback_dev && idev != NULL && idev->dev == dev) { struct inet6_dev *loopback_idev = @@ -433,7 +433,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) RT6_TRACE("%s() => %p\n", __func__, match); - net = rt0->rt6i_dev->nd_net; + net = dev_net(rt0->rt6i_dev); return (match ? match : net->ipv6.ip6_null_entry); } @@ -441,7 +441,7 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, struct in6_addr *gwaddr) { - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct route_info *rinfo = (struct route_info *) opt; struct in6_addr prefix_buf, *prefix; unsigned int pref; @@ -607,7 +607,7 @@ static int __ip6_ins_rt(struct rt6_info *rt, struct nl_info *info) int ip6_ins_rt(struct rt6_info *rt) { struct nl_info info = { - .nl_net = rt->rt6i_dev->nd_net, + .nl_net = dev_net(rt->rt6i_dev), }; return __ip6_ins_rt(rt, &info); } @@ -745,7 +745,7 @@ static struct rt6_info *ip6_pol_route_input(struct net *net, struct fib6_table * void ip6_route_input(struct sk_buff *skb) { struct ipv6hdr *iph = ipv6_hdr(skb); - struct net *net = skb->dev->nd_net; + struct net *net = dev_net(skb->dev); int flags = RT6_LOOKUP_F_HAS_SADDR; struct flowi fl = { .iif = skb->dev->ifindex, @@ -928,7 +928,7 @@ struct dst_entry *icmp6_dst_alloc(struct net_device *dev, { struct rt6_info *rt; struct inet6_dev *idev = in6_dev_get(dev); - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); if (unlikely(idev == NULL)) return NULL; @@ -1252,7 +1252,7 @@ install_route: rt->rt6i_idev = idev; rt->rt6i_table = table; - cfg->fc_nlinfo.nl_net = dev->nd_net; + cfg->fc_nlinfo.nl_net = dev_net(dev); return __ip6_ins_rt(rt, &cfg->fc_nlinfo); @@ -1270,7 +1270,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) { int err; struct fib6_table *table; - struct net *net = rt->rt6i_dev->nd_net; + struct net *net = dev_net(rt->rt6i_dev); if (rt == net->ipv6.ip6_null_entry) return -ENOENT; @@ -1289,7 +1289,7 @@ static int __ip6_del_rt(struct rt6_info *rt, struct nl_info *info) int ip6_del_rt(struct rt6_info *rt) { struct nl_info info = { - .nl_net = rt->rt6i_dev->nd_net, + .nl_net = dev_net(rt->rt6i_dev), }; return __ip6_del_rt(rt, &info); } @@ -1401,7 +1401,7 @@ static struct rt6_info *ip6_route_redirect(struct in6_addr *dest, struct net_device *dev) { int flags = RT6_LOOKUP_F_HAS_SADDR; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); struct ip6rd_flowi rdfl = { .fl = { .oif = dev->ifindex, @@ -1428,7 +1428,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, { struct rt6_info *rt, *nrt = NULL; struct netevent_redirect netevent; - struct net *net = neigh->dev->nd_net; + struct net *net = dev_net(neigh->dev); rt = ip6_route_redirect(dest, src, saddr, neigh->dev); @@ -1477,7 +1477,7 @@ void rt6_redirect(struct in6_addr *dest, struct in6_addr *src, nrt->rt6i_nexthop = neigh_clone(neigh); /* Reset pmtu, it may be better */ nrt->u.dst.metrics[RTAX_MTU-1] = ipv6_get_mtu(neigh->dev); - nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(neigh->dev->nd_net, + nrt->u.dst.metrics[RTAX_ADVMSS-1] = ipv6_advmss(dev_net(neigh->dev), dst_mtu(&nrt->u.dst)); if (ip6_ins_rt(nrt)) @@ -1506,7 +1506,7 @@ void rt6_pmtu_discovery(struct in6_addr *daddr, struct in6_addr *saddr, struct net_device *dev, u32 pmtu) { struct rt6_info *rt, *nrt; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); int allfrag = 0; rt = rt6_lookup(net, daddr, saddr, dev->ifindex, 0); @@ -1583,7 +1583,7 @@ out: static struct rt6_info * ip6_rt_copy(struct rt6_info *ort) { - struct net *net = ort->rt6i_dev->nd_net; + struct net *net = dev_net(ort->rt6i_dev); struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); if (rt) { @@ -1682,7 +1682,7 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d struct rt6_info *rt; struct fib6_table *table; - table = fib6_get_table(dev->nd_net, RT6_TABLE_DFLT); + table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT); if (table == NULL) return NULL; @@ -1713,7 +1713,7 @@ struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr, RTF_UP | RTF_EXPIRES | RTF_PREF(pref), .fc_nlinfo.pid = 0, .fc_nlinfo.nlh = NULL, - .fc_nlinfo.nl_net = dev->nd_net, + .fc_nlinfo.nl_net = dev_net(dev), }; ipv6_addr_copy(&cfg.fc_gateway, gwaddr); @@ -1862,7 +1862,7 @@ struct rt6_info *addrconf_dst_alloc(struct inet6_dev *idev, const struct in6_addr *addr, int anycast) { - struct net *net = idev->dev->nd_net; + struct net *net = dev_net(idev->dev); struct rt6_info *rt = ip6_dst_alloc(net->ipv6.ip6_dst_ops); if (rt == NULL) @@ -1939,7 +1939,7 @@ static int rt6_mtu_change_route(struct rt6_info *rt, void *p_arg) { struct rt6_mtu_change_arg *arg = (struct rt6_mtu_change_arg *) p_arg; struct inet6_dev *idev; - struct net *net = arg->dev->nd_net; + struct net *net = dev_net(arg->dev); /* In IPv6 pmtu discovery is not optional, so that RTAX_MTU lock cannot disable it. @@ -1983,7 +1983,7 @@ void rt6_mtu_change(struct net_device *dev, unsigned mtu) .mtu = mtu, }; - fib6_clean_all(dev->nd_net, rt6_mtu_change_route, 0, &arg); + fib6_clean_all(dev_net(dev), rt6_mtu_change_route, 0, &arg); } static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = { @@ -2321,7 +2321,7 @@ static int ip6_route_dev_notify(struct notifier_block *this, unsigned long event, void *data) { struct net_device *dev = (struct net_device *)data; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); if (event == NETDEV_REGISTER && (dev->flags & IFF_LOOPBACK)) { net->ipv6.ip6_null_entry->u.dst.dev = dev; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 8dd72966ff78..086deffff9c9 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -321,7 +321,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct tcp_sock *tp; __u32 seq; - sk = inet6_lookup(skb->dev->nd_net, &tcp_hashinfo, &hdr->daddr, + sk = inet6_lookup(dev_net(skb->dev), &tcp_hashinfo, &hdr->daddr, th->dest, &hdr->saddr, th->source, skb->dev->ifindex); if (sk == NULL) { @@ -988,7 +988,7 @@ static void tcp_v6_send_reset(struct sock *sk, struct sk_buff *skb) struct tcphdr *th = tcp_hdr(skb), *t1; struct sk_buff *buff; struct flowi fl; - struct net *net = skb->dst->dev->nd_net; + struct net *net = dev_net(skb->dst->dev); struct sock *ctl_sk = net->ipv6.tcp_sk; unsigned int tot_len = sizeof(*th); #ifdef CONFIG_TCP_MD5SIG @@ -1093,7 +1093,7 @@ static void tcp_v6_send_ack(struct tcp_timewait_sock *tw, struct tcphdr *th = tcp_hdr(skb), *t1; struct sk_buff *buff; struct flowi fl; - struct net *net = skb->dev->nd_net; + struct net *net = dev_net(skb->dev); struct sock *ctl_sk = net->ipv6.tcp_sk; unsigned int tot_len = sizeof(struct tcphdr); __be32 *topt; @@ -1739,7 +1739,7 @@ static int tcp_v6_rcv(struct sk_buff *skb) TCP_SKB_CB(skb)->flags = ipv6_get_dsfield(ipv6_hdr(skb)); TCP_SKB_CB(skb)->sacked = 0; - sk = __inet6_lookup(skb->dev->nd_net, &tcp_hashinfo, + sk = __inet6_lookup(dev_net(skb->dev), &tcp_hashinfo, &ipv6_hdr(skb)->saddr, th->source, &ipv6_hdr(skb)->daddr, ntohs(th->dest), inet6_iif(skb)); @@ -1822,7 +1822,7 @@ do_time_wait: { struct sock *sk2; - sk2 = inet6_lookup_listener(skb->dev->nd_net, &tcp_hashinfo, + sk2 = inet6_lookup_listener(dev_net(skb->dev), &tcp_hashinfo, &ipv6_hdr(skb)->daddr, ntohs(th->dest), inet6_iif(skb)); if (sk2 != NULL) { diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 593d3efadaf9..6683c04b427e 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -235,7 +235,7 @@ void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt, struct sock *sk; int err; - sk = __udp6_lib_lookup(skb->dev->nd_net, daddr, uh->dest, + sk = __udp6_lib_lookup(dev_net(skb->dev), daddr, uh->dest, saddr, uh->source, inet6_iif(skb), udptable); if (sk == NULL) return; @@ -483,7 +483,7 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[], * check socket cache ... must talk to Alan about his plans * for sock caches... i'll skip this for now. */ - sk = __udp6_lib_lookup(skb->dev->nd_net, saddr, uh->source, + sk = __udp6_lib_lookup(dev_net(skb->dev), saddr, uh->source, daddr, uh->dest, inet6_iif(skb), udptable); if (sk == NULL) { diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index d92d1fceb8cf..8f1e0543b3c4 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -247,7 +247,7 @@ static void xfrm6_dst_ifdown(struct dst_entry *dst, struct net_device *dev, xdst = (struct xfrm_dst *)dst; if (xdst->u.rt6.rt6i_idev->dev == dev) { struct inet6_dev *loopback_idev = - in6_dev_get(dev->nd_net->loopback_dev); + in6_dev_get(dev_net(dev)->loopback_dev); BUG_ON(!loopback_idev); do { diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index c76a9523091b..81ae8735f5e3 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -335,7 +335,7 @@ static int ipxitf_device_event(struct notifier_block *notifier, struct net_device *dev = ptr; struct ipx_interface *i, *tmp; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event != NETDEV_DOWN && event != NETDEV_UP) @@ -1636,7 +1636,7 @@ static int ipx_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty u16 ipx_pktsize; int rc = 0; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; /* Not ours */ diff --git a/net/irda/irlap_frame.c b/net/irda/irlap_frame.c index a38b231c8689..90894534f3cc 100644 --- a/net/irda/irlap_frame.c +++ b/net/irda/irlap_frame.c @@ -1326,7 +1326,7 @@ int irlap_driver_rcv(struct sk_buff *skb, struct net_device *dev, int command; __u8 control; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto out; /* FIXME: should we get our own field? */ diff --git a/net/llc/llc_input.c b/net/llc/llc_input.c index b9143d2a04e1..a69c5c427fe3 100644 --- a/net/llc/llc_input.c +++ b/net/llc/llc_input.c @@ -146,7 +146,7 @@ int llc_rcv(struct sk_buff *skb, struct net_device *dev, int (*rcv)(struct sk_buff *, struct net_device *, struct packet_type *, struct net_device *); - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; /* diff --git a/net/netfilter/core.c b/net/netfilter/core.c index ec05684c56d7..292fa28146fb 100644 --- a/net/netfilter/core.c +++ b/net/netfilter/core.c @@ -168,7 +168,7 @@ int nf_hook_slow(int pf, unsigned int hook, struct sk_buff *skb, #ifdef CONFIG_NET_NS struct net *net; - net = indev == NULL ? outdev->nd_net : indev->nd_net; + net = indev == NULL ? dev_net(outdev) : dev_net(indev); if (net != &init_net) return 1; #endif diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 012cb6910820..81fb048add88 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -557,7 +557,7 @@ nfqnl_rcv_dev_event(struct notifier_block *this, { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* Drop any packets associated with the downed device */ diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 4478f2f6079d..a547c6320eb3 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -954,7 +954,7 @@ static int netlbl_unlhsh_netdev_handler(struct notifier_block *this, struct net_device *dev = ptr; struct netlbl_unlhsh_iface *iface = NULL; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; /* XXX - should this be a check for NETDEV_DOWN or _UNREGISTER? */ diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index 972250c974f1..a270ebf9f765 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -106,7 +106,7 @@ static int nr_device_event(struct notifier_block *this, unsigned long event, voi { struct net_device *dev = (struct net_device *)ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event != NETDEV_DOWN) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index a56ed2120e07..baa290d3444a 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -263,7 +263,7 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev, struct if (skb->pkt_type == PACKET_LOOPBACK) goto out; - if (dev->nd_net != sk->sk_net) + if (dev_net(dev) != sk->sk_net) goto out; if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) @@ -451,7 +451,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet sk = pt->af_packet_priv; po = pkt_sk(sk); - if (dev->nd_net != sk->sk_net) + if (dev_net(dev) != sk->sk_net) goto drop; skb->dev = dev; @@ -568,7 +568,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe sk = pt->af_packet_priv; po = pkt_sk(sk); - if (dev->nd_net != sk->sk_net) + if (dev_net(dev) != sk->sk_net) goto drop; if (dev->header_ops) { @@ -1450,7 +1450,7 @@ static int packet_notifier(struct notifier_block *this, unsigned long msg, void struct sock *sk; struct hlist_node *node; struct net_device *dev = data; - struct net *net = dev->nd_net; + struct net *net = dev_net(dev); read_lock(&net->packet.sklist_lock); sk_for_each(sk, node, &net->packet.sklist) { diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 4a31a81059ab..1a7f143cf741 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -197,7 +197,7 @@ static int rose_device_event(struct notifier_block *this, unsigned long event, { struct net_device *dev = (struct net_device *)ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event != NETDEV_DOWN) diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index beea2fb18b15..2faa0d8839eb 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -630,7 +630,7 @@ static int sctp_inetaddr_event(struct notifier_block *this, unsigned long ev, struct sctp_sockaddr_entry *temp; int found = 0; - if (ifa->ifa_dev->dev->nd_net != &init_net) + if (dev_net(ifa->ifa_dev->dev) != &init_net) return NOTIFY_DONE; switch (ev) { diff --git a/net/tipc/eth_media.c b/net/tipc/eth_media.c index 3bbef2ab22ae..9cd35eec3e7f 100644 --- a/net/tipc/eth_media.c +++ b/net/tipc/eth_media.c @@ -101,7 +101,7 @@ static int recv_msg(struct sk_buff *buf, struct net_device *dev, struct eth_bearer *eb_ptr = (struct eth_bearer *)pt->af_packet_priv; u32 size; - if (dev->nd_net != &init_net) { + if (dev_net(dev) != &init_net) { kfree_skb(buf); return 0; } @@ -198,7 +198,7 @@ static int recv_notification(struct notifier_block *nb, unsigned long evt, struct eth_bearer *eb_ptr = ð_bearers[0]; struct eth_bearer *stop = ð_bearers[MAX_ETH_BEARERS]; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; while ((eb_ptr->dev != dev)) { diff --git a/net/wireless/wext.c b/net/wireless/wext.c index 2c569b63e7d8..947188a5b937 100644 --- a/net/wireless/wext.c +++ b/net/wireless/wext.c @@ -1157,7 +1157,7 @@ static void rtmsg_iwinfo(struct net_device *dev, char *event, int event_len) struct sk_buff *skb; int err; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return; skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC); diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 339ca4a8e89e..7a46ea73fe2d 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -191,7 +191,7 @@ static int x25_device_event(struct notifier_block *this, unsigned long event, struct net_device *dev = ptr; struct x25_neigh *nb; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (dev->type == ARPHRD_X25 diff --git a/net/x25/x25_dev.c b/net/x25/x25_dev.c index f0679d283110..3ff206c0ae94 100644 --- a/net/x25/x25_dev.c +++ b/net/x25/x25_dev.c @@ -95,7 +95,7 @@ int x25_lapb_receive_frame(struct sk_buff *skb, struct net_device *dev, struct sk_buff *nskb; struct x25_neigh *nb; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) goto drop; nskb = skb_copy(skb, GFP_ATOMIC); diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 8e588f20c60c..15d73e47cc2c 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -2079,7 +2079,7 @@ static int stale_bundle(struct dst_entry *dst) void xfrm_dst_ifdown(struct dst_entry *dst, struct net_device *dev) { while ((dst = dst->child) && dst->xfrm && dst->dev == dev) { - dst->dev = dev->nd_net->loopback_dev; + dst->dev = dev_net(dev)->loopback_dev; dev_hold(dst->dev); dev_put(dev); } @@ -2350,7 +2350,7 @@ static int xfrm_dev_event(struct notifier_block *this, unsigned long event, void { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; switch (event) { diff --git a/security/selinux/netif.c b/security/selinux/netif.c index 013d3117a86b..9c8a82aa8baf 100644 --- a/security/selinux/netif.c +++ b/security/selinux/netif.c @@ -281,7 +281,7 @@ static int sel_netif_netdev_notifier_handler(struct notifier_block *this, { struct net_device *dev = ptr; - if (dev->nd_net != &init_net) + if (dev_net(dev) != &init_net) return NOTIFY_DONE; if (event == NETDEV_DOWN) -- cgit v1.2.3 From 3b1e0a655f8eba44ab1ee2a1068d169ccfb853b9 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 26 Mar 2008 02:26:21 +0900 Subject: [NET] NETNS: Omit sock->sk_net without CONFIG_NET_NS. Introduce per-sock inlines: sock_net(), sock_net_set() and per-inet_timewait_sock inlines: twsk_net(), twsk_net_set(). Without CONFIG_NET_NS, no namespace other than &init_net exists. Let's explicitly define them to help compiler optimizations. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/ipv6.h | 4 ++-- include/net/inet_hashtables.h | 8 ++++---- include/net/inet_timewait_sock.h | 18 ++++++++++++++++++ include/net/route.h | 4 ++-- include/net/sock.h | 24 ++++++++++++++++++++++-- net/atm/svc.c | 2 +- net/ax25/af_ax25.c | 2 +- net/bluetooth/l2cap.c | 2 +- net/bluetooth/rfcomm/sock.c | 2 +- net/bluetooth/sco.c | 2 +- net/bridge/br_netlink.c | 4 ++-- net/core/fib_rules.c | 6 +++--- net/core/neighbour.c | 10 +++++----- net/core/rtnetlink.c | 12 ++++++------ net/core/sock.c | 10 +++++----- net/decnet/af_decnet.c | 2 +- net/decnet/dn_dev.c | 6 +++--- net/decnet/dn_fib.c | 4 ++-- net/decnet/dn_route.c | 4 ++-- net/decnet/dn_table.c | 2 +- net/ipv4/af_inet.c | 6 +++--- net/ipv4/devinet.c | 6 +++--- net/ipv4/fib_frontend.c | 8 ++++---- net/ipv4/fib_rules.c | 2 +- net/ipv4/igmp.c | 14 +++++++------- net/ipv4/inet_connection_sock.c | 4 ++-- net/ipv4/inet_hashtables.c | 8 ++++---- net/ipv4/inet_timewait_sock.c | 2 +- net/ipv4/ip_input.c | 2 +- net/ipv4/ip_output.c | 4 ++-- net/ipv4/ip_sockglue.c | 6 +++--- net/ipv4/ipmr.c | 4 ++-- net/ipv4/netfilter/arp_tables.c | 16 ++++++++-------- net/ipv4/netfilter/ip_tables.c | 16 ++++++++-------- net/ipv4/raw.c | 12 ++++++------ net/ipv4/route.c | 4 ++-- net/ipv4/tcp_ipv4.c | 14 +++++++------- net/ipv4/udp.c | 16 ++++++++-------- net/ipv6/addrconf.c | 10 +++++----- net/ipv6/addrlabel.c | 6 +++--- net/ipv6/af_inet6.c | 4 ++-- net/ipv6/fib6_rules.c | 2 +- net/ipv6/icmp.c | 2 +- net/ipv6/inet6_hashtables.c | 4 ++-- net/ipv6/ip6_fib.c | 2 +- net/ipv6/ip6_output.c | 2 +- net/ipv6/ipv6_sockglue.c | 2 +- net/ipv6/mcast.c | 12 ++++++------ net/ipv6/netfilter/ip6_tables.c | 16 ++++++++-------- net/ipv6/raw.c | 6 +++--- net/ipv6/route.c | 4 ++-- net/ipv6/tcp_ipv6.c | 2 +- net/ipv6/udp.c | 4 ++-- net/irda/af_irda.c | 2 +- net/llc/llc_conn.c | 2 +- net/netfilter/nf_sockopt.c | 2 +- net/netlink/af_netlink.c | 30 +++++++++++++++--------------- net/netrom/af_netrom.c | 2 +- net/packet/af_packet.c | 28 ++++++++++++++-------------- net/rose/af_rose.c | 2 +- net/sched/act_api.c | 4 ++-- net/sched/cls_api.c | 4 ++-- net/sched/sch_api.c | 10 +++++----- net/sctp/ipv6.c | 2 +- net/sctp/protocol.c | 2 +- net/socket.c | 4 ++-- net/tipc/socket.c | 2 +- net/unix/af_unix.c | 20 ++++++++++---------- net/x25/af_x25.c | 2 +- 69 files changed, 253 insertions(+), 215 deletions(-) (limited to 'include') diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index c9ba0da16ce9..b90d3d461d4e 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -481,7 +481,7 @@ static inline struct raw6_sock *raw6_sk(const struct sock *sk) #endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ #define INET6_MATCH(__sk, __net, __hash, __saddr, __daddr, __ports, __dif)\ - (((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net)) && \ + (((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net) && \ ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports)) && \ ((__sk)->sk_family == AF_INET6) && \ ipv6_addr_equal(&inet6_sk(__sk)->daddr, (__saddr)) && \ @@ -489,7 +489,7 @@ static inline struct raw6_sock *raw6_sk(const struct sock *sk) (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) #define INET6_TW_MATCH(__sk, __net, __hash, __saddr, __daddr, __ports, __dif) \ - (((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net)) && \ + (((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net) && \ (*((__portpair *)&(inet_twsk(__sk)->tw_dport)) == (__ports)) && \ ((__sk)->sk_family == PF_INET6) && \ (ipv6_addr_equal(&inet6_twsk(__sk)->tw_v6_daddr, (__saddr))) && \ diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index d99c1ba2ece0..5525227c5e92 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -314,25 +314,25 @@ typedef __u64 __bitwise __addrpair; ((__force __u64)(__be32)(__saddr))); #endif /* __BIG_ENDIAN */ #define INET_MATCH(__sk, __net, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ - (((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net)) && \ + (((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net) && \ ((*((__addrpair *)&(inet_sk(__sk)->daddr))) == (__cookie)) && \ ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) #define INET_TW_MATCH(__sk, __net, __hash, __cookie, __saddr, __daddr, __ports, __dif)\ - (((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net)) && \ + (((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net) && \ ((*((__addrpair *)&(inet_twsk(__sk)->tw_daddr))) == (__cookie)) && \ ((*((__portpair *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) #else /* 32-bit arch */ #define INET_ADDR_COOKIE(__name, __saddr, __daddr) #define INET_MATCH(__sk, __net, __hash, __cookie, __saddr, __daddr, __ports, __dif) \ - (((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net)) && \ + (((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net) && \ (inet_sk(__sk)->daddr == (__saddr)) && \ (inet_sk(__sk)->rcv_saddr == (__daddr)) && \ ((*((__portpair *)&(inet_sk(__sk)->dport))) == (__ports)) && \ (!((__sk)->sk_bound_dev_if) || ((__sk)->sk_bound_dev_if == (__dif)))) #define INET_TW_MATCH(__sk, __net, __hash,__cookie, __saddr, __daddr, __ports, __dif) \ - (((__sk)->sk_hash == (__hash)) && ((__sk)->sk_net == (__net)) && \ + (((__sk)->sk_hash == (__hash)) && sock_net((__sk)) == (__net) && \ (inet_twsk(__sk)->tw_daddr == (__saddr)) && \ (inet_twsk(__sk)->tw_rcv_saddr == (__daddr)) && \ ((*((__portpair *)&(inet_twsk(__sk)->tw_dport))) == (__ports)) && \ diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index 296547bfb0b7..07fe0d1a4f03 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -207,4 +207,22 @@ extern void inet_twsk_schedule(struct inet_timewait_sock *tw, const int timeo, const int timewait_len); extern void inet_twsk_deschedule(struct inet_timewait_sock *tw, struct inet_timewait_death_row *twdr); + +static inline +struct net *twsk_net(const struct inet_timewait_sock *twsk) +{ +#ifdef CONFIG_NET_NS + return twsk->tw_net; +#else + return &init_net; +#endif +} + +static inline +void twsk_net_set(struct inet_timewait_sock *twsk, const struct net *net) +{ +#ifdef CONFIG_NET_NS + twsk->tw_net = net; +#endif +} #endif /* _INET_TIMEWAIT_SOCK_ */ diff --git a/include/net/route.h b/include/net/route.h index 28dba925663c..c6338802e8f1 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -160,7 +160,7 @@ static inline int ip_route_connect(struct rtable **rp, __be32 dst, .dport = dport } } }; int err; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); if (!dst || !src) { err = __ip_route_output_key(net, rp, &fl); if (err) @@ -188,7 +188,7 @@ static inline int ip_route_newports(struct rtable **rp, u8 protocol, ip_rt_put(*rp); *rp = NULL; security_sk_classify_flow(sk, &fl); - return ip_route_output_flow(sk->sk_net, rp, &fl, sk, 0); + return ip_route_output_flow(sock_net(sk), rp, &fl, sk, 0); } return 0; } diff --git a/include/net/sock.h b/include/net/sock.h index b433b1ed203d..7e0d4a0c4d12 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -126,7 +126,9 @@ struct sock_common { atomic_t skc_refcnt; unsigned int skc_hash; struct proto *skc_prot; +#ifdef CONFIG_NET_NS struct net *skc_net; +#endif }; /** @@ -1345,6 +1347,24 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_e } #endif +static inline +struct net *sock_net(const struct sock *sk) +{ +#ifdef CONFIG_NET_NS + return sk->sk_net; +#else + return &init_net; +#endif +} + +static inline +void sock_net_set(struct sock *sk, const struct net *net) +{ +#ifdef CONFIG_NET_NS + sk->sk_net = net; +#endif +} + /* * Kernel sockets, f.e. rtnl or icmp_socket, are a part of a namespace. * They should not hold a referrence to a namespace in order to allow @@ -1353,8 +1373,8 @@ static inline void sk_eat_skb(struct sock *sk, struct sk_buff *skb, int copied_e */ static inline void sk_change_net(struct sock *sk, struct net *net) { - put_net(sk->sk_net); - sk->sk_net = net; + put_net(sock_net(sk)); + sock_net_set(sk, net); } extern void sock_enable_timestamp(struct sock *sk); diff --git a/net/atm/svc.c b/net/atm/svc.c index daf9a48a7db0..de1e4f2f3a43 100644 --- a/net/atm/svc.c +++ b/net/atm/svc.c @@ -326,7 +326,7 @@ static int svc_accept(struct socket *sock,struct socket *newsock,int flags) lock_sock(sk); - error = svc_create(sk->sk_net, newsock,0); + error = svc_create(sock_net(sk), newsock,0); if (error) goto out; diff --git a/net/ax25/af_ax25.c b/net/ax25/af_ax25.c index ee9dd83e7561..2712544cf0ca 100644 --- a/net/ax25/af_ax25.c +++ b/net/ax25/af_ax25.c @@ -869,7 +869,7 @@ struct sock *ax25_make_new(struct sock *osk, struct ax25_dev *ax25_dev) struct sock *sk; ax25_cb *ax25, *oax25; - sk = sk_alloc(osk->sk_net, PF_AX25, GFP_ATOMIC, osk->sk_prot); + sk = sk_alloc(sock_net(osk), PF_AX25, GFP_ATOMIC, osk->sk_prot); if (sk == NULL) return NULL; diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c index 34f8bf98bc05..6b995ac832f5 100644 --- a/net/bluetooth/l2cap.c +++ b/net/bluetooth/l2cap.c @@ -1499,7 +1499,7 @@ static inline int l2cap_connect_req(struct l2cap_conn *conn, struct l2cap_cmd_hd goto response; } - sk = l2cap_sock_alloc(parent->sk_net, NULL, BTPROTO_L2CAP, GFP_ATOMIC); + sk = l2cap_sock_alloc(sock_net(parent), NULL, BTPROTO_L2CAP, GFP_ATOMIC); if (!sk) goto response; diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index c46d51035e77..c103fa02893b 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -868,7 +868,7 @@ int rfcomm_connect_ind(struct rfcomm_session *s, u8 channel, struct rfcomm_dlc * goto done; } - sk = rfcomm_sock_alloc(parent->sk_net, NULL, BTPROTO_RFCOMM, GFP_ATOMIC); + sk = rfcomm_sock_alloc(sock_net(parent), NULL, BTPROTO_RFCOMM, GFP_ATOMIC); if (!sk) goto done; diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index b91d3c81a73c..2a5953b4405d 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -803,7 +803,7 @@ static void sco_conn_ready(struct sco_conn *conn) bh_lock_sock(parent); - sk = sco_sock_alloc(parent->sk_net, NULL, BTPROTO_SCO, GFP_ATOMIC); + sk = sco_sock_alloc(sock_net(parent), NULL, BTPROTO_SCO, GFP_ATOMIC); if (!sk) { bh_unlock_sock(parent); goto done; diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index f5d69336d97b..f155e6ce8a21 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -108,7 +108,7 @@ errout: */ static int br_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct net_device *dev; int idx; @@ -140,7 +140,7 @@ skip: */ static int br_rtm_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ifinfomsg *ifm; struct nlattr *protinfo; struct net_device *dev; diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 942be93a2eb0..540c07283e31 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -214,7 +214,7 @@ errout: static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct fib_rule_hdr *frh = nlmsg_data(nlh); struct fib_rules_ops *ops = NULL; struct fib_rule *rule, *r, *last = NULL; @@ -352,7 +352,7 @@ errout: static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct fib_rule_hdr *frh = nlmsg_data(nlh); struct fib_rules_ops *ops = NULL; struct fib_rule *rule, *tmp; @@ -534,7 +534,7 @@ skip: static int fib_nl_dumprule(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct fib_rules_ops *ops; int idx = 0, family; diff --git a/net/core/neighbour.c b/net/core/neighbour.c index c978bd1cd659..065fbac7ecd3 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -1478,7 +1478,7 @@ int neigh_table_clear(struct neigh_table *tbl) static int neigh_delete(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ndmsg *ndm; struct nlattr *dst_attr; struct neigh_table *tbl; @@ -1544,7 +1544,7 @@ out: static int neigh_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ndmsg *ndm; struct nlattr *tb[NDA_MAX+1]; struct neigh_table *tbl; @@ -1812,7 +1812,7 @@ static const struct nla_policy nl_ntbl_parm_policy[NDTPA_MAX+1] = { static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct neigh_table *tbl; struct ndtmsg *ndtmsg; struct nlattr *tb[NDTA_MAX+1]; @@ -1937,7 +1937,7 @@ errout: static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int family, tidx, nidx = 0; int tbl_skip = cb->args[0]; int neigh_skip = cb->args[1]; @@ -2037,7 +2037,7 @@ static void neigh_update_notify(struct neighbour *neigh) static int neigh_dump_table(struct neigh_table *tbl, struct sk_buff *skb, struct netlink_callback *cb) { - struct net * net = skb->sk->sk_net; + struct net * net = sock_net(skb->sk); struct neighbour *n; int rc, h, s_h = cb->args[1]; int idx, s_idx = idx = cb->args[2]; diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 09250a0800f6..da99ac0871bf 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -662,7 +662,7 @@ nla_put_failure: static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int idx; int s_idx = cb->args[0]; struct net_device *dev; @@ -879,7 +879,7 @@ errout: static int rtnl_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ifinfomsg *ifm; struct net_device *dev; int err; @@ -921,7 +921,7 @@ errout: static int rtnl_dellink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); const struct rtnl_link_ops *ops; struct net_device *dev; struct ifinfomsg *ifm; @@ -1000,7 +1000,7 @@ err: static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); const struct rtnl_link_ops *ops; struct net_device *dev; struct ifinfomsg *ifm; @@ -1132,7 +1132,7 @@ replay: static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ifinfomsg *ifm; struct nlattr *tb[IFLA_MAX+1]; struct net_device *dev = NULL; @@ -1227,7 +1227,7 @@ static int rtattr_max; static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); rtnl_doit_func doit; int sz_idx, kind; int min_len; diff --git a/net/core/sock.c b/net/core/sock.c index b1a6ed4d33c1..3ee95060dbd0 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -372,7 +372,7 @@ static int sock_bindtodevice(struct sock *sk, char __user *optval, int optlen) { int ret = -ENOPROTOOPT; #ifdef CONFIG_NETDEVICES - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); char devname[IFNAMSIZ]; int index; @@ -958,7 +958,7 @@ struct sock *sk_alloc(struct net *net, int family, gfp_t priority, */ sk->sk_prot = sk->sk_prot_creator = prot; sock_lock_init(sk); - sk->sk_net = get_net(net); + sock_net_set(sk, get_net(net)); } return sk; @@ -983,7 +983,7 @@ void sk_free(struct sock *sk) printk(KERN_DEBUG "%s: optmem leakage (%d bytes) detected.\n", __func__, atomic_read(&sk->sk_omem_alloc)); - put_net(sk->sk_net); + put_net(sock_net(sk)); sk_prot_free(sk->sk_prot_creator, sk); } @@ -1001,7 +1001,7 @@ void sk_release_kernel(struct sock *sk) sock_hold(sk); sock_release(sk->sk_socket); - sk->sk_net = get_net(&init_net); + sock_net_set(sk, get_net(&init_net)); sock_put(sk); } EXPORT_SYMBOL(sk_release_kernel); @@ -1017,7 +1017,7 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority) sock_copy(newsk, sk); /* SANITY */ - get_net(newsk->sk_net); + get_net(sock_net(newsk)); sk_node_init(&newsk->sk_node); sock_lock_init(newsk); bh_lock_sock(newsk); diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index 3554fb3d251c..fc2efe899e91 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -1094,7 +1094,7 @@ static int dn_accept(struct socket *sock, struct socket *newsock, int flags) cb = DN_SKB_CB(skb); sk->sk_ack_backlog--; - newsk = dn_alloc_sock(sk->sk_net, newsock, sk->sk_allocation); + newsk = dn_alloc_sock(sock_net(sk), newsock, sk->sk_allocation); if (newsk == NULL) { release_sock(sk); kfree_skb(skb); diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 1bbfce5f7a2d..2f0ac3c3eb71 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -625,7 +625,7 @@ static const struct nla_policy dn_ifa_policy[IFA_MAX+1] = { static int dn_nl_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct nlattr *tb[IFA_MAX+1]; struct dn_dev *dn_db; struct ifaddrmsg *ifm; @@ -663,7 +663,7 @@ errout: static int dn_nl_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct nlattr *tb[IFA_MAX+1]; struct net_device *dev; struct dn_dev *dn_db; @@ -779,7 +779,7 @@ errout: static int dn_nl_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int idx, dn_idx = 0, skip_ndevs, skip_naddr; struct net_device *dev; struct dn_dev *dn_db; diff --git a/net/decnet/dn_fib.c b/net/decnet/dn_fib.c index 4aa9a423e606..27ea2e9b080a 100644 --- a/net/decnet/dn_fib.c +++ b/net/decnet/dn_fib.c @@ -504,7 +504,7 @@ static int dn_fib_check_attr(struct rtmsg *r, struct rtattr **rta) static int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct dn_fib_table *tb; struct rtattr **rta = arg; struct rtmsg *r = NLMSG_DATA(nlh); @@ -524,7 +524,7 @@ static int dn_fib_rtm_delroute(struct sk_buff *skb, struct nlmsghdr *nlh, void * static int dn_fib_rtm_newroute(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct dn_fib_table *tb; struct rtattr **rta = arg; struct rtmsg *r = NLMSG_DATA(nlh); diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index 0a46b6c10e51..2f665a516476 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -1512,7 +1512,7 @@ rtattr_failure: */ static int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = in_skb->sk->sk_net; + struct net *net = sock_net(in_skb->sk); struct rtattr **rta = arg; struct rtmsg *rtm = NLMSG_DATA(nlh); struct dn_route *rt = NULL; @@ -1601,7 +1601,7 @@ out_free: */ int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct dn_route *rt; int h, s_h; int idx, s_idx; diff --git a/net/decnet/dn_table.c b/net/decnet/dn_table.c index e09d915dbd77..3a2830ac89c2 100644 --- a/net/decnet/dn_table.c +++ b/net/decnet/dn_table.c @@ -463,7 +463,7 @@ static int dn_fib_table_dump(struct dn_fib_table *tb, struct sk_buff *skb, int dn_fib_dump(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); unsigned int h, s_h; unsigned int e = 0, s_e; struct dn_fib_table *tb; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 06cfb0bed631..5882a1316441 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -464,7 +464,7 @@ int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) if (addr_len < sizeof(struct sockaddr_in)) goto out; - chk_addr_ret = inet_addr_type(sk->sk_net, addr->sin_addr.s_addr); + chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr); /* Not specified by any standard per-se, however it breaks too * many applications when removed. It is unfortunate since @@ -802,7 +802,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { struct sock *sk = sock->sk; int err = 0; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); switch (cmd) { case SIOCGSTAMP: @@ -1132,7 +1132,7 @@ int inet_sk_rebuild_header(struct sock *sk) }; security_sk_classify_flow(sk, &fl); - err = ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 0); + err = ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 0); } if (!err) sk_setup_caps(sk, &rt->u.dst); diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 823c724a8593..6848e4760f34 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -437,7 +437,7 @@ struct in_ifaddr *inet_ifa_byprefix(struct in_device *in_dev, __be32 prefix, static int inet_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct nlattr *tb[IFA_MAX+1]; struct in_device *in_dev; struct ifaddrmsg *ifm; @@ -552,7 +552,7 @@ errout: static int inet_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct in_ifaddr *ifa; ASSERT_RTNL(); @@ -1158,7 +1158,7 @@ nla_put_failure: static int inet_dump_ifaddr(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int idx, ip_idx; struct net_device *dev; struct in_device *in_dev; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 0e4b34b07cb5..0f1557a4ac7a 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -583,7 +583,7 @@ errout: static int inet_rtm_delroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct fib_config cfg; struct fib_table *tb; int err; @@ -605,7 +605,7 @@ errout: static int inet_rtm_newroute(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct fib_config cfg; struct fib_table *tb; int err; @@ -627,7 +627,7 @@ errout: static int inet_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); unsigned int h, s_h; unsigned int e = 0, s_e; struct fib_table *tb; @@ -857,7 +857,7 @@ static void nl_fib_input(struct sk_buff *skb) struct fib_table *tb; u32 pid; - net = skb->sk->sk_net; + net = sock_net(skb->sk); nlh = nlmsg_hdr(skb); if (skb->len < NLMSG_SPACE(0) || skb->len < nlh->nlmsg_len || nlh->nlmsg_len < NLMSG_LENGTH(sizeof(*frn))) diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index 19274d01afa4..1fb56876be54 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -137,7 +137,7 @@ static int fib4_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct nlmsghdr *nlh, struct fib_rule_hdr *frh, struct nlattr **tb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int err = -EINVAL; struct fib4_rule *rule4 = (struct fib4_rule *) rule; diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 682f632bfb77..6250f4239b61 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -1762,7 +1762,7 @@ int ip_mc_join_group(struct sock *sk , struct ip_mreqn *imr) if (!ipv4_is_multicast(addr)) return -EINVAL; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -EPROTONOSUPPORT; rtnl_lock(); @@ -1833,7 +1833,7 @@ int ip_mc_leave_group(struct sock *sk, struct ip_mreqn *imr) u32 ifindex; int ret = -EADDRNOTAVAIL; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -EPROTONOSUPPORT; rtnl_lock(); @@ -1881,7 +1881,7 @@ int ip_mc_source(int add, int omode, struct sock *sk, struct if (!ipv4_is_multicast(addr)) return -EINVAL; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -EPROTONOSUPPORT; rtnl_lock(); @@ -2017,7 +2017,7 @@ int ip_mc_msfilter(struct sock *sk, struct ip_msfilter *msf, int ifindex) msf->imsf_fmode != MCAST_EXCLUDE) return -EINVAL; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -EPROTONOSUPPORT; rtnl_lock(); @@ -2100,7 +2100,7 @@ int ip_mc_msfget(struct sock *sk, struct ip_msfilter *msf, if (!ipv4_is_multicast(addr)) return -EINVAL; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -EPROTONOSUPPORT; rtnl_lock(); @@ -2165,7 +2165,7 @@ int ip_mc_gsfget(struct sock *sk, struct group_filter *gsf, if (!ipv4_is_multicast(addr)) return -EINVAL; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -EPROTONOSUPPORT; rtnl_lock(); @@ -2252,7 +2252,7 @@ void ip_mc_drop_socket(struct sock *sk) if (inet->mc_list == NULL) return; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return; rtnl_lock(); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index d13c5f12bb32..a7fcaf205644 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -85,7 +85,7 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum) struct hlist_node *node; struct inet_bind_bucket *tb; int ret; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); local_bh_disable(); if (!snum) { @@ -333,7 +333,7 @@ struct dst_entry* inet_csk_route_req(struct sock *sk, .dport = ireq->rmt_port } } }; security_req_classify_flow(req, &fl); - if (ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 0)) { + if (ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 0)) { IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); return NULL; } diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 8cd1ad9b9111..1064111e5b96 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -139,7 +139,7 @@ static struct sock *inet_lookup_listener_slow(struct net *net, sk_for_each(sk, node, head) { const struct inet_sock *inet = inet_sk(sk); - if (sk->sk_net == net && inet->num == hnum && + if (sock_net(sk) == net && inet->num == hnum && !ipv6_only_sock(sk)) { const __be32 rcv_saddr = inet->rcv_saddr; int score = sk->sk_family == PF_INET ? 1 : 0; @@ -182,7 +182,7 @@ struct sock *__inet_lookup_listener(struct net *net, if (inet->num == hnum && !sk->sk_node.next && (!inet->rcv_saddr || inet->rcv_saddr == daddr) && (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) && - !sk->sk_bound_dev_if && sk->sk_net == net) + !sk->sk_bound_dev_if && sock_net(sk) == net) goto sherry_cache; sk = inet_lookup_listener_slow(net, head, daddr, hnum, dif); } @@ -254,7 +254,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, struct sock *sk2; const struct hlist_node *node; struct inet_timewait_sock *tw; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); prefetch(head->chain.first); write_lock(lock); @@ -406,7 +406,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, struct inet_bind_hashbucket *head; struct inet_bind_bucket *tb; int ret; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); if (!snum) { int i, remaining, low, high, port; diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index 717c411a5c6b..f12bc24de46f 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -124,7 +124,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk, const int stat tw->tw_hash = sk->sk_hash; tw->tw_ipv6only = 0; tw->tw_prot = sk->sk_prot_creator; - tw->tw_net = sk->sk_net; + twsk_net_set(tw, sock_net(sk)); atomic_set(&tw->tw_refcnt, 1); inet_twsk_dead_node_init(tw); __module_get(tw->tw_prot->owner); diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index 26685c83a146..4be00959b748 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -172,7 +172,7 @@ int ip_call_ra_chain(struct sk_buff *skb) if (sk && inet_sk(sk)->num == protocol && (!sk->sk_bound_dev_if || sk->sk_bound_dev_if == dev->ifindex) && - sk->sk_net == dev_net(dev)) { + sock_net(sk) == dev_net(dev)) { if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { if (ip_defrag(skb, IP_DEFRAG_CALL_RA_CHAIN)) { read_unlock(&ip_ra_lock); diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 913266cd9902..08349267ceb4 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -351,7 +351,7 @@ int ip_queue_xmit(struct sk_buff *skb, int ipfragok) * itself out. */ security_sk_classify_flow(sk, &fl); - if (ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 0)) + if (ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 0)) goto no_route; } sk_setup_caps(sk, &rt->u.dst); @@ -1382,7 +1382,7 @@ void ip_send_reply(struct sock *sk, struct sk_buff *skb, struct ip_reply_arg *ar .dport = tcp_hdr(skb)->source } }, .proto = sk->sk_protocol }; security_skb_classify_flow(skb, &fl); - if (ip_route_output_key(sk->sk_net, &rt, &fl)) + if (ip_route_output_key(sock_net(sk), &rt, &fl)) return; } diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index b854431047a4..d6e76f5229cc 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -449,7 +449,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, struct ip_options * opt = NULL; if (optlen > 40 || optlen < 0) goto e_inval; - err = ip_options_get_from_user(sk->sk_net, &opt, + err = ip_options_get_from_user(sock_net(sk), &opt, optval, optlen); if (err) break; @@ -590,13 +590,13 @@ static int do_ip_setsockopt(struct sock *sk, int level, err = 0; break; } - dev = ip_dev_find(sk->sk_net, mreq.imr_address.s_addr); + dev = ip_dev_find(sock_net(sk), mreq.imr_address.s_addr); if (dev) { mreq.imr_ifindex = dev->ifindex; dev_put(dev); } } else - dev = __dev_get_by_index(sk->sk_net, mreq.imr_ifindex); + dev = __dev_get_by_index(sock_net(sk), mreq.imr_ifindex); err = -EADDRNOTAVAIL; diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index e54bc1364473..11700a4dcd95 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -849,7 +849,7 @@ static void mrtsock_destruct(struct sock *sk) { rtnl_lock(); if (sk == mroute_socket) { - IPV4_DEVCONF_ALL(sk->sk_net, MC_FORWARDING)--; + IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)--; write_lock_bh(&mrt_lock); mroute_socket=NULL; @@ -898,7 +898,7 @@ int ip_mroute_setsockopt(struct sock *sk,int optname,char __user *optval,int opt mroute_socket=sk; write_unlock_bh(&mrt_lock); - IPV4_DEVCONF_ALL(sk->sk_net, MC_FORWARDING)++; + IPV4_DEVCONF_ALL(sock_net(sk), MC_FORWARDING)++; } rtnl_unlock(); return ret; diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 756bc0e1a7c6..1563f29b5117 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -1496,11 +1496,11 @@ static int compat_do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, switch (cmd) { case ARPT_SO_SET_REPLACE: - ret = compat_do_replace(sk->sk_net, user, len); + ret = compat_do_replace(sock_net(sk), user, len); break; case ARPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sk->sk_net, user, len, 1); + ret = do_add_counters(sock_net(sk), user, len, 1); break; default: @@ -1644,10 +1644,10 @@ static int compat_do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, switch (cmd) { case ARPT_SO_GET_INFO: - ret = get_info(sk->sk_net, user, len, 1); + ret = get_info(sock_net(sk), user, len, 1); break; case ARPT_SO_GET_ENTRIES: - ret = compat_get_entries(sk->sk_net, user, len); + ret = compat_get_entries(sock_net(sk), user, len); break; default: ret = do_arpt_get_ctl(sk, cmd, user, len); @@ -1665,11 +1665,11 @@ static int do_arpt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned switch (cmd) { case ARPT_SO_SET_REPLACE: - ret = do_replace(sk->sk_net, user, len); + ret = do_replace(sock_net(sk), user, len); break; case ARPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sk->sk_net, user, len, 0); + ret = do_add_counters(sock_net(sk), user, len, 0); break; default: @@ -1689,11 +1689,11 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len switch (cmd) { case ARPT_SO_GET_INFO: - ret = get_info(sk->sk_net, user, len, 0); + ret = get_info(sock_net(sk), user, len, 0); break; case ARPT_SO_GET_ENTRIES: - ret = get_entries(sk->sk_net, user, len); + ret = get_entries(sock_net(sk), user, len); break; case ARPT_SO_GET_REVISION_TARGET: { diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 85a75e186b4b..a819d191e1aa 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -1852,11 +1852,11 @@ compat_do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, switch (cmd) { case IPT_SO_SET_REPLACE: - ret = compat_do_replace(sk->sk_net, user, len); + ret = compat_do_replace(sock_net(sk), user, len); break; case IPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sk->sk_net, user, len, 1); + ret = do_add_counters(sock_net(sk), user, len, 1); break; default: @@ -1963,10 +1963,10 @@ compat_do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) switch (cmd) { case IPT_SO_GET_INFO: - ret = get_info(sk->sk_net, user, len, 1); + ret = get_info(sock_net(sk), user, len, 1); break; case IPT_SO_GET_ENTRIES: - ret = compat_get_entries(sk->sk_net, user, len); + ret = compat_get_entries(sock_net(sk), user, len); break; default: ret = do_ipt_get_ctl(sk, cmd, user, len); @@ -1985,11 +1985,11 @@ do_ipt_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) switch (cmd) { case IPT_SO_SET_REPLACE: - ret = do_replace(sk->sk_net, user, len); + ret = do_replace(sock_net(sk), user, len); break; case IPT_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sk->sk_net, user, len, 0); + ret = do_add_counters(sock_net(sk), user, len, 0); break; default: @@ -2010,11 +2010,11 @@ do_ipt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) switch (cmd) { case IPT_SO_GET_INFO: - ret = get_info(sk->sk_net, user, len, 0); + ret = get_info(sock_net(sk), user, len, 0); break; case IPT_SO_GET_ENTRIES: - ret = get_entries(sk->sk_net, user, len); + ret = get_entries(sock_net(sk), user, len); break; case IPT_SO_GET_REVISION_MATCH: diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 8756d502a47f..be19a4048d7c 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -117,7 +117,7 @@ static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk, sk_for_each_from(sk, node) { struct inet_sock *inet = inet_sk(sk); - if (sk->sk_net == net && inet->num == num && + if (sock_net(sk) == net && inet->num == num && !(inet->daddr && inet->daddr != raddr) && !(inet->rcv_saddr && inet->rcv_saddr != laddr) && !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)) @@ -499,7 +499,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.oif = sk->sk_bound_dev_if; if (msg->msg_controllen) { - err = ip_cmsg_send(sk->sk_net, msg, &ipc); + err = ip_cmsg_send(sock_net(sk), msg, &ipc); if (err) goto out; if (ipc.opt) @@ -553,7 +553,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, } security_sk_classify_flow(sk, &fl); - err = ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 1); + err = ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 1); } if (err) goto done; @@ -620,7 +620,7 @@ static int raw_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (sk->sk_state != TCP_CLOSE || addr_len < sizeof(struct sockaddr_in)) goto out; - chk_addr_ret = inet_addr_type(sk->sk_net, addr->sin_addr.s_addr); + chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr); ret = -EADDRNOTAVAIL; if (addr->sin_addr.s_addr && chk_addr_ret != RTN_LOCAL && chk_addr_ret != RTN_MULTICAST && chk_addr_ret != RTN_BROADCAST) @@ -856,7 +856,7 @@ static struct sock *raw_get_first(struct seq_file *seq) struct hlist_node *node; sk_for_each(sk, node, &state->h->ht[state->bucket]) - if (sk->sk_net == state->p.net) + if (sock_net(sk) == state->p.net) goto found; } sk = NULL; @@ -872,7 +872,7 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk) sk = sk_next(sk); try_again: ; - } while (sk && sk->sk_net != state->p.net); + } while (sk && sock_net(sk) != state->p.net); if (!sk && ++state->bucket < RAW_HTABLE_SIZE) { sk = sk_head(&state->h->ht[state->bucket]); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 7768d718e199..194f5cca3121 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2689,7 +2689,7 @@ nla_put_failure: static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = in_skb->sk->sk_net; + struct net *net = sock_net(in_skb->sk); struct rtmsg *rtm; struct nlattr *tb[RTA_MAX+1]; struct rtable *rt = NULL; @@ -2785,7 +2785,7 @@ int ip_rt_dump(struct sk_buff *skb, struct netlink_callback *cb) int idx, s_idx; struct net *net; - net = skb->sk->sk_net; + net = sock_net(skb->sk); s_h = cb->args[0]; if (s_h < 0) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 28bece6f281b..46847e600a46 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1486,7 +1486,7 @@ static struct sock *tcp_v4_hnd_req(struct sock *sk, struct sk_buff *skb) if (req) return tcp_check_req(sk, skb, req, prev); - nsk = inet_lookup_established(sk->sk_net, &tcp_hashinfo, iph->saddr, + nsk = inet_lookup_established(sock_net(sk), &tcp_hashinfo, iph->saddr, th->source, iph->daddr, th->dest, inet_iif(skb)); if (nsk) { @@ -1974,7 +1974,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur) while (1) { while (req) { if (req->rsk_ops->family == st->family && - req->sk->sk_net == net) { + sock_net(req->sk) == net) { cur = req; goto out; } @@ -1998,7 +1998,7 @@ get_req: } get_sk: sk_for_each_from(sk, node) { - if (sk->sk_family == st->family && sk->sk_net == net) { + if (sk->sk_family == st->family && sock_net(sk) == net) { cur = sk; goto out; } @@ -2049,7 +2049,7 @@ static void *established_get_first(struct seq_file *seq) read_lock_bh(lock); sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) { if (sk->sk_family != st->family || - sk->sk_net != net) { + sock_net(sk) != net) { continue; } rc = sk; @@ -2059,7 +2059,7 @@ static void *established_get_first(struct seq_file *seq) inet_twsk_for_each(tw, node, &tcp_hashinfo.ehash[st->bucket].twchain) { if (tw->tw_family != st->family || - tw->tw_net != net) { + twsk_net(tw) != net) { continue; } rc = tw; @@ -2086,7 +2086,7 @@ static void *established_get_next(struct seq_file *seq, void *cur) tw = cur; tw = tw_next(tw); get_tw: - while (tw && (tw->tw_family != st->family || tw->tw_net != net)) { + while (tw && (tw->tw_family != st->family || twsk_net(tw) != net)) { tw = tw_next(tw); } if (tw) { @@ -2107,7 +2107,7 @@ get_tw: sk = sk_next(sk); sk_for_each_from(sk, node) { - if (sk->sk_family == st->family && sk->sk_net == net) + if (sk->sk_family == st->family && sock_net(sk) == net) goto found; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index e2cd93481359..76d52d37d6ac 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -137,7 +137,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num, struct hlist_node *node; sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)]) - if (sk->sk_net == net && sk->sk_hash == num) + if (sock_net(sk) == net && sk->sk_hash == num) return 1; return 0; } @@ -158,7 +158,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum, struct hlist_head *head; struct sock *sk2; int error = 1; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); write_lock_bh(&udp_hash_lock); @@ -218,7 +218,7 @@ gotit: sk_for_each(sk2, node, head) if (sk2->sk_hash == snum && sk2 != sk && - sk2->sk_net == net && + sock_net(sk2) == net && (!sk2->sk_reuse || !sk->sk_reuse) && (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && @@ -269,7 +269,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); - if (sk->sk_net == net && sk->sk_hash == hnum && + if (sock_net(sk) == net && sk->sk_hash == hnum && !ipv6_only_sock(sk)) { int score = (sk->sk_family == PF_INET ? 1 : 0); if (inet->rcv_saddr) { @@ -607,7 +607,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.oif = sk->sk_bound_dev_if; if (msg->msg_controllen) { - err = ip_cmsg_send(sk->sk_net, msg, &ipc); + err = ip_cmsg_send(sock_net(sk), msg, &ipc); if (err) return err; if (ipc.opt) @@ -656,7 +656,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, { .sport = inet->sport, .dport = dport } } }; security_sk_classify_flow(sk, &fl); - err = ip_route_output_flow(sk->sk_net, &rt, &fl, sk, 1); + err = ip_route_output_flow(sock_net(sk), &rt, &fl, sk, 1); if (err) { if (err == -ENETUNREACH) IP_INC_STATS_BH(IPSTATS_MIB_OUTNOROUTES); @@ -1511,7 +1511,7 @@ static struct sock *udp_get_first(struct seq_file *seq) for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) { struct hlist_node *node; sk_for_each(sk, node, state->hashtable + state->bucket) { - if (sk->sk_net != net) + if (sock_net(sk) != net) continue; if (sk->sk_family == state->family) goto found; @@ -1531,7 +1531,7 @@ static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk) sk = sk_next(sk); try_again: ; - } while (sk && (sk->sk_net != net || sk->sk_family != state->family)); + } while (sk && (sock_net(sk) != net || sk->sk_family != state->family)); if (!sk && ++state->bucket < UDP_HTABLE_SIZE) { sk = sk_head(state->hashtable + state->bucket); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index d1de9ec74261..f2c90f145cbb 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3054,7 +3054,7 @@ static const struct nla_policy ifa_ipv6_policy[IFA_MAX+1] = { static int inet6_rtm_deladdr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ifaddrmsg *ifm; struct nlattr *tb[IFA_MAX+1]; struct in6_addr *pfx; @@ -3112,7 +3112,7 @@ static int inet6_addr_modify(struct inet6_ifaddr *ifp, u8 ifa_flags, static int inet6_rtm_newaddr(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ifaddrmsg *ifm; struct nlattr *tb[IFA_MAX+1]; struct in6_addr *pfx; @@ -3322,7 +3322,7 @@ static int inet6_dump_addr(struct sk_buff *skb, struct netlink_callback *cb, struct inet6_ifaddr *ifa; struct ifmcaddr6 *ifmca; struct ifacaddr6 *ifaca; - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); s_idx = cb->args[0]; s_ip_idx = ip_idx = cb->args[1]; @@ -3418,7 +3418,7 @@ static int inet6_dump_ifacaddr(struct sk_buff *skb, struct netlink_callback *cb) static int inet6_rtm_getaddr(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = in_skb->sk->sk_net; + struct net *net = sock_net(in_skb->sk); struct ifaddrmsg *ifm; struct nlattr *tb[IFA_MAX+1]; struct in6_addr *addr = NULL; @@ -3645,7 +3645,7 @@ nla_put_failure: static int inet6_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int idx, err; int s_idx = cb->args[0]; struct net_device *dev; diff --git a/net/ipv6/addrlabel.c b/net/ipv6/addrlabel.c index de371b5997fe..9bfa8846f262 100644 --- a/net/ipv6/addrlabel.c +++ b/net/ipv6/addrlabel.c @@ -364,7 +364,7 @@ static const struct nla_policy ifal_policy[IFAL_MAX+1] = { static int ip6addrlbl_newdel(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ifaddrlblmsg *ifal; struct nlattr *tb[IFAL_MAX+1]; struct in6_addr *pfx; @@ -452,7 +452,7 @@ static int ip6addrlbl_fill(struct sk_buff *skb, static int ip6addrlbl_dump(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct ip6addrlbl_entry *p; struct hlist_node *pos; int idx = 0, s_idx = cb->args[0]; @@ -490,7 +490,7 @@ static inline int ip6addrlbl_msgsize(void) static int ip6addrlbl_get(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = in_skb->sk->sk_net; + struct net *net = sock_net(in_skb->sk); struct ifaddrlblmsg *ifal; struct nlattr *tb[IFAL_MAX+1]; struct in6_addr *addr; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index f52bdaed8a1b..12f04e9d3e88 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -245,7 +245,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) struct sock *sk = sock->sk; struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); __be32 v4addr = 0; unsigned short snum; int addr_type = 0; @@ -438,7 +438,7 @@ EXPORT_SYMBOL(inet6_getname); int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); switch(cmd) { diff --git a/net/ipv6/fib6_rules.c b/net/ipv6/fib6_rules.c index e7a7fe26cebf..cac580749ebe 100644 --- a/net/ipv6/fib6_rules.c +++ b/net/ipv6/fib6_rules.c @@ -154,7 +154,7 @@ static int fib6_rule_configure(struct fib_rule *rule, struct sk_buff *skb, struct nlattr **tb) { int err = -EINVAL; - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct fib6_rule *rule6 = (struct fib6_rule *) rule; if (rule->action == FR_ACT_TO_TBL) { diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 50857662e6b7..63309d10df3a 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -163,7 +163,7 @@ static inline int icmpv6_xrlim_allow(struct sock *sk, int type, struct flowi *fl) { struct dst_entry *dst; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); int res = 0; /* Informational messages are not limited. */ diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index c0c8d2d17682..21c467675412 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -105,7 +105,7 @@ struct sock *inet6_lookup_listener(struct net *net, read_lock(&hashinfo->lhash_lock); sk_for_each(sk, node, &hashinfo->listening_hash[inet_lhashfn(hnum)]) { - if (sk->sk_net == net && inet_sk(sk)->num == hnum && + if (sock_net(sk) == net && inet_sk(sk)->num == hnum && sk->sk_family == PF_INET6) { const struct ipv6_pinfo *np = inet6_sk(sk); @@ -172,7 +172,7 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, struct sock *sk2; const struct hlist_node *node; struct inet_timewait_sock *tw; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); prefetch(head->chain.first); write_lock(lock); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index b0814b0082e7..b3f6e03c454c 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -346,7 +346,7 @@ end: static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); unsigned int h, s_h; unsigned int e = 0, s_e; struct rt6_rtnl_dump_arg arg; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 556300f0eba5..a8b4da25b0a7 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -910,7 +910,7 @@ static int ip6_dst_lookup_tail(struct sock *sk, struct dst_entry **dst, struct flowi *fl) { int err; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); if (*dst == NULL) *dst = ip6_route_output(net, sk, fl); diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index dc6695cc5767..d3d93d752e10 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -107,7 +107,7 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, char __user *optval, int optlen) { struct ipv6_pinfo *np = inet6_sk(sk); - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); int val, valbool; int retv = -ENOPROTOOPT; diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 0357de8e78c8..20a3d8e2f6c6 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -181,7 +181,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr) struct net_device *dev = NULL; struct ipv6_mc_socklist *mc_lst; struct ipv6_pinfo *np = inet6_sk(sk); - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); int err; if (!ipv6_addr_is_multicast(addr)) @@ -255,7 +255,7 @@ int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr) { struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6_mc_socklist *mc_lst, **lnk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); write_lock_bh(&ipv6_sk_mc_lock); for (lnk = &np->ipv6_mc_list; (mc_lst = *lnk) !=NULL ; lnk = &mc_lst->next) { @@ -327,7 +327,7 @@ void ipv6_sock_mc_close(struct sock *sk) { struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6_mc_socklist *mc_lst; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); write_lock_bh(&ipv6_sk_mc_lock); while ((mc_lst = np->ipv6_mc_list) != NULL) { @@ -365,7 +365,7 @@ int ip6_mc_source(int add, int omode, struct sock *sk, struct inet6_dev *idev; struct ipv6_pinfo *inet6 = inet6_sk(sk); struct ip6_sf_socklist *psl; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); int i, j, rv; int leavegroup = 0; int pmclocked = 0; @@ -505,7 +505,7 @@ int ip6_mc_msfilter(struct sock *sk, struct group_filter *gsf) struct inet6_dev *idev; struct ipv6_pinfo *inet6 = inet6_sk(sk); struct ip6_sf_socklist *newpsl, *psl; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); int leavegroup = 0; int i, err; @@ -598,7 +598,7 @@ int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf, struct net_device *dev; struct ipv6_pinfo *inet6 = inet6_sk(sk); struct ip6_sf_socklist *psl; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); group = &((struct sockaddr_in6 *)&gsf->gf_group)->sin6_addr; diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index af1ec7ba757c..70ef0d276cc0 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -1879,11 +1879,11 @@ compat_do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, switch (cmd) { case IP6T_SO_SET_REPLACE: - ret = compat_do_replace(sk->sk_net, user, len); + ret = compat_do_replace(sock_net(sk), user, len); break; case IP6T_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sk->sk_net, user, len, 1); + ret = do_add_counters(sock_net(sk), user, len, 1); break; default: @@ -1990,10 +1990,10 @@ compat_do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) switch (cmd) { case IP6T_SO_GET_INFO: - ret = get_info(sk->sk_net, user, len, 1); + ret = get_info(sock_net(sk), user, len, 1); break; case IP6T_SO_GET_ENTRIES: - ret = compat_get_entries(sk->sk_net, user, len); + ret = compat_get_entries(sock_net(sk), user, len); break; default: ret = do_ip6t_get_ctl(sk, cmd, user, len); @@ -2012,11 +2012,11 @@ do_ip6t_set_ctl(struct sock *sk, int cmd, void __user *user, unsigned int len) switch (cmd) { case IP6T_SO_SET_REPLACE: - ret = do_replace(sk->sk_net, user, len); + ret = do_replace(sock_net(sk), user, len); break; case IP6T_SO_SET_ADD_COUNTERS: - ret = do_add_counters(sk->sk_net, user, len, 0); + ret = do_add_counters(sock_net(sk), user, len, 0); break; default: @@ -2037,11 +2037,11 @@ do_ip6t_get_ctl(struct sock *sk, int cmd, void __user *user, int *len) switch (cmd) { case IP6T_SO_GET_INFO: - ret = get_info(sk->sk_net, user, len, 0); + ret = get_info(sock_net(sk), user, len, 0); break; case IP6T_SO_GET_ENTRIES: - ret = get_entries(sk->sk_net, user, len); + ret = get_entries(sock_net(sk), user, len); break; case IP6T_SO_GET_REVISION_MATCH: diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index efb0047f6880..12c7a1560977 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -76,7 +76,7 @@ static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, if (inet_sk(sk)->num == num) { struct ipv6_pinfo *np = inet6_sk(sk); - if (sk->sk_net != net) + if (sock_net(sk) != net) continue; if (!ipv6_addr_any(&np->daddr) && @@ -280,7 +280,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) if (!sk->sk_bound_dev_if) goto out; - dev = dev_get_by_index(sk->sk_net, sk->sk_bound_dev_if); + dev = dev_get_by_index(sock_net(sk), sk->sk_bound_dev_if); if (!dev) { err = -ENODEV; goto out; @@ -293,7 +293,7 @@ static int rawv6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) v4addr = LOOPBACK4_IPV6; if (!(addr_type & IPV6_ADDR_MULTICAST)) { err = -EADDRNOTAVAIL; - if (!ipv6_chk_addr(sk->sk_net, &addr->sin6_addr, + if (!ipv6_chk_addr(sock_net(sk), &addr->sin6_addr, dev, 0)) { if (dev) dev_put(dev); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 65053fba8c1a..ac4428371432 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2020,7 +2020,7 @@ static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, cfg->fc_nlinfo.pid = NETLINK_CB(skb).pid; cfg->fc_nlinfo.nlh = nlh; - cfg->fc_nlinfo.nl_net = skb->sk->sk_net; + cfg->fc_nlinfo.nl_net = sock_net(skb->sk); if (tb[RTA_GATEWAY]) { nla_memcpy(&cfg->fc_gateway, tb[RTA_GATEWAY], 16); @@ -2216,7 +2216,7 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg) static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) { - struct net *net = in_skb->sk->sk_net; + struct net *net = sock_net(in_skb->sk); struct nlattr *tb[RTA_MAX+1]; struct rt6_info *rt; struct sk_buff *skb; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 086deffff9c9..323c7e06ef43 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1218,7 +1218,7 @@ static struct sock *tcp_v6_hnd_req(struct sock *sk,struct sk_buff *skb) if (req) return tcp_check_req(sk, skb, req, prev); - nsk = __inet6_lookup_established(sk->sk_net, &tcp_hashinfo, + nsk = __inet6_lookup_established(sock_net(sk), &tcp_hashinfo, &ipv6_hdr(skb)->saddr, th->source, &ipv6_hdr(skb)->daddr, ntohs(th->dest), inet6_iif(skb)); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 6683c04b427e..db266ff297e5 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -70,7 +70,7 @@ static struct sock *__udp6_lib_lookup(struct net *net, sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); - if (sk->sk_net == net && sk->sk_hash == hnum && + if (sock_net(sk) == net && sk->sk_hash == hnum && sk->sk_family == PF_INET6) { struct ipv6_pinfo *np = inet6_sk(sk); int score = 0; @@ -323,7 +323,7 @@ static struct sock *udp_v6_mcast_next(struct sock *sk, sk_for_each_from(s, node) { struct inet_sock *inet = inet_sk(s); - if (s->sk_net != sk->sk_net) + if (sock_net(s) != sock_net(sk)) continue; if (s->sk_hash == num && s->sk_family == PF_INET6) { diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c index 6f21a53cb3e7..ae54b20d0470 100644 --- a/net/irda/af_irda.c +++ b/net/irda/af_irda.c @@ -837,7 +837,7 @@ static int irda_accept(struct socket *sock, struct socket *newsock, int flags) IRDA_DEBUG(2, "%s()\n", __func__); - err = irda_create(sk->sk_net, newsock, sk->sk_protocol); + err = irda_create(sock_net(sk), newsock, sk->sk_protocol); if (err) return err; diff --git a/net/llc/llc_conn.c b/net/llc/llc_conn.c index 5ebfd93ff5e7..5c6d89c6d51d 100644 --- a/net/llc/llc_conn.c +++ b/net/llc/llc_conn.c @@ -700,7 +700,7 @@ static struct sock *llc_create_incoming_sock(struct sock *sk, struct llc_addr *saddr, struct llc_addr *daddr) { - struct sock *newsk = llc_sk_alloc(sk->sk_net, sk->sk_family, GFP_ATOMIC, + struct sock *newsk = llc_sk_alloc(sock_net(sk), sk->sk_family, GFP_ATOMIC, sk->sk_prot); struct llc_sock *newllc, *llc = llc_sk(sk); diff --git a/net/netfilter/nf_sockopt.c b/net/netfilter/nf_sockopt.c index 3dd4b3c76d81..69d699f95f4c 100644 --- a/net/netfilter/nf_sockopt.c +++ b/net/netfilter/nf_sockopt.c @@ -65,7 +65,7 @@ static struct nf_sockopt_ops *nf_sockopt_find(struct sock *sk, int pf, { struct nf_sockopt_ops *ops; - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return ERR_PTR(-ENOPROTOOPT); if (mutex_lock_interruptible(&nf_sockopt_mutex) != 0) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 86bd8660a8f2..712a7bff8560 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -228,7 +228,7 @@ static inline struct sock *netlink_lookup(struct net *net, int protocol, read_lock(&nl_table_lock); head = nl_pid_hashfn(hash, pid); sk_for_each(sk, node, head) { - if ((sk->sk_net == net) && (nlk_sk(sk)->pid == pid)) { + if (sock_net(sk) == net && (nlk_sk(sk)->pid == pid)) { sock_hold(sk); goto found; } @@ -348,7 +348,7 @@ static int netlink_insert(struct sock *sk, struct net *net, u32 pid) head = nl_pid_hashfn(hash, pid); len = 0; sk_for_each(osk, node, head) { - if ((osk->sk_net == net) && (nlk_sk(osk)->pid == pid)) + if (sock_net(osk) == net && (nlk_sk(osk)->pid == pid)) break; len++; } @@ -486,7 +486,7 @@ static int netlink_release(struct socket *sock) if (nlk->pid && !nlk->subscriptions) { struct netlink_notify n = { - .net = sk->sk_net, + .net = sock_net(sk), .protocol = sk->sk_protocol, .pid = nlk->pid, }; @@ -518,7 +518,7 @@ static int netlink_release(struct socket *sock) static int netlink_autobind(struct socket *sock) { struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct nl_pid_hash *hash = &nl_table[sk->sk_protocol].hash; struct hlist_head *head; struct sock *osk; @@ -532,7 +532,7 @@ retry: netlink_table_grab(); head = nl_pid_hashfn(hash, pid); sk_for_each(osk, node, head) { - if ((osk->sk_net != net)) + if (sock_net(osk) != net) continue; if (nlk_sk(osk)->pid == pid) { /* Bind collision, search negative pid values. */ @@ -611,7 +611,7 @@ static int netlink_bind(struct socket *sock, struct sockaddr *addr, int addr_len) { struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct netlink_sock *nlk = nlk_sk(sk); struct sockaddr_nl *nladdr = (struct sockaddr_nl *)addr; int err; @@ -720,7 +720,7 @@ static struct sock *netlink_getsockbypid(struct sock *ssk, u32 pid) struct sock *sock; struct netlink_sock *nlk; - sock = netlink_lookup(ssk->sk_net, ssk->sk_protocol, pid); + sock = netlink_lookup(sock_net(ssk), ssk->sk_protocol, pid); if (!sock) return ERR_PTR(-ECONNREFUSED); @@ -962,7 +962,7 @@ static inline int do_one_broadcast(struct sock *sk, !test_bit(p->group - 1, nlk->groups)) goto out; - if ((sk->sk_net != p->net)) + if (sock_net(sk) != p->net) goto out; if (p->failure) { @@ -1006,7 +1006,7 @@ out: int netlink_broadcast(struct sock *ssk, struct sk_buff *skb, u32 pid, u32 group, gfp_t allocation) { - struct net *net = ssk->sk_net; + struct net *net = sock_net(ssk); struct netlink_broadcast_data info; struct hlist_node *node; struct sock *sk; @@ -1064,7 +1064,7 @@ static inline int do_one_set_err(struct sock *sk, if (sk == p->exclude_sk) goto out; - if (sk->sk_net != p->exclude_sk->sk_net) + if (sock_net(sk) != sock_net(p->exclude_sk)) goto out; if (nlk->pid == p->pid || p->group - 1 >= nlk->ngroups || @@ -1601,7 +1601,7 @@ int netlink_dump_start(struct sock *ssk, struct sk_buff *skb, atomic_inc(&skb->users); cb->skb = skb; - sk = netlink_lookup(ssk->sk_net, ssk->sk_protocol, NETLINK_CB(skb).pid); + sk = netlink_lookup(sock_net(ssk), ssk->sk_protocol, NETLINK_CB(skb).pid); if (sk == NULL) { netlink_destroy_callback(cb); return -ECONNREFUSED; @@ -1643,7 +1643,7 @@ void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err) if (!skb) { struct sock *sk; - sk = netlink_lookup(in_skb->sk->sk_net, + sk = netlink_lookup(sock_net(in_skb->sk), in_skb->sk->sk_protocol, NETLINK_CB(in_skb).pid); if (sk) { @@ -1758,7 +1758,7 @@ static struct sock *netlink_seq_socket_idx(struct seq_file *seq, loff_t pos) for (j = 0; j <= hash->mask; j++) { sk_for_each(s, node, &hash->table[j]) { - if (iter->p.net != s->sk_net) + if (sock_net(s) != iter->p.net) continue; if (off == pos) { iter->link = i; @@ -1794,7 +1794,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos) s = v; do { s = sk_next(s); - } while (s && (iter->p.net != s->sk_net)); + } while (s && (sock_net(s) != iter->p.net)); if (s) return s; @@ -1806,7 +1806,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos) for (; j <= hash->mask; j++) { s = sk_head(&hash->table[j]); - while (s && (iter->p.net != s->sk_net)) + while (s && sock_net(s) != iter->p.net) s = sk_next(s); if (s) { iter->link = i; diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c index a270ebf9f765..4bae8b998cab 100644 --- a/net/netrom/af_netrom.c +++ b/net/netrom/af_netrom.c @@ -466,7 +466,7 @@ static struct sock *nr_make_new(struct sock *osk) if (osk->sk_type != SOCK_SEQPACKET) return NULL; - sk = sk_alloc(osk->sk_net, PF_NETROM, GFP_ATOMIC, osk->sk_prot); + sk = sk_alloc(sock_net(osk), PF_NETROM, GFP_ATOMIC, osk->sk_prot); if (sk == NULL) return NULL; diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index baa290d3444a..25070240d4ae 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -263,7 +263,7 @@ static int packet_rcv_spkt(struct sk_buff *skb, struct net_device *dev, struct if (skb->pkt_type == PACKET_LOOPBACK) goto out; - if (dev_net(dev) != sk->sk_net) + if (dev_net(dev) != sock_net(sk)) goto out; if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) @@ -337,7 +337,7 @@ static int packet_sendmsg_spkt(struct kiocb *iocb, struct socket *sock, */ saddr->spkt_device[13] = 0; - dev = dev_get_by_name(sk->sk_net, saddr->spkt_device); + dev = dev_get_by_name(sock_net(sk), saddr->spkt_device); err = -ENODEV; if (dev == NULL) goto out_unlock; @@ -451,7 +451,7 @@ static int packet_rcv(struct sk_buff *skb, struct net_device *dev, struct packet sk = pt->af_packet_priv; po = pkt_sk(sk); - if (dev_net(dev) != sk->sk_net) + if (dev_net(dev) != sock_net(sk)) goto drop; skb->dev = dev; @@ -568,7 +568,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe sk = pt->af_packet_priv; po = pkt_sk(sk); - if (dev_net(dev) != sk->sk_net) + if (dev_net(dev) != sock_net(sk)) goto drop; if (dev->header_ops) { @@ -728,7 +728,7 @@ static int packet_sendmsg(struct kiocb *iocb, struct socket *sock, } - dev = dev_get_by_index(sk->sk_net, ifindex); + dev = dev_get_by_index(sock_net(sk), ifindex); err = -ENXIO; if (dev == NULL) goto out_unlock; @@ -800,7 +800,7 @@ static int packet_release(struct socket *sock) if (!sk) return 0; - net = sk->sk_net; + net = sock_net(sk); po = pkt_sk(sk); write_lock_bh(&net->packet.sklist_lock); @@ -914,7 +914,7 @@ static int packet_bind_spkt(struct socket *sock, struct sockaddr *uaddr, int add return -EINVAL; strlcpy(name,uaddr->sa_data,sizeof(name)); - dev = dev_get_by_name(sk->sk_net, name); + dev = dev_get_by_name(sock_net(sk), name); if (dev) { err = packet_do_bind(sk, dev, pkt_sk(sk)->num); dev_put(dev); @@ -941,7 +941,7 @@ static int packet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len if (sll->sll_ifindex) { err = -ENODEV; - dev = dev_get_by_index(sk->sk_net, sll->sll_ifindex); + dev = dev_get_by_index(sock_net(sk), sll->sll_ifindex); if (dev == NULL) goto out; } @@ -1135,7 +1135,7 @@ static int packet_getname_spkt(struct socket *sock, struct sockaddr *uaddr, return -EOPNOTSUPP; uaddr->sa_family = AF_PACKET; - dev = dev_get_by_index(sk->sk_net, pkt_sk(sk)->ifindex); + dev = dev_get_by_index(sock_net(sk), pkt_sk(sk)->ifindex); if (dev) { strlcpy(uaddr->sa_data, dev->name, 15); dev_put(dev); @@ -1160,7 +1160,7 @@ static int packet_getname(struct socket *sock, struct sockaddr *uaddr, sll->sll_family = AF_PACKET; sll->sll_ifindex = po->ifindex; sll->sll_protocol = po->num; - dev = dev_get_by_index(sk->sk_net, po->ifindex); + dev = dev_get_by_index(sock_net(sk), po->ifindex); if (dev) { sll->sll_hatype = dev->type; sll->sll_halen = dev->addr_len; @@ -1212,7 +1212,7 @@ static int packet_mc_add(struct sock *sk, struct packet_mreq_max *mreq) rtnl_lock(); err = -ENODEV; - dev = __dev_get_by_index(sk->sk_net, mreq->mr_ifindex); + dev = __dev_get_by_index(sock_net(sk), mreq->mr_ifindex); if (!dev) goto done; @@ -1266,7 +1266,7 @@ static int packet_mc_drop(struct sock *sk, struct packet_mreq_max *mreq) if (--ml->count == 0) { struct net_device *dev; *mlp = ml->next; - dev = dev_get_by_index(sk->sk_net, ml->ifindex); + dev = dev_get_by_index(sock_net(sk), ml->ifindex); if (dev) { packet_dev_mc(dev, ml, -1); dev_put(dev); @@ -1294,7 +1294,7 @@ static void packet_flush_mclist(struct sock *sk) struct net_device *dev; po->mclist = ml->next; - if ((dev = dev_get_by_index(sk->sk_net, ml->ifindex)) != NULL) { + if ((dev = dev_get_by_index(sock_net(sk), ml->ifindex)) != NULL) { packet_dev_mc(dev, ml, -1); dev_put(dev); } @@ -1540,7 +1540,7 @@ static int packet_ioctl(struct socket *sock, unsigned int cmd, case SIOCGIFDSTADDR: case SIOCSIFDSTADDR: case SIOCSIFFLAGS: - if (sk->sk_net != &init_net) + if (sock_net(sk) != &init_net) return -ENOIOCTLCMD; return inet_dgram_ops.ioctl(sock, cmd, arg); #endif diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c index 1a7f143cf741..92d85c38e4d2 100644 --- a/net/rose/af_rose.c +++ b/net/rose/af_rose.c @@ -551,7 +551,7 @@ static struct sock *rose_make_new(struct sock *osk) if (osk->sk_type != SOCK_SEQPACKET) return NULL; - sk = sk_alloc(osk->sk_net, PF_ROSE, GFP_ATOMIC, &rose_proto); + sk = sk_alloc(sock_net(osk), PF_ROSE, GFP_ATOMIC, &rose_proto); if (sk == NULL) return NULL; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 0b8eb235bc13..74e662cbb2c5 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -951,7 +951,7 @@ done: static int tc_ctl_action(struct sk_buff *skb, struct nlmsghdr *n, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct nlattr *tca[TCA_ACT_MAX + 1]; u32 pid = skb ? NETLINK_CB(skb).pid : 0; int ret = 0, ovr = 0; @@ -1029,7 +1029,7 @@ find_dump_kind(struct nlmsghdr *n) static int tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct nlmsghdr *nlh; unsigned char *b = skb_tail_pointer(skb); struct nlattr *nest; diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 0fbedcabf111..1086df7478bc 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -118,7 +118,7 @@ static inline u32 tcf_auto_prio(struct tcf_proto *tp) static int tc_ctl_tfilter(struct sk_buff *skb, struct nlmsghdr *n, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct nlattr *tca[TCA_MAX + 1]; struct tcmsg *t; u32 protocol; @@ -389,7 +389,7 @@ static int tcf_node_dump(struct tcf_proto *tp, unsigned long n, static int tc_dump_tfilter(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int t; int s_t; struct net_device *dev; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 7e3c048ba9b1..15b91a9ee8e8 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -605,7 +605,7 @@ check_loop_fn(struct Qdisc *q, unsigned long cl, struct qdisc_walker *w) static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct tcmsg *tcm = NLMSG_DATA(n); struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; @@ -674,7 +674,7 @@ static int tc_get_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) static int tc_modify_qdisc(struct sk_buff *skb, struct nlmsghdr *n, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct tcmsg *tcm; struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; @@ -893,7 +893,7 @@ err_out: static int tc_dump_qdisc(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int idx, q_idx; int s_idx, s_q_idx; struct net_device *dev; @@ -945,7 +945,7 @@ done: static int tc_ctl_tclass(struct sk_buff *skb, struct nlmsghdr *n, void *arg) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); struct tcmsg *tcm = NLMSG_DATA(n); struct nlattr *tca[TCA_MAX + 1]; struct net_device *dev; @@ -1139,7 +1139,7 @@ static int qdisc_class_dump(struct Qdisc *q, unsigned long cl, struct qdisc_walk static int tc_dump_tclass(struct sk_buff *skb, struct netlink_callback *cb) { - struct net *net = skb->sk->sk_net; + struct net *net = sock_net(skb->sk); int t; int s_t; struct net_device *dev; diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c index dc71d0d83753..036bfcc8d15b 100644 --- a/net/sctp/ipv6.c +++ b/net/sctp/ipv6.c @@ -636,7 +636,7 @@ static struct sock *sctp_v6_create_accept_sk(struct sock *sk, struct ipv6_pinfo *newnp, *np = inet6_sk(sk); struct sctp6_sock *newsctp6sk; - newsk = sk_alloc(sk->sk_net, PF_INET6, GFP_KERNEL, sk->sk_prot); + newsk = sk_alloc(sock_net(sk), PF_INET6, GFP_KERNEL, sk->sk_prot); if (!newsk) goto out; diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 2faa0d8839eb..5aea91137fbb 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -554,7 +554,7 @@ 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(sk->sk_net, PF_INET, GFP_KERNEL, + struct sock *newsk = sk_alloc(sock_net(sk), PF_INET, GFP_KERNEL, sk->sk_prot); if (!newsk) diff --git a/net/socket.c b/net/socket.c index 9d3fbfbc8535..79e5382fd110 100644 --- a/net/socket.c +++ b/net/socket.c @@ -857,7 +857,7 @@ static long sock_ioctl(struct file *file, unsigned cmd, unsigned long arg) sock = file->private_data; sk = sock->sk; - net = sk->sk_net; + net = sock_net(sk); if (cmd >= SIOCDEVPRIVATE && cmd <= (SIOCDEVPRIVATE + 15)) { err = dev_ioctl(net, cmd, argp); } else @@ -1375,7 +1375,7 @@ asmlinkage long sys_listen(int fd, int backlog) sock = sockfd_lookup_light(fd, &err, &fput_needed); if (sock) { - somaxconn = sock->sk->sk_net->sysctl_somaxconn; + somaxconn = sock_net(sock->sk)->sysctl_somaxconn; if ((unsigned)backlog > somaxconn) backlog = somaxconn; diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 3220d5cb5b5d..ae45df060e3a 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -1375,7 +1375,7 @@ static int accept(struct socket *sock, struct socket *newsock, int flags) } buf = skb_peek(&sock->sk->sk_receive_queue); - res = tipc_create(sock->sk->sk_net, newsock, 0); + res = tipc_create(sock_net(sock->sk), newsock, 0); if (!res) { struct tipc_sock *new_tsock = tipc_sk(newsock->sk); struct tipc_portid id; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index ae584356852c..cb9d0cb5f270 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -252,7 +252,7 @@ static struct sock *__unix_find_socket_byname(struct net *net, sk_for_each(s, node, &unix_socket_table[hash ^ type]) { struct unix_sock *u = unix_sk(s); - if (s->sk_net != net) + if (sock_net(s) != net) continue; if (u->addr->len == len && @@ -289,7 +289,7 @@ static struct sock *unix_find_socket_byinode(struct net *net, struct inode *i) &unix_socket_table[i->i_ino & (UNIX_HASH_SIZE - 1)]) { struct dentry *dentry = unix_sk(s)->dentry; - if (s->sk_net != net) + if (sock_net(s) != net) continue; if(dentry && dentry->d_inode == i) @@ -654,7 +654,7 @@ static int unix_release(struct socket *sock) static int unix_autobind(struct socket *sock) { struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct unix_sock *u = unix_sk(sk); static u32 ordernum = 1; struct unix_address * addr; @@ -758,7 +758,7 @@ fail: static int unix_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) { struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct unix_sock *u = unix_sk(sk); struct sockaddr_un *sunaddr=(struct sockaddr_un *)uaddr; struct dentry * dentry = NULL; @@ -899,7 +899,7 @@ static int unix_dgram_connect(struct socket *sock, struct sockaddr *addr, int alen, int flags) { struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct sockaddr_un *sunaddr=(struct sockaddr_un*)addr; struct sock *other; unsigned hash; @@ -996,7 +996,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr, { struct sockaddr_un *sunaddr=(struct sockaddr_un *)uaddr; struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct unix_sock *u = unix_sk(sk), *newu, *otheru; struct sock *newsk = NULL; struct sock *other = NULL; @@ -1025,7 +1025,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr, err = -ENOMEM; /* create new sock for complete connection */ - newsk = unix_create1(sk->sk_net, NULL); + newsk = unix_create1(sock_net(sk), NULL); if (newsk == NULL) goto out; @@ -1312,7 +1312,7 @@ static int unix_dgram_sendmsg(struct kiocb *kiocb, struct socket *sock, { struct sock_iocb *siocb = kiocb_to_siocb(kiocb); struct sock *sk = sock->sk; - struct net *net = sk->sk_net; + struct net *net = sock_net(sk); struct unix_sock *u = unix_sk(sk); struct sockaddr_un *sunaddr=msg->msg_name; struct sock *other = NULL; @@ -2022,7 +2022,7 @@ static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos) struct sock *s; for (s = first_unix_socket(&iter->i); s; s = next_unix_socket(&iter->i, s)) { - if (s->sk_net != iter->p.net) + if (sock_net(s) != iter->p.net) continue; if (off == pos) return s; @@ -2050,7 +2050,7 @@ static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos) sk = first_unix_socket(&iter->i); else sk = next_unix_socket(&iter->i, sk); - while (sk && (sk->sk_net != iter->p.net)) + while (sk && (sock_net(sk) != iter->p.net)) sk = next_unix_socket(&iter->i, sk); return sk; } diff --git a/net/x25/af_x25.c b/net/x25/af_x25.c index 7a46ea73fe2d..6ba67c523c16 100644 --- a/net/x25/af_x25.c +++ b/net/x25/af_x25.c @@ -549,7 +549,7 @@ static struct sock *x25_make_new(struct sock *osk) if (osk->sk_type != SOCK_SEQPACKET) goto out; - if ((sk = x25_alloc_socket(osk->sk_net)) == NULL) + if ((sk = x25_alloc_socket(sock_net(osk))) == NULL) goto out; x25 = x25_sk(sk); -- cgit v1.2.3 From 1218854afa6f659be90b748cf1bc7badee954a35 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 26 Mar 2008 02:36:06 +0900 Subject: [NET] NETNS: Omit seq_net_private->net without CONFIG_NET_NS. Without CONFIG_NET_NS, no namespace other than &init_net exists, no need to store net in seq_net_private. Signed-off-by: YOSHIFUJI Hideaki --- fs/proc/proc_net.c | 6 +++--- include/linux/seq_file.h | 7 +++++++ net/core/neighbour.c | 8 ++++---- net/ipv4/fib_hash.c | 5 ++--- net/ipv4/fib_trie.c | 13 ++++++------- net/ipv4/raw.c | 4 ++-- net/ipv4/route.c | 29 +++++++++++++++-------------- net/ipv6/addrconf.c | 4 ++-- net/ipv6/mcast.c | 4 ++-- net/netfilter/x_tables.c | 4 ++-- net/netlink/af_netlink.c | 6 +++--- net/unix/af_unix.c | 10 +++++----- 12 files changed, 53 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 4caa5f774fb7..13cd7835d0df 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -44,7 +44,9 @@ int seq_open_net(struct inode *ino, struct file *f, put_net(net); return -ENOMEM; } +#ifdef CONFIG_NET_NS p->net = net; +#endif return 0; } EXPORT_SYMBOL_GPL(seq_open_net); @@ -52,12 +54,10 @@ EXPORT_SYMBOL_GPL(seq_open_net); int seq_release_net(struct inode *ino, struct file *f) { struct seq_file *seq; - struct seq_net_private *p; seq = f->private_data; - p = seq->private; - put_net(p->net); + put_net(seq_file_net(seq)); seq_release_private(ino, f); return 0; } diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index 67c2563961f3..d870a8253769 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -5,6 +5,7 @@ #include #include #include +#include struct seq_operations; struct file; @@ -64,7 +65,9 @@ extern struct list_head *seq_list_next(void *v, struct list_head *head, struct net; struct seq_net_private { +#ifdef CONFIG_NET_NS struct net *net; +#endif }; int seq_open_net(struct inode *, struct file *, @@ -72,7 +75,11 @@ int seq_open_net(struct inode *, struct file *, int seq_release_net(struct inode *, struct file *); static inline struct net *seq_file_net(struct seq_file *seq) { +#ifdef CONFIG_NET_NS return ((struct seq_net_private *)seq->private)->net; +#else + return &init_net; +#endif } #endif diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 065fbac7ecd3..b8d491fb4b42 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -2145,7 +2145,7 @@ EXPORT_SYMBOL(__neigh_for_each_release); static struct neighbour *neigh_get_first(struct seq_file *seq) { struct neigh_seq_state *state = seq->private; - struct net *net = state->p.net; + struct net *net = seq_file_net(seq); struct neigh_table *tbl = state->tbl; struct neighbour *n = NULL; int bucket = state->bucket; @@ -2186,7 +2186,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, loff_t *pos) { struct neigh_seq_state *state = seq->private; - struct net *net = state->p.net; + struct net *net = seq_file_net(seq); struct neigh_table *tbl = state->tbl; if (state->neigh_sub_iter) { @@ -2246,7 +2246,7 @@ static struct neighbour *neigh_get_idx(struct seq_file *seq, loff_t *pos) static struct pneigh_entry *pneigh_get_first(struct seq_file *seq) { struct neigh_seq_state *state = seq->private; - struct net * net = state->p.net; + struct net *net = seq_file_net(seq); struct neigh_table *tbl = state->tbl; struct pneigh_entry *pn = NULL; int bucket = state->bucket; @@ -2269,7 +2269,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq, loff_t *pos) { struct neigh_seq_state *state = seq->private; - struct net * net = state->p.net; + struct net *net = seq_file_net(seq); struct neigh_table *tbl = state->tbl; pn = pn->next; diff --git a/net/ipv4/fib_hash.c b/net/ipv4/fib_hash.c index 8d58d85dfac6..02088deb0461 100644 --- a/net/ipv4/fib_hash.c +++ b/net/ipv4/fib_hash.c @@ -821,7 +821,7 @@ static struct fib_alias *fib_get_first(struct seq_file *seq) struct fib_table *main_table; struct fn_hash *table; - main_table = fib_get_table(iter->p.net, RT_TABLE_MAIN); + main_table = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN); table = (struct fn_hash *)main_table->tb_data; iter->bucket = 0; @@ -959,11 +959,10 @@ static struct fib_alias *fib_get_idx(struct seq_file *seq, loff_t pos) static void *fib_seq_start(struct seq_file *seq, loff_t *pos) __acquires(fib_hash_lock) { - struct fib_iter_state *iter = seq->private; void *v = NULL; read_lock(&fib_hash_lock); - if (fib_get_table(iter->p.net, RT_TABLE_MAIN)) + if (fib_get_table(seq_file_net(seq), RT_TABLE_MAIN)) v = *pos ? fib_get_idx(seq, *pos - 1) : SEQ_START_TOKEN; return v; } diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index ce6cb34e28e1..9e491e70e855 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -2279,9 +2279,10 @@ static const struct file_operations fib_triestat_fops = { .release = fib_triestat_seq_release, }; -static struct node *fib_trie_get_idx(struct fib_trie_iter *iter, loff_t pos) +static struct node *fib_trie_get_idx(struct seq_file *seq, loff_t pos) { - struct net *net = iter->p.net; + struct fib_trie_iter *iter = seq->private; + struct net *net = seq_file_net(seq); loff_t idx = 0; unsigned int h; @@ -2309,16 +2310,14 @@ static struct node *fib_trie_get_idx(struct fib_trie_iter *iter, loff_t pos) static void *fib_trie_seq_start(struct seq_file *seq, loff_t *pos) __acquires(RCU) { - struct fib_trie_iter *iter = seq->private; - rcu_read_lock(); - return fib_trie_get_idx(iter, *pos); + return fib_trie_get_idx(seq, *pos); } static void *fib_trie_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct fib_trie_iter *iter = seq->private; - struct net *net = iter->p.net; + struct net *net = seq_file_net(seq); struct fib_table *tb = iter->tb; struct hlist_node *tb_node; unsigned int h; @@ -2513,7 +2512,7 @@ static void *fib_route_seq_start(struct seq_file *seq, loff_t *pos) struct fib_table *tb; rcu_read_lock(); - tb = fib_get_table(iter->p.net, RT_TABLE_MAIN); + tb = fib_get_table(seq_file_net(seq), RT_TABLE_MAIN); if (!tb) return NULL; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index be19a4048d7c..25dc8b38cac3 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -856,7 +856,7 @@ static struct sock *raw_get_first(struct seq_file *seq) struct hlist_node *node; sk_for_each(sk, node, &state->h->ht[state->bucket]) - if (sock_net(sk) == state->p.net) + if (sock_net(sk) == seq_file_net(seq)) goto found; } sk = NULL; @@ -872,7 +872,7 @@ static struct sock *raw_get_next(struct seq_file *seq, struct sock *sk) sk = sk_next(sk); try_again: ; - } while (sk && sock_net(sk) != state->p.net); + } while (sk && sock_net(sk) != seq_file_net(seq)); if (!sk && ++state->bucket < RAW_HTABLE_SIZE) { sk = sk_head(&state->h->ht[state->bucket]); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 194f5cca3121..eab8d75e5222 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -276,15 +276,16 @@ struct rt_cache_iter_state { int genid; }; -static struct rtable *rt_cache_get_first(struct rt_cache_iter_state *st) +static struct rtable *rt_cache_get_first(struct seq_file *seq) { + struct rt_cache_iter_state *st = seq->private; struct rtable *r = NULL; for (st->bucket = rt_hash_mask; st->bucket >= 0; --st->bucket) { rcu_read_lock_bh(); r = rcu_dereference(rt_hash_table[st->bucket].chain); while (r) { - if (dev_net(r->u.dst.dev) == st->p.net && + if (dev_net(r->u.dst.dev) == seq_file_net(seq) && r->rt_genid == st->genid) return r; r = rcu_dereference(r->u.dst.rt_next); @@ -294,9 +295,10 @@ static struct rtable *rt_cache_get_first(struct rt_cache_iter_state *st) return r; } -static struct rtable *__rt_cache_get_next(struct rt_cache_iter_state *st, +static struct rtable *__rt_cache_get_next(struct seq_file *seq, struct rtable *r) { + struct rt_cache_iter_state *st = seq->private; r = r->u.dst.rt_next; while (!r) { rcu_read_unlock_bh(); @@ -308,11 +310,12 @@ static struct rtable *__rt_cache_get_next(struct rt_cache_iter_state *st, return rcu_dereference(r); } -static struct rtable *rt_cache_get_next(struct rt_cache_iter_state *st, +static struct rtable *rt_cache_get_next(struct seq_file *seq, struct rtable *r) { - while ((r = __rt_cache_get_next(st, r)) != NULL) { - if (dev_net(r->u.dst.dev) != st->p.net) + struct rt_cache_iter_state *st = seq->private; + while ((r = __rt_cache_get_next(seq, r)) != NULL) { + if (dev_net(r->u.dst.dev) != seq_file_net(seq)) continue; if (r->rt_genid == st->genid) break; @@ -320,12 +323,12 @@ static struct rtable *rt_cache_get_next(struct rt_cache_iter_state *st, return r; } -static struct rtable *rt_cache_get_idx(struct rt_cache_iter_state *st, loff_t pos) +static struct rtable *rt_cache_get_idx(struct seq_file *seq, loff_t pos) { - struct rtable *r = rt_cache_get_first(st); + struct rtable *r = rt_cache_get_first(seq); if (r) - while (pos && (r = rt_cache_get_next(st, r))) + while (pos && (r = rt_cache_get_next(seq, r))) --pos; return pos ? NULL : r; } @@ -333,9 +336,8 @@ static struct rtable *rt_cache_get_idx(struct rt_cache_iter_state *st, loff_t po static void *rt_cache_seq_start(struct seq_file *seq, loff_t *pos) { struct rt_cache_iter_state *st = seq->private; - if (*pos) - return rt_cache_get_idx(st, *pos - 1); + return rt_cache_get_idx(seq, *pos - 1); st->genid = atomic_read(&rt_genid); return SEQ_START_TOKEN; } @@ -343,12 +345,11 @@ static void *rt_cache_seq_start(struct seq_file *seq, loff_t *pos) static void *rt_cache_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct rtable *r; - struct rt_cache_iter_state *st = seq->private; if (v == SEQ_START_TOKEN) - r = rt_cache_get_first(st); + r = rt_cache_get_first(seq); else - r = rt_cache_get_next(st, v); + r = rt_cache_get_next(seq, v); ++*pos; return r; } diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index f2c90f145cbb..ac5d4f4b6312 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2766,7 +2766,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq) { struct inet6_ifaddr *ifa = NULL; struct if6_iter_state *state = seq->private; - struct net *net = state->p.net; + struct net *net = seq_file_net(seq); for (state->bucket = 0; state->bucket < IN6_ADDR_HSIZE; ++state->bucket) { ifa = inet6_addr_lst[state->bucket]; @@ -2782,7 +2782,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq) static struct inet6_ifaddr *if6_get_next(struct seq_file *seq, struct inet6_ifaddr *ifa) { struct if6_iter_state *state = seq->private; - struct net *net = state->p.net; + struct net *net = seq_file_net(seq); ifa = ifa->lst_next; try_again: diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 20a3d8e2f6c6..d810cff818cf 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -2355,7 +2355,7 @@ static inline struct ifmcaddr6 *igmp6_mc_get_first(struct seq_file *seq) { struct ifmcaddr6 *im = NULL; struct igmp6_mc_iter_state *state = igmp6_mc_seq_private(seq); - struct net *net = state->p.net; + struct net *net = seq_file_net(seq); state->idev = NULL; for_each_netdev(net, state->dev) { @@ -2486,7 +2486,7 @@ static inline struct ip6_sf_list *igmp6_mcf_get_first(struct seq_file *seq) struct ip6_sf_list *psf = NULL; struct ifmcaddr6 *im = NULL; struct igmp6_mcf_iter_state *state = igmp6_mcf_seq_private(seq); - struct net *net = state->p.net; + struct net *net = seq_file_net(seq); state->idev = NULL; state->im = NULL; diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index a6792089fcf9..0bd95680a494 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -727,7 +727,7 @@ struct xt_names_priv { static void *xt_table_seq_start(struct seq_file *seq, loff_t *pos) { struct xt_names_priv *priv = seq->private; - struct net *net = priv->p.net; + struct net *net = seq_file_net(seq); int af = priv->af; mutex_lock(&xt[af].mutex); @@ -737,7 +737,7 @@ static void *xt_table_seq_start(struct seq_file *seq, loff_t *pos) static void *xt_table_seq_next(struct seq_file *seq, void *v, loff_t *pos) { struct xt_names_priv *priv = seq->private; - struct net *net = priv->p.net; + struct net *net = seq_file_net(seq); int af = priv->af; return seq_list_next(v, &net->xt.tables[af], pos); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 712a7bff8560..1d16d95dfaaf 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1758,7 +1758,7 @@ static struct sock *netlink_seq_socket_idx(struct seq_file *seq, loff_t pos) for (j = 0; j <= hash->mask; j++) { sk_for_each(s, node, &hash->table[j]) { - if (sock_net(s) != iter->p.net) + if (sock_net(s) != seq_file_net(seq)) continue; if (off == pos) { iter->link = i; @@ -1794,7 +1794,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos) s = v; do { s = sk_next(s); - } while (s && (sock_net(s) != iter->p.net)); + } while (s && sock_net(s) != seq_file_net(seq)); if (s) return s; @@ -1806,7 +1806,7 @@ static void *netlink_seq_next(struct seq_file *seq, void *v, loff_t *pos) for (; j <= hash->mask; j++) { s = sk_head(&hash->table[j]); - while (s && sock_net(s) != iter->p.net) + while (s && sock_net(s) != seq_file_net(seq)) s = sk_next(s); if (s) { iter->link = i; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index cb9d0cb5f270..4a4793051bcb 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2016,13 +2016,14 @@ struct unix_iter_state { struct seq_net_private p; int i; }; -static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos) +static struct sock *unix_seq_idx(struct seq_file *seq, loff_t pos) { + struct unix_iter_state *iter = seq->private; loff_t off = 0; struct sock *s; for (s = first_unix_socket(&iter->i); s; s = next_unix_socket(&iter->i, s)) { - if (sock_net(s) != iter->p.net) + if (sock_net(s) != seq_file_net(seq)) continue; if (off == pos) return s; @@ -2035,9 +2036,8 @@ static struct sock *unix_seq_idx(struct unix_iter_state *iter, loff_t pos) static void *unix_seq_start(struct seq_file *seq, loff_t *pos) __acquires(unix_table_lock) { - struct unix_iter_state *iter = seq->private; spin_lock(&unix_table_lock); - return *pos ? unix_seq_idx(iter, *pos - 1) : ((void *) 1); + return *pos ? unix_seq_idx(seq, *pos - 1) : ((void *) 1); } static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos) @@ -2050,7 +2050,7 @@ static void *unix_seq_next(struct seq_file *seq, void *v, loff_t *pos) sk = first_unix_socket(&iter->i); else sk = next_unix_socket(&iter->i, sk); - while (sk && (sock_net(sk) != iter->p.net)) + while (sk && (sock_net(sk) != seq_file_net(seq))) sk = next_unix_socket(&iter->i, sk); return sk; } -- cgit v1.2.3 From 57da52c1e62c6c13875e97de6c69d3156f8416da Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 26 Mar 2008 03:49:59 +0900 Subject: [NET] NETNS: Omit neigh_parms->net and pneigh_entry->net without CONFIG_NET_NS. Introduce neigh_parms/pneigh_entry inlines: neigh_parms_net(), pneigh_net(). Without CONFIG_NET_NS, no namespace other than &init_net exists. Let's explicitly define them to help compiler optimizations. Signed-off-by: YOSHIFUJI Hideaki --- include/net/neighbour.h | 25 +++++++++++++++++++++++++ net/core/neighbour.c | 26 ++++++++++++++++---------- 2 files changed, 41 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 062281872064..8bec0d69b270 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -38,7 +38,9 @@ struct neighbour; struct neigh_parms { +#ifdef CONFIG_NET_NS struct net *net; +#endif struct net_device *dev; struct neigh_parms *next; int (*neigh_setup)(struct neighbour *); @@ -131,7 +133,9 @@ struct neigh_ops struct pneigh_entry { struct pneigh_entry *next; +#ifdef CONFIG_NET_NS struct net *net; +#endif struct net_device *dev; u8 flags; u8 key[0]; @@ -213,6 +217,17 @@ extern struct neighbour *neigh_event_ns(struct neigh_table *tbl, extern struct neigh_parms *neigh_parms_alloc(struct net_device *dev, struct neigh_table *tbl); extern void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms); + +static inline +struct net *neigh_parms_net(const struct neigh_parms *parms) +{ +#ifdef CONFIG_NET_NS + return parms->net; +#else + return &init_net; +#endif +} + extern unsigned long neigh_rand_reach_time(unsigned long base); extern void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p, @@ -220,6 +235,16 @@ extern void pneigh_enqueue(struct neigh_table *tbl, struct neigh_parms *p, extern struct pneigh_entry *pneigh_lookup(struct neigh_table *tbl, struct net *net, const void *key, struct net_device *dev, int creat); extern int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *key, struct net_device *dev); +static inline +struct net *pneigh_net(const struct pneigh_entry *pneigh) +{ +#ifdef CONFIG_NET_NS + return pneigh->net; +#else + return &init_net; +#endif +} + extern void neigh_app_ns(struct neighbour *n); extern void neigh_for_each(struct neigh_table *tbl, void (*cb)(struct neighbour *, void *), void *cookie); extern void __neigh_for_each_release(struct neigh_table *tbl, int (*cb)(struct neighbour *)); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index b8d491fb4b42..de654ea8a944 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -483,7 +483,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, for (n = tbl->phash_buckets[hash_val]; n; n = n->next) { if (!memcmp(n->key, pkey, key_len) && - (n->net == net) && + (pneigh_net(n) == net) && (n->dev == dev || !n->dev)) { read_unlock_bh(&tbl->lock); goto out; @@ -500,7 +500,9 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, if (!n) goto out; +#ifdef CONFIG_NET_NS n->net = hold_net(net); +#endif memcpy(n->key, pkey, key_len); n->dev = dev; if (dev) @@ -540,14 +542,14 @@ int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey, for (np = &tbl->phash_buckets[hash_val]; (n = *np) != NULL; np = &n->next) { if (!memcmp(n->key, pkey, key_len) && n->dev == dev && - (n->net == net)) { + (pneigh_net(n) == net)) { *np = n->next; write_unlock_bh(&tbl->lock); if (tbl->pdestructor) tbl->pdestructor(n); if (n->dev) dev_put(n->dev); - release_net(n->net); + release_net(pneigh_net(n)); kfree(n); return 0; } @@ -570,7 +572,7 @@ static int pneigh_ifdown(struct neigh_table *tbl, struct net_device *dev) tbl->pdestructor(n); if (n->dev) dev_put(n->dev); - release_net(n->net); + release_net(pneigh_net(n)); kfree(n); continue; } @@ -1284,7 +1286,7 @@ static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl, struct neigh_parms *p; for (p = &tbl->parms; p; p = p->next) { - if ((p->dev && p->dev->ifindex == ifindex && p->net == net) || + if ((p->dev && p->dev->ifindex == ifindex && neigh_parms_net(p) == net) || (!p->dev && !ifindex)) return p; } @@ -1318,7 +1320,9 @@ struct neigh_parms *neigh_parms_alloc(struct net_device *dev, dev_hold(dev); p->dev = dev; +#ifdef CONFIG_NET_NS p->net = hold_net(net); +#endif p->sysctl_table = NULL; write_lock_bh(&tbl->lock); p->next = tbl->parms.next; @@ -1360,7 +1364,7 @@ void neigh_parms_release(struct neigh_table *tbl, struct neigh_parms *parms) static void neigh_parms_destroy(struct neigh_parms *parms) { - release_net(parms->net); + release_net(neigh_parms_net(parms)); kfree(parms); } @@ -1371,7 +1375,9 @@ void neigh_table_init_no_netlink(struct neigh_table *tbl) unsigned long now = jiffies; unsigned long phsize; +#ifdef CONFIG_NET_NS tbl->parms.net = &init_net; +#endif atomic_set(&tbl->parms.refcnt, 1); INIT_RCU_HEAD(&tbl->parms.rcu_head); tbl->parms.reachable_time = @@ -1958,7 +1964,7 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) break; for (nidx = 0, p = tbl->parms.next; p; p = p->next) { - if (net != p->net) + if (net != neigh_parms_net(p)) continue; if (nidx++ < neigh_skip) @@ -2254,7 +2260,7 @@ static struct pneigh_entry *pneigh_get_first(struct seq_file *seq) state->flags |= NEIGH_SEQ_IS_PNEIGH; for (bucket = 0; bucket <= PNEIGH_HASHMASK; bucket++) { pn = tbl->phash_buckets[bucket]; - while (pn && (pn->net != net)) + while (pn && (pneigh_net(pn) != net)) pn = pn->next; if (pn) break; @@ -2277,7 +2283,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq, if (++state->bucket > PNEIGH_HASHMASK) break; pn = tbl->phash_buckets[state->bucket]; - while (pn && (pn->net != net)) + while (pn && (pneigh_net(pn) != net)) pn = pn->next; if (pn) break; @@ -2740,7 +2746,7 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, neigh_path[NEIGH_CTL_PATH_PROTO].ctl_name = p_id; t->sysctl_header = - register_net_sysctl_table(p->net, neigh_path, t->neigh_vars); + register_net_sysctl_table(neigh_parms_net(p), neigh_path, t->neigh_vars); if (!t->sysctl_header) goto free_procname; -- cgit v1.2.3 From 878628fbf2589eb24357e42027d5f54b1dafd3c8 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Wed, 26 Mar 2008 03:57:35 +0900 Subject: [NET] NETNS: Omit namespace comparision without CONFIG_NET_NS. Introduce an inline net_eq() to compare two namespaces. Without CONFIG_NET_NS, since no namespace other than &init_net exists, it is always 1. We do not need to convert 1) inline vs inline and 2) inline vs &init_net comparisons. Signed-off-by: YOSHIFUJI Hideaki --- include/net/net_namespace.h | 12 ++++++++++++ net/core/dev.c | 2 +- net/core/neighbour.c | 18 +++++++++--------- net/ipv4/inet_hashtables.c | 4 ++-- net/ipv4/raw.c | 2 +- net/ipv4/route.c | 10 +++++----- net/ipv4/tcp_ipv4.c | 12 ++++++------ net/ipv4/udp.c | 10 +++++----- net/ipv6/addrconf.c | 12 ++++++------ net/ipv6/inet6_hashtables.c | 2 +- net/ipv6/raw.c | 2 +- net/ipv6/udp.c | 2 +- net/netlink/af_netlink.c | 8 ++++---- net/unix/af_unix.c | 4 ++-- 14 files changed, 56 insertions(+), 44 deletions(-) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 923f2b8b9096..f8f3d1a5fc35 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -118,6 +118,12 @@ static inline void release_net(struct net *net) { atomic_dec(&net->use_count); } + +static inline +int net_eq(const struct net *net1, const struct net *net2) +{ + return net1 == net2; +} #else static inline struct net *get_net(struct net *net) { @@ -141,6 +147,12 @@ static inline struct net *maybe_get_net(struct net *net) { return net; } + +static inline +int net_eq(const struct net *net1, const struct net *net2) +{ + return 1; +} #endif #define for_each_net(VAR) \ diff --git a/net/core/dev.c b/net/core/dev.c index 812534828914..75c3f7f4edd5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4136,7 +4136,7 @@ int dev_change_net_namespace(struct net_device *dev, struct net *net, const char /* Get out if there is nothing todo */ err = 0; - if (dev_net(dev) == net) + if (net_eq(dev_net(dev), net)) goto out; /* Pick the destination device name, and ensure diff --git a/net/core/neighbour.c b/net/core/neighbour.c index de654ea8a944..857915a12c15 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -388,7 +388,7 @@ struct neighbour *neigh_lookup_nodev(struct neigh_table *tbl, struct net *net, hash_val = tbl->hash(pkey, NULL); for (n = tbl->hash_buckets[hash_val & tbl->hash_mask]; n; n = n->next) { if (!memcmp(n->primary_key, pkey, key_len) && - dev_net(n->dev) == net) { + net_eq(dev_net(n->dev), net)) { neigh_hold(n); NEIGH_CACHE_STAT_INC(tbl, hits); break; @@ -483,7 +483,7 @@ struct pneigh_entry * pneigh_lookup(struct neigh_table *tbl, for (n = tbl->phash_buckets[hash_val]; n; n = n->next) { if (!memcmp(n->key, pkey, key_len) && - (pneigh_net(n) == net) && + net_eq(pneigh_net(n), net) && (n->dev == dev || !n->dev)) { read_unlock_bh(&tbl->lock); goto out; @@ -542,7 +542,7 @@ int pneigh_delete(struct neigh_table *tbl, struct net *net, const void *pkey, for (np = &tbl->phash_buckets[hash_val]; (n = *np) != NULL; np = &n->next) { if (!memcmp(n->key, pkey, key_len) && n->dev == dev && - (pneigh_net(n) == net)) { + net_eq(pneigh_net(n), net)) { *np = n->next; write_unlock_bh(&tbl->lock); if (tbl->pdestructor) @@ -1286,7 +1286,7 @@ static inline struct neigh_parms *lookup_neigh_params(struct neigh_table *tbl, struct neigh_parms *p; for (p = &tbl->parms; p; p = p->next) { - if ((p->dev && p->dev->ifindex == ifindex && neigh_parms_net(p) == net) || + if ((p->dev && p->dev->ifindex == ifindex && net_eq(neigh_parms_net(p), net)) || (!p->dev && !ifindex)) return p; } @@ -1964,7 +1964,7 @@ static int neightbl_dump_info(struct sk_buff *skb, struct netlink_callback *cb) break; for (nidx = 0, p = tbl->parms.next; p; p = p->next) { - if (net != neigh_parms_net(p)) + if (!net_eq(neigh_parms_net(p), net)) continue; if (nidx++ < neigh_skip) @@ -2161,7 +2161,7 @@ static struct neighbour *neigh_get_first(struct seq_file *seq) n = tbl->hash_buckets[bucket]; while (n) { - if (dev_net(n->dev) != net) + if (!net_eq(dev_net(n->dev), net)) goto next; if (state->neigh_sub_iter) { loff_t fakep = 0; @@ -2204,7 +2204,7 @@ static struct neighbour *neigh_get_next(struct seq_file *seq, while (1) { while (n) { - if (dev_net(n->dev) != net) + if (!net_eq(dev_net(n->dev), net)) goto next; if (state->neigh_sub_iter) { void *v = state->neigh_sub_iter(state, n, pos); @@ -2260,7 +2260,7 @@ static struct pneigh_entry *pneigh_get_first(struct seq_file *seq) state->flags |= NEIGH_SEQ_IS_PNEIGH; for (bucket = 0; bucket <= PNEIGH_HASHMASK; bucket++) { pn = tbl->phash_buckets[bucket]; - while (pn && (pneigh_net(pn) != net)) + while (pn && !net_eq(pneigh_net(pn), net)) pn = pn->next; if (pn) break; @@ -2283,7 +2283,7 @@ static struct pneigh_entry *pneigh_get_next(struct seq_file *seq, if (++state->bucket > PNEIGH_HASHMASK) break; pn = tbl->phash_buckets[state->bucket]; - while (pn && (pneigh_net(pn) != net)) + while (pn && !net_eq(pneigh_net(pn), net)) pn = pn->next; if (pn) break; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 1064111e5b96..1b6ff513c75d 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -139,7 +139,7 @@ static struct sock *inet_lookup_listener_slow(struct net *net, sk_for_each(sk, node, head) { const struct inet_sock *inet = inet_sk(sk); - if (sock_net(sk) == net && inet->num == hnum && + if (net_eq(sock_net(sk), net) && inet->num == hnum && !ipv6_only_sock(sk)) { const __be32 rcv_saddr = inet->rcv_saddr; int score = sk->sk_family == PF_INET ? 1 : 0; @@ -182,7 +182,7 @@ struct sock *__inet_lookup_listener(struct net *net, if (inet->num == hnum && !sk->sk_node.next && (!inet->rcv_saddr || inet->rcv_saddr == daddr) && (sk->sk_family == PF_INET || !ipv6_only_sock(sk)) && - !sk->sk_bound_dev_if && sock_net(sk) == net) + !sk->sk_bound_dev_if && net_eq(sock_net(sk), net)) goto sherry_cache; sk = inet_lookup_listener_slow(net, head, daddr, hnum, dif); } diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 25dc8b38cac3..d965f0a39c84 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -117,7 +117,7 @@ static struct sock *__raw_v4_lookup(struct net *net, struct sock *sk, sk_for_each_from(sk, node) { struct inet_sock *inet = inet_sk(sk); - if (sock_net(sk) == net && inet->num == num && + if (net_eq(sock_net(sk), net) && inet->num == num && !(inet->daddr && inet->daddr != raddr) && !(inet->rcv_saddr && inet->rcv_saddr != laddr) && !(sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif)) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index eab8d75e5222..230716c2dfe0 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1196,7 +1196,7 @@ void ip_rt_redirect(__be32 old_gw, __be32 daddr, __be32 new_gw, rth->fl.oif != ikeys[k] || rth->fl.iif != 0 || rth->rt_genid != atomic_read(&rt_genid) || - dev_net(rth->u.dst.dev) != net) { + !net_eq(dev_net(rth->u.dst.dev), net)) { rthp = &rth->u.dst.rt_next; continue; } @@ -1455,7 +1455,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph, rth->rt_src == iph->saddr && rth->fl.iif == 0 && !(dst_metric_locked(&rth->u.dst, RTAX_MTU)) && - dev_net(rth->u.dst.dev) == net && + net_eq(dev_net(rth->u.dst.dev), net) && rth->rt_genid == atomic_read(&rt_genid)) { unsigned short mtu = new_mtu; @@ -2085,7 +2085,7 @@ int ip_route_input(struct sk_buff *skb, __be32 daddr, __be32 saddr, rth->fl.oif == 0 && rth->fl.mark == skb->mark && rth->fl.fl4_tos == tos && - dev_net(rth->u.dst.dev) == net && + net_eq(dev_net(rth->u.dst.dev), net) && rth->rt_genid == atomic_read(&rt_genid)) { dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(in_hit); @@ -2487,7 +2487,7 @@ int __ip_route_output_key(struct net *net, struct rtable **rp, rth->fl.mark == flp->mark && !((rth->fl.fl4_tos ^ flp->fl4_tos) & (IPTOS_RT_MASK | RTO_ONLINK)) && - dev_net(rth->u.dst.dev) == net && + net_eq(dev_net(rth->u.dst.dev), net) && rth->rt_genid == atomic_read(&rt_genid)) { dst_use(&rth->u.dst, jiffies); RT_CACHE_STAT_INC(out_hit); @@ -2796,7 +2796,7 @@ int ip_rt_dump(struct sk_buff *skb, struct netlink_callback *cb) rcu_read_lock_bh(); for (rt = rcu_dereference(rt_hash_table[h].chain), idx = 0; rt; rt = rcu_dereference(rt->u.dst.rt_next), idx++) { - if (dev_net(rt->u.dst.dev) != net || idx < s_idx) + if (!net_eq(dev_net(rt->u.dst.dev), net) || idx < s_idx) continue; if (rt->rt_genid != atomic_read(&rt_genid)) continue; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 46847e600a46..2a5881c81778 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1974,7 +1974,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur) while (1) { while (req) { if (req->rsk_ops->family == st->family && - sock_net(req->sk) == net) { + net_eq(sock_net(req->sk), net)) { cur = req; goto out; } @@ -1998,7 +1998,7 @@ get_req: } get_sk: sk_for_each_from(sk, node) { - if (sk->sk_family == st->family && sock_net(sk) == net) { + if (sk->sk_family == st->family && net_eq(sock_net(sk), net)) { cur = sk; goto out; } @@ -2049,7 +2049,7 @@ static void *established_get_first(struct seq_file *seq) read_lock_bh(lock); sk_for_each(sk, node, &tcp_hashinfo.ehash[st->bucket].chain) { if (sk->sk_family != st->family || - sock_net(sk) != net) { + !net_eq(sock_net(sk), net)) { continue; } rc = sk; @@ -2059,7 +2059,7 @@ static void *established_get_first(struct seq_file *seq) inet_twsk_for_each(tw, node, &tcp_hashinfo.ehash[st->bucket].twchain) { if (tw->tw_family != st->family || - twsk_net(tw) != net) { + !net_eq(twsk_net(tw), net)) { continue; } rc = tw; @@ -2086,7 +2086,7 @@ static void *established_get_next(struct seq_file *seq, void *cur) tw = cur; tw = tw_next(tw); get_tw: - while (tw && (tw->tw_family != st->family || twsk_net(tw) != net)) { + while (tw && (tw->tw_family != st->family || !net_eq(twsk_net(tw), net))) { tw = tw_next(tw); } if (tw) { @@ -2107,7 +2107,7 @@ get_tw: sk = sk_next(sk); sk_for_each_from(sk, node) { - if (sk->sk_family == st->family && sock_net(sk) == net) + if (sk->sk_family == st->family && net_eq(sock_net(sk), net)) goto found; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 76d52d37d6ac..80007c79f12f 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -137,7 +137,7 @@ static inline int __udp_lib_lport_inuse(struct net *net, __u16 num, struct hlist_node *node; sk_for_each(sk, node, &udptable[num & (UDP_HTABLE_SIZE - 1)]) - if (sock_net(sk) == net && sk->sk_hash == num) + if (net_eq(sock_net(sk), net) && sk->sk_hash == num) return 1; return 0; } @@ -218,7 +218,7 @@ gotit: sk_for_each(sk2, node, head) if (sk2->sk_hash == snum && sk2 != sk && - sock_net(sk2) == net && + net_eq(sock_net(sk2), net) && (!sk2->sk_reuse || !sk->sk_reuse) && (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if || sk2->sk_bound_dev_if == sk->sk_bound_dev_if) && @@ -269,7 +269,7 @@ static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr, sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); - if (sock_net(sk) == net && sk->sk_hash == hnum && + if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum && !ipv6_only_sock(sk)) { int score = (sk->sk_family == PF_INET ? 1 : 0); if (inet->rcv_saddr) { @@ -1511,7 +1511,7 @@ static struct sock *udp_get_first(struct seq_file *seq) for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) { struct hlist_node *node; sk_for_each(sk, node, state->hashtable + state->bucket) { - if (sock_net(sk) != net) + if (!net_eq(sock_net(sk), net)) continue; if (sk->sk_family == state->family) goto found; @@ -1531,7 +1531,7 @@ static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk) sk = sk_next(sk); try_again: ; - } while (sk && (sock_net(sk) != net || sk->sk_family != state->family)); + } while (sk && (!net_eq(sock_net(sk), net) || sk->sk_family != state->family)); if (!sk && ++state->bucket < UDP_HTABLE_SIZE) { sk = sk_head(state->hashtable + state->bucket); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index ac5d4f4b6312..5ab9973571ef 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1217,7 +1217,7 @@ int ipv6_chk_addr(struct net *net, struct in6_addr *addr, read_lock_bh(&addrconf_hash_lock); for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) { - if (dev_net(ifp->idev->dev) != net) + if (!net_eq(dev_net(ifp->idev->dev), net)) continue; if (ipv6_addr_equal(&ifp->addr, addr) && !(ifp->flags&IFA_F_TENTATIVE)) { @@ -1239,7 +1239,7 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr, u8 hash = ipv6_addr_hash(addr); for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) { - if (dev_net(ifp->idev->dev) != net) + if (!net_eq(dev_net(ifp->idev->dev), net)) continue; if (ipv6_addr_equal(&ifp->addr, addr)) { if (dev == NULL || ifp->idev->dev == dev) @@ -1257,7 +1257,7 @@ struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr, read_lock_bh(&addrconf_hash_lock); for(ifp = inet6_addr_lst[hash]; ifp; ifp=ifp->lst_next) { - if (dev_net(ifp->idev->dev) != net) + if (!net_eq(dev_net(ifp->idev->dev), net)) continue; if (ipv6_addr_equal(&ifp->addr, addr)) { if (dev == NULL || ifp->idev->dev == dev || @@ -2771,7 +2771,7 @@ static struct inet6_ifaddr *if6_get_first(struct seq_file *seq) for (state->bucket = 0; state->bucket < IN6_ADDR_HSIZE; ++state->bucket) { ifa = inet6_addr_lst[state->bucket]; - while (ifa && dev_net(ifa->idev->dev) != net) + while (ifa && !net_eq(dev_net(ifa->idev->dev), net)) ifa = ifa->lst_next; if (ifa) break; @@ -2787,7 +2787,7 @@ static struct inet6_ifaddr *if6_get_next(struct seq_file *seq, struct inet6_ifad ifa = ifa->lst_next; try_again: if (ifa) { - if (dev_net(ifa->idev->dev) != net) { + if (!net_eq(dev_net(ifa->idev->dev), net)) { ifa = ifa->lst_next; goto try_again; } @@ -2905,7 +2905,7 @@ int ipv6_chk_home_addr(struct net *net, struct in6_addr *addr) u8 hash = ipv6_addr_hash(addr); read_lock_bh(&addrconf_hash_lock); for (ifp = inet6_addr_lst[hash]; ifp; ifp = ifp->lst_next) { - if (dev_net(ifp->idev->dev) != net) + if (!net_eq(dev_net(ifp->idev->dev), net)) continue; if (ipv6_addr_cmp(&ifp->addr, addr) == 0 && (ifp->flags & IFA_F_HOMEADDRESS)) { diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 21c467675412..340c7d42b83a 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -105,7 +105,7 @@ struct sock *inet6_lookup_listener(struct net *net, read_lock(&hashinfo->lhash_lock); sk_for_each(sk, node, &hashinfo->listening_hash[inet_lhashfn(hnum)]) { - if (sock_net(sk) == net && inet_sk(sk)->num == hnum && + if (net_eq(sock_net(sk), net) && inet_sk(sk)->num == hnum && sk->sk_family == PF_INET6) { const struct ipv6_pinfo *np = inet6_sk(sk); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 12c7a1560977..830da4603697 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -76,7 +76,7 @@ static struct sock *__raw_v6_lookup(struct net *net, struct sock *sk, if (inet_sk(sk)->num == num) { struct ipv6_pinfo *np = inet6_sk(sk); - if (sock_net(sk) != net) + if (!net_eq(sock_net(sk), net)) continue; if (!ipv6_addr_any(&np->daddr) && diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index db266ff297e5..aacbc82ecf0f 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -70,7 +70,7 @@ static struct sock *__udp6_lib_lookup(struct net *net, sk_for_each(sk, node, &udptable[hnum & (UDP_HTABLE_SIZE - 1)]) { struct inet_sock *inet = inet_sk(sk); - if (sock_net(sk) == net && sk->sk_hash == hnum && + if (net_eq(sock_net(sk), net) && sk->sk_hash == hnum && sk->sk_family == PF_INET6) { struct ipv6_pinfo *np = inet6_sk(sk); int score = 0; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 1d16d95dfaaf..36f75d873898 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -228,7 +228,7 @@ static inline struct sock *netlink_lookup(struct net *net, int protocol, read_lock(&nl_table_lock); head = nl_pid_hashfn(hash, pid); sk_for_each(sk, node, head) { - if (sock_net(sk) == net && (nlk_sk(sk)->pid == pid)) { + if (net_eq(sock_net(sk), net) && (nlk_sk(sk)->pid == pid)) { sock_hold(sk); goto found; } @@ -348,7 +348,7 @@ static int netlink_insert(struct sock *sk, struct net *net, u32 pid) head = nl_pid_hashfn(hash, pid); len = 0; sk_for_each(osk, node, head) { - if (sock_net(osk) == net && (nlk_sk(osk)->pid == pid)) + if (net_eq(sock_net(osk), net) && (nlk_sk(osk)->pid == pid)) break; len++; } @@ -532,7 +532,7 @@ retry: netlink_table_grab(); head = nl_pid_hashfn(hash, pid); sk_for_each(osk, node, head) { - if (sock_net(osk) != net) + if (!net_eq(sock_net(osk), net)) continue; if (nlk_sk(osk)->pid == pid) { /* Bind collision, search negative pid values. */ @@ -962,7 +962,7 @@ static inline int do_one_broadcast(struct sock *sk, !test_bit(p->group - 1, nlk->groups)) goto out; - if (sock_net(sk) != p->net) + if (!net_eq(sock_net(sk), p->net)) goto out; if (p->failure) { diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 4a4793051bcb..50bbf6bb1a22 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -252,7 +252,7 @@ static struct sock *__unix_find_socket_byname(struct net *net, sk_for_each(s, node, &unix_socket_table[hash ^ type]) { struct unix_sock *u = unix_sk(s); - if (sock_net(s) != net) + if (!net_eq(sock_net(s), net)) continue; if (u->addr->len == len && @@ -289,7 +289,7 @@ static struct sock *unix_find_socket_byinode(struct net *net, struct inode *i) &unix_socket_table[i->i_ino & (UNIX_HASH_SIZE - 1)]) { struct dentry *dentry = unix_sk(s)->dentry; - if (sock_net(s) != net) + if (!net_eq(sock_net(s), net)) continue; if(dentry && dentry->d_inode == i) -- cgit v1.2.3 From 5d2cdcd4e85c5187db30a6b29f79fbbe59f39f78 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 20 Mar 2008 15:06:41 +0200 Subject: mac80211: get a TKIP phase key from skb This patch makes mac80211 able to compute a TKIP key from an skb. The requested key can be a phase 1 or a phase 2 key. This is useful for drivers who need to provide tkip key to their HW to enable HW encryption. Signed-off-by: Emmanuel Grumbach Signed-off-by: Tomas Winkler Signed-off-by: John W. Linville --- include/net/mac80211.h | 30 ++++++++++++++++++++++++++++ net/mac80211/tkip.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 83 insertions(+) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 5ab6a350ee6d..2a134582fc16 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -643,6 +643,21 @@ enum sta_notify_cmd { STA_NOTIFY_ADD, STA_NOTIFY_REMOVE }; +/** + * enum ieee80211_tkip_key_type - get tkip key + * + * Used by drivers which need to get a tkip key for skb. Some drivers need a + * phase 1 key, others need a phase 2 key. A single function allows the driver + * to get the key, this enum indicates what type of key is required. + * + * @IEEE80211_TKIP_P1_KEY: the driver needs a phase 1 key + * @IEEE80211_TKIP_P2_KEY: the driver needs a phase 2 key + */ +enum ieee80211_tkip_key_type { + IEEE80211_TKIP_P1_KEY, + IEEE80211_TKIP_P2_KEY, +}; + /** * enum ieee80211_hw_flags - hardware flags * @@ -1471,6 +1486,21 @@ int ieee80211_get_hdrlen_from_skb(const struct sk_buff *skb); */ int ieee80211_get_hdrlen(u16 fc); +/** + * ieee80211_get_tkip_key - get a TKIP rc4 for skb + * + * This function computes a TKIP rc4 key for an skb. It computes + * a phase 1 key if needed (iv16 wraps around). This function is to + * be used by drivers which can do HW encryption but need to compute + * to phase 1/2 key in SW. + * + * @keyconf: the parameter passed with the set key + * @skb: the skb for which the key is needed + * @rc4key: a buffer to which the key will be written + */ +void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf, + struct sk_buff *skb, + enum ieee80211_tkip_key_type type, u8 *key); /** * ieee80211_wake_queue - wake specific queue * @hw: pointer as obtained from ieee80211_alloc_hw(). diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c index 3abe194e4d55..5c36b2df3faf 100644 --- a/net/mac80211/tkip.c +++ b/net/mac80211/tkip.c @@ -214,6 +214,59 @@ void ieee80211_tkip_gen_rc4key(struct ieee80211_key *key, u8 *ta, key->u.tkip.iv16, rc4key); } +void ieee80211_get_tkip_key(struct ieee80211_key_conf *keyconf, + struct sk_buff *skb, enum ieee80211_tkip_key_type type, + u8 *outkey) +{ + struct ieee80211_key *key = (struct ieee80211_key *) + container_of(keyconf, struct ieee80211_key, conf); + struct ieee80211_hdr *hdr = (struct ieee80211_hdr *) skb->data; + u8 *data = (u8 *) hdr; + u16 fc = le16_to_cpu(hdr->frame_control); + int hdr_len = ieee80211_get_hdrlen(fc); + u8 *ta = hdr->addr2; + u16 iv16; + u32 iv32; + + iv16 = data[hdr_len] << 8; + iv16 += data[hdr_len + 2]; + iv32 = data[hdr_len + 4] + + (data[hdr_len + 5] >> 8) + + (data[hdr_len + 6] >> 16) + + (data[hdr_len + 7] >> 24); + +#ifdef CONFIG_TKIP_DEBUG + printk(KERN_DEBUG "TKIP encrypt: iv16 = 0x%04x, iv32 = 0x%08x\n", + iv16, iv32); + + if (iv32 != key->u.tkip.iv32) { + printk(KERN_DEBUG "skb: iv32 = 0x%08x key: iv32 = 0x%08x\n", + iv32, key->u.tkip.iv32); + printk(KERN_DEBUG "Wrap around of iv16 in the middle of a " + "fragmented packet\n"); + } +#endif /* CONFIG_TKIP_DEBUG */ + + /* Update the p1k only when the iv16 in the packet wraps around, this + * might occur after the wrap around of iv16 in the key in case of + * fragmented packets. */ + if (iv16 == 0 || !key->u.tkip.tx_initialized) { + /* IV16 wrapped around - perform TKIP phase 1 */ + tkip_mixing_phase1(ta, &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY], + iv32, key->u.tkip.p1k); + key->u.tkip.tx_initialized = 1; + } + + if (type == IEEE80211_TKIP_P1_KEY) { + memcpy(outkey, key->u.tkip.p1k, sizeof(u16) * 5); + return; + } + + tkip_mixing_phase2(key->u.tkip.p1k, + &key->conf.key[ALG_TKIP_TEMP_ENCR_KEY], iv16, outkey); +} +EXPORT_SYMBOL(ieee80211_get_tkip_key); + /* Encrypt packet payload with TKIP using @key. @pos is a pointer to the * beginning of the buffer containing payload. This payload must include * headroom of eight octets for IV and Ext. IV and taildroom of four octets -- cgit v1.2.3 From 9ae4fda332df616ef47d5bb710c39681641d4303 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Thu, 20 Mar 2008 15:06:42 +0200 Subject: mac80211: allows driver to request a Phase 1 RX key This patch makes mac80211 able to send a phase1 key for TKIP decryption. This is needed for drivers that don't do the rekeying by themselves (i.e. iwlwifi). Upon IV16 wrap around, the packet is decrypted in SW, if decryption is ok, mac80211 calls to update_tkip_key with a new phase 1 RX key. Signed-off-by: Emmanuel Grumbach Signed-off-by: Tomas Winkler Signed-off-by: John W. Linville --- include/net/mac80211.h | 17 +++++++++++++++++ net/mac80211/tkip.c | 15 ++++++++++++++- net/mac80211/tkip.h | 2 +- net/mac80211/wpa.c | 2 +- 4 files changed, 33 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 2a134582fc16..48428a6b9109 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -827,6 +827,16 @@ static inline void SET_IEEE80211_PERM_ADDR(struct ieee80211_hw *hw, u8 *addr) * parameter is guaranteed to be valid until another call to set_key() * removes it, but it can only be used as a cookie to differentiate * keys. + * + * In TKIP some HW need to be provided a phase 1 key, for RX decryption + * acceleration (i.e. iwlwifi). Those drivers should provide update_tkip_key + * handler. + * The update_tkip_key() call updates the driver with the new phase 1 key. + * This happens everytime the iv16 wraps around (every 65536 packets). The + * set_key() call will happen only once for each key (unless the AP did + * rekeying), it will not include a valid phase 1 key. The valid phase 1 key is + * provided by udpate_tkip_key only. The trigger that makes mac80211 call this + * handler is software decryption with wrap around of iv16. */ /** @@ -1003,6 +1013,10 @@ enum ieee80211_ampdu_mlme_action { * and remove_interface calls, i.e. while the interface with the * given local_address is enabled. * + * @update_tkip_key: See the section "Hardware crypto acceleration" + * This callback will be called in the context of Rx. Called for drivers + * which set IEEE80211_KEY_FLAG_TKIP_REQ_RX_P1_KEY. + * * @hw_scan: Ask the hardware to service the scan request, no need to start * the scan state machine in stack. The scan must honour the channel * configuration done by the regulatory agent in the wiphy's registered @@ -1094,6 +1108,9 @@ struct ieee80211_ops { int (*set_key)(struct ieee80211_hw *hw, enum set_key_cmd cmd, const u8 *local_address, const u8 *address, struct ieee80211_key_conf *key); + void (*update_tkip_key)(struct ieee80211_hw *hw, + struct ieee80211_key_conf *conf, const u8 *address, + u32 iv32, u16 *phase1key); int (*hw_scan)(struct ieee80211_hw *hw, u8 *ssid, size_t len); int (*get_stats)(struct ieee80211_hw *hw, struct ieee80211_low_level_stats *stats); diff --git a/net/mac80211/tkip.c b/net/mac80211/tkip.c index 5c36b2df3faf..45d59f19c29f 100644 --- a/net/mac80211/tkip.c +++ b/net/mac80211/tkip.c @@ -291,7 +291,7 @@ void ieee80211_tkip_encrypt_data(struct crypto_blkcipher *tfm, int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm, struct ieee80211_key *key, u8 *payload, size_t payload_len, u8 *ta, - int only_iv, int queue, + u8 *ra, int only_iv, int queue, u32 *out_iv32, u16 *out_iv16) { u32 iv32; @@ -368,6 +368,19 @@ int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm, printk("\n"); } #endif /* CONFIG_TKIP_DEBUG */ + if (key->local->ops->update_tkip_key && + key->flags & KEY_FLAG_UPLOADED_TO_HARDWARE) { + u8 bcast[ETH_ALEN] = + {0xff, 0xff, 0xff, 0xff, 0xff, 0xff}; + u8 *sta_addr = key->sta->addr; + + if (is_multicast_ether_addr(ra)) + sta_addr = bcast; + + key->local->ops->update_tkip_key( + local_to_hw(key->local), &key->conf, + sta_addr, iv32, key->u.tkip.p1k_rx[queue]); + } } tkip_mixing_phase2(key->u.tkip.p1k_rx[queue], diff --git a/net/mac80211/tkip.h b/net/mac80211/tkip.h index 73d8ef2a93b0..ffaee3253e19 100644 --- a/net/mac80211/tkip.h +++ b/net/mac80211/tkip.h @@ -31,7 +31,7 @@ enum { int ieee80211_tkip_decrypt_data(struct crypto_blkcipher *tfm, struct ieee80211_key *key, u8 *payload, size_t payload_len, u8 *ta, - int only_iv, int queue, + u8 *ra, int only_iv, int queue, u32 *out_iv32, u16 *out_iv16); #endif /* TKIP_H */ diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index df0b7341efc8..45709ada8fee 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -312,7 +312,7 @@ ieee80211_crypto_tkip_decrypt(struct ieee80211_rx_data *rx) res = ieee80211_tkip_decrypt_data(rx->local->wep_rx_tfm, key, skb->data + hdrlen, skb->len - hdrlen, rx->sta->addr, - hwaccel, rx->queue, + hdr->addr1, hwaccel, rx->queue, &rx->tkip_iv32, &rx->tkip_iv16); if (res != TKIP_DECRYPT_OK || wpa_test) { -- cgit v1.2.3 From 906c730a2db950b7bce4ef17d65399acd791c360 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sun, 16 Mar 2008 18:34:33 +0100 Subject: wireless: add wiphy channel freq to channel struct lookup helper Add ieee80211_get_channel() which gets you a channel struct for a specific wiphy if that channel is present in that wiphy. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/wireless.h | 6 ++++++ net/wireless/util.c | 23 +++++++++++++++++++++++ 2 files changed, 29 insertions(+) (limited to 'include') diff --git a/include/net/wireless.h b/include/net/wireless.h index c7f805ee5545..f4b77ab66bae 100644 --- a/include/net/wireless.h +++ b/include/net/wireless.h @@ -304,4 +304,10 @@ extern int ieee80211_channel_to_frequency(int chan); */ extern int ieee80211_frequency_to_channel(int freq); +/** + * ieee80211_get_channel - get channel struct from wiphy for specified frequency + */ +extern struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy, + int freq); + #endif /* __NET_WIRELESS_H */ diff --git a/net/wireless/util.c b/net/wireless/util.c index 77336c22fcf2..f3e623df3515 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -33,6 +33,29 @@ int ieee80211_frequency_to_channel(int freq) } EXPORT_SYMBOL(ieee80211_frequency_to_channel); +struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy, + int freq) +{ + enum ieee80211_band band; + struct ieee80211_supported_band *sband; + int i; + + for (band = 0; band < IEEE80211_NUM_BANDS; band++) { + sband = wiphy->bands[band]; + + if (!sband) + continue; + + for (i = 0; i < sband->n_channels; i++) { + if (sband->channels[i].center_freq == freq) + return &sband->channels[i]; + } + } + + return NULL; +} +EXPORT_SYMBOL(ieee80211_get_channel); + static void set_mandatory_flags_band(struct ieee80211_supported_band *sband, enum ieee80211_band band) { -- cgit v1.2.3 From ef27559b70bd5312dfcbeab3b9ab0296206413c4 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:07:38 -0700 Subject: [NETFILTER]: nf_conntrack: fix NF_CT_TUPLE_DUMP for IPv4 NF_CT_TUPLE_DUMP prints IPv4 addresses as IPv6, fix this and use printk (guarded by #ifdef DEBUG) directly instead of pr_debug since the tuple is usually printed at the end of line and we don't want to include a log-level. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_tuple.h | 38 ++++++++++++++++++++++++++---- 1 file changed, 33 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index e69ab2e87597..d9b53dd601ad 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -113,11 +113,39 @@ struct nf_conntrack_tuple_mask #ifdef __KERNEL__ -#define NF_CT_DUMP_TUPLE(tp) \ -pr_debug("tuple %p: %u %u " NIP6_FMT " %hu -> " NIP6_FMT " %hu\n", \ - (tp), (tp)->src.l3num, (tp)->dst.protonum, \ - NIP6(*(struct in6_addr *)(tp)->src.u3.all), ntohs((tp)->src.u.all), \ - NIP6(*(struct in6_addr *)(tp)->dst.u3.all), ntohs((tp)->dst.u.all)) +static inline void nf_ct_dump_tuple_ip(const struct nf_conntrack_tuple *t) +{ +#ifdef DEBUG + printk("tuple %p: %u " NIPQUAD_FMT ":%hu -> " NIPQUAD_FMT ":%hu\n", + t, t->dst.protonum, + NIPQUAD(t->src.u3.ip), ntohs(t->src.u.all), + NIPQUAD(t->dst.u3.ip), ntohs(t->dst.u.all)); +#endif +} + +static inline void nf_ct_dump_tuple_ipv6(const struct nf_conntrack_tuple *t) +{ +#ifdef DEBUG + printk("tuple %p: %u " NIP6_FMT " %hu -> " NIP6_FMT " %hu\n", + t, t->dst.protonum, + NIP6(*(struct in6_addr *)t->src.u3.all), ntohs(t->src.u.all), + NIP6(*(struct in6_addr *)t->dst.u3.all), ntohs(t->dst.u.all)); +#endif +} + +static inline void nf_ct_dump_tuple(const struct nf_conntrack_tuple *t) +{ + switch (t->src.l3num) { + case AF_INET: + nf_ct_dump_tuple_ip(t); + break; + case AF_INET6: + nf_ct_dump_tuple_ipv6(t); + break; + } +} + +#define NF_CT_DUMP_TUPLE(tp) nf_ct_dump_tuple(tp) /* If we're the first tuple, it's the original dir. */ #define NF_CT_DIRECTION(h) \ -- cgit v1.2.3 From 1d9d752259ab2e16b99f84b7e7bd4a30c884c289 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:07:58 -0700 Subject: [NETFILTER]: nf_conntrack_expect: constify nf_ct_expect_init arguments Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_expect.h | 6 +++--- net/netfilter/nf_conntrack_expect.c | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index cb608a1b44e5..f1bdcb4f3f2a 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -75,9 +75,9 @@ void nf_ct_unexpect_related(struct nf_conntrack_expect *exp); nf_ct_expect_related. You will have to call put afterwards. */ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me); void nf_ct_expect_init(struct nf_conntrack_expect *, int, - union nf_inet_addr *, - union nf_inet_addr *, - u_int8_t, __be16 *, __be16 *); + const union nf_inet_addr *, + const union nf_inet_addr *, + u_int8_t, const __be16 *, const __be16 *); void nf_ct_expect_put(struct nf_conntrack_expect *exp); int nf_ct_expect_related(struct nf_conntrack_expect *expect); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 684ec9c1ad38..740d94d30091 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -229,9 +229,9 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me) EXPORT_SYMBOL_GPL(nf_ct_expect_alloc); void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family, - union nf_inet_addr *saddr, - union nf_inet_addr *daddr, - u_int8_t proto, __be16 *src, __be16 *dst) + const union nf_inet_addr *saddr, + const union nf_inet_addr *daddr, + u_int8_t proto, const __be16 *src, const __be16 *dst) { int len; -- cgit v1.2.3 From 359b9ab614aba71c2c3bc047efbd6d12dd4a2b9e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:08:37 -0700 Subject: [NETFILTER]: nf_conntrack_expect: support inactive expectations This is useful for the SIP helper and signalling expectations. We don't want to create a full-blown expectation with a wildcard as source based on a single UDP packet, but need to know the final port anyways. With inactive expectations we can register the expectation and reserve the tuple, but wait for confirmation from the registrar before activating it. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack_expect.h | 3 ++- net/netfilter/nf_conntrack_expect.c | 25 +++++++++++++++++++++---- 2 files changed, 23 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index f1bdcb4f3f2a..47c28dd07896 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -53,7 +53,8 @@ struct nf_conntrack_expect struct rcu_head rcu; }; -#define NF_CT_EXPECT_PERMANENT 0x1 +#define NF_CT_EXPECT_PERMANENT 0x1 +#define NF_CT_EXPECT_INACTIVE 0x2 int nf_conntrack_expect_init(void); void nf_conntrack_expect_fini(void); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 4c05a588116f..882602f1c0ef 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -126,9 +126,21 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_find_get); struct nf_conntrack_expect * nf_ct_find_expectation(const struct nf_conntrack_tuple *tuple) { - struct nf_conntrack_expect *exp; + struct nf_conntrack_expect *i, *exp = NULL; + struct hlist_node *n; + unsigned int h; + + if (!nf_ct_expect_count) + return NULL; - exp = __nf_ct_expect_find(tuple); + h = nf_ct_expect_dst_hash(tuple); + hlist_for_each_entry(i, n, &nf_ct_expect_hash[h], hnode) { + if (!(i->flags & NF_CT_EXPECT_INACTIVE) && + nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask)) { + exp = i; + break; + } + } if (!exp) return NULL; @@ -460,6 +472,7 @@ static int exp_seq_show(struct seq_file *s, void *v) { struct nf_conntrack_expect *expect; struct hlist_node *n = v; + char *delim = ""; expect = hlist_entry(n, struct nf_conntrack_expect, hnode); @@ -476,8 +489,12 @@ static int exp_seq_show(struct seq_file *s, void *v) __nf_ct_l4proto_find(expect->tuple.src.l3num, expect->tuple.dst.protonum)); - if (expect->flags & NF_CT_EXPECT_PERMANENT) - seq_printf(s, "PERMANENT "); + if (expect->flags & NF_CT_EXPECT_PERMANENT) { + seq_printf(s, "PERMANENT"); + delim = ","; + } + if (expect->flags & NF_CT_EXPECT_INACTIVE) + seq_printf(s, "%sINACTIVE", delim); return seq_putc(s, '\n'); } -- cgit v1.2.3 From 6002f266b3e7f0acc2d5158cddbed41730b02e82 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:09:15 -0700 Subject: [NETFILTER]: nf_conntrack: introduce expectation classes and policies Introduce expectation classes and policies. An expectation class is used to distinguish different types of expectations by the same helper (for example audio/video/t.120). The expectation policy is used to hold the maximum number of expectations and the initial timeout for each class. The individual classes are isolated from each other, which means that for example an audio expectation will only evict other audio expectations. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/netfilter/nf_conntrack.h | 5 ++- include/net/netfilter/nf_conntrack_expect.h | 13 +++++- include/net/netfilter/nf_conntrack_helper.h | 5 +-- net/ipv4/netfilter/nf_nat_snmp_basic.c | 12 ++++-- net/netfilter/nf_conntrack_amanda.c | 14 +++--- net/netfilter/nf_conntrack_expect.c | 50 ++++++++++++++-------- net/netfilter/nf_conntrack_ftp.c | 10 +++-- net/netfilter/nf_conntrack_h323_main.c | 66 ++++++++++++++++++----------- net/netfilter/nf_conntrack_helper.c | 3 +- net/netfilter/nf_conntrack_irc.c | 10 +++-- net/netfilter/nf_conntrack_netbios_ns.c | 9 +++- net/netfilter/nf_conntrack_pptp.c | 14 ++++-- net/netfilter/nf_conntrack_sane.c | 11 +++-- net/netfilter/nf_conntrack_sip.c | 10 +++-- net/netfilter/nf_conntrack_tftp.c | 11 +++-- 15 files changed, 166 insertions(+), 77 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 90b3e7f5df5f..922877133598 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -75,6 +75,9 @@ do { \ struct nf_conntrack_helper; +/* Must be kept in sync with the classes defined by helpers */ +#define NF_CT_MAX_EXPECT_CLASSES 1 + /* nf_conn feature for connections that have a helper */ struct nf_conn_help { /* Helper. if any */ @@ -85,7 +88,7 @@ struct nf_conn_help { struct hlist_head expectations; /* Current number of expected connections */ - unsigned int expecting; + u8 expecting[NF_CT_MAX_EXPECT_CLASSES]; }; diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 47c28dd07896..dfdf4b459475 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -41,6 +41,9 @@ struct nf_conntrack_expect /* Flags */ unsigned int flags; + /* Expectation class */ + unsigned int class; + #ifdef CONFIG_NF_NAT_NEEDED __be32 saved_ip; /* This is the original per-proto part, used to map the @@ -53,6 +56,14 @@ struct nf_conntrack_expect struct rcu_head rcu; }; +struct nf_conntrack_expect_policy +{ + unsigned int max_expected; + unsigned int timeout; +}; + +#define NF_CT_EXPECT_CLASS_DEFAULT 0 + #define NF_CT_EXPECT_PERMANENT 0x1 #define NF_CT_EXPECT_INACTIVE 0x2 @@ -75,7 +86,7 @@ void nf_ct_unexpect_related(struct nf_conntrack_expect *exp); /* Allocate space for an expectation: this is mandatory before calling nf_ct_expect_related. You will have to call put afterwards. */ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me); -void nf_ct_expect_init(struct nf_conntrack_expect *, int, +void nf_ct_expect_init(struct nf_conntrack_expect *, unsigned int, int, const union nf_inet_addr *, const union nf_inet_addr *, u_int8_t, const __be16 *, const __be16 *); diff --git a/include/net/netfilter/nf_conntrack_helper.h b/include/net/netfilter/nf_conntrack_helper.h index 4ca125e9b3ce..f8060ab5a083 100644 --- a/include/net/netfilter/nf_conntrack_helper.h +++ b/include/net/netfilter/nf_conntrack_helper.h @@ -20,9 +20,7 @@ struct nf_conntrack_helper const char *name; /* name of the module */ struct module *me; /* pointer to self */ - unsigned int max_expected; /* Maximum number of concurrent - * expected connections */ - unsigned int timeout; /* timeout for expecteds */ + const struct nf_conntrack_expect_policy *expect_policy; /* Tuple of things we will help (compared against server response) */ struct nf_conntrack_tuple tuple; @@ -37,6 +35,7 @@ struct nf_conntrack_helper void (*destroy)(struct nf_conn *ct); int (*to_nlattr)(struct sk_buff *skb, const struct nf_conn *ct); + unsigned int expect_class_max; }; extern struct nf_conntrack_helper * diff --git a/net/ipv4/netfilter/nf_nat_snmp_basic.c b/net/ipv4/netfilter/nf_nat_snmp_basic.c index 540ce6ae887c..000e080bac5c 100644 --- a/net/ipv4/netfilter/nf_nat_snmp_basic.c +++ b/net/ipv4/netfilter/nf_nat_snmp_basic.c @@ -50,6 +50,7 @@ #include #include +#include #include #include @@ -1267,11 +1268,15 @@ static int help(struct sk_buff *skb, unsigned int protoff, return ret; } +static const struct nf_conntrack_expect_policy snmp_exp_policy = { + .max_expected = 0, + .timeout = 180, +}; + static struct nf_conntrack_helper snmp_helper __read_mostly = { - .max_expected = 0, - .timeout = 180, .me = THIS_MODULE, .help = help, + .expect_policy = &snmp_exp_policy, .name = "snmp", .tuple.src.l3num = AF_INET, .tuple.src.u.udp.port = __constant_htons(SNMP_PORT), @@ -1279,10 +1284,9 @@ static struct nf_conntrack_helper snmp_helper __read_mostly = { }; static struct nf_conntrack_helper snmp_trap_helper __read_mostly = { - .max_expected = 0, - .timeout = 180, .me = THIS_MODULE, .help = help, + .expect_policy = &snmp_exp_policy, .name = "snmp_trap", .tuple.src.l3num = AF_INET, .tuple.src.u.udp.port = __constant_htons(SNMP_TRAP_PORT), diff --git a/net/netfilter/nf_conntrack_amanda.c b/net/netfilter/nf_conntrack_amanda.c index 7b8239c0cd5e..d14585a19b7d 100644 --- a/net/netfilter/nf_conntrack_amanda.c +++ b/net/netfilter/nf_conntrack_amanda.c @@ -148,7 +148,8 @@ static int amanda_help(struct sk_buff *skb, goto out; } tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - nf_ct_expect_init(exp, family, &tuple->src.u3, &tuple->dst.u3, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + &tuple->src.u3, &tuple->dst.u3, IPPROTO_TCP, NULL, &port); nf_nat_amanda = rcu_dereference(nf_nat_amanda_hook); @@ -164,26 +165,29 @@ out: return ret; } +static const struct nf_conntrack_expect_policy amanda_exp_policy = { + .max_expected = 3, + .timeout = 180, +}; + static struct nf_conntrack_helper amanda_helper[2] __read_mostly = { { .name = "amanda", - .max_expected = 3, - .timeout = 180, .me = THIS_MODULE, .help = amanda_help, .tuple.src.l3num = AF_INET, .tuple.src.u.udp.port = __constant_htons(10080), .tuple.dst.protonum = IPPROTO_UDP, + .expect_policy = &amanda_exp_policy, }, { .name = "amanda", - .max_expected = 3, - .timeout = 180, .me = THIS_MODULE, .help = amanda_help, .tuple.src.l3num = AF_INET6, .tuple.src.u.udp.port = __constant_htons(10080), .tuple.dst.protonum = IPPROTO_UDP, + .expect_policy = &amanda_exp_policy, }, }; diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 882602f1c0ef..e31beeb33b2b 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -54,7 +54,7 @@ void nf_ct_unlink_expect(struct nf_conntrack_expect *exp) nf_ct_expect_count--; hlist_del(&exp->lnode); - master_help->expecting--; + master_help->expecting[exp->class]--; nf_ct_expect_put(exp); NF_CT_STAT_INC(expect_delete); @@ -171,7 +171,7 @@ void nf_ct_remove_expectations(struct nf_conn *ct) struct hlist_node *n, *next; /* Optimization: most connection never expect any others. */ - if (!help || help->expecting == 0) + if (!help) return; hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) { @@ -205,7 +205,7 @@ static inline int expect_clash(const struct nf_conntrack_expect *a, static inline int expect_matches(const struct nf_conntrack_expect *a, const struct nf_conntrack_expect *b) { - return a->master == b->master + return a->master == b->master && a->class == b->class && nf_ct_tuple_equal(&a->tuple, &b->tuple) && nf_ct_tuple_mask_equal(&a->mask, &b->mask); } @@ -240,7 +240,8 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me) } EXPORT_SYMBOL_GPL(nf_ct_expect_alloc); -void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family, +void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, + int family, const union nf_inet_addr *saddr, const union nf_inet_addr *daddr, u_int8_t proto, const __be16 *src, const __be16 *dst) @@ -253,6 +254,7 @@ void nf_ct_expect_init(struct nf_conntrack_expect *exp, int family, len = 16; exp->flags = 0; + exp->class = class; exp->expectfn = NULL; exp->helper = NULL; exp->tuple.src.l3num = family; @@ -309,19 +311,21 @@ EXPORT_SYMBOL_GPL(nf_ct_expect_put); static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) { struct nf_conn_help *master_help = nfct_help(exp->master); + const struct nf_conntrack_expect_policy *p; unsigned int h = nf_ct_expect_dst_hash(&exp->tuple); atomic_inc(&exp->use); hlist_add_head(&exp->lnode, &master_help->expectations); - master_help->expecting++; + master_help->expecting[exp->class]++; hlist_add_head_rcu(&exp->hnode, &nf_ct_expect_hash[h]); nf_ct_expect_count++; setup_timer(&exp->timeout, nf_ct_expectation_timed_out, (unsigned long)exp); - exp->timeout.expires = jiffies + master_help->helper->timeout * HZ; + p = &master_help->helper->expect_policy[exp->class]; + exp->timeout.expires = jiffies + p->timeout * HZ; add_timer(&exp->timeout); atomic_inc(&exp->use); @@ -329,35 +333,41 @@ static void nf_ct_expect_insert(struct nf_conntrack_expect *exp) } /* Race with expectations being used means we could have none to find; OK. */ -static void evict_oldest_expect(struct nf_conn *master) +static void evict_oldest_expect(struct nf_conn *master, + struct nf_conntrack_expect *new) { struct nf_conn_help *master_help = nfct_help(master); - struct nf_conntrack_expect *exp = NULL; + struct nf_conntrack_expect *exp, *last = NULL; struct hlist_node *n; - hlist_for_each_entry(exp, n, &master_help->expectations, lnode) - ; /* nothing */ + hlist_for_each_entry(exp, n, &master_help->expectations, lnode) { + if (exp->class == new->class) + last = exp; + } - if (exp && del_timer(&exp->timeout)) { - nf_ct_unlink_expect(exp); - nf_ct_expect_put(exp); + if (last && del_timer(&last->timeout)) { + nf_ct_unlink_expect(last); + nf_ct_expect_put(last); } } static inline int refresh_timer(struct nf_conntrack_expect *i) { struct nf_conn_help *master_help = nfct_help(i->master); + const struct nf_conntrack_expect_policy *p; if (!del_timer(&i->timeout)) return 0; - i->timeout.expires = jiffies + master_help->helper->timeout*HZ; + p = &master_help->helper->expect_policy[i->class]; + i->timeout.expires = jiffies + p->timeout * HZ; add_timer(&i->timeout); return 1; } int nf_ct_expect_related(struct nf_conntrack_expect *expect) { + const struct nf_conntrack_expect_policy *p; struct nf_conntrack_expect *i; struct nf_conn *master = expect->master; struct nf_conn_help *master_help = nfct_help(master); @@ -386,9 +396,15 @@ int nf_ct_expect_related(struct nf_conntrack_expect *expect) } } /* Will be over limit? */ - if (master_help->helper->max_expected && - master_help->expecting >= master_help->helper->max_expected) - evict_oldest_expect(master); + p = &master_help->helper->expect_policy[expect->class]; + if (p->max_expected && + master_help->expecting[expect->class] >= p->max_expected) { + evict_oldest_expect(master, expect); + if (master_help->expecting[expect->class] >= p->max_expected) { + ret = -EMFILE; + goto out; + } + } if (nf_ct_expect_count >= nf_ct_expect_max) { if (net_ratelimit()) diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index 6770baf2e845..7eff876bb8bc 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -483,7 +483,7 @@ static int help(struct sk_buff *skb, daddr = &cmd.u3; } - nf_ct_expect_init(exp, cmd.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, cmd.l3num, &ct->tuplehash[!dir].tuple.src.u3, daddr, IPPROTO_TCP, NULL, &cmd.u.tcp.port); @@ -517,6 +517,11 @@ out_update_nl: static struct nf_conntrack_helper ftp[MAX_PORTS][2] __read_mostly; static char ftp_names[MAX_PORTS][2][sizeof("ftp-65535")] __read_mostly; +static const struct nf_conntrack_expect_policy ftp_exp_policy = { + .max_expected = 1, + .timeout = 5 * 60, +}; + /* don't make this __exit, since it's called from __init ! */ static void nf_conntrack_ftp_fini(void) { @@ -556,8 +561,7 @@ static int __init nf_conntrack_ftp_init(void) for (j = 0; j < 2; j++) { ftp[i][j].tuple.src.u.tcp.port = htons(ports[i]); ftp[i][j].tuple.dst.protonum = IPPROTO_TCP; - ftp[i][j].max_expected = 1; - ftp[i][j].timeout = 5 * 60; /* 5 Minutes */ + ftp[i][j].expect_policy = &ftp_exp_policy; ftp[i][j].me = THIS_MODULE; ftp[i][j].help = help; tmpname = &ftp_names[i][j][0]; diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index 898f1922b5b8..505052d495cf 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -277,7 +277,8 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, /* Create expect for RTP */ if ((rtp_exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(rtp_exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(rtp_exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_UDP, NULL, &rtp_port); @@ -287,7 +288,8 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, nf_ct_expect_put(rtp_exp); return -1; } - nf_ct_expect_init(rtcp_exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(rtcp_exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_UDP, NULL, &rtcp_port); @@ -344,7 +346,8 @@ static int expect_t120(struct sk_buff *skb, /* Create expect for T.120 connections */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); @@ -612,13 +615,17 @@ static int h245_help(struct sk_buff *skb, unsigned int protoff, } /****************************************************************************/ +static const struct nf_conntrack_expect_policy h245_exp_policy = { + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 2 /* T.120 */, + .timeout = 240, +}; + static struct nf_conntrack_helper nf_conntrack_helper_h245 __read_mostly = { .name = "H.245", .me = THIS_MODULE, - .max_expected = H323_RTP_CHANNEL_MAX * 4 + 2 /* T.120 */, - .timeout = 240, .tuple.dst.protonum = IPPROTO_UDP, - .help = h245_help + .help = h245_help, + .expect_policy = &h245_exp_policy, }; /****************************************************************************/ @@ -676,7 +683,8 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct, /* Create expect for h245 connection */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); @@ -792,7 +800,8 @@ static int expect_callforwarding(struct sk_buff *skb, /* Create expect for the second call leg */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->helper = nf_conntrack_helper_q931; @@ -1156,28 +1165,30 @@ static int q931_help(struct sk_buff *skb, unsigned int protoff, } /****************************************************************************/ +static const struct nf_conntrack_expect_policy q931_exp_policy = { + /* T.120 and H.245 */ + .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, + .timeout = 240, +}; + static struct nf_conntrack_helper nf_conntrack_helper_q931[] __read_mostly = { { .name = "Q.931", .me = THIS_MODULE, - /* T.120 and H.245 */ - .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, - .timeout = 240, .tuple.src.l3num = AF_INET, .tuple.src.u.tcp.port = __constant_htons(Q931_PORT), .tuple.dst.protonum = IPPROTO_TCP, - .help = q931_help + .help = q931_help, + .expect_policy = &q931_exp_policy, }, { .name = "Q.931", .me = THIS_MODULE, - /* T.120 and H.245 */ - .max_expected = H323_RTP_CHANNEL_MAX * 4 + 4, - .timeout = 240, .tuple.src.l3num = AF_INET6, .tuple.src.u.tcp.port = __constant_htons(Q931_PORT), .tuple.dst.protonum = IPPROTO_TCP, - .help = q931_help + .help = q931_help, + .expect_policy = &q931_exp_policy, }, }; @@ -1261,7 +1272,8 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct, /* Create expect for Q.931 */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, gkrouted_only ? /* only accept calls from GK? */ &ct->tuplehash[!dir].tuple.src.u3 : NULL, &ct->tuplehash[!dir].tuple.dst.u3, @@ -1332,7 +1344,8 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct, /* Need new expect */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_UDP, NULL, &port); exp->helper = nf_conntrack_helper_ras; @@ -1536,7 +1549,8 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct, /* Need new expect */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; @@ -1589,7 +1603,8 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct, /* Need new expect for call signal */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[!dir].tuple.src.l3num, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; @@ -1728,26 +1743,29 @@ static int ras_help(struct sk_buff *skb, unsigned int protoff, } /****************************************************************************/ +static const struct nf_conntrack_expect_policy ras_exp_policy = { + .max_expected = 32, + .timeout = 240, +}; + static struct nf_conntrack_helper nf_conntrack_helper_ras[] __read_mostly = { { .name = "RAS", .me = THIS_MODULE, - .max_expected = 32, - .timeout = 240, .tuple.src.l3num = AF_INET, .tuple.src.u.udp.port = __constant_htons(RAS_PORT), .tuple.dst.protonum = IPPROTO_UDP, .help = ras_help, + .expect_policy = &ras_exp_policy, }, { .name = "RAS", .me = THIS_MODULE, - .max_expected = 32, - .timeout = 240, .tuple.src.l3num = AF_INET6, .tuple.src.u.udp.port = __constant_htons(RAS_PORT), .tuple.dst.protonum = IPPROTO_UDP, .help = ras_help, + .expect_policy = &ras_exp_policy, }, }; diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index b1fd21cc1dbc..e350f56d43c9 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -110,7 +110,8 @@ int nf_conntrack_helper_register(struct nf_conntrack_helper *me) { unsigned int h = helper_hash(&me->tuple); - BUG_ON(me->timeout == 0); + BUG_ON(me->expect_policy == NULL); + BUG_ON(me->expect_class_max >= NF_CT_MAX_EXPECT_CLASSES); mutex_lock(&nf_ct_helper_mutex); hlist_add_head_rcu(&me->hnode, &nf_ct_helper_hash[h]); diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c index c336b07a0d4c..02f21cbe5ae7 100644 --- a/net/netfilter/nf_conntrack_irc.c +++ b/net/netfilter/nf_conntrack_irc.c @@ -187,7 +187,8 @@ static int help(struct sk_buff *skb, unsigned int protoff, } tuple = &ct->tuplehash[!dir].tuple; port = htons(dcc_port); - nf_ct_expect_init(exp, tuple->src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + tuple->src.l3num, NULL, &tuple->dst.u3, IPPROTO_TCP, NULL, &port); @@ -210,6 +211,7 @@ static int help(struct sk_buff *skb, unsigned int protoff, static struct nf_conntrack_helper irc[MAX_PORTS] __read_mostly; static char irc_names[MAX_PORTS][sizeof("irc-65535")] __read_mostly; +static struct nf_conntrack_expect_policy irc_exp_policy; static void nf_conntrack_irc_fini(void); @@ -223,6 +225,9 @@ static int __init nf_conntrack_irc_init(void) return -EINVAL; } + irc_exp_policy.max_expected = max_dcc_channels; + irc_exp_policy.timeout = dcc_timeout; + irc_buffer = kmalloc(65536, GFP_KERNEL); if (!irc_buffer) return -ENOMEM; @@ -235,8 +240,7 @@ static int __init nf_conntrack_irc_init(void) irc[i].tuple.src.l3num = AF_INET; irc[i].tuple.src.u.tcp.port = htons(ports[i]); irc[i].tuple.dst.protonum = IPPROTO_TCP; - irc[i].max_expected = max_dcc_channels; - irc[i].timeout = dcc_timeout; + irc[i].expect_policy = &irc_exp_policy; irc[i].me = THIS_MODULE; irc[i].help = help; diff --git a/net/netfilter/nf_conntrack_netbios_ns.c b/net/netfilter/nf_conntrack_netbios_ns.c index 60dedaded84e..08404e6755fb 100644 --- a/net/netfilter/nf_conntrack_netbios_ns.c +++ b/net/netfilter/nf_conntrack_netbios_ns.c @@ -86,6 +86,7 @@ static int help(struct sk_buff *skb, unsigned int protoff, exp->expectfn = NULL; exp->flags = NF_CT_EXPECT_PERMANENT; + exp->class = NF_CT_EXPECT_CLASS_DEFAULT; exp->helper = NULL; nf_ct_expect_related(exp); @@ -96,19 +97,23 @@ out: return NF_ACCEPT; } +static struct nf_conntrack_expect_policy exp_policy = { + .max_expected = 1, +}; + static struct nf_conntrack_helper helper __read_mostly = { .name = "netbios-ns", .tuple.src.l3num = AF_INET, .tuple.src.u.udp.port = __constant_htons(NMBD_PORT), .tuple.dst.protonum = IPPROTO_UDP, - .max_expected = 1, .me = THIS_MODULE, .help = help, + .expect_policy = &exp_policy, }; static int __init nf_conntrack_netbios_ns_init(void) { - helper.timeout = timeout; + exp_policy.timeout = timeout; return nf_conntrack_helper_register(&helper); } diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index b5cb8e831230..8fd83470d1b3 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -208,7 +208,8 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) /* original direction, PNS->PAC */ dir = IP_CT_DIR_ORIGINAL; - nf_ct_expect_init(exp_orig, ct->tuplehash[dir].tuple.src.l3num, + nf_ct_expect_init(exp_orig, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[dir].tuple.src.l3num, &ct->tuplehash[dir].tuple.src.u3, &ct->tuplehash[dir].tuple.dst.u3, IPPROTO_GRE, &peer_callid, &callid); @@ -216,7 +217,8 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) /* reply direction, PAC->PNS */ dir = IP_CT_DIR_REPLY; - nf_ct_expect_init(exp_reply, ct->tuplehash[dir].tuple.src.l3num, + nf_ct_expect_init(exp_reply, NF_CT_EXPECT_CLASS_DEFAULT, + ct->tuplehash[dir].tuple.src.l3num, &ct->tuplehash[dir].tuple.src.u3, &ct->tuplehash[dir].tuple.dst.u3, IPPROTO_GRE, &callid, &peer_callid); @@ -575,17 +577,21 @@ conntrack_pptp_help(struct sk_buff *skb, unsigned int protoff, return ret; } +static const struct nf_conntrack_expect_policy pptp_exp_policy = { + .max_expected = 2, + .timeout = 5 * 60, +}; + /* control protocol helper */ static struct nf_conntrack_helper pptp __read_mostly = { .name = "pptp", .me = THIS_MODULE, - .max_expected = 2, - .timeout = 5 * 60, .tuple.src.l3num = AF_INET, .tuple.src.u.tcp.port = __constant_htons(PPTP_CONTROL_PORT), .tuple.dst.protonum = IPPROTO_TCP, .help = conntrack_pptp_help, .destroy = pptp_destroy_siblings, + .expect_policy = &pptp_exp_policy, }; static int __init nf_conntrack_pptp_init(void) diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c index a70051d741a7..7542e25eede3 100644 --- a/net/netfilter/nf_conntrack_sane.c +++ b/net/netfilter/nf_conntrack_sane.c @@ -143,7 +143,8 @@ static int help(struct sk_buff *skb, } tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - nf_ct_expect_init(exp, family, &tuple->src.u3, &tuple->dst.u3, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + &tuple->src.u3, &tuple->dst.u3, IPPROTO_TCP, NULL, &reply->port); pr_debug("nf_ct_sane: expect: "); @@ -163,6 +164,11 @@ out: static struct nf_conntrack_helper sane[MAX_PORTS][2] __read_mostly; static char sane_names[MAX_PORTS][2][sizeof("sane-65535")] __read_mostly; +static const struct nf_conntrack_expect_policy sane_exp_policy = { + .max_expected = 1, + .timeout = 5 * 60, +}; + /* don't make this __exit, since it's called from __init ! */ static void nf_conntrack_sane_fini(void) { @@ -200,8 +206,7 @@ static int __init nf_conntrack_sane_init(void) for (j = 0; j < 2; j++) { sane[i][j].tuple.src.u.tcp.port = htons(ports[i]); sane[i][j].tuple.dst.protonum = IPPROTO_TCP; - sane[i][j].max_expected = 1; - sane[i][j].timeout = 5 * 60; /* 5 Minutes */ + sane[i][j].expect_policy = &sane_exp_policy; sane[i][j].me = THIS_MODULE; sane[i][j].help = help; tmpname = &sane_names[i][j][0]; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index c521c891d351..0021d5b60cec 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -380,7 +380,7 @@ static int set_expected_rtp(struct sk_buff *skb, exp = nf_ct_expect_alloc(ct); if (exp == NULL) return NF_DROP; - nf_ct_expect_init(exp, family, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, &ct->tuplehash[!dir].tuple.src.u3, addr, IPPROTO_UDP, NULL, &port); @@ -476,6 +476,11 @@ out: static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly; static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly; +static const struct nf_conntrack_expect_policy sip_exp_policy = { + .max_expected = 2, + .timeout = 3 * 60, +}; + static void nf_conntrack_sip_fini(void) { int i, j; @@ -505,8 +510,7 @@ static int __init nf_conntrack_sip_init(void) for (j = 0; j < 2; j++) { sip[i][j].tuple.dst.protonum = IPPROTO_UDP; sip[i][j].tuple.src.u.udp.port = htons(ports[i]); - sip[i][j].max_expected = 2; - sip[i][j].timeout = 3 * 60; /* 3 minutes */ + sip[i][j].expect_policy = &sip_exp_policy; sip[i][j].me = THIS_MODULE; sip[i][j].help = sip_help; diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c index bd2e800f23cc..a28341b30f21 100644 --- a/net/netfilter/nf_conntrack_tftp.c +++ b/net/netfilter/nf_conntrack_tftp.c @@ -63,7 +63,8 @@ static int tftp_help(struct sk_buff *skb, if (exp == NULL) return NF_DROP; tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; - nf_ct_expect_init(exp, family, &tuple->src.u3, &tuple->dst.u3, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + &tuple->src.u3, &tuple->dst.u3, IPPROTO_UDP, NULL, &tuple->dst.u.udp.port); pr_debug("expect: "); @@ -92,6 +93,11 @@ static int tftp_help(struct sk_buff *skb, static struct nf_conntrack_helper tftp[MAX_PORTS][2] __read_mostly; static char tftp_names[MAX_PORTS][2][sizeof("tftp-65535")] __read_mostly; +static const struct nf_conntrack_expect_policy tftp_exp_policy = { + .max_expected = 1, + .timeout = 5 * 60, +}; + static void nf_conntrack_tftp_fini(void) { int i, j; @@ -118,8 +124,7 @@ static int __init nf_conntrack_tftp_init(void) for (j = 0; j < 2; j++) { tftp[i][j].tuple.dst.protonum = IPPROTO_UDP; tftp[i][j].tuple.src.u.udp.port = htons(ports[i]); - tftp[i][j].max_expected = 1; - tftp[i][j].timeout = 5 * 60; /* 5 minutes */ + tftp[i][j].expect_policy = &tftp_exp_policy; tftp[i][j].me = THIS_MODULE; tftp[i][j].help = tftp_help; -- cgit v1.2.3 From b8beedd25d3913d45b8330a08ab88fdf90eb54b8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:09:33 -0700 Subject: [NETFILTER]: Add nf_inet_addr_cmp() Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter.h | 9 +++++++++ include/net/netfilter/nf_conntrack_tuple.h | 15 +++------------ 2 files changed, 12 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index f0680c2bee73..89e6c72ad295 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -61,6 +61,15 @@ union nf_inet_addr { #ifdef __KERNEL__ #ifdef CONFIG_NETFILTER +static inline int nf_inet_addr_cmp(const union nf_inet_addr *a1, + const union nf_inet_addr *a2) +{ + return a1->all[0] == a2->all[0] && + a1->all[1] == a2->all[1] && + a1->all[2] == a2->all[2] && + a1->all[3] == a2->all[3]; +} + extern void netfilter_init(void); /* Largest hook number + 1 */ diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index d9b53dd601ad..168c91754d89 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -163,10 +163,7 @@ struct nf_conntrack_tuple_hash static inline int __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1, const struct nf_conntrack_tuple *t2) { - return (t1->src.u3.all[0] == t2->src.u3.all[0] && - t1->src.u3.all[1] == t2->src.u3.all[1] && - t1->src.u3.all[2] == t2->src.u3.all[2] && - t1->src.u3.all[3] == t2->src.u3.all[3] && + return (nf_inet_addr_cmp(&t1->src.u3, &t2->src.u3) && t1->src.u.all == t2->src.u.all && t1->src.l3num == t2->src.l3num); } @@ -174,10 +171,7 @@ static inline int __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1, static inline int __nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1, const struct nf_conntrack_tuple *t2) { - return (t1->dst.u3.all[0] == t2->dst.u3.all[0] && - t1->dst.u3.all[1] == t2->dst.u3.all[1] && - t1->dst.u3.all[2] == t2->dst.u3.all[2] && - t1->dst.u3.all[3] == t2->dst.u3.all[3] && + return (nf_inet_addr_cmp(&t1->dst.u3, &t2->dst.u3) && t1->dst.u.all == t2->dst.u.all && t1->dst.protonum == t2->dst.protonum); } @@ -192,10 +186,7 @@ static inline int nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1, static inline int nf_ct_tuple_mask_equal(const struct nf_conntrack_tuple_mask *m1, const struct nf_conntrack_tuple_mask *m2) { - return (m1->src.u3.all[0] == m2->src.u3.all[0] && - m1->src.u3.all[1] == m2->src.u3.all[1] && - m1->src.u3.all[2] == m2->src.u3.all[2] && - m1->src.u3.all[3] == m2->src.u3.all[3] && + return (nf_inet_addr_cmp(&m1->src.u3, &m2->src.u3) && m1->src.u.all == m2->src.u.all); } -- cgit v1.2.3 From 2a6cfb22ae002330d445f734668d9158db9e90de Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:16:54 -0700 Subject: [NETFILTER]: nf_conntrack_sip: adjust dptr and datalen after packet mangling After mangling the packet, the pointer to the data and the length of the data portion may change and need to be adjusted. Use double data pointers and a pointer to the length everywhere and add a helper function to the NAT helper for performing the adjustments. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 6 +- net/ipv4/netfilter/nf_nat_sip.c | 91 ++++++++++++++++-------------- net/netfilter/nf_conntrack_sip.c | 14 +++-- 3 files changed, 60 insertions(+), 51 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 8e5ce1ca7bfc..9d0dbfb26300 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -24,11 +24,13 @@ enum sip_header_pos { extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conn *ct, - const char **dptr); + const char **dptr, + unsigned int *datalen); extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conntrack_expect *exp, - const char *dptr); + const char **dptr, + unsigned int *datalen); extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr, size_t dlen, unsigned int *matchoff, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 84d8b4982cdf..e77122e65283 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -60,15 +60,35 @@ static void addr_map_init(const struct nf_conn *ct, struct addr_map *map) } } +static unsigned int mangle_packet(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int matchoff, unsigned int matchlen, + const char *buffer, unsigned int buflen) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + + if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo, matchoff, matchlen, + buffer, buflen)) + return 0; + + /* Reload data pointer and adjust datalen value */ + *dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr); + *datalen += buflen - matchlen; + return 1; +} + static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo, - struct nf_conn *ct, const char **dptr, size_t dlen, + struct nf_conn *ct, + const char **dptr, unsigned int *datalen, enum sip_header_pos pos, struct addr_map *map) { enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); unsigned int matchlen, matchoff, addrlen; char *addr; - if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0) + if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, + pos) <= 0) return 1; if ((matchlen == map->addr[dir].srciplen || @@ -84,26 +104,19 @@ static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo, } else return 1; - if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo, - matchoff, matchlen, addr, addrlen)) - return 0; - *dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr); - return 1; - + return mangle_packet(skb, dptr, datalen, matchoff, matchlen, + addr, addrlen); } static unsigned int ip_nat_sip(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conn *ct, - const char **dptr) + const char **dptr, unsigned int *datalen) { enum sip_header_pos pos; struct addr_map map; - int dataoff, datalen; - dataoff = ip_hdrlen(skb) + sizeof(struct udphdr); - datalen = skb->len - dataoff; - if (datalen < sizeof("SIP/2.0") - 1) + if (*datalen < sizeof("SIP/2.0") - 1) return NF_ACCEPT; addr_map_init(ct, &map); @@ -115,7 +128,7 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, * The "userinfo" and "@" components of the SIP URI MUST NOT * be present. */ - if (datalen >= sizeof("REGISTER") - 1 && + if (*datalen >= sizeof("REGISTER") - 1 && strncmp(*dptr, "REGISTER", sizeof("REGISTER") - 1) == 0) pos = POS_REG_REQ_URI; else @@ -136,51 +149,45 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, static unsigned int mangle_sip_packet(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conn *ct, - const char **dptr, size_t dlen, + const char **dptr, unsigned int *datalen, char *buffer, int bufflen, enum sip_header_pos pos) { unsigned int matchlen, matchoff; - if (ct_sip_get_info(ct, *dptr, dlen, &matchoff, &matchlen, pos) <= 0) - return 0; - - if (!nf_nat_mangle_udp_packet(skb, ct, ctinfo, - matchoff, matchlen, buffer, bufflen)) + if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, + pos) <= 0) return 0; - /* We need to reload this. Thanks Patrick. */ - *dptr = skb->data + ip_hdrlen(skb) + sizeof(struct udphdr); - return 1; + return mangle_packet(skb, dptr, datalen, matchoff, matchlen, + buffer, bufflen); } static int mangle_content_len(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conn *ct, - const char *dptr) + const char **dptr, unsigned int *datalen) { - unsigned int dataoff, matchoff, matchlen; + unsigned int matchoff, matchlen; char buffer[sizeof("65536")]; int bufflen; - dataoff = ip_hdrlen(skb) + sizeof(struct udphdr); - /* Get actual SDP length */ - if (ct_sip_get_info(ct, dptr, skb->len - dataoff, &matchoff, + if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, POS_SDP_HEADER) > 0) { /* since ct_sip_get_info() give us a pointer passing 'v=' we need to add 2 bytes in this count. */ - int c_len = skb->len - dataoff - matchoff + 2; + int c_len = *datalen - matchoff + 2; /* Now, update SDP length */ - if (ct_sip_get_info(ct, dptr, skb->len - dataoff, &matchoff, + if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, POS_CONTENT) > 0) { bufflen = sprintf(buffer, "%u", c_len); - return nf_nat_mangle_udp_packet(skb, ct, ctinfo, - matchoff, matchlen, - buffer, bufflen); + return mangle_packet(skb, dptr, datalen, + matchoff, matchlen, + buffer, bufflen); } } return 0; @@ -190,30 +197,28 @@ static unsigned int mangle_sdp(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conn *ct, __be32 newip, u_int16_t port, - const char *dptr) + const char **dptr, unsigned int *datalen) { char buffer[sizeof("nnn.nnn.nnn.nnn")]; - unsigned int dataoff, bufflen; - - dataoff = ip_hdrlen(skb) + sizeof(struct udphdr); + unsigned int bufflen; /* Mangle owner and contact info. */ bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); - if (!mangle_sip_packet(skb, ctinfo, ct, &dptr, skb->len - dataoff, + if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen, buffer, bufflen, POS_OWNER_IP4)) return 0; - if (!mangle_sip_packet(skb, ctinfo, ct, &dptr, skb->len - dataoff, + if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen, buffer, bufflen, POS_CONNECTION_IP4)) return 0; /* Mangle media port. */ bufflen = sprintf(buffer, "%u", port); - if (!mangle_sip_packet(skb, ctinfo, ct, &dptr, skb->len - dataoff, + if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen, buffer, bufflen, POS_MEDIA)) return 0; - return mangle_content_len(skb, ctinfo, ct, dptr); + return mangle_content_len(skb, ctinfo, ct, dptr, datalen); } static void ip_nat_sdp_expect(struct nf_conn *ct, @@ -242,7 +247,7 @@ static void ip_nat_sdp_expect(struct nf_conn *ct, static unsigned int ip_nat_sdp(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conntrack_expect *exp, - const char *dptr) + const char **dptr, unsigned int *datalen) { struct nf_conn *ct = exp->master; enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); @@ -275,7 +280,7 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb, if (port == 0) return NF_DROP; - if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr)) { + if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr, datalen)) { nf_ct_unexpect_related(exp); return NF_DROP; } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 016e1c1aafe4..fa0d5599ff24 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -39,13 +39,15 @@ MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session"); unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conn *ct, - const char **dptr) __read_mostly; + const char **dptr, + unsigned int *datalen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_hook); unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, enum ip_conntrack_info ctinfo, struct nf_conntrack_expect *exp, - const char *dptr) __read_mostly; + const char **dptr, + unsigned int *datalen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); static int digits_len(const struct nf_conn *, const char *, const char *, int *); @@ -369,7 +371,7 @@ static int set_expected_rtp(struct sk_buff *skb, enum ip_conntrack_info ctinfo, union nf_inet_addr *addr, __be16 port, - const char *dptr) + const char **dptr, unsigned int *datalen) { struct nf_conntrack_expect *exp; enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); @@ -386,7 +388,7 @@ static int set_expected_rtp(struct sk_buff *skb, nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook); if (nf_nat_sdp && ct->status & IPS_NAT_MASK) - ret = nf_nat_sdp(skb, ctinfo, exp, dptr); + ret = nf_nat_sdp(skb, ctinfo, exp, dptr, datalen); else { if (nf_ct_expect_related(exp) != 0) ret = NF_DROP; @@ -429,7 +431,7 @@ static int sip_help(struct sk_buff *skb, nf_nat_sip = rcu_dereference(nf_nat_sip_hook); if (nf_nat_sip && ct->status & IPS_NAT_MASK) { - if (!nf_nat_sip(skb, ctinfo, ct, &dptr)) { + if (!nf_nat_sip(skb, ctinfo, ct, &dptr, &datalen)) { ret = NF_DROP; goto out; } @@ -466,7 +468,7 @@ static int sip_help(struct sk_buff *skb, goto out; } ret = set_expected_rtp(skb, ct, ctinfo, &addr, - htons(port), dptr); + htons(port), &dptr, &datalen); } } out: -- cgit v1.2.3 From 212440a7d04a12ee13787afecc6c86c7fc4e6184 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:17:13 -0700 Subject: [NETFILTER]: nf_conntrack_sip: remove redundant function arguments The conntrack reference and ctinfo can be derived from the packet. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 7 ++--- net/ipv4/netfilter/nf_nat_sip.c | 49 +++++++++++++++--------------- net/netfilter/nf_conntrack_sip.c | 24 ++++++--------- 3 files changed, 37 insertions(+), 43 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 9d0dbfb26300..b94de3d60303 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -22,15 +22,12 @@ enum sip_header_pos { }; extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conn *ct, const char **dptr, unsigned int *datalen); extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conntrack_expect *exp, const char **dptr, - unsigned int *datalen); + unsigned int *datalen, + struct nf_conntrack_expect *exp); extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr, size_t dlen, unsigned int *matchoff, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index e77122e65283..acaa7d4569fa 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -78,11 +78,12 @@ static unsigned int mangle_packet(struct sk_buff *skb, return 1; } -static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo, - struct nf_conn *ct, +static int map_sip_addr(struct sk_buff *skb, const char **dptr, unsigned int *datalen, enum sip_header_pos pos, struct addr_map *map) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); unsigned int matchlen, matchoff, addrlen; char *addr; @@ -109,10 +110,10 @@ static int map_sip_addr(struct sk_buff *skb, enum ip_conntrack_info ctinfo, } static unsigned int ip_nat_sip(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conn *ct, const char **dptr, unsigned int *datalen) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum sip_header_pos pos; struct addr_map map; @@ -134,25 +135,25 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, else pos = POS_REQ_URI; - if (!map_sip_addr(skb, ctinfo, ct, dptr, datalen, pos, &map)) + if (!map_sip_addr(skb, dptr, datalen, pos, &map)) return NF_DROP; } - if (!map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_FROM, &map) || - !map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_TO, &map) || - !map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_VIA, &map) || - !map_sip_addr(skb, ctinfo, ct, dptr, datalen, POS_CONTACT, &map)) + if (!map_sip_addr(skb, dptr, datalen, POS_FROM, &map) || + !map_sip_addr(skb, dptr, datalen, POS_TO, &map) || + !map_sip_addr(skb, dptr, datalen, POS_VIA, &map) || + !map_sip_addr(skb, dptr, datalen, POS_CONTACT, &map)) return NF_DROP; return NF_ACCEPT; } static unsigned int mangle_sip_packet(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conn *ct, const char **dptr, unsigned int *datalen, char *buffer, int bufflen, enum sip_header_pos pos) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchlen, matchoff; if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, @@ -164,10 +165,10 @@ static unsigned int mangle_sip_packet(struct sk_buff *skb, } static int mangle_content_len(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conn *ct, const char **dptr, unsigned int *datalen) { + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchoff, matchlen; char buffer[sizeof("65536")]; int bufflen; @@ -204,21 +205,21 @@ static unsigned int mangle_sdp(struct sk_buff *skb, /* Mangle owner and contact info. */ bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); - if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen, - buffer, bufflen, POS_OWNER_IP4)) + if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen, + POS_OWNER_IP4)) return 0; - if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen, - buffer, bufflen, POS_CONNECTION_IP4)) + if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen, + POS_CONNECTION_IP4)) return 0; /* Mangle media port. */ bufflen = sprintf(buffer, "%u", port); - if (!mangle_sip_packet(skb, ctinfo, ct, dptr, datalen, - buffer, bufflen, POS_MEDIA)) + if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen, + POS_MEDIA)) return 0; - return mangle_content_len(skb, ctinfo, ct, dptr, datalen); + return mangle_content_len(skb, dptr, datalen); } static void ip_nat_sdp_expect(struct nf_conn *ct, @@ -245,11 +246,11 @@ static void ip_nat_sdp_expect(struct nf_conn *ct, /* So, this packet has hit the connection tracking matching code. Mangle it, and change the expectation to match the new version. */ static unsigned int ip_nat_sdp(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conntrack_expect *exp, - const char **dptr, unsigned int *datalen) + const char **dptr, unsigned int *datalen, + struct nf_conntrack_expect *exp) { - struct nf_conn *ct = exp->master; + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); __be32 newip; u_int16_t port; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index fa0d5599ff24..38e1e7a05334 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -37,17 +37,14 @@ module_param(sip_timeout, uint, 0600); MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session"); unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conn *ct, const char **dptr, unsigned int *datalen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_hook); unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conntrack_expect *exp, const char **dptr, - unsigned int *datalen) __read_mostly; + unsigned int *datalen, + struct nf_conntrack_expect *exp) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); static int digits_len(const struct nf_conn *, const char *, const char *, int *); @@ -367,13 +364,12 @@ int ct_sip_get_info(const struct nf_conn *ct, EXPORT_SYMBOL_GPL(ct_sip_get_info); static int set_expected_rtp(struct sk_buff *skb, - struct nf_conn *ct, - enum ip_conntrack_info ctinfo, - union nf_inet_addr *addr, - __be16 port, - const char **dptr, unsigned int *datalen) + const char **dptr, unsigned int *datalen, + union nf_inet_addr *addr, __be16 port) { struct nf_conntrack_expect *exp; + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); int family = ct->tuplehash[!dir].tuple.src.l3num; int ret; @@ -388,7 +384,7 @@ static int set_expected_rtp(struct sk_buff *skb, nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook); if (nf_nat_sdp && ct->status & IPS_NAT_MASK) - ret = nf_nat_sdp(skb, ctinfo, exp, dptr, datalen); + ret = nf_nat_sdp(skb, dptr, datalen, exp); else { if (nf_ct_expect_related(exp) != 0) ret = NF_DROP; @@ -431,7 +427,7 @@ static int sip_help(struct sk_buff *skb, nf_nat_sip = rcu_dereference(nf_nat_sip_hook); if (nf_nat_sip && ct->status & IPS_NAT_MASK) { - if (!nf_nat_sip(skb, ctinfo, ct, &dptr, &datalen)) { + if (!nf_nat_sip(skb, &dptr, &datalen)) { ret = NF_DROP; goto out; } @@ -467,8 +463,8 @@ static int sip_help(struct sk_buff *skb, ret = NF_DROP; goto out; } - ret = set_expected_rtp(skb, ct, ctinfo, &addr, - htons(port), &dptr, &datalen); + ret = set_expected_rtp(skb, &dptr, &datalen, + &addr, htons(port)); } } out: -- cgit v1.2.3 From 3e9b4600b4e71beaa9d943251bfe9c25f6a97b8c Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:17:55 -0700 Subject: [NETFILTER]: nf_conntrack_sip: add seperate SDP header parsing function SDP and SIP headers are quite different, SIP can have continuation lines, leading and trailing whitespace after the colon and is mostly case-insensitive while SDP headers always begin on a new line and are followed by an equal sign and the value, without any whitespace. Introduce new SDP header parsing function and convert all users that used the SIP header parsing function. This will allow to properly deal with the special SIP cases in the SIP header parsing function later. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 49 +++++++-- net/ipv4/netfilter/nf_nat_sip.c | 69 ++++++------- net/netfilter/nf_conntrack_sip.c | 159 +++++++++++++++++------------ 3 files changed, 169 insertions(+), 108 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index b94de3d60303..9131cbc9b9de 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -13,12 +13,42 @@ enum sip_header_pos { POS_VIA, POS_CONTACT, POS_CONTENT, - POS_MEDIA, - POS_OWNER_IP4, - POS_CONNECTION_IP4, - POS_OWNER_IP6, - POS_CONNECTION_IP6, - POS_SDP_HEADER, +}; + +struct sip_header { + const char *name; + const char *cname; + const char *search; + unsigned int len; + unsigned int clen; + unsigned int slen; + int (*match_len)(const struct nf_conn *ct, + const char *dptr, const char *limit, + int *shift); +}; + +#define __SIP_HDR(__name, __cname, __search, __match) \ +{ \ + .name = (__name), \ + .len = sizeof(__name) - 1, \ + .cname = (__cname), \ + .clen = (__cname) ? sizeof(__cname) - 1 : 0, \ + .search = (__search), \ + .slen = (__search) ? sizeof(__search) - 1 : 0, \ + .match_len = (__match), \ +} + +#define SDP_HDR(__name, __search, __match) \ + __SIP_HDR(__name, NULL, __search, __match) + +enum sdp_header_types { + SDP_HDR_UNSPEC, + SDP_HDR_VERSION, + SDP_HDR_OWNER_IP4, + SDP_HDR_CONNECTION_IP4, + SDP_HDR_OWNER_IP6, + SDP_HDR_CONNECTION_IP6, + SDP_HDR_MEDIA, }; extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, @@ -36,5 +66,12 @@ extern int ct_sip_lnlen(const char *line, const char *limit); extern const char *ct_sip_search(const char *needle, const char *haystack, size_t needle_len, size_t haystack_len, int case_sensitive); + +extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sdp_header_types type, + enum sdp_header_types term, + unsigned int *matchoff, unsigned int *matchlen); + #endif /* __KERNEL__ */ #endif /* __NF_CONNTRACK_SIP_H__ */ diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index dd1b2d86deee..aa8a4f492baf 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -147,51 +147,46 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, return NF_ACCEPT; } -static unsigned int mangle_sip_packet(struct sk_buff *skb, - const char **dptr, unsigned int *datalen, - char *buffer, int bufflen, - enum sip_header_pos pos) +static int mangle_content_len(struct sk_buff *skb, + const char **dptr, unsigned int *datalen) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - unsigned int matchlen, matchoff; + unsigned int matchoff, matchlen; + char buffer[sizeof("65536")]; + int buflen, c_len; + /* Get actual SDP length */ + if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, + SDP_HDR_VERSION, SDP_HDR_UNSPEC, + &matchoff, &matchlen) <= 0) + return 0; + c_len = *datalen - matchoff + strlen("v="); + + /* Now, update SDP length */ if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, - pos) <= 0) + POS_CONTENT) <= 0) return 0; + buflen = sprintf(buffer, "%u", c_len); return mangle_packet(skb, dptr, datalen, matchoff, matchlen, - buffer, bufflen); + buffer, buflen); } -static int mangle_content_len(struct sk_buff *skb, - const char **dptr, unsigned int *datalen) +static unsigned mangle_sdp_packet(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + enum sdp_header_types type, + char *buffer, int buflen) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - unsigned int matchoff, matchlen; - char buffer[sizeof("65536")]; - int bufflen; + unsigned int matchlen, matchoff; - /* Get actual SDP length */ - if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, - &matchlen, POS_SDP_HEADER) > 0) { - - /* since ct_sip_get_info() give us a pointer passing 'v=' - we need to add 2 bytes in this count. */ - int c_len = *datalen - matchoff + 2; - - /* Now, update SDP length */ - if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, - &matchlen, POS_CONTENT) > 0) { - - bufflen = sprintf(buffer, "%u", c_len); - return mangle_packet(skb, dptr, datalen, - matchoff, matchlen, - buffer, bufflen); - } - } - return 0; + if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, type, SDP_HDR_UNSPEC, + &matchoff, &matchlen) <= 0) + return 0; + return mangle_packet(skb, dptr, datalen, matchoff, matchlen, + buffer, buflen); } static unsigned int mangle_sdp(struct sk_buff *skb, @@ -205,18 +200,18 @@ static unsigned int mangle_sdp(struct sk_buff *skb, /* Mangle owner and contact info. */ bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); - if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen, - POS_OWNER_IP4)) + if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_OWNER_IP4, + buffer, bufflen)) return 0; - if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen, - POS_CONNECTION_IP4)) + if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_CONNECTION_IP4, + buffer, bufflen)) return 0; /* Mangle media port. */ bufflen = sprintf(buffer, "%u", port); - if (!mangle_sip_packet(skb, dptr, datalen, buffer, bufflen, - POS_MEDIA)) + if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_MEDIA, + buffer, bufflen)) return 0; return mangle_content_len(skb, dptr, datalen); diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index cf19a7082a75..801fcb3c749f 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -124,66 +124,6 @@ static const struct sip_header_nfo ct_sip_hdrs[] = { .ln_strlen = sizeof(":") - 1, .match_len = skp_digits_len }, - [POS_MEDIA] = { /* SDP media info */ - .case_sensitive = 1, - .lname = "\nm=", - .lnlen = sizeof("\nm=") - 1, - .sname = "\rm=", - .snlen = sizeof("\rm=") - 1, - .ln_str = "audio ", - .ln_strlen = sizeof("audio ") - 1, - .match_len = digits_len - }, - [POS_OWNER_IP4] = { /* SDP owner address*/ - .case_sensitive = 1, - .lname = "\no=", - .lnlen = sizeof("\no=") - 1, - .sname = "\ro=", - .snlen = sizeof("\ro=") - 1, - .ln_str = "IN IP4 ", - .ln_strlen = sizeof("IN IP4 ") - 1, - .match_len = epaddr_len - }, - [POS_CONNECTION_IP4] = {/* SDP connection info */ - .case_sensitive = 1, - .lname = "\nc=", - .lnlen = sizeof("\nc=") - 1, - .sname = "\rc=", - .snlen = sizeof("\rc=") - 1, - .ln_str = "IN IP4 ", - .ln_strlen = sizeof("IN IP4 ") - 1, - .match_len = epaddr_len - }, - [POS_OWNER_IP6] = { /* SDP owner address*/ - .case_sensitive = 1, - .lname = "\no=", - .lnlen = sizeof("\no=") - 1, - .sname = "\ro=", - .snlen = sizeof("\ro=") - 1, - .ln_str = "IN IP6 ", - .ln_strlen = sizeof("IN IP6 ") - 1, - .match_len = epaddr_len - }, - [POS_CONNECTION_IP6] = {/* SDP connection info */ - .case_sensitive = 1, - .lname = "\nc=", - .lnlen = sizeof("\nc=") - 1, - .sname = "\rc=", - .snlen = sizeof("\rc=") - 1, - .ln_str = "IN IP6 ", - .ln_strlen = sizeof("IN IP6 ") - 1, - .match_len = epaddr_len - }, - [POS_SDP_HEADER] = { /* SDP version header */ - .case_sensitive = 1, - .lname = "\nv=", - .lnlen = sizeof("\nv=") - 1, - .sname = "\rv=", - .snlen = sizeof("\rv=") - 1, - .ln_str = "=", - .ln_strlen = sizeof("=") - 1, - .match_len = digits_len - } }; /* get line length until first CR or LF seen. */ @@ -363,6 +303,92 @@ int ct_sip_get_info(const struct nf_conn *ct, } EXPORT_SYMBOL_GPL(ct_sip_get_info); +/* SDP header parsing: a SDP session description contains an ordered set of + * headers, starting with a section containing general session parameters, + * optionally followed by multiple media descriptions. + * + * SDP headers always start at the beginning of a line. According to RFC 2327: + * "The sequence CRLF (0x0d0a) is used to end a record, although parsers should + * be tolerant and also accept records terminated with a single newline + * character". We handle both cases. + */ +static const struct sip_header ct_sdp_hdrs[] = { + [SDP_HDR_VERSION] = SDP_HDR("v=", NULL, digits_len), + [SDP_HDR_OWNER_IP4] = SDP_HDR("o=", "IN IP4 ", epaddr_len), + [SDP_HDR_CONNECTION_IP4] = SDP_HDR("c=", "IN IP4 ", epaddr_len), + [SDP_HDR_OWNER_IP6] = SDP_HDR("o=", "IN IP6 ", epaddr_len), + [SDP_HDR_CONNECTION_IP6] = SDP_HDR("c=", "IN IP6 ", epaddr_len), + [SDP_HDR_MEDIA] = SDP_HDR("m=", "audio ", digits_len), +}; + +/* Linear string search within SDP header values */ +static const char *ct_sdp_header_search(const char *dptr, const char *limit, + const char *needle, unsigned int len) +{ + for (limit -= len; dptr < limit; dptr++) { + if (*dptr == '\r' || *dptr == '\n') + break; + if (strncmp(dptr, needle, len) == 0) + return dptr; + } + return NULL; +} + +/* Locate a SDP header (optionally a substring within the header value), + * optionally stopping at the first occurence of the term header, parse + * it and return the offset and length of the data we're interested in. + */ +int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sdp_header_types type, + enum sdp_header_types term, + unsigned int *matchoff, unsigned int *matchlen) +{ + const struct sip_header *hdr = &ct_sdp_hdrs[type]; + const struct sip_header *thdr = &ct_sdp_hdrs[term]; + const char *start = dptr, *limit = dptr + datalen; + int shift = 0; + + for (dptr += dataoff; dptr < limit; dptr++) { + /* Find beginning of line */ + if (*dptr != '\r' && *dptr != '\n') + continue; + if (++dptr >= limit) + break; + if (*(dptr - 1) == '\r' && *dptr == '\n') { + if (++dptr >= limit) + break; + } + + if (term != SDP_HDR_UNSPEC && + limit - dptr >= thdr->len && + strnicmp(dptr, thdr->name, thdr->len) == 0) + break; + else if (limit - dptr >= hdr->len && + strnicmp(dptr, hdr->name, hdr->len) == 0) + dptr += hdr->len; + else + continue; + + *matchoff = dptr - start; + if (hdr->search) { + dptr = ct_sdp_header_search(dptr, limit, hdr->search, + hdr->slen); + if (!dptr) + return -1; + dptr += hdr->slen; + } + + *matchlen = hdr->match_len(ct, dptr, limit, &shift); + if (!*matchlen) + return -1; + *matchoff = dptr - start + shift; + return 1; + } + return 0; +} +EXPORT_SYMBOL_GPL(ct_sip_get_sdp_header); + static int set_expected_rtp(struct sk_buff *skb, const char **dptr, unsigned int *datalen, union nf_inet_addr *addr, __be16 port) @@ -408,7 +434,7 @@ static int sip_help(struct sk_buff *skb, int ret = NF_ACCEPT; unsigned int matchoff, matchlen; u_int16_t port; - enum sip_header_pos pos; + enum sdp_header_types type; typeof(nf_nat_sip_hook) nf_nat_sip; /* No Data ? */ @@ -446,8 +472,10 @@ static int sip_help(struct sk_buff *skb, goto out; } /* Get address and port from SDP packet. */ - pos = family == AF_INET ? POS_CONNECTION_IP4 : POS_CONNECTION_IP6; - if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, pos) > 0) { + type = family == AF_INET ? SDP_HDR_CONNECTION_IP4 : + SDP_HDR_CONNECTION_IP6; + if (ct_sip_get_sdp_header(ct, dptr, 0, datalen, type, SDP_HDR_UNSPEC, + &matchoff, &matchlen) > 0) { /* We'll drop only if there are parse problems. */ if (!parse_addr(ct, dptr + matchoff, NULL, &addr, @@ -455,8 +483,9 @@ static int sip_help(struct sk_buff *skb, ret = NF_DROP; goto out; } - if (ct_sip_get_info(ct, dptr, datalen, &matchoff, &matchlen, - POS_MEDIA) > 0) { + if (ct_sip_get_sdp_header(ct, dptr, 0, datalen, + SDP_HDR_MEDIA, SDP_HDR_UNSPEC, + &matchoff, &matchlen) > 0) { port = simple_strtoul(dptr + matchoff, NULL, 10); if (port < 1024) { -- cgit v1.2.3 From ac3677406d4e36e86b1eb5a453997a3b3e0c089a Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:18:40 -0700 Subject: [NETFILTER]: nf_conntrack_sip: kill request URI "header" definitions The request URI is not a header and needs to be treated differently than real SIP headers. Add a seperate function for parsing it and get rid of the POS_REQ_URI/POS_REG_REQ_URI definitions. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 5 ++- net/ipv4/netfilter/nf_nat_sip.c | 46 +++++++++++---------- net/netfilter/nf_conntrack_sip.c | 64 +++++++++++++++++++++++------- 3 files changed, 77 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 9131cbc9b9de..480b26f40ce4 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -6,8 +6,6 @@ #define SIP_TIMEOUT 3600 enum sip_header_pos { - POS_REG_REQ_URI, - POS_REQ_URI, POS_FROM, POS_TO, POS_VIA, @@ -59,6 +57,9 @@ extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, unsigned int *datalen, struct nf_conntrack_expect *exp); +extern int ct_sip_parse_request(const struct nf_conn *ct, + const char *dptr, unsigned int datalen, + unsigned int *matchoff, unsigned int *matchlen); extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr, size_t dlen, unsigned int *matchoff, unsigned int *matchlen, enum sip_header_pos pos); diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index aa8a4f492baf..60151b5901a5 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -78,20 +78,17 @@ static unsigned int mangle_packet(struct sk_buff *skb, return 1; } -static int map_sip_addr(struct sk_buff *skb, - const char **dptr, unsigned int *datalen, - enum sip_header_pos pos, struct addr_map *map) +static int map_addr(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int matchoff, unsigned int matchlen, + struct addr_map *map) { enum ip_conntrack_info ctinfo; - struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn *ct __maybe_unused = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - unsigned int matchlen, matchoff, addrlen; + unsigned int addrlen; char *addr; - if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, - pos) <= 0) - return 1; - if ((matchlen == map->addr[dir].srciplen || matchlen == map->addr[dir].srclen) && strncmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) { @@ -109,13 +106,27 @@ static int map_sip_addr(struct sk_buff *skb, addr, addrlen); } +static int map_sip_addr(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + enum sip_header_pos pos, struct addr_map *map) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + unsigned int matchlen, matchoff; + + if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, + pos) <= 0) + return 1; + return map_addr(skb, dptr, datalen, matchoff, matchlen, map); +} + static unsigned int ip_nat_sip(struct sk_buff *skb, const char **dptr, unsigned int *datalen) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - enum sip_header_pos pos; struct addr_map map; + unsigned int matchoff, matchlen; if (*datalen < strlen("SIP/2.0")) return NF_ACCEPT; @@ -124,18 +135,9 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, /* Basic rules: requests and responses. */ if (strnicmp(*dptr, "SIP/2.0", strlen("SIP/2.0")) != 0) { - /* 10.2: Constructing the REGISTER Request: - * - * The "userinfo" and "@" components of the SIP URI MUST NOT - * be present. - */ - if (*datalen >= strlen("REGISTER") && - strnicmp(*dptr, "REGISTER", strlen("REGISTER")) == 0) - pos = POS_REG_REQ_URI; - else - pos = POS_REQ_URI; - - if (!map_sip_addr(skb, dptr, datalen, pos, &map)) + if (ct_sip_parse_request(ct, *dptr, *datalen, + &matchoff, &matchlen) > 0 && + !map_addr(skb, dptr, datalen, matchoff, matchlen, &map)) return NF_DROP; } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 801fcb3c749f..bb4396155681 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -65,20 +65,6 @@ struct sip_header_nfo { }; static const struct sip_header_nfo ct_sip_hdrs[] = { - [POS_REG_REQ_URI] = { /* SIP REGISTER request URI */ - .lname = "sip:", - .lnlen = sizeof("sip:") - 1, - .ln_str = ":", - .ln_strlen = sizeof(":") - 1, - .match_len = epaddr_len, - }, - [POS_REQ_URI] = { /* SIP request URI */ - .lname = "sip:", - .lnlen = sizeof("sip:") - 1, - .ln_str = "@", - .ln_strlen = sizeof("@") - 1, - .match_len = epaddr_len, - }, [POS_FROM] = { /* SIP From header */ .lname = "From:", .lnlen = sizeof("From:") - 1, @@ -164,6 +150,18 @@ const char *ct_sip_search(const char *needle, const char *haystack, } EXPORT_SYMBOL_GPL(ct_sip_search); +static int string_len(const struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int len = 0; + + while (dptr < limit && isalpha(*dptr)) { + dptr++; + len++; + } + return len; +} + static int digits_len(const struct nf_conn *ct, const char *dptr, const char *limit, int *shift) { @@ -258,6 +256,44 @@ static int skp_epaddr_len(const struct nf_conn *ct, const char *dptr, return epaddr_len(ct, dptr, limit, shift); } +/* Parse a SIP request line of the form: + * + * Request-Line = Method SP Request-URI SP SIP-Version CRLF + * + * and return the offset and length of the address contained in the Request-URI. + */ +int ct_sip_parse_request(const struct nf_conn *ct, + const char *dptr, unsigned int datalen, + unsigned int *matchoff, unsigned int *matchlen) +{ + const char *start = dptr, *limit = dptr + datalen; + unsigned int mlen; + int shift = 0; + + /* Skip method and following whitespace */ + mlen = string_len(ct, dptr, limit, NULL); + if (!mlen) + return 0; + dptr += mlen; + if (++dptr >= limit) + return 0; + + /* Find SIP URI */ + limit -= strlen("sip:"); + for (; dptr < limit; dptr++) { + if (*dptr == '\r' || *dptr == '\n') + return -1; + if (strnicmp(dptr, "sip:", strlen("sip:")) == 0) + break; + } + *matchlen = skp_epaddr_len(ct, dptr, limit, &shift); + if (!*matchlen) + return 0; + *matchoff = dptr - start + shift; + return 1; +} +EXPORT_SYMBOL_GPL(ct_sip_parse_request); + /* Returns 0 if not found, -1 error parsing. */ int ct_sip_get_info(const struct nf_conn *ct, const char *dptr, size_t dlen, -- cgit v1.2.3 From ea45f12a2766dae54e5426a23e8f4bafdbe2782e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:18:57 -0700 Subject: [NETFILTER]: nf_conntrack_sip: parse SIP headers properly Introduce new function for SIP header parsing that properly deals with continuation lines and whitespace in headers and use it. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 30 ++-- net/ipv4/netfilter/nf_nat_sip.c | 18 +- net/netfilter/nf_conntrack_sip.c | 271 ++++++++++++++--------------- 3 files changed, 151 insertions(+), 168 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 480b26f40ce4..ccc701422963 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -5,14 +5,6 @@ #define SIP_PORT 5060 #define SIP_TIMEOUT 3600 -enum sip_header_pos { - POS_FROM, - POS_TO, - POS_VIA, - POS_CONTACT, - POS_CONTENT, -}; - struct sip_header { const char *name; const char *cname; @@ -36,9 +28,20 @@ struct sip_header { .match_len = (__match), \ } +#define SIP_HDR(__name, __cname, __search, __match) \ + __SIP_HDR(__name, __cname, __search, __match) + #define SDP_HDR(__name, __search, __match) \ __SIP_HDR(__name, NULL, __search, __match) +enum sip_header_types { + SIP_HDR_FROM, + SIP_HDR_TO, + SIP_HDR_CONTACT, + SIP_HDR_VIA, + SIP_HDR_CONTENT_LENGTH, +}; + enum sdp_header_types { SDP_HDR_UNSPEC, SDP_HDR_VERSION, @@ -60,13 +63,10 @@ extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, extern int ct_sip_parse_request(const struct nf_conn *ct, const char *dptr, unsigned int datalen, unsigned int *matchoff, unsigned int *matchlen); -extern int ct_sip_get_info(const struct nf_conn *ct, const char *dptr, - size_t dlen, unsigned int *matchoff, - unsigned int *matchlen, enum sip_header_pos pos); -extern int ct_sip_lnlen(const char *line, const char *limit); -extern const char *ct_sip_search(const char *needle, const char *haystack, - size_t needle_len, size_t haystack_len, - int case_sensitive); +extern int ct_sip_get_header(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sip_header_types type, + unsigned int *matchoff, unsigned int *matchlen); extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 60151b5901a5..c13e43862361 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -108,14 +108,14 @@ static int map_addr(struct sk_buff *skb, static int map_sip_addr(struct sk_buff *skb, const char **dptr, unsigned int *datalen, - enum sip_header_pos pos, struct addr_map *map) + enum sip_header_types type, struct addr_map *map) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchlen, matchoff; - if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, - pos) <= 0) + if (ct_sip_get_header(ct, *dptr, 0, *datalen, type, + &matchoff, &matchlen) <= 0) return 1; return map_addr(skb, dptr, datalen, matchoff, matchlen, map); } @@ -141,10 +141,10 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, return NF_DROP; } - if (!map_sip_addr(skb, dptr, datalen, POS_FROM, &map) || - !map_sip_addr(skb, dptr, datalen, POS_TO, &map) || - !map_sip_addr(skb, dptr, datalen, POS_VIA, &map) || - !map_sip_addr(skb, dptr, datalen, POS_CONTACT, &map)) + if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM, &map) || + !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO, &map) || + !map_sip_addr(skb, dptr, datalen, SIP_HDR_VIA, &map) || + !map_sip_addr(skb, dptr, datalen, SIP_HDR_CONTACT, &map)) return NF_DROP; return NF_ACCEPT; } @@ -166,8 +166,8 @@ static int mangle_content_len(struct sk_buff *skb, c_len = *datalen - matchoff + strlen("v="); /* Now, update SDP length */ - if (ct_sip_get_info(ct, *dptr, *datalen, &matchoff, &matchlen, - POS_CONTENT) <= 0) + if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_CONTENT_LENGTH, + &matchoff, &matchlen) <= 0) return 0; buflen = sprintf(buffer, "%u", c_len); diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index bb4396155681..cbc91598acee 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -47,109 +47,6 @@ unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, struct nf_conntrack_expect *exp) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); -static int digits_len(const struct nf_conn *, const char *, const char *, int *); -static int epaddr_len(const struct nf_conn *, const char *, const char *, int *); -static int skp_digits_len(const struct nf_conn *, const char *, const char *, int *); -static int skp_epaddr_len(const struct nf_conn *, const char *, const char *, int *); - -struct sip_header_nfo { - const char *lname; - const char *sname; - const char *ln_str; - size_t lnlen; - size_t snlen; - size_t ln_strlen; - int case_sensitive; - int (*match_len)(const struct nf_conn *, const char *, - const char *, int *); -}; - -static const struct sip_header_nfo ct_sip_hdrs[] = { - [POS_FROM] = { /* SIP From header */ - .lname = "From:", - .lnlen = sizeof("From:") - 1, - .sname = "\r\nf:", - .snlen = sizeof("\r\nf:") - 1, - .ln_str = "sip:", - .ln_strlen = sizeof("sip:") - 1, - .match_len = skp_epaddr_len, - }, - [POS_TO] = { /* SIP To header */ - .lname = "To:", - .lnlen = sizeof("To:") - 1, - .sname = "\r\nt:", - .snlen = sizeof("\r\nt:") - 1, - .ln_str = "sip:", - .ln_strlen = sizeof("sip:") - 1, - .match_len = skp_epaddr_len - }, - [POS_VIA] = { /* SIP Via header */ - .lname = "Via:", - .lnlen = sizeof("Via:") - 1, - .sname = "\r\nv:", - .snlen = sizeof("\r\nv:") - 1, /* rfc3261 "\r\n" */ - .ln_str = "UDP ", - .ln_strlen = sizeof("UDP ") - 1, - .match_len = epaddr_len, - }, - [POS_CONTACT] = { /* SIP Contact header */ - .lname = "Contact:", - .lnlen = sizeof("Contact:") - 1, - .sname = "\r\nm:", - .snlen = sizeof("\r\nm:") - 1, - .ln_str = "sip:", - .ln_strlen = sizeof("sip:") - 1, - .match_len = skp_epaddr_len - }, - [POS_CONTENT] = { /* SIP Content length header */ - .lname = "Content-Length:", - .lnlen = sizeof("Content-Length:") - 1, - .sname = "\r\nl:", - .snlen = sizeof("\r\nl:") - 1, - .ln_str = ":", - .ln_strlen = sizeof(":") - 1, - .match_len = skp_digits_len - }, -}; - -/* get line length until first CR or LF seen. */ -int ct_sip_lnlen(const char *line, const char *limit) -{ - const char *k = line; - - while ((line < limit) && (*line == '\r' || *line == '\n')) - line++; - - while (line < limit) { - if (*line == '\r' || *line == '\n') - break; - line++; - } - return line - k; -} -EXPORT_SYMBOL_GPL(ct_sip_lnlen); - -/* Linear string search, case sensitive. */ -const char *ct_sip_search(const char *needle, const char *haystack, - size_t needle_len, size_t haystack_len, - int case_sensitive) -{ - const char *limit = haystack + (haystack_len - needle_len); - - while (haystack < limit) { - if (case_sensitive) { - if (strncmp(haystack, needle, needle_len) == 0) - return haystack; - } else { - if (strnicmp(haystack, needle, needle_len) == 0) - return haystack; - } - haystack++; - } - return NULL; -} -EXPORT_SYMBOL_GPL(ct_sip_search); - static int string_len(const struct nf_conn *ct, const char *dptr, const char *limit, int *shift) { @@ -173,16 +70,6 @@ static int digits_len(const struct nf_conn *ct, const char *dptr, return len; } -/* get digits length, skipping blank spaces. */ -static int skp_digits_len(const struct nf_conn *ct, const char *dptr, - const char *limit, int *shift) -{ - for (; dptr < limit && *dptr == ' '; dptr++) - (*shift)++; - - return digits_len(ct, dptr, limit, shift); -} - static int parse_addr(const struct nf_conn *ct, const char *cp, const char **endp, union nf_inet_addr *addr, const char *limit) @@ -294,50 +181,146 @@ int ct_sip_parse_request(const struct nf_conn *ct, } EXPORT_SYMBOL_GPL(ct_sip_parse_request); -/* Returns 0 if not found, -1 error parsing. */ -int ct_sip_get_info(const struct nf_conn *ct, - const char *dptr, size_t dlen, - unsigned int *matchoff, - unsigned int *matchlen, - enum sip_header_pos pos) +/* SIP header parsing: SIP headers are located at the beginning of a line, but + * may span several lines, in which case the continuation lines begin with a + * whitespace character. RFC 2543 allows lines to be terminated with CR, LF or + * CRLF, RFC 3261 allows only CRLF, we support both. + * + * Headers are followed by (optionally) whitespace, a colon, again (optionally) + * whitespace and the values. Whitespace in this context means any amount of + * tabs, spaces and continuation lines, which are treated as a single whitespace + * character. + */ +static const struct sip_header ct_sip_hdrs[] = { + [SIP_HDR_FROM] = SIP_HDR("From", "f", "sip:", skp_epaddr_len), + [SIP_HDR_TO] = SIP_HDR("To", "t", "sip:", skp_epaddr_len), + [SIP_HDR_CONTACT] = SIP_HDR("Contact", "m", "sip:", skp_epaddr_len), + [SIP_HDR_VIA] = SIP_HDR("Via", "v", "UDP ", epaddr_len), + [SIP_HDR_CONTENT_LENGTH] = SIP_HDR("Content-Length", "l", NULL, digits_len), +}; + +static const char *sip_follow_continuation(const char *dptr, const char *limit) { - const struct sip_header_nfo *hnfo = &ct_sip_hdrs[pos]; - const char *limit, *aux, *k = dptr; - int shift = 0; + /* Walk past newline */ + if (++dptr >= limit) + return NULL; + + /* Skip '\n' in CR LF */ + if (*(dptr - 1) == '\r' && *dptr == '\n') { + if (++dptr >= limit) + return NULL; + } + + /* Continuation line? */ + if (*dptr != ' ' && *dptr != '\t') + return NULL; + + /* skip leading whitespace */ + for (; dptr < limit; dptr++) { + if (*dptr != ' ' && *dptr != '\t') + break; + } + return dptr; +} + +static const char *sip_skip_whitespace(const char *dptr, const char *limit) +{ + for (; dptr < limit; dptr++) { + if (*dptr == ' ') + continue; + if (*dptr != '\r' && *dptr != '\n') + break; + dptr = sip_follow_continuation(dptr, limit); + if (dptr == NULL) + return NULL; + } + return dptr; +} - limit = dptr + (dlen - hnfo->lnlen); +/* Search within a SIP header value, dealing with continuation lines */ +static const char *ct_sip_header_search(const char *dptr, const char *limit, + const char *needle, unsigned int len) +{ + for (limit -= len; dptr < limit; dptr++) { + if (*dptr == '\r' || *dptr == '\n') { + dptr = sip_follow_continuation(dptr, limit); + if (dptr == NULL) + break; + continue; + } - while (dptr < limit) { - if ((strncmp(dptr, hnfo->lname, hnfo->lnlen) != 0) && - (hnfo->sname == NULL || - strncmp(dptr, hnfo->sname, hnfo->snlen) != 0)) { - dptr++; + if (strnicmp(dptr, needle, len) == 0) + return dptr; + } + return NULL; +} + +int ct_sip_get_header(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sip_header_types type, + unsigned int *matchoff, unsigned int *matchlen) +{ + const struct sip_header *hdr = &ct_sip_hdrs[type]; + const char *start = dptr, *limit = dptr + datalen; + int shift = 0; + + for (dptr += dataoff; dptr < limit; dptr++) { + /* Find beginning of line */ + if (*dptr != '\r' && *dptr != '\n') continue; + if (++dptr >= limit) + break; + if (*(dptr - 1) == '\r' && *dptr == '\n') { + if (++dptr >= limit) + break; } - aux = ct_sip_search(hnfo->ln_str, dptr, hnfo->ln_strlen, - ct_sip_lnlen(dptr, limit), - hnfo->case_sensitive); - if (!aux) { - pr_debug("'%s' not found in '%s'.\n", hnfo->ln_str, - hnfo->lname); - return -1; + + /* Skip continuation lines */ + if (*dptr == ' ' || *dptr == '\t') + continue; + + /* Find header. Compact headers must be followed by a + * non-alphabetic character to avoid mismatches. */ + if (limit - dptr >= hdr->len && + strnicmp(dptr, hdr->name, hdr->len) == 0) + dptr += hdr->len; + else if (hdr->cname && limit - dptr >= hdr->clen + 1 && + strnicmp(dptr, hdr->cname, hdr->clen) == 0 && + !isalpha(*(dptr + hdr->clen + 1))) + dptr += hdr->clen; + else + continue; + + /* Find and skip colon */ + dptr = sip_skip_whitespace(dptr, limit); + if (dptr == NULL) + break; + if (*dptr != ':' || ++dptr >= limit) + break; + + /* Skip whitespace after colon */ + dptr = sip_skip_whitespace(dptr, limit); + if (dptr == NULL) + break; + + *matchoff = dptr - start; + if (hdr->search) { + dptr = ct_sip_header_search(dptr, limit, hdr->search, + hdr->slen); + if (!dptr) + return -1; + dptr += hdr->slen; } - aux += hnfo->ln_strlen; - *matchlen = hnfo->match_len(ct, aux, limit, &shift); + *matchlen = hdr->match_len(ct, dptr, limit, &shift); if (!*matchlen) return -1; - - *matchoff = (aux - k) + shift; - - pr_debug("%s match succeeded! - len: %u\n", hnfo->lname, - *matchlen); + *matchoff = dptr - start + shift; return 1; } - pr_debug("%s header not found.\n", hnfo->lname); return 0; } -EXPORT_SYMBOL_GPL(ct_sip_get_info); +EXPORT_SYMBOL_GPL(ct_sip_get_header); /* SDP header parsing: a SDP session description contains an ordered set of * headers, starting with a section containing general session parameters, -- cgit v1.2.3 From 05e3ced297fe755093140e7487e292fb7603316e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:19:13 -0700 Subject: [NETFILTER]: nf_conntrack_sip: introduce SIP-URI parsing helper Introduce a helper function to parse a SIP-URI in a header value, optionally iterating through all headers of this kind. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 5 ++ net/netfilter/nf_conntrack_sip.c | 107 +++++++++++++++++++++++++++++ 2 files changed, 112 insertions(+) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index ccc701422963..87bc6f79efc4 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -67,6 +67,11 @@ extern int ct_sip_get_header(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, enum sip_header_types type, unsigned int *matchoff, unsigned int *matchlen); +extern int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, + unsigned int *dataoff, unsigned int datalen, + enum sip_header_types type, int *in_header, + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr, __be16 *port); extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index cbc91598acee..a74d76a97312 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -190,6 +190,9 @@ EXPORT_SYMBOL_GPL(ct_sip_parse_request); * whitespace and the values. Whitespace in this context means any amount of * tabs, spaces and continuation lines, which are treated as a single whitespace * character. + * + * Some headers may appear multiple times. A comma seperated list of values is + * equivalent to multiple headers. */ static const struct sip_header ct_sip_hdrs[] = { [SIP_HDR_FROM] = SIP_HDR("From", "f", "sip:", skp_epaddr_len), @@ -322,6 +325,110 @@ int ct_sip_get_header(const struct nf_conn *ct, const char *dptr, } EXPORT_SYMBOL_GPL(ct_sip_get_header); +/* Get next header field in a list of comma seperated values */ +static int ct_sip_next_header(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sip_header_types type, + unsigned int *matchoff, unsigned int *matchlen) +{ + const struct sip_header *hdr = &ct_sip_hdrs[type]; + const char *start = dptr, *limit = dptr + datalen; + int shift = 0; + + dptr += dataoff; + + dptr = ct_sip_header_search(dptr, limit, ",", strlen(",")); + if (!dptr) + return 0; + + dptr = ct_sip_header_search(dptr, limit, hdr->search, hdr->slen); + if (!dptr) + return 0; + dptr += hdr->slen; + + *matchoff = dptr - start; + *matchlen = hdr->match_len(ct, dptr, limit, &shift); + if (!*matchlen) + return -1; + *matchoff += shift; + return 1; +} + +/* Walk through headers until a parsable one is found or no header of the + * given type is left. */ +static int ct_sip_walk_headers(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sip_header_types type, int *in_header, + unsigned int *matchoff, unsigned int *matchlen) +{ + int ret; + + if (in_header && *in_header) { + while (1) { + ret = ct_sip_next_header(ct, dptr, dataoff, datalen, + type, matchoff, matchlen); + if (ret > 0) + return ret; + if (ret == 0) + break; + dataoff += *matchoff; + } + *in_header = 0; + } + + while (1) { + ret = ct_sip_get_header(ct, dptr, dataoff, datalen, + type, matchoff, matchlen); + if (ret > 0) + break; + if (ret == 0) + return ret; + dataoff += *matchoff; + } + + if (in_header) + *in_header = 1; + return 1; +} + +/* Locate a SIP header, parse the URI and return the offset and length of + * the address as well as the address and port themselves. A stream of + * headers can be parsed by handing in a non-NULL datalen and in_header + * pointer. + */ +int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, + unsigned int *dataoff, unsigned int datalen, + enum sip_header_types type, int *in_header, + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr, __be16 *port) +{ + const char *c, *limit = dptr + datalen; + unsigned int p; + int ret; + + ret = ct_sip_walk_headers(ct, dptr, dataoff ? *dataoff : 0, datalen, + type, in_header, matchoff, matchlen); + WARN_ON(ret < 0); + if (ret == 0) + return ret; + + if (!parse_addr(ct, dptr + *matchoff, &c, addr, limit)) + return -1; + if (*c == ':') { + c++; + p = simple_strtoul(c, (char **)&c, 10); + if (p < 1024 || p > 65535) + return -1; + *port = htons(p); + } else + *port = htons(SIP_PORT); + + if (dataoff) + *dataoff = c - dptr; + return 1; +} +EXPORT_SYMBOL_GPL(ct_sip_parse_header_uri); + /* SDP header parsing: a SDP session description contains an ordered set of * headers, starting with a section containing general session parameters, * optionally followed by multiple media descriptions. -- cgit v1.2.3 From 624f8b7bba98c27a1464f5f858c4a861d5d3e8d7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:19:30 -0700 Subject: [NETFILTER]: nf_nat_sip: get rid of text based header translation Use the URI parsing helper to get the numerical addresses and get rid of the text based header translation. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 3 +- net/ipv4/netfilter/nf_nat_sip.c | 100 +++++++++++------------------ net/netfilter/nf_conntrack_sip.c | 27 ++++++-- 3 files changed, 62 insertions(+), 68 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 87bc6f79efc4..68a0d6a41733 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -62,7 +62,8 @@ extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, extern int ct_sip_parse_request(const struct nf_conn *ct, const char *dptr, unsigned int datalen, - unsigned int *matchoff, unsigned int *matchlen); + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr, __be16 *port); extern int ct_sip_get_header(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, enum sip_header_types type, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index c13e43862361..5b4a5cd23f39 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -26,39 +26,6 @@ MODULE_AUTHOR("Christian Hentschel "); MODULE_DESCRIPTION("SIP NAT helper"); MODULE_ALIAS("ip_nat_sip"); -struct addr_map { - struct { - char src[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; - char dst[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; - unsigned int srclen, srciplen; - unsigned int dstlen, dstiplen; - } addr[IP_CT_DIR_MAX]; -}; - -static void addr_map_init(const struct nf_conn *ct, struct addr_map *map) -{ - const struct nf_conntrack_tuple *t; - enum ip_conntrack_dir dir; - unsigned int n; - - for (dir = 0; dir < IP_CT_DIR_MAX; dir++) { - t = &ct->tuplehash[dir].tuple; - - n = sprintf(map->addr[dir].src, "%u.%u.%u.%u", - NIPQUAD(t->src.u3.ip)); - map->addr[dir].srciplen = n; - n += sprintf(map->addr[dir].src + n, ":%u", - ntohs(t->src.u.udp.port)); - map->addr[dir].srclen = n; - - n = sprintf(map->addr[dir].dst, "%u.%u.%u.%u", - NIPQUAD(t->dst.u3.ip)); - map->addr[dir].dstiplen = n; - n += sprintf(map->addr[dir].dst + n, ":%u", - ntohs(t->dst.u.udp.port)); - map->addr[dir].dstlen = n; - } -} static unsigned int mangle_packet(struct sk_buff *skb, const char **dptr, unsigned int *datalen, @@ -81,43 +48,51 @@ static unsigned int mangle_packet(struct sk_buff *skb, static int map_addr(struct sk_buff *skb, const char **dptr, unsigned int *datalen, unsigned int matchoff, unsigned int matchlen, - struct addr_map *map) + union nf_inet_addr *addr, __be16 port) { enum ip_conntrack_info ctinfo; - struct nf_conn *ct __maybe_unused = nf_ct_get(skb, &ctinfo); + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - unsigned int addrlen; - char *addr; - - if ((matchlen == map->addr[dir].srciplen || - matchlen == map->addr[dir].srclen) && - strncmp(*dptr + matchoff, map->addr[dir].src, matchlen) == 0) { - addr = map->addr[!dir].dst; - addrlen = map->addr[!dir].dstlen; - } else if ((matchlen == map->addr[dir].dstiplen || - matchlen == map->addr[dir].dstlen) && - strncmp(*dptr + matchoff, map->addr[dir].dst, matchlen) == 0) { - addr = map->addr[!dir].src; - addrlen = map->addr[!dir].srclen; + char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + unsigned int buflen; + __be32 newaddr; + __be16 newport; + + if (ct->tuplehash[dir].tuple.src.u3.ip == addr->ip && + ct->tuplehash[dir].tuple.src.u.udp.port == port) { + newaddr = ct->tuplehash[!dir].tuple.dst.u3.ip; + newport = ct->tuplehash[!dir].tuple.dst.u.udp.port; + } else if (ct->tuplehash[dir].tuple.dst.u3.ip == addr->ip && + ct->tuplehash[dir].tuple.dst.u.udp.port == port) { + newaddr = ct->tuplehash[!dir].tuple.src.u3.ip; + newport = ct->tuplehash[!dir].tuple.src.u.udp.port; } else return 1; + if (newaddr == addr->ip && newport == port) + return 1; + + buflen = sprintf(buffer, "%u.%u.%u.%u:%u", + NIPQUAD(newaddr), ntohs(newport)); + return mangle_packet(skb, dptr, datalen, matchoff, matchlen, - addr, addrlen); + buffer, buflen); } static int map_sip_addr(struct sk_buff *skb, const char **dptr, unsigned int *datalen, - enum sip_header_types type, struct addr_map *map) + enum sip_header_types type) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchlen, matchoff; + union nf_inet_addr addr; + __be16 port; - if (ct_sip_get_header(ct, *dptr, 0, *datalen, type, - &matchoff, &matchlen) <= 0) + if (ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, type, NULL, + &matchoff, &matchlen, &addr, &port) <= 0) return 1; - return map_addr(skb, dptr, datalen, matchoff, matchlen, map); + return map_addr(skb, dptr, datalen, matchoff, matchlen, &addr, port); } static unsigned int ip_nat_sip(struct sk_buff *skb, @@ -125,26 +100,27 @@ static unsigned int ip_nat_sip(struct sk_buff *skb, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - struct addr_map map; unsigned int matchoff, matchlen; + union nf_inet_addr addr; + __be16 port; if (*datalen < strlen("SIP/2.0")) return NF_ACCEPT; - addr_map_init(ct, &map); - /* Basic rules: requests and responses. */ if (strnicmp(*dptr, "SIP/2.0", strlen("SIP/2.0")) != 0) { if (ct_sip_parse_request(ct, *dptr, *datalen, - &matchoff, &matchlen) > 0 && - !map_addr(skb, dptr, datalen, matchoff, matchlen, &map)) + &matchoff, &matchlen, + &addr, &port) > 0 && + !map_addr(skb, dptr, datalen, matchoff, matchlen, + &addr, port)) return NF_DROP; } - if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM, &map) || - !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO, &map) || - !map_sip_addr(skb, dptr, datalen, SIP_HDR_VIA, &map) || - !map_sip_addr(skb, dptr, datalen, SIP_HDR_CONTACT, &map)) + if (!map_sip_addr(skb, dptr, datalen, SIP_HDR_FROM) || + !map_sip_addr(skb, dptr, datalen, SIP_HDR_TO) || + !map_sip_addr(skb, dptr, datalen, SIP_HDR_VIA) || + !map_sip_addr(skb, dptr, datalen, SIP_HDR_CONTACT)) return NF_DROP; return NF_ACCEPT; } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index a74d76a97312..f20fa2d94c0a 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -151,10 +151,12 @@ static int skp_epaddr_len(const struct nf_conn *ct, const char *dptr, */ int ct_sip_parse_request(const struct nf_conn *ct, const char *dptr, unsigned int datalen, - unsigned int *matchoff, unsigned int *matchlen) + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr, __be16 *port) { - const char *start = dptr, *limit = dptr + datalen; + const char *start = dptr, *limit = dptr + datalen, *end; unsigned int mlen; + unsigned int p; int shift = 0; /* Skip method and following whitespace */ @@ -173,10 +175,25 @@ int ct_sip_parse_request(const struct nf_conn *ct, if (strnicmp(dptr, "sip:", strlen("sip:")) == 0) break; } - *matchlen = skp_epaddr_len(ct, dptr, limit, &shift); - if (!*matchlen) + if (!skp_epaddr_len(ct, dptr, limit, &shift)) return 0; - *matchoff = dptr - start + shift; + dptr += shift; + + if (!parse_addr(ct, dptr, &end, addr, limit)) + return -1; + if (end < limit && *end == ':') { + end++; + p = simple_strtoul(end, (char **)&end, 10); + if (p < 1024 || p > 65535) + return -1; + *port = htons(p); + } else + *port = htons(SIP_PORT); + + if (end == dptr) + return 0; + *matchoff = dptr - start; + *matchlen = end - dptr; return 1; } EXPORT_SYMBOL_GPL(ct_sip_parse_request); -- cgit v1.2.3 From 30f33e6dee80c6ded917f978e4f377d1069d519d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:22:20 -0700 Subject: [NETFILTER]: nf_conntrack_sip: support method specific request/response handling Add support for per-method request/response handlers and perform SDP parsing for INVITE/UPDATE requests and for all informational and successful responses. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 20 ++++++ net/netfilter/nf_conntrack_sip.c | 107 ++++++++++++++++++++++++++--- 2 files changed, 117 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 68a0d6a41733..da93e80804c2 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -5,6 +5,25 @@ #define SIP_PORT 5060 #define SIP_TIMEOUT 3600 +struct sip_handler { + const char *method; + unsigned int len; + int (*request)(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int cseq); + int (*response)(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int cseq, unsigned int code); +}; + +#define SIP_HANDLER(__method, __request, __response) \ +{ \ + .method = (__method), \ + .len = sizeof(__method) - 1, \ + .request = (__request), \ + .response = (__response), \ +} + struct sip_header { const char *name; const char *cname; @@ -35,6 +54,7 @@ struct sip_header { __SIP_HDR(__name, NULL, __search, __match) enum sip_header_types { + SIP_HDR_CSEQ, SIP_HDR_FROM, SIP_HDR_TO, SIP_HDR_CONTACT, diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 96bedb52bd4b..1be949febab7 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -212,6 +212,7 @@ EXPORT_SYMBOL_GPL(ct_sip_parse_request); * equivalent to multiple headers. */ static const struct sip_header ct_sip_hdrs[] = { + [SIP_HDR_CSEQ] = SIP_HDR("CSeq", NULL, NULL, digits_len), [SIP_HDR_FROM] = SIP_HDR("From", "f", "sip:", skp_epaddr_len), [SIP_HDR_TO] = SIP_HDR("To", "t", "sip:", skp_epaddr_len), [SIP_HDR_CONTACT] = SIP_HDR("Contact", "m", "sip:", skp_epaddr_len), @@ -566,7 +567,8 @@ static int set_expected_rtp(struct sk_buff *skb, } static int process_sdp(struct sk_buff *skb, - const char **dptr, unsigned int *datalen) + const char **dptr, unsigned int *datalen, + unsigned int cseq) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); @@ -600,6 +602,96 @@ static int process_sdp(struct sk_buff *skb, return set_expected_rtp(skb, dptr, datalen, &addr, htons(port)); } +static int process_invite_response(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int cseq, unsigned int code) +{ + if ((code >= 100 && code <= 199) || + (code >= 200 && code <= 299)) + return process_sdp(skb, dptr, datalen, cseq); + + return NF_ACCEPT; +} + +static int process_update_response(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int cseq, unsigned int code) +{ + if ((code >= 100 && code <= 199) || + (code >= 200 && code <= 299)) + return process_sdp(skb, dptr, datalen, cseq); + + return NF_ACCEPT; +} + +static const struct sip_handler sip_handlers[] = { + SIP_HANDLER("INVITE", process_sdp, process_invite_response), + SIP_HANDLER("UPDATE", process_sdp, process_update_response), +}; + +static int process_sip_response(struct sk_buff *skb, + const char **dptr, unsigned int *datalen) +{ + static const struct sip_handler *handler; + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + unsigned int matchoff, matchlen; + unsigned int code, cseq, dataoff, i; + + if (*datalen < strlen("SIP/2.0 200")) + return NF_ACCEPT; + code = simple_strtoul(*dptr + strlen("SIP/2.0 "), NULL, 10); + if (!code) + return NF_DROP; + + if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_CSEQ, + &matchoff, &matchlen) <= 0) + return NF_DROP; + cseq = simple_strtoul(*dptr + matchoff, NULL, 10); + if (!cseq) + return NF_DROP; + dataoff = matchoff + matchlen + 1; + + for (i = 0; i < ARRAY_SIZE(sip_handlers); i++) { + handler = &sip_handlers[i]; + if (handler->response == NULL) + continue; + if (*datalen < dataoff + handler->len || + strnicmp(*dptr + dataoff, handler->method, handler->len)) + continue; + return handler->response(skb, dptr, datalen, cseq, code); + } + return NF_ACCEPT; +} + +static int process_sip_request(struct sk_buff *skb, + const char **dptr, unsigned int *datalen) +{ + static const struct sip_handler *handler; + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + unsigned int matchoff, matchlen; + unsigned int cseq, i; + + for (i = 0; i < ARRAY_SIZE(sip_handlers); i++) { + handler = &sip_handlers[i]; + if (handler->request == NULL) + continue; + if (*datalen < handler->len || + strnicmp(*dptr, handler->method, handler->len)) + continue; + + if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_CSEQ, + &matchoff, &matchlen) <= 0) + return NF_DROP; + cseq = simple_strtoul(*dptr + matchoff, NULL, 10); + if (!cseq) + return NF_DROP; + + return handler->request(skb, dptr, datalen, cseq); + } + return NF_ACCEPT; +} static int sip_help(struct sk_buff *skb, unsigned int protoff, @@ -634,15 +726,10 @@ static int sip_help(struct sk_buff *skb, if (datalen < strlen("SIP/2.0 200")) return NF_ACCEPT; - /* RTP info only in some SDP pkts */ - if (strnicmp(dptr, "INVITE", strlen("INVITE")) != 0 && - strnicmp(dptr, "UPDATE", strlen("UPDATE")) != 0 && - strnicmp(dptr, "SIP/2.0 180", strlen("SIP/2.0 180")) != 0 && - strnicmp(dptr, "SIP/2.0 183", strlen("SIP/2.0 183")) != 0 && - strnicmp(dptr, "SIP/2.0 200", strlen("SIP/2.0 200")) != 0) - return NF_ACCEPT; - - return process_sdp(skb, &dptr, &datalen); + if (strnicmp(dptr, "SIP/2.0 ", strlen("SIP/2.0 ")) != 0) + return process_sip_request(skb, &dptr, &datalen); + else + return process_sip_response(skb, &dptr, &datalen); } static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly; -- cgit v1.2.3 From 2bbb21168a90c788e12fe722eb66f27e611e7df7 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:24:24 -0700 Subject: [NETFILTER]: nf_conntrack_sip: introduce URI and header parameter parsing helpers Introduce URI and header parameter parsing helpers. These are needed by the conntrack helper to parse expiration values in Contact: header parameters and by the NAT helper to properly update the Via-header rport=, received= and maddr= parameters. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 10 ++++++ net/netfilter/nf_conntrack_sip.c | 58 ++++++++++++++++++++++++++++++ 2 files changed, 68 insertions(+) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index da93e80804c2..87e402825dba 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -93,6 +93,16 @@ extern int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, enum sip_header_types type, int *in_header, unsigned int *matchoff, unsigned int *matchlen, union nf_inet_addr *addr, __be16 *port); +extern int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + const char *name, + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr); +extern int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr, + unsigned int off, unsigned int datalen, + const char *name, + unsigned int *matchoff, unsigned int *matchen, + unsigned int *val); extern int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, unsigned int dataoff, unsigned int datalen, diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 8e7e5b465ffb..126f30842d60 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -448,6 +448,64 @@ int ct_sip_parse_header_uri(const struct nf_conn *ct, const char *dptr, } EXPORT_SYMBOL_GPL(ct_sip_parse_header_uri); +/* Parse address from header parameter and return address, offset and length */ +int ct_sip_parse_address_param(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + const char *name, + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr) +{ + const char *limit = dptr + datalen; + const char *start, *end; + + limit = ct_sip_header_search(dptr + dataoff, limit, ",", strlen(",")); + if (!limit) + limit = dptr + datalen; + + start = ct_sip_header_search(dptr + dataoff, limit, name, strlen(name)); + if (!start) + return 0; + + start += strlen(name); + if (!parse_addr(ct, start, &end, addr, limit)) + return 0; + *matchoff = start - dptr; + *matchlen = end - start; + return 1; +} +EXPORT_SYMBOL_GPL(ct_sip_parse_address_param); + +/* Parse numerical header parameter and return value, offset and length */ +int ct_sip_parse_numerical_param(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + const char *name, + unsigned int *matchoff, unsigned int *matchlen, + unsigned int *val) +{ + const char *limit = dptr + datalen; + const char *start; + char *end; + + limit = ct_sip_header_search(dptr + dataoff, limit, ",", strlen(",")); + if (!limit) + limit = dptr + datalen; + + start = ct_sip_header_search(dptr + dataoff, limit, name, strlen(name)); + if (!start) + return 0; + + start += strlen(name); + *val = simple_strtoul(start, &end, 0); + if (start == end) + return 0; + if (matchoff && matchlen) { + *matchoff = start - dptr; + *matchlen = end - start; + } + return 1; +} +EXPORT_SYMBOL_GPL(ct_sip_parse_numerical_param); + /* SDP header parsing: a SDP session description contains an ordered set of * headers, starting with a section containing general session parameters, * optionally followed by multiple media descriptions. -- cgit v1.2.3 From 0f32a40fc91a9ebbbf66e826ac2a829ab37d9cf8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:25:13 -0700 Subject: [NETFILTER]: nf_conntrack_sip: create signalling expectations Create expectations for incoming signalling connections when seeing a REGISTER request. This is needed when the registrar uses a different source port number for signalling messages and for receiving incoming calls from other endpoints than the registrar. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 18 +++ include/net/netfilter/nf_conntrack.h | 4 +- net/ipv4/netfilter/nf_nat_sip.c | 111 +++++++++++--- net/netfilter/nf_conntrack_sip.c | 232 +++++++++++++++++++++++++++-- 4 files changed, 332 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 87e402825dba..7cc84ed0c5da 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -5,6 +5,17 @@ #define SIP_PORT 5060 #define SIP_TIMEOUT 3600 +struct nf_ct_sip_master { + unsigned int register_cseq; +}; + +enum sip_expectation_classes { + SIP_EXPECT_SIGNALLING, + SIP_EXPECT_AUDIO, + __SIP_EXPECT_MAX +}; +#define SIP_EXPECT_MAX (__SIP_EXPECT_MAX - 1) + struct sip_handler { const char *method; unsigned int len; @@ -59,6 +70,7 @@ enum sip_header_types { SIP_HDR_TO, SIP_HDR_CONTACT, SIP_HDR_VIA, + SIP_HDR_EXPIRES, SIP_HDR_CONTENT_LENGTH, }; @@ -75,6 +87,12 @@ enum sdp_header_types { extern unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, const char **dptr, unsigned int *datalen); +extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + struct nf_conntrack_expect *exp, + unsigned int matchoff, + unsigned int matchlen); extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, const char **dptr, unsigned int *datalen, diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 922877133598..4a4f870d2a5e 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -46,6 +46,7 @@ union nf_conntrack_expect_proto { #include #include #include +#include /* per conntrack: application helper private data */ union nf_conntrack_help { @@ -54,6 +55,7 @@ union nf_conntrack_help { struct nf_ct_pptp_master ct_pptp_info; struct nf_ct_h323_master ct_h323_info; struct nf_ct_sane_master ct_sane_info; + struct nf_ct_sip_master ct_sip_info; }; #include @@ -76,7 +78,7 @@ do { \ struct nf_conntrack_helper; /* Must be kept in sync with the classes defined by helpers */ -#define NF_CT_MAX_EXPECT_CLASSES 1 +#define NF_CT_MAX_EXPECT_CLASSES 2 /* nf_conn feature for connections that have a helper */ struct nf_conn_help { diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index b443618a857f..4b85e21a2a4a 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -205,6 +205,91 @@ next: return NF_ACCEPT; } +/* Handles expected signalling connections and media streams */ +static void ip_nat_sip_expected(struct nf_conn *ct, + struct nf_conntrack_expect *exp) +{ + struct nf_nat_range range; + + /* This must be a fresh one. */ + BUG_ON(ct->status & IPS_NAT_DONE_MASK); + + /* For DST manip, map port here to where it's expected. */ + range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); + range.min = range.max = exp->saved_proto; + range.min_ip = range.max_ip = exp->saved_ip; + nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST); + + /* Change src to where master sends to, but only if the connection + * actually came from the same source. */ + if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip == + ct->master->tuplehash[exp->dir].tuple.src.u3.ip) { + range.flags = IP_NAT_RANGE_MAP_IPS; + range.min_ip = range.max_ip + = ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip; + nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC); + } +} + +static unsigned int ip_nat_sip_expect(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + struct nf_conntrack_expect *exp, + unsigned int matchoff, + unsigned int matchlen) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + __be32 newip; + u_int16_t port; + char buffer[sizeof("nnn.nnn.nnn.nnn:nnnnn")]; + unsigned buflen; + + /* Connection will come from reply */ + if (ct->tuplehash[dir].tuple.src.u3.ip == ct->tuplehash[!dir].tuple.dst.u3.ip) + newip = exp->tuple.dst.u3.ip; + else + newip = ct->tuplehash[!dir].tuple.dst.u3.ip; + + /* If the signalling port matches the connection's source port in the + * original direction, try to use the destination port in the opposite + * direction. */ + if (exp->tuple.dst.u.udp.port == + ct->tuplehash[dir].tuple.src.u.udp.port) + port = ntohs(ct->tuplehash[!dir].tuple.dst.u.udp.port); + else + port = ntohs(exp->tuple.dst.u.udp.port); + + exp->saved_ip = exp->tuple.dst.u3.ip; + exp->tuple.dst.u3.ip = newip; + exp->saved_proto.udp.port = exp->tuple.dst.u.udp.port; + exp->dir = !dir; + exp->expectfn = ip_nat_sip_expected; + + for (; port != 0; port++) { + exp->tuple.dst.u.udp.port = htons(port); + if (nf_ct_expect_related(exp) == 0) + break; + } + + if (port == 0) + return NF_DROP; + + if (exp->tuple.dst.u3.ip != exp->saved_ip || + exp->tuple.dst.u.udp.port != exp->saved_proto.udp.port) { + buflen = sprintf(buffer, "%u.%u.%u.%u:%u", + NIPQUAD(newip), port); + if (!mangle_packet(skb, dptr, datalen, matchoff, matchlen, + buffer, buflen)) + goto err; + } + return NF_ACCEPT; + +err: + nf_ct_unexpect_related(exp); + return NF_DROP; +} + static int mangle_content_len(struct sk_buff *skb, const char **dptr, unsigned int *datalen) { @@ -275,27 +360,6 @@ static unsigned int mangle_sdp(struct sk_buff *skb, return mangle_content_len(skb, dptr, datalen); } -static void ip_nat_sdp_expect(struct nf_conn *ct, - struct nf_conntrack_expect *exp) -{ - struct nf_nat_range range; - - /* This must be a fresh one. */ - BUG_ON(ct->status & IPS_NAT_DONE_MASK); - - /* For DST manip, map port here to where it's expected. */ - range.flags = (IP_NAT_RANGE_MAP_IPS | IP_NAT_RANGE_PROTO_SPECIFIED); - range.min = range.max = exp->saved_proto; - range.min_ip = range.max_ip = exp->saved_ip; - nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST); - - /* Change src to where master sends to */ - range.flags = IP_NAT_RANGE_MAP_IPS; - range.min_ip = range.max_ip - = ct->master->tuplehash[!exp->dir].tuple.dst.u3.ip; - nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC); -} - /* So, this packet has hit the connection tracking matching code. Mangle it, and change the expectation to match the new version. */ static unsigned int ip_nat_sdp(struct sk_buff *skb, @@ -322,7 +386,7 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb, /* When you see the packet, we need to NAT it the same as the this one. */ - exp->expectfn = ip_nat_sdp_expect; + exp->expectfn = ip_nat_sip_expected; /* Try to get same port: if not, try to change it. */ for (port = ntohs(exp->saved_proto.udp.port); port != 0; port++) { @@ -344,6 +408,7 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb, static void __exit nf_nat_sip_fini(void) { rcu_assign_pointer(nf_nat_sip_hook, NULL); + rcu_assign_pointer(nf_nat_sip_expect_hook, NULL); rcu_assign_pointer(nf_nat_sdp_hook, NULL); synchronize_rcu(); } @@ -351,8 +416,10 @@ static void __exit nf_nat_sip_fini(void) static int __init nf_nat_sip_init(void) { BUG_ON(nf_nat_sip_hook != NULL); + BUG_ON(nf_nat_sip_expect_hook != NULL); BUG_ON(nf_nat_sdp_hook != NULL); rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip); + rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect); rcu_assign_pointer(nf_nat_sdp_hook, ip_nat_sdp); return 0; } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 126f30842d60..043aa557e7a8 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -37,11 +37,24 @@ static unsigned int sip_timeout __read_mostly = SIP_TIMEOUT; module_param(sip_timeout, uint, 0600); MODULE_PARM_DESC(sip_timeout, "timeout for the master SIP session"); +static int sip_direct_signalling __read_mostly = 1; +module_param(sip_direct_signalling, int, 0600); +MODULE_PARM_DESC(sip_direct_signalling, "expect incoming calls from registrar " + "only (default 1)"); + unsigned int (*nf_nat_sip_hook)(struct sk_buff *skb, const char **dptr, unsigned int *datalen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_hook); +unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + struct nf_conntrack_expect *exp, + unsigned int matchoff, + unsigned int matchlen) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook); + unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, const char **dptr, unsigned int *datalen, @@ -218,6 +231,7 @@ static const struct sip_header ct_sip_hdrs[] = { [SIP_HDR_TO] = SIP_HDR("To", "t", "sip:", skp_epaddr_len), [SIP_HDR_CONTACT] = SIP_HDR("Contact", "m", "sip:", skp_epaddr_len), [SIP_HDR_VIA] = SIP_HDR("Via", "v", "UDP ", epaddr_len), + [SIP_HDR_EXPIRES] = SIP_HDR("Expires", NULL, NULL, digits_len), [SIP_HDR_CONTENT_LENGTH] = SIP_HDR("Content-Length", "l", NULL, digits_len), }; @@ -592,18 +606,50 @@ int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, } EXPORT_SYMBOL_GPL(ct_sip_get_sdp_header); -static void flush_expectations(struct nf_conn *ct) +static int refresh_signalling_expectation(struct nf_conn *ct, + union nf_inet_addr *addr, + __be16 port, + unsigned int expires) { struct nf_conn_help *help = nfct_help(ct); struct nf_conntrack_expect *exp; struct hlist_node *n, *next; + int found = 0; spin_lock_bh(&nf_conntrack_lock); hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) { + if (exp->class != SIP_EXPECT_SIGNALLING || + !nf_inet_addr_cmp(&exp->tuple.dst.u3, addr) || + exp->tuple.dst.u.udp.port != port) + continue; + if (!del_timer(&exp->timeout)) + continue; + exp->flags &= ~NF_CT_EXPECT_INACTIVE; + exp->timeout.expires = jiffies + expires * HZ; + add_timer(&exp->timeout); + found = 1; + break; + } + spin_unlock_bh(&nf_conntrack_lock); + return found; +} + +static void flush_expectations(struct nf_conn *ct, bool media) +{ + struct nf_conn_help *help = nfct_help(ct); + struct nf_conntrack_expect *exp; + struct hlist_node *n, *next; + + spin_lock_bh(&nf_conntrack_lock); + hlist_for_each_entry_safe(exp, n, next, &help->expectations, lnode) { + if ((exp->class != SIP_EXPECT_SIGNALLING) ^ media) + continue; if (!del_timer(&exp->timeout)) continue; nf_ct_unlink_expect(exp); nf_ct_expect_put(exp); + if (!media) + break; } spin_unlock_bh(&nf_conntrack_lock); } @@ -623,7 +669,7 @@ static int set_expected_rtp(struct sk_buff *skb, exp = nf_ct_expect_alloc(ct); if (exp == NULL) return NF_DROP; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + nf_ct_expect_init(exp, SIP_EXPECT_AUDIO, family, &ct->tuplehash[!dir].tuple.src.u3, addr, IPPROTO_UDP, NULL, &port); @@ -688,7 +734,7 @@ static int process_invite_response(struct sk_buff *skb, (code >= 200 && code <= 299)) return process_sdp(skb, dptr, datalen, cseq); else { - flush_expectations(ct); + flush_expectations(ct, true); return NF_ACCEPT; } } @@ -704,7 +750,7 @@ static int process_update_response(struct sk_buff *skb, (code >= 200 && code <= 299)) return process_sdp(skb, dptr, datalen, cseq); else { - flush_expectations(ct); + flush_expectations(ct, true); return NF_ACCEPT; } } @@ -720,7 +766,7 @@ static int process_prack_response(struct sk_buff *skb, (code >= 200 && code <= 299)) return process_sdp(skb, dptr, datalen, cseq); else { - flush_expectations(ct); + flush_expectations(ct, true); return NF_ACCEPT; } } @@ -732,7 +778,165 @@ static int process_bye_request(struct sk_buff *skb, enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - flush_expectations(ct); + flush_expectations(ct, true); + return NF_ACCEPT; +} + +/* Parse a REGISTER request and create a permanent expectation for incoming + * signalling connections. The expectation is marked inactive and is activated + * when receiving a response indicating success from the registrar. + */ +static int process_register_request(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int cseq) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + unsigned int matchoff, matchlen; + struct nf_conntrack_expect *exp; + union nf_inet_addr *saddr, daddr; + __be16 port; + unsigned int expires = 0; + int ret; + typeof(nf_nat_sip_expect_hook) nf_nat_sip_expect; + + /* Expected connections can not register again. */ + if (ct->status & IPS_EXPECTED) + return NF_ACCEPT; + + /* We must check the expiration time: a value of zero signals the + * registrar to release the binding. We'll remove our expectation + * when receiving the new bindings in the response, but we don't + * want to create new ones. + * + * The expiration time may be contained in Expires: header, the + * Contact: header parameters or the URI parameters. + */ + if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_EXPIRES, + &matchoff, &matchlen) > 0) + expires = simple_strtoul(*dptr + matchoff, NULL, 10); + + ret = ct_sip_parse_header_uri(ct, *dptr, NULL, *datalen, + SIP_HDR_CONTACT, NULL, + &matchoff, &matchlen, &daddr, &port); + if (ret < 0) + return NF_DROP; + else if (ret == 0) + return NF_ACCEPT; + + /* We don't support third-party registrations */ + if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.src.u3, &daddr)) + return NF_ACCEPT; + + if (ct_sip_parse_numerical_param(ct, *dptr, + matchoff + matchlen, *datalen, + "expires=", NULL, NULL, &expires) < 0) + return NF_DROP; + + if (expires == 0) { + ret = NF_ACCEPT; + goto store_cseq; + } + + exp = nf_ct_expect_alloc(ct); + if (!exp) + return NF_DROP; + + saddr = NULL; + if (sip_direct_signalling) + saddr = &ct->tuplehash[!dir].tuple.src.u3; + + nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, family, saddr, &daddr, + IPPROTO_UDP, NULL, &port); + exp->timeout.expires = sip_timeout * HZ; + exp->helper = nfct_help(ct)->helper; + exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE; + + nf_nat_sip_expect = rcu_dereference(nf_nat_sip_expect_hook); + if (nf_nat_sip_expect && ct->status & IPS_NAT_MASK) + ret = nf_nat_sip_expect(skb, dptr, datalen, exp, + matchoff, matchlen); + else { + if (nf_ct_expect_related(exp) != 0) + ret = NF_DROP; + else + ret = NF_ACCEPT; + } + nf_ct_expect_put(exp); + +store_cseq: + if (ret == NF_ACCEPT) + help->help.ct_sip_info.register_cseq = cseq; + return ret; +} + +static int process_register_response(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + unsigned int cseq, unsigned int code) +{ + enum ip_conntrack_info ctinfo; + struct nf_conn *ct = nf_ct_get(skb, &ctinfo); + struct nf_conn_help *help = nfct_help(ct); + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + union nf_inet_addr addr; + __be16 port; + unsigned int matchoff, matchlen, dataoff = 0; + unsigned int expires = 0; + int in_contact = 0, ret; + + /* According to RFC 3261, "UAs MUST NOT send a new registration until + * they have received a final response from the registrar for the + * previous one or the previous REGISTER request has timed out". + * + * However, some servers fail to detect retransmissions and send late + * responses, so we store the sequence number of the last valid + * request and compare it here. + */ + if (help->help.ct_sip_info.register_cseq != cseq) + return NF_ACCEPT; + + if (code >= 100 && code <= 199) + return NF_ACCEPT; + if (code < 200 || code > 299) + goto flush; + + if (ct_sip_get_header(ct, *dptr, 0, *datalen, SIP_HDR_EXPIRES, + &matchoff, &matchlen) > 0) + expires = simple_strtoul(*dptr + matchoff, NULL, 10); + + while (1) { + unsigned int c_expires = expires; + + ret = ct_sip_parse_header_uri(ct, *dptr, &dataoff, *datalen, + SIP_HDR_CONTACT, &in_contact, + &matchoff, &matchlen, + &addr, &port); + if (ret < 0) + return NF_DROP; + else if (ret == 0) + break; + + /* We don't support third-party registrations */ + if (!nf_inet_addr_cmp(&ct->tuplehash[dir].tuple.dst.u3, &addr)) + continue; + + ret = ct_sip_parse_numerical_param(ct, *dptr, + matchoff + matchlen, + *datalen, "expires=", + NULL, NULL, &c_expires); + if (ret < 0) + return NF_DROP; + if (c_expires == 0) + break; + if (refresh_signalling_expectation(ct, &addr, port, c_expires)) + return NF_ACCEPT; + } + +flush: + flush_expectations(ct, false); return NF_ACCEPT; } @@ -742,6 +946,7 @@ static const struct sip_handler sip_handlers[] = { SIP_HANDLER("ACK", process_sdp, NULL), SIP_HANDLER("PRACK", process_sdp, process_prack_response), SIP_HANDLER("BYE", process_bye_request, NULL), + SIP_HANDLER("REGISTER", process_register_request, process_register_response), }; static int process_sip_response(struct sk_buff *skb, @@ -853,9 +1058,15 @@ static int sip_help(struct sk_buff *skb, static struct nf_conntrack_helper sip[MAX_PORTS][2] __read_mostly; static char sip_names[MAX_PORTS][2][sizeof("sip-65535")] __read_mostly; -static const struct nf_conntrack_expect_policy sip_exp_policy = { - .max_expected = 2, - .timeout = 3 * 60, +static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1] = { + [SIP_EXPECT_SIGNALLING] = { + .max_expected = 1, + .timeout = 3 * 60, + }, + [SIP_EXPECT_AUDIO] = { + .max_expected = IP_CT_DIR_MAX, + .timeout = 3 * 60, + }, }; static void nf_conntrack_sip_fini(void) @@ -887,7 +1098,8 @@ static int __init nf_conntrack_sip_init(void) for (j = 0; j < 2; j++) { sip[i][j].tuple.dst.protonum = IPPROTO_UDP; sip[i][j].tuple.src.u.udp.port = htons(ports[i]); - sip[i][j].expect_policy = &sip_exp_policy; + sip[i][j].expect_policy = sip_exp_policy; + sip[i][j].expect_class_max = SIP_EXPECT_MAX; sip[i][j].me = THIS_MODULE; sip[i][j].help = sip_help; -- cgit v1.2.3 From a9c1d35917c0c95c8f95a8e497fb91e301419693 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:25:49 -0700 Subject: [NETFILTER]: nf_conntrack_sip: create RTCP expectations Create expectations for the RTCP connections in addition to RTP connections. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 3 +- net/ipv4/netfilter/nf_nat_sip.c | 42 +++++++++++++--------- net/netfilter/nf_conntrack_sip.c | 58 +++++++++++++++++++----------- 3 files changed, 66 insertions(+), 37 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 7cc84ed0c5da..6ddf95f51fb5 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -96,7 +96,8 @@ extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, const char **dptr, unsigned int *datalen, - struct nf_conntrack_expect *exp); + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp); extern int ct_sip_parse_request(const struct nf_conn *ct, const char *dptr, unsigned int datalen, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 4b85e21a2a4a..f73ab4883b75 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -364,7 +364,8 @@ static unsigned int mangle_sdp(struct sk_buff *skb, Mangle it, and change the expectation to match the new version. */ static unsigned int ip_nat_sdp(struct sk_buff *skb, const char **dptr, unsigned int *datalen, - struct nf_conntrack_expect *exp) + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); @@ -375,31 +376,40 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb, /* Connection will come from reply */ if (ct->tuplehash[dir].tuple.src.u3.ip == ct->tuplehash[!dir].tuple.dst.u3.ip) - newip = exp->tuple.dst.u3.ip; + newip = rtp_exp->tuple.dst.u3.ip; else newip = ct->tuplehash[!dir].tuple.dst.u3.ip; - exp->saved_ip = exp->tuple.dst.u3.ip; - exp->tuple.dst.u3.ip = newip; - exp->saved_proto.udp.port = exp->tuple.dst.u.udp.port; - exp->dir = !dir; - - /* When you see the packet, we need to NAT it the same as the - this one. */ - exp->expectfn = ip_nat_sip_expected; - - /* Try to get same port: if not, try to change it. */ - for (port = ntohs(exp->saved_proto.udp.port); port != 0; port++) { - exp->tuple.dst.u.udp.port = htons(port); - if (nf_ct_expect_related(exp) == 0) + rtp_exp->saved_ip = rtp_exp->tuple.dst.u3.ip; + rtp_exp->tuple.dst.u3.ip = newip; + rtp_exp->saved_proto.udp.port = rtp_exp->tuple.dst.u.udp.port; + rtp_exp->dir = !dir; + rtp_exp->expectfn = ip_nat_sip_expected; + + rtcp_exp->saved_ip = rtcp_exp->tuple.dst.u3.ip; + rtcp_exp->tuple.dst.u3.ip = newip; + rtcp_exp->saved_proto.udp.port = rtcp_exp->tuple.dst.u.udp.port; + rtcp_exp->dir = !dir; + rtcp_exp->expectfn = ip_nat_sip_expected; + + /* Try to get same pair of ports: if not, try to change them. */ + for (port = ntohs(rtp_exp->tuple.dst.u.udp.port); + port != 0; port += 2) { + rtp_exp->tuple.dst.u.udp.port = htons(port); + if (nf_ct_expect_related(rtp_exp) != 0) + continue; + rtcp_exp->tuple.dst.u.udp.port = htons(port + 1); + if (nf_ct_expect_related(rtcp_exp) == 0) break; + nf_ct_unexpect_related(rtp_exp); } if (port == 0) return NF_DROP; if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr, datalen)) { - nf_ct_unexpect_related(exp); + nf_ct_unexpect_related(rtp_exp); + nf_ct_unexpect_related(rtcp_exp); return NF_DROP; } return NF_ACCEPT; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 813aa8c67e4c..217262e23403 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -63,7 +63,9 @@ EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook); unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, const char **dptr, unsigned int *datalen, - struct nf_conntrack_expect *exp) __read_mostly; + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp) + __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); static int string_len(const struct nf_conn *ct, const char *dptr, @@ -659,18 +661,20 @@ static void flush_expectations(struct nf_conn *ct, bool media) spin_unlock_bh(&nf_conntrack_lock); } -static int set_expected_rtp(struct sk_buff *skb, - const char **dptr, unsigned int *datalen, - union nf_inet_addr *daddr, __be16 port) +static int set_expected_rtp_rtcp(struct sk_buff *skb, + const char **dptr, unsigned int *datalen, + union nf_inet_addr *daddr, __be16 port) { - struct nf_conntrack_expect *exp; + struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp; enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); union nf_inet_addr *saddr; struct nf_conntrack_tuple tuple; int family = ct->tuplehash[!dir].tuple.src.l3num; - int skip_expect = 0, ret; + int skip_expect = 0, ret = NF_DROP; + u_int16_t base_port; + __be16 rtp_port, rtcp_port; typeof(nf_nat_sdp_hook) nf_nat_sdp; saddr = NULL; @@ -704,23 +708,37 @@ static int set_expected_rtp(struct sk_buff *skb, if (skip_expect) return NF_ACCEPT; - exp = nf_ct_expect_alloc(ct); - if (exp == NULL) - return NF_DROP; - nf_ct_expect_init(exp, SIP_EXPECT_AUDIO, family, saddr, daddr, - IPPROTO_UDP, NULL, &port); + base_port = ntohs(tuple.dst.u.udp.port) & ~1; + rtp_port = htons(base_port); + rtcp_port = htons(base_port + 1); + + rtp_exp = nf_ct_expect_alloc(ct); + if (rtp_exp == NULL) + goto err1; + nf_ct_expect_init(rtp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr, + IPPROTO_UDP, NULL, &rtp_port); + + rtcp_exp = nf_ct_expect_alloc(ct); + if (rtcp_exp == NULL) + goto err2; + nf_ct_expect_init(rtcp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr, + IPPROTO_UDP, NULL, &rtcp_port); nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook); if (nf_nat_sdp && ct->status & IPS_NAT_MASK) - ret = nf_nat_sdp(skb, dptr, datalen, exp); + ret = nf_nat_sdp(skb, dptr, datalen, rtp_exp, rtcp_exp); else { - if (nf_ct_expect_related(exp) != 0) - ret = NF_DROP; - else - ret = NF_ACCEPT; + if (nf_ct_expect_related(rtp_exp) == 0) { + if (nf_ct_expect_related(rtcp_exp) != 0) + nf_ct_unexpect_related(rtp_exp); + else + ret = NF_ACCEPT; + } } - nf_ct_expect_put(exp); - + nf_ct_expect_put(rtcp_exp); +err2: + nf_ct_expect_put(rtp_exp); +err1: return ret; } @@ -758,7 +776,7 @@ static int process_sdp(struct sk_buff *skb, if (port < 1024 || port > 65535) return NF_DROP; - return set_expected_rtp(skb, dptr, datalen, &addr, htons(port)); + return set_expected_rtp_rtcp(skb, dptr, datalen, &addr, htons(port)); } static int process_invite_response(struct sk_buff *skb, const char **dptr, unsigned int *datalen, @@ -1101,7 +1119,7 @@ static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1 .timeout = 3 * 60, }, [SIP_EXPECT_AUDIO] = { - .max_expected = IP_CT_DIR_MAX, + .max_expected = 2 * IP_CT_DIR_MAX, .timeout = 3 * 60, }, }; -- cgit v1.2.3 From 4ab9e64e5e3c0516577818804aaf13a630d67bc9 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:26:08 -0700 Subject: [NETFILTER]: nf_nat_sip: split up SDP mangling The SDP connection addresses may be contained in the payload multiple times (in the session description and/or once per media description), currently only the session description is properly updated. Split up SDP mangling so the function setting up expectations only updates the media port, update connection addresses from media descriptions while parsing them and at the end update the session description when the final addresses are known. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 25 ++++- net/ipv4/netfilter/nf_nat_sip.c | 121 ++++++++++++++++-------- net/netfilter/nf_conntrack_sip.c | 142 +++++++++++++++++++++++------ 3 files changed, 219 insertions(+), 69 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 6ddf95f51fb5..eca3ad3f28dc 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -93,11 +93,26 @@ extern unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, struct nf_conntrack_expect *exp, unsigned int matchoff, unsigned int matchlen); -extern unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, - const char **dptr, - unsigned int *datalen, - struct nf_conntrack_expect *rtp_exp, - struct nf_conntrack_expect *rtcp_exp); +extern unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int dataoff, + unsigned int *datalen, + enum sdp_header_types type, + enum sdp_header_types term, + const union nf_inet_addr *addr); +extern unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int dataoff, + unsigned int *datalen, + const union nf_inet_addr *addr); +extern unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp, + unsigned int mediaoff, + unsigned int medialen, + union nf_inet_addr *rtp_addr); extern int ct_sip_parse_request(const struct nf_conn *ct, const char *dptr, unsigned int datalen, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index f73ab4883b75..4429069d9b42 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -316,45 +316,77 @@ static int mangle_content_len(struct sk_buff *skb, buffer, buflen); } -static unsigned mangle_sdp_packet(struct sk_buff *skb, - const char **dptr, unsigned int *datalen, +static unsigned mangle_sdp_packet(struct sk_buff *skb, const char **dptr, + unsigned int dataoff, unsigned int *datalen, enum sdp_header_types type, + enum sdp_header_types term, char *buffer, int buflen) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); unsigned int matchlen, matchoff; - if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, type, SDP_HDR_UNSPEC, + if (ct_sip_get_sdp_header(ct, *dptr, dataoff, *datalen, type, term, &matchoff, &matchlen) <= 0) return 0; return mangle_packet(skb, dptr, datalen, matchoff, matchlen, buffer, buflen); } -static unsigned int mangle_sdp(struct sk_buff *skb, - enum ip_conntrack_info ctinfo, - struct nf_conn *ct, - __be32 newip, u_int16_t port, - const char **dptr, unsigned int *datalen) +static unsigned int ip_nat_sdp_addr(struct sk_buff *skb, const char **dptr, + unsigned int dataoff, + unsigned int *datalen, + enum sdp_header_types type, + enum sdp_header_types term, + const union nf_inet_addr *addr) { char buffer[sizeof("nnn.nnn.nnn.nnn")]; - unsigned int bufflen; + unsigned int buflen; - /* Mangle owner and contact info. */ - bufflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(newip)); - if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_OWNER_IP4, - buffer, bufflen)) + buflen = sprintf(buffer, NIPQUAD_FMT, NIPQUAD(addr->ip)); + if (!mangle_sdp_packet(skb, dptr, dataoff, datalen, type, term, + buffer, buflen)) return 0; - if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_CONNECTION_IP4, - buffer, bufflen)) + return mangle_content_len(skb, dptr, datalen); +} + +static unsigned int ip_nat_sdp_port(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + unsigned int matchoff, + unsigned int matchlen, + u_int16_t port) +{ + char buffer[sizeof("nnnnn")]; + unsigned int buflen; + + buflen = sprintf(buffer, "%u", port); + if (!mangle_packet(skb, dptr, datalen, matchoff, matchlen, + buffer, buflen)) return 0; - /* Mangle media port. */ - bufflen = sprintf(buffer, "%u", port); - if (!mangle_sdp_packet(skb, dptr, datalen, SDP_HDR_MEDIA, - buffer, bufflen)) + return mangle_content_len(skb, dptr, datalen); +} + +static unsigned int ip_nat_sdp_session(struct sk_buff *skb, const char **dptr, + unsigned int dataoff, + unsigned int *datalen, + const union nf_inet_addr *addr) +{ + char buffer[sizeof("nnn.nnn.nnn.nnn")]; + unsigned int buflen; + + /* Mangle session description owner and contact addresses */ + buflen = sprintf(buffer, "%u.%u.%u.%u", NIPQUAD(addr->ip)); + if (!mangle_sdp_packet(skb, dptr, dataoff, datalen, + SDP_HDR_OWNER_IP4, SDP_HDR_MEDIA, + buffer, buflen)) + return 0; + + if (!mangle_sdp_packet(skb, dptr, dataoff, datalen, + SDP_HDR_CONNECTION_IP4, SDP_HDR_MEDIA, + buffer, buflen)) return 0; return mangle_content_len(skb, dptr, datalen); @@ -362,32 +394,35 @@ static unsigned int mangle_sdp(struct sk_buff *skb, /* So, this packet has hit the connection tracking matching code. Mangle it, and change the expectation to match the new version. */ -static unsigned int ip_nat_sdp(struct sk_buff *skb, - const char **dptr, unsigned int *datalen, - struct nf_conntrack_expect *rtp_exp, - struct nf_conntrack_expect *rtcp_exp) +static unsigned int ip_nat_sdp_media(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp, + unsigned int mediaoff, + unsigned int medialen, + union nf_inet_addr *rtp_addr) { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - __be32 newip; u_int16_t port; /* Connection will come from reply */ if (ct->tuplehash[dir].tuple.src.u3.ip == ct->tuplehash[!dir].tuple.dst.u3.ip) - newip = rtp_exp->tuple.dst.u3.ip; + rtp_addr->ip = rtp_exp->tuple.dst.u3.ip; else - newip = ct->tuplehash[!dir].tuple.dst.u3.ip; + rtp_addr->ip = ct->tuplehash[!dir].tuple.dst.u3.ip; rtp_exp->saved_ip = rtp_exp->tuple.dst.u3.ip; - rtp_exp->tuple.dst.u3.ip = newip; + rtp_exp->tuple.dst.u3.ip = rtp_addr->ip; rtp_exp->saved_proto.udp.port = rtp_exp->tuple.dst.u.udp.port; rtp_exp->dir = !dir; rtp_exp->expectfn = ip_nat_sip_expected; rtcp_exp->saved_ip = rtcp_exp->tuple.dst.u3.ip; - rtcp_exp->tuple.dst.u3.ip = newip; + rtcp_exp->tuple.dst.u3.ip = rtp_addr->ip; rtcp_exp->saved_proto.udp.port = rtcp_exp->tuple.dst.u.udp.port; rtcp_exp->dir = !dir; rtcp_exp->expectfn = ip_nat_sip_expected; @@ -405,21 +440,29 @@ static unsigned int ip_nat_sdp(struct sk_buff *skb, } if (port == 0) - return NF_DROP; + goto err1; + + /* Update media port. */ + if (rtp_exp->tuple.dst.u.udp.port != rtp_exp->saved_proto.udp.port && + !ip_nat_sdp_port(skb, dptr, datalen, mediaoff, medialen, port)) + goto err2; - if (!mangle_sdp(skb, ctinfo, ct, newip, port, dptr, datalen)) { - nf_ct_unexpect_related(rtp_exp); - nf_ct_unexpect_related(rtcp_exp); - return NF_DROP; - } return NF_ACCEPT; + +err2: + nf_ct_unexpect_related(rtp_exp); + nf_ct_unexpect_related(rtcp_exp); +err1: + return NF_DROP; } static void __exit nf_nat_sip_fini(void) { rcu_assign_pointer(nf_nat_sip_hook, NULL); rcu_assign_pointer(nf_nat_sip_expect_hook, NULL); - rcu_assign_pointer(nf_nat_sdp_hook, NULL); + rcu_assign_pointer(nf_nat_sdp_addr_hook, NULL); + rcu_assign_pointer(nf_nat_sdp_session_hook, NULL); + rcu_assign_pointer(nf_nat_sdp_media_hook, NULL); synchronize_rcu(); } @@ -427,10 +470,14 @@ static int __init nf_nat_sip_init(void) { BUG_ON(nf_nat_sip_hook != NULL); BUG_ON(nf_nat_sip_expect_hook != NULL); - BUG_ON(nf_nat_sdp_hook != NULL); + BUG_ON(nf_nat_sdp_addr_hook != NULL); + BUG_ON(nf_nat_sdp_session_hook != NULL); + BUG_ON(nf_nat_sdp_media_hook != NULL); rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip); rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect); - rcu_assign_pointer(nf_nat_sdp_hook, ip_nat_sdp); + rcu_assign_pointer(nf_nat_sdp_addr_hook, ip_nat_sdp_addr); + rcu_assign_pointer(nf_nat_sdp_session_hook, ip_nat_sdp_session); + rcu_assign_pointer(nf_nat_sdp_media_hook, ip_nat_sdp_media); return 0; } diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 217262e23403..f929add324f3 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -60,13 +60,34 @@ unsigned int (*nf_nat_sip_expect_hook)(struct sk_buff *skb, unsigned int matchlen) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sip_expect_hook); -unsigned int (*nf_nat_sdp_hook)(struct sk_buff *skb, - const char **dptr, - unsigned int *datalen, - struct nf_conntrack_expect *rtp_exp, - struct nf_conntrack_expect *rtcp_exp) - __read_mostly; -EXPORT_SYMBOL_GPL(nf_nat_sdp_hook); +unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int dataoff, + unsigned int *datalen, + enum sdp_header_types type, + enum sdp_header_types term, + const union nf_inet_addr *addr) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sdp_addr_hook); + +unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int dataoff, + unsigned int *datalen, + const union nf_inet_addr *addr) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sdp_session_hook); + +unsigned int (*nf_nat_sdp_media_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + struct nf_conntrack_expect *rtp_exp, + struct nf_conntrack_expect *rtcp_exp, + unsigned int mediaoff, + unsigned int medialen, + union nf_inet_addr *rtp_addr) + __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sdp_media_hook); static int string_len(const struct nf_conn *ct, const char *dptr, const char *limit, int *shift) @@ -613,6 +634,26 @@ int ct_sip_get_sdp_header(const struct nf_conn *ct, const char *dptr, } EXPORT_SYMBOL_GPL(ct_sip_get_sdp_header); +static int ct_sip_parse_sdp_addr(const struct nf_conn *ct, const char *dptr, + unsigned int dataoff, unsigned int datalen, + enum sdp_header_types type, + enum sdp_header_types term, + unsigned int *matchoff, unsigned int *matchlen, + union nf_inet_addr *addr) +{ + int ret; + + ret = ct_sip_get_sdp_header(ct, dptr, dataoff, datalen, type, term, + matchoff, matchlen); + if (ret <= 0) + return ret; + + if (!parse_addr(ct, dptr + *matchoff, NULL, addr, + dptr + *matchoff + *matchlen)) + return -1; + return 1; +} + static int refresh_signalling_expectation(struct nf_conn *ct, union nf_inet_addr *addr, __be16 port, @@ -663,7 +704,8 @@ static void flush_expectations(struct nf_conn *ct, bool media) static int set_expected_rtp_rtcp(struct sk_buff *skb, const char **dptr, unsigned int *datalen, - union nf_inet_addr *daddr, __be16 port) + union nf_inet_addr *daddr, __be16 port, + unsigned int mediaoff, unsigned int medialen) { struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp; enum ip_conntrack_info ctinfo; @@ -675,7 +717,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, int skip_expect = 0, ret = NF_DROP; u_int16_t base_port; __be16 rtp_port, rtcp_port; - typeof(nf_nat_sdp_hook) nf_nat_sdp; + typeof(nf_nat_sdp_media_hook) nf_nat_sdp_media; saddr = NULL; if (sip_direct_media) { @@ -724,9 +766,10 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, nf_ct_expect_init(rtcp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr, IPPROTO_UDP, NULL, &rtcp_port); - nf_nat_sdp = rcu_dereference(nf_nat_sdp_hook); - if (nf_nat_sdp && ct->status & IPS_NAT_MASK) - ret = nf_nat_sdp(skb, dptr, datalen, rtp_exp, rtcp_exp); + nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook); + if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK) + ret = nf_nat_sdp_media(skb, dptr, datalen, rtp_exp, rtcp_exp, + mediaoff, medialen, daddr); else { if (nf_ct_expect_related(rtp_exp) == 0) { if (nf_ct_expect_related(rtcp_exp) != 0) @@ -750,33 +793,78 @@ static int process_sdp(struct sk_buff *skb, struct nf_conn *ct = nf_ct_get(skb, &ctinfo); int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; unsigned int matchoff, matchlen; - union nf_inet_addr addr; + unsigned int mediaoff, medialen; + unsigned int sdpoff; + unsigned int caddr_len, maddr_len; + union nf_inet_addr caddr, maddr, rtp_addr; unsigned int port; - enum sdp_header_types type; + enum sdp_header_types c_hdr; + int ret; + typeof(nf_nat_sdp_addr_hook) nf_nat_sdp_addr; + typeof(nf_nat_sdp_session_hook) nf_nat_sdp_session; - /* Get address and port from SDP packet. */ - type = family == AF_INET ? SDP_HDR_CONNECTION_IP4 : - SDP_HDR_CONNECTION_IP6; + c_hdr = family == AF_INET ? SDP_HDR_CONNECTION_IP4 : + SDP_HDR_CONNECTION_IP6; + /* Find beginning of session description */ if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, - type, SDP_HDR_UNSPEC, + SDP_HDR_VERSION, SDP_HDR_UNSPEC, &matchoff, &matchlen) <= 0) return NF_ACCEPT; - - /* We'll drop only if there are parse problems. */ - if (!parse_addr(ct, *dptr + matchoff, NULL, &addr, *dptr + *datalen)) - return NF_DROP; - - if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, + sdpoff = matchoff; + + /* The connection information is contained in the session description + * and/or once per media description. The first media description marks + * the end of the session description. */ + caddr_len = 0; + if (ct_sip_parse_sdp_addr(ct, *dptr, sdpoff, *datalen, + c_hdr, SDP_HDR_MEDIA, + &matchoff, &matchlen, &caddr) > 0) + caddr_len = matchlen; + + if (ct_sip_get_sdp_header(ct, *dptr, sdpoff, *datalen, SDP_HDR_MEDIA, SDP_HDR_UNSPEC, - &matchoff, &matchlen) <= 0) + &mediaoff, &medialen) <= 0) return NF_ACCEPT; - port = simple_strtoul(*dptr + matchoff, NULL, 10); + port = simple_strtoul(*dptr + mediaoff, NULL, 10); if (port < 1024 || port > 65535) return NF_DROP; - return set_expected_rtp_rtcp(skb, dptr, datalen, &addr, htons(port)); + /* The media description overrides the session description. */ + maddr_len = 0; + if (ct_sip_parse_sdp_addr(ct, *dptr, mediaoff, *datalen, + c_hdr, SDP_HDR_MEDIA, + &matchoff, &matchlen, &maddr) > 0) { + maddr_len = matchlen; + memcpy(&rtp_addr, &maddr, sizeof(rtp_addr)); + } else if (caddr_len) + memcpy(&rtp_addr, &caddr, sizeof(rtp_addr)); + else + return NF_DROP; + + ret = set_expected_rtp_rtcp(skb, dptr, datalen, &rtp_addr, htons(port), + mediaoff, medialen); + if (ret != NF_ACCEPT) + return ret; + + /* Update media connection address if present */ + if (maddr_len) { + nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook); + if (nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) { + ret = nf_nat_sdp_addr(skb, dptr, mediaoff, datalen, + c_hdr, SDP_HDR_MEDIA, &rtp_addr); + if (ret != NF_ACCEPT) + return ret; + } + } + + /* Update session connection and owner addresses */ + nf_nat_sdp_session = rcu_dereference(nf_nat_sdp_session_hook); + if (nf_nat_sdp_session && ct->status & IPS_NAT_MASK) + ret = nf_nat_sdp_session(skb, dptr, sdpoff, datalen, &rtp_addr); + + return ret; } static int process_invite_response(struct sk_buff *skb, const char **dptr, unsigned int *datalen, -- cgit v1.2.3 From 0d0ab0378d67517a4f4ae3497706c13d9dd24af1 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:26:24 -0700 Subject: [NETFILTER]: nf_conntrack_sip: support multiple media channels Add support for multiple media channels and use it to create expectations for video streams when present. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 14 ++++ include/net/netfilter/nf_conntrack.h | 2 +- net/netfilter/nf_conntrack_sip.c | 121 +++++++++++++++++++++-------- 3 files changed, 105 insertions(+), 32 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index eca3ad3f28dc..71fa3eb5f485 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -12,10 +12,24 @@ struct nf_ct_sip_master { enum sip_expectation_classes { SIP_EXPECT_SIGNALLING, SIP_EXPECT_AUDIO, + SIP_EXPECT_VIDEO, __SIP_EXPECT_MAX }; #define SIP_EXPECT_MAX (__SIP_EXPECT_MAX - 1) +struct sdp_media_type { + const char *name; + unsigned int len; + enum sip_expectation_classes class; +}; + +#define SDP_MEDIA_TYPE(__name, __class) \ +{ \ + .name = (__name), \ + .len = sizeof(__name) - 1, \ + .class = (__class), \ +} + struct sip_handler { const char *method; unsigned int len; diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 4a4f870d2a5e..a3567a7a6d67 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -78,7 +78,7 @@ do { \ struct nf_conntrack_helper; /* Must be kept in sync with the classes defined by helpers */ -#define NF_CT_MAX_EXPECT_CLASSES 2 +#define NF_CT_MAX_EXPECT_CLASSES 3 /* nf_conn feature for connections that have a helper */ struct nf_conn_help { diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index f929add324f3..f40a525732d1 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -112,6 +112,21 @@ static int digits_len(const struct nf_conn *ct, const char *dptr, return len; } +/* get media type + port length */ +static int media_len(const struct nf_conn *ct, const char *dptr, + const char *limit, int *shift) +{ + int len = string_len(ct, dptr, limit, shift); + + dptr += len; + if (dptr >= limit || *dptr != ' ') + return 0; + len++; + dptr++; + + return len + digits_len(ct, dptr, limit, shift); +} + static int parse_addr(const struct nf_conn *ct, const char *cp, const char **endp, union nf_inet_addr *addr, const char *limit) @@ -563,7 +578,7 @@ static const struct sip_header ct_sdp_hdrs[] = { [SDP_HDR_CONNECTION_IP4] = SDP_HDR("c=", "IN IP4 ", epaddr_len), [SDP_HDR_OWNER_IP6] = SDP_HDR("o=", "IN IP6 ", epaddr_len), [SDP_HDR_CONNECTION_IP6] = SDP_HDR("c=", "IN IP6 ", epaddr_len), - [SDP_HDR_MEDIA] = SDP_HDR("m=", "audio ", digits_len), + [SDP_HDR_MEDIA] = SDP_HDR("m=", NULL, media_len), }; /* Linear string search within SDP header values */ @@ -705,6 +720,7 @@ static void flush_expectations(struct nf_conn *ct, bool media) static int set_expected_rtp_rtcp(struct sk_buff *skb, const char **dptr, unsigned int *datalen, union nf_inet_addr *daddr, __be16 port, + enum sip_expectation_classes class, unsigned int mediaoff, unsigned int medialen) { struct nf_conntrack_expect *exp, *rtp_exp, *rtcp_exp; @@ -743,7 +759,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, exp = __nf_ct_expect_find(&tuple); if (exp && exp->master != ct && nfct_help(exp->master)->helper == nfct_help(ct)->helper && - exp->class == SIP_EXPECT_AUDIO) + exp->class == class) skip_expect = 1; rcu_read_unlock(); @@ -757,13 +773,13 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, rtp_exp = nf_ct_expect_alloc(ct); if (rtp_exp == NULL) goto err1; - nf_ct_expect_init(rtp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr, + nf_ct_expect_init(rtp_exp, class, family, saddr, daddr, IPPROTO_UDP, NULL, &rtp_port); rtcp_exp = nf_ct_expect_alloc(ct); if (rtcp_exp == NULL) goto err2; - nf_ct_expect_init(rtcp_exp, SIP_EXPECT_AUDIO, family, saddr, daddr, + nf_ct_expect_init(rtcp_exp, class, family, saddr, daddr, IPPROTO_UDP, NULL, &rtcp_port); nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook); @@ -785,6 +801,28 @@ err1: return ret; } +static const struct sdp_media_type sdp_media_types[] = { + SDP_MEDIA_TYPE("audio ", SIP_EXPECT_AUDIO), + SDP_MEDIA_TYPE("video ", SIP_EXPECT_VIDEO), +}; + +static const struct sdp_media_type *sdp_media_type(const char *dptr, + unsigned int matchoff, + unsigned int matchlen) +{ + const struct sdp_media_type *t; + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(sdp_media_types); i++) { + t = &sdp_media_types[i]; + if (matchlen < t->len || + strncmp(dptr + matchoff, t->name, t->len)) + continue; + return t; + } + return NULL; +} + static int process_sdp(struct sk_buff *skb, const char **dptr, unsigned int *datalen, unsigned int cseq) @@ -796,13 +834,16 @@ static int process_sdp(struct sk_buff *skb, unsigned int mediaoff, medialen; unsigned int sdpoff; unsigned int caddr_len, maddr_len; + unsigned int i; union nf_inet_addr caddr, maddr, rtp_addr; unsigned int port; enum sdp_header_types c_hdr; - int ret; + const struct sdp_media_type *t; + int ret = NF_ACCEPT; typeof(nf_nat_sdp_addr_hook) nf_nat_sdp_addr; typeof(nf_nat_sdp_session_hook) nf_nat_sdp_session; + nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook); c_hdr = family == AF_INET ? SDP_HDR_CONNECTION_IP4 : SDP_HDR_CONNECTION_IP6; @@ -822,41 +863,55 @@ static int process_sdp(struct sk_buff *skb, &matchoff, &matchlen, &caddr) > 0) caddr_len = matchlen; - if (ct_sip_get_sdp_header(ct, *dptr, sdpoff, *datalen, - SDP_HDR_MEDIA, SDP_HDR_UNSPEC, - &mediaoff, &medialen) <= 0) - return NF_ACCEPT; + mediaoff = sdpoff; + for (i = 0; i < ARRAY_SIZE(sdp_media_types); ) { + if (ct_sip_get_sdp_header(ct, *dptr, mediaoff, *datalen, + SDP_HDR_MEDIA, SDP_HDR_UNSPEC, + &mediaoff, &medialen) <= 0) + break; - port = simple_strtoul(*dptr + mediaoff, NULL, 10); - if (port < 1024 || port > 65535) - return NF_DROP; + /* Get media type and port number. A media port value of zero + * indicates an inactive stream. */ + t = sdp_media_type(*dptr, mediaoff, medialen); + if (!t) { + mediaoff += medialen; + continue; + } + mediaoff += t->len; + medialen -= t->len; - /* The media description overrides the session description. */ - maddr_len = 0; - if (ct_sip_parse_sdp_addr(ct, *dptr, mediaoff, *datalen, - c_hdr, SDP_HDR_MEDIA, - &matchoff, &matchlen, &maddr) > 0) { - maddr_len = matchlen; - memcpy(&rtp_addr, &maddr, sizeof(rtp_addr)); - } else if (caddr_len) - memcpy(&rtp_addr, &caddr, sizeof(rtp_addr)); - else - return NF_DROP; + port = simple_strtoul(*dptr + mediaoff, NULL, 10); + if (port == 0) + continue; + if (port < 1024 || port > 65535) + return NF_DROP; - ret = set_expected_rtp_rtcp(skb, dptr, datalen, &rtp_addr, htons(port), - mediaoff, medialen); - if (ret != NF_ACCEPT) - return ret; + /* The media description overrides the session description. */ + maddr_len = 0; + if (ct_sip_parse_sdp_addr(ct, *dptr, mediaoff, *datalen, + c_hdr, SDP_HDR_MEDIA, + &matchoff, &matchlen, &maddr) > 0) { + maddr_len = matchlen; + memcpy(&rtp_addr, &maddr, sizeof(rtp_addr)); + } else if (caddr_len) + memcpy(&rtp_addr, &caddr, sizeof(rtp_addr)); + else + return NF_DROP; + + ret = set_expected_rtp_rtcp(skb, dptr, datalen, + &rtp_addr, htons(port), t->class, + mediaoff, medialen); + if (ret != NF_ACCEPT) + return ret; - /* Update media connection address if present */ - if (maddr_len) { - nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook); - if (nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) { + /* Update media connection address if present */ + if (maddr_len && nf_nat_sdp_addr && ct->status & IPS_NAT_MASK) { ret = nf_nat_sdp_addr(skb, dptr, mediaoff, datalen, c_hdr, SDP_HDR_MEDIA, &rtp_addr); if (ret != NF_ACCEPT) return ret; } + i++; } /* Update session connection and owner addresses */ @@ -1210,6 +1265,10 @@ static const struct nf_conntrack_expect_policy sip_exp_policy[SIP_EXPECT_MAX + 1 .max_expected = 2 * IP_CT_DIR_MAX, .timeout = 3 * 60, }, + [SIP_EXPECT_VIDEO] = { + .max_expected = 2 * IP_CT_DIR_MAX, + .timeout = 3 * 60, + }, }; static void nf_conntrack_sip_fini(void) -- cgit v1.2.3 From c7f485abd618e0d249bdd1abdc586bd10fee1954 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Tue, 25 Mar 2008 20:26:43 -0700 Subject: [NETFILTER]: nf_conntrack_sip: RTP routing optimization Optimize call routing between NATed endpoints: when an external registrar sends a media description that contains an existing RTP expectation from a different SNATed connection, the gatekeeper is trying to route the call directly between the two endpoints. We assume both endpoints can reach each other directly and "un-NAT" the addresses, which makes the media stream go between the two endpoints directly. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/netfilter/nf_conntrack_sip.h | 6 +++ net/ipv4/netfilter/nf_nat_sip.c | 3 ++ net/netfilter/nf_conntrack_sip.c | 59 +++++++++++++++++++++++++----- 3 files changed, 58 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_sip.h b/include/linux/netfilter/nf_conntrack_sip.h index 71fa3eb5f485..5da04e586a3f 100644 --- a/include/linux/netfilter/nf_conntrack_sip.h +++ b/include/linux/netfilter/nf_conntrack_sip.h @@ -114,6 +114,12 @@ extern unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, enum sdp_header_types type, enum sdp_header_types term, const union nf_inet_addr *addr); +extern unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + unsigned int matchoff, + unsigned int matchlen, + u_int16_t port); extern unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb, const char **dptr, unsigned int dataoff, diff --git a/net/ipv4/netfilter/nf_nat_sip.c b/net/ipv4/netfilter/nf_nat_sip.c index 4429069d9b42..bcddccddf768 100644 --- a/net/ipv4/netfilter/nf_nat_sip.c +++ b/net/ipv4/netfilter/nf_nat_sip.c @@ -461,6 +461,7 @@ static void __exit nf_nat_sip_fini(void) rcu_assign_pointer(nf_nat_sip_hook, NULL); rcu_assign_pointer(nf_nat_sip_expect_hook, NULL); rcu_assign_pointer(nf_nat_sdp_addr_hook, NULL); + rcu_assign_pointer(nf_nat_sdp_port_hook, NULL); rcu_assign_pointer(nf_nat_sdp_session_hook, NULL); rcu_assign_pointer(nf_nat_sdp_media_hook, NULL); synchronize_rcu(); @@ -471,11 +472,13 @@ static int __init nf_nat_sip_init(void) BUG_ON(nf_nat_sip_hook != NULL); BUG_ON(nf_nat_sip_expect_hook != NULL); BUG_ON(nf_nat_sdp_addr_hook != NULL); + BUG_ON(nf_nat_sdp_port_hook != NULL); BUG_ON(nf_nat_sdp_session_hook != NULL); BUG_ON(nf_nat_sdp_media_hook != NULL); rcu_assign_pointer(nf_nat_sip_hook, ip_nat_sip); rcu_assign_pointer(nf_nat_sip_expect_hook, ip_nat_sip_expect); rcu_assign_pointer(nf_nat_sdp_addr_hook, ip_nat_sdp_addr); + rcu_assign_pointer(nf_nat_sdp_port_hook, ip_nat_sdp_port); rcu_assign_pointer(nf_nat_sdp_session_hook, ip_nat_sdp_session); rcu_assign_pointer(nf_nat_sdp_media_hook, ip_nat_sdp_media); return 0; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index f40a525732d1..57de22c770a3 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -70,6 +70,14 @@ unsigned int (*nf_nat_sdp_addr_hook)(struct sk_buff *skb, __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_sdp_addr_hook); +unsigned int (*nf_nat_sdp_port_hook)(struct sk_buff *skb, + const char **dptr, + unsigned int *datalen, + unsigned int matchoff, + unsigned int matchlen, + u_int16_t port) __read_mostly; +EXPORT_SYMBOL_GPL(nf_nat_sdp_port_hook); + unsigned int (*nf_nat_sdp_session_hook)(struct sk_buff *skb, const char **dptr, unsigned int dataoff, @@ -730,9 +738,10 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, union nf_inet_addr *saddr; struct nf_conntrack_tuple tuple; int family = ct->tuplehash[!dir].tuple.src.l3num; - int skip_expect = 0, ret = NF_DROP; + int direct_rtp = 0, skip_expect = 0, ret = NF_DROP; u_int16_t base_port; __be16 rtp_port, rtcp_port; + typeof(nf_nat_sdp_port_hook) nf_nat_sdp_port; typeof(nf_nat_sdp_media_hook) nf_nat_sdp_media; saddr = NULL; @@ -746,6 +755,14 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, * to register it since we can see the same media description multiple * times on different connections in case multiple endpoints receive * the same call. + * + * RTP optimization: if we find a matching media channel expectation + * and both the expectation and this connection are SNATed, we assume + * both sides can reach each other directly and use the final + * destination address from the expectation. We still need to keep + * the NATed expectations for media that might arrive from the + * outside, and additionally need to expect the direct RTP stream + * in case it passes through us even without NAT. */ memset(&tuple, 0, sizeof(tuple)); if (saddr) @@ -756,20 +773,42 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, tuple.dst.u.udp.port = port; rcu_read_lock(); - exp = __nf_ct_expect_find(&tuple); - if (exp && exp->master != ct && - nfct_help(exp->master)->helper == nfct_help(ct)->helper && - exp->class == class) - skip_expect = 1; - rcu_read_unlock(); + do { + exp = __nf_ct_expect_find(&tuple); - if (skip_expect) - return NF_ACCEPT; + if (!exp || exp->master == ct || + nfct_help(exp->master)->helper != nfct_help(ct)->helper || + exp->class != class) + break; + + if (exp->tuple.src.l3num == AF_INET && !direct_rtp && + (exp->saved_ip != exp->tuple.dst.u3.ip || + exp->saved_proto.udp.port != exp->tuple.dst.u.udp.port) && + ct->status & IPS_NAT_MASK) { + daddr->ip = exp->saved_ip; + tuple.dst.u3.ip = exp->saved_ip; + tuple.dst.u.udp.port = exp->saved_proto.udp.port; + direct_rtp = 1; + } else + skip_expect = 1; + } while (!skip_expect); + rcu_read_unlock(); base_port = ntohs(tuple.dst.u.udp.port) & ~1; rtp_port = htons(base_port); rtcp_port = htons(base_port + 1); + if (direct_rtp) { + nf_nat_sdp_port = rcu_dereference(nf_nat_sdp_port_hook); + if (nf_nat_sdp_port && + !nf_nat_sdp_port(skb, dptr, datalen, + mediaoff, medialen, ntohs(rtp_port))) + goto err1; + } + + if (skip_expect) + return NF_ACCEPT; + rtp_exp = nf_ct_expect_alloc(ct); if (rtp_exp == NULL) goto err1; @@ -783,7 +822,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, IPPROTO_UDP, NULL, &rtcp_port); nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook); - if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK) + if (nf_nat_sdp_media && ct->status & IPS_NAT_MASK && !direct_rtp) ret = nf_nat_sdp_media(skb, dptr, datalen, rtp_exp, rtcp_exp, mediaoff, medialen, daddr); else { -- cgit v1.2.3 From 9c2f5746b9cd536f0007709196d85a7e7d0070fa Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 26 Mar 2008 00:47:14 -0700 Subject: [NETNS]: Compilation fix for include/linux/netdevice.h. Commit commit c346dca10840a874240c78efe3f39acf4312a1f2 ([NET] NETNS: Omit net_device->nd_net without CONFIG_NET_NS) breaks compilation with CONFIG_NET_NS set. Fix the typo. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d146be40f46c..06ca84d71db4 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -756,7 +756,7 @@ static inline void dev_net_set(struct net_device *dev, const struct net *net) { #ifdef CONFIG_NET_NS - dev->nd_dev = net; + dev->nd_net = net; #endif } -- cgit v1.2.3 From f5aa23fd49063745f85644dd7a9330acd706add6 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 26 Mar 2008 00:48:17 -0700 Subject: [NETNS]: Compilation warnings under CONFIG_NET_NS. Recent commits from YOSHIFUJI Hideaki have been introduced a several compilation warnings 'assignment discards qualifiers from pointer target type' due to extra const modifier in the inline call parameters of {dev|sock|twsk}_net_set. Drop it. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 +- include/net/inet_timewait_sock.h | 2 +- include/net/sock.h | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 06ca84d71db4..15fa84a15c27 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -753,7 +753,7 @@ struct net *dev_net(const struct net_device *dev) } static inline -void dev_net_set(struct net_device *dev, const struct net *net) +void dev_net_set(struct net_device *dev, struct net *net) { #ifdef CONFIG_NET_NS dev->nd_net = net; diff --git a/include/net/inet_timewait_sock.h b/include/net/inet_timewait_sock.h index 07fe0d1a4f03..95c660c9719b 100644 --- a/include/net/inet_timewait_sock.h +++ b/include/net/inet_timewait_sock.h @@ -219,7 +219,7 @@ struct net *twsk_net(const struct inet_timewait_sock *twsk) } static inline -void twsk_net_set(struct inet_timewait_sock *twsk, const struct net *net) +void twsk_net_set(struct inet_timewait_sock *twsk, struct net *net) { #ifdef CONFIG_NET_NS twsk->tw_net = net; diff --git a/include/net/sock.h b/include/net/sock.h index 7e0d4a0c4d12..1c9d059223ee 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1358,7 +1358,7 @@ struct net *sock_net(const struct sock *sk) } static inline -void sock_net_set(struct sock *sk, const struct net *net) +void sock_net_set(struct sock *sk, struct net *net) { #ifdef CONFIG_NET_NS sk->sk_net = net; -- cgit v1.2.3 From a24022e1887978decaa28fb11d1ddff63e31497f Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 26 Mar 2008 01:55:37 -0700 Subject: [NETNS][ICMP]: Move ICMP sysctls on struct net. Initialization is moved to icmp_sk_init, all the places, that refer to them use init_net for now. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/icmp.h | 7 ------ include/net/netns/ipv4.h | 7 ++++++ net/ipv4/icmp.c | 60 ++++++++++++++++++++++++---------------------- net/ipv4/sysctl_net_ipv4.c | 12 +++++----- 4 files changed, 44 insertions(+), 42 deletions(-) (limited to 'include') diff --git a/include/net/icmp.h b/include/net/icmp.h index faba64db8ff0..dddb839ff4b5 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -65,11 +65,4 @@ static inline struct raw_sock *raw_sk(const struct sock *sk) return (struct raw_sock *)sk; } -extern int sysctl_icmp_echo_ignore_all; -extern int sysctl_icmp_echo_ignore_broadcasts; -extern int sysctl_icmp_ignore_bogus_error_responses; -extern int sysctl_icmp_errors_use_inbound_ifaddr; -extern int sysctl_icmp_ratelimit; -extern int sysctl_icmp_ratemask; - #endif /* _ICMP_H */ diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 504fde174525..e3de0ff8ea4b 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -35,5 +35,12 @@ struct netns_ipv4 { struct xt_table *iptable_raw; struct xt_table *arptable_filter; #endif + + int sysctl_icmp_echo_ignore_all; + int sysctl_icmp_echo_ignore_broadcasts; + int sysctl_icmp_ignore_bogus_error_responses; + int sysctl_icmp_ratelimit; + int sysctl_icmp_ratemask; + int sysctl_icmp_errors_use_inbound_ifaddr; }; #endif diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index f38f093ef751..958a38445120 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -188,29 +188,6 @@ struct icmp_err icmp_err_convert[] = { }, }; -/* Control parameters for ECHO replies. */ -int sysctl_icmp_echo_ignore_all __read_mostly; -int sysctl_icmp_echo_ignore_broadcasts __read_mostly = 1; - -/* Control parameter - ignore bogus broadcast responses? */ -int sysctl_icmp_ignore_bogus_error_responses __read_mostly = 1; - -/* - * Configurable global rate limit. - * - * ratelimit defines tokens/packet consumed for dst->rate_token bucket - * ratemask defines which icmp types are ratelimited by setting - * it's bit position. - * - * default: - * dest unreachable (3), source quench (4), - * time exceeded (11), parameter problem (12) - */ - -int sysctl_icmp_ratelimit __read_mostly = 1 * HZ; -int sysctl_icmp_ratemask __read_mostly = 0x1818; -int sysctl_icmp_errors_use_inbound_ifaddr __read_mostly; - /* * ICMP control array. This specifies what to do with each ICMP. */ @@ -310,8 +287,8 @@ static inline int icmpv4_xrlim_allow(struct rtable *rt, int type, int code) goto out; /* Limit if icmp type is enabled in ratemask. */ - if ((1 << type) & sysctl_icmp_ratemask) - rc = xrlim_allow(dst, sysctl_icmp_ratelimit); + if ((1 << type) & init_net.ipv4.sysctl_icmp_ratemask) + rc = xrlim_allow(dst, init_net.ipv4.sysctl_icmp_ratelimit); out: return rc; } @@ -523,7 +500,8 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) if (!(rt->rt_flags & RTCF_LOCAL)) { struct net_device *dev = NULL; - if (rt->fl.iif && sysctl_icmp_errors_use_inbound_ifaddr) + if (rt->fl.iif && + init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr) dev = dev_get_by_index(net, rt->fl.iif); if (dev) { @@ -745,7 +723,7 @@ static void icmp_unreach(struct sk_buff *skb) * get the other vendor to fix their kit. */ - if (!sysctl_icmp_ignore_bogus_error_responses && + if (!init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses && inet_addr_type(net, iph->daddr) == RTN_BROADCAST) { if (net_ratelimit()) printk(KERN_WARNING "%u.%u.%u.%u sent an invalid ICMP " @@ -840,7 +818,7 @@ out_err: static void icmp_echo(struct sk_buff *skb) { - if (!sysctl_icmp_echo_ignore_all) { + if (!init_net.ipv4.sysctl_icmp_echo_ignore_all) { struct icmp_bxm icmp_param; icmp_param.data.icmph = *icmp_hdr(skb); @@ -1051,7 +1029,7 @@ int icmp_rcv(struct sk_buff *skb) */ if ((icmph->type == ICMP_ECHO || icmph->type == ICMP_TIMESTAMP) && - sysctl_icmp_echo_ignore_broadcasts) { + init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts) { goto error; } if (icmph->type != ICMP_ECHO && @@ -1195,6 +1173,30 @@ int __net_init icmp_sk_init(struct net *net) */ sk->sk_prot->unhash(sk); } + + /* Control parameters for ECHO replies. */ + net->ipv4.sysctl_icmp_echo_ignore_all = 0; + net->ipv4.sysctl_icmp_echo_ignore_broadcasts = 1; + + /* Control parameter - ignore bogus broadcast responses? */ + net->ipv4.sysctl_icmp_ignore_bogus_error_responses = 1; + + /* + * Configurable global rate limit. + * + * ratelimit defines tokens/packet consumed for dst->rate_token + * bucket ratemask defines which icmp types are ratelimited by + * setting it's bit position. + * + * default: + * dest unreachable (3), source quench (4), + * time exceeded (11), parameter problem (12) + */ + + net->ipv4.sysctl_icmp_ratelimit = 1 * HZ; + net->ipv4.sysctl_icmp_ratemask = 0x1818; + net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr = 0; + return 0; fail: diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index c2fca3024d2e..e9585c02f243 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -406,7 +406,7 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_ICMP_ECHO_IGNORE_ALL, .procname = "icmp_echo_ignore_all", - .data = &sysctl_icmp_echo_ignore_all, + .data = &init_net.ipv4.sysctl_icmp_echo_ignore_all, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec @@ -414,7 +414,7 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS, .procname = "icmp_echo_ignore_broadcasts", - .data = &sysctl_icmp_echo_ignore_broadcasts, + .data = &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec @@ -422,7 +422,7 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES, .procname = "icmp_ignore_bogus_error_responses", - .data = &sysctl_icmp_ignore_bogus_error_responses, + .data = &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec @@ -430,7 +430,7 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, .procname = "icmp_errors_use_inbound_ifaddr", - .data = &sysctl_icmp_errors_use_inbound_ifaddr, + .data = &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec @@ -588,7 +588,7 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_ICMP_RATELIMIT, .procname = "icmp_ratelimit", - .data = &sysctl_icmp_ratelimit, + .data = &init_net.ipv4.sysctl_icmp_ratelimit, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec @@ -596,7 +596,7 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = NET_IPV4_ICMP_RATEMASK, .procname = "icmp_ratemask", - .data = &sysctl_icmp_ratemask, + .data = &init_net.ipv4.sysctl_icmp_ratemask, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec -- cgit v1.2.3 From 68528f09980a60c8df046d16336333cac4fc0c32 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 26 Mar 2008 01:56:24 -0700 Subject: [NETNS][ICMP]: Make ctl tables for ICMP sysctls per-net. Add some flesh to ipv4_sysctl_init_net and ipv4_sysctl_exit_net, i.e. copy the table, alter .data pointers and register it per-net. Other ipv4_table's sysctls are now global, but this is going to change once sysctl permissions patches migrate from -mm tree to mainline in 2.6.26 merge window :) Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + net/ipv4/sysctl_net_ipv4.c | 138 +++++++++++++++++++++++++++++---------------- 2 files changed, 91 insertions(+), 48 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index e3de0ff8ea4b..af685f71f4b5 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -17,6 +17,7 @@ struct netns_ipv4 { #ifdef CONFIG_SYSCTL struct ctl_table_header *forw_hdr; struct ctl_table_header *frags_hdr; + struct ctl_table_header *ipv4_hdr; #endif struct ipv4_devconf *devconf_all; struct ipv4_devconf *devconf_dflt; diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index e9585c02f243..c437f804ee38 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -403,38 +403,6 @@ static struct ctl_table ipv4_table[] = { .proc_handler = &ipv4_local_port_range, .strategy = &ipv4_sysctl_local_port_range, }, - { - .ctl_name = NET_IPV4_ICMP_ECHO_IGNORE_ALL, - .procname = "icmp_echo_ignore_all", - .data = &init_net.ipv4.sysctl_icmp_echo_ignore_all, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS, - .procname = "icmp_echo_ignore_broadcasts", - .data = &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES, - .procname = "icmp_ignore_bogus_error_responses", - .data = &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, - .procname = "icmp_errors_use_inbound_ifaddr", - .data = &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, { .ctl_name = NET_IPV4_ROUTE, .procname = "route", @@ -585,22 +553,6 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = &proc_dointvec }, - { - .ctl_name = NET_IPV4_ICMP_RATELIMIT, - .procname = "icmp_ratelimit", - .data = &init_net.ipv4.sysctl_icmp_ratelimit, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, - { - .ctl_name = NET_IPV4_ICMP_RATEMASK, - .procname = "icmp_ratemask", - .data = &init_net.ipv4.sysctl_icmp_ratemask, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = &proc_dointvec - }, { .ctl_name = NET_TCP_TW_REUSE, .procname = "tcp_tw_reuse", @@ -804,6 +756,58 @@ static struct ctl_table ipv4_table[] = { { .ctl_name = 0 } }; +static struct ctl_table ipv4_net_table[] = { + { + .ctl_name = NET_IPV4_ICMP_ECHO_IGNORE_ALL, + .procname = "icmp_echo_ignore_all", + .data = &init_net.ipv4.sysctl_icmp_echo_ignore_all, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_IPV4_ICMP_ECHO_IGNORE_BROADCASTS, + .procname = "icmp_echo_ignore_broadcasts", + .data = &init_net.ipv4.sysctl_icmp_echo_ignore_broadcasts, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_IPV4_ICMP_IGNORE_BOGUS_ERROR_RESPONSES, + .procname = "icmp_ignore_bogus_error_responses", + .data = &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_IPV4_ICMP_ERRORS_USE_INBOUND_IFADDR, + .procname = "icmp_errors_use_inbound_ifaddr", + .data = &init_net.ipv4.sysctl_icmp_errors_use_inbound_ifaddr, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_IPV4_ICMP_RATELIMIT, + .procname = "icmp_ratelimit", + .data = &init_net.ipv4.sysctl_icmp_ratelimit, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { + .ctl_name = NET_IPV4_ICMP_RATEMASK, + .procname = "icmp_ratemask", + .data = &init_net.ipv4.sysctl_icmp_ratemask, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { } +}; + struct ctl_path net_ipv4_ctl_path[] = { { .procname = "net", .ctl_name = CTL_NET, }, { .procname = "ipv4", .ctl_name = NET_IPV4, }, @@ -813,11 +817,49 @@ EXPORT_SYMBOL_GPL(net_ipv4_ctl_path); static __net_init int ipv4_sysctl_init_net(struct net *net) { + struct ctl_table *table; + + table = ipv4_net_table; + if (net != &init_net) { + table = kmemdup(table, sizeof(ipv4_net_table), GFP_KERNEL); + if (table == NULL) + goto err_alloc; + + table[0].data = + &net->ipv4.sysctl_icmp_echo_ignore_all; + table[1].data = + &net->ipv4.sysctl_icmp_echo_ignore_broadcasts; + table[2].data = + &net->ipv4.sysctl_icmp_ignore_bogus_error_responses; + table[3].data = + &net->ipv4.sysctl_icmp_errors_use_inbound_ifaddr; + table[4].data = + &net->ipv4.sysctl_icmp_ratelimit; + table[5].data = + &net->ipv4.sysctl_icmp_ratemask; + } + + net->ipv4.ipv4_hdr = register_net_sysctl_table(net, + net_ipv4_ctl_path, table); + if (net->ipv4.ipv4_hdr == NULL) + goto err_reg; + return 0; + +err_reg: + if (net != &init_net) + kfree(table); +err_alloc: + return -ENOMEM; } static __net_exit void ipv4_sysctl_exit_net(struct net *net) { + struct ctl_table *table; + + table = net->ipv4.ipv4_hdr->ctl_table_arg; + unregister_net_sysctl_table(net->ipv4.ipv4_hdr); + kfree(table); } static __net_initdata struct pernet_operations ipv4_sysctl_ops = { -- cgit v1.2.3 From 67727184f28c38d06013c6659560bb046c1d9f9c Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 26 Mar 2008 16:27:22 -0700 Subject: [VLAN]: Reduce memory consumed by vlan_groups Currently each vlan_groupd contains 8 pointers on arrays with 512 pointers on struct net_device each :) Such a construction "in many cases ... wastes memory". My proposal is to allow for some of these arrays pointers be NULL, meaning that there are no devices in it. When a new device is added to the vlan_group, the appropriate array is allocated. The check in vlan_group_get_device's is safe, since the pointer vg->vlan_devices_arrays[x] can only switch from NULL to not-NULL. The vlan_group_prealloc_vid() is guarded with rtnl lock and is also safe. I've checked (I hope that) all the places, that use these arrays and found, that the register_vlan_dev is the only place, that can put a vlan device on an empty vlan_group. Rough calculations shows, that after the patch a setup with a single vlan dev (or up to 512 vlans with sequential vids) will occupy approximately 8 times less memory. The question I have is - does this patch makes sense, or a totally new structures are required to store the vlan_devs? Signed-off-by: Pavel Emelyanov Signed-off-by: Patrick McHardy --- include/linux/if_vlan.h | 2 +- net/8021q/vlan.c | 36 +++++++++++++++++++++++------------- 2 files changed, 24 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index 79504b22a932..edd55af7ebd6 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -93,7 +93,7 @@ static inline struct net_device *vlan_group_get_device(struct vlan_group *vg, { struct net_device **array; array = vg->vlan_devices_arrays[vlan_id / VLAN_GROUP_ARRAY_PART_LEN]; - return array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN]; + return array ? array[vlan_id % VLAN_GROUP_ARRAY_PART_LEN] : NULL; } static inline void vlan_group_set_device(struct vlan_group *vg, diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index c35dc230365c..694be86e4490 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -106,29 +106,35 @@ static void vlan_group_free(struct vlan_group *grp) static struct vlan_group *vlan_group_alloc(int ifindex) { struct vlan_group *grp; - unsigned int size; - unsigned int i; grp = kzalloc(sizeof(struct vlan_group), GFP_KERNEL); if (!grp) return NULL; - size = sizeof(struct net_device *) * VLAN_GROUP_ARRAY_PART_LEN; - - for (i = 0; i < VLAN_GROUP_ARRAY_SPLIT_PARTS; i++) { - grp->vlan_devices_arrays[i] = kzalloc(size, GFP_KERNEL); - if (!grp->vlan_devices_arrays[i]) - goto err; - } - grp->real_dev_ifindex = ifindex; hlist_add_head_rcu(&grp->hlist, &vlan_group_hash[vlan_grp_hashfn(ifindex)]); return grp; +} -err: - vlan_group_free(grp); - return NULL; +static int vlan_group_prealloc_vid(struct vlan_group *vg, int vid) +{ + struct net_device **array; + unsigned int size; + + ASSERT_RTNL(); + + array = vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN]; + if (array != NULL) + return 0; + + size = sizeof(struct net_device *) * VLAN_GROUP_ARRAY_PART_LEN; + array = kzalloc(size, GFP_KERNEL); + if (array == NULL) + return -ENOBUFS; + + vg->vlan_devices_arrays[vid / VLAN_GROUP_ARRAY_PART_LEN] = array; + return 0; } static void vlan_rcu_free(struct rcu_head *rcu) @@ -247,6 +253,10 @@ int register_vlan_dev(struct net_device *dev) return -ENOBUFS; } + err = vlan_group_prealloc_vid(grp, vlan_id); + if (err < 0) + goto out_free_group; + err = register_netdevice(dev); if (err < 0) goto out_free_group; -- cgit v1.2.3 From 6ab57e7e7fa316552d0f94eaebf1def1d49f18da Mon Sep 17 00:00:00 2001 From: Daniel Lezcano Date: Wed, 26 Mar 2008 16:52:32 -0700 Subject: [NETNS][IPV6] anycast - handle several network namespace Make use of the network namespace information to have this protocol to handle several network namespace. Signed-off-by: Daniel Lezcano Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- include/net/addrconf.h | 3 ++- include/net/ipv6.h | 17 ++++++----------- net/ipv6/af_inet6.c | 12 ++++++------ net/ipv6/anycast.c | 38 ++++++++++++++++++++++---------------- net/ipv6/ndisc.c | 2 +- 5 files changed, 37 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index c9276c72764d..d0c47c306046 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -134,7 +134,8 @@ extern int inet6_ac_check(struct sock *sk, struct in6_addr *addr, int ifindex); extern int ipv6_dev_ac_inc(struct net_device *dev, struct in6_addr *addr); extern int __ipv6_dev_ac_dec(struct inet6_dev *idev, struct in6_addr *addr); -extern int ipv6_chk_acast_addr(struct net_device *dev, struct in6_addr *addr); +extern int ipv6_chk_acast_addr(struct net *net, struct net_device *dev, + struct in6_addr *addr); /* Device notifier */ diff --git a/include/net/ipv6.h b/include/net/ipv6.h index e82f1814d96b..1c98e737dbd0 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -591,8 +591,8 @@ extern int ip6_mc_msfget(struct sock *sk, struct group_filter *gsf, int __user *optlen); #ifdef CONFIG_PROC_FS -extern int ac6_proc_init(void); -extern void ac6_proc_exit(void); +extern int ac6_proc_init(struct net *net); +extern void ac6_proc_exit(struct net *net); extern int raw6_proc_init(void); extern void raw6_proc_exit(void); extern int tcp6_proc_init(struct net *net); @@ -607,15 +607,10 @@ extern int snmp6_register_dev(struct inet6_dev *idev); extern int snmp6_unregister_dev(struct inet6_dev *idev); #else -static inline int snmp6_register_dev(struct inet6_dev *idev) -{ - return 0; -} - -static inline int snmp6_unregister_dev(struct inet6_dev *idev) -{ - return 0; -} +static inline int ac6_proc_init(struct net *net) { return 0; } +static inline void ac6_proc_exit(struct net *net) { } +static inline int snmp6_register_dev(struct inet6_dev *idev) { return 0; } +static inline int snmp6_unregister_dev(struct inet6_dev *idev) { return 0; } #endif #ifdef CONFIG_SYSCTL diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 12f04e9d3e88..1731b0abf7f5 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -862,11 +862,16 @@ static int inet6_net_init(struct net *net) err = tcp6_proc_init(net); if (err) goto proc_tcp6_fail; + err = ac6_proc_init(net); + if (err) + goto proc_ac6_fail; out: #endif return err; #ifdef CONFIG_PROC_FS +proc_ac6_fail: + tcp6_proc_exit(net); proc_tcp6_fail: udp6_proc_exit(net); goto out; @@ -878,6 +883,7 @@ static void inet6_net_exit(struct net *net) #ifdef CONFIG_PROC_FS udp6_proc_exit(net); tcp6_proc_exit(net); + ac6_proc_exit(net); #endif } @@ -965,9 +971,6 @@ static int __init inet6_init(void) goto proc_udplite6_fail; if (ipv6_misc_proc_init()) goto proc_misc6_fail; - - if (ac6_proc_init()) - goto proc_anycast6_fail; if (if6_proc_init()) goto proc_if6_fail; #endif @@ -1039,8 +1042,6 @@ ip6_route_fail: #ifdef CONFIG_PROC_FS if6_proc_exit(); proc_if6_fail: - ac6_proc_exit(); -proc_anycast6_fail: ipv6_misc_proc_exit(); proc_misc6_fail: udplite6_proc_exit(); @@ -1101,7 +1102,6 @@ static void __exit inet6_exit(void) /* Cleanup code parts. */ if6_proc_exit(); - ac6_proc_exit(); ipv6_misc_proc_exit(); udplite6_proc_exit(); raw6_proc_exit(); diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 96868b994b37..463bd95d6b13 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -82,6 +82,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr) struct net_device *dev = NULL; struct inet6_dev *idev; struct ipv6_ac_socklist *pac; + struct net *net = sock_net(sk); int ishost = !ipv6_devconf.forwarding; int err = 0; @@ -89,7 +90,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr) return -EPERM; if (ipv6_addr_is_multicast(addr)) return -EINVAL; - if (ipv6_chk_addr(&init_net, addr, NULL, 0)) + if (ipv6_chk_addr(net, addr, NULL, 0)) return -EINVAL; pac = sock_kmalloc(sk, sizeof(struct ipv6_ac_socklist), GFP_KERNEL); @@ -101,7 +102,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr) if (ifindex == 0) { struct rt6_info *rt; - rt = rt6_lookup(&init_net, addr, NULL, 0, 0); + rt = rt6_lookup(net, addr, NULL, 0, 0); if (rt) { dev = rt->rt6i_dev; dev_hold(dev); @@ -112,10 +113,10 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr) } else { /* router, no matching interface: just pick one */ - dev = dev_get_by_flags(&init_net, IFF_UP, IFF_UP|IFF_LOOPBACK); + dev = dev_get_by_flags(net, IFF_UP, IFF_UP|IFF_LOOPBACK); } } else - dev = dev_get_by_index(&init_net, ifindex); + dev = dev_get_by_index(net, ifindex); if (dev == NULL) { err = -ENODEV; @@ -176,6 +177,7 @@ int ipv6_sock_ac_drop(struct sock *sk, int ifindex, struct in6_addr *addr) struct ipv6_pinfo *np = inet6_sk(sk); struct net_device *dev; struct ipv6_ac_socklist *pac, *prev_pac; + struct net *net = sock_net(sk); write_lock_bh(&ipv6_sk_ac_lock); prev_pac = NULL; @@ -196,7 +198,7 @@ int ipv6_sock_ac_drop(struct sock *sk, int ifindex, struct in6_addr *addr) write_unlock_bh(&ipv6_sk_ac_lock); - dev = dev_get_by_index(&init_net, pac->acl_ifindex); + dev = dev_get_by_index(net, pac->acl_ifindex); if (dev) { ipv6_dev_ac_dec(dev, &pac->acl_addr); dev_put(dev); @@ -210,6 +212,7 @@ void ipv6_sock_ac_close(struct sock *sk) struct ipv6_pinfo *np = inet6_sk(sk); struct net_device *dev = NULL; struct ipv6_ac_socklist *pac; + struct net *net = sock_net(sk); int prev_index; write_lock_bh(&ipv6_sk_ac_lock); @@ -224,7 +227,7 @@ void ipv6_sock_ac_close(struct sock *sk) if (pac->acl_ifindex != prev_index) { if (dev) dev_put(dev); - dev = dev_get_by_index(&init_net, pac->acl_ifindex); + dev = dev_get_by_index(net, pac->acl_ifindex); prev_index = pac->acl_ifindex; } if (dev) @@ -422,14 +425,15 @@ static int ipv6_chk_acast_dev(struct net_device *dev, struct in6_addr *addr) /* * check if given interface (or any, if dev==0) has this anycast address */ -int ipv6_chk_acast_addr(struct net_device *dev, struct in6_addr *addr) +int ipv6_chk_acast_addr(struct net *net, struct net_device *dev, + struct in6_addr *addr) { int found = 0; if (dev) return ipv6_chk_acast_dev(dev, addr); read_lock(&dev_base_lock); - for_each_netdev(&init_net, dev) + for_each_netdev(net, dev) if (ipv6_chk_acast_dev(dev, addr)) { found = 1; break; @@ -441,6 +445,7 @@ int ipv6_chk_acast_addr(struct net_device *dev, struct in6_addr *addr) #ifdef CONFIG_PROC_FS struct ac6_iter_state { + struct seq_net_private p; struct net_device *dev; struct inet6_dev *idev; }; @@ -451,9 +456,10 @@ static inline struct ifacaddr6 *ac6_get_first(struct seq_file *seq) { struct ifacaddr6 *im = NULL; struct ac6_iter_state *state = ac6_seq_private(seq); + struct net *net = seq_file_net(seq); state->idev = NULL; - for_each_netdev(&init_net, state->dev) { + for_each_netdev(net, state->dev) { struct inet6_dev *idev; idev = in6_dev_get(state->dev); if (!idev) @@ -551,8 +557,8 @@ static const struct seq_operations ac6_seq_ops = { static int ac6_seq_open(struct inode *inode, struct file *file) { - return seq_open_private(file, &ac6_seq_ops, - sizeof(struct ac6_iter_state)); + return seq_open_net(inode, file, &ac6_seq_ops, + sizeof(struct ac6_iter_state)); } static const struct file_operations ac6_seq_fops = { @@ -560,20 +566,20 @@ static const struct file_operations ac6_seq_fops = { .open = ac6_seq_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release_private, + .release = seq_release_net, }; -int __init ac6_proc_init(void) +int ac6_proc_init(struct net *net) { - if (!proc_net_fops_create(&init_net, "anycast6", S_IRUGO, &ac6_seq_fops)) + if (!proc_net_fops_create(net, "anycast6", S_IRUGO, &ac6_seq_fops)) return -ENOMEM; return 0; } -void ac6_proc_exit(void) +void ac6_proc_exit(struct net *net) { - proc_net_remove(&init_net, "anycast6"); + proc_net_remove(net, "anycast6"); } #endif diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 79af57f586e8..b4d8e331432e 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -773,7 +773,7 @@ static void ndisc_recv_ns(struct sk_buff *skb) return; } - if (ipv6_chk_acast_addr(dev, &msg->target) || + if (ipv6_chk_acast_addr(dev_net(dev), dev, &msg->target) || (idev->cnf.forwarding && (ipv6_devconf.proxy_ndp || idev->cnf.proxy_ndp) && (pneigh = pneigh_lookup(&nd_tbl, dev_net(dev), -- cgit v1.2.3 From 60e8fbc4c53d3ef0cbffa393a9e7b77e2a1bae58 Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Wed, 26 Mar 2008 16:53:08 -0700 Subject: [NETNS][IPV6] flowlabels - make flowlabels per namespace This patch introduces a new member, fl_net, in struct ip6_flowlabel. This allows to create labels with the same value in different namespaces. Signed-off-by: Benjamin Thery Signed-off-by: Daniel Lezcano Signed-off-by: David S. Miller --- include/net/ipv6.h | 1 + net/ipv6/ip6_flowlabel.c | 72 +++++++++++++++++++++++++++++++++++++----------- 2 files changed, 57 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 1c98e737dbd0..296f61d84709 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -202,6 +202,7 @@ struct ip6_flowlabel u32 owner; unsigned long lastuse; unsigned long expires; + struct net *fl_net; }; #define IPV6_FLOWINFO_MASK __constant_htonl(0x0FFFFFFF) diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c index 2b7d9ee98832..78d1d913e36c 100644 --- a/net/ipv6/ip6_flowlabel.c +++ b/net/ipv6/ip6_flowlabel.c @@ -62,23 +62,23 @@ static DEFINE_RWLOCK(ip6_fl_lock); static DEFINE_RWLOCK(ip6_sk_fl_lock); -static __inline__ struct ip6_flowlabel * __fl_lookup(__be32 label) +static inline struct ip6_flowlabel *__fl_lookup(struct net *net, __be32 label) { struct ip6_flowlabel *fl; for (fl=fl_ht[FL_HASH(label)]; fl; fl = fl->next) { - if (fl->label == label) + if (fl->label == label && fl->fl_net == net) return fl; } return NULL; } -static struct ip6_flowlabel * fl_lookup(__be32 label) +static struct ip6_flowlabel *fl_lookup(struct net *net, __be32 label) { struct ip6_flowlabel *fl; read_lock_bh(&ip6_fl_lock); - fl = __fl_lookup(label); + fl = __fl_lookup(net, label); if (fl) atomic_inc(&fl->users); read_unlock_bh(&ip6_fl_lock); @@ -88,8 +88,10 @@ static struct ip6_flowlabel * fl_lookup(__be32 label) static void fl_free(struct ip6_flowlabel *fl) { - if (fl) + if (fl) { + release_net(fl->fl_net); kfree(fl->opt); + } kfree(fl); } @@ -112,7 +114,6 @@ static void fl_release(struct ip6_flowlabel *fl) time_after(ip6_fl_gc_timer.expires, ttd)) mod_timer(&ip6_fl_gc_timer, ttd); } - write_unlock_bh(&ip6_fl_lock); } @@ -148,13 +149,34 @@ static void ip6_fl_gc(unsigned long dummy) if (!sched && atomic_read(&fl_size)) sched = now + FL_MAX_LINGER; if (sched) { - ip6_fl_gc_timer.expires = sched; - add_timer(&ip6_fl_gc_timer); + mod_timer(&ip6_fl_gc_timer, sched); } write_unlock(&ip6_fl_lock); } -static struct ip6_flowlabel *fl_intern(struct ip6_flowlabel *fl, __be32 label) +static void ip6_fl_purge(struct net *net) +{ + int i; + + write_lock(&ip6_fl_lock); + for (i = 0; i <= FL_HASH_MASK; i++) { + struct ip6_flowlabel *fl, **flp; + flp = &fl_ht[i]; + while ((fl = *flp) != NULL) { + if (fl->fl_net == net && atomic_read(&fl->users) == 0) { + *flp = fl->next; + fl_free(fl); + atomic_dec(&fl_size); + continue; + } + flp = &fl->next; + } + } + write_unlock(&ip6_fl_lock); +} + +static struct ip6_flowlabel *fl_intern(struct net *net, + struct ip6_flowlabel *fl, __be32 label) { struct ip6_flowlabel *lfl; @@ -165,7 +187,7 @@ static struct ip6_flowlabel *fl_intern(struct ip6_flowlabel *fl, __be32 label) for (;;) { fl->label = htonl(net_random())&IPV6_FLOWLABEL_MASK; if (fl->label) { - lfl = __fl_lookup(fl->label); + lfl = __fl_lookup(net, fl->label); if (lfl == NULL) break; } @@ -179,7 +201,7 @@ static struct ip6_flowlabel *fl_intern(struct ip6_flowlabel *fl, __be32 label) * done in ipv6_flowlabel_opt - sock is locked, so new entry * with the same label can only appear on another sock */ - lfl = __fl_lookup(fl->label); + lfl = __fl_lookup(net, fl->label); if (lfl != NULL) { atomic_inc(&lfl->users); write_unlock_bh(&ip6_fl_lock); @@ -298,7 +320,8 @@ static int fl6_renew(struct ip6_flowlabel *fl, unsigned long linger, unsigned lo } static struct ip6_flowlabel * -fl_create(struct in6_flowlabel_req *freq, char __user *optval, int optlen, int *err_p) +fl_create(struct net *net, struct in6_flowlabel_req *freq, char __user *optval, + int optlen, int *err_p) { struct ip6_flowlabel *fl; int olen; @@ -343,6 +366,7 @@ fl_create(struct in6_flowlabel_req *freq, char __user *optval, int optlen, int * } } + fl->fl_net = hold_net(net); fl->expires = jiffies; err = fl6_renew(fl, freq->flr_linger, freq->flr_expires); if (err) @@ -441,6 +465,7 @@ static inline void fl_link(struct ipv6_pinfo *np, struct ipv6_fl_socklist *sfl, int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen) { int err; + struct net *net = sock_net(sk); struct ipv6_pinfo *np = inet6_sk(sk); struct in6_flowlabel_req freq; struct ipv6_fl_socklist *sfl1=NULL; @@ -483,7 +508,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen) read_unlock_bh(&ip6_sk_fl_lock); if (freq.flr_share == IPV6_FL_S_NONE && capable(CAP_NET_ADMIN)) { - fl = fl_lookup(freq.flr_label); + fl = fl_lookup(net, freq.flr_label); if (fl) { err = fl6_renew(fl, freq.flr_linger, freq.flr_expires); fl_release(fl); @@ -496,7 +521,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen) if (freq.flr_label & ~IPV6_FLOWLABEL_MASK) return -EINVAL; - fl = fl_create(&freq, optval, optlen, &err); + fl = fl_create(net, &freq, optval, optlen, &err); if (fl == NULL) return err; sfl1 = kmalloc(sizeof(*sfl1), GFP_KERNEL); @@ -518,7 +543,7 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen) read_unlock_bh(&ip6_sk_fl_lock); if (fl1 == NULL) - fl1 = fl_lookup(freq.flr_label); + fl1 = fl_lookup(net, freq.flr_label); if (fl1) { recheck: err = -EEXIST; @@ -559,7 +584,7 @@ release: if (sfl1 == NULL || (err = mem_check(sk)) != 0) goto done; - fl1 = fl_intern(fl, freq.flr_label); + fl1 = fl_intern(net, fl, freq.flr_label); if (fl1 != NULL) goto recheck; @@ -717,13 +742,28 @@ static inline void ip6_flowlabel_proc_fini(struct net *net) } #endif +static inline void ip6_flowlabel_net_exit(struct net *net) +{ + ip6_fl_purge(net); +} + +static struct pernet_operations ip6_flowlabel_net_ops = { + .exit = ip6_flowlabel_net_exit, +}; + int ip6_flowlabel_init(void) { + int err; + + err = register_pernet_subsys(&ip6_flowlabel_net_ops); + if (err) + return err; return ip6_flowlabel_proc_init(&init_net); } void ip6_flowlabel_cleanup(void) { del_timer(&ip6_fl_gc_timer); + unregister_pernet_subsys(&ip6_flowlabel_net_ops); ip6_flowlabel_proc_fini(&init_net); } -- cgit v1.2.3 From 6c507cd0400cb51dd2ee251c1b8756b9375a1128 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 26 Mar 2008 14:14:55 +0100 Subject: cfg80211: don't export ieee80211_get_channel This patch makes ieee80211_get_channel a static inline defined in cfg80211's header file which simply calls __ieee80211_get_channel to avoid symbol clashes with the ieee80211 code. The problem was pointed out by David Miller, thanks! Signed-off-by: Johannes Berg Cc: David Miller Signed-off-by: John W. Linville --- include/net/wireless.h | 18 +++++++++++++++--- net/wireless/util.c | 6 +++--- 2 files changed, 18 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/wireless.h b/include/net/wireless.h index f4b77ab66bae..667b4080d30f 100644 --- a/include/net/wireless.h +++ b/include/net/wireless.h @@ -304,10 +304,22 @@ extern int ieee80211_channel_to_frequency(int chan); */ extern int ieee80211_frequency_to_channel(int freq); +/* + * Name indirection necessary because the ieee80211 code also has + * a function named "ieee80211_get_channel", so if you include + * cfg80211's header file you get cfg80211's version, if you try + * to include both header files you'll (rightfully!) get a symbol + * clash. + */ +extern struct ieee80211_channel *__ieee80211_get_channel(struct wiphy *wiphy, + int freq); + /** * ieee80211_get_channel - get channel struct from wiphy for specified frequency */ -extern struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy, - int freq); - +static inline struct ieee80211_channel * +ieee80211_get_channel(struct wiphy *wiphy, int freq) +{ + return __ieee80211_get_channel(wiphy, freq); +} #endif /* __NET_WIRELESS_H */ diff --git a/net/wireless/util.c b/net/wireless/util.c index f3e623df3515..f54424693a38 100644 --- a/net/wireless/util.c +++ b/net/wireless/util.c @@ -33,8 +33,8 @@ int ieee80211_frequency_to_channel(int freq) } EXPORT_SYMBOL(ieee80211_frequency_to_channel); -struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy, - int freq) +struct ieee80211_channel *__ieee80211_get_channel(struct wiphy *wiphy, + int freq) { enum ieee80211_band band; struct ieee80211_supported_band *sband; @@ -54,7 +54,7 @@ struct ieee80211_channel *ieee80211_get_channel(struct wiphy *wiphy, return NULL; } -EXPORT_SYMBOL(ieee80211_get_channel); +EXPORT_SYMBOL(__ieee80211_get_channel); static void set_mandatory_flags_band(struct ieee80211_supported_band *sband, enum ieee80211_band band) -- cgit v1.2.3 From 0e5f8be1388093edc324a78ebf241170b258eba3 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 27 Mar 2008 14:25:53 -0700 Subject: [NETNS]: Compile NET /proc support only if CONFIG_NET is set. This fix broken compilation for 'allnoconfig'. This was introduced by Introduced by commit 1218854afa6f659be90b748cf1bc7badee954a35 ("[NET] NETNS: Omit seq_net_private->net without CONFIG_NET_NS.") Signed-off-by: Denis V. Lunev Acked-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- fs/proc/proc_net.c | 2 ++ include/linux/seq_file.h | 2 ++ 2 files changed, 4 insertions(+) (limited to 'include') diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 13cd7835d0df..7034facf8b8f 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -51,6 +51,7 @@ int seq_open_net(struct inode *ino, struct file *f, } EXPORT_SYMBOL_GPL(seq_open_net); +#ifdef CONFIG_NET int seq_release_net(struct inode *ino, struct file *f) { struct seq_file *seq; @@ -218,3 +219,4 @@ int __init proc_net_init(void) return register_pernet_subsys(&proc_net_ns_ops); } +#endif /* CONFIG_NET */ diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index d870a8253769..5da70c3f4417 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -63,6 +63,7 @@ extern struct list_head *seq_list_start_head(struct list_head *head, extern struct list_head *seq_list_next(void *v, struct list_head *head, loff_t *ppos); +#ifdef CONFIG_NET struct net; struct seq_net_private { #ifdef CONFIG_NET_NS @@ -81,6 +82,7 @@ static inline struct net *seq_file_net(struct seq_file *seq) return &init_net; #endif } +#endif /* CONFIG_NET */ #endif #endif -- cgit v1.2.3 From 09382bac667821017144de6b733a26f29a1c185b Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 27 Mar 2008 16:53:37 -0700 Subject: [PKT_SCHED]: Pass real namespace in net scheduler classifiers. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/pkt_cls.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index d349c66ef828..aa9e282db485 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -353,7 +353,7 @@ tcf_match_indev(struct sk_buff *skb, char *indev) if (indev[0]) { if (!skb->iif) return 0; - dev = __dev_get_by_index(&init_net, skb->iif); + dev = __dev_get_by_index(dev_net(skb->dev), skb->iif); if (!dev || strcmp(indev, dev->name)) return 0; } -- cgit v1.2.3 From 4f95165d4bf6369d57052d60cc0266c569c6b077 Mon Sep 17 00:00:00 2001 From: Rami Rosen Date: Thu, 27 Mar 2008 17:39:19 -0700 Subject: [IPV6]: Remove three unused method declarations in include/net/ipv6.h This patch removes three unused method declarations in include/net/ipv6.h: inet_getfrag_t(), ipv6_build_nfrag_opts() and ipv6_build_frag_opts(). Signed-off-by: Rami Rosen Signed-off-by: David S. Miller --- include/net/ipv6.h | 17 ----------------- 1 file changed, 17 deletions(-) (limited to 'include') diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 296f61d84709..5738c1c73ac1 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -250,15 +250,6 @@ int ip6_frag_mem(struct net *net); #define IPV6_FRAG_TIMEOUT (60*HZ) /* 60 seconds */ -/* - * Function prototype for build_xmit - */ - -typedef int (*inet_getfrag_t) (const void *data, - struct in6_addr *addr, - char *, - unsigned int, unsigned int); - extern int __ipv6_addr_type(const struct in6_addr *addr); static inline int ipv6_addr_type(const struct in6_addr *addr) { @@ -510,14 +501,6 @@ extern int ip6_local_out(struct sk_buff *skb); * Extension header (options) processing */ -extern u8 * ipv6_build_nfrag_opts(struct sk_buff *skb, - u8 *prev_hdr, - struct ipv6_txoptions *opt, - struct in6_addr *daddr, - u32 jumbolen); -extern u8 * ipv6_build_frag_opts(struct sk_buff *skb, - u8 *prev_hdr, - struct ipv6_txoptions *opt); extern void ipv6_push_nfrag_opts(struct sk_buff *skb, struct ipv6_txoptions *opt, u8 *proto, -- cgit v1.2.3 From 0dde3e16485dca16eb682dd59da1a598bf62e284 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:43:41 -0700 Subject: [NET]: uninline skb_put, de-bloats a lot MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Allyesconfig (v2.6.24-mm1): ~500 files changed ... 869 funcs, 198 +, 111003 -, diff: -110805 --- skb_put skb_put | +104 Without number of debug related CONFIGs (v2.6.25-rc2-mm1): -60744 855 funcs, 861 +, 61605 -, diff: -60744 --- skb_put skb_put | +57 Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/linux/skbuff.h | 21 +-------------------- net/core/skbuff.c | 21 +++++++++++++++++++++ 2 files changed, 22 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 7beb239d2ee0..f085955cb5a7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -892,6 +892,7 @@ static inline void skb_set_tail_pointer(struct sk_buff *skb, const int offset) /* * Add data to an sk_buff */ +extern unsigned char *skb_put(struct sk_buff *skb, unsigned int len); static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len) { unsigned char *tmp = skb_tail_pointer(skb); @@ -901,26 +902,6 @@ static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len) return tmp; } -/** - * skb_put - add data to a buffer - * @skb: buffer to use - * @len: amount of data to add - * - * This function extends the used data area of the buffer. If this would - * exceed the total buffer size the kernel will panic. A pointer to the - * first byte of the extra data is returned. - */ -static inline unsigned char *skb_put(struct sk_buff *skb, unsigned int len) -{ - unsigned char *tmp = skb_tail_pointer(skb); - SKB_LINEAR_ASSERT(skb); - skb->tail += len; - skb->len += len; - if (unlikely(skb->tail > skb->end)) - skb_over_panic(skb, len, current_text_addr()); - return tmp; -} - static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len) { skb->data -= len; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 0d0fd28a9041..3402eca768f8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -857,6 +857,27 @@ free_skb: return err; } +/** + * skb_put - add data to a buffer + * @skb: buffer to use + * @len: amount of data to add + * + * This function extends the used data area of the buffer. If this would + * exceed the total buffer size the kernel will panic. A pointer to the + * first byte of the extra data is returned. + */ +unsigned char *skb_put(struct sk_buff *skb, unsigned int len) +{ + unsigned char *tmp = skb_tail_pointer(skb); + SKB_LINEAR_ASSERT(skb); + skb->tail += len; + skb->len += len; + if (unlikely(skb->tail > skb->end)) + skb_over_panic(skb, len, __builtin_return_address(0)); + return tmp; +} +EXPORT_SYMBOL(skb_put); + /* Trims skb to length len. It can change skb pointers. */ -- cgit v1.2.3 From 6be8ac2fdc5e69dec53913a42312a92dbfbd4907 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:47:24 -0700 Subject: [NET]: uninline skb_pull, de-bloats a lot MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Allyesconfig (v2.6.24-mm1): -28162 354 funcs, 3005 +, 31167 -, diff: -28162 --- skb_pull Without number of debug related CONFIGs (v2.6.25-rc2-mm1): -9697 338 funcs, 221 +, 9918 -, diff: -9697 --- skb_pull skb_pull | +44 Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/linux/skbuff.h | 16 +--------------- net/core/skbuff.c | 16 ++++++++++++++++ 2 files changed, 17 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index f085955cb5a7..6d6cde7b243c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -927,6 +927,7 @@ static inline unsigned char *skb_push(struct sk_buff *skb, unsigned int len) return skb->data; } +extern unsigned char *skb_pull(struct sk_buff *skb, unsigned int len); static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len) { skb->len -= len; @@ -934,21 +935,6 @@ static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len) return skb->data += len; } -/** - * skb_pull - remove data from the start of a buffer - * @skb: buffer to use - * @len: amount of data to remove - * - * This function removes data from the start of a buffer, returning - * the memory to the headroom. A pointer to the next data in the buffer - * is returned. Once the data has been pulled future pushes will overwrite - * the old data. - */ -static inline unsigned char *skb_pull(struct sk_buff *skb, unsigned int len) -{ - return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len); -} - extern unsigned char *__pskb_pull_tail(struct sk_buff *skb, int delta); static inline unsigned char *__pskb_pull(struct sk_buff *skb, unsigned int len) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3402eca768f8..cf489b6329e8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -878,6 +878,22 @@ unsigned char *skb_put(struct sk_buff *skb, unsigned int len) } EXPORT_SYMBOL(skb_put); +/** + * skb_pull - remove data from the start of a buffer + * @skb: buffer to use + * @len: amount of data to remove + * + * This function removes data from the start of a buffer, returning + * the memory to the headroom. A pointer to the next data in the buffer + * is returned. Once the data has been pulled future pushes will overwrite + * the old data. + */ +unsigned char *skb_pull(struct sk_buff *skb, unsigned int len) +{ + return unlikely(len > skb->len) ? NULL : __skb_pull(skb, len); +} +EXPORT_SYMBOL(skb_pull); + /* Trims skb to length len. It can change skb pointers. */ -- cgit v1.2.3 From f58518e678e5eef430c8d5cdcc7cd28d285f1980 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:51:31 -0700 Subject: [NET]: uninline dev_alloc_skb, de-bloats a lot MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Allyesconfig (v2.6.24-mm1): -23668 392 funcs, 104 +, 23772 -, diff: -23668 --- dev_alloc_skb Without many debug CONFIGs (v2.6.25-rc2-mm1): -12178 382 funcs, 157 +, 12335 -, diff: -12178 --- dev_alloc_skb dev_alloc_skb | +37 Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/linux/skbuff.h | 17 +---------------- net/core/skbuff.c | 18 ++++++++++++++++++ 2 files changed, 19 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 6d6cde7b243c..01a11b0c0291 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1272,22 +1272,7 @@ static inline struct sk_buff *__dev_alloc_skb(unsigned int length, return skb; } -/** - * dev_alloc_skb - allocate an skbuff for receiving - * @length: length to allocate - * - * Allocate a new &sk_buff and assign it a usage count of one. The - * buffer has unspecified headroom built in. Users should allocate - * the headroom they think they need without accounting for the - * built in space. The built in space is used for optimisations. - * - * %NULL is returned if there is no free memory. Although this function - * allocates memory it can be called from an interrupt. - */ -static inline struct sk_buff *dev_alloc_skb(unsigned int length) -{ - return __dev_alloc_skb(length, GFP_ATOMIC); -} +extern struct sk_buff *dev_alloc_skb(unsigned int length); extern struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int length, gfp_t gfp_mask); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index cf489b6329e8..0daf5c0e5b8d 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -263,6 +263,24 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev, return skb; } +/** + * dev_alloc_skb - allocate an skbuff for receiving + * @length: length to allocate + * + * Allocate a new &sk_buff and assign it a usage count of one. The + * buffer has unspecified headroom built in. Users should allocate + * the headroom they think they need without accounting for the + * built in space. The built in space is used for optimisations. + * + * %NULL is returned if there is no free memory. Although this function + * allocates memory it can be called from an interrupt. + */ +struct sk_buff *dev_alloc_skb(unsigned int length) +{ + return __dev_alloc_skb(length, GFP_ATOMIC); +} +EXPORT_SYMBOL(dev_alloc_skb); + static void skb_drop_list(struct sk_buff **listp) { struct sk_buff *list = *listp; -- cgit v1.2.3 From c2aa270ad73d385bd6cdebf5d741bdf18a3e17ad Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:52:40 -0700 Subject: [NET]: uninline skb_push, de-bloats a lot MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Allyesconfig (v2.6.24-mm1): -21593 356 funcs, 2418 +, 24011 -, diff: -21593 --- skb_push Without many debug related CONFIGs (v2.6.25-rc2-mm1): -13890 341 funcs, 189 +, 14079 -, diff: -13890 --- skb_push skb_push | +46 Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/linux/skbuff.h | 19 +------------------ net/core/skbuff.c | 19 +++++++++++++++++++ 2 files changed, 20 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 01a11b0c0291..1baf4d43bb2d 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -902,6 +902,7 @@ static inline unsigned char *__skb_put(struct sk_buff *skb, unsigned int len) return tmp; } +extern unsigned char *skb_push(struct sk_buff *skb, unsigned int len); static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len) { skb->data -= len; @@ -909,24 +910,6 @@ static inline unsigned char *__skb_push(struct sk_buff *skb, unsigned int len) return skb->data; } -/** - * skb_push - add data to the start of a buffer - * @skb: buffer to use - * @len: amount of data to add - * - * This function extends the used data area of the buffer at the buffer - * start. If this would exceed the total buffer headroom the kernel will - * panic. A pointer to the first byte of the extra data is returned. - */ -static inline unsigned char *skb_push(struct sk_buff *skb, unsigned int len) -{ - skb->data -= len; - skb->len += len; - if (unlikely(skb->datahead)) - skb_under_panic(skb, len, current_text_addr()); - return skb->data; -} - extern unsigned char *skb_pull(struct sk_buff *skb, unsigned int len); static inline unsigned char *__skb_pull(struct sk_buff *skb, unsigned int len) { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 0daf5c0e5b8d..a37127b5899c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -896,6 +896,25 @@ unsigned char *skb_put(struct sk_buff *skb, unsigned int len) } EXPORT_SYMBOL(skb_put); +/** + * skb_push - add data to the start of a buffer + * @skb: buffer to use + * @len: amount of data to add + * + * This function extends the used data area of the buffer at the buffer + * start. If this would exceed the total buffer headroom the kernel will + * panic. A pointer to the first byte of the extra data is returned. + */ +unsigned char *skb_push(struct sk_buff *skb, unsigned int len) +{ + skb->data -= len; + skb->len += len; + if (unlikely(skb->datahead)) + skb_under_panic(skb, len, __builtin_return_address(0)); + return skb->data; +} +EXPORT_SYMBOL(skb_push); + /** * skb_pull - remove data from the start of a buffer * @skb: buffer to use -- cgit v1.2.3 From 8d3308687f7f1eaa1bb5d202d14752d5f90068eb Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:53:31 -0700 Subject: [NET]: uninline dst_release MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Codiff stats (allyesconfig, v2.6.24-mm1): -16420 187 funcs, 103 +, 16523 -, diff: -16420 --- dst_release Without number of debug related CONFIGs (v2.6.25-rc2-mm1): -7257 186 funcs, 70 +, 7327 -, diff: -7257 --- dst_release dst_release | +40 Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/net/dst.h | 10 +--------- net/core/dst.c | 10 ++++++++++ 2 files changed, 11 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/dst.h b/include/net/dst.h index ae13370e8484..002500e631f5 100644 --- a/include/net/dst.h +++ b/include/net/dst.h @@ -163,15 +163,7 @@ struct dst_entry * dst_clone(struct dst_entry * dst) return dst; } -static inline -void dst_release(struct dst_entry * dst) -{ - if (dst) { - WARN_ON(atomic_read(&dst->__refcnt) < 1); - smp_mb__before_atomic_dec(); - atomic_dec(&dst->__refcnt); - } -} +extern void dst_release(struct dst_entry *dst); /* Children define the path of the packet through the * Linux networking. Thus, destinations are stackable. diff --git a/net/core/dst.c b/net/core/dst.c index 694cd2a3f6d2..fe03266130b6 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -259,6 +259,16 @@ again: return NULL; } +void dst_release(struct dst_entry *dst) +{ + if (dst) { + WARN_ON(atomic_read(&dst->__refcnt) < 1); + smp_mb__before_atomic_dec(); + atomic_dec(&dst->__refcnt); + } +} +EXPORT_SYMBOL(dst_release); + /* Dirty hack. We did it in 2.2 (in __dst_free), * we have _very_ good reasons not to repeat * this mistake in 2.3, but we have no choice -- cgit v1.2.3 From 419ae74ecc9494e58928a5c6652f4c072f3ca744 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:54:01 -0700 Subject: [NET]: uninline skb_trim, de-bloats MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Allyesconfig (v2.6.24-mm1): -10976 209 funcs, 123 +, 11099 -, diff: -10976 --- skb_trim Without number of debug related CONFIGs (v2.6.25-rc2-mm1): -7360 192 funcs, 131 +, 7491 -, diff: -7360 --- skb_trim skb_trim | +42 Signed-off-by: Ilpo Järvinen Signed-off-by: David S. Miller --- include/linux/skbuff.h | 16 +--------------- net/core/skbuff.c | 16 ++++++++++++++++ 2 files changed, 17 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 1baf4d43bb2d..ff72145d5d9e 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -1158,21 +1158,7 @@ static inline void __skb_trim(struct sk_buff *skb, unsigned int len) skb_set_tail_pointer(skb, len); } -/** - * skb_trim - remove end from a buffer - * @skb: buffer to alter - * @len: new length - * - * Cut the length of a buffer down by removing data from the tail. If - * the buffer is already under the length specified it is not modified. - * The skb must be linear. - */ -static inline void skb_trim(struct sk_buff *skb, unsigned int len) -{ - if (skb->len > len) - __skb_trim(skb, len); -} - +extern void skb_trim(struct sk_buff *skb, unsigned int len); static inline int __pskb_trim(struct sk_buff *skb, unsigned int len) { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index a37127b5899c..86e5682728be 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -931,6 +931,22 @@ unsigned char *skb_pull(struct sk_buff *skb, unsigned int len) } EXPORT_SYMBOL(skb_pull); +/** + * skb_trim - remove end from a buffer + * @skb: buffer to alter + * @len: new length + * + * Cut the length of a buffer down by removing data from the tail. If + * the buffer is already under the length specified it is not modified. + * The skb must be linear. + */ +void skb_trim(struct sk_buff *skb, unsigned int len) +{ + if (skb->len > len) + __skb_trim(skb, len); +} +EXPORT_SYMBOL(skb_trim); + /* Trims skb to length len. It can change skb pointers. */ -- cgit v1.2.3 From bc09dff198e67a98a82c42000006b39f6d502031 Mon Sep 17 00:00:00 2001 From: Ilpo Järvinen Date: Thu, 27 Mar 2008 17:54:29 -0700 Subject: [SCTP]: Remove sctp_add_cmd_sf wrapper bloat MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With a was number of callsites sctp_add_cmd_sf wrapper bloats kernel by some amount. Due to unlikely tracking allyesconfig, with the initial result were around ~7kB (thus caught my attention) while a non-debug config produced only ~2.3kB effect. I (ij) proposed first a patch to uninline it but Vlad responded with a patch that removed the only sctp_add_cmd call which is wrapped by sctp_add_cmd_sf (I wasn't sure if I could do that). I did minor cleanup to Vlad's patch. Signed-off-by: Ilpo Järvinen Signed-off-by: Vlad Yasevich Signed-off-by: David S. Miller --- include/net/sctp/command.h | 3 +-- include/net/sctp/sm.h | 8 -------- net/sctp/command.c | 10 ++-------- net/sctp/sm_statefuns.c | 8 ++------ 4 files changed, 5 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index 10ae2da6f93b..4263af857794 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -205,12 +205,11 @@ typedef struct { int sctp_init_cmd_seq(sctp_cmd_seq_t *seq); /* Add a command to an sctp_cmd_seq_t. - * Return 0 if the command sequence is full. * * Use the SCTP_* constructors defined by SCTP_ARG_CONSTRUCTOR() above * to wrap data which goes in the obj argument. */ -int sctp_add_cmd(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj); +void sctp_add_cmd_sf(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj); /* Return the next command structure in an sctp_cmd_seq. * Return NULL at the end of the sequence. diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index ef9e7ed2c82e..24811732bdb2 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -385,14 +385,6 @@ static inline int ADDIP_SERIAL_gte(__u16 s, __u16 t) return (((s) == (t)) || (((t) - (s)) & ADDIP_SERIAL_SIGN_BIT)); } - -/* Run sctp_add_cmd() generating a BUG() if there is a failure. */ -static inline void sctp_add_cmd_sf(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj) -{ - if (unlikely(!sctp_add_cmd(seq, verb, obj))) - BUG(); -} - /* Check VTAG of the packet matches the sender's own tag. */ static inline int sctp_vtag_verify(const struct sctp_chunk *chunk, diff --git a/net/sctp/command.c b/net/sctp/command.c index bb977330002a..c0044019db9e 100644 --- a/net/sctp/command.c +++ b/net/sctp/command.c @@ -52,18 +52,12 @@ int sctp_init_cmd_seq(sctp_cmd_seq_t *seq) /* 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) +void sctp_add_cmd_sf(sctp_cmd_seq_t *seq, sctp_verb_t verb, sctp_arg_t obj) { - if (seq->next_free_slot >= SCTP_MAX_NUM_COMMANDS) - goto fail; + BUG_ON(seq->next_free_slot >= SCTP_MAX_NUM_COMMANDS); 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. diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 6545b5fcbc73..b534dbef864f 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -3135,12 +3135,8 @@ sctp_disposition_t sctp_sf_operr_notify(const struct sctp_endpoint *ep, 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_EVENT_ULP, + SCTP_ULPEVENT(ev)); sctp_add_cmd_sf(commands, SCTP_CMD_PROCESS_OPERR, SCTP_CHUNK(chunk)); } -- cgit v1.2.3 From 1567ca7eec7664b8be3b07755ac59dc1b1ec76cb Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 28 Mar 2008 15:53:11 -0700 Subject: [NET]: Protect device namespace inlines with CONFIG_NET Include sites should not be bothered by whether CONFIG_NET is set or not when trying to include benign files like linux/etherdevice.h et al. From a report by Stephen Rothwell. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 3b54f8a2c055..8576ca928dae 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -741,6 +741,7 @@ struct net_device #define NETDEV_ALIGN 32 #define NETDEV_ALIGN_CONST (NETDEV_ALIGN - 1) +#ifdef CONFIG_NET /* * Net namespace inlines */ @@ -761,6 +762,7 @@ void dev_net_set(struct net_device *dev, struct net *net) dev->nd_net = net; #endif } +#endif /** * netdev_priv - access network device private data -- cgit v1.2.3 From be2ce06b4962658d807410e58f7c6b739dc6a0c3 Mon Sep 17 00:00:00 2001 From: Rami Rosen Date: Fri, 28 Mar 2008 16:26:45 -0700 Subject: [IPV6]: Remove unused method declaration in include/net/addrconf.h. This patches removes unused declaration of addrconf_forwarding_on() method in include/net/addrconf.h. Signed-off-by: Rami Rosen Signed-off-by: David S. Miller --- include/net/addrconf.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index d0c47c306046..d89b0bc7ab75 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -185,7 +185,6 @@ static inline void in6_ifa_put(struct inet6_ifaddr *ifp) #define in6_ifa_hold(ifp) atomic_inc(&(ifp)->refcnt) -extern void addrconf_forwarding_on(void); /* * Hash function taken from net_alias.c */ -- cgit v1.2.3 From bc578a54f0fd489d0722303f9a52508495ccaf9a Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Fri, 28 Mar 2008 16:35:27 -0700 Subject: [NET]: Rename inet_frag.h identifiers COMPLETE, FIRST_IN, LAST_IN to INET_FRAG_* On Fri, 2008-03-28 at 03:24 -0700, Andrew Morton wrote: > they should all be renamed. Done for include/net and net Signed-off-by: Joe Perches Signed-off-by: David S. Miller --- include/net/inet_frag.h | 6 +++--- net/ipv4/inet_fragment.c | 10 +++++----- net/ipv4/ip_fragment.c | 17 +++++++++-------- net/ipv6/netfilter/nf_conntrack_reasm.c | 15 ++++++++------- net/ipv6/reassembly.c | 17 +++++++++-------- 5 files changed, 34 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/net/inet_frag.h b/include/net/inet_frag.h index 7374251b9787..e081eefd6f47 100644 --- a/include/net/inet_frag.h +++ b/include/net/inet_frag.h @@ -25,9 +25,9 @@ struct inet_frag_queue { int meat; __u8 last_in; /* first/last segment arrived? */ -#define COMPLETE 4 -#define FIRST_IN 2 -#define LAST_IN 1 +#define INET_FRAG_COMPLETE 4 +#define INET_FRAG_FIRST_IN 2 +#define INET_FRAG_LAST_IN 1 }; #define INETFRAGS_HASHSZ 64 diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c index 724d69aed031..93170bfcc22e 100644 --- a/net/ipv4/inet_fragment.c +++ b/net/ipv4/inet_fragment.c @@ -104,10 +104,10 @@ void inet_frag_kill(struct inet_frag_queue *fq, struct inet_frags *f) if (del_timer(&fq->timer)) atomic_dec(&fq->refcnt); - if (!(fq->last_in & COMPLETE)) { + if (!(fq->last_in & INET_FRAG_COMPLETE)) { fq_unlink(fq, f); atomic_dec(&fq->refcnt); - fq->last_in |= COMPLETE; + fq->last_in |= INET_FRAG_COMPLETE; } } @@ -131,7 +131,7 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f, struct sk_buff *fp; struct netns_frags *nf; - BUG_TRAP(q->last_in & COMPLETE); + BUG_TRAP(q->last_in & INET_FRAG_COMPLETE); BUG_TRAP(del_timer(&q->timer) == 0); /* Release all fragment data. */ @@ -174,7 +174,7 @@ int inet_frag_evictor(struct netns_frags *nf, struct inet_frags *f) read_unlock(&f->lock); spin_lock(&q->lock); - if (!(q->last_in & COMPLETE)) + if (!(q->last_in & INET_FRAG_COMPLETE)) inet_frag_kill(q, f); spin_unlock(&q->lock); @@ -206,7 +206,7 @@ static struct inet_frag_queue *inet_frag_intern(struct netns_frags *nf, if (qp->net == nf && f->match(qp, arg)) { atomic_inc(&qp->refcnt); write_unlock(&f->lock); - qp_in->last_in |= COMPLETE; + qp_in->last_in |= INET_FRAG_COMPLETE; inet_frag_put(qp_in, f); return qp; } diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index fcb60e76b234..02ae470fe60c 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -194,7 +194,7 @@ static void ip_expire(unsigned long arg) spin_lock(&qp->q.lock); - if (qp->q.last_in & COMPLETE) + if (qp->q.last_in & INET_FRAG_COMPLETE) goto out; ipq_kill(qp); @@ -202,7 +202,7 @@ static void ip_expire(unsigned long arg) IP_INC_STATS_BH(IPSTATS_MIB_REASMTIMEOUT); IP_INC_STATS_BH(IPSTATS_MIB_REASMFAILS); - if ((qp->q.last_in&FIRST_IN) && qp->q.fragments != NULL) { + if ((qp->q.last_in & INET_FRAG_FIRST_IN) && qp->q.fragments != NULL) { struct sk_buff *head = qp->q.fragments; struct net *net; @@ -301,7 +301,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) int ihl, end; int err = -ENOENT; - if (qp->q.last_in & COMPLETE) + if (qp->q.last_in & INET_FRAG_COMPLETE) goto err; if (!(IPCB(skb)->flags & IPSKB_FRAG_COMPLETE) && @@ -327,9 +327,9 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) * or have different end, the segment is corrrupted. */ if (end < qp->q.len || - ((qp->q.last_in & LAST_IN) && end != qp->q.len)) + ((qp->q.last_in & INET_FRAG_LAST_IN) && end != qp->q.len)) goto err; - qp->q.last_in |= LAST_IN; + qp->q.last_in |= INET_FRAG_LAST_IN; qp->q.len = end; } else { if (end&7) { @@ -339,7 +339,7 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) } if (end > qp->q.len) { /* Some bits beyond end -> corruption. */ - if (qp->q.last_in & LAST_IN) + if (qp->q.last_in & INET_FRAG_LAST_IN) goto err; qp->q.len = end; } @@ -438,9 +438,10 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb) qp->q.meat += skb->len; atomic_add(skb->truesize, &qp->q.net->mem); if (offset == 0) - qp->q.last_in |= FIRST_IN; + qp->q.last_in |= INET_FRAG_FIRST_IN; - if (qp->q.last_in == (FIRST_IN | LAST_IN) && qp->q.meat == qp->q.len) + if (qp->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && + qp->q.meat == qp->q.len) return ip_frag_reasm(qp, prev, dev); write_lock(&ip4_frags.lock); diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 2a0d698b24d5..ad8066200f9d 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -183,7 +183,7 @@ static void nf_ct_frag6_expire(unsigned long data) spin_lock(&fq->q.lock); - if (fq->q.last_in & COMPLETE) + if (fq->q.last_in & INET_FRAG_COMPLETE) goto out; fq_kill(fq); @@ -225,7 +225,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, struct sk_buff *prev, *next; int offset, end; - if (fq->q.last_in & COMPLETE) { + if (fq->q.last_in & INET_FRAG_COMPLETE) { pr_debug("Allready completed\n"); goto err; } @@ -252,11 +252,11 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, * or have different end, the segment is corrupted. */ if (end < fq->q.len || - ((fq->q.last_in & LAST_IN) && end != fq->q.len)) { + ((fq->q.last_in & INET_FRAG_LAST_IN) && end != fq->q.len)) { pr_debug("already received last fragment\n"); goto err; } - fq->q.last_in |= LAST_IN; + fq->q.last_in |= INET_FRAG_LAST_IN; fq->q.len = end; } else { /* Check if the fragment is rounded to 8 bytes. @@ -271,7 +271,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, } if (end > fq->q.len) { /* Some bits beyond end -> corruption. */ - if (fq->q.last_in & LAST_IN) { + if (fq->q.last_in & INET_FRAG_LAST_IN) { pr_debug("last packet already reached.\n"); goto err; } @@ -383,7 +383,7 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, */ if (offset == 0) { fq->nhoffset = nhoff; - fq->q.last_in |= FIRST_IN; + fq->q.last_in |= INET_FRAG_FIRST_IN; } write_lock(&nf_frags.lock); list_move_tail(&fq->q.lru_list, &nf_init_frags.lru_list); @@ -645,7 +645,8 @@ struct sk_buff *nf_ct_frag6_gather(struct sk_buff *skb) goto ret_orig; } - if (fq->q.last_in == (FIRST_IN|LAST_IN) && fq->q.meat == fq->q.len) { + if (fq->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && + fq->q.meat == fq->q.len) { ret_skb = nf_ct_frag6_reasm(fq, dev); if (ret_skb == NULL) pr_debug("Can't reassemble fragmented packets\n"); diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 4e1447634f36..7b247e3a16fe 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -202,7 +202,7 @@ static void ip6_frag_expire(unsigned long data) spin_lock(&fq->q.lock); - if (fq->q.last_in & COMPLETE) + if (fq->q.last_in & INET_FRAG_COMPLETE) goto out; fq_kill(fq); @@ -217,7 +217,7 @@ static void ip6_frag_expire(unsigned long data) rcu_read_unlock(); /* Don't send error if the first segment did not arrive. */ - if (!(fq->q.last_in&FIRST_IN) || !fq->q.fragments) + if (!(fq->q.last_in & INET_FRAG_FIRST_IN) || !fq->q.fragments) goto out; /* @@ -265,7 +265,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, struct net_device *dev; int offset, end; - if (fq->q.last_in & COMPLETE) + if (fq->q.last_in & INET_FRAG_COMPLETE) goto err; offset = ntohs(fhdr->frag_off) & ~0x7; @@ -294,9 +294,9 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, * or have different end, the segment is corrupted. */ if (end < fq->q.len || - ((fq->q.last_in & LAST_IN) && end != fq->q.len)) + ((fq->q.last_in & INET_FRAG_LAST_IN) && end != fq->q.len)) goto err; - fq->q.last_in |= LAST_IN; + fq->q.last_in |= INET_FRAG_LAST_IN; fq->q.len = end; } else { /* Check if the fragment is rounded to 8 bytes. @@ -314,7 +314,7 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, } if (end > fq->q.len) { /* Some bits beyond end -> corruption. */ - if (fq->q.last_in & LAST_IN) + if (fq->q.last_in & INET_FRAG_LAST_IN) goto err; fq->q.len = end; } @@ -417,10 +417,11 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, */ if (offset == 0) { fq->nhoffset = nhoff; - fq->q.last_in |= FIRST_IN; + fq->q.last_in |= INET_FRAG_FIRST_IN; } - if (fq->q.last_in == (FIRST_IN | LAST_IN) && fq->q.meat == fq->q.len) + if (fq->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) && + fq->q.meat == fq->q.len) return ip6_frag_reasm(fq, prev, dev); write_lock(&ip6_frags.lock); -- cgit v1.2.3 From 13ff3d6fa4e6d8b6ee7c64245a0078e6a0e6f977 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Fri, 28 Mar 2008 16:38:17 -0700 Subject: [SOCK]: Enumerate struct proto-s to facilitate percpu inuse accounting (v2). The inuse counters are going to become a per-cpu array. Introduce an index for this array on the struct proto. To handle the case of proto register-unregister-register loop the bitmap is used. All its bits manipulations are protected with proto_list_lock and a sanity check for the bitmap being exhausted is also added. Signed-off-by: Pavel Emelyanov Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 1 + net/core/sock.c | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 1c9d059223ee..abc6341f536f 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -562,6 +562,7 @@ struct proto { /* Keeping track of sockets in use */ #ifdef CONFIG_PROC_FS + unsigned int inuse_idx; struct pcounter inuse; #endif diff --git a/net/core/sock.c b/net/core/sock.c index 3ee95060dbd0..7d2c8add5f5a 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1940,6 +1940,38 @@ EXPORT_SYMBOL(sk_common_release); static DEFINE_RWLOCK(proto_list_lock); static LIST_HEAD(proto_list); +#ifdef CONFIG_PROC_FS +#define PROTO_INUSE_NR 64 /* should be enough for the first time */ + +static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR); + +static void assign_proto_idx(struct proto *prot) +{ + prot->inuse_idx = find_first_zero_bit(proto_inuse_idx, PROTO_INUSE_NR); + + if (unlikely(prot->inuse_idx == PROTO_INUSE_NR - 1)) { + printk(KERN_ERR "PROTO_INUSE_NR exhausted\n"); + return; + } + + set_bit(prot->inuse_idx, proto_inuse_idx); +} + +static void release_proto_idx(struct proto *prot) +{ + if (prot->inuse_idx != PROTO_INUSE_NR - 1) + clear_bit(prot->inuse_idx, proto_inuse_idx); +} +#else +static inline void assign_proto_idx(struct proto *prot) +{ +} + +static inline void release_proto_idx(struct proto *prot) +{ +} +#endif + int proto_register(struct proto *prot, int alloc_slab) { char *request_sock_slab_name = NULL; @@ -2000,6 +2032,7 @@ int proto_register(struct proto *prot, int alloc_slab) write_lock(&proto_list_lock); list_add(&prot->node, &proto_list); + assign_proto_idx(prot); write_unlock(&proto_list_lock); return 0; @@ -2026,6 +2059,7 @@ EXPORT_SYMBOL(proto_register); void proto_unregister(struct proto *prot) { write_lock(&proto_list_lock); + release_proto_idx(prot); list_del(&prot->node); write_unlock(&proto_list_lock); -- cgit v1.2.3 From 1338d466d9c3f8a65cc6d83c629cd906f2a989f8 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Fri, 28 Mar 2008 16:38:43 -0700 Subject: [SOCK]: Introduce a percpu inuse counters array (v2). And redirect sock_prot_inuse_add and _get to use one. As far as the dereferences are concerned. Before the patch we made 1 dereference to proto->inuse.add call, the call itself and then called the __get_cpu_var() on a static variable. After the patch we make a direct call, then one dereference to proto->inuse_idx and then the same __get_cpu_var() on a still static variable. So this patch doesn't seem to produce performance penalty on SMP. This is not per-net yet, but I will deliberately make NET_NS=y case separated from NET_NS=n one, since it'll cost us one-or-two more dereferences to get the struct net and the inuse counter. Signed-off-by: Pavel Emelyanov Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 13 +++++-------- net/core/sock.c | 22 ++++++++++++++++++++++ 2 files changed, 27 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index abc6341f536f..ebf9552664b2 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -639,18 +639,15 @@ static inline void sk_refcnt_debug_release(const struct sock *sk) # define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME) # define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse) /* Called with local bh disabled */ -static inline void sock_prot_inuse_add(struct proto *prot, int inc) -{ - pcounter_add(&prot->inuse, inc); -} +extern void sock_prot_inuse_add(struct proto *prot, int inc); + static inline int sock_prot_inuse_init(struct proto *proto) { return pcounter_alloc(&proto->inuse); } -static inline int sock_prot_inuse_get(struct proto *proto) -{ - return pcounter_getval(&proto->inuse); -} + +extern int sock_prot_inuse_get(struct proto *proto); + static inline void sock_prot_inuse_free(struct proto *proto) { pcounter_free(&proto->inuse); diff --git a/net/core/sock.c b/net/core/sock.c index 7d2c8add5f5a..174c64bc7a43 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1942,8 +1942,30 @@ static LIST_HEAD(proto_list); #ifdef CONFIG_PROC_FS #define PROTO_INUSE_NR 64 /* should be enough for the first time */ +struct prot_inuse { + int val[PROTO_INUSE_NR]; +}; static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR); +static DEFINE_PER_CPU(struct prot_inuse, prot_inuse); + +void sock_prot_inuse_add(struct proto *prot, int val) +{ + __get_cpu_var(prot_inuse).val[prot->inuse_idx] += val; +} +EXPORT_SYMBOL_GPL(sock_prot_inuse_add); + +int sock_prot_inuse_get(struct proto *prot) +{ + int cpu, idx = prot->inuse_idx; + int res = 0; + + for_each_possible_cpu(cpu) + res += per_cpu(prot_inuse, cpu).val[idx]; + + return res >= 0 ? res : 0; +} +EXPORT_SYMBOL_GPL(sock_prot_inuse_get); static void assign_proto_idx(struct proto *prot) { -- cgit v1.2.3 From 60e7663d462af3994f292cb3691ea4f7371a9220 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Fri, 28 Mar 2008 16:39:10 -0700 Subject: [SOCK]: Drop per-proto inuse init and fre functions (v2). Constructive part of the set is finished here. We have to remove the pcounter, so start with its init and free functions. Signed-off-by: Pavel Emelyanov Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 18 ------------------ net/core/sock.c | 11 +---------- 2 files changed, 1 insertion(+), 28 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index ebf9552664b2..1f4294252dd7 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -640,31 +640,13 @@ static inline void sk_refcnt_debug_release(const struct sock *sk) # define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse) /* Called with local bh disabled */ extern void sock_prot_inuse_add(struct proto *prot, int inc); - -static inline int sock_prot_inuse_init(struct proto *proto) -{ - return pcounter_alloc(&proto->inuse); -} - extern int sock_prot_inuse_get(struct proto *proto); - -static inline void sock_prot_inuse_free(struct proto *proto) -{ - pcounter_free(&proto->inuse); -} #else # define DEFINE_PROTO_INUSE(NAME) # define REF_PROTO_INUSE(NAME) static void inline sock_prot_inuse_add(struct proto *prot, int inc) { } -static int inline sock_prot_inuse_init(struct proto *proto) -{ - return 0; -} -static void inline sock_prot_inuse_free(struct proto *proto) -{ -} #endif diff --git a/net/core/sock.c b/net/core/sock.c index 174c64bc7a43..c1ae56eb96ec 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1999,11 +1999,6 @@ int proto_register(struct proto *prot, int alloc_slab) char *request_sock_slab_name = NULL; char *timewait_sock_slab_name; - if (sock_prot_inuse_init(prot) != 0) { - printk(KERN_CRIT "%s: Can't alloc inuse counters!\n", prot->name); - goto out; - } - if (alloc_slab) { prot->slab = kmem_cache_create(prot->name, prot->obj_size, 0, SLAB_HWCACHE_ALIGN, NULL); @@ -2011,7 +2006,7 @@ int proto_register(struct proto *prot, int alloc_slab) if (prot->slab == NULL) { printk(KERN_CRIT "%s: Can't create sock SLAB cache!\n", prot->name); - goto out_free_inuse; + goto out; } if (prot->rsk_prot != NULL) { @@ -2070,8 +2065,6 @@ out_free_request_sock_slab_name: out_free_sock_slab: kmem_cache_destroy(prot->slab); prot->slab = NULL; -out_free_inuse: - sock_prot_inuse_free(prot); out: return -ENOBUFS; } @@ -2085,8 +2078,6 @@ void proto_unregister(struct proto *prot) list_del(&prot->node); write_unlock(&proto_list_lock); - sock_prot_inuse_free(prot); - if (prot->slab != NULL) { kmem_cache_destroy(prot->slab); prot->slab = NULL; -- cgit v1.2.3 From bdcde3d71a67e97f25e851f3ca97c9bb5ef03e7f Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Fri, 28 Mar 2008 16:39:33 -0700 Subject: [SOCK]: Drop inuse pcounter from struct proto (v2). An uppercut - do not use the pcounter on struct proto. Signed-off-by: Pavel Emelyanov Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/sock.h | 6 ------ net/dccp/ipv4.c | 3 --- net/dccp/ipv6.c | 3 --- net/ipv4/raw.c | 3 --- net/ipv4/tcp_ipv4.c | 3 --- net/ipv4/udp.c | 3 --- net/ipv4/udplite.c | 3 --- net/ipv6/raw.c | 3 --- net/ipv6/tcp_ipv6.c | 3 --- net/ipv6/udp.c | 3 --- net/ipv6/udplite.c | 3 --- net/sctp/socket.c | 5 ----- 12 files changed, 41 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 1f4294252dd7..2a3344f666aa 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -47,7 +47,6 @@ #include #include #include -#include #include /* struct sk_buff */ #include #include @@ -563,7 +562,6 @@ struct proto { /* Keeping track of sockets in use */ #ifdef CONFIG_PROC_FS unsigned int inuse_idx; - struct pcounter inuse; #endif /* Memory pressure */ @@ -636,14 +634,10 @@ static inline void sk_refcnt_debug_release(const struct sock *sk) #ifdef CONFIG_PROC_FS -# define DEFINE_PROTO_INUSE(NAME) DEFINE_PCOUNTER(NAME) -# define REF_PROTO_INUSE(NAME) PCOUNTER_MEMBER_INITIALIZER(NAME, .inuse) /* Called with local bh disabled */ extern void sock_prot_inuse_add(struct proto *prot, int inc); extern int sock_prot_inuse_get(struct proto *proto); #else -# define DEFINE_PROTO_INUSE(NAME) -# define REF_PROTO_INUSE(NAME) static void inline sock_prot_inuse_add(struct proto *prot, int inc) { } diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 4ca8b0c93c80..7d62f7eb6134 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -915,8 +915,6 @@ static struct timewait_sock_ops dccp_timewait_sock_ops = { .twsk_obj_size = sizeof(struct inet_timewait_sock), }; -DEFINE_PROTO_INUSE(dccp_v4) - static struct proto dccp_v4_prot = { .name = "DCCP", .owner = THIS_MODULE, @@ -946,7 +944,6 @@ static struct proto dccp_v4_prot = { .compat_setsockopt = compat_dccp_setsockopt, .compat_getsockopt = compat_dccp_getsockopt, #endif - REF_PROTO_INUSE(dccp_v4) }; static struct net_protocol dccp_v4_protocol = { diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 2fec1af1a8c3..ea3f32648618 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -1100,8 +1100,6 @@ static struct timewait_sock_ops dccp6_timewait_sock_ops = { .twsk_obj_size = sizeof(struct dccp6_timewait_sock), }; -DEFINE_PROTO_INUSE(dccp_v6) - static struct proto dccp_v6_prot = { .name = "DCCPv6", .owner = THIS_MODULE, @@ -1131,7 +1129,6 @@ static struct proto dccp_v6_prot = { .compat_setsockopt = compat_dccp_setsockopt, .compat_getsockopt = compat_dccp_getsockopt, #endif - REF_PROTO_INUSE(dccp_v6) }; static struct inet6_protocol dccp_v6_protocol = { diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index d965f0a39c84..349205048557 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -818,8 +818,6 @@ static int raw_ioctl(struct sock *sk, int cmd, unsigned long arg) } } -DEFINE_PROTO_INUSE(raw) - struct proto raw_prot = { .name = "RAW", .owner = THIS_MODULE, @@ -842,7 +840,6 @@ struct proto raw_prot = { .compat_setsockopt = compat_raw_setsockopt, .compat_getsockopt = compat_raw_getsockopt, #endif - REF_PROTO_INUSE(raw) }; #ifdef CONFIG_PROC_FS diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 2a5881c81778..ef141b841d2f 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2451,8 +2451,6 @@ void tcp4_proc_exit(void) } #endif /* CONFIG_PROC_FS */ -DEFINE_PROTO_INUSE(tcp) - struct proto tcp_prot = { .name = "TCP", .owner = THIS_MODULE, @@ -2488,7 +2486,6 @@ struct proto tcp_prot = { .compat_setsockopt = compat_tcp_setsockopt, .compat_getsockopt = compat_tcp_getsockopt, #endif - REF_PROTO_INUSE(tcp) }; void __init tcp_v4_init(void) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 80007c79f12f..0255f373b3e5 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1467,8 +1467,6 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait) } -DEFINE_PROTO_INUSE(udp) - struct proto udp_prot = { .name = "UDP", .owner = THIS_MODULE, @@ -1496,7 +1494,6 @@ struct proto udp_prot = { .compat_setsockopt = compat_udp_setsockopt, .compat_getsockopt = compat_udp_getsockopt, #endif - REF_PROTO_INUSE(udp) }; /* ------------------------------------------------------------------------ */ diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 53f3ed496328..7c1e0271c0db 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -34,8 +34,6 @@ static struct net_protocol udplite_protocol = { .netns_ok = 1, }; -DEFINE_PROTO_INUSE(udplite) - struct proto udplite_prot = { .name = "UDP-Lite", .owner = THIS_MODULE, @@ -60,7 +58,6 @@ struct proto udplite_prot = { .compat_setsockopt = compat_udp_setsockopt, .compat_getsockopt = compat_udp_getsockopt, #endif - REF_PROTO_INUSE(udplite) }; static struct inet_protosw udplite4_protosw = { diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 830da4603697..c1e495f105ad 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -1171,8 +1171,6 @@ static int rawv6_init_sk(struct sock *sk) return(0); } -DEFINE_PROTO_INUSE(rawv6) - struct proto rawv6_prot = { .name = "RAWv6", .owner = THIS_MODULE, @@ -1196,7 +1194,6 @@ struct proto rawv6_prot = { .compat_setsockopt = compat_rawv6_setsockopt, .compat_getsockopt = compat_rawv6_getsockopt, #endif - REF_PROTO_INUSE(rawv6) }; #ifdef CONFIG_PROC_FS diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 323c7e06ef43..6d851c3c3db9 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2140,8 +2140,6 @@ void tcp6_proc_exit(struct net *net) } #endif -DEFINE_PROTO_INUSE(tcpv6) - struct proto tcpv6_prot = { .name = "TCPv6", .owner = THIS_MODULE, @@ -2177,7 +2175,6 @@ struct proto tcpv6_prot = { .compat_setsockopt = compat_tcp_setsockopt, .compat_getsockopt = compat_tcp_getsockopt, #endif - REF_PROTO_INUSE(tcpv6) }; static struct inet6_protocol tcpv6_protocol = { diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index aacbc82ecf0f..463ae4d448a3 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -999,8 +999,6 @@ void udp6_proc_exit(struct net *net) { /* ------------------------------------------------------------------------ */ -DEFINE_PROTO_INUSE(udpv6) - struct proto udpv6_prot = { .name = "UDPv6", .owner = THIS_MODULE, @@ -1027,7 +1025,6 @@ struct proto udpv6_prot = { .compat_setsockopt = compat_udpv6_setsockopt, .compat_getsockopt = compat_udpv6_getsockopt, #endif - REF_PROTO_INUSE(udpv6) }; static struct inet_protosw udpv6_protosw = { diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 706c5c375a01..c5f5357d115d 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -35,8 +35,6 @@ static struct inet6_protocol udplitev6_protocol = { .flags = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL, }; -DEFINE_PROTO_INUSE(udplitev6) - struct proto udplitev6_prot = { .name = "UDPLITEv6", .owner = THIS_MODULE, @@ -60,7 +58,6 @@ struct proto udplitev6_prot = { .compat_setsockopt = compat_udpv6_setsockopt, .compat_getsockopt = compat_udpv6_getsockopt, #endif - REF_PROTO_INUSE(udplitev6) }; static struct inet_protosw udplite6_protosw = { diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 76c747056dd7..00ebd0610be2 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -6513,8 +6513,6 @@ static void sctp_sock_migrate(struct sock *oldsk, struct sock *newsk, } -DEFINE_PROTO_INUSE(sctp) - /* This proto struct describes the ULP interface for SCTP. */ struct proto sctp_prot = { .name = "SCTP", @@ -6544,11 +6542,9 @@ struct proto sctp_prot = { .enter_memory_pressure = sctp_enter_memory_pressure, .memory_allocated = &sctp_memory_allocated, .sockets_allocated = &sctp_sockets_allocated, - REF_PROTO_INUSE(sctp) }; #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -DEFINE_PROTO_INUSE(sctpv6) struct proto sctpv6_prot = { .name = "SCTPv6", @@ -6578,6 +6574,5 @@ struct proto sctpv6_prot = { .enter_memory_pressure = sctp_enter_memory_pressure, .memory_allocated = &sctp_memory_allocated, .sockets_allocated = &sctp_sockets_allocated, - REF_PROTO_INUSE(sctpv6) }; #endif /* defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) */ -- cgit v1.2.3 From 095d911201b0741e7f326d269a005dba55985acf Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Fri, 28 Mar 2008 16:39:58 -0700 Subject: [LIB]: Drop the pcounter itself. The knock-out. The pcounter abstraction is not used any longer in the kernel. Not sure whether this should go via netdev tree, but as far as I remember it was added via this one, and besides Eric thinks that Andrew shouldn't mind this. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/linux/pcounter.h | 74 ------------------------------------------------ lib/Makefile | 1 - lib/pcounter.c | 58 ------------------------------------- 3 files changed, 133 deletions(-) delete mode 100644 include/linux/pcounter.h delete mode 100644 lib/pcounter.c (limited to 'include') diff --git a/include/linux/pcounter.h b/include/linux/pcounter.h deleted file mode 100644 index a82d9f2628ca..000000000000 --- a/include/linux/pcounter.h +++ /dev/null @@ -1,74 +0,0 @@ -#ifndef __LINUX_PCOUNTER_H -#define __LINUX_PCOUNTER_H -/* - * Using a dynamic percpu 'int' variable has a cost : - * 1) Extra dereference - * Current per_cpu_ptr() implementation uses an array per 'percpu variable'. - * 2) memory cost of NR_CPUS*(32+sizeof(void *)) instead of num_possible_cpus()*4 - * - * This pcounter implementation is an abstraction to be able to use - * either a static or a dynamic per cpu variable. - * One dynamic per cpu variable gets a fast & cheap implementation, we can - * change pcounter implementation too. - */ -struct pcounter { -#ifdef CONFIG_SMP - void (*add)(struct pcounter *self, int inc); - int (*getval)(const struct pcounter *self, int cpu); - int *per_cpu_values; -#else - int val; -#endif -}; - -#ifdef CONFIG_SMP -#include - -#define DEFINE_PCOUNTER(NAME) \ -static DEFINE_PER_CPU(int, NAME##_pcounter_values); \ -static void NAME##_pcounter_add(struct pcounter *self, int val) \ -{ \ - __get_cpu_var(NAME##_pcounter_values) += val; \ -} \ -static int NAME##_pcounter_getval(const struct pcounter *self, int cpu) \ -{ \ - return per_cpu(NAME##_pcounter_values, cpu); \ -} \ - -#define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER) \ - MEMBER = { \ - .add = NAME##_pcounter_add, \ - .getval = NAME##_pcounter_getval, \ - } - - -static inline void pcounter_add(struct pcounter *self, int inc) -{ - self->add(self, inc); -} - -extern int pcounter_getval(const struct pcounter *self); -extern int pcounter_alloc(struct pcounter *self); -extern void pcounter_free(struct pcounter *self); - - -#else /* CONFIG_SMP */ - -static inline void pcounter_add(struct pcounter *self, int inc) -{ - self->val += inc; -} - -static inline int pcounter_getval(const struct pcounter *self) -{ - return self->val; -} - -#define DEFINE_PCOUNTER(NAME) -#define PCOUNTER_MEMBER_INITIALIZER(NAME, MEMBER) -#define pcounter_alloc(self) 0 -#define pcounter_free(self) - -#endif /* CONFIG_SMP */ - -#endif /* __LINUX_PCOUNTER_H */ diff --git a/lib/Makefile b/lib/Makefile index 23de261a4c83..4d059d469554 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -61,7 +61,6 @@ obj-$(CONFIG_TEXTSEARCH_KMP) += ts_kmp.o obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o obj-$(CONFIG_SMP) += percpu_counter.o -obj-$(CONFIG_SMP) += pcounter.o obj-$(CONFIG_AUDIT_GENERIC) += audit.o obj-$(CONFIG_SWIOTLB) += swiotlb.o diff --git a/lib/pcounter.c b/lib/pcounter.c deleted file mode 100644 index 9b56807da93b..000000000000 --- a/lib/pcounter.c +++ /dev/null @@ -1,58 +0,0 @@ -/* - * Define default pcounter functions - * Note that often used pcounters use dedicated functions to get a speed increase. - * (see DEFINE_PCOUNTER/REF_PCOUNTER_MEMBER) - */ - -#include -#include -#include -#include - -static void pcounter_dyn_add(struct pcounter *self, int inc) -{ - per_cpu_ptr(self->per_cpu_values, smp_processor_id())[0] += inc; -} - -static int pcounter_dyn_getval(const struct pcounter *self, int cpu) -{ - return per_cpu_ptr(self->per_cpu_values, cpu)[0]; -} - -int pcounter_getval(const struct pcounter *self) -{ - int res = 0, cpu; - - for_each_possible_cpu(cpu) - res += self->getval(self, cpu); - - return res; -} -EXPORT_SYMBOL_GPL(pcounter_getval); - -int pcounter_alloc(struct pcounter *self) -{ - int rc = 0; - if (self->add == NULL) { - self->per_cpu_values = alloc_percpu(int); - if (self->per_cpu_values != NULL) { - self->add = pcounter_dyn_add; - self->getval = pcounter_dyn_getval; - } else - rc = 1; - } - return rc; -} -EXPORT_SYMBOL_GPL(pcounter_alloc); - -void pcounter_free(struct pcounter *self) -{ - if (self->per_cpu_values != NULL) { - free_percpu(self->per_cpu_values); - self->per_cpu_values = NULL; - self->getval = NULL; - self->add = NULL; - } -} -EXPORT_SYMBOL_GPL(pcounter_free); - -- cgit v1.2.3 From 6f191efe48af62dd5f8917dd21d187f896cd6c81 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 28 Mar 2008 18:23:33 -0700 Subject: [UDP]: Replace struct net on udp_iter_state with seq_net_private. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/udp.h | 2 +- net/ipv4/udp.c | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index 635940d374ab..77af7d46d868 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -194,7 +194,7 @@ struct udp_seq_afinfo { }; struct udp_iter_state { - struct net *net; + struct seq_net_private p; sa_family_t family; struct hlist_head *hashtable; int bucket; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 0255f373b3e5..d695097be2dc 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1503,7 +1503,7 @@ static struct sock *udp_get_first(struct seq_file *seq) { struct sock *sk; struct udp_iter_state *state = seq->private; - struct net *net = state->net; + struct net *net = seq_file_net(seq); for (state->bucket = 0; state->bucket < UDP_HTABLE_SIZE; ++state->bucket) { struct hlist_node *node; @@ -1522,7 +1522,7 @@ found: static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk) { struct udp_iter_state *state = seq->private; - struct net *net = state->net; + struct net *net = seq_file_net(seq); do { sk = sk_next(sk); @@ -1595,7 +1595,7 @@ static int udp_seq_open(struct inode *inode, struct file *file) s->seq_ops.next = udp_seq_next; s->seq_ops.show = afinfo->seq_show; s->seq_ops.stop = udp_seq_stop; - s->net = net; + s->p.net = net; rc = seq_open(file, &s->seq_ops); if (rc) @@ -1617,7 +1617,7 @@ static int udp_seq_release(struct inode *inode, struct file *file) struct seq_file *seq = file->private_data; struct udp_iter_state *s = seq->private; - put_net(s->net); + put_net(s->p.net); seq_release_private(inode, file); return 0; } -- cgit v1.2.3 From dda61925f84d89e2f2a4597d6298a05a2bc05c20 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 28 Mar 2008 18:24:26 -0700 Subject: [UDP]: Move seq_ops from udp_iter_state to udp_seq_afinfo. No need to create seq_operations for each instance of 'netstat'. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/udp.h | 3 +-- net/ipv4/udp.c | 14 ++++++++------ net/ipv4/udplite.c | 4 +++- net/ipv6/udp.c | 4 +++- net/ipv6/udplite.c | 4 +++- 5 files changed, 18 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index 77af7d46d868..0079d17fd3a3 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -189,8 +189,8 @@ struct udp_seq_afinfo { char *name; sa_family_t family; struct hlist_head *hashtable; - int (*seq_show) (struct seq_file *m, void *v); struct file_operations *seq_fops; + struct seq_operations seq_ops; }; struct udp_iter_state { @@ -198,7 +198,6 @@ struct udp_iter_state { sa_family_t family; struct hlist_head *hashtable; int bucket; - struct seq_operations seq_ops; }; #ifdef CONFIG_PROC_FS diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index eb0c64da2634..bae94b3eaa43 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1591,13 +1591,9 @@ static int udp_seq_open(struct inode *inode, struct file *file) s->family = afinfo->family; s->hashtable = afinfo->hashtable; - s->seq_ops.start = udp_seq_start; - s->seq_ops.next = udp_seq_next; - s->seq_ops.show = afinfo->seq_show; - s->seq_ops.stop = udp_seq_stop; s->p.net = net; - rc = seq_open(file, &s->seq_ops); + rc = seq_open(file, &afinfo->seq_ops); if (rc) goto out_put_net; @@ -1634,6 +1630,10 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) afinfo->seq_fops->llseek = seq_lseek; afinfo->seq_fops->release = udp_seq_release; + afinfo->seq_ops.start = udp_seq_start; + afinfo->seq_ops.next = udp_seq_next; + afinfo->seq_ops.stop = udp_seq_stop; + p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops); if (p) p->data = afinfo; @@ -1690,8 +1690,10 @@ static struct udp_seq_afinfo udp4_seq_afinfo = { .name = "udp", .family = AF_INET, .hashtable = udp_hash, - .seq_show = udp4_seq_show, .seq_fops = &udp4_seq_fops, + .seq_ops = { + .show = udp4_seq_show, + }, }; static int udp4_proc_init_net(struct net *net) diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 7c1e0271c0db..4add87553bce 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -77,8 +77,10 @@ static struct udp_seq_afinfo udplite4_seq_afinfo = { .name = "udplite", .family = AF_INET, .hashtable = udplite_hash, - .seq_show = udp4_seq_show, .seq_fops = &udplite4_seq_fops, + .seq_ops = { + .show = udp4_seq_show, + }, }; static int udplite4_proc_init_net(struct net *net) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 463ae4d448a3..e35921a46689 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -983,8 +983,10 @@ static struct udp_seq_afinfo udp6_seq_afinfo = { .name = "udp6", .family = AF_INET6, .hashtable = udp_hash, - .seq_show = udp6_seq_show, .seq_fops = &udp6_seq_fops, + .seq_ops = { + .show = udp6_seq_show, + }, }; int udp6_proc_init(struct net *net) diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index c5f5357d115d..710366c3f0a6 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -102,8 +102,10 @@ static struct udp_seq_afinfo udplite6_seq_afinfo = { .name = "udplite6", .family = AF_INET6, .hashtable = udplite_hash, - .seq_show = udp6_seq_show, .seq_fops = &udplite6_seq_fops, + .seq_ops = { + .show = udp6_seq_show, + }, }; static int udplite6_proc_init_net(struct net *net) -- cgit v1.2.3 From 3ba9441bdf07370670a684e6d95dfc523476677f Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 28 Mar 2008 18:25:32 -0700 Subject: [UDP]: Place file operations directly into udp_seq_afinfo. No need to have separate never-used variable. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/udp.h | 2 +- net/ipv4/udp.c | 15 ++++++--------- net/ipv4/udplite.c | 2 -- net/ipv6/udp.c | 2 -- net/ipv6/udplite.c | 2 -- 5 files changed, 7 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index 0079d17fd3a3..5cf0e593dda7 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -189,7 +189,7 @@ struct udp_seq_afinfo { char *name; sa_family_t family; struct hlist_head *hashtable; - struct file_operations *seq_fops; + struct file_operations seq_fops; struct seq_operations seq_ops; }; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index e716d2d55e1e..a95dff8063a4 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1596,17 +1596,17 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) struct proc_dir_entry *p; int rc = 0; - afinfo->seq_fops->owner = afinfo->owner; - afinfo->seq_fops->open = udp_seq_open; - afinfo->seq_fops->read = seq_read; - afinfo->seq_fops->llseek = seq_lseek; - afinfo->seq_fops->release = seq_release_net; + afinfo->seq_fops.owner = afinfo->owner; + afinfo->seq_fops.open = udp_seq_open; + afinfo->seq_fops.read = seq_read; + afinfo->seq_fops.llseek = seq_lseek; + afinfo->seq_fops.release = seq_release_net; afinfo->seq_ops.start = udp_seq_start; afinfo->seq_ops.next = udp_seq_next; afinfo->seq_ops.stop = udp_seq_stop; - p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops); + p = proc_net_fops_create(net, afinfo->name, S_IRUGO, &afinfo->seq_fops); if (p) p->data = afinfo; else @@ -1617,7 +1617,6 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo) { proc_net_remove(net, afinfo->name); - memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } /* ------------------------------------------------------------------------ */ @@ -1656,13 +1655,11 @@ int udp4_seq_show(struct seq_file *seq, void *v) } /* ------------------------------------------------------------------------ */ -static struct file_operations udp4_seq_fops; static struct udp_seq_afinfo udp4_seq_afinfo = { .owner = THIS_MODULE, .name = "udp", .family = AF_INET, .hashtable = udp_hash, - .seq_fops = &udp4_seq_fops, .seq_ops = { .show = udp4_seq_show, }, diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 4add87553bce..148ae72af111 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -71,13 +71,11 @@ static struct inet_protosw udplite4_protosw = { }; #ifdef CONFIG_PROC_FS -static struct file_operations udplite4_seq_fops; static struct udp_seq_afinfo udplite4_seq_afinfo = { .owner = THIS_MODULE, .name = "udplite", .family = AF_INET, .hashtable = udplite_hash, - .seq_fops = &udplite4_seq_fops, .seq_ops = { .show = udp4_seq_show, }, diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index e35921a46689..ff8d53ecdd60 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -977,13 +977,11 @@ int udp6_seq_show(struct seq_file *seq, void *v) return 0; } -static struct file_operations udp6_seq_fops; static struct udp_seq_afinfo udp6_seq_afinfo = { .owner = THIS_MODULE, .name = "udp6", .family = AF_INET6, .hashtable = udp_hash, - .seq_fops = &udp6_seq_fops, .seq_ops = { .show = udp6_seq_show, }, diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 710366c3f0a6..5adf651d0721 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -96,13 +96,11 @@ void udplitev6_exit(void) } #ifdef CONFIG_PROC_FS -static struct file_operations udplite6_seq_fops; static struct udp_seq_afinfo udplite6_seq_afinfo = { .owner = THIS_MODULE, .name = "udplite6", .family = AF_INET6, .hashtable = udplite_hash, - .seq_fops = &udplite6_seq_fops, .seq_ops = { .show = udp6_seq_show, }, -- cgit v1.2.3 From 4ad96d39a2d74c1b2e400b602da2594f5098fc26 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Fri, 28 Mar 2008 18:25:53 -0700 Subject: [UDP]: Remove owner from udp_seq_afinfo. Move it to udp_seq_afinfo->seq_fops as should be. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/udp.h | 1 - net/ipv4/udp.c | 5 +++-- net/ipv4/udplite.c | 4 +++- net/ipv6/udp.c | 4 +++- net/ipv6/udplite.c | 4 +++- 5 files changed, 12 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/udp.h b/include/net/udp.h index 5cf0e593dda7..24a41fa31641 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -185,7 +185,6 @@ DECLARE_SNMP_STAT(struct udp_mib, udplite_stats_in6); /* /proc */ struct udp_seq_afinfo { - struct module *owner; char *name; sa_family_t family; struct hlist_head *hashtable; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index a95dff8063a4..9143645f9a1b 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1596,7 +1596,6 @@ int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo) struct proc_dir_entry *p; int rc = 0; - afinfo->seq_fops.owner = afinfo->owner; afinfo->seq_fops.open = udp_seq_open; afinfo->seq_fops.read = seq_read; afinfo->seq_fops.llseek = seq_lseek; @@ -1656,10 +1655,12 @@ int udp4_seq_show(struct seq_file *seq, void *v) /* ------------------------------------------------------------------------ */ static struct udp_seq_afinfo udp4_seq_afinfo = { - .owner = THIS_MODULE, .name = "udp", .family = AF_INET, .hashtable = udp_hash, + .seq_fops = { + .owner = THIS_MODULE, + }, .seq_ops = { .show = udp4_seq_show, }, diff --git a/net/ipv4/udplite.c b/net/ipv4/udplite.c index 148ae72af111..72ce26b6c4d3 100644 --- a/net/ipv4/udplite.c +++ b/net/ipv4/udplite.c @@ -72,10 +72,12 @@ static struct inet_protosw udplite4_protosw = { #ifdef CONFIG_PROC_FS static struct udp_seq_afinfo udplite4_seq_afinfo = { - .owner = THIS_MODULE, .name = "udplite", .family = AF_INET, .hashtable = udplite_hash, + .seq_fops = { + .owner = THIS_MODULE, + }, .seq_ops = { .show = udp4_seq_show, }, diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index ff8d53ecdd60..30ef7dc5d403 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -978,10 +978,12 @@ int udp6_seq_show(struct seq_file *seq, void *v) } static struct udp_seq_afinfo udp6_seq_afinfo = { - .owner = THIS_MODULE, .name = "udp6", .family = AF_INET6, .hashtable = udp_hash, + .seq_fops = { + .owner = THIS_MODULE, + }, .seq_ops = { .show = udp6_seq_show, }, diff --git a/net/ipv6/udplite.c b/net/ipv6/udplite.c index 5adf651d0721..491efd00a866 100644 --- a/net/ipv6/udplite.c +++ b/net/ipv6/udplite.c @@ -97,10 +97,12 @@ void udplitev6_exit(void) #ifdef CONFIG_PROC_FS static struct udp_seq_afinfo udplite6_seq_afinfo = { - .owner = THIS_MODULE, .name = "udplite6", .family = AF_INET6, .hashtable = udplite_hash, + .seq_fops = { + .owner = THIS_MODULE, + }, .seq_ops = { .show = udp6_seq_show, }, -- cgit v1.2.3 From 9307b570a745da4f2d83195f5337927e98221bb2 Mon Sep 17 00:00:00 2001 From: "S.Caglar Onur" Date: Fri, 28 Mar 2008 14:41:24 -0700 Subject: drivers/net/arcnet/arcnet.c: use time_* macros The functions time_before, time_before_eq, time_after, and time_after_eq are more robust for comparing jiffies against other values. So use the time_after() macro, defined in linux/jiffies.h, which deals with wrapping correctly. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: S.Caglar Onur Signed-off-by: Andrew Morton Signed-off-by: Jeff Garzik --- drivers/net/arcnet/arcnet.c | 5 +++-- include/linux/arcdevice.h | 4 ++-- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/drivers/net/arcnet/arcnet.c b/drivers/net/arcnet/arcnet.c index c59c8067de99..bdc4c0bb56d9 100644 --- a/drivers/net/arcnet/arcnet.c +++ b/drivers/net/arcnet/arcnet.c @@ -940,7 +940,7 @@ irqreturn_t arcnet_interrupt(int irq, void *dev_id) /* is the RECON info empty or old? */ if (!lp->first_recon || !lp->last_recon || - jiffies - lp->last_recon > HZ * 10) { + time_after(jiffies, lp->last_recon + HZ * 10)) { if (lp->network_down) BUGMSG(D_NORMAL, "reconfiguration detected: cabling restored?\n"); lp->first_recon = lp->last_recon = jiffies; @@ -974,7 +974,8 @@ irqreturn_t arcnet_interrupt(int irq, void *dev_id) lp->num_recons = 1; } } - } else if (lp->network_down && jiffies - lp->last_recon > HZ * 10) { + } else if (lp->network_down && + time_after(jiffies, lp->last_recon + HZ * 10)) { if (lp->network_down) BUGMSG(D_NORMAL, "cabling restored?\n"); lp->first_recon = lp->last_recon = 0; diff --git a/include/linux/arcdevice.h b/include/linux/arcdevice.h index fde675872c56..a1916078fd08 100644 --- a/include/linux/arcdevice.h +++ b/include/linux/arcdevice.h @@ -283,8 +283,8 @@ struct arcnet_local { int next_buf, first_free_buf; /* network "reconfiguration" handling */ - time_t first_recon, /* time of "first" RECON message to count */ - last_recon; /* time of most recent RECON */ + unsigned long first_recon; /* time of "first" RECON message to count */ + unsigned long last_recon; /* time of most recent RECON */ int num_recons; /* number of RECONs between first and last. */ bool network_down; /* do we think the network is down? */ -- cgit v1.2.3 From 3edf8fa5ccf10688a9280b5cbca8ed3947c42866 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 31 Mar 2008 00:28:14 -0700 Subject: [NET]: Fix allnoconfig build on powerpc and avr32 As reported by Haavard Skinnemoen and Stephen Rothwell: > allnoconfig fails with > > include/linux/netdevice.h:843: error: implicit declaration of function 'dev_net' > > which seems to be because the definition of dev_net is inside #ifdef > CONFIG_NET, while next_net_device, which calls it, is not. Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 8576ca928dae..993758f924be 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -827,6 +827,7 @@ struct packet_type { extern rwlock_t dev_base_lock; /* Device list lock */ +#ifdef CONFIG_NET #define for_each_netdev(net, d) \ list_for_each_entry(d, &(net)->dev_base_head, dev_list) #define for_each_netdev_safe(net, d, n) \ @@ -850,6 +851,7 @@ static inline struct net_device *first_net_device(struct net *net) return list_empty(&net->dev_base_head) ? NULL : net_device_entry(net->dev_base_head.next); } +#endif extern int netdev_boot_setup_check(struct net_device *dev); extern unsigned long netdev_boot_base(const char *prefix, int unit); -- cgit v1.2.3 From 8efa6e93cb2666dceafc4844057fdcb9aa324fb7 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Mon, 31 Mar 2008 19:41:14 -0700 Subject: [NETNS]: Introduce a netns_core structure. There's already some stuff on the struct net, that should better be folded into netns_core structure. I'm making the per-proto inuse counter be per-net also, which is also a candidate for this, so introduce this structure and populate it a bit. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/net_namespace.h | 6 ++---- include/net/netns/core.h | 13 +++++++++++++ net/core/sysctl_net_core.c | 12 ++++++------ net/socket.c | 2 +- 4 files changed, 22 insertions(+), 11 deletions(-) create mode 100644 include/net/netns/core.h (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index f8f3d1a5fc35..c01d45fe08c3 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -8,6 +8,7 @@ #include #include +#include #include #include #include @@ -46,10 +47,7 @@ struct net { struct sock *rtnl; /* rtnetlink socket */ - /* core sysctls */ - struct ctl_table_header *sysctl_core_hdr; - int sysctl_somaxconn; - + struct netns_core core; struct netns_packet packet; struct netns_unix unx; struct netns_ipv4 ipv4; diff --git a/include/net/netns/core.h b/include/net/netns/core.h new file mode 100644 index 000000000000..0e8c0f8435d4 --- /dev/null +++ b/include/net/netns/core.h @@ -0,0 +1,13 @@ +#ifndef __NETNS_CORE_H__ +#define __NETNS_CORE_H__ + +struct ctl_table_header; + +struct netns_core { + /* core sysctls */ + struct ctl_table_header *sysctl_hdr; + + int sysctl_somaxconn; +}; + +#endif diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index 130338f83ae5..5fc801057244 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -127,7 +127,7 @@ static struct ctl_table net_core_table[] = { { .ctl_name = NET_CORE_SOMAXCONN, .procname = "somaxconn", - .data = &init_net.sysctl_somaxconn, + .data = &init_net.core.sysctl_somaxconn, .maxlen = sizeof(int), .mode = 0644, .proc_handler = &proc_dointvec @@ -161,7 +161,7 @@ static __net_init int sysctl_core_net_init(struct net *net) { struct ctl_table *tbl, *tmp; - net->sysctl_somaxconn = SOMAXCONN; + net->core.sysctl_somaxconn = SOMAXCONN; tbl = net_core_table; if (net != &init_net) { @@ -178,9 +178,9 @@ static __net_init int sysctl_core_net_init(struct net *net) } } - net->sysctl_core_hdr = register_net_sysctl_table(net, + net->core.sysctl_hdr = register_net_sysctl_table(net, net_core_path, tbl); - if (net->sysctl_core_hdr == NULL) + if (net->core.sysctl_hdr == NULL) goto err_reg; return 0; @@ -196,8 +196,8 @@ static __net_exit void sysctl_core_net_exit(struct net *net) { struct ctl_table *tbl; - tbl = net->sysctl_core_hdr->ctl_table_arg; - unregister_net_sysctl_table(net->sysctl_core_hdr); + tbl = net->core.sysctl_hdr->ctl_table_arg; + unregister_net_sysctl_table(net->core.sysctl_hdr); BUG_ON(tbl == net_core_table); kfree(tbl); } diff --git a/net/socket.c b/net/socket.c index 79e5382fd110..9b5c917f8a6b 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1375,7 +1375,7 @@ asmlinkage long sys_listen(int fd, int backlog) sock = sockfd_lookup_light(fd, &err, &fput_needed); if (sock) { - somaxconn = sock_net(sock->sk)->sysctl_somaxconn; + somaxconn = sock_net(sock->sk)->core.sysctl_somaxconn; if ((unsigned)backlog > somaxconn) backlog = somaxconn; -- cgit v1.2.3 From c29a0bc4dfc4d833eb702b1929cec96a3eeb9f7a Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Mon, 31 Mar 2008 19:41:46 -0700 Subject: [SOCK][NETNS]: Add a struct net argument to sock_prot_inuse_add and _get. This counter is about to become per-proto-and-per-net, so we'll need two arguments to determine which cell in this "table" to work with. All the places, but proc already pass proper net to it - proc will be tuned a bit later. Some indentation with spaces in proc files is done to keep the file coding style consistent. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/sock.h | 7 ++++--- include/net/udp.h | 2 +- net/core/sock.c | 4 ++-- net/ipv4/inet_hashtables.c | 8 ++++---- net/ipv4/inet_timewait_sock.c | 2 +- net/ipv4/proc.c | 11 +++++++---- net/ipv4/raw.c | 4 ++-- net/ipv4/udp.c | 2 +- net/ipv6/inet6_hashtables.c | 4 ++-- net/ipv6/ipv6_sockglue.c | 10 ++++++---- net/ipv6/proc.c | 8 ++++---- 11 files changed, 34 insertions(+), 28 deletions(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 2a3344f666aa..f4fdd101c9a2 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -635,10 +635,11 @@ static inline void sk_refcnt_debug_release(const struct sock *sk) #ifdef CONFIG_PROC_FS /* Called with local bh disabled */ -extern void sock_prot_inuse_add(struct proto *prot, int inc); -extern int sock_prot_inuse_get(struct proto *proto); +extern void sock_prot_inuse_add(struct net *net, struct proto *prot, int inc); +extern int sock_prot_inuse_get(struct net *net, struct proto *proto); #else -static void inline sock_prot_inuse_add(struct proto *prot, int inc) +static void inline sock_prot_inuse_add(struct net *net, struct proto *prot, + int inc) { } #endif diff --git a/include/net/udp.h b/include/net/udp.h index 24a41fa31641..3e55a99b0ba3 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -115,7 +115,7 @@ static inline void udp_lib_unhash(struct sock *sk) write_lock_bh(&udp_hash_lock); if (sk_del_node_init(sk)) { inet_sk(sk)->num = 0; - sock_prot_inuse_add(sk->sk_prot, -1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); } write_unlock_bh(&udp_hash_lock); } diff --git a/net/core/sock.c b/net/core/sock.c index c1ae56eb96ec..6f36ab91bb59 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1949,13 +1949,13 @@ struct prot_inuse { static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR); static DEFINE_PER_CPU(struct prot_inuse, prot_inuse); -void sock_prot_inuse_add(struct proto *prot, int val) +void sock_prot_inuse_add(struct net *net, struct proto *prot, int val) { __get_cpu_var(prot_inuse).val[prot->inuse_idx] += val; } EXPORT_SYMBOL_GPL(sock_prot_inuse_add); -int sock_prot_inuse_get(struct proto *prot) +int sock_prot_inuse_get(struct net *net, struct proto *prot) { int cpu, idx = prot->inuse_idx; int res = 0; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 1b6ff513c75d..32ca2f8b581c 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -288,7 +288,7 @@ unique: sk->sk_hash = hash; BUG_TRAP(sk_unhashed(sk)); __sk_add_node(sk, &head->chain); - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); write_unlock(lock); if (twp) { @@ -332,7 +332,7 @@ void __inet_hash_nolisten(struct sock *sk) write_lock(lock); __sk_add_node(sk, list); - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); write_unlock(lock); } EXPORT_SYMBOL_GPL(__inet_hash_nolisten); @@ -354,7 +354,7 @@ static void __inet_hash(struct sock *sk) inet_listen_wlock(hashinfo); __sk_add_node(sk, list); - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); write_unlock(lock); wake_up(&hashinfo->lhash_wait); } @@ -387,7 +387,7 @@ void inet_unhash(struct sock *sk) } if (__sk_del_node_init(sk)) - sock_prot_inuse_add(sk->sk_prot, -1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); write_unlock_bh(lock); out: if (sk->sk_state == TCP_LISTEN) diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index f12bc24de46f..a74137866fbc 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -91,7 +91,7 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk, /* Step 2: Remove SK from established hash. */ if (__sk_del_node_init(sk)) - sock_prot_inuse_add(sk->sk_prot, -1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); /* Step 3: Hash TW into TIMEWAIT chain. */ inet_twsk_add_node(tw, &ehead->twchain); diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index d63474c6b400..8156c26f9337 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -53,14 +53,17 @@ static int sockstat_seq_show(struct seq_file *seq, void *v) { socket_seq_show(seq); seq_printf(seq, "TCP: inuse %d orphan %d tw %d alloc %d mem %d\n", - sock_prot_inuse_get(&tcp_prot), + sock_prot_inuse_get(&init_net, &tcp_prot), atomic_read(&tcp_orphan_count), tcp_death_row.tw_count, atomic_read(&tcp_sockets_allocated), atomic_read(&tcp_memory_allocated)); - seq_printf(seq, "UDP: inuse %d mem %d\n", sock_prot_inuse_get(&udp_prot), + seq_printf(seq, "UDP: inuse %d mem %d\n", + sock_prot_inuse_get(&init_net, &udp_prot), atomic_read(&udp_memory_allocated)); - seq_printf(seq, "UDPLITE: inuse %d\n", sock_prot_inuse_get(&udplite_prot)); - seq_printf(seq, "RAW: inuse %d\n", sock_prot_inuse_get(&raw_prot)); + seq_printf(seq, "UDPLITE: inuse %d\n", + sock_prot_inuse_get(&init_net, &udplite_prot)); + seq_printf(seq, "RAW: inuse %d\n", + sock_prot_inuse_get(&init_net, &raw_prot)); seq_printf(seq, "FRAG: inuse %d memory %d\n", ip_frag_nqueues(&init_net), ip_frag_mem(&init_net)); return 0; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 349205048557..11d7f753a820 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -93,7 +93,7 @@ void raw_hash_sk(struct sock *sk) write_lock_bh(&h->lock); sk_add_node(sk, head); - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); write_unlock_bh(&h->lock); } EXPORT_SYMBOL_GPL(raw_hash_sk); @@ -104,7 +104,7 @@ void raw_unhash_sk(struct sock *sk) write_lock_bh(&h->lock); if (sk_del_node_init(sk)) - sock_prot_inuse_add(sk->sk_prot, -1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1); write_unlock_bh(&h->lock); } EXPORT_SYMBOL_GPL(raw_unhash_sk); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 9143645f9a1b..03bd70697481 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -231,7 +231,7 @@ gotit: if (sk_unhashed(sk)) { head = &udptable[snum & (UDP_HTABLE_SIZE - 1)]; sk_add_node(sk, head); - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); } error = 0; fail: diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 340c7d42b83a..580014aea4d6 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -43,7 +43,7 @@ void __inet6_hash(struct sock *sk) } __sk_add_node(sk, list); - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); write_unlock(lock); } EXPORT_SYMBOL(__inet6_hash); @@ -204,7 +204,7 @@ unique: BUG_TRAP(sk_unhashed(sk)); __sk_add_node(sk, &head->chain); sk->sk_hash = hash; - sock_prot_inuse_add(sk->sk_prot, 1); + sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1); write_unlock(lock); if (twp != NULL) { diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index d3d93d752e10..4195ac92345e 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -155,10 +155,11 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, if (sk->sk_protocol == IPPROTO_TCP) { struct inet_connection_sock *icsk = inet_csk(sk); + struct net *net = sock_net(sk); local_bh_disable(); - sock_prot_inuse_add(sk->sk_prot, -1); - sock_prot_inuse_add(&tcp_prot, 1); + sock_prot_inuse_add(net, sk->sk_prot, -1); + sock_prot_inuse_add(net, &tcp_prot, 1); local_bh_enable(); sk->sk_prot = &tcp_prot; icsk->icsk_af_ops = &ipv4_specific; @@ -167,12 +168,13 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, tcp_sync_mss(sk, icsk->icsk_pmtu_cookie); } else { struct proto *prot = &udp_prot; + struct net *net = sock_net(sk); if (sk->sk_protocol == IPPROTO_UDPLITE) prot = &udplite_prot; local_bh_disable(); - sock_prot_inuse_add(sk->sk_prot, -1); - sock_prot_inuse_add(prot, 1); + sock_prot_inuse_add(net, sk->sk_prot, -1); + sock_prot_inuse_add(net, prot, 1); local_bh_enable(); sk->sk_prot = prot; sk->sk_socket->ops = &inet_dgram_ops; diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 364dc332532c..4b9d5a905725 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -36,13 +36,13 @@ static struct proc_dir_entry *proc_net_devsnmp6; static int sockstat6_seq_show(struct seq_file *seq, void *v) { seq_printf(seq, "TCP6: inuse %d\n", - sock_prot_inuse_get(&tcpv6_prot)); + sock_prot_inuse_get(&init_net, &tcpv6_prot)); seq_printf(seq, "UDP6: inuse %d\n", - sock_prot_inuse_get(&udpv6_prot)); + sock_prot_inuse_get(&init_net, &udpv6_prot)); seq_printf(seq, "UDPLITE6: inuse %d\n", - sock_prot_inuse_get(&udplitev6_prot)); + sock_prot_inuse_get(&init_net, &udplitev6_prot)); seq_printf(seq, "RAW6: inuse %d\n", - sock_prot_inuse_get(&rawv6_prot)); + sock_prot_inuse_get(&init_net, &rawv6_prot)); seq_printf(seq, "FRAG6: inuse %d memory %d\n", ip6_frag_nqueues(&init_net), ip6_frag_mem(&init_net)); return 0; -- cgit v1.2.3 From 70ee115942be6ce52ff10e5e813fb4da82cdb25a Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Mon, 31 Mar 2008 19:42:16 -0700 Subject: [SOCK][NETNS]: Add the percpu prot_inuse counter in the struct net. Such an accounting would cost us two more dereferences to get the percpu variable from the struct net, so I make sock_prot_inuse_get and _add calls work differently depending on CONFIG_NET_NS - without it old optimized routines are used. The per-cpu counter for init_net is prepared in core_initcall, so that even af_inet, that starts as fs_initcall, will already have the init_net prepared. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/netns/core.h | 3 +++ net/core/sock.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 51 insertions(+) (limited to 'include') diff --git a/include/net/netns/core.h b/include/net/netns/core.h index 0e8c0f8435d4..24d4be76bbd1 100644 --- a/include/net/netns/core.h +++ b/include/net/netns/core.h @@ -2,12 +2,15 @@ #define __NETNS_CORE_H__ struct ctl_table_header; +struct prot_inuse; struct netns_core { /* core sysctls */ struct ctl_table_header *sysctl_hdr; int sysctl_somaxconn; + + struct prot_inuse *inuse; }; #endif diff --git a/net/core/sock.c b/net/core/sock.c index 6f36ab91bb59..83e11f7260ff 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1947,6 +1947,53 @@ struct prot_inuse { }; static DECLARE_BITMAP(proto_inuse_idx, PROTO_INUSE_NR); + +#ifdef CONFIG_NET_NS +void sock_prot_inuse_add(struct net *net, struct proto *prot, int val) +{ + int cpu = smp_processor_id(); + per_cpu_ptr(net->core.inuse, cpu)->val[prot->inuse_idx] += val; +} +EXPORT_SYMBOL_GPL(sock_prot_inuse_add); + +int sock_prot_inuse_get(struct net *net, struct proto *prot) +{ + int cpu, idx = prot->inuse_idx; + int res = 0; + + for_each_possible_cpu(cpu) + res += per_cpu_ptr(net->core.inuse, cpu)->val[idx]; + + return res >= 0 ? res : 0; +} +EXPORT_SYMBOL_GPL(sock_prot_inuse_get); + +static int sock_inuse_init_net(struct net *net) +{ + net->core.inuse = alloc_percpu(struct prot_inuse); + return net->core.inuse ? 0 : -ENOMEM; +} + +static void sock_inuse_exit_net(struct net *net) +{ + free_percpu(net->core.inuse); +} + +static struct pernet_operations net_inuse_ops = { + .init = sock_inuse_init_net, + .exit = sock_inuse_exit_net, +}; + +static __init int net_inuse_init(void) +{ + if (register_pernet_subsys(&net_inuse_ops)) + panic("Cannot initialize net inuse counters"); + + return 0; +} + +core_initcall(net_inuse_init); +#else static DEFINE_PER_CPU(struct prot_inuse, prot_inuse); void sock_prot_inuse_add(struct net *net, struct proto *prot, int val) @@ -1966,6 +2013,7 @@ int sock_prot_inuse_get(struct net *net, struct proto *prot) return res >= 0 ? res : 0; } EXPORT_SYMBOL_GPL(sock_prot_inuse_get); +#endif static void assign_proto_idx(struct proto *prot) { -- cgit v1.2.3 From 225c0a0107735597100dc4133cd88c5ed10d9e63 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 2 Apr 2008 00:09:29 -0700 Subject: [NETNS]: Merge ifdef CONFIG_NET in include/net/net_namespace.h. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/net_namespace.h | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index c01d45fe08c3..4a37037b1d17 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -59,25 +59,27 @@ struct net { #endif }; + #ifdef CONFIG_NET /* Init's network namespace */ extern struct net init_net; #define INIT_NET_NS(net_ns) .net_ns = &init_net, -#else -#define INIT_NET_NS(net_ns) -#endif -extern struct list_head net_namespace_list; - -#ifdef CONFIG_NET extern struct net *copy_net_ns(unsigned long flags, struct net *net_ns); -#else + +#else /* CONFIG_NET */ + +#define INIT_NET_NS(net_ns) + static inline struct net *copy_net_ns(unsigned long flags, struct net *net_ns) { /* There is nothing to copy so this is a noop */ return net_ns; } -#endif +#endif /* CONFIG_NET */ + + +extern struct list_head net_namespace_list; #ifdef CONFIG_NET_NS extern void __put_net(struct net *net); -- cgit v1.2.3 From c0f39322c335412339dec16ebfd2a05ceba5ebcf Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 2 Apr 2008 00:10:28 -0700 Subject: [NETNS]: Do not include net/net_namespace.h from seq_file.h Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/linux/seq_file.h | 22 ---------------------- include/linux/seq_file_net.h | 27 +++++++++++++++++++++++++++ include/net/net_namespace.h | 2 ++ 3 files changed, 29 insertions(+), 22 deletions(-) create mode 100644 include/linux/seq_file_net.h (limited to 'include') diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index 5da70c3f4417..1da1e6208a0a 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -5,7 +5,6 @@ #include #include #include -#include struct seq_operations; struct file; @@ -63,26 +62,5 @@ extern struct list_head *seq_list_start_head(struct list_head *head, extern struct list_head *seq_list_next(void *v, struct list_head *head, loff_t *ppos); -#ifdef CONFIG_NET -struct net; -struct seq_net_private { -#ifdef CONFIG_NET_NS - struct net *net; -#endif -}; - -int seq_open_net(struct inode *, struct file *, - const struct seq_operations *, int); -int seq_release_net(struct inode *, struct file *); -static inline struct net *seq_file_net(struct seq_file *seq) -{ -#ifdef CONFIG_NET_NS - return ((struct seq_net_private *)seq->private)->net; -#else - return &init_net; -#endif -} -#endif /* CONFIG_NET */ - #endif #endif diff --git a/include/linux/seq_file_net.h b/include/linux/seq_file_net.h new file mode 100644 index 000000000000..4ac52542a563 --- /dev/null +++ b/include/linux/seq_file_net.h @@ -0,0 +1,27 @@ +#ifndef __SEQ_FILE_NET_H__ +#define __SEQ_FILE_NET_H__ + +#include + +struct net; +extern struct net init_net; + +struct seq_net_private { +#ifdef CONFIG_NET_NS + struct net *net; +#endif +}; + +int seq_open_net(struct inode *, struct file *, + const struct seq_operations *, int); +int seq_release_net(struct inode *, struct file *); +static inline struct net *seq_file_net(struct seq_file *seq) +{ +#ifdef CONFIG_NET_NS + return ((struct seq_net_private *)seq->private)->net; +#else + return &init_net; +#endif +} + +#endif diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 4a37037b1d17..6c9a48a46685 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -61,6 +61,8 @@ struct net { #ifdef CONFIG_NET +#include + /* Init's network namespace */ extern struct net init_net; #define INIT_NET_NS(net_ns) .net_ns = &init_net, -- cgit v1.2.3 From fadf6bf06069138f8e97c9a963be38348ba2708b Mon Sep 17 00:00:00 2001 From: "Templin, Fred L" Date: Tue, 11 Mar 2008 18:35:59 -0400 Subject: [IPV6] SIT: Add PRL management for ISATAP. This patch updates the Linux the Intra-Site Automatic Tunnel Addressing Protocol (ISATAP) implementation. It places the ISATAP potential router list (PRL) in the kernel and adds three new private ioctls for PRL management. [Add several changes of structure name, constant names etc. - yoshfuji] Signed-off-by: Fred L. Templin Signed-off-by: YOSHIFUJI Hideaki --- include/linux/if_tunnel.h | 18 ++++- include/linux/skbuff.h | 3 +- include/net/ipip.h | 7 ++ include/net/ndisc.h | 9 +++ net/ipv6/ndisc.c | 24 ++++++ net/ipv6/route.c | 2 - net/ipv6/sit.c | 186 +++++++++++++++++++++++++++++++++++----------- 7 files changed, 199 insertions(+), 50 deletions(-) (limited to 'include') diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index 228eb4eb3129..f20c224d544c 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -7,6 +7,9 @@ #define SIOCADDTUNNEL (SIOCDEVPRIVATE + 1) #define SIOCDELTUNNEL (SIOCDEVPRIVATE + 2) #define SIOCCHGTUNNEL (SIOCDEVPRIVATE + 3) +#define SIOCADDPRL (SIOCDEVPRIVATE + 5) +#define SIOCDELPRL (SIOCDEVPRIVATE + 6) +#define SIOCCHGPRL (SIOCDEVPRIVATE + 7) #define GRE_CSUM __constant_htons(0x8000) #define GRE_ROUTING __constant_htons(0x4000) @@ -17,9 +20,6 @@ #define GRE_FLAGS __constant_htons(0x00F8) #define GRE_VERSION __constant_htons(0x0007) -/* i_flags values for SIT mode */ -#define SIT_ISATAP 0x0001 - struct ip_tunnel_parm { char name[IFNAMSIZ]; @@ -31,4 +31,16 @@ struct ip_tunnel_parm struct iphdr iph; }; +/* SIT-mode i_flags */ +#define SIT_ISATAP 0x0001 + +struct ip_tunnel_prl { + __be32 addr; + __u16 flags; + __u16 __reserved; +}; + +/* PRL flags */ +#define PRL_DEFAULT 0x0001 + #endif /* _IF_TUNNEL_H_ */ diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ff72145d5d9e..e10e55c9b081 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -313,7 +313,8 @@ struct sk_buff { __u16 tc_verd; /* traffic control verdict */ #endif #endif - /* 2 byte hole */ + __u8 ndisc_nodetype:2; + /* 14 bit hole */ #ifdef CONFIG_NET_DMA dma_cookie_t dma_cookie; diff --git a/include/net/ipip.h b/include/net/ipip.h index 549e132bca9c..205536a014e8 100644 --- a/include/net/ipip.h +++ b/include/net/ipip.h @@ -24,6 +24,13 @@ struct ip_tunnel int mlink; struct ip_tunnel_parm parms; + struct ip_tunnel_prl_entry *prl; /* potential router list */ +}; + +struct ip_tunnel_prl_entry +{ + struct ip_tunnel_prl_entry *next; + struct ip_tunnel_prl entry; }; #define IPTUNNEL_XMIT() do { \ diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 5aedf324de66..9f2bae68d28c 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -11,6 +11,15 @@ #define NDISC_NEIGHBOUR_ADVERTISEMENT 136 #define NDISC_REDIRECT 137 +/* + * Router type: cross-layer information from link-layer to + * IPv6 layer reported by certain link types (e.g., RFC4214). + */ +#define NDISC_NODETYPE_UNSPEC 0 /* unspecified (default) */ +#define NDISC_NODETYPE_HOST 1 /* host or unauthorized router */ +#define NDISC_NODETYPE_NODEFAULT 2 /* non-default router */ +#define NDISC_NODETYPE_DEFAULT 3 /* default router */ + /* * ndisc options */ diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 510aa747a404..53b546019fd5 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1092,6 +1092,12 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } + if (skb->ndisc_nodetype == NDISC_NODETYPE_HOST) { + ND_PRINTK2(KERN_WARNING + "ICMPv6 RA: from host or unauthorized router\n"); + return; + } + /* * set the RA_RECV flag in the interface */ @@ -1115,6 +1121,10 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } + /* skip link-specific parameters from interior routers */ + if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT) + goto skip_linkparms; + if (in6_dev->if_flags & IF_RS_SENT) { /* * flag that an RA was received after an RS was sent @@ -1229,6 +1239,8 @@ skip_defrtr: } } +skip_linkparms: + /* * Process options. */ @@ -1268,6 +1280,10 @@ skip_defrtr: } #endif + /* skip link-specific ndopts from interior routers */ + if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT) + goto out; + if (in6_dev->cnf.accept_ra_pinfo && ndopts.nd_opts_pi) { struct nd_opt_hdr *p; for (p = ndopts.nd_opts_pi; @@ -1331,6 +1347,14 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) int optlen; u8 *lladdr = NULL; + switch (skb->ndisc_nodetype) { + case NDISC_NODETYPE_HOST: + case NDISC_NODETYPE_NODEFAULT: + ND_PRINTK2(KERN_WARNING + "ICMPv6 Redirect: from host or unauthorized router\n"); + return; + } + if (!(ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL)) { ND_PRINTK2(KERN_WARNING "ICMPv6 Redirect: source address is not link-local.\n"); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index cd82b6db35ff..f17b2f61891e 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1699,8 +1699,6 @@ struct rt6_info *rt6_get_dflt_router(struct in6_addr *addr, struct net_device *d return rt; } -EXPORT_SYMBOL(rt6_get_dflt_router); - struct rt6_info *rt6_add_dflt_router(struct in6_addr *gwaddr, struct net_device *dev, unsigned int pref) diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 1b8196c8d145..4786419ade0e 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -16,7 +16,7 @@ * Changes: * Roger Venning : 6to4 support * Nate Thompson : 6to4 support - * Fred L. Templin : isatap support + * Fred Templin : isatap support */ #include @@ -197,6 +197,119 @@ failed: return NULL; } +static struct ip_tunnel_prl_entry * +ipip6_tunnel_locate_prl(struct ip_tunnel *t, __be32 addr) +{ + struct ip_tunnel_prl_entry *p = (struct ip_tunnel_prl_entry *)NULL; + + for (p = t->prl; p; p = p->next) + if (p->entry.addr == addr) + break; + return p; + +} + +static int +ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg) +{ + struct ip_tunnel_prl_entry *p; + + for (p = t->prl; p; p = p->next) { + if (p->entry.addr == a->addr) { + if (chg) { + p->entry = *a; + return 0; + } + return -EEXIST; + } + } + + if (chg) + return -ENXIO; + + p = kzalloc(sizeof(struct ip_tunnel_prl_entry), GFP_KERNEL); + if (!p) + return -ENOBUFS; + + p->entry = *a; + p->next = t->prl; + t->prl = p; + return 0; +} + +static int +ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a) +{ + struct ip_tunnel_prl_entry *x, **p; + + if (a) { + for (p = &t->prl; *p; p = &(*p)->next) { + if ((*p)->entry.addr == a->addr) { + x = *p; + *p = x->next; + kfree(x); + return 0; + } + } + return -ENXIO; + } else { + while (t->prl) { + x = t->prl; + t->prl = t->prl->next; + kfree(x); + } + } + return 0; +} + +/* copied directly from anycast.c */ +static int +ipip6_onlink(struct in6_addr *addr, struct net_device *dev) +{ + struct inet6_dev *idev; + struct inet6_ifaddr *ifa; + int onlink; + + onlink = 0; + rcu_read_lock(); + idev = __in6_dev_get(dev); + if (idev) { + read_lock_bh(&idev->lock); + for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) { + onlink = ipv6_prefix_equal(addr, &ifa->addr, + ifa->prefix_len); + if (onlink) + break; + } + read_unlock_bh(&idev->lock); + } + rcu_read_unlock(); + return onlink; +} + +static int +isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t) +{ + struct ip_tunnel_prl_entry *p = ipip6_tunnel_locate_prl(t, iph->saddr); + int ok = 1; + + if (p) { + if (p->entry.flags & PRL_DEFAULT) + skb->ndisc_nodetype = NDISC_NODETYPE_DEFAULT; + else + skb->ndisc_nodetype = NDISC_NODETYPE_NODEFAULT; + } else { + struct in6_addr *addr6 = &ipv6_hdr(skb)->saddr; + if (ipv6_addr_is_isatap(addr6) && + (addr6->s6_addr32[3] == iph->saddr) && + ipip6_onlink(addr6, t->dev)) + skb->ndisc_nodetype = NDISC_NODETYPE_HOST; + else + ok = 0; + } + return ok; +} + static void ipip6_tunnel_uninit(struct net_device *dev) { if (dev == ipip6_fb_tunnel_dev) { @@ -206,6 +319,7 @@ static void ipip6_tunnel_uninit(struct net_device *dev) dev_put(dev); } else { ipip6_tunnel_unlink(netdev_priv(dev)); + ipip6_tunnel_del_prl(netdev_priv(dev), 0); dev_put(dev); } } @@ -365,48 +479,6 @@ static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb) IP6_ECN_set_ce(ipv6_hdr(skb)); } -/* ISATAP (RFC4214) - check source address */ -static int -isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev) -{ - struct neighbour *neigh; - struct dst_entry *dst; - struct rt6_info *rt; - struct flowi fl; - struct in6_addr *addr6; - struct in6_addr rtr; - struct ipv6hdr *iph6; - int ok = 0; - - /* from onlink default router */ - ipv6_addr_set(&rtr, htonl(0xFE800000), 0, 0, 0); - ipv6_isatap_eui64(rtr.s6_addr + 8, iph->saddr); - if ((rt = rt6_get_dflt_router(&rtr, dev))) { - dst_release(&rt->u.dst); - return 1; - } - - iph6 = ipv6_hdr(skb); - memset(&fl, 0, sizeof(fl)); - fl.proto = iph6->nexthdr; - ipv6_addr_copy(&fl.fl6_dst, &iph6->saddr); - fl.oif = dev->ifindex; - security_skb_classify_flow(skb, &fl); - - dst = ip6_route_output(&init_net, NULL, &fl); - if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) { - - addr6 = (struct in6_addr*)&neigh->primary_key; - - /* from correct previous hop */ - if (ipv6_addr_is_isatap(addr6) && - (addr6->s6_addr32[3] == iph->saddr)) - ok = 1; - } - dst_release(dst); - return ok; -} - static int ipip6_rcv(struct sk_buff *skb) { struct iphdr *iph; @@ -427,7 +499,7 @@ static int ipip6_rcv(struct sk_buff *skb) skb->pkt_type = PACKET_HOST; if ((tunnel->dev->priv_flags & IFF_ISATAP) && - !isatap_srcok(skb, iph, tunnel->dev)) { + !isatap_chksrc(skb, iph, tunnel)) { tunnel->stat.rx_errors++; read_unlock(&ipip6_lock); kfree_skb(skb); @@ -707,6 +779,7 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd) { int err = 0; struct ip_tunnel_parm p; + struct ip_tunnel_prl prl; struct ip_tunnel *t; switch (cmd) { @@ -806,6 +879,31 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd) err = 0; break; + case SIOCADDPRL: + case SIOCDELPRL: + case SIOCCHGPRL: + err = -EPERM; + if (!capable(CAP_NET_ADMIN)) + goto done; + err = -EINVAL; + if (dev == ipip6_fb_tunnel_dev) + goto done; + err = -EFAULT; + if (copy_from_user(&prl, ifr->ifr_ifru.ifru_data, sizeof(prl))) + goto done; + err = -ENOENT; + if (!(t = netdev_priv(dev))) + goto done; + + ipip6_tunnel_unlink(t); + if (cmd == SIOCDELPRL) + err = ipip6_tunnel_del_prl(t, &prl); + else + err = ipip6_tunnel_add_prl(t, &prl, cmd == SIOCCHGPRL); + ipip6_tunnel_link(t); + netdev_state_change(dev); + break; + default: err = -EINVAL; } -- cgit v1.2.3 From 300aaeeaab5f447fcf40e911afe96df3de28f0db Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Mon, 24 Mar 2008 18:28:39 +0900 Subject: [IPV6] SIT: Add SIOCGETPRL ioctl to get/dump PRL. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/if_tunnel.h | 4 ++ include/net/ipip.h | 5 ++- net/ipv6/sit.c | 96 ++++++++++++++++++++++++++++++++++++++++++----- 3 files changed, 95 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/if_tunnel.h b/include/linux/if_tunnel.h index f20c224d544c..f1fbe9c930d7 100644 --- a/include/linux/if_tunnel.h +++ b/include/linux/if_tunnel.h @@ -7,6 +7,7 @@ #define SIOCADDTUNNEL (SIOCDEVPRIVATE + 1) #define SIOCDELTUNNEL (SIOCDEVPRIVATE + 2) #define SIOCCHGTUNNEL (SIOCDEVPRIVATE + 3) +#define SIOCGETPRL (SIOCDEVPRIVATE + 4) #define SIOCADDPRL (SIOCDEVPRIVATE + 5) #define SIOCDELPRL (SIOCDEVPRIVATE + 6) #define SIOCCHGPRL (SIOCDEVPRIVATE + 7) @@ -38,6 +39,9 @@ struct ip_tunnel_prl { __be32 addr; __u16 flags; __u16 __reserved; + __u32 datalen; + __u32 __reserved2; + void __user *data; }; /* PRL flags */ diff --git a/include/net/ipip.h b/include/net/ipip.h index 205536a014e8..633ed4def8e3 100644 --- a/include/net/ipip.h +++ b/include/net/ipip.h @@ -24,13 +24,16 @@ struct ip_tunnel int mlink; struct ip_tunnel_parm parms; + struct ip_tunnel_prl_entry *prl; /* potential router list */ + unsigned int prl_count; /* # of entries in PRL */ }; struct ip_tunnel_prl_entry { struct ip_tunnel_prl_entry *next; - struct ip_tunnel_prl entry; + __be32 addr; + u16 flags; }; #define IPTUNNEL_XMIT() do { \ diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 84c1ed246afb..08a483a8de50 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -203,12 +203,73 @@ __ipip6_tunnel_locate_prl(struct ip_tunnel *t, __be32 addr) struct ip_tunnel_prl_entry *p = (struct ip_tunnel_prl_entry *)NULL; for (p = t->prl; p; p = p->next) - if (p->entry.addr == addr) + if (p->addr == addr) break; return p; } +static int ipip6_tunnel_get_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a) +{ + struct ip_tunnel_prl *kp; + struct ip_tunnel_prl_entry *prl; + unsigned int cmax, c = 0, ca, len; + int ret = 0; + + cmax = a->datalen / sizeof(*a); + if (cmax > 1 && a->addr != htonl(INADDR_ANY)) + cmax = 1; + + /* For simple GET or for root users, + * we try harder to allocate. + */ + kp = (cmax <= 1 || capable(CAP_NET_ADMIN)) ? + kcalloc(cmax, sizeof(*kp), GFP_KERNEL) : + NULL; + + read_lock(&ipip6_lock); + + ca = t->prl_count < cmax ? t->prl_count : cmax; + + if (!kp) { + /* We don't try hard to allocate much memory for + * non-root users. + * For root users, retry allocating enough memory for + * the answer. + */ + kp = kcalloc(ca, sizeof(*kp), GFP_ATOMIC); + if (!kp) { + ret = -ENOMEM; + goto out; + } + } + + c = 0; + for (prl = t->prl; prl; prl = prl->next) { + if (c > cmax) + break; + if (a->addr != htonl(INADDR_ANY) && prl->addr != a->addr) + continue; + kp[c].addr = prl->addr; + kp[c].flags = prl->flags; + c++; + if (a->addr != htonl(INADDR_ANY)) + break; + } +out: + read_unlock(&ipip6_lock); + + len = sizeof(*kp) * c; + ret = len ? copy_to_user(a->data, kp, len) : 0; + + kfree(kp); + if (ret) + return -EFAULT; + + a->datalen = len; + return 0; +} + static int ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg) { @@ -221,7 +282,7 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg) write_lock(&ipip6_lock); for (p = t->prl; p; p = p->next) { - if (p->entry.addr == a->addr) { + if (p->addr == a->addr) { if (chg) goto update; err = -EEXIST; @@ -242,8 +303,10 @@ ipip6_tunnel_add_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a, int chg) p->next = t->prl; t->prl = p; + t->prl_count++; update: - p->entry = *a; + p->addr = a->addr; + p->flags = a->flags; out: write_unlock(&ipip6_lock); return err; @@ -259,10 +322,11 @@ ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a) if (a && a->addr != htonl(INADDR_ANY)) { for (p = &t->prl; *p; p = &(*p)->next) { - if ((*p)->entry.addr == a->addr) { + if ((*p)->addr == a->addr) { x = *p; *p = x->next; kfree(x); + t->prl_count--; goto out; } } @@ -272,6 +336,7 @@ ipip6_tunnel_del_prl(struct ip_tunnel *t, struct ip_tunnel_prl *a) x = t->prl; t->prl = t->prl->next; kfree(x); + t->prl_count--; } } out: @@ -313,7 +378,7 @@ isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t) read_lock(&ipip6_lock); p = __ipip6_tunnel_locate_prl(t, iph->saddr); if (p) { - if (p->entry.flags & PRL_DEFAULT) + if (p->flags & PRL_DEFAULT) skb->ndisc_nodetype = NDISC_NODETYPE_DEFAULT; else skb->ndisc_nodetype = NDISC_NODETYPE_NODEFAULT; @@ -899,11 +964,12 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd) err = 0; break; + case SIOCGETPRL: case SIOCADDPRL: case SIOCDELPRL: case SIOCCHGPRL: err = -EPERM; - if (!capable(CAP_NET_ADMIN)) + if (cmd != SIOCGETPRL && !capable(CAP_NET_ADMIN)) goto done; err = -EINVAL; if (dev == ipip6_fb_tunnel_dev) @@ -915,11 +981,23 @@ ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd) if (!(t = netdev_priv(dev))) goto done; - if (cmd == SIOCDELPRL) + switch (cmd) { + case SIOCGETPRL: + err = ipip6_tunnel_get_prl(t, &prl); + if (!err && copy_to_user(ifr->ifr_ifru.ifru_data, + &prl, sizeof(prl))) + err = -EFAULT; + break; + case SIOCDELPRL: err = ipip6_tunnel_del_prl(t, &prl); - else + break; + case SIOCADDPRL: + case SIOCCHGPRL: err = ipip6_tunnel_add_prl(t, &prl, cmd == SIOCCHGPRL); - netdev_state_change(dev); + break; + } + if (cmd != SIOCGETPRL) + netdev_state_change(dev); break; default: -- cgit v1.2.3 From 52eeeb8481d705e61e2e9aae974e7799a93783e9 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sat, 15 Mar 2008 22:54:23 -0400 Subject: [IPV6]: Unify ip6_onlink() and ipip6_onlink(). Both are identical, let's create ipv6_chk_prefix() and use it in both places. --- include/net/addrconf.h | 4 ++++ net/ipv6/addrconf.c | 25 +++++++++++++++++++++++++ net/ipv6/anycast.c | 25 +------------------------ net/ipv6/sit.c | 27 +-------------------------- 4 files changed, 31 insertions(+), 50 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index d89b0bc7ab75..bdcc863a60a4 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -71,6 +71,10 @@ extern int ipv6_chk_addr(struct net *net, extern int ipv6_chk_home_addr(struct net *net, struct in6_addr *addr); #endif + +extern int ipv6_chk_prefix(struct in6_addr *addr, + struct net_device *dev); + extern struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr, struct net_device *dev, diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5ab9973571ef..c17f8c0b933e 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1249,6 +1249,31 @@ int ipv6_chk_same_addr(struct net *net, const struct in6_addr *addr, return ifp != NULL; } +int ipv6_chk_prefix(struct in6_addr *addr, struct net_device *dev) +{ + struct inet6_dev *idev; + struct inet6_ifaddr *ifa; + int onlink; + + onlink = 0; + rcu_read_lock(); + idev = __in6_dev_get(dev); + if (idev) { + read_lock_bh(&idev->lock); + for (ifa = idev->addr_list; ifa; ifa = ifa->if_next) { + onlink = ipv6_prefix_equal(addr, &ifa->addr, + ifa->prefix_len); + if (onlink) + break; + } + read_unlock_bh(&idev->lock); + } + rcu_read_unlock(); + return onlink; +} + +EXPORT_SYMBOL(ipv6_chk_prefix); + struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr, struct net_device *dev, int strict) { diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 463bd95d6b13..36e817492095 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -48,29 +48,6 @@ static int ipv6_dev_ac_dec(struct net_device *dev, struct in6_addr *addr); /* Big ac list lock for all the sockets */ static DEFINE_RWLOCK(ipv6_sk_ac_lock); -static int -ip6_onlink(struct in6_addr *addr, struct net_device *dev) -{ - struct inet6_dev *idev; - struct inet6_ifaddr *ifa; - int onlink; - - onlink = 0; - rcu_read_lock(); - idev = __in6_dev_get(dev); - if (idev) { - read_lock_bh(&idev->lock); - for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) { - onlink = ipv6_prefix_equal(addr, &ifa->addr, - ifa->prefix_len); - if (onlink) - break; - } - read_unlock_bh(&idev->lock); - } - rcu_read_unlock(); - return onlink; -} /* * socket join an anycast group @@ -142,7 +119,7 @@ int ipv6_sock_ac_join(struct sock *sk, int ifindex, struct in6_addr *addr) * This obviates the need for propagating anycast routes while * still allowing some non-router anycast participation. */ - if (!ip6_onlink(addr, dev)) { + if (!ipv6_chk_prefix(addr, dev)) { if (ishost) err = -EADDRNOTAVAIL; if (err) diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 08a483a8de50..cc16fe07bbff 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -344,31 +344,6 @@ out: return 0; } -/* copied directly from anycast.c */ -static int -ipip6_onlink(struct in6_addr *addr, struct net_device *dev) -{ - struct inet6_dev *idev; - struct inet6_ifaddr *ifa; - int onlink; - - onlink = 0; - rcu_read_lock(); - idev = __in6_dev_get(dev); - if (idev) { - read_lock_bh(&idev->lock); - for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) { - onlink = ipv6_prefix_equal(addr, &ifa->addr, - ifa->prefix_len); - if (onlink) - break; - } - read_unlock_bh(&idev->lock); - } - rcu_read_unlock(); - return onlink; -} - static int isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t) { @@ -386,7 +361,7 @@ isatap_chksrc(struct sk_buff *skb, struct iphdr *iph, struct ip_tunnel *t) struct in6_addr *addr6 = &ipv6_hdr(skb)->saddr; if (ipv6_addr_is_isatap(addr6) && (addr6->s6_addr32[3] == iph->saddr) && - ipip6_onlink(addr6, t->dev)) + ipv6_chk_prefix(addr6, t->dev)) skb->ndisc_nodetype = NDISC_NODETYPE_HOST; else ok = 0; -- cgit v1.2.3 From de357cc01334a468e4d5b7ba66a17b0d3ca9d63e Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sat, 15 Mar 2008 23:59:18 -0400 Subject: [IPV6] NDISC: Don't rely on node-type hint from L2 unless required. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/skbuff.h | 2 ++ net/ipv6/Kconfig | 4 ++++ net/ipv6/ndisc.c | 10 ++++++++++ 3 files changed, 16 insertions(+) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index e10e55c9b081..e517701c25ba 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -313,7 +313,9 @@ struct sk_buff { __u16 tc_verd; /* traffic control verdict */ #endif #endif +#ifdef CONFIG_IPV6_NDISC_NODETYPE __u8 ndisc_nodetype:2; +#endif /* 14 bit hole */ #ifdef CONFIG_NET_DMA diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 47263e45bacb..7d2e7f0941ac 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -168,6 +168,7 @@ config IPV6_SIT tristate "IPv6: IPv6-in-IPv4 tunnel (SIT driver)" depends on IPV6 select INET_TUNNEL + select IPV6_NDISC_NODETYPE default y ---help--- Tunneling means encapsulating data of one protocol type within @@ -178,6 +179,9 @@ config IPV6_SIT Saying M here will produce a module called sit.ko. If unsure, say Y. +config IPV6_NDISC_NODETYPE + bool + config IPV6_TUNNEL tristate "IPv6: IP-in-IPv6 tunnel (RFC2473)" select INET6_TUNNEL diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 16273e11e53d..c400b874097a 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1092,11 +1092,13 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } +#ifdef CONFIG_IPV6_NDISC_NODETYPE if (skb->ndisc_nodetype == NDISC_NODETYPE_HOST) { ND_PRINTK2(KERN_WARNING "ICMPv6 RA: from host or unauthorized router\n"); return; } +#endif /* * set the RA_RECV flag in the interface @@ -1121,9 +1123,11 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } +#ifdef CONFIG_IPV6_NDISC_NODETYPE /* skip link-specific parameters from interior routers */ if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT) goto skip_linkparms; +#endif if (in6_dev->if_flags & IF_RS_SENT) { /* @@ -1239,7 +1243,9 @@ skip_defrtr: } } +#ifdef CONFIG_IPV6_NDISC_NODETYPE skip_linkparms: +#endif /* * Process options. @@ -1286,9 +1292,11 @@ skip_linkparms: } #endif +#ifdef CONFIG_IPV6_NDISC_NODETYPE /* skip link-specific ndopts from interior routers */ if (skb->ndisc_nodetype == NDISC_NODETYPE_NODEFAULT) goto out; +#endif if (in6_dev->cnf.accept_ra_pinfo && ndopts.nd_opts_pi) { struct nd_opt_hdr *p; @@ -1353,6 +1361,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) int optlen; u8 *lladdr = NULL; +#ifdef CONFIG_IPV6_NDISC_NODETYPE switch (skb->ndisc_nodetype) { case NDISC_NODETYPE_HOST: case NDISC_NODETYPE_NODEFAULT: @@ -1360,6 +1369,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) "ICMPv6 Redirect: from host or unauthorized router\n"); return; } +#endif if (!(ipv6_addr_type(&ipv6_hdr(skb)->saddr) & IPV6_ADDR_LINKLOCAL)) { ND_PRINTK2(KERN_WARNING -- cgit v1.2.3 From a4aa834a9165150252c5cd953faab4de29d51b87 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 3 Apr 2008 13:04:33 -0700 Subject: [NETNS]: Declare init_net even without CONFIG_NET defined. This does not look good, but there is no other choice. The compilation without CONFIG_NET is broken and can not be fixed with ease. After that there is no need for the following commits: 1567ca7eec7664b8be3b07755ac59dc1b1ec76cb 3edf8fa5ccf10688a9280b5cbca8ed3947c42866 2d38f9a4f8d2ebdc799f03eecf82345825495711 Revert them. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/linux/netdevice.h | 4 ---- include/net/net_namespace.h | 3 ++- lib/kobject_uevent.c | 2 -- 3 files changed, 2 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index c36c76caf20b..8b17ed40dea2 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -741,7 +741,6 @@ struct net_device #define NETDEV_ALIGN 32 #define NETDEV_ALIGN_CONST (NETDEV_ALIGN - 1) -#ifdef CONFIG_NET /* * Net namespace inlines */ @@ -762,7 +761,6 @@ void dev_net_set(struct net_device *dev, struct net *net) dev->nd_net = net; #endif } -#endif /** * netdev_priv - access network device private data @@ -827,7 +825,6 @@ struct packet_type { extern rwlock_t dev_base_lock; /* Device list lock */ -#ifdef CONFIG_NET #define for_each_netdev(net, d) \ list_for_each_entry(d, &(net)->dev_base_head, dev_list) #define for_each_netdev_safe(net, d, n) \ @@ -851,7 +848,6 @@ static inline struct net_device *first_net_device(struct net *net) return list_empty(&net->dev_base_head) ? NULL : net_device_entry(net->dev_base_head.next); } -#endif extern int netdev_boot_setup_check(struct net_device *dev); extern unsigned long netdev_boot_base(const char *prefix, int unit); diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 6c9a48a46685..0ab62ed2fdef 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -60,11 +60,12 @@ struct net { }; -#ifdef CONFIG_NET #include /* Init's network namespace */ extern struct net init_net; + +#ifdef CONFIG_NET #define INIT_NET_NS(net_ns) .net_ns = &init_net, extern struct net *copy_net_ns(unsigned long flags, struct net *net_ns); diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c index 0d56dad319ad..b06185ed1895 100644 --- a/lib/kobject_uevent.c +++ b/lib/kobject_uevent.c @@ -19,12 +19,10 @@ #include #include -#ifdef CONFIG_NET #include #include #include #include -#endif u64 uevent_seqnum; -- cgit v1.2.3 From 3d58b5fa8e4c461ab09afdacd3d1754fccca06ad Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 3 Apr 2008 14:22:32 -0700 Subject: [INET]: Rename inet_csk_ctl_sock_create to inet_ctl_sock_create. This call is nothing common with INET connection sockets code. It simply creates an unhashes kernel sockets for protocol messages. Move the new call into af_inet.c after the rename. Signed-off-by: Denis V. Lunev Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/inet_common.h | 5 +++++ include/net/inet_connection_sock.h | 5 ----- net/dccp/ipv4.c | 4 ++-- net/dccp/ipv6.c | 4 ++-- net/ipv4/af_inet.c | 19 +++++++++++++++++++ net/ipv4/inet_connection_sock.c | 19 ------------------- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv6/tcp_ipv6.c | 3 ++- 8 files changed, 32 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/net/inet_common.h b/include/net/inet_common.h index 38d5a1e9980d..d6238bdefbaf 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -39,6 +39,11 @@ extern int inet_getname(struct socket *sock, extern int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg); +extern int inet_ctl_sock_create(struct socket **sock, + unsigned short family, + unsigned short type, + unsigned char protocol); + #endif diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index f00f0573627b..2ff545a56fb5 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -327,11 +327,6 @@ extern void inet_csk_listen_stop(struct sock *sk); extern void inet_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr); -extern int inet_csk_ctl_sock_create(struct socket **sock, - unsigned short family, - unsigned short type, - unsigned char protocol); - extern int inet_csk_compat_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen); extern int inet_csk_compat_setsockopt(struct sock *sk, int level, int optname, diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 6d8f684a7a49..feb3fa5b7141 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1003,8 +1003,8 @@ static int __init dccp_v4_init(void) inet_register_protosw(&dccp_v4_protosw); - err = inet_csk_ctl_sock_create(&socket, PF_INET, - SOCK_DCCP, IPPROTO_DCCP); + err = inet_ctl_sock_create(&socket, PF_INET, + SOCK_DCCP, IPPROTO_DCCP); if (err) goto out_unregister_protosw; dccp_v4_ctl_sk = socket->sk; diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index c5d9d1be56f3..5690fbd3bf68 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -1185,8 +1185,8 @@ static int __init dccp_v6_init(void) inet6_register_protosw(&dccp_v6_protosw); - err = inet_csk_ctl_sock_create(&socket, PF_INET6, - SOCK_DCCP, IPPROTO_DCCP); + err = inet_ctl_sock_create(&socket, PF_INET6, + SOCK_DCCP, IPPROTO_DCCP); if (err != 0) goto out_unregister_protosw; dccp_v6_ctl_sk = socket->sk; diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 5882a1316441..7ab0bd64c9d1 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1250,6 +1250,25 @@ out: return segs; } +int inet_ctl_sock_create(struct socket **sock, unsigned short family, + unsigned short type, unsigned char protocol) +{ + int rc = sock_create_kern(family, type, protocol, sock); + + if (rc == 0) { + (*sock)->sk->sk_allocation = GFP_ATOMIC; + inet_sk((*sock)->sk)->uc_ttl = -1; + /* + * Unhash it so that IP input processing does not even see it, + * we do not wish this socket to see incoming packets. + */ + (*sock)->sk->sk_prot->unhash((*sock)->sk); + } + return rc; +} + +EXPORT_SYMBOL_GPL(inet_ctl_sock_create); + unsigned long snmp_fold_field(void *mib[], int offt) { unsigned long res = 0; diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index a7fcaf205644..ee55678a987d 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -651,25 +651,6 @@ void inet_csk_addr2sockaddr(struct sock *sk, struct sockaddr *uaddr) EXPORT_SYMBOL_GPL(inet_csk_addr2sockaddr); -int inet_csk_ctl_sock_create(struct socket **sock, unsigned short family, - unsigned short type, unsigned char protocol) -{ - int rc = sock_create_kern(family, type, protocol, sock); - - if (rc == 0) { - (*sock)->sk->sk_allocation = GFP_ATOMIC; - inet_sk((*sock)->sk)->uc_ttl = -1; - /* - * Unhash it so that IP input processing does not even see it, - * we do not wish this socket to see incoming packets. - */ - (*sock)->sk->sk_prot->unhash((*sock)->sk); - } - return rc; -} - -EXPORT_SYMBOL_GPL(inet_csk_ctl_sock_create); - #ifdef CONFIG_COMPAT int inet_csk_compat_getsockopt(struct sock *sk, int level, int optname, char __user *optval, int __user *optlen) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 1d77f37d7708..edf5a37bb5c3 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2491,8 +2491,8 @@ struct proto tcp_prot = { void __init tcp_v4_init(void) { struct socket *__tcp_socket; - if (inet_csk_ctl_sock_create(&__tcp_socket, PF_INET, SOCK_RAW, - IPPROTO_TCP) < 0) + if (inet_ctl_sock_create(&__tcp_socket, PF_INET, SOCK_RAW, + IPPROTO_TCP) < 0) panic("Failed to create the TCP control socket.\n"); tcp_sock = __tcp_socket->sk; } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 6d851c3c3db9..d98222fba041 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -60,6 +60,7 @@ #include #include #include +#include #include @@ -2202,7 +2203,7 @@ static int tcpv6_net_init(struct net *net) struct socket *sock; struct sock *sk; - err = inet_csk_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP); + err = inet_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP); if (err) return err; -- cgit v1.2.3 From eee4fe4ded6e9c196168aee8f9787771f4df9c90 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 3 Apr 2008 14:27:58 -0700 Subject: [INET]: Let inet_ctl_sock_create return sock rather than socket. All upper protocol layers are already use sock internally. Signed-off-by: Denis V. Lunev Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/inet_common.h | 2 +- net/dccp/ipv4.c | 4 +--- net/dccp/ipv6.c | 4 +--- net/ipv4/af_inet.c | 12 +++++++----- net/ipv4/tcp_ipv4.c | 4 +--- net/ipv6/tcp_ipv6.c | 5 ++--- net/sctp/protocol.c | 4 +--- 7 files changed, 14 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/net/inet_common.h b/include/net/inet_common.h index d6238bdefbaf..4bfcf3f3555f 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -39,7 +39,7 @@ extern int inet_getname(struct socket *sock, extern int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg); -extern int inet_ctl_sock_create(struct socket **sock, +extern int inet_ctl_sock_create(struct sock **sk, unsigned short family, unsigned short type, unsigned char protocol); diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index feb3fa5b7141..5669c895c873 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -991,7 +991,6 @@ static struct inet_protosw dccp_v4_protosw = { static int __init dccp_v4_init(void) { - struct socket *socket; int err = proto_register(&dccp_v4_prot, 1); if (err != 0) @@ -1003,11 +1002,10 @@ static int __init dccp_v4_init(void) inet_register_protosw(&dccp_v4_protosw); - err = inet_ctl_sock_create(&socket, PF_INET, + err = inet_ctl_sock_create(&dccp_v4_ctl_sk, PF_INET, SOCK_DCCP, IPPROTO_DCCP); if (err) goto out_unregister_protosw; - dccp_v4_ctl_sk = socket->sk; out: return err; out_unregister_protosw: diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 5690fbd3bf68..cf598bfc6a18 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -1173,7 +1173,6 @@ static struct inet_protosw dccp_v6_protosw = { static int __init dccp_v6_init(void) { - struct socket *socket; int err = proto_register(&dccp_v6_prot, 1); if (err != 0) @@ -1185,11 +1184,10 @@ static int __init dccp_v6_init(void) inet6_register_protosw(&dccp_v6_protosw); - err = inet_ctl_sock_create(&socket, PF_INET6, + err = inet_ctl_sock_create(&dccp_v6_ctl_sk, PF_INET6, SOCK_DCCP, IPPROTO_DCCP); if (err != 0) goto out_unregister_protosw; - dccp_v6_ctl_sk = socket->sk; out: return err; out_unregister_protosw: diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 7ab0bd64c9d1..cad664bf3f2e 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1250,19 +1250,21 @@ out: return segs; } -int inet_ctl_sock_create(struct socket **sock, unsigned short family, +int inet_ctl_sock_create(struct sock **sk, unsigned short family, unsigned short type, unsigned char protocol) { - int rc = sock_create_kern(family, type, protocol, sock); + struct socket *sock; + int rc = sock_create_kern(family, type, protocol, &sock); if (rc == 0) { - (*sock)->sk->sk_allocation = GFP_ATOMIC; - inet_sk((*sock)->sk)->uc_ttl = -1; + *sk = sock->sk; + (*sk)->sk_allocation = GFP_ATOMIC; + inet_sk(*sk)->uc_ttl = -1; /* * Unhash it so that IP input processing does not even see it, * we do not wish this socket to see incoming packets. */ - (*sock)->sk->sk_prot->unhash((*sock)->sk); + (*sk)->sk_prot->unhash(*sk); } return rc; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index edf5a37bb5c3..cfe5df76e14b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2490,11 +2490,9 @@ struct proto tcp_prot = { void __init tcp_v4_init(void) { - struct socket *__tcp_socket; - if (inet_ctl_sock_create(&__tcp_socket, PF_INET, SOCK_RAW, + if (inet_ctl_sock_create(&tcp_sock, PF_INET, SOCK_RAW, IPPROTO_TCP) < 0) panic("Failed to create the TCP control socket.\n"); - tcp_sock = __tcp_socket->sk; } EXPORT_SYMBOL(ipv4_specific); diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index d98222fba041..2882cc51669e 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2200,14 +2200,13 @@ static struct inet_protosw tcpv6_protosw = { static int tcpv6_net_init(struct net *net) { int err; - struct socket *sock; struct sock *sk; - err = inet_ctl_sock_create(&sock, PF_INET6, SOCK_RAW, IPPROTO_TCP); + err = inet_ctl_sock_create(&sk, PF_INET6, SOCK_RAW, IPPROTO_TCP); if (err) return err; - net->ipv6.tcp_sk = sk = sock->sk; + net->ipv6.tcp_sk = sk; sk_change_net(sk, net); return err; } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index ac0833c19450..3c08d334d4a8 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -674,21 +674,19 @@ static int sctp_ctl_sock_init(void) { int err; sa_family_t family; - struct socket *socket; if (sctp_get_pf_specific(PF_INET6)) family = PF_INET6; else family = PF_INET; - err = inet_ctl_sock_create(&socket, family, + err = inet_ctl_sock_create(&sctp_ctl_sock, family, SOCK_SEQPACKET, IPPROTO_SCTP); if (err < 0) { printk(KERN_ERR "SCTP: Failed to create the SCTP control socket.\n"); return err; } - sctp_ctl_sock = socket->sk; return 0; } -- cgit v1.2.3 From 5677242f432102dea9e6eceec1dc089e2f709ca4 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 3 Apr 2008 14:28:30 -0700 Subject: [NETNS]: Inet control socket should not hold a namespace. This is a generic requirement, so make inet_ctl_sock_create namespace aware and create a inet_ctl_sock_destroy wrapper around sk_release_kernel. Signed-off-by: Denis V. Lunev Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/inet_common.h | 8 +++++++- net/dccp/ipv4.c | 4 ++-- net/dccp/ipv6.c | 4 ++-- net/ipv4/af_inet.c | 5 ++++- net/ipv4/tcp_ipv4.c | 2 +- net/ipv6/tcp_ipv6.c | 14 +++----------- net/sctp/protocol.c | 6 +++--- 7 files changed, 22 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/net/inet_common.h b/include/net/inet_common.h index 4bfcf3f3555f..18c773286b91 100644 --- a/include/net/inet_common.h +++ b/include/net/inet_common.h @@ -42,7 +42,13 @@ extern int inet_ioctl(struct socket *sock, extern int inet_ctl_sock_create(struct sock **sk, unsigned short family, unsigned short type, - unsigned char protocol); + unsigned char protocol, + struct net *net); + +static inline void inet_ctl_sock_destroy(struct sock *sk) +{ + sk_release_kernel(sk); +} #endif diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 5669c895c873..b12803bcba56 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -1003,7 +1003,7 @@ static int __init dccp_v4_init(void) inet_register_protosw(&dccp_v4_protosw); err = inet_ctl_sock_create(&dccp_v4_ctl_sk, PF_INET, - SOCK_DCCP, IPPROTO_DCCP); + SOCK_DCCP, IPPROTO_DCCP, &init_net); if (err) goto out_unregister_protosw; out: @@ -1018,7 +1018,7 @@ out_proto_unregister: static void __exit dccp_v4_exit(void) { - sock_release(dccp_v4_ctl_sk->sk_socket); + inet_ctl_sock_destroy(dccp_v4_ctl_sk); inet_unregister_protosw(&dccp_v4_protosw); inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP); proto_unregister(&dccp_v4_prot); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index cf598bfc6a18..94d749e6d494 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -1185,7 +1185,7 @@ static int __init dccp_v6_init(void) inet6_register_protosw(&dccp_v6_protosw); err = inet_ctl_sock_create(&dccp_v6_ctl_sk, PF_INET6, - SOCK_DCCP, IPPROTO_DCCP); + SOCK_DCCP, IPPROTO_DCCP, &init_net); if (err != 0) goto out_unregister_protosw; out: @@ -1200,7 +1200,7 @@ out_unregister_proto: static void __exit dccp_v6_exit(void) { - sock_release(dccp_v6_ctl_sk->sk_socket); + inet_ctl_sock_destroy(dccp_v6_ctl_sk); inet6_del_protocol(&dccp_v6_protocol, IPPROTO_DCCP); inet6_unregister_protosw(&dccp_v6_protosw); proto_unregister(&dccp_v6_prot); diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index cad664bf3f2e..cf766ad15776 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1251,7 +1251,8 @@ out: } int inet_ctl_sock_create(struct sock **sk, unsigned short family, - unsigned short type, unsigned char protocol) + unsigned short type, unsigned char protocol, + struct net *net) { struct socket *sock; int rc = sock_create_kern(family, type, protocol, &sock); @@ -1265,6 +1266,8 @@ int inet_ctl_sock_create(struct sock **sk, unsigned short family, * we do not wish this socket to see incoming packets. */ (*sk)->sk_prot->unhash(*sk); + + sk_change_net(*sk, net); } return rc; } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index cfe5df76e14b..dc8c3dc75fe5 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2491,7 +2491,7 @@ struct proto tcp_prot = { void __init tcp_v4_init(void) { if (inet_ctl_sock_create(&tcp_sock, PF_INET, SOCK_RAW, - IPPROTO_TCP) < 0) + IPPROTO_TCP, &init_net) < 0) panic("Failed to create the TCP control socket.\n"); } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 2882cc51669e..378cc4002a76 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2199,21 +2199,13 @@ static struct inet_protosw tcpv6_protosw = { static int tcpv6_net_init(struct net *net) { - int err; - struct sock *sk; - - err = inet_ctl_sock_create(&sk, PF_INET6, SOCK_RAW, IPPROTO_TCP); - if (err) - return err; - - net->ipv6.tcp_sk = sk; - sk_change_net(sk, net); - return err; + return inet_ctl_sock_create(&net->ipv6.tcp_sk, PF_INET6, + SOCK_RAW, IPPROTO_TCP, net); } static void tcpv6_net_exit(struct net *net) { - sk_release_kernel(net->ipv6.tcp_sk); + inet_ctl_sock_destroy(net->ipv6.tcp_sk); } static struct pernet_operations tcpv6_net_ops = { diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index 3c08d334d4a8..067c8a1658d6 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -681,7 +681,7 @@ static int sctp_ctl_sock_init(void) family = PF_INET; err = inet_ctl_sock_create(&sctp_ctl_sock, family, - SOCK_SEQPACKET, IPPROTO_SCTP); + SOCK_SEQPACKET, IPPROTO_SCTP, &init_net); if (err < 0) { printk(KERN_ERR "SCTP: Failed to create the SCTP control socket.\n"); @@ -1284,7 +1284,7 @@ err_v6_add_protocol: sctp_v6_del_protocol(); err_add_protocol: sctp_v4_del_protocol(); - sock_release(sctp_ctl_sock->sk_socket); + inet_ctl_sock_destroy(sctp_ctl_sock); err_ctl_sock_init: sctp_v6_protosw_exit(); err_v6_protosw_init: @@ -1328,7 +1328,7 @@ SCTP_STATIC __exit void sctp_exit(void) sctp_v4_del_protocol(); /* Free the control endpoint. */ - sock_release(sctp_ctl_sock->sk_socket); + inet_ctl_sock_destroy(sctp_ctl_sock); /* Free protosw registrations */ sctp_v6_protosw_exit(); -- cgit v1.2.3 From 046ee902357adc046d041441956ec7eeb30c77c4 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Thu, 3 Apr 2008 14:31:33 -0700 Subject: [NETNS]: Create tcp control socket in the each namespace. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/netns/ipv4.h | 1 + net/ipv4/tcp_ipv4.c | 21 +++++++++++++++++++-- 2 files changed, 20 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index af685f71f4b5..34ee348a2cf2 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -28,6 +28,7 @@ struct netns_ipv4 { struct sock *fibnl; struct sock **icmp_sk; + struct sock *tcp_sock; struct netns_frags frags; #ifdef CONFIG_NETFILTER diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index dc8c3dc75fe5..1d4a77acdc0b 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2488,11 +2488,28 @@ struct proto tcp_prot = { #endif }; + +static int __net_init tcp_sk_init(struct net *net) +{ + return inet_ctl_sock_create(&net->ipv4.tcp_sock, + PF_INET, SOCK_RAW, IPPROTO_TCP, net); +} + +static void __net_exit tcp_sk_exit(struct net *net) +{ + inet_ctl_sock_destroy(net->ipv4.tcp_sock); +} + +static struct pernet_operations __net_initdata tcp_sk_ops = { + .init = tcp_sk_init, + .exit = tcp_sk_exit, +}; + void __init tcp_v4_init(void) { - if (inet_ctl_sock_create(&tcp_sock, PF_INET, SOCK_RAW, - IPPROTO_TCP, &init_net) < 0) + if (register_pernet_device(&tcp_sk_ops)) panic("Failed to create the TCP control socket.\n"); + tcp_sock = init_net.ipv4.tcp_sock; } EXPORT_SYMBOL(ipv4_specific); -- cgit v1.2.3 From 2e8046271f68198dd37451017c1a4a2432e4ec68 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 3 Apr 2008 09:22:09 +0900 Subject: [IPV4] MROUTE: Move PIM definitions to . Signed-off-by: YOSHIFUJI Hideaki --- include/linux/Kbuild | 1 + include/linux/mroute.h | 22 +--------------------- include/linux/pim.h | 29 +++++++++++++++++++++++++++++ 3 files changed, 31 insertions(+), 21 deletions(-) create mode 100644 include/linux/pim.h (limited to 'include') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 9cdd12a9e843..84736acb4b99 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -289,6 +289,7 @@ unifdef-y += parport.h unifdef-y += patchkey.h unifdef-y += pci.h unifdef-y += personality.h +unifdef-y += pim.h unifdef-y += pktcdvd.h unifdef-y += pmu.h unifdef-y += poll.h diff --git a/include/linux/mroute.h b/include/linux/mroute.h index 35a8277ec1bd..c41b4217ae3b 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -3,6 +3,7 @@ #include #include +#include /* * Based on the MROUTING 3.5 defines primarily to keep @@ -210,27 +211,6 @@ struct mfc_cache #define IGMPMSG_WHOLEPKT 3 /* For PIM Register processing */ #ifdef __KERNEL__ - -#define PIM_V1_VERSION __constant_htonl(0x10000000) -#define PIM_V1_REGISTER 1 - -#define PIM_VERSION 2 -#define PIM_REGISTER 1 - -#define PIM_NULL_REGISTER __constant_htonl(0x40000000) - -/* PIMv2 register message header layout (ietf-draft-idmr-pimvsm-v2-00.ps */ - -struct pimreghdr -{ - __u8 type; - __u8 reserved; - __be16 csum; - __be32 flags; -}; - -extern int pim_rcv_v1(struct sk_buff *); - struct rtmsg; extern int ipmr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait); #endif diff --git a/include/linux/pim.h b/include/linux/pim.h new file mode 100644 index 000000000000..6f689dc85503 --- /dev/null +++ b/include/linux/pim.h @@ -0,0 +1,29 @@ +#ifndef __LINUX_PIM_H +#define __LINUX_PIM_H + +#include + +/* Message types - V1 */ +#define PIM_V1_VERSION __constant_htonl(0x10000000) +#define PIM_V1_REGISTER 1 + +/* Message types - V2 */ +#define PIM_VERSION 2 +#define PIM_REGISTER 1 + +#if defined(__KERNEL__) +#define PIM_NULL_REGISTER __constant_htonl(0x40000000) + +/* PIMv2 register message header layout (ietf-draft-idmr-pimvsm-v2-00.ps */ +struct pimreghdr +{ + __u8 type; + __u8 reserved; + __be16 csum; + __be32 flags; +}; + +struct sk_buff; +extern int pim_rcv_v1(struct sk_buff *); +#endif +#endif -- cgit v1.2.3 From 80a9492a33dd7d852465625022d56ff76d62174d Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 3 Apr 2008 09:22:52 +0900 Subject: [IPV4] MROUTE: Adjust include files for user-space. needs . Avoid including in user-space, which conflicts with standard . Add basic struct and constant in . Signed-off-by: YOSHIFUJI Hideaki --- include/linux/mroute.h | 3 +++ include/linux/pim.h | 16 ++++++++++++++++ 2 files changed, 19 insertions(+) (limited to 'include') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index c41b4217ae3b..de4decfa1bfc 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -2,7 +2,10 @@ #define __LINUX_MROUTE_H #include +#include +#ifdef __KERNEL__ #include +#endif #include /* diff --git a/include/linux/pim.h b/include/linux/pim.h index 6f689dc85503..236ffd317394 100644 --- a/include/linux/pim.h +++ b/include/linux/pim.h @@ -3,6 +3,22 @@ #include +#ifndef __KERNEL__ +struct pim { +#if defined(__LITTLE_ENDIAN_BITFIELD) + __u8 pim_type:4, /* PIM message type */ + pim_ver:4; /* PIM version */ +#elif defined(__BIG_ENDIAN_BITFIELD) + __u8 pim_ver:4; /* PIM version */ + pim_type:4; /* PIM message type */ +#endif + __u8 pim_rsv; /* Reserved */ + __be16 pim_cksum; /* Checksum */ +}; + +#define PIM_MINLEN 8 +#endif + /* Message types - V1 */ #define PIM_V1_VERSION __constant_htonl(0x10000000) #define PIM_V1_REGISTER 1 -- cgit v1.2.3 From 7bc570c8b4f75ddb3fd5dbeb38127cdc4acbcc9c Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 3 Apr 2008 09:22:53 +0900 Subject: [IPV6] MROUTE: Support multicast forwarding. Based on ancient patch by Mickael Hoerdt , which is available at . Signed-off-by: YOSHIFUJI Hideaki --- include/linux/Kbuild | 1 + include/linux/ipv6.h | 5 + include/linux/mroute6.h | 227 ++++++++ net/ipv6/Kconfig | 7 + net/ipv6/Makefile | 2 + net/ipv6/addrconf.c | 15 +- net/ipv6/af_inet6.c | 6 + net/ipv6/ip6_input.c | 87 ++- net/ipv6/ip6_output.c | 6 +- net/ipv6/ip6mr.c | 1384 ++++++++++++++++++++++++++++++++++++++++++++++ net/ipv6/ipv6_sockglue.c | 7 + net/ipv6/raw.c | 7 +- net/ipv6/route.c | 30 +- 13 files changed, 1754 insertions(+), 30 deletions(-) create mode 100644 include/linux/mroute6.h create mode 100644 net/ipv6/ip6mr.c (limited to 'include') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 84736acb4b99..29ab9b95d376 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -261,6 +261,7 @@ unifdef-y += mempolicy.h unifdef-y += mii.h unifdef-y += mman.h unifdef-y += mroute.h +unifdef-y += mroute6.h unifdef-y += msdos_fs.h unifdef-y += msg.h unifdef-y += nbd.h diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index b90d3d461d4e..f53e4764fc05 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -159,6 +159,9 @@ struct ipv6_devconf { __s32 accept_source_route; #ifdef CONFIG_IPV6_OPTIMISTIC_DAD __s32 optimistic_dad; +#endif +#ifdef CONFIG_IPV6_MROUTE + __s32 mc_forwarding; #endif void *sysctl; }; @@ -190,6 +193,7 @@ enum { DEVCONF_PROXY_NDP, DEVCONF_OPTIMISTIC_DAD, DEVCONF_ACCEPT_SOURCE_ROUTE, + DEVCONF_MC_FORWARDING, DEVCONF_MAX }; @@ -230,6 +234,7 @@ struct inet6_skb_parm { #endif #define IP6SKB_XFRM_TRANSFORMED 1 +#define IP6SKB_FORWARDED 2 }; #define IP6CB(skb) ((struct inet6_skb_parm*)((skb)->cb)) diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h new file mode 100644 index 000000000000..b92190304e0b --- /dev/null +++ b/include/linux/mroute6.h @@ -0,0 +1,227 @@ +#ifndef __LINUX_MROUTE6_H +#define __LINUX_MROUTE6_H + +#include +#include + +/* + * Based on the MROUTING 3.5 defines primarily to keep + * source compatibility with BSD. + * + * See the pim6sd code for the original history. + * + * Protocol Independent Multicast (PIM) data structures included + * Carlos Picoto (cap@di.fc.ul.pt) + * + */ + +#define MRT6_BASE 200 +#define MRT6_INIT (MRT6_BASE) /* Activate the kernel mroute code */ +#define MRT6_DONE (MRT6_BASE+1) /* Shutdown the kernel mroute */ +#define MRT6_ADD_MIF (MRT6_BASE+2) /* Add a virtual interface */ +#define MRT6_DEL_MIF (MRT6_BASE+3) /* Delete a virtual interface */ +#define MRT6_ADD_MFC (MRT6_BASE+4) /* Add a multicast forwarding entry */ +#define MRT6_DEL_MFC (MRT6_BASE+5) /* Delete a multicast forwarding entry */ +#define MRT6_VERSION (MRT6_BASE+6) /* Get the kernel multicast version */ + +#define SIOCGETMIFCNT_IN6 SIOCPROTOPRIVATE /* IP protocol privates */ +#define SIOCGETSGCNT_IN6 (SIOCPROTOPRIVATE+1) +#define SIOCGETRPF (SIOCPROTOPRIVATE+2) + +#define MAXMIFS 32 +typedef unsigned long mifbitmap_t; /* User mode code depends on this lot */ +typedef unsigned short mifi_t; +#define ALL_MIFS ((mifi_t)(-1)) + +#ifndef IF_SETSIZE +#define IF_SETSIZE 256 +#endif + +typedef __u32 if_mask; +#define NIFBITS (sizeof(if_mask) * 8) /* bits per mask */ + +#if !defined(__KERNEL__) && !defined(DIV_ROUND_UP) +#define DIV_ROUND_UP(x,y) (((x) + ((y) - 1)) / (y)) +#endif + +typedef struct if_set { + if_mask ifs_bits[DIV_ROUND_UP(IF_SETSIZE, NIFBITS)]; +} if_set; + +#define IF_SET(n, p) ((p)->ifs_bits[(n)/NIFBITS] |= (1 << ((n) % NIFBITS))) +#define IF_CLR(n, p) ((p)->ifs_bits[(n)/NIFBITS] &= ~(1 << ((n) % NIFBITS))) +#define IF_ISSET(n, p) ((p)->ifs_bits[(n)/NIFBITS] & (1 << ((n) % NIFBITS))) +#define IF_COPY(f, t) bcopy(f, t, sizeof(*(f))) +#define IF_ZERO(p) bzero(p, sizeof(*(p))) + +/* + * Passed by mrouted for an MRT_ADD_MIF - again we use the + * mrouted 3.6 structures for compatibility + */ + +struct mif6ctl { + mifi_t mif6c_mifi; /* Index of MIF */ + unsigned char mif6c_flags; /* MIFF_ flags */ + unsigned char vifc_threshold; /* ttl limit */ + u_short mif6c_pifi; /* the index of the physical IF */ + unsigned int vifc_rate_limit; /* Rate limiter values (NI) */ +}; + +#define MIFF_REGISTER 0x1 /* register vif */ + +/* + * Cache manipulation structures for mrouted and PIMd + */ + +struct mf6cctl +{ + struct sockaddr_in6 mf6cc_origin; /* Origin of mcast */ + struct sockaddr_in6 mf6cc_mcastgrp; /* Group in question */ + mifi_t mf6cc_parent; /* Where it arrived */ + struct if_set mf6cc_ifset; /* Where it is going */ +}; + +/* + * Group count retrieval for pim6sd + */ + +struct sioc_sg_req6 +{ + struct sockaddr_in6 src; + struct sockaddr_in6 grp; + unsigned long pktcnt; + unsigned long bytecnt; + unsigned long wrong_if; +}; + +/* + * To get vif packet counts + */ + +struct sioc_mif_req6 +{ + mifi_t mifi; /* Which iface */ + unsigned long icount; /* In packets */ + unsigned long ocount; /* Out packets */ + unsigned long ibytes; /* In bytes */ + unsigned long obytes; /* Out bytes */ +}; + +/* + * That's all usermode folks + */ + +#ifdef __KERNEL__ + +#include /* for struct sk_buff_head */ + +struct net_device; +struct inet6_dev *ipv6_find_idev(struct net_device *dev); + +#ifdef CONFIG_IPV6_MROUTE +static inline int ip6_mroute_opt(int opt) +{ + return (opt >= MRT6_BASE) && (opt <= MRT6_BASE + 10); +} +#else +static inline int ip6_mroute_opt(int opt) +{ + return 0; +} +#endif + +struct sock; + +extern int ip6_mroute_setsockopt(struct sock *, int, char __user *, int); +extern int ip6_mroute_getsockopt(struct sock *, int, char __user *, int __user *); +extern int ip6_mr_input(struct sk_buff *skb); +extern int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg); +extern void ip6_mr_init(void); + +struct mif_device +{ + struct net_device *dev; /* Device we are using */ + unsigned long bytes_in,bytes_out; + unsigned long pkt_in,pkt_out; /* Statistics */ + unsigned long rate_limit; /* Traffic shaping (NI) */ + unsigned char threshold; /* TTL threshold */ + unsigned short flags; /* Control flags */ + int link; /* Physical interface index */ +}; + +#define VIFF_STATIC 0x8000 + +struct mfc6_cache +{ + struct mfc6_cache *next; /* Next entry on cache line */ + struct in6_addr mf6c_mcastgrp; /* Group the entry belongs to */ + struct in6_addr mf6c_origin; /* Source of packet */ + mifi_t mf6c_parent; /* Source interface */ + int mfc_flags; /* Flags on line */ + + union { + struct { + unsigned long expires; + struct sk_buff_head unresolved; /* Unresolved buffers */ + } unres; + struct { + unsigned long last_assert; + int minvif; + int maxvif; + unsigned long bytes; + unsigned long pkt; + unsigned long wrong_if; + unsigned char ttls[MAXMIFS]; /* TTL thresholds */ + } res; + } mfc_un; +}; + +#define MFC_STATIC 1 +#define MFC_NOTIFY 2 + +#define MFC6_LINES 64 + +#define MFC6_HASH(a, g) (((__force u32)(a)->s6_addr32[0] ^ \ + (__force u32)(a)->s6_addr32[1] ^ \ + (__force u32)(a)->s6_addr32[2] ^ \ + (__force u32)(a)->s6_addr32[3] ^ \ + (__force u32)(g)->s6_addr32[0] ^ \ + (__force u32)(g)->s6_addr32[1] ^ \ + (__force u32)(g)->s6_addr32[2] ^ \ + (__force u32)(g)->s6_addr32[3]) % MFC6_LINES) + +#define MFC_ASSERT_THRESH (3*HZ) /* Maximal freq. of asserts */ + +#endif + +#ifdef __KERNEL__ +struct rtmsg; +extern int ip6mr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait); + +#ifdef CONFIG_IPV6_MROUTE +extern struct sock *mroute6_socket; +extern int ip6mr_sk_done(struct sock *sk); +#else +#define mroute6_socket NULL +static inline int ip6mr_sk_done(struct sock *sk) { return 0; } +#endif +#endif + +/* + * Structure used to communicate from kernel to multicast router. + * We'll overlay the structure onto an MLD header (not an IPv6 heder like igmpmsg{} + * used for IPv4 implementation). This is because this structure will be passed via an + * IPv6 raw socket, on wich an application will only receiver the payload i.e the data after + * the IPv6 header and all the extension headers. (See section 3 of RFC 3542) + */ + +struct mrt6msg { +#define MRT6MSG_NOCACHE 1 + __u8 im6_mbz; /* must be zero */ + __u8 im6_msgtype; /* what type of message */ + __u16 im6_mif; /* mif rec'd on */ + __u32 im6_pad; /* padding for 64 bit arch */ + struct in6_addr im6_src, im6_dst; +}; + +#endif diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 47263e45bacb..9a2ea81e499f 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -209,3 +209,10 @@ config IPV6_SUBTREES If unsure, say N. +config IPV6_MROUTE + bool "IPv6: multicast routing (EXPERIMENTAL)" + depends on IPV6 && EXPERIMENTAL + ---help--- + Experimental support for IPv6 multicast forwarding. + If unsure, say N. + diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index ae14617e607f..686934acfac1 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -11,6 +11,8 @@ ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o ipv6-$(CONFIG_SYSCTL) = sysctl_net_ipv6.o +ipv6-$(CONFIG_IPV6_MROUTE) += ip6mr.o + ipv6-$(CONFIG_XFRM) += xfrm6_policy.o xfrm6_state.o xfrm6_input.o \ xfrm6_output.o ipv6-$(CONFIG_NETFILTER) += netfilter.o diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 11037615bc73..dbc51af69017 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -412,7 +412,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) return ndev; } -static struct inet6_dev * ipv6_find_idev(struct net_device *dev) +struct inet6_dev * ipv6_find_idev(struct net_device *dev) { struct inet6_dev *idev; @@ -3547,6 +3547,9 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf, #ifdef CONFIG_IPV6_OPTIMISTIC_DAD array[DEVCONF_OPTIMISTIC_DAD] = cnf->optimistic_dad; #endif +#ifdef CONFIG_IPV6_MROUTE + array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding; +#endif } static inline size_t inet6_if_nlmsg_size(void) @@ -4094,6 +4097,16 @@ static struct addrconf_sysctl_table .proc_handler = &proc_dointvec, }, +#endif +#ifdef CONFIG_IPV6_MROUTE + { + .ctl_name = CTL_UNNUMBERED, + .procname = "mc_forwarding", + .data = &ipv6_devconf.mc_forwarding, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, #endif { .ctl_name = 0, /* sentinel */ diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 1731b0abf7f5..3c6aafb02183 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -61,6 +61,9 @@ #include #include +#ifdef CONFIG_IPV6_MROUTE +#include +#endif MODULE_AUTHOR("Cast of dozens"); MODULE_DESCRIPTION("IPv6 protocol stack for Linux"); @@ -953,6 +956,9 @@ static int __init inet6_init(void) err = icmpv6_init(); if (err) goto icmp_fail; +#ifdef CONFIG_IPV6_MROUTE + ip6_mr_init(); +#endif err = ndisc_init(); if (err) goto ndisc_fail; diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 43a617e2268b..09a3201e408a 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -236,36 +237,84 @@ int ip6_mc_input(struct sk_buff *skb) hdr = ipv6_hdr(skb); deliver = ipv6_chk_mcast_addr(skb->dev, &hdr->daddr, NULL); +#ifdef CONFIG_IPV6_MROUTE /* - * IPv6 multicast router mode isnt currently supported. + * IPv6 multicast router mode is now supported ;) */ -#if 0 - if (ipv6_config.multicast_route) { - int addr_type; - - addr_type = ipv6_addr_type(&hdr->daddr); - - if (!(addr_type & (IPV6_ADDR_LOOPBACK | IPV6_ADDR_LINKLOCAL))) { - struct sk_buff *skb2; - struct dst_entry *dst; + if (ipv6_devconf.mc_forwarding && + likely(!(IP6CB(skb)->flags & IP6SKB_FORWARDED))) { + /* + * Okay, we try to forward - split and duplicate + * packets. + */ + struct sk_buff *skb2; + struct inet6_skb_parm *opt = IP6CB(skb); + + /* Check for MLD */ + if (unlikely(opt->ra)) { + /* Check if this is a mld message */ + u8 *ptr = skb_network_header(skb) + opt->ra; + struct icmp6hdr *icmp6; + u8 nexthdr = hdr->nexthdr; + int offset; + + /* Check if the value of Router Alert + * is for MLD (0x0000). + */ + if ((ptr[2] | ptr[3]) == 0) { + if (!ipv6_ext_hdr(nexthdr)) { + /* BUG */ + goto discard; + } + offset = ipv6_skip_exthdr(skb, sizeof(*hdr), + &nexthdr); + if (offset < 0) + goto discard; + + if (nexthdr != IPPROTO_ICMPV6) + goto discard; + + if (!pskb_may_pull(skb, (skb_network_header(skb) + + offset + 1 - skb->data))) + goto discard; + + icmp6 = (struct icmp6hdr *)(skb_network_header(skb) + offset); + + switch (icmp6->icmp6_type) { + case ICMPV6_MGM_QUERY: + case ICMPV6_MGM_REPORT: + case ICMPV6_MGM_REDUCTION: + case ICMPV6_MLD2_REPORT: + break; + default: + /* Bogus */ + goto discard; + } + deliver = 1; + goto out; + } + /* unknown RA - process it normally */ + } - dst = skb->dst; + if (deliver) + skb2 = skb_clone(skb, GFP_ATOMIC); + else { + skb2 = skb; + skb = NULL; + } - if (deliver) { - skb2 = skb_clone(skb, GFP_ATOMIC); - dst_output(skb2); - } else { - dst_output(skb); - return 0; - } + if (skb2) { + skb2->dev = skb2->dst->dev; + ip6_mr_input(skb2); } } #endif - +out: if (likely(deliver)) { ip6_input(skb); return 0; } +discard: /* discard */ kfree_skb(skb); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index a8b4da25b0a7..c0dbe549cc42 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -55,6 +55,7 @@ #include #include #include +#include static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *)); @@ -137,8 +138,9 @@ static int ip6_output2(struct sk_buff *skb) struct inet6_dev *idev = ip6_dst_idev(skb->dst); if (!(dev->flags & IFF_LOOPBACK) && (!np || np->mc_loop) && - ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr, - &ipv6_hdr(skb)->saddr)) { + ((mroute6_socket && !(IP6CB(skb)->flags & IP6SKB_FORWARDED)) || + ipv6_chk_mcast_addr(dev, &ipv6_hdr(skb)->daddr, + &ipv6_hdr(skb)->saddr))) { struct sk_buff *newskb = skb_clone(skb, GFP_ATOMIC); /* Do not check for IFF_ALLMULTI; multicast routing diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c new file mode 100644 index 000000000000..1bdf3c177d58 --- /dev/null +++ b/net/ipv6/ip6mr.c @@ -0,0 +1,1384 @@ +/* + * Linux IPv6 multicast routing support for BSD pim6sd + * Based on net/ipv4/ipmr.c. + * + * (c) 2004 Mickael Hoerdt, + * LSIIT Laboratory, Strasbourg, France + * (c) 2004 Jean-Philippe Andriot, + * 6WIND, Paris, France + * Copyright (C)2007,2008 USAGI/WIDE Project + * YOSHIFUJI Hideaki + * + * This program 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 of the License, or (at your option) any later version. + * + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include + +struct sock *mroute6_socket; + + +/* Big lock, protecting vif table, mrt cache and mroute socket state. + Note that the changes are semaphored via rtnl_lock. + */ + +static DEFINE_RWLOCK(mrt_lock); + +/* + * Multicast router control variables + */ + +static struct mif_device vif6_table[MAXMIFS]; /* Devices */ +static int maxvif; + +#define MIF_EXISTS(idx) (vif6_table[idx].dev != NULL) + +static struct mfc6_cache *mfc6_cache_array[MFC_LINES]; /* Forwarding cache */ + +static struct mfc6_cache *mfc_unres_queue; /* Queue of unresolved entries */ +static atomic_t cache_resolve_queue_len; /* Size of unresolved */ + +/* Special spinlock for queue of unresolved entries */ +static DEFINE_SPINLOCK(mfc_unres_lock); + +/* We return to original Alan's scheme. Hash table of resolved + entries is changed only in process context and protected + with weak lock mrt_lock. Queue of unresolved entries is protected + with strong spinlock mfc_unres_lock. + + In this case data path is free of exclusive locks at all. + */ + +static struct kmem_cache *mrt_cachep __read_mostly; + +static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache); +static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert); +static int ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm); + +static struct timer_list ipmr_expire_timer; + + +#ifdef CONFIG_PROC_FS + +struct ipmr_mfc_iter { + struct mfc6_cache **cache; + int ct; +}; + + +static struct mfc6_cache *ipmr_mfc_seq_idx(struct ipmr_mfc_iter *it, loff_t pos) +{ + struct mfc6_cache *mfc; + + it->cache = mfc6_cache_array; + read_lock(&mrt_lock); + for (it->ct = 0; it->ct < ARRAY_SIZE(mfc6_cache_array); it->ct++) + for (mfc = mfc6_cache_array[it->ct]; mfc; mfc = mfc->next) + if (pos-- == 0) + return mfc; + read_unlock(&mrt_lock); + + it->cache = &mfc_unres_queue; + spin_lock_bh(&mfc_unres_lock); + for (mfc = mfc_unres_queue; mfc; mfc = mfc->next) + if (pos-- == 0) + return mfc; + spin_unlock_bh(&mfc_unres_lock); + + it->cache = NULL; + return NULL; +} + + + + +/* + * The /proc interfaces to multicast routing /proc/ip6_mr_cache /proc/ip6_mr_vif + */ + +struct ipmr_vif_iter { + int ct; +}; + +static struct mif_device *ip6mr_vif_seq_idx(struct ipmr_vif_iter *iter, + loff_t pos) +{ + for (iter->ct = 0; iter->ct < maxvif; ++iter->ct) { + if (!MIF_EXISTS(iter->ct)) + continue; + if (pos-- == 0) + return &vif6_table[iter->ct]; + } + return NULL; +} + +static void *ip6mr_vif_seq_start(struct seq_file *seq, loff_t *pos) + __acquires(mrt_lock) +{ + read_lock(&mrt_lock); + return (*pos ? ip6mr_vif_seq_idx(seq->private, *pos - 1) + : SEQ_START_TOKEN); +} + +static void *ip6mr_vif_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct ipmr_vif_iter *iter = seq->private; + + ++*pos; + if (v == SEQ_START_TOKEN) + return ip6mr_vif_seq_idx(iter, 0); + + while (++iter->ct < maxvif) { + if (!MIF_EXISTS(iter->ct)) + continue; + return &vif6_table[iter->ct]; + } + return NULL; +} + +static void ip6mr_vif_seq_stop(struct seq_file *seq, void *v) + __releases(mrt_lock) +{ + read_unlock(&mrt_lock); +} + +static int ip6mr_vif_seq_show(struct seq_file *seq, void *v) +{ + if (v == SEQ_START_TOKEN) { + seq_puts(seq, + "Interface BytesIn PktsIn BytesOut PktsOut Flags\n"); + } else { + const struct mif_device *vif = v; + const char *name = vif->dev ? vif->dev->name : "none"; + + seq_printf(seq, + "%2Zd %-10s %8ld %7ld %8ld %7ld %05X\n", + vif - vif6_table, + name, vif->bytes_in, vif->pkt_in, + vif->bytes_out, vif->pkt_out, + vif->flags); + } + return 0; +} + +static struct seq_operations ip6mr_vif_seq_ops = { + .start = ip6mr_vif_seq_start, + .next = ip6mr_vif_seq_next, + .stop = ip6mr_vif_seq_stop, + .show = ip6mr_vif_seq_show, +}; + +static int ip6mr_vif_open(struct inode *inode, struct file *file) +{ + return seq_open_private(file, &ip6mr_vif_seq_ops, + sizeof(struct ipmr_vif_iter)); +} + +static struct file_operations ip6mr_vif_fops = { + .owner = THIS_MODULE, + .open = ip6mr_vif_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static void *ipmr_mfc_seq_start(struct seq_file *seq, loff_t *pos) +{ + return (*pos ? ipmr_mfc_seq_idx(seq->private, *pos - 1) + : SEQ_START_TOKEN); +} + +static void *ipmr_mfc_seq_next(struct seq_file *seq, void *v, loff_t *pos) +{ + struct mfc6_cache *mfc = v; + struct ipmr_mfc_iter *it = seq->private; + + ++*pos; + + if (v == SEQ_START_TOKEN) + return ipmr_mfc_seq_idx(seq->private, 0); + + if (mfc->next) + return mfc->next; + + if (it->cache == &mfc_unres_queue) + goto end_of_list; + + BUG_ON(it->cache != mfc6_cache_array); + + while (++it->ct < ARRAY_SIZE(mfc6_cache_array)) { + mfc = mfc6_cache_array[it->ct]; + if (mfc) + return mfc; + } + + /* exhausted cache_array, show unresolved */ + read_unlock(&mrt_lock); + it->cache = &mfc_unres_queue; + it->ct = 0; + + spin_lock_bh(&mfc_unres_lock); + mfc = mfc_unres_queue; + if (mfc) + return mfc; + + end_of_list: + spin_unlock_bh(&mfc_unres_lock); + it->cache = NULL; + + return NULL; +} + +static void ipmr_mfc_seq_stop(struct seq_file *seq, void *v) +{ + struct ipmr_mfc_iter *it = seq->private; + + if (it->cache == &mfc_unres_queue) + spin_unlock_bh(&mfc_unres_lock); + else if (it->cache == mfc6_cache_array) + read_unlock(&mrt_lock); +} + +static int ipmr_mfc_seq_show(struct seq_file *seq, void *v) +{ + int n; + + if (v == SEQ_START_TOKEN) { + seq_puts(seq, + "Group " + "Origin " + "Iif Pkts Bytes Wrong Oifs\n"); + } else { + const struct mfc6_cache *mfc = v; + const struct ipmr_mfc_iter *it = seq->private; + + seq_printf(seq, + NIP6_FMT " " NIP6_FMT " %-3d %8ld %8ld %8ld", + NIP6(mfc->mf6c_mcastgrp), NIP6(mfc->mf6c_origin), + mfc->mf6c_parent, + mfc->mfc_un.res.pkt, + mfc->mfc_un.res.bytes, + mfc->mfc_un.res.wrong_if); + + if (it->cache != &mfc_unres_queue) { + for (n = mfc->mfc_un.res.minvif; + n < mfc->mfc_un.res.maxvif; n++) { + if (MIF_EXISTS(n) && + mfc->mfc_un.res.ttls[n] < 255) + seq_printf(seq, + " %2d:%-3d", + n, mfc->mfc_un.res.ttls[n]); + } + } + seq_putc(seq, '\n'); + } + return 0; +} + +static struct seq_operations ipmr_mfc_seq_ops = { + .start = ipmr_mfc_seq_start, + .next = ipmr_mfc_seq_next, + .stop = ipmr_mfc_seq_stop, + .show = ipmr_mfc_seq_show, +}; + +static int ipmr_mfc_open(struct inode *inode, struct file *file) +{ + return seq_open_private(file, &ipmr_mfc_seq_ops, + sizeof(struct ipmr_mfc_iter)); +} + +static struct file_operations ip6mr_mfc_fops = { + .owner = THIS_MODULE, + .open = ipmr_mfc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; +#endif + +/* + * Delete a VIF entry + */ + +static int mif6_delete(int vifi) +{ + struct mif_device *v; + struct net_device *dev; + if (vifi < 0 || vifi >= maxvif) + return -EADDRNOTAVAIL; + + v = &vif6_table[vifi]; + + write_lock_bh(&mrt_lock); + dev = v->dev; + v->dev = NULL; + + if (!dev) { + write_unlock_bh(&mrt_lock); + return -EADDRNOTAVAIL; + } + + if (vifi + 1 == maxvif) { + int tmp; + for (tmp = vifi - 1; tmp >= 0; tmp--) { + if (MIF_EXISTS(tmp)) + break; + } + maxvif = tmp + 1; + } + + write_unlock_bh(&mrt_lock); + + dev_set_allmulti(dev, -1); + + if (v->flags & MIFF_REGISTER) + unregister_netdevice(dev); + + dev_put(dev); + return 0; +} + +/* Destroy an unresolved cache entry, killing queued skbs + and reporting error to netlink readers. + */ + +static void ip6mr_destroy_unres(struct mfc6_cache *c) +{ + struct sk_buff *skb; + + atomic_dec(&cache_resolve_queue_len); + + while((skb = skb_dequeue(&c->mfc_un.unres.unresolved)) != NULL) { + if (ipv6_hdr(skb)->version == 0) { + struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct ipv6hdr)); + nlh->nlmsg_type = NLMSG_ERROR; + nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr)); + skb_trim(skb, nlh->nlmsg_len); + ((struct nlmsgerr *)NLMSG_DATA(nlh))->error = -ETIMEDOUT; + rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid); + } else + kfree_skb(skb); + } + + kmem_cache_free(mrt_cachep, c); +} + + +/* Single timer process for all the unresolved queue. */ + +static void ipmr_do_expire_process(unsigned long dummy) +{ + unsigned long now = jiffies; + unsigned long expires = 10 * HZ; + struct mfc6_cache *c, **cp; + + cp = &mfc_unres_queue; + + while ((c = *cp) != NULL) { + if (time_after(c->mfc_un.unres.expires, now)) { + /* not yet... */ + unsigned long interval = c->mfc_un.unres.expires - now; + if (interval < expires) + expires = interval; + cp = &c->next; + continue; + } + + *cp = c->next; + ip6mr_destroy_unres(c); + } + + if (atomic_read(&cache_resolve_queue_len)) + mod_timer(&ipmr_expire_timer, jiffies + expires); +} + +static void ipmr_expire_process(unsigned long dummy) +{ + if (!spin_trylock(&mfc_unres_lock)) { + mod_timer(&ipmr_expire_timer, jiffies + 1); + return; + } + + if (atomic_read(&cache_resolve_queue_len)) + ipmr_do_expire_process(dummy); + + spin_unlock(&mfc_unres_lock); +} + +/* Fill oifs list. It is called under write locked mrt_lock. */ + +static void ip6mr_update_thresholds(struct mfc6_cache *cache, unsigned char *ttls) +{ + int vifi; + + cache->mfc_un.res.minvif = MAXVIFS; + cache->mfc_un.res.maxvif = 0; + memset(cache->mfc_un.res.ttls, 255, MAXVIFS); + + for (vifi = 0; vifi < maxvif; vifi++) { + if (MIF_EXISTS(vifi) && ttls[vifi] && ttls[vifi] < 255) { + cache->mfc_un.res.ttls[vifi] = ttls[vifi]; + if (cache->mfc_un.res.minvif > vifi) + cache->mfc_un.res.minvif = vifi; + if (cache->mfc_un.res.maxvif <= vifi) + cache->mfc_un.res.maxvif = vifi + 1; + } + } +} + +static int mif6_add(struct mif6ctl *vifc, int mrtsock) +{ + int vifi = vifc->mif6c_mifi; + struct mif_device *v = &vif6_table[vifi]; + struct net_device *dev; + + /* Is vif busy ? */ + if (MIF_EXISTS(vifi)) + return -EADDRINUSE; + + switch (vifc->mif6c_flags) { + case 0: + dev = dev_get_by_index(&init_net, vifc->mif6c_pifi); + if (!dev) + return -EADDRNOTAVAIL; + dev_put(dev); + break; + default: + return -EINVAL; + } + + dev_set_allmulti(dev, 1); + + /* + * Fill in the VIF structures + */ + v->rate_limit = vifc->vifc_rate_limit; + v->flags = vifc->mif6c_flags; + if (!mrtsock) + v->flags |= VIFF_STATIC; + v->threshold = vifc->vifc_threshold; + v->bytes_in = 0; + v->bytes_out = 0; + v->pkt_in = 0; + v->pkt_out = 0; + v->link = dev->ifindex; + if (v->flags & MIFF_REGISTER) + v->link = dev->iflink; + + /* And finish update writing critical data */ + write_lock_bh(&mrt_lock); + dev_hold(dev); + v->dev = dev; + if (vifi + 1 > maxvif) + maxvif = vifi + 1; + write_unlock_bh(&mrt_lock); + return 0; +} + +static struct mfc6_cache *ip6mr_cache_find(struct in6_addr *origin, struct in6_addr *mcastgrp) +{ + int line = MFC6_HASH(mcastgrp, origin); + struct mfc6_cache *c; + + for (c = mfc6_cache_array[line]; c; c = c->next) { + if (ipv6_addr_equal(&c->mf6c_origin, origin) && + ipv6_addr_equal(&c->mf6c_mcastgrp, mcastgrp)) + break; + } + return c; +} + +/* + * Allocate a multicast cache entry + */ +static struct mfc6_cache *ip6mr_cache_alloc(void) +{ + struct mfc6_cache *c = kmem_cache_alloc(mrt_cachep, GFP_KERNEL); + if (c == NULL) + return NULL; + memset(c, 0, sizeof(*c)); + c->mfc_un.res.minvif = MAXVIFS; + return c; +} + +static struct mfc6_cache *ip6mr_cache_alloc_unres(void) +{ + struct mfc6_cache *c = kmem_cache_alloc(mrt_cachep, GFP_ATOMIC); + if (c == NULL) + return NULL; + memset(c, 0, sizeof(*c)); + skb_queue_head_init(&c->mfc_un.unres.unresolved); + c->mfc_un.unres.expires = jiffies + 10 * HZ; + return c; +} + +/* + * A cache entry has gone into a resolved state from queued + */ + +static void ip6mr_cache_resolve(struct mfc6_cache *uc, struct mfc6_cache *c) +{ + struct sk_buff *skb; + + /* + * Play the pending entries through our router + */ + + while((skb = __skb_dequeue(&uc->mfc_un.unres.unresolved))) { + if (ipv6_hdr(skb)->version == 0) { + int err; + struct nlmsghdr *nlh = (struct nlmsghdr *)skb_pull(skb, sizeof(struct ipv6hdr)); + + if (ip6mr_fill_mroute(skb, c, NLMSG_DATA(nlh)) > 0) { + nlh->nlmsg_len = skb->tail - (u8 *)nlh; + } else { + nlh->nlmsg_type = NLMSG_ERROR; + nlh->nlmsg_len = NLMSG_LENGTH(sizeof(struct nlmsgerr)); + skb_trim(skb, nlh->nlmsg_len); + ((struct nlmsgerr *)NLMSG_DATA(nlh))->error = -EMSGSIZE; + } + err = rtnl_unicast(skb, &init_net, NETLINK_CB(skb).pid); + } else + ip6_mr_forward(skb, c); + } +} + +/* + * Bounce a cache query up to pim6sd. We could use netlink for this but pim6sd + * expects the following bizarre scheme. + * + * Called under mrt_lock. + */ + +static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) +{ + struct sk_buff *skb; + struct mrt6msg *msg; + int ret; + + skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC); + + if (!skb) + return -ENOBUFS; + + /* I suppose that internal messages + * do not require checksums */ + + skb->ip_summed = CHECKSUM_UNNECESSARY; + + /* + * Copy the IP header + */ + + skb_put(skb, sizeof(struct ipv6hdr)); + skb_reset_network_header(skb); + skb_copy_to_linear_data(skb, ipv6_hdr(pkt), sizeof(struct ipv6hdr)); + + /* + * Add our header + */ + skb_put(skb, sizeof(*msg)); + skb_reset_transport_header(skb); + msg = (struct mrt6msg *)skb_transport_header(skb); + + msg->im6_mbz = 0; + msg->im6_msgtype = assert; + msg->im6_mif = vifi; + msg->im6_pad = 0; + ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr); + ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr); + + skb->dst = dst_clone(pkt->dst); + skb->ip_summed = CHECKSUM_UNNECESSARY; + + skb_pull(skb, sizeof(struct ipv6hdr)); + + if (mroute6_socket == NULL) { + kfree_skb(skb); + return -EINVAL; + } + + /* + * Deliver to user space multicast routing algorithms + */ + if ((ret = sock_queue_rcv_skb(mroute6_socket, skb)) < 0) { + if (net_ratelimit()) + printk(KERN_WARNING "mroute6: pending queue full, dropping entries.\n"); + kfree_skb(skb); + } + + return ret; +} + +/* + * Queue a packet for resolution. It gets locked cache entry! + */ + +static int +ip6mr_cache_unresolved(vifi_t vifi, struct sk_buff *skb) +{ + int err; + struct mfc6_cache *c; + + spin_lock_bh(&mfc_unres_lock); + for (c = mfc_unres_queue; c; c = c->next) { + if (ipv6_addr_equal(&c->mf6c_mcastgrp, &ipv6_hdr(skb)->daddr) && + ipv6_addr_equal(&c->mf6c_origin, &ipv6_hdr(skb)->saddr)) + break; + } + + if (c == NULL) { + /* + * Create a new entry if allowable + */ + + if (atomic_read(&cache_resolve_queue_len) >= 10 || + (c = ip6mr_cache_alloc_unres()) == NULL) { + spin_unlock_bh(&mfc_unres_lock); + + kfree_skb(skb); + return -ENOBUFS; + } + + /* + * Fill in the new cache entry + */ + c->mf6c_parent = -1; + c->mf6c_origin = ipv6_hdr(skb)->saddr; + c->mf6c_mcastgrp = ipv6_hdr(skb)->daddr; + + /* + * Reflect first query at pim6sd + */ + if ((err = ip6mr_cache_report(skb, vifi, MRT6MSG_NOCACHE)) < 0) { + /* If the report failed throw the cache entry + out - Brad Parker + */ + spin_unlock_bh(&mfc_unres_lock); + + kmem_cache_free(mrt_cachep, c); + kfree_skb(skb); + return err; + } + + atomic_inc(&cache_resolve_queue_len); + c->next = mfc_unres_queue; + mfc_unres_queue = c; + + ipmr_do_expire_process(1); + } + + /* + * See if we can append the packet + */ + if (c->mfc_un.unres.unresolved.qlen > 3) { + kfree_skb(skb); + err = -ENOBUFS; + } else { + skb_queue_tail(&c->mfc_un.unres.unresolved, skb); + err = 0; + } + + spin_unlock_bh(&mfc_unres_lock); + return err; +} + +/* + * MFC6 cache manipulation by user space + */ + +static int ip6mr_mfc_delete(struct mf6cctl *mfc) +{ + int line; + struct mfc6_cache *c, **cp; + + line = MFC6_HASH(&mfc->mf6cc_mcastgrp.sin6_addr, &mfc->mf6cc_origin.sin6_addr); + + for (cp = &mfc6_cache_array[line]; (c = *cp) != NULL; cp = &c->next) { + if (ipv6_addr_equal(&c->mf6c_origin, &mfc->mf6cc_origin.sin6_addr) && + ipv6_addr_equal(&c->mf6c_mcastgrp, &mfc->mf6cc_mcastgrp.sin6_addr)) { + write_lock_bh(&mrt_lock); + *cp = c->next; + write_unlock_bh(&mrt_lock); + + kmem_cache_free(mrt_cachep, c); + return 0; + } + } + return -ENOENT; +} + +static int ip6mr_device_event(struct notifier_block *this, + unsigned long event, void *ptr) +{ + struct net_device *dev = ptr; + struct mif_device *v; + int ct; + + if (dev_net(dev) != &init_net) + return NOTIFY_DONE; + + if (event != NETDEV_UNREGISTER) + return NOTIFY_DONE; + + v = &vif6_table[0]; + for (ct = 0; ct < maxvif; ct++, v++) { + if (v->dev == dev) + mif6_delete(ct); + } + return NOTIFY_DONE; +} + +static struct notifier_block ip6_mr_notifier = { + .notifier_call = ip6mr_device_event +}; + +/* + * Setup for IP multicast routing + */ + +void __init ip6_mr_init(void) +{ + mrt_cachep = kmem_cache_create("ip6_mrt_cache", + sizeof(struct mfc6_cache), + 0, SLAB_HWCACHE_ALIGN, + NULL); + if (!mrt_cachep) + panic("cannot allocate ip6_mrt_cache"); + + setup_timer(&ipmr_expire_timer, ipmr_expire_process, 0); + register_netdevice_notifier(&ip6_mr_notifier); +#ifdef CONFIG_PROC_FS + proc_net_fops_create(&init_net, "ip6_mr_vif", 0, &ip6mr_vif_fops); + proc_net_fops_create(&init_net, "ip6_mr_cache", 0, &ip6mr_mfc_fops); +#endif +} + + +static int ip6mr_mfc_add(struct mf6cctl *mfc, int mrtsock) +{ + int line; + struct mfc6_cache *uc, *c, **cp; + unsigned char ttls[MAXVIFS]; + int i; + + memset(ttls, 255, MAXVIFS); + for (i = 0; i < MAXVIFS; i++) { + if (IF_ISSET(i, &mfc->mf6cc_ifset)) + ttls[i] = 1; + + } + + line = MFC6_HASH(&mfc->mf6cc_mcastgrp.sin6_addr, &mfc->mf6cc_origin.sin6_addr); + + for (cp = &mfc6_cache_array[line]; (c = *cp) != NULL; cp = &c->next) { + if (ipv6_addr_equal(&c->mf6c_origin, &mfc->mf6cc_origin.sin6_addr) && + ipv6_addr_equal(&c->mf6c_mcastgrp, &mfc->mf6cc_mcastgrp.sin6_addr)) + break; + } + + if (c != NULL) { + write_lock_bh(&mrt_lock); + c->mf6c_parent = mfc->mf6cc_parent; + ip6mr_update_thresholds(c, ttls); + if (!mrtsock) + c->mfc_flags |= MFC_STATIC; + write_unlock_bh(&mrt_lock); + return 0; + } + + if (!ipv6_addr_is_multicast(&mfc->mf6cc_mcastgrp.sin6_addr)) + return -EINVAL; + + c = ip6mr_cache_alloc(); + if (c == NULL) + return -ENOMEM; + + c->mf6c_origin = mfc->mf6cc_origin.sin6_addr; + c->mf6c_mcastgrp = mfc->mf6cc_mcastgrp.sin6_addr; + c->mf6c_parent = mfc->mf6cc_parent; + ip6mr_update_thresholds(c, ttls); + if (!mrtsock) + c->mfc_flags |= MFC_STATIC; + + write_lock_bh(&mrt_lock); + c->next = mfc6_cache_array[line]; + mfc6_cache_array[line] = c; + write_unlock_bh(&mrt_lock); + + /* + * Check to see if we resolved a queued list. If so we + * need to send on the frames and tidy up. + */ + spin_lock_bh(&mfc_unres_lock); + for (cp = &mfc_unres_queue; (uc = *cp) != NULL; + cp = &uc->next) { + if (ipv6_addr_equal(&uc->mf6c_origin, &c->mf6c_origin) && + ipv6_addr_equal(&uc->mf6c_mcastgrp, &c->mf6c_mcastgrp)) { + *cp = uc->next; + if (atomic_dec_and_test(&cache_resolve_queue_len)) + del_timer(&ipmr_expire_timer); + break; + } + } + spin_unlock_bh(&mfc_unres_lock); + + if (uc) { + ip6mr_cache_resolve(uc, c); + kmem_cache_free(mrt_cachep, uc); + } + return 0; +} + +/* + * Close the multicast socket, and clear the vif tables etc + */ + +static void mroute_clean_tables(struct sock *sk) +{ + int i; + + /* + * Shut down all active vif entries + */ + for (i = 0; i < maxvif; i++) { + if (!(vif6_table[i].flags & VIFF_STATIC)) + mif6_delete(i); + } + + /* + * Wipe the cache + */ + for (i = 0; i < ARRAY_SIZE(mfc6_cache_array); i++) { + struct mfc6_cache *c, **cp; + + cp = &mfc6_cache_array[i]; + while ((c = *cp) != NULL) { + if (c->mfc_flags & MFC_STATIC) { + cp = &c->next; + continue; + } + write_lock_bh(&mrt_lock); + *cp = c->next; + write_unlock_bh(&mrt_lock); + + kmem_cache_free(mrt_cachep, c); + } + } + + if (atomic_read(&cache_resolve_queue_len) != 0) { + struct mfc6_cache *c; + + spin_lock_bh(&mfc_unres_lock); + while (mfc_unres_queue != NULL) { + c = mfc_unres_queue; + mfc_unres_queue = c->next; + spin_unlock_bh(&mfc_unres_lock); + + ip6mr_destroy_unres(c); + + spin_lock_bh(&mfc_unres_lock); + } + spin_unlock_bh(&mfc_unres_lock); + } +} + +static int ip6mr_sk_init(struct sock *sk) +{ + int err = 0; + + rtnl_lock(); + write_lock_bh(&mrt_lock); + if (likely(mroute6_socket == NULL)) + mroute6_socket = sk; + else + err = -EADDRINUSE; + write_unlock_bh(&mrt_lock); + + rtnl_unlock(); + + return err; +} + +int ip6mr_sk_done(struct sock *sk) +{ + int err = 0; + + rtnl_lock(); + if (sk == mroute6_socket) { + write_lock_bh(&mrt_lock); + mroute6_socket = NULL; + write_unlock_bh(&mrt_lock); + + mroute_clean_tables(sk); + } else + err = -EACCES; + rtnl_unlock(); + + return err; +} + +/* + * Socket options and virtual interface manipulation. The whole + * virtual interface system is a complete heap, but unfortunately + * that's how BSD mrouted happens to think. Maybe one day with a proper + * MOSPF/PIM router set up we can clean this up. + */ + +int ip6_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int optlen) +{ + int ret; + struct mif6ctl vif; + struct mf6cctl mfc; + mifi_t mifi; + + if (optname != MRT6_INIT) { + if (sk != mroute6_socket && !capable(CAP_NET_ADMIN)) + return -EACCES; + } + + switch (optname) { + case MRT6_INIT: + if (sk->sk_type != SOCK_RAW || + inet_sk(sk)->num != IPPROTO_ICMPV6) + return -EOPNOTSUPP; + if (optlen < sizeof(int)) + return -EINVAL; + + return ip6mr_sk_init(sk); + + case MRT6_DONE: + return ip6mr_sk_done(sk); + + case MRT6_ADD_MIF: + if (optlen < sizeof(vif)) + return -EINVAL; + if (copy_from_user(&vif, optval, sizeof(vif))) + return -EFAULT; + if (vif.mif6c_mifi >= MAXVIFS) + return -ENFILE; + rtnl_lock(); + ret = mif6_add(&vif, sk == mroute6_socket); + rtnl_unlock(); + return ret; + + case MRT6_DEL_MIF: + if (optlen < sizeof(mifi_t)) + return -EINVAL; + if (copy_from_user(&mifi, optval, sizeof(mifi_t))) + return -EFAULT; + rtnl_lock(); + ret = mif6_delete(mifi); + rtnl_unlock(); + return ret; + + /* + * Manipulate the forwarding caches. These live + * in a sort of kernel/user symbiosis. + */ + case MRT6_ADD_MFC: + case MRT6_DEL_MFC: + if (optlen < sizeof(mfc)) + return -EINVAL; + if (copy_from_user(&mfc, optval, sizeof(mfc))) + return -EFAULT; + rtnl_lock(); + if (optname == MRT6_DEL_MFC) + ret = ip6mr_mfc_delete(&mfc); + else + ret = ip6mr_mfc_add(&mfc, sk == mroute6_socket); + rtnl_unlock(); + return ret; + + /* + * Spurious command, or MRT_VERSION which you cannot + * set. + */ + default: + return -ENOPROTOOPT; + } +} + +/* + * Getsock opt support for the multicast routing system. + */ + +int ip6_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, + int __user *optlen) +{ + int olr; + int val; + + switch (optname) { + case MRT6_VERSION: + val = 0x0305; + break; + default: + return -ENOPROTOOPT; + } + + if (get_user(olr, optlen)) + return -EFAULT; + + olr = min_t(int, olr, sizeof(int)); + if (olr < 0) + return -EINVAL; + + if (put_user(olr, optlen)) + return -EFAULT; + if (copy_to_user(optval, &val, olr)) + return -EFAULT; + return 0; +} + +/* + * The IP multicast ioctl support routines. + */ + +int ip6mr_ioctl(struct sock *sk, int cmd, void __user *arg) +{ + struct sioc_sg_req6 sr; + struct sioc_mif_req6 vr; + struct mif_device *vif; + struct mfc6_cache *c; + + switch (cmd) { + case SIOCGETMIFCNT_IN6: + if (copy_from_user(&vr, arg, sizeof(vr))) + return -EFAULT; + if (vr.mifi >= maxvif) + return -EINVAL; + read_lock(&mrt_lock); + vif = &vif6_table[vr.mifi]; + if (MIF_EXISTS(vr.mifi)) { + vr.icount = vif->pkt_in; + vr.ocount = vif->pkt_out; + vr.ibytes = vif->bytes_in; + vr.obytes = vif->bytes_out; + read_unlock(&mrt_lock); + + if (copy_to_user(arg, &vr, sizeof(vr))) + return -EFAULT; + return 0; + } + read_unlock(&mrt_lock); + return -EADDRNOTAVAIL; + case SIOCGETSGCNT_IN6: + if (copy_from_user(&sr, arg, sizeof(sr))) + return -EFAULT; + + read_lock(&mrt_lock); + c = ip6mr_cache_find(&sr.src.sin6_addr, &sr.grp.sin6_addr); + if (c) { + sr.pktcnt = c->mfc_un.res.pkt; + sr.bytecnt = c->mfc_un.res.bytes; + sr.wrong_if = c->mfc_un.res.wrong_if; + read_unlock(&mrt_lock); + + if (copy_to_user(arg, &sr, sizeof(sr))) + return -EFAULT; + return 0; + } + read_unlock(&mrt_lock); + return -EADDRNOTAVAIL; + default: + return -ENOIOCTLCMD; + } +} + + +static inline int ip6mr_forward2_finish(struct sk_buff *skb) +{ + /* XXX stats */ + return dst_output(skb); +} + +/* + * Processing handlers for ip6mr_forward + */ + +static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi) +{ + struct ipv6hdr *ipv6h; + struct mif_device *vif = &vif6_table[vifi]; + struct net_device *dev; + struct dst_entry *dst; + struct flowi fl; + + if (vif->dev == NULL) + goto out_free; + + ipv6h = ipv6_hdr(skb); + + fl = (struct flowi) { + .oif = vif->link, + .nl_u = { .ip6_u = + { .daddr = ipv6h->daddr, } + } + }; + + dst = ip6_route_output(&init_net, NULL, &fl); + if (!dst) + goto out_free; + + dst_release(skb->dst); + skb->dst = dst; + + /* + * RFC1584 teaches, that DVMRP/PIM router must deliver packets locally + * not only before forwarding, but after forwarding on all output + * interfaces. It is clear, if mrouter runs a multicasting + * program, it should receive packets not depending to what interface + * program is joined. + * If we will not make it, the program will have to join on all + * interfaces. On the other hand, multihoming host (or router, but + * not mrouter) cannot join to more than one interface - it will + * result in receiving multiple packets. + */ + dev = vif->dev; + skb->dev = dev; + vif->pkt_out++; + vif->bytes_out += skb->len; + + /* We are about to write */ + /* XXX: extension headers? */ + if (skb_cow(skb, sizeof(*ipv6h) + LL_RESERVED_SPACE(dev))) + goto out_free; + + ipv6h = ipv6_hdr(skb); + ipv6h->hop_limit--; + + IP6CB(skb)->flags |= IP6SKB_FORWARDED; + + return NF_HOOK(PF_INET6, NF_INET_FORWARD, skb, skb->dev, dev, + ip6mr_forward2_finish); + +out_free: + kfree_skb(skb); + return 0; +} + +static int ip6mr_find_vif(struct net_device *dev) +{ + int ct; + for (ct = maxvif - 1; ct >= 0; ct--) { + if (vif6_table[ct].dev == dev) + break; + } + return ct; +} + +static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache) +{ + int psend = -1; + int vif, ct; + + vif = cache->mf6c_parent; + cache->mfc_un.res.pkt++; + cache->mfc_un.res.bytes += skb->len; + + vif6_table[vif].pkt_in++; + vif6_table[vif].bytes_in += skb->len; + + /* + * Forward the frame + */ + for (ct = cache->mfc_un.res.maxvif - 1; ct >= cache->mfc_un.res.minvif; ct--) { + if (ipv6_hdr(skb)->hop_limit > cache->mfc_un.res.ttls[ct]) { + if (psend != -1) { + struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); + if (skb2) + ip6mr_forward2(skb2, cache, psend); + } + psend = ct; + } + } + if (psend != -1) { + ip6mr_forward2(skb, cache, psend); + return 0; + } + + kfree_skb(skb); + return 0; +} + + +/* + * Multicast packets for forwarding arrive here + */ + +int ip6_mr_input(struct sk_buff *skb) +{ + struct mfc6_cache *cache; + + read_lock(&mrt_lock); + cache = ip6mr_cache_find(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr); + + /* + * No usable cache entry + */ + if (cache == NULL) { + int vif; + + vif = ip6mr_find_vif(skb->dev); + if (vif >= 0) { + int err = ip6mr_cache_unresolved(vif, skb); + read_unlock(&mrt_lock); + + return err; + } + read_unlock(&mrt_lock); + kfree_skb(skb); + return -ENODEV; + } + + ip6_mr_forward(skb, cache); + + read_unlock(&mrt_lock); + + return 0; +} + + +static int +ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm) +{ + int ct; + struct rtnexthop *nhp; + struct net_device *dev = vif6_table[c->mf6c_parent].dev; + u8 *b = skb->tail; + struct rtattr *mp_head; + + if (dev) + RTA_PUT(skb, RTA_IIF, 4, &dev->ifindex); + + mp_head = (struct rtattr *)skb_put(skb, RTA_LENGTH(0)); + + for (ct = c->mfc_un.res.minvif; ct < c->mfc_un.res.maxvif; ct++) { + if (c->mfc_un.res.ttls[ct] < 255) { + if (skb_tailroom(skb) < RTA_ALIGN(RTA_ALIGN(sizeof(*nhp)) + 4)) + goto rtattr_failure; + nhp = (struct rtnexthop *)skb_put(skb, RTA_ALIGN(sizeof(*nhp))); + nhp->rtnh_flags = 0; + nhp->rtnh_hops = c->mfc_un.res.ttls[ct]; + nhp->rtnh_ifindex = vif6_table[ct].dev->ifindex; + nhp->rtnh_len = sizeof(*nhp); + } + } + mp_head->rta_type = RTA_MULTIPATH; + mp_head->rta_len = skb->tail - (u8 *)mp_head; + rtm->rtm_type = RTN_MULTICAST; + return 1; + +rtattr_failure: + nlmsg_trim(skb, b); + return -EMSGSIZE; +} + +int ip6mr_get_route(struct sk_buff *skb, struct rtmsg *rtm, int nowait) +{ + int err; + struct mfc6_cache *cache; + struct rt6_info *rt = (struct rt6_info *)skb->dst; + + read_lock(&mrt_lock); + cache = ip6mr_cache_find(&rt->rt6i_src.addr, &rt->rt6i_dst.addr); + + if (!cache) { + struct sk_buff *skb2; + struct ipv6hdr *iph; + struct net_device *dev; + int vif; + + if (nowait) { + read_unlock(&mrt_lock); + return -EAGAIN; + } + + dev = skb->dev; + if (dev == NULL || (vif = ip6mr_find_vif(dev)) < 0) { + read_unlock(&mrt_lock); + return -ENODEV; + } + + /* really correct? */ + skb2 = alloc_skb(sizeof(struct ipv6hdr), GFP_ATOMIC); + if (!skb2) { + read_unlock(&mrt_lock); + return -ENOMEM; + } + + skb_reset_transport_header(skb2); + + skb_put(skb2, sizeof(struct ipv6hdr)); + skb_reset_network_header(skb2); + + iph = ipv6_hdr(skb2); + iph->version = 0; + iph->priority = 0; + iph->flow_lbl[0] = 0; + iph->flow_lbl[1] = 0; + iph->flow_lbl[2] = 0; + iph->payload_len = 0; + iph->nexthdr = IPPROTO_NONE; + iph->hop_limit = 0; + ipv6_addr_copy(&iph->saddr, &rt->rt6i_src.addr); + ipv6_addr_copy(&iph->daddr, &rt->rt6i_dst.addr); + + err = ip6mr_cache_unresolved(vif, skb2); + read_unlock(&mrt_lock); + + return err; + } + + if (!nowait && (rtm->rtm_flags&RTM_F_NOTIFY)) + cache->mfc_flags |= MFC_NOTIFY; + + err = ip6mr_fill_mroute(skb, cache, rtm); + read_unlock(&mrt_lock); + return err; +} + diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c index 4195ac92345e..99624109c010 100644 --- a/net/ipv6/ipv6_sockglue.c +++ b/net/ipv6/ipv6_sockglue.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include @@ -118,6 +119,9 @@ static int do_ipv6_setsockopt(struct sock *sk, int level, int optname, valbool = (val!=0); + if (ip6_mroute_opt(optname)) + return ip6_mroute_setsockopt(sk, optname, optval, optlen); + lock_sock(sk); switch (optname) { @@ -790,6 +794,9 @@ static int do_ipv6_getsockopt(struct sock *sk, int level, int optname, int len; int val; + if (ip6_mroute_opt(optname)) + return ip6_mroute_getsockopt(sk, optname, optval, optlen); + if (get_user(len, optlen)) return -EFAULT; switch (optname) { diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index aae6cedf1709..088b80b4ce74 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -53,6 +53,7 @@ #if defined(CONFIG_IPV6_MIP6) || defined(CONFIG_IPV6_MIP6_MODULE) #include #endif +#include #include #include @@ -1135,7 +1136,11 @@ static int rawv6_ioctl(struct sock *sk, int cmd, unsigned long arg) } default: +#ifdef CONFIG_IPV6_MROUTE + return ip6mr_ioctl(sk, cmd, (void __user *)arg); +#else return -ENOIOCTLCMD; +#endif } } @@ -1143,7 +1148,7 @@ static void rawv6_close(struct sock *sk, long timeout) { if (inet_sk(sk)->num == IPPROTO_RAW) ip6_ra_control(sk, -1, NULL); - + ip6mr_sk_done(sk); sk_common_release(sk); } diff --git a/net/ipv6/route.c b/net/ipv6/route.c index cd82b6db35ff..3c314d5f46c6 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -2106,7 +2107,7 @@ static inline size_t rt6_nlmsg_size(void) static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, struct in6_addr *dst, struct in6_addr *src, int iif, int type, u32 pid, u32 seq, - int prefix, unsigned int flags) + int prefix, int nowait, unsigned int flags) { struct rtmsg *rtm; struct nlmsghdr *nlh; @@ -2166,9 +2167,24 @@ static int rt6_fill_node(struct sk_buff *skb, struct rt6_info *rt, } else if (rtm->rtm_src_len) NLA_PUT(skb, RTA_SRC, 16, &rt->rt6i_src.addr); #endif - if (iif) - NLA_PUT_U32(skb, RTA_IIF, iif); - else if (dst) { + if (iif) { +#ifdef CONFIG_IPV6_MROUTE + if (ipv6_addr_is_multicast(&rt->rt6i_dst.addr)) { + int err = ip6mr_get_route(skb, rtm, nowait); + if (err <= 0) { + if (!nowait) { + if (err == 0) + return 0; + goto nla_put_failure; + } else { + if (err == -EMSGSIZE) + goto nla_put_failure; + } + } + } else +#endif + NLA_PUT_U32(skb, RTA_IIF, iif); + } else if (dst) { struct in6_addr saddr_buf; if (ipv6_dev_get_saddr(ip6_dst_idev(&rt->u.dst)->dev, dst, 0, &saddr_buf) == 0) @@ -2211,7 +2227,7 @@ int rt6_dump_route(struct rt6_info *rt, void *p_arg) return rt6_fill_node(arg->skb, rt, NULL, NULL, 0, RTM_NEWROUTE, NETLINK_CB(arg->cb->skb).pid, arg->cb->nlh->nlmsg_seq, - prefix, NLM_F_MULTI); + prefix, 0, NLM_F_MULTI); } static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void *arg) @@ -2277,7 +2293,7 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh, void err = rt6_fill_node(skb, rt, &fl.fl6_dst, &fl.fl6_src, iif, RTM_NEWROUTE, NETLINK_CB(in_skb).pid, - nlh->nlmsg_seq, 0, 0); + nlh->nlmsg_seq, 0, 0, 0); if (err < 0) { kfree_skb(skb); goto errout; @@ -2303,7 +2319,7 @@ void inet6_rt_notify(int event, struct rt6_info *rt, struct nl_info *info) goto errout; err = rt6_fill_node(skb, rt, NULL, NULL, 0, - event, info->pid, seq, 0, 0); + event, info->pid, seq, 0, 0, 0); if (err < 0) { /* -EMSGSIZE implies BUG in rt6_nlmsg_size() */ WARN_ON(err == -EMSGSIZE); -- cgit v1.2.3 From 14fb64e1f449ef6666f1c3a3fa4e13aec669b98d Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 3 Apr 2008 09:22:54 +0900 Subject: [IPV6] MROUTE: Support PIM-SM (SSM). Based on ancient patch by Mickael Hoerdt , which is available at . Signed-off-by: YOSHIFUJI Hideaki --- include/linux/mroute6.h | 4 + net/ipv6/Kconfig | 7 ++ net/ipv6/ip6mr.c | 275 +++++++++++++++++++++++++++++++++++++++++++++++- 3 files changed, 285 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h index b92190304e0b..f6469fb90840 100644 --- a/include/linux/mroute6.h +++ b/include/linux/mroute6.h @@ -23,6 +23,8 @@ #define MRT6_ADD_MFC (MRT6_BASE+4) /* Add a multicast forwarding entry */ #define MRT6_DEL_MFC (MRT6_BASE+5) /* Delete a multicast forwarding entry */ #define MRT6_VERSION (MRT6_BASE+6) /* Get the kernel multicast version */ +#define MRT6_ASSERT (MRT6_BASE+7) /* Activate PIM assert mode */ +#define MRT6_PIM (MRT6_BASE+8) /* enable PIM code */ #define SIOCGETMIFCNT_IN6 SIOCPROTOPRIVATE /* IP protocol privates */ #define SIOCGETSGCNT_IN6 (SIOCPROTOPRIVATE+1) @@ -217,6 +219,8 @@ static inline int ip6mr_sk_done(struct sock *sk) { return 0; } struct mrt6msg { #define MRT6MSG_NOCACHE 1 +#define MRT6MSG_WRONGMIF 2 +#define MRT6MSG_WHOLEPKT 3 /* used for use level encap */ __u8 im6_mbz; /* must be zero */ __u8 im6_msgtype; /* what type of message */ __u16 im6_mif; /* mif rec'd on */ diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig index 9a2ea81e499f..82f987b4ef84 100644 --- a/net/ipv6/Kconfig +++ b/net/ipv6/Kconfig @@ -216,3 +216,10 @@ config IPV6_MROUTE Experimental support for IPv6 multicast forwarding. If unsure, say N. +config IPV6_PIMSM_V2 + bool "IPv6: PIM-SM version 2 support (EXPERIMENTAL)" + depends on IPV6_MROUTE + ---help--- + Support for IPv6 PIM multicast routing protocol PIM-SMv2. + If unsure, say N. + diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 1bdf3c177d58..2b70774be61f 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -54,6 +54,7 @@ #include #include #include +#include #include #include @@ -75,6 +76,13 @@ static int maxvif; #define MIF_EXISTS(idx) (vif6_table[idx].dev != NULL) +static int mroute_do_assert; /* Set in PIM assert */ +#ifdef CONFIG_IPV6_PIMSM_V2 +static int mroute_do_pim; +#else +#define mroute_do_pim 0 +#endif + static struct mfc6_cache *mfc6_cache_array[MFC_LINES]; /* Forwarding cache */ static struct mfc6_cache *mfc_unres_queue; /* Queue of unresolved entries */ @@ -97,6 +105,10 @@ static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache); static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert); static int ip6mr_fill_mroute(struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm); +#ifdef CONFIG_IPV6_PIMSM_V2 +static struct inet6_protocol pim6_protocol; +#endif + static struct timer_list ipmr_expire_timer; @@ -339,6 +351,132 @@ static struct file_operations ip6mr_mfc_fops = { }; #endif +#ifdef CONFIG_IPV6_PIMSM_V2 +static int reg_vif_num = -1; + +static int pim6_rcv(struct sk_buff *skb) +{ + struct pimreghdr *pim; + struct ipv6hdr *encap; + struct net_device *reg_dev = NULL; + + if (!pskb_may_pull(skb, sizeof(*pim) + sizeof(*encap))) + goto drop; + + pim = (struct pimreghdr *)skb_transport_header(skb); + if (pim->type != ((PIM_VERSION << 4) | PIM_REGISTER) || + (pim->flags & PIM_NULL_REGISTER) || + (ip_compute_csum((void *)pim, sizeof(*pim)) != 0 && + (u16)csum_fold(skb_checksum(skb, 0, skb->len, 0)))) + goto drop; + + /* check if the inner packet is destined to mcast group */ + encap = (struct ipv6hdr *)(skb_transport_header(skb) + + sizeof(*pim)); + + if (!ipv6_addr_is_multicast(&encap->daddr) || + encap->payload_len == 0 || + ntohs(encap->payload_len) + sizeof(*pim) > skb->len) + goto drop; + + read_lock(&mrt_lock); + if (reg_vif_num >= 0) + reg_dev = vif6_table[reg_vif_num].dev; + if (reg_dev) + dev_hold(reg_dev); + read_unlock(&mrt_lock); + + if (reg_dev == NULL) + goto drop; + + skb->mac_header = skb->network_header; + skb_pull(skb, (u8 *)encap - skb->data); + skb_reset_network_header(skb); + skb->dev = reg_dev; + skb->protocol = htons(ETH_P_IP); + skb->ip_summed = 0; + skb->pkt_type = PACKET_HOST; + dst_release(skb->dst); + ((struct net_device_stats *)netdev_priv(reg_dev))->rx_bytes += skb->len; + ((struct net_device_stats *)netdev_priv(reg_dev))->rx_packets++; + skb->dst = NULL; + nf_reset(skb); + netif_rx(skb); + dev_put(reg_dev); + return 0; + drop: + kfree_skb(skb); + return 0; +} + +static struct inet6_protocol pim6_protocol = { + .handler = pim6_rcv, +}; + +/* Service routines creating virtual interfaces: PIMREG */ + +static int reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) +{ + read_lock(&mrt_lock); + ((struct net_device_stats *)netdev_priv(dev))->tx_bytes += skb->len; + ((struct net_device_stats *)netdev_priv(dev))->tx_packets++; + ip6mr_cache_report(skb, reg_vif_num, MRT6MSG_WHOLEPKT); + read_unlock(&mrt_lock); + kfree_skb(skb); + return 0; +} + +static struct net_device_stats *reg_vif_get_stats(struct net_device *dev) +{ + return (struct net_device_stats *)netdev_priv(dev); +} + +static void reg_vif_setup(struct net_device *dev) +{ + dev->type = ARPHRD_PIMREG; + dev->mtu = 1500 - sizeof(struct ipv6hdr) - 8; + dev->flags = IFF_NOARP; + dev->hard_start_xmit = reg_vif_xmit; + dev->get_stats = reg_vif_get_stats; + dev->destructor = free_netdev; +} + +static struct net_device *ip6mr_reg_vif(void) +{ + struct net_device *dev; + struct inet6_dev *in_dev; + + dev = alloc_netdev(sizeof(struct net_device_stats), "pim6reg", + reg_vif_setup); + + if (dev == NULL) + return NULL; + + if (register_netdevice(dev)) { + free_netdev(dev); + return NULL; + } + dev->iflink = 0; + + in_dev = ipv6_find_idev(dev); + if (!in_dev) + goto failure; + + if (dev_open(dev)) + goto failure; + + return dev; + +failure: + /* allow the register to be completed before unregistering. */ + rtnl_unlock(); + rtnl_lock(); + + unregister_netdevice(dev); + return NULL; +} +#endif + /* * Delete a VIF entry */ @@ -361,6 +499,11 @@ static int mif6_delete(int vifi) return -EADDRNOTAVAIL; } +#ifdef CONFIG_IPV6_PIMSM_V2 + if (vifi == reg_vif_num) + reg_vif_num = -1; +#endif + if (vifi + 1 == maxvif) { int tmp; for (tmp = vifi - 1; tmp >= 0; tmp--) { @@ -480,6 +623,19 @@ static int mif6_add(struct mif6ctl *vifc, int mrtsock) return -EADDRINUSE; switch (vifc->mif6c_flags) { +#ifdef CONFIG_IPV6_PIMSM_V2 + case MIFF_REGISTER: + /* + * Special Purpose VIF in PIM + * All the packets will be sent to the daemon + */ + if (reg_vif_num >= 0) + return -EADDRINUSE; + dev = ip6mr_reg_vif(); + if (!dev) + return -ENOBUFS; + break; +#endif case 0: dev = dev_get_by_index(&init_net, vifc->mif6c_pifi); if (!dev) @@ -512,6 +668,10 @@ static int mif6_add(struct mif6ctl *vifc, int mrtsock) write_lock_bh(&mrt_lock); dev_hold(dev); v->dev = dev; +#ifdef CONFIG_IPV6_PIMSM_V2 + if (v->flags & MIFF_REGISTER) + reg_vif_num = vifi; +#endif if (vifi + 1 > maxvif) maxvif = vifi + 1; write_unlock_bh(&mrt_lock); @@ -599,7 +759,13 @@ static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) struct mrt6msg *msg; int ret; - skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC); +#ifdef CONFIG_IPV6_PIMSM_V2 + if (assert == MRT6MSG_WHOLEPKT) + skb = skb_realloc_headroom(pkt, -skb_network_offset(pkt) + +sizeof(*msg)); + else +#endif + skb = alloc_skb(sizeof(struct ipv6hdr) + sizeof(*msg), GFP_ATOMIC); if (!skb) return -ENOBUFS; @@ -609,6 +775,29 @@ static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) skb->ip_summed = CHECKSUM_UNNECESSARY; +#ifdef CONFIG_IPV6_PIMSM_V2 + if (assert == MRT6MSG_WHOLEPKT) { + /* Ugly, but we have no choice with this interface. + Duplicate old header, fix length etc. + And all this only to mangle msg->im6_msgtype and + to set msg->im6_mbz to "mbz" :-) + */ + skb_push(skb, -skb_network_offset(pkt)); + + skb_push(skb, sizeof(*msg)); + skb_reset_transport_header(skb); + msg = (struct mrt6msg *)skb_transport_header(skb); + msg->im6_mbz = 0; + msg->im6_msgtype = MRT6MSG_WHOLEPKT; + msg->im6_mif = reg_vif_num; + msg->im6_pad = 0; + ipv6_addr_copy(&msg->im6_src, &ipv6_hdr(pkt)->saddr); + ipv6_addr_copy(&msg->im6_dst, &ipv6_hdr(pkt)->daddr); + + skb->ip_summed = CHECKSUM_UNNECESSARY; + } else +#endif + { /* * Copy the IP header */ @@ -635,6 +824,7 @@ static int ip6mr_cache_report(struct sk_buff *pkt, vifi_t vifi, int assert) skb->ip_summed = CHECKSUM_UNNECESSARY; skb_pull(skb, sizeof(struct ipv6hdr)); + } if (mroute6_socket == NULL) { kfree_skb(skb); @@ -1033,6 +1223,44 @@ int ip6_mroute_setsockopt(struct sock *sk, int optname, char __user *optval, int rtnl_unlock(); return ret; + /* + * Control PIM assert (to activate pim will activate assert) + */ + case MRT6_ASSERT: + { + int v; + if (get_user(v, (int __user *)optval)) + return -EFAULT; + mroute_do_assert = !!v; + return 0; + } + +#ifdef CONFIG_IPV6_PIMSM_V2 + case MRT6_PIM: + { + int v, ret; + if (get_user(v, (int __user *)optval)) + return -EFAULT; + v = !!v; + rtnl_lock(); + ret = 0; + if (v != mroute_do_pim) { + mroute_do_pim = v; + mroute_do_assert = v; + if (mroute_do_pim) + ret = inet6_add_protocol(&pim6_protocol, + IPPROTO_PIM); + else + ret = inet6_del_protocol(&pim6_protocol, + IPPROTO_PIM); + if (ret < 0) + ret = -EAGAIN; + } + rtnl_unlock(); + return ret; + } + +#endif /* * Spurious command, or MRT_VERSION which you cannot * set. @@ -1056,6 +1284,14 @@ int ip6_mroute_getsockopt(struct sock *sk, int optname, char __user *optval, case MRT6_VERSION: val = 0x0305; break; +#ifdef CONFIG_IPV6_PIMSM_V2 + case MRT6_PIM: + val = mroute_do_pim; + break; +#endif + case MRT6_ASSERT: + val = mroute_do_assert; + break; default: return -ENOPROTOOPT; } @@ -1151,6 +1387,18 @@ static int ip6mr_forward2(struct sk_buff *skb, struct mfc6_cache *c, int vifi) if (vif->dev == NULL) goto out_free; +#ifdef CONFIG_IPV6_PIMSM_V2 + if (vif->flags & MIFF_REGISTER) { + vif->pkt_out++; + vif->bytes_out += skb->len; + ((struct net_device_stats *)netdev_priv(vif->dev))->tx_bytes += skb->len; + ((struct net_device_stats *)netdev_priv(vif->dev))->tx_packets++; + ip6mr_cache_report(skb, vifi, MRT6MSG_WHOLEPKT); + kfree_skb(skb); + return 0; + } +#endif + ipv6h = ipv6_hdr(skb); fl = (struct flowi) { @@ -1220,6 +1468,30 @@ static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache) cache->mfc_un.res.pkt++; cache->mfc_un.res.bytes += skb->len; + /* + * Wrong interface: drop packet and (maybe) send PIM assert. + */ + if (vif6_table[vif].dev != skb->dev) { + int true_vifi; + + cache->mfc_un.res.wrong_if++; + true_vifi = ip6mr_find_vif(skb->dev); + + if (true_vifi >= 0 && mroute_do_assert && + /* pimsm uses asserts, when switching from RPT to SPT, + so that we cannot check that packet arrived on an oif. + It is bad, but otherwise we would need to move pretty + large chunk of pimd to kernel. Ough... --ANK + */ + (mroute_do_pim || cache->mfc_un.res.ttls[true_vifi] < 255) && + time_after(jiffies, + cache->mfc_un.res.last_assert + MFC_ASSERT_THRESH)) { + cache->mfc_un.res.last_assert = jiffies; + ip6mr_cache_report(skb, true_vifi, MRT6MSG_WRONGMIF); + } + goto dont_forward; + } + vif6_table[vif].pkt_in++; vif6_table[vif].bytes_in += skb->len; @@ -1241,6 +1513,7 @@ static int ip6_mr_forward(struct sk_buff *skb, struct mfc6_cache *cache) return 0; } +dont_forward: kfree_skb(skb); return 0; } -- cgit v1.2.3 From 12802d058a003048104fe405a8d283b94ac50801 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 3 Apr 2008 09:22:56 +0900 Subject: [IPV6]: Comment MRT6_xxx sockopts in include/linux/in6.h. Signed-off-by: YOSHIFUJI Hideaki --- include/linux/in6.h | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'include') diff --git a/include/linux/in6.h b/include/linux/in6.h index f674000c6c99..e6aa8de2b939 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -260,4 +260,19 @@ struct in6_flowlabel_req #define IPV6_PREFER_SRC_CGA 0x0008 #define IPV6_PREFER_SRC_NONCGA 0x0800 +/* + * Multicast Routing: + * see include/linux/mroute6.h. + * + * MRT6_INIT 200 + * MRT6_DONE 201 + * MRT6_ADD_MIF 202 + * MRT6_DEL_MIF 203 + * MRT6_ADD_MFC 204 + * MRT6_DEL_MFC 205 + * MRT6_VERSION 206 + * MRT6_ASSERT 207 + * MRT6_PIM 208 + * (reserved) 209 + */ #endif -- cgit v1.2.3 From 38668c059f5202f5fd9612391f9aa1b38a97241b Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Fri, 28 Mar 2008 16:33:32 -0700 Subject: mac80211: eliminate conf_ht This patch eliminates the use of conf_ht, replacing it with bss_info_changed. Signed-off-by: Tomas Winkler Signed-off-by: Ron Rindjunsky Signed-off-by: John W. Linville --- include/net/mac80211.h | 10 +++++- net/mac80211/ieee80211.c | 73 ++++++++++++++++++++++++++------------------ net/mac80211/ieee80211_i.h | 6 ++-- net/mac80211/ieee80211_sta.c | 33 +++++++++++--------- 4 files changed, 75 insertions(+), 47 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 48428a6b9109..5174eaa89500 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -178,11 +178,13 @@ struct ieee80211_low_level_stats { * also implies a change in the AID. * @BSS_CHANGED_ERP_CTS_PROT: CTS protection changed * @BSS_CHANGED_ERP_PREAMBLE: preamble changed + * @BSS_CHANGED_HT: 802.11n parameters changed */ enum ieee80211_bss_change { BSS_CHANGED_ASSOC = 1<<0, BSS_CHANGED_ERP_CTS_PROT = 1<<1, BSS_CHANGED_ERP_PREAMBLE = 1<<2, + BSS_CHANGED_HT = 1<<4, }; /** @@ -195,6 +197,9 @@ enum ieee80211_bss_change { * @aid: association ID number, valid only when @assoc is true * @use_cts_prot: use CTS protection * @use_short_preamble: use 802.11b short preamble + * @assoc_ht: association in HT mode + * @ht_conf: ht capabilities + * @ht_bss_conf: ht extended capabilities */ struct ieee80211_bss_conf { /* association related data */ @@ -203,6 +208,10 @@ struct ieee80211_bss_conf { /* erp related data */ bool use_cts_prot; bool use_short_preamble; + /* ht related data */ + bool assoc_ht; + struct ieee80211_ht_info *ht_conf; + struct ieee80211_ht_bss_info *ht_bss_conf; }; /** @@ -1132,7 +1141,6 @@ struct ieee80211_ops { struct sk_buff *skb, struct ieee80211_tx_control *control); int (*tx_last_beacon)(struct ieee80211_hw *hw); - int (*conf_ht)(struct ieee80211_hw *hw, struct ieee80211_conf *conf); int (*ampdu_action)(struct ieee80211_hw *hw, enum ieee80211_ampdu_mlme_action action, const u8 *addr, u16 tid, u16 *ssn); diff --git a/net/mac80211/ieee80211.c b/net/mac80211/ieee80211.c index aaa5480e204a..5d30dd463f22 100644 --- a/net/mac80211/ieee80211.c +++ b/net/mac80211/ieee80211.c @@ -1046,54 +1046,69 @@ int ieee80211_hw_config(struct ieee80211_local *local) } /** - * ieee80211_hw_config_ht should be used only after legacy configuration - * has been determined, as ht configuration depends upon the hardware's - * HT abilities for a _specific_ band. + * ieee80211_handle_ht should be used only after legacy configuration + * has been determined namely band, as ht configuration depends upon + * the hardware's HT abilities for a _specific_ band. */ -int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht, +u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht, struct ieee80211_ht_info *req_ht_cap, struct ieee80211_ht_bss_info *req_bss_cap) { struct ieee80211_conf *conf = &local->hw.conf; struct ieee80211_supported_band *sband; + struct ieee80211_ht_info ht_conf; + struct ieee80211_ht_bss_info ht_bss_conf; int i; + u32 changed = 0; sband = local->hw.wiphy->bands[conf->channel->band]; /* HT is not supported */ if (!sband->ht_info.ht_supported) { conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE; - return -EOPNOTSUPP; + return 0; } - /* disable HT */ - if (!enable_ht) { - conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE; - } else { + memset(&ht_conf, 0, sizeof(struct ieee80211_ht_info)); + memset(&ht_bss_conf, 0, sizeof(struct ieee80211_ht_bss_info)); + + if (enable_ht) { + if (!(conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE)) + changed |= BSS_CHANGED_HT; + conf->flags |= IEEE80211_CONF_SUPPORT_HT_MODE; - conf->ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap; - conf->ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS); - conf->ht_conf.cap |= - sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS; - conf->ht_bss_conf.primary_channel = - req_bss_cap->primary_channel; - conf->ht_bss_conf.bss_cap = req_bss_cap->bss_cap; - conf->ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode; + ht_conf.ht_supported = 1; + + ht_conf.cap = req_ht_cap->cap & sband->ht_info.cap; + ht_conf.cap &= ~(IEEE80211_HT_CAP_MIMO_PS); + ht_conf.cap |= sband->ht_info.cap & IEEE80211_HT_CAP_MIMO_PS; + for (i = 0; i < SUPP_MCS_SET_LEN; i++) - conf->ht_conf.supp_mcs_set[i] = - sband->ht_info.supp_mcs_set[i] & - req_ht_cap->supp_mcs_set[i]; - - /* In STA mode, this gives us indication - * to the AP's mode of operation */ - conf->ht_conf.ht_supported = 1; - conf->ht_conf.ampdu_factor = req_ht_cap->ampdu_factor; - conf->ht_conf.ampdu_density = req_ht_cap->ampdu_density; + ht_conf.supp_mcs_set[i] = + sband->ht_info.supp_mcs_set[i] & + req_ht_cap->supp_mcs_set[i]; + + ht_bss_conf.primary_channel = req_bss_cap->primary_channel; + ht_bss_conf.bss_cap = req_bss_cap->bss_cap; + ht_bss_conf.bss_op_mode = req_bss_cap->bss_op_mode; + + ht_conf.ampdu_factor = req_ht_cap->ampdu_factor; + ht_conf.ampdu_density = req_ht_cap->ampdu_density; + + /* if bss configuration changed store the new one */ + if (memcmp(&conf->ht_conf, &ht_conf, sizeof(ht_conf)) || + memcmp(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf))) { + changed |= BSS_CHANGED_HT; + memcpy(&conf->ht_conf, &ht_conf, sizeof(ht_conf)); + memcpy(&conf->ht_bss_conf, &ht_bss_conf, sizeof(ht_bss_conf)); + } + } else { + if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) + changed |= BSS_CHANGED_HT; + conf->flags &= ~IEEE80211_CONF_SUPPORT_HT_MODE; } - local->ops->conf_ht(local_to_hw(local), &local->hw.conf); - - return 0; + return changed; } void ieee80211_bss_info_change_notify(struct ieee80211_sub_if_data *sdata, diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 0997a0f96203..377c448a0556 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -865,9 +865,9 @@ int ieee80211_if_config(struct net_device *dev); int ieee80211_if_config_beacon(struct net_device *dev); void ieee80211_tx_set_protected(struct ieee80211_tx_data *tx); void ieee80211_if_setup(struct net_device *dev); -int ieee80211_hw_config_ht(struct ieee80211_local *local, int enable_ht, - struct ieee80211_ht_info *req_ht_cap, - struct ieee80211_ht_bss_info *req_bss_cap); +u32 ieee80211_handle_ht(struct ieee80211_local *local, int enable_ht, + struct ieee80211_ht_info *req_ht_cap, + struct ieee80211_ht_bss_info *req_bss_cap); /* ieee80211_ioctl.c */ extern const struct iw_handler_def ieee80211_iw_handler_def; diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index 1ee07f0b02e2..0e936233dac5 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -493,6 +493,7 @@ static void ieee80211_set_associated(struct net_device *dev, { struct ieee80211_sub_if_data *sdata = IEEE80211_DEV_TO_SUB_IF(dev); struct ieee80211_local *local = sdata->local; + struct ieee80211_conf *conf = &local_to_hw(local)->conf; union iwreq_data wrqu; u32 changed = BSS_CHANGED_ASSOC; @@ -505,7 +506,7 @@ static void ieee80211_set_associated(struct net_device *dev, return; bss = ieee80211_rx_bss_get(dev, ifsta->bssid, - local->hw.conf.channel->center_freq, + conf->channel->center_freq, ifsta->ssid, ifsta->ssid_len); if (bss) { if (bss->has_erp_value) @@ -514,6 +515,13 @@ static void ieee80211_set_associated(struct net_device *dev, ieee80211_rx_bss_put(dev, bss); } + if (conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) { + changed |= BSS_CHANGED_HT; + sdata->bss_conf.assoc_ht = 1; + sdata->bss_conf.ht_conf = &conf->ht_conf; + sdata->bss_conf.ht_bss_conf = &conf->ht_bss_conf; + } + netif_carrier_on(dev); ifsta->flags |= IEEE80211_STA_PREV_BSSID_SET; memcpy(ifsta->prev_bssid, sdata->u.sta.bssid, ETH_ALEN); @@ -524,6 +532,11 @@ static void ieee80211_set_associated(struct net_device *dev, ifsta->flags &= ~IEEE80211_STA_ASSOCIATED; netif_carrier_off(dev); ieee80211_reset_erp_info(dev); + + sdata->bss_conf.assoc_ht = 0; + sdata->bss_conf.ht_conf = NULL; + sdata->bss_conf.ht_bss_conf = NULL; + memset(wrqu.ap_addr.sa_data, 0, ETH_ALEN); } wrqu.ap_addr.sa_family = ARPHRD_ETHER; @@ -1999,17 +2012,15 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, else sdata->flags &= ~IEEE80211_SDATA_OPERATING_GMODE; - if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param && - local->ops->conf_ht) { + if (elems.ht_cap_elem && elems.ht_info_elem && elems.wmm_param) { struct ieee80211_ht_bss_info bss_info; - ieee80211_ht_cap_ie_to_ht_info( (struct ieee80211_ht_cap *) elems.ht_cap_elem, &sta->ht_info); ieee80211_ht_addt_info_ie_to_ht_bss_info( (struct ieee80211_ht_addt_info *) elems.ht_info_elem, &bss_info); - ieee80211_hw_config_ht(local, 1, &sta->ht_info, &bss_info); + ieee80211_handle_ht(local, 1, &sta->ht_info, &bss_info); } rate_control_rate_init(sta, local); @@ -2760,20 +2771,14 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev, changed |= ieee80211_handle_erp_ie(sdata, elems.erp_info[0]); if (elems.ht_cap_elem && elems.ht_info_elem && - elems.wmm_param && local->ops->conf_ht && - conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) { + elems.wmm_param && conf->flags & IEEE80211_CONF_SUPPORT_HT_MODE) { struct ieee80211_ht_bss_info bss_info; ieee80211_ht_addt_info_ie_to_ht_bss_info( (struct ieee80211_ht_addt_info *) elems.ht_info_elem, &bss_info); - /* check if AP changed bss inforamation */ - if ((conf->ht_bss_conf.primary_channel != - bss_info.primary_channel) || - (conf->ht_bss_conf.bss_cap != bss_info.bss_cap) || - (conf->ht_bss_conf.bss_op_mode != bss_info.bss_op_mode)) - ieee80211_hw_config_ht(local, 1, &conf->ht_conf, - &bss_info); + changed |= ieee80211_handle_ht(local, 1, &conf->ht_conf, + &bss_info); } if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) { -- cgit v1.2.3 From 21c0cbe760ca6b5d4c6927c3ec1352a843a8c11c Mon Sep 17 00:00:00 2001 From: Tomas Winkler Date: Fri, 28 Mar 2008 16:33:34 -0700 Subject: mac80211: add association capabilty and timing info into bss_conf This patch adds assocation capability, timestamp (tsf) and beacon interval to bss_conf. This is required for successful assocation of iwlwifi drivers Signed-off-by: Tomas Winkler Signed-off-by: Gregory Greenman Signed-off-by: John W. Linville --- include/net/mac80211.h | 6 ++++++ net/mac80211/ieee80211_sta.c | 9 ++++++++- 2 files changed, 14 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 5174eaa89500..01b32152b89e 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -197,6 +197,9 @@ enum ieee80211_bss_change { * @aid: association ID number, valid only when @assoc is true * @use_cts_prot: use CTS protection * @use_short_preamble: use 802.11b short preamble + * @timestamp: beacon timestamp + * @beacon_int: beacon interval + * @assoc_capability: capabbilities taken from assoc resp * @assoc_ht: association in HT mode * @ht_conf: ht capabilities * @ht_bss_conf: ht extended capabilities @@ -208,6 +211,9 @@ struct ieee80211_bss_conf { /* erp related data */ bool use_cts_prot; bool use_short_preamble; + u16 beacon_int; + u16 assoc_capability; + u64 timestamp; /* ht related data */ bool assoc_ht; struct ieee80211_ht_info *ht_conf; diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index 0e936233dac5..cb119d3b6353 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -509,9 +509,14 @@ static void ieee80211_set_associated(struct net_device *dev, conf->channel->center_freq, ifsta->ssid, ifsta->ssid_len); if (bss) { + /* set timing information */ + sdata->bss_conf.beacon_int = bss->beacon_int; + sdata->bss_conf.timestamp = bss->timestamp; + if (bss->has_erp_value) changed |= ieee80211_handle_erp_ie( sdata, bss->erp_value); + ieee80211_rx_bss_put(dev, bss); } @@ -2033,8 +2038,10 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, } else rcu_read_unlock(); - /* set AID, ieee80211_set_associated() will tell the driver */ + /* set AID and assoc capability, + * ieee80211_set_associated() will tell the driver */ bss_conf->aid = aid; + bss_conf->assoc_capability = capab_info; ieee80211_set_associated(dev, ifsta, 1); ieee80211_associated(dev, ifsta); -- cgit v1.2.3 From 8fe2b65a18e49bfde56a59ed4ab3fc7aa0c2f325 Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Sun, 30 Mar 2008 00:10:50 +0100 Subject: ssb: Turn suspend/resume upside down Turn the SSB bus suspend mechanism upside down. Instead of deciding by an internal reference count when to suspend/resume, let the parent bus call us in their suspend/resume routine. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/net/wireless/b43/pcmcia.c | 10 ++-- drivers/ssb/driver_chipcommon.c | 2 +- drivers/ssb/main.c | 81 +++++++++++++------------------ drivers/ssb/pcihost_wrapper.c | 10 ++++ drivers/ssb/pcmcia.c | 34 +++++++++---- drivers/ssb/ssb_private.h | 5 ++ include/linux/ssb/ssb.h | 10 ++-- include/linux/ssb/ssb_driver_chipcommon.h | 3 +- 8 files changed, 88 insertions(+), 67 deletions(-) (limited to 'include') diff --git a/drivers/net/wireless/b43/pcmcia.c b/drivers/net/wireless/b43/pcmcia.c index 371e4a119511..b8aa16307f79 100644 --- a/drivers/net/wireless/b43/pcmcia.c +++ b/drivers/net/wireless/b43/pcmcia.c @@ -43,14 +43,16 @@ MODULE_DEVICE_TABLE(pcmcia, b43_pcmcia_tbl); #ifdef CONFIG_PM static int b43_pcmcia_suspend(struct pcmcia_device *dev) { - //TODO - return 0; + struct ssb_bus *ssb = dev->priv; + + return ssb_bus_suspend(ssb); } static int b43_pcmcia_resume(struct pcmcia_device *dev) { - //TODO - return 0; + struct ssb_bus *ssb = dev->priv; + + return ssb_bus_resume(ssb); } #else /* CONFIG_PM */ # define b43_pcmcia_suspend NULL diff --git a/drivers/ssb/driver_chipcommon.c b/drivers/ssb/driver_chipcommon.c index 45b672a69003..571f4fd55236 100644 --- a/drivers/ssb/driver_chipcommon.c +++ b/drivers/ssb/driver_chipcommon.c @@ -251,7 +251,7 @@ void ssb_chipcommon_init(struct ssb_chipcommon *cc) calc_fast_powerup_delay(cc); } -void ssb_chipco_suspend(struct ssb_chipcommon *cc, pm_message_t state) +void ssb_chipco_suspend(struct ssb_chipcommon *cc) { if (!cc->dev) return; diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c index 2fcfd73b3b6e..c0cbdba07aee 100644 --- a/drivers/ssb/main.c +++ b/drivers/ssb/main.c @@ -120,35 +120,12 @@ static void ssb_device_put(struct ssb_device *dev) put_device(dev->dev); } -static int ssb_bus_resume(struct ssb_bus *bus) -{ - int err; - - ssb_pci_xtal(bus, SSB_GPIO_XTAL | SSB_GPIO_PLL, 1); - err = ssb_pcmcia_init(bus); - if (err) { - /* No need to disable XTAL, as we don't have one on PCMCIA. */ - return err; - } - ssb_chipco_resume(&bus->chipco); - - return 0; -} - static int ssb_device_resume(struct device *dev) { struct ssb_device *ssb_dev = dev_to_ssb_dev(dev); struct ssb_driver *ssb_drv; - struct ssb_bus *bus; int err = 0; - bus = ssb_dev->bus; - if (bus->suspend_cnt == bus->nr_devices) { - err = ssb_bus_resume(bus); - if (err) - return err; - } - bus->suspend_cnt--; if (dev->driver) { ssb_drv = drv_to_ssb_drv(dev->driver); if (ssb_drv && ssb_drv->resume) @@ -160,27 +137,10 @@ out: return err; } -static void ssb_bus_suspend(struct ssb_bus *bus, pm_message_t state) -{ - ssb_chipco_suspend(&bus->chipco, state); - ssb_pci_xtal(bus, SSB_GPIO_XTAL | SSB_GPIO_PLL, 0); - - /* Reset HW state information in memory, so that HW is - * completely reinitialized on resume. */ - bus->mapped_device = NULL; -#ifdef CONFIG_SSB_DRIVER_PCICORE - bus->pcicore.setup_done = 0; -#endif -#ifdef CONFIG_SSB_DEBUG - bus->powered_up = 0; -#endif -} - static int ssb_device_suspend(struct device *dev, pm_message_t state) { struct ssb_device *ssb_dev = dev_to_ssb_dev(dev); struct ssb_driver *ssb_drv; - struct ssb_bus *bus; int err = 0; if (dev->driver) { @@ -190,17 +150,44 @@ static int ssb_device_suspend(struct device *dev, pm_message_t state) if (err) goto out; } +out: + return err; +} + +int ssb_bus_resume(struct ssb_bus *bus) +{ + int err; - bus = ssb_dev->bus; - bus->suspend_cnt++; - if (bus->suspend_cnt == bus->nr_devices) { - /* All devices suspended. Shutdown the bus. */ - ssb_bus_suspend(bus, state); + /* Reset HW state information in memory, so that HW is + * completely reinitialized. */ + bus->mapped_device = NULL; +#ifdef CONFIG_SSB_DRIVER_PCICORE + bus->pcicore.setup_done = 0; +#endif + + err = ssb_bus_powerup(bus, 0); + if (err) + return err; + err = ssb_pcmcia_hardware_setup(bus); + if (err) { + ssb_bus_may_powerdown(bus); + return err; } + ssb_chipco_resume(&bus->chipco); + ssb_bus_may_powerdown(bus); -out: - return err; + return 0; +} +EXPORT_SYMBOL(ssb_bus_resume); + +int ssb_bus_suspend(struct ssb_bus *bus) +{ + ssb_chipco_suspend(&bus->chipco); + ssb_pci_xtal(bus, SSB_GPIO_XTAL | SSB_GPIO_PLL, 0); + + return 0; } +EXPORT_SYMBOL(ssb_bus_suspend); #ifdef CONFIG_SSB_PCIHOST int ssb_devices_freeze(struct ssb_bus *bus) diff --git a/drivers/ssb/pcihost_wrapper.c b/drivers/ssb/pcihost_wrapper.c index 82a10abef640..e82db4aaa050 100644 --- a/drivers/ssb/pcihost_wrapper.c +++ b/drivers/ssb/pcihost_wrapper.c @@ -18,6 +18,12 @@ #ifdef CONFIG_PM static int ssb_pcihost_suspend(struct pci_dev *dev, pm_message_t state) { + struct ssb_bus *ssb = pci_get_drvdata(dev); + int err; + + err = ssb_bus_suspend(ssb); + if (err) + return err; pci_save_state(dev); pci_disable_device(dev); pci_set_power_state(dev, pci_choose_state(dev, state)); @@ -27,6 +33,7 @@ static int ssb_pcihost_suspend(struct pci_dev *dev, pm_message_t state) static int ssb_pcihost_resume(struct pci_dev *dev) { + struct ssb_bus *ssb = pci_get_drvdata(dev); int err; pci_set_power_state(dev, 0); @@ -34,6 +41,9 @@ static int ssb_pcihost_resume(struct pci_dev *dev) if (err) return err; pci_restore_state(dev); + err = ssb_bus_resume(ssb); + if (err) + return err; return 0; } diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c index d674cef7210d..dcaf2412bea7 100644 --- a/drivers/ssb/pcmcia.c +++ b/drivers/ssb/pcmcia.c @@ -684,6 +684,29 @@ static int ssb_pcmcia_cor_setup(struct ssb_bus *bus, u8 cor) return 0; } +/* Initialize the PCMCIA hardware. This is called on Init and Resume. */ +int ssb_pcmcia_hardware_setup(struct ssb_bus *bus) +{ + int err; + + if (bus->bustype != SSB_BUSTYPE_PCMCIA) + return 0; + + /* Switch segment to a known state and sync + * bus->mapped_pcmcia_seg with hardware state. */ + ssb_pcmcia_switch_segment(bus, 0); + /* Init the COR register. */ + err = ssb_pcmcia_cor_setup(bus, CISREG_COR); + if (err) + return err; + /* Some cards also need this register to get poked. */ + err = ssb_pcmcia_cor_setup(bus, CISREG_COR + 0x80); + if (err) + return err; + + return 0; +} + void ssb_pcmcia_exit(struct ssb_bus *bus) { if (bus->bustype != SSB_BUSTYPE_PCMCIA) @@ -699,16 +722,7 @@ int ssb_pcmcia_init(struct ssb_bus *bus) if (bus->bustype != SSB_BUSTYPE_PCMCIA) return 0; - /* Switch segment to a known state and sync - * bus->mapped_pcmcia_seg with hardware state. */ - ssb_pcmcia_switch_segment(bus, 0); - - /* Init the COR register. */ - err = ssb_pcmcia_cor_setup(bus, CISREG_COR); - if (err) - goto error; - /* Some cards also need this register to get poked. */ - err = ssb_pcmcia_cor_setup(bus, CISREG_COR + 0x80); + err = ssb_pcmcia_hardware_setup(bus); if (err) goto error; diff --git a/drivers/ssb/ssb_private.h b/drivers/ssb/ssb_private.h index a83bf7a4d80b..ebc32d8fe15f 100644 --- a/drivers/ssb/ssb_private.h +++ b/drivers/ssb/ssb_private.h @@ -81,6 +81,7 @@ extern int ssb_pcmcia_switch_segment(struct ssb_bus *bus, u8 seg); extern int ssb_pcmcia_get_invariants(struct ssb_bus *bus, struct ssb_init_invariants *iv); +extern int ssb_pcmcia_hardware_setup(struct ssb_bus *bus); extern void ssb_pcmcia_exit(struct ssb_bus *bus); extern int ssb_pcmcia_init(struct ssb_bus *bus); extern const struct ssb_bus_ops ssb_pcmcia_ops; @@ -100,6 +101,10 @@ static inline int ssb_pcmcia_switch_segment(struct ssb_bus *bus, { return 0; } +static inline int ssb_pcmcia_hardware_setup(struct ssb_bus *bus) +{ + return 0; +} static inline void ssb_pcmcia_exit(struct ssb_bus *bus) { } diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h index 8644e03cf588..a8ca396f810a 100644 --- a/include/linux/ssb/ssb.h +++ b/include/linux/ssb/ssb.h @@ -260,9 +260,6 @@ struct ssb_bus { struct ssb_device devices[SSB_MAX_NR_CORES]; u8 nr_devices; - /* Reference count. Number of suspended devices. */ - u8 suspend_cnt; - /* Software ID number for this bus. */ unsigned int busnumber; @@ -334,6 +331,13 @@ extern int ssb_bus_pcmciabus_register(struct ssb_bus *bus, extern void ssb_bus_unregister(struct ssb_bus *bus); +/* Suspend a SSB bus. + * Call this from the parent bus suspend routine. */ +extern int ssb_bus_suspend(struct ssb_bus *bus); +/* Resume a SSB bus. + * Call this from the parent bus resume routine. */ +extern int ssb_bus_resume(struct ssb_bus *bus); + extern u32 ssb_clockspeed(struct ssb_bus *bus); /* Is the device enabled in hardware? */ diff --git a/include/linux/ssb/ssb_driver_chipcommon.h b/include/linux/ssb/ssb_driver_chipcommon.h index b548a54ff1f5..7d7e03dcf77c 100644 --- a/include/linux/ssb/ssb_driver_chipcommon.h +++ b/include/linux/ssb/ssb_driver_chipcommon.h @@ -367,8 +367,7 @@ static inline bool ssb_chipco_available(struct ssb_chipcommon *cc) extern void ssb_chipcommon_init(struct ssb_chipcommon *cc); -#include -extern void ssb_chipco_suspend(struct ssb_chipcommon *cc, pm_message_t state); +extern void ssb_chipco_suspend(struct ssb_chipcommon *cc); extern void ssb_chipco_resume(struct ssb_chipcommon *cc); extern void ssb_chipco_get_clockcpu(struct ssb_chipcommon *cc, -- cgit v1.2.3 From fff7710937f755099209357e5b5740d42a2c9f97 Mon Sep 17 00:00:00 2001 From: Chr Date: Tue, 1 Apr 2008 21:45:18 +0200 Subject: mac80211: add station aid into ieee80211_tx_control This patch is necessary for the upcoming Accesspoint patch for p54. Signed-off-by: Christian Lamparter Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 1 + net/mac80211/tx.c | 1 + 2 files changed, 2 insertions(+) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 01b32152b89e..999f970da6ba 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -302,6 +302,7 @@ struct ieee80211_tx_control { u8 iv_len; /* length of the IV field in octets */ u8 queue; /* hardware queue to use for this frame; * 0 = highest, hw->queues-1 = lowest */ + u16 aid; /* Station AID */ int type; /* internal */ }; diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 80f4343a3007..ea3fa0f91906 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -741,6 +741,7 @@ ieee80211_tx_h_misc(struct ieee80211_tx_data *tx) } if (tx->sta) { + control->aid = tx->sta->aid; tx->sta->tx_packets++; tx->sta->tx_fragments++; tx->sta->tx_bytes += tx->skb->len; -- cgit v1.2.3 From d625a29ba649a4df6027520ffc378f23c0e6883e Mon Sep 17 00:00:00 2001 From: Michael Buesch Date: Wed, 2 Apr 2008 19:46:56 +0200 Subject: ssb: Add support for block-I/O This adds support for block based I/O to SSB. This is needed in order to efficiently support PIO data transfers to the card. The block-I/O support is only compiled, if it's selected by the weird driver that needs it. So there's no overhead for sane devices. Signed-off-by: Michael Buesch Signed-off-by: John W. Linville --- drivers/ssb/Kconfig | 5 ++ drivers/ssb/main.c | 102 +++++++++++++++++++++++++++++++++++++++++ drivers/ssb/pci.c | 70 ++++++++++++++++++++++++++++ drivers/ssb/pcmcia.c | 119 ++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/ssb/ssb.h | 19 ++++++++ 5 files changed, 315 insertions(+) (limited to 'include') diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig index 0f7cce2560d1..cd845b8acd17 100644 --- a/drivers/ssb/Kconfig +++ b/drivers/ssb/Kconfig @@ -24,6 +24,11 @@ config SSB config SSB_SPROM bool +# Support for Block-I/O. SELECT this from the driver that needs it. +config SSB_BLOCKIO + bool + depends on SSB + config SSB_PCIHOST_POSSIBLE bool depends on SSB && (PCI = y || PCI = SSB) diff --git a/drivers/ssb/main.c b/drivers/ssb/main.c index 3e58db7ef608..19ddd2bd1057 100644 --- a/drivers/ssb/main.c +++ b/drivers/ssb/main.c @@ -555,6 +555,55 @@ static u32 ssb_ssb_read32(struct ssb_device *dev, u16 offset) return readl(bus->mmio + offset); } +#ifdef CONFIG_SSB_BLOCKIO +static void ssb_ssb_block_read(struct ssb_device *dev, void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + struct ssb_bus *bus = dev->bus; + void __iomem *addr; + + offset += dev->core_index * SSB_CORE_SIZE; + addr = bus->mmio + offset; + + switch (reg_width) { + case sizeof(u8): { + u8 *buf = buffer; + + while (count) { + *buf = __raw_readb(addr); + buf++; + count--; + } + break; + } + case sizeof(u16): { + __le16 *buf = buffer; + + SSB_WARN_ON(count & 1); + while (count) { + *buf = (__force __le16)__raw_readw(addr); + buf++; + count -= 2; + } + break; + } + case sizeof(u32): { + __le32 *buf = buffer; + + SSB_WARN_ON(count & 3); + while (count) { + *buf = (__force __le32)__raw_readl(addr); + buf++; + count -= 4; + } + break; + } + default: + SSB_WARN_ON(1); + } +} +#endif /* CONFIG_SSB_BLOCKIO */ + static void ssb_ssb_write8(struct ssb_device *dev, u16 offset, u8 value) { struct ssb_bus *bus = dev->bus; @@ -579,6 +628,55 @@ static void ssb_ssb_write32(struct ssb_device *dev, u16 offset, u32 value) writel(value, bus->mmio + offset); } +#ifdef CONFIG_SSB_BLOCKIO +static void ssb_ssb_block_write(struct ssb_device *dev, const void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + struct ssb_bus *bus = dev->bus; + void __iomem *addr; + + offset += dev->core_index * SSB_CORE_SIZE; + addr = bus->mmio + offset; + + switch (reg_width) { + case sizeof(u8): { + const u8 *buf = buffer; + + while (count) { + __raw_writeb(*buf, addr); + buf++; + count--; + } + break; + } + case sizeof(u16): { + const __le16 *buf = buffer; + + SSB_WARN_ON(count & 1); + while (count) { + __raw_writew((__force u16)(*buf), addr); + buf++; + count -= 2; + } + break; + } + case sizeof(u32): { + const __le32 *buf = buffer; + + SSB_WARN_ON(count & 3); + while (count) { + __raw_writel((__force u32)(*buf), addr); + buf++; + count -= 4; + } + break; + } + default: + SSB_WARN_ON(1); + } +} +#endif /* CONFIG_SSB_BLOCKIO */ + /* Ops for the plain SSB bus without a host-device (no PCI or PCMCIA). */ static const struct ssb_bus_ops ssb_ssb_ops = { .read8 = ssb_ssb_read8, @@ -587,6 +685,10 @@ static const struct ssb_bus_ops ssb_ssb_ops = { .write8 = ssb_ssb_write8, .write16 = ssb_ssb_write16, .write32 = ssb_ssb_write32, +#ifdef CONFIG_SSB_BLOCKIO + .block_read = ssb_ssb_block_read, + .block_write = ssb_ssb_block_write, +#endif }; static int ssb_fetch_invariants(struct ssb_bus *bus, diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c index f1514b33cfae..904b1a8d0885 100644 --- a/drivers/ssb/pci.c +++ b/drivers/ssb/pci.c @@ -613,6 +613,41 @@ static u32 ssb_pci_read32(struct ssb_device *dev, u16 offset) return ioread32(bus->mmio + offset); } +#ifdef CONFIG_SSB_BLOCKIO +static void ssb_pci_block_read(struct ssb_device *dev, void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + struct ssb_bus *bus = dev->bus; + void __iomem *addr = bus->mmio + offset; + + if (unlikely(ssb_pci_assert_buspower(bus))) + goto error; + if (unlikely(bus->mapped_device != dev)) { + if (unlikely(ssb_pci_switch_core(bus, dev))) + goto error; + } + switch (reg_width) { + case sizeof(u8): + ioread8_rep(addr, buffer, count); + break; + case sizeof(u16): + SSB_WARN_ON(count & 1); + ioread16_rep(addr, buffer, count >> 1); + break; + case sizeof(u32): + SSB_WARN_ON(count & 3); + ioread32_rep(addr, buffer, count >> 2); + break; + default: + SSB_WARN_ON(1); + } + + return; +error: + memset(buffer, 0xFF, count); +} +#endif /* CONFIG_SSB_BLOCKIO */ + static void ssb_pci_write8(struct ssb_device *dev, u16 offset, u8 value) { struct ssb_bus *bus = dev->bus; @@ -652,6 +687,37 @@ static void ssb_pci_write32(struct ssb_device *dev, u16 offset, u32 value) iowrite32(value, bus->mmio + offset); } +#ifdef CONFIG_SSB_BLOCKIO +static void ssb_pci_block_write(struct ssb_device *dev, const void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + struct ssb_bus *bus = dev->bus; + void __iomem *addr = bus->mmio + offset; + + if (unlikely(ssb_pci_assert_buspower(bus))) + return; + if (unlikely(bus->mapped_device != dev)) { + if (unlikely(ssb_pci_switch_core(bus, dev))) + return; + } + switch (reg_width) { + case sizeof(u8): + iowrite8_rep(addr, buffer, count); + break; + case sizeof(u16): + SSB_WARN_ON(count & 1); + iowrite16_rep(addr, buffer, count >> 1); + break; + case sizeof(u32): + SSB_WARN_ON(count & 3); + iowrite32_rep(addr, buffer, count >> 2); + break; + default: + SSB_WARN_ON(1); + } +} +#endif /* CONFIG_SSB_BLOCKIO */ + /* Not "static", as it's used in main.c */ const struct ssb_bus_ops ssb_pci_ops = { .read8 = ssb_pci_read8, @@ -660,6 +726,10 @@ const struct ssb_bus_ops ssb_pci_ops = { .write8 = ssb_pci_write8, .write16 = ssb_pci_write16, .write32 = ssb_pci_write32, +#ifdef CONFIG_SSB_BLOCKIO + .block_read = ssb_pci_block_read, + .block_write = ssb_pci_block_write, +#endif }; static ssize_t ssb_pci_attr_sprom_show(struct device *pcidev, diff --git a/drivers/ssb/pcmcia.c b/drivers/ssb/pcmcia.c index dcaf2412bea7..24c2a46c1476 100644 --- a/drivers/ssb/pcmcia.c +++ b/drivers/ssb/pcmcia.c @@ -285,6 +285,64 @@ static u32 ssb_pcmcia_read32(struct ssb_device *dev, u16 offset) return (lo | (hi << 16)); } +#ifdef CONFIG_SSB_BLOCKIO +static void ssb_pcmcia_block_read(struct ssb_device *dev, void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + struct ssb_bus *bus = dev->bus; + unsigned long flags; + void __iomem *addr = bus->mmio + offset; + int err; + + spin_lock_irqsave(&bus->bar_lock, flags); + err = select_core_and_segment(dev, &offset); + if (unlikely(err)) { + memset(buffer, 0xFF, count); + goto unlock; + } + switch (reg_width) { + case sizeof(u8): { + u8 *buf = buffer; + + while (count) { + *buf = __raw_readb(addr); + buf++; + count--; + } + break; + } + case sizeof(u16): { + __le16 *buf = buffer; + + SSB_WARN_ON(count & 1); + while (count) { + *buf = (__force __le16)__raw_readw(addr); + buf++; + count -= 2; + } + break; + } + case sizeof(u32): { + __le16 *buf = buffer; + + SSB_WARN_ON(count & 3); + while (count) { + *buf = (__force __le16)__raw_readw(addr); + buf++; + *buf = (__force __le16)__raw_readw(addr + 2); + buf++; + count -= 4; + } + break; + } + default: + SSB_WARN_ON(1); + } +unlock: + spin_unlock_irqrestore(&bus->bar_lock, flags); +} +#endif /* CONFIG_SSB_BLOCKIO */ + static void ssb_pcmcia_write8(struct ssb_device *dev, u16 offset, u8 value) { struct ssb_bus *bus = dev->bus; @@ -329,6 +387,63 @@ static void ssb_pcmcia_write32(struct ssb_device *dev, u16 offset, u32 value) spin_unlock_irqrestore(&bus->bar_lock, flags); } +#ifdef CONFIG_SSB_BLOCKIO +static void ssb_pcmcia_block_write(struct ssb_device *dev, const void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + struct ssb_bus *bus = dev->bus; + unsigned long flags; + void __iomem *addr = bus->mmio + offset; + int err; + + spin_lock_irqsave(&bus->bar_lock, flags); + err = select_core_and_segment(dev, &offset); + if (unlikely(err)) + goto unlock; + switch (reg_width) { + case sizeof(u8): { + const u8 *buf = buffer; + + while (count) { + __raw_writeb(*buf, addr); + buf++; + count--; + } + break; + } + case sizeof(u16): { + const __le16 *buf = buffer; + + SSB_WARN_ON(count & 1); + while (count) { + __raw_writew((__force u16)(*buf), addr); + buf++; + count -= 2; + } + break; + } + case sizeof(u32): { + const __le16 *buf = buffer; + + SSB_WARN_ON(count & 3); + while (count) { + __raw_writew((__force u16)(*buf), addr); + buf++; + __raw_writew((__force u16)(*buf), addr + 2); + buf++; + count -= 4; + } + break; + } + default: + SSB_WARN_ON(1); + } +unlock: + mmiowb(); + spin_unlock_irqrestore(&bus->bar_lock, flags); +} +#endif /* CONFIG_SSB_BLOCKIO */ + /* Not "static", as it's used in main.c */ const struct ssb_bus_ops ssb_pcmcia_ops = { .read8 = ssb_pcmcia_read8, @@ -337,6 +452,10 @@ const struct ssb_bus_ops ssb_pcmcia_ops = { .write8 = ssb_pcmcia_write8, .write16 = ssb_pcmcia_write16, .write32 = ssb_pcmcia_write32, +#ifdef CONFIG_SSB_BLOCKIO + .block_read = ssb_pcmcia_block_read, + .block_write = ssb_pcmcia_block_write, +#endif }; static int ssb_pcmcia_sprom_command(struct ssb_bus *bus, u8 command) diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h index a8ca396f810a..9f95afd0e9e3 100644 --- a/include/linux/ssb/ssb.h +++ b/include/linux/ssb/ssb.h @@ -78,6 +78,12 @@ struct ssb_bus_ops { void (*write8)(struct ssb_device *dev, u16 offset, u8 value); void (*write16)(struct ssb_device *dev, u16 offset, u16 value); void (*write32)(struct ssb_device *dev, u16 offset, u32 value); +#ifdef CONFIG_SSB_BLOCKIO + void (*block_read)(struct ssb_device *dev, void *buffer, + size_t count, u16 offset, u8 reg_width); + void (*block_write)(struct ssb_device *dev, const void *buffer, + size_t count, u16 offset, u8 reg_width); +#endif }; @@ -374,6 +380,19 @@ static inline void ssb_write32(struct ssb_device *dev, u16 offset, u32 value) { dev->ops->write32(dev, offset, value); } +#ifdef CONFIG_SSB_BLOCKIO +static inline void ssb_block_read(struct ssb_device *dev, void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + dev->ops->block_read(dev, buffer, count, offset, reg_width); +} + +static inline void ssb_block_write(struct ssb_device *dev, const void *buffer, + size_t count, u16 offset, u8 reg_width) +{ + dev->ops->block_write(dev, buffer, count, offset, reg_width); +} +#endif /* CONFIG_SSB_BLOCKIO */ /* Translation (routing) bits that need to be ORed to DMA -- cgit v1.2.3 From 84363e6e07f17f8cc580065260907ee3f0520485 Mon Sep 17 00:00:00 2001 From: Mohamed Abbas Date: Fri, 4 Apr 2008 16:59:58 -0700 Subject: mac80211: notify mac from low level driver (iwlwifi) Add new API to MAC80211 to allow low level driver to notify MAC with driver status. Signed-off-by: Mohamed Abbas Signed-off-by: Reinette Chatre Acked-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl3945-base.c | 1 + drivers/net/wireless/iwlwifi/iwl4965-base.c | 1 + include/net/mac80211.h | 19 +++++++++++++++++++ net/mac80211/ieee80211_sta.c | 23 +++++++++++++++++++++++ 4 files changed, 44 insertions(+) (limited to 'include') diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c index 5e51cfcda39f..29a9ecdcbf35 100644 --- a/drivers/net/wireless/iwlwifi/iwl3945-base.c +++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c @@ -5886,6 +5886,7 @@ static void iwl3945_alive_start(struct iwl3945_priv *priv) if (priv->error_recovering) iwl3945_error_recovery(priv); + ieee80211_notify_mac(priv->hw, IEEE80211_NOTIFY_RE_ASSOC); return; restart: diff --git a/drivers/net/wireless/iwlwifi/iwl4965-base.c b/drivers/net/wireless/iwlwifi/iwl4965-base.c index b043871d53f1..06e44dad5f02 100644 --- a/drivers/net/wireless/iwlwifi/iwl4965-base.c +++ b/drivers/net/wireless/iwlwifi/iwl4965-base.c @@ -5714,6 +5714,7 @@ static void iwl4965_alive_start(struct iwl_priv *priv) iwl4965_error_recovery(priv); iwlcore_low_level_notify(priv, IWLCORE_START_EVT); + ieee80211_notify_mac(priv->hw, IEEE80211_NOTIFY_RE_ASSOC); return; restart: diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 999f970da6ba..079e7bd86c90 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -73,6 +73,14 @@ * not do so then mac80211 may add this under certain circumstances. */ +/** + * enum ieee80211_notification_type - Low level driver notification + * @IEEE80211_NOTIFY_RE_ASSOC: start the re-association sequence + */ +enum ieee80211_notification_types { + IEEE80211_NOTIFY_RE_ASSOC, +}; + /** * struct ieee80211_ht_bss_info - describing BSS's HT characteristics * @@ -1678,4 +1686,15 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid); void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra, u16 tid); +/** + * ieee80211_notify_mac - low level driver notification + * @hw: pointer as obtained from ieee80211_alloc_hw(). + * @notification_types: enum ieee80211_notification_types + * + * This function must be called by low level driver to inform mac80211 of + * low level driver status change or force mac80211 to re-assoc for low + * level driver internal error that require re-assoc. + */ +void ieee80211_notify_mac(struct ieee80211_hw *hw, + enum ieee80211_notification_types notif_type); #endif /* MAC80211_H */ diff --git a/net/mac80211/ieee80211_sta.c b/net/mac80211/ieee80211_sta.c index 9e30333aa81e..89481c919cb6 100644 --- a/net/mac80211/ieee80211_sta.c +++ b/net/mac80211/ieee80211_sta.c @@ -4225,3 +4225,26 @@ int ieee80211_sta_disassociate(struct net_device *dev, u16 reason) ieee80211_set_disassoc(dev, ifsta, 0); return 0; } + +void ieee80211_notify_mac(struct ieee80211_hw *hw, + enum ieee80211_notification_types notif_type) +{ + struct ieee80211_local *local = hw_to_local(hw); + struct ieee80211_sub_if_data *sdata; + + switch (notif_type) { + case IEEE80211_NOTIFY_RE_ASSOC: + rcu_read_lock(); + list_for_each_entry_rcu(sdata, &local->interfaces, list) { + + if (sdata->vif.type == IEEE80211_IF_TYPE_STA) { + ieee80211_sta_req_auth(sdata->dev, + &sdata->u.sta); + } + + } + rcu_read_unlock(); + break; + } +} +EXPORT_SYMBOL(ieee80211_notify_mac); -- cgit v1.2.3 From b715631fad3ed320b85d386a84a6fb0b3f86b0b9 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 10 Apr 2008 01:33:47 -0700 Subject: socket: sk_filter minor cleanups Some minor style cleanups: * Move __KERNEL__ definitions to one place in filter.h * Use const for sk_filter_len * Line wrapping * Put EXPORT_SYMBOL next to function definition Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/filter.h | 31 +++++++++++++++---------------- net/core/filter.c | 5 ++--- 2 files changed, 17 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/linux/filter.h b/include/linux/filter.h index ddfa0372a3b7..bfc5d319b946 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -37,21 +37,6 @@ struct sock_fprog /* Required for SO_ATTACH_FILTER. */ struct sock_filter __user *filter; }; -#ifdef __KERNEL__ -struct sk_filter -{ - atomic_t refcnt; - unsigned int len; /* Number of filter blocks */ - struct rcu_head rcu; - struct sock_filter insns[0]; -}; - -static inline unsigned int sk_filter_len(struct sk_filter *fp) -{ - return fp->len*sizeof(struct sock_filter) + sizeof(*fp); -} -#endif - /* * Instruction classes */ @@ -141,10 +126,24 @@ static inline unsigned int sk_filter_len(struct sk_filter *fp) #define SKF_LL_OFF (-0x200000) #ifdef __KERNEL__ +struct sk_filter +{ + atomic_t refcnt; + unsigned int len; /* Number of filter blocks */ + struct rcu_head rcu; + struct sock_filter insns[0]; +}; + +static inline unsigned int sk_filter_len(const struct sk_filter *fp) +{ + return fp->len * sizeof(struct sock_filter) + sizeof(*fp); +} + struct sk_buff; struct sock; -extern unsigned int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int flen); +extern unsigned int sk_run_filter(struct sk_buff *skb, + struct sock_filter *filter, int flen); extern int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk); extern int sk_detach_filter(struct sock *sk); extern int sk_chk_filter(struct sock_filter *filter, int flen); diff --git a/net/core/filter.c b/net/core/filter.c index e0a06942c025..85a5febab567 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -275,6 +275,7 @@ load_b: return 0; } +EXPORT_SYMBOL(sk_run_filter); /** * sk_chk_filter - verify socket filter code @@ -385,6 +386,7 @@ int sk_chk_filter(struct sock_filter *filter, int flen) return (BPF_CLASS(filter[flen - 1].code) == BPF_RET) ? 0 : -EINVAL; } +EXPORT_SYMBOL(sk_chk_filter); /** * sk_filter_rcu_release: Release a socket filter by rcu_head @@ -467,6 +469,3 @@ int sk_detach_filter(struct sock *sk) rcu_read_unlock_bh(); return ret; } - -EXPORT_SYMBOL(sk_chk_filter); -EXPORT_SYMBOL(sk_run_filter); -- cgit v1.2.3 From 43db6d65e0ef943a361cb91f8baa49132009227b Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Thu, 10 Apr 2008 01:43:09 -0700 Subject: socket: sk_filter deinline The sk_filter function is too big to be inlined. This saves 2296 bytes of text on allyesconfig. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- include/linux/filter.h | 1 + include/net/sock.h | 35 ----------------------------------- net/core/filter.c | 35 +++++++++++++++++++++++++++++++++++ 3 files changed, 36 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/include/linux/filter.h b/include/linux/filter.h index bfc5d319b946..673e5677ebcc 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -142,6 +142,7 @@ static inline unsigned int sk_filter_len(const struct sk_filter *fp) struct sk_buff; struct sock; +extern int sk_filter(struct sock *sk, struct sk_buff *skb); extern unsigned int sk_run_filter(struct sk_buff *skb, struct sock_filter *filter, int flen); extern int sk_attach_filter(struct sock_fprog *fprog, struct sock *sk); diff --git a/include/net/sock.h b/include/net/sock.h index f4fdd101c9a2..09255eae93e9 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -927,41 +927,6 @@ extern void sk_common_release(struct sock *sk); /* Initialise core socket variables */ extern void sock_init_data(struct socket *sock, struct sock *sk); -/** - * sk_filter - run a packet through a socket filter - * @sk: sock associated with &sk_buff - * @skb: buffer to filter - * @needlock: set to 1 if the sock is not locked by caller. - * - * Run the filter code and then cut skb->data to correct size returned by - * sk_run_filter. If pkt_len is 0 we toss packet. If skb->len is smaller - * than pkt_len we keep whole skb->data. This is the socket level - * wrapper to sk_run_filter. It returns 0 if the packet should - * be accepted or -EPERM if the packet should be tossed. - * - */ - -static inline int sk_filter(struct sock *sk, struct sk_buff *skb) -{ - int err; - struct sk_filter *filter; - - err = security_sock_rcv_skb(sk, skb); - if (err) - return err; - - rcu_read_lock_bh(); - filter = rcu_dereference(sk->sk_filter); - if (filter) { - unsigned int pkt_len = sk_run_filter(skb, filter->insns, - filter->len); - err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM; - } - rcu_read_unlock_bh(); - - return err; -} - /** * sk_filter_release: Release a socket filter * @sk: socket diff --git a/net/core/filter.c b/net/core/filter.c index 85a5febab567..bbb53c69857c 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -63,6 +63,41 @@ static inline void *load_pointer(struct sk_buff *skb, int k, } } +/** + * sk_filter - run a packet through a socket filter + * @sk: sock associated with &sk_buff + * @skb: buffer to filter + * @needlock: set to 1 if the sock is not locked by caller. + * + * Run the filter code and then cut skb->data to correct size returned by + * sk_run_filter. If pkt_len is 0 we toss packet. If skb->len is smaller + * than pkt_len we keep whole skb->data. This is the socket level + * wrapper to sk_run_filter. It returns 0 if the packet should + * be accepted or -EPERM if the packet should be tossed. + * + */ +int sk_filter(struct sock *sk, struct sk_buff *skb) +{ + int err; + struct sk_filter *filter; + + err = security_sock_rcv_skb(sk, skb); + if (err) + return err; + + rcu_read_lock_bh(); + filter = rcu_dereference(sk->sk_filter); + if (filter) { + unsigned int pkt_len = sk_run_filter(skb, filter->insns, + filter->len); + err = pkt_len ? pskb_trim(skb, pkt_len) : -EPERM; + } + rcu_read_unlock_bh(); + + return err; +} +EXPORT_SYMBOL(sk_filter); + /** * sk_run_filter - run a filter on a socket * @skb: buffer to run the filter on -- cgit v1.2.3 From 3cccd6078413e9707f0ef3652b4e6e9cb84e9fa0 Mon Sep 17 00:00:00 2001 From: Rami Rosen Date: Thu, 10 Apr 2008 02:01:21 -0700 Subject: [IPV6] Remove three method declarations in include/net/ndisc.h. This patch removes two unused method declarations in include/net/ndisc.h: ndisc_forwarding_on(void) and ndisc_forwarding_off(void); Also igmp6_cleanup(void) appears twice in this header, so one igmp6_cleanup(void) declaration is removed. Signed-off-by: Rami Rosen Signed-off-by: David S. Miller --- include/net/ndisc.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include') diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 9f2bae68d28c..16424236fe2f 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -102,9 +102,6 @@ extern void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr, struct in6_addr *daddr); -extern void ndisc_forwarding_on(void); -extern void ndisc_forwarding_off(void); - extern void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, struct in6_addr *target); @@ -124,7 +121,6 @@ extern int igmp6_event_query(struct sk_buff *skb); extern int igmp6_event_report(struct sk_buff *skb); -extern void igmp6_cleanup(void); #ifdef CONFIG_SYSCTL extern int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, -- cgit v1.2.3 From 4738c1db1593687713869fa69e733eebc7b0d6d8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 10 Apr 2008 02:02:28 -0700 Subject: [SKFILTER]: Add SKF_ADF_NLATTR instruction SKF_ADF_NLATTR searches for a netlink attribute, which avoids manually parsing and walking attributes. It takes the offset at which to start searching in the 'A' register and the attribute type in the 'X' register and returns the offset in the 'A' register. When the attribute is not found it returns zero. A top-level attribute can be located using a filter like this (example for nfnetlink, using struct nfgenmsg): ... { /* A = offset of first attribute */ .code = BPF_LD | BPF_IMM, .k = sizeof(struct nlmsghdr) + sizeof(struct nfgenmsg) }, { /* X = CTA_PROTOINFO */ .code = BPF_LDX | BPF_IMM, .k = CTA_PROTOINFO, }, { /* A = netlink attribute offset */ .code = BPF_LD | BPF_B | BPF_ABS, .k = SKF_AD_OFF + SKF_AD_NLATTR }, { /* Exit if not found */ .code = BPF_JMP | BPF_JEQ | BPF_K, .k = 0, .jt = }, ... A nested attribute below the CTA_PROTOINFO attribute would then be parsed like this: ... { /* A += sizeof(struct nlattr) */ .code = BPF_ALU | BPF_ADD | BPF_K, .k = sizeof(struct nlattr), }, { /* X = CTA_PROTOINFO_TCP */ .code = BPF_LDX | BPF_IMM, .k = CTA_PROTOINFO_TCP, }, { /* A = netlink attribute offset */ .code = BPF_LD | BPF_B | BPF_ABS, .k = SKF_AD_OFF + SKF_AD_NLATTR }, ... The data of an attribute can be loaded into 'A' like this: ... { /* X = A (attribute offset) */ .code = BPF_MISC | BPF_TAX, }, { /* A = skb->data[X + k] */ .code = BPF_LD | BPF_B | BPF_IND, .k = sizeof(struct nlattr), }, ... Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/filter.h | 3 ++- net/core/filter.c | 17 +++++++++++++++++ 2 files changed, 19 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/filter.h b/include/linux/filter.h index 673e5677ebcc..b6ea9aa9e853 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -121,7 +121,8 @@ struct sock_fprog /* Required for SO_ATTACH_FILTER. */ #define SKF_AD_PROTOCOL 0 #define SKF_AD_PKTTYPE 4 #define SKF_AD_IFINDEX 8 -#define SKF_AD_MAX 12 +#define SKF_AD_NLATTR 12 +#define SKF_AD_MAX 16 #define SKF_NET_OFF (-0x100000) #define SKF_LL_OFF (-0x200000) diff --git a/net/core/filter.c b/net/core/filter.c index bbb53c69857c..f5f3cf603064 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -303,6 +304,22 @@ load_b: case SKF_AD_IFINDEX: A = skb->dev->ifindex; continue; + case SKF_AD_NLATTR: { + struct nlattr *nla; + + if (skb_is_nonlinear(skb)) + return 0; + if (A > skb->len - sizeof(struct nlattr)) + return 0; + + nla = nla_find((struct nlattr *)&skb->data[A], + skb->len - A, X); + if (nla) + A = (void *)nla - (void *)skb->data; + else + A = 0; + continue; + } default: return 0; } -- cgit v1.2.3 From 5c06f510a25153ff79e8c2dca312b732a367c5bb Mon Sep 17 00:00:00 2001 From: Rami Rosen Date: Thu, 10 Apr 2008 02:31:20 -0700 Subject: [IPV6]: Remove unused declarations in include/net/ip6_route.h. 1) Standlaone ip6_null_entry is no longer needed as it is replaced by the ip6_null_entry member of ipv6 (instance of struct netns_ipv6) in struct net (as a result of Network Namespaces patches). 2) These 3 methods from this same header are not defined anywhere: ip6_rt_addr_add(), ip6_rt_addr_del(), rt6_sndmsg() Signed-off-by: Rami Rosen Signed-off-by: David S. Miller --- include/net/ip6_route.h | 15 --------------- 1 file changed, 15 deletions(-) (limited to 'include') diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 3ae6799c2b14..9080076ce0e5 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -37,7 +37,6 @@ struct route_info { #define RT6_LOOKUP_F_SRCPREF_PUBLIC 0x00000010 #define RT6_LOOKUP_F_SRCPREF_COA 0x00000020 -extern struct rt6_info *ip6_null_entry; #ifdef CONFIG_IPV6_MULTIPLE_TABLES extern struct rt6_info *ip6_prohibit_entry; @@ -61,20 +60,6 @@ extern int ip6_route_add(struct fib6_config *cfg); extern int ip6_ins_rt(struct rt6_info *); extern int ip6_del_rt(struct rt6_info *); -extern int ip6_rt_addr_add(struct in6_addr *addr, - struct net_device *dev, - int anycast); - -extern int ip6_rt_addr_del(struct in6_addr *addr, - struct net_device *dev); - -extern void rt6_sndmsg(int type, struct in6_addr *dst, - struct in6_addr *src, - struct in6_addr *gw, - struct net_device *dev, - int dstlen, int srclen, - int metric, __u32 flags); - extern struct rt6_info *rt6_lookup(struct net *net, struct in6_addr *daddr, struct in6_addr *saddr, -- cgit v1.2.3 From 4dfc2817025965a2fc78a18c50f540736a6b5c24 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 10 Apr 2008 03:12:40 -0700 Subject: [Syncookies]: Add support for TCP options via timestamps. Allow the use of SACK and window scaling when syncookies are used and the client supports tcp timestamps. Options are encoded into the timestamp sent in the syn-ack and restored from the timestamp echo when the ack is received. Based on earlier work by Glenn Griffin. This patch avoids increasing the size of structs by encoding TCP options into the least significant bits of the timestamp and by not using any 'timestamp offset'. The downside is that the timestamp sent in the packet after the synack will increase by several seconds. changes since v1: don't duplicate timestamp echo decoding function, put it into ipv4/syncookie.c and have ipv6/syncookies.c use it. Feedback from Glenn Griffin: fix line indented with spaces, kill redundant if () Reviewed-by: Hagen Paul Pfeifer Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/net/request_sock.h | 2 +- include/net/tcp.h | 4 +++ net/ipv4/syncookies.c | 89 +++++++++++++++++++++++++++++++++++++++++++--- net/ipv4/tcp_ipv4.c | 5 ++- net/ipv4/tcp_output.c | 6 +++- net/ipv6/syncookies.c | 20 ++++++++--- net/ipv6/tcp_ipv6.c | 5 ++- 7 files changed, 114 insertions(+), 17 deletions(-) (limited to 'include') diff --git a/include/net/request_sock.h b/include/net/request_sock.h index 0369f98e9f3a..b220b5f624de 100644 --- a/include/net/request_sock.h +++ b/include/net/request_sock.h @@ -45,7 +45,7 @@ struct request_sock { struct request_sock *dl_next; /* Must be first member! */ u16 mss; u8 retrans; - u8 __pad; + u8 cookie_ts; /* syncookie: encode tcpopts in timestamp */ /* The following two fields can be easily recomputed I think -AK */ u32 window_clamp; /* window clamp at creation time */ u32 rcv_wnd; /* rcv_wnd offered first time */ diff --git a/include/net/tcp.h b/include/net/tcp.h index 723b36851dde..7b41bb962b9c 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -442,6 +442,9 @@ extern struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, extern __u32 cookie_v4_init_sequence(struct sock *sk, struct sk_buff *skb, __u16 *mss); +extern __u32 cookie_init_timestamp(struct request_sock *req); +extern void cookie_check_timestamp(struct tcp_options_received *tcp_opt); + /* From net/ipv6/syncookies.c */ extern struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb); extern __u32 cookie_v6_init_sequence(struct sock *sk, struct sk_buff *skb, @@ -956,6 +959,7 @@ static inline void tcp_openreq_init(struct request_sock *req, struct inet_request_sock *ireq = inet_rsk(req); req->rcv_wnd = 0; /* So that tcp_send_synack() knows! */ + req->cookie_ts = 0; tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq; req->mss = rx_opt->mss_clamp; req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index abc752d45cf7..73ba98921d64 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -19,6 +19,10 @@ #include #include +/* Timestamps: lowest 9 bits store TCP options */ +#define TSBITS 9 +#define TSMASK (((__u32)1 << TSBITS) - 1) + extern int sysctl_tcp_syncookies; __u32 syncookie_secret[2][16-4+SHA_DIGEST_WORDS]; @@ -51,6 +55,39 @@ static u32 cookie_hash(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport, return tmp[17]; } + +/* + * when syncookies are in effect and tcp timestamps are enabled we encode + * tcp options in the lowest 9 bits of the timestamp value that will be + * sent in the syn-ack. + * Since subsequent timestamps use the normal tcp_time_stamp value, we + * must make sure that the resulting initial timestamp is <= tcp_time_stamp. + */ +__u32 cookie_init_timestamp(struct request_sock *req) +{ + struct inet_request_sock *ireq; + u32 ts, ts_now = tcp_time_stamp; + u32 options = 0; + + ireq = inet_rsk(req); + if (ireq->wscale_ok) { + options = ireq->snd_wscale; + options |= ireq->rcv_wscale << 4; + } + options |= ireq->sack_ok << 8; + + ts = ts_now & ~TSMASK; + ts |= options; + if (ts > ts_now) { + ts >>= TSBITS; + ts--; + ts <<= TSBITS; + ts |= options; + } + return ts; +} + + static __u32 secure_tcp_syn_cookie(__be32 saddr, __be32 daddr, __be16 sport, __be16 dport, __u32 sseq, __u32 count, __u32 data) @@ -185,6 +222,35 @@ static inline struct sock *get_cookie_sock(struct sock *sk, struct sk_buff *skb, return child; } + +/* + * when syncookies are in effect and tcp timestamps are enabled we stored + * additional tcp options in the timestamp. + * This extracts these options from the timestamp echo. + * + * The lowest 4 bits are for snd_wscale + * The next 4 lsb are for rcv_wscale + * The next lsb is for sack_ok + */ +void cookie_check_timestamp(struct tcp_options_received *tcp_opt) +{ + /* echoed timestamp, 9 lowest bits contain options */ + u32 options = tcp_opt->rcv_tsecr & TSMASK; + + tcp_opt->snd_wscale = options & 0xf; + options >>= 4; + tcp_opt->rcv_wscale = options & 0xf; + + tcp_opt->sack_ok = (options >> 4) & 0x1; + + if (tcp_opt->sack_ok) + tcp_sack_reset(tcp_opt); + + if (tcp_opt->snd_wscale || tcp_opt->rcv_wscale) + tcp_opt->wscale_ok = 1; +} +EXPORT_SYMBOL(cookie_check_timestamp); + struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, struct ip_options *opt) { @@ -198,6 +264,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, int mss; struct rtable *rt; __u8 rcv_wscale; + struct tcp_options_received tcp_opt; if (!sysctl_tcp_syncookies || !th->ack) goto out; @@ -210,6 +277,13 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); + /* check for timestamp cookie support */ + memset(&tcp_opt, 0, sizeof(tcp_opt)); + tcp_parse_options(skb, &tcp_opt, 0); + + if (tcp_opt.saw_tstamp) + cookie_check_timestamp(&tcp_opt); + ret = NULL; req = reqsk_alloc(&tcp_request_sock_ops); /* for safety */ if (!req) @@ -228,6 +302,12 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, ireq->loc_addr = ip_hdr(skb)->daddr; ireq->rmt_addr = ip_hdr(skb)->saddr; ireq->opt = NULL; + ireq->snd_wscale = tcp_opt.snd_wscale; + ireq->rcv_wscale = tcp_opt.rcv_wscale; + ireq->sack_ok = tcp_opt.sack_ok; + ireq->wscale_ok = tcp_opt.wscale_ok; + ireq->tstamp_ok = tcp_opt.saw_tstamp; + req->ts_recent = tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0; /* We throwed the options of the initial SYN away, so we hope * the ACK carries the same options again (see RFC1122 4.2.3.8) @@ -242,8 +322,6 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, } } - ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0; - ireq->wscale_ok = ireq->sack_ok = 0; req->expires = 0UL; req->retrans = 0; @@ -272,11 +350,12 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, } /* Try to redo what tcp_v4_send_synack did. */ - req->window_clamp = dst_metric(&rt->u.dst, RTAX_WINDOW); + req->window_clamp = tp->window_clamp ? :dst_metric(&rt->u.dst, RTAX_WINDOW); + tcp_select_initial_window(tcp_full_space(sk), req->mss, &req->rcv_wnd, &req->window_clamp, - 0, &rcv_wscale); - /* BTW win scale with syncookies is 0 by definition */ + ireq->wscale_ok, &rcv_wscale); + ireq->rcv_wscale = rcv_wscale; ret = get_cookie_sock(sk, skb, req, &rt->u.dst); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index df89a566a5a1..52e3ae603ca9 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1299,10 +1299,8 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) tcp_parse_options(skb, &tmp_opt, 0); - if (want_cookie) { + if (want_cookie && !tmp_opt.saw_tstamp) tcp_clear_options(&tmp_opt); - tmp_opt.saw_tstamp = 0; - } if (tmp_opt.saw_tstamp && !tmp_opt.rcv_tsval) { /* Some OSes (unknown ones, but I see them on web server, which @@ -1330,6 +1328,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) if (want_cookie) { #ifdef CONFIG_SYN_COOKIES syn_flood_warning(skb); + req->cookie_ts = tmp_opt.tstamp_ok; #endif isn = cookie_v4_init_sequence(sk, skb, &req->mss); } else if (!isn) { diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index a627616314ba..76b3653e9b4c 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2233,7 +2233,11 @@ struct sk_buff *tcp_make_synack(struct sock *sk, struct dst_entry *dst, /* RFC1323: The window in SYN & SYN/ACK segments is never scaled. */ th->window = htons(min(req->rcv_wnd, 65535U)); - +#ifdef CONFIG_SYN_COOKIES + if (unlikely(req->cookie_ts)) + TCP_SKB_CB(skb)->when = cookie_init_timestamp(req); + else +#endif TCP_SKB_CB(skb)->when = tcp_time_stamp; tcp_syn_build_options((__be32 *)(th + 1), dst_metric(dst, RTAX_ADVMSS), ireq->tstamp_ok, ireq->sack_ok, ireq->wscale_ok, ireq->rcv_wscale, diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index 3a622e7abc02..938ce4ecde55 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -170,6 +170,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) int mss; struct dst_entry *dst; __u8 rcv_wscale; + struct tcp_options_received tcp_opt; if (!sysctl_tcp_syncookies || !th->ack) goto out; @@ -182,6 +183,13 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) NET_INC_STATS_BH(LINUX_MIB_SYNCOOKIESRECV); + /* check for timestamp cookie support */ + memset(&tcp_opt, 0, sizeof(tcp_opt)); + tcp_parse_options(skb, &tcp_opt, 0); + + if (tcp_opt.saw_tstamp) + cookie_check_timestamp(&tcp_opt); + ret = NULL; req = inet6_reqsk_alloc(&tcp6_request_sock_ops); if (!req) @@ -216,8 +224,12 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) req->expires = 0UL; req->retrans = 0; - ireq->snd_wscale = ireq->rcv_wscale = ireq->tstamp_ok = 0; - ireq->wscale_ok = ireq->sack_ok = 0; + ireq->snd_wscale = tcp_opt.snd_wscale; + ireq->rcv_wscale = tcp_opt.rcv_wscale; + ireq->sack_ok = tcp_opt.sack_ok; + ireq->wscale_ok = tcp_opt.wscale_ok; + ireq->tstamp_ok = tcp_opt.saw_tstamp; + req->ts_recent = tcp_opt.saw_tstamp ? tcp_opt.rcv_tsval : 0; treq->rcv_isn = ntohl(th->seq) - 1; treq->snt_isn = cookie; @@ -253,10 +265,10 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) goto out; } - req->window_clamp = dst_metric(dst, RTAX_WINDOW); + req->window_clamp = tp->window_clamp ? :dst_metric(dst, RTAX_WINDOW); tcp_select_initial_window(tcp_full_space(sk), req->mss, &req->rcv_wnd, &req->window_clamp, - 0, &rcv_wscale); + ireq->wscale_ok, &rcv_wscale); ireq->rcv_wscale = rcv_wscale; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 378cc4002a76..8ebf6de29562 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1290,10 +1290,8 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) tcp_parse_options(skb, &tmp_opt, 0); - if (want_cookie) { + if (want_cookie && !tmp_opt.saw_tstamp) tcp_clear_options(&tmp_opt); - tmp_opt.saw_tstamp = 0; - } tmp_opt.tstamp_ok = tmp_opt.saw_tstamp; tcp_openreq_init(req, &tmp_opt, skb); @@ -1307,6 +1305,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) if (want_cookie) { isn = cookie_v6_init_sequence(sk, skb, &req->mss); + req->cookie_ts = tmp_opt.tstamp_ok; } else if (!isn) { if (ipv6_opt_accepted(sk, skb) || np->rxopt.bits.rxinfo || np->rxopt.bits.rxoinfo || -- cgit v1.2.3 From fed85383ac34d82e96f227ce49ce68117cec23a0 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Fri, 11 Apr 2008 20:17:55 +0900 Subject: [IPV6]: Use XOR and OR rather than mutiple ands for ipv6 address comparisons. ipv6_addr_equal(), ipv6_addr_v4mapped(), ipv6_addr_is_ll_all_{nodes,routers}(), ipv6_masked_addr_cmp() Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 14 ++++++-------- include/net/ipv6.h | 22 ++++++++++------------ 2 files changed, 16 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index bdcc863a60a4..1dc9d03372d9 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -240,18 +240,16 @@ static inline int ipv6_addr_is_multicast(const struct in6_addr *addr) static inline int ipv6_addr_is_ll_all_nodes(const struct in6_addr *addr) { - return (addr->s6_addr32[0] == htonl(0xff020000) && - addr->s6_addr32[1] == 0 && - addr->s6_addr32[2] == 0 && - addr->s6_addr32[3] == htonl(0x00000001)); + return (((addr->s6_addr32[0] ^ htonl(0xff020000)) | + addr->s6_addr32[1] | addr->s6_addr32[2] | + (addr->s6_addr32[3] ^ htonl(0x00000001))) == 0); } static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *addr) { - return (addr->s6_addr32[0] == htonl(0xff020000) && - addr->s6_addr32[1] == 0 && - addr->s6_addr32[2] == 0 && - addr->s6_addr32[3] == htonl(0x00000002)); + return (((addr->s6_addr32[0] ^ htonl(0xff020000)) | + addr->s6_addr32[1] | addr->s6_addr32[2] | + (addr->s6_addr32[3] ^ htonl(0x00000002))) == 0); } static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr) diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 5738c1c73ac1..a0c285b6311e 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -280,12 +280,10 @@ static inline int ipv6_masked_addr_cmp(const struct in6_addr *a1, const struct in6_addr *m, const struct in6_addr *a2) { - unsigned int i; - - for (i = 0; i < 4; i++) - if ((a1->s6_addr32[i] ^ a2->s6_addr32[i]) & m->s6_addr32[i]) - return 1; - return 0; + return (!!(((a1->s6_addr32[0] ^ a2->s6_addr32[0]) & m->s6_addr32[0]) | + ((a1->s6_addr32[1] ^ a2->s6_addr32[1]) & m->s6_addr32[1]) | + ((a1->s6_addr32[2] ^ a2->s6_addr32[2]) & m->s6_addr32[2]) | + ((a1->s6_addr32[3] ^ a2->s6_addr32[3]) & m->s6_addr32[3]))); } static inline void ipv6_addr_copy(struct in6_addr *a1, const struct in6_addr *a2) @@ -320,10 +318,10 @@ static inline void ipv6_addr_set(struct in6_addr *addr, static inline int ipv6_addr_equal(const struct in6_addr *a1, const struct in6_addr *a2) { - return (a1->s6_addr32[0] == a2->s6_addr32[0] && - a1->s6_addr32[1] == a2->s6_addr32[1] && - a1->s6_addr32[2] == a2->s6_addr32[2] && - a1->s6_addr32[3] == a2->s6_addr32[3]); + return (((a1->s6_addr32[0] ^ a2->s6_addr32[0]) | + (a1->s6_addr32[1] ^ a2->s6_addr32[1]) | + (a1->s6_addr32[2] ^ a2->s6_addr32[2]) | + (a1->s6_addr32[3] ^ a2->s6_addr32[3])) == 0); } static inline int __ipv6_prefix_equal(const __be32 *a1, const __be32 *a2, @@ -371,8 +369,8 @@ static inline int ipv6_addr_any(const struct in6_addr *a) static inline int ipv6_addr_v4mapped(const struct in6_addr *a) { - return ((a->s6_addr32[0] | a->s6_addr32[1]) == 0 && - a->s6_addr32[2] == htonl(0x0000ffff)); + return ((a->s6_addr32[0] | a->s6_addr32[1] | + (a->s6_addr32[2] ^ htonl(0x0000ffff))) == 0); } /* -- cgit v1.2.3 From 3eb84f49290461e2b83d6e8ee1f3f0e504340c8b Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 10 Apr 2008 15:42:08 +0900 Subject: [IPV6] ADDRCONF: Uninline ipv6_addr_hash(). The function is only used in net/ipv6/addrconf.c. Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 19 ------------------- net/ipv6/addrconf.c | 19 +++++++++++++++++++ 2 files changed, 19 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 1dc9d03372d9..1ba4e5bf5801 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -189,25 +189,6 @@ static inline void in6_ifa_put(struct inet6_ifaddr *ifp) #define in6_ifa_hold(ifp) atomic_inc(&(ifp)->refcnt) -/* - * Hash function taken from net_alias.c - */ - -static __inline__ u8 ipv6_addr_hash(const struct in6_addr *addr) -{ - __u32 word; - - /* - * We perform the hash function over the last 64 bits of the address - * This will include the IEEE address token on links that support it. - */ - - word = (__force u32)(addr->s6_addr32[2] ^ addr->s6_addr32[3]); - word ^= (word >> 16); - word ^= (word >> 8); - - return ((word ^ (word >> 4)) & 0x0f); -} /* * compute link-local solicited-node multicast address diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index ed9938014cf9..b17fafc56241 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -541,6 +541,25 @@ ipv6_link_dev_addr(struct inet6_dev *idev, struct inet6_ifaddr *ifp) *ifap = ifp; } +/* + * Hash function taken from net_alias.c + */ +static u8 ipv6_addr_hash(const struct in6_addr *addr) +{ + __u32 word; + + /* + * We perform the hash function over the last 64 bits of the address + * This will include the IEEE address token on links that support it. + */ + + word = (__force u32)(addr->s6_addr32[2] ^ addr->s6_addr32[3]); + word ^= (word >> 16); + word ^= (word >> 8); + + return ((word ^ (word >> 4)) & 0x0f); +} + /* On success it returns ifp with increased reference count */ static struct inet6_ifaddr * -- cgit v1.2.3 From dfd982baff01c18e3e1717c97fdac79c28f105ce Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 10 Apr 2008 15:42:09 +0900 Subject: [IPV6] ADDRCONF: Uninline ipv6_isatap_eui64(). Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 15 +-------------- net/ipv6/addrconf.c | 26 ++++++++++++++++++++++++-- 2 files changed, 25 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 1ba4e5bf5801..8317c1bcd86c 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -233,20 +233,7 @@ static inline int ipv6_addr_is_ll_all_routers(const struct in6_addr *addr) (addr->s6_addr32[3] ^ htonl(0x00000002))) == 0); } -static inline int ipv6_isatap_eui64(u8 *eui, __be32 addr) -{ - eui[0] = (ipv4_is_zeronet(addr) || ipv4_is_private_10(addr) || - ipv4_is_loopback(addr) || ipv4_is_linklocal_169(addr) || - ipv4_is_private_172(addr) || ipv4_is_test_192(addr) || - ipv4_is_anycast_6to4(addr) || ipv4_is_private_192(addr) || - ipv4_is_test_198(addr) || ipv4_is_multicast(addr) || - ipv4_is_lbcast(addr)) ? 0x00 : 0x02; - eui[1] = 0; - eui[2] = 0x5E; - eui[3] = 0xFE; - memcpy (eui+4, &addr, 4); - return 0; -} +extern int __ipv6_isatap_ifid(u8 *eui, __be32 addr); static inline int ipv6_addr_is_isatap(const struct in6_addr *addr) { diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index b17fafc56241..d15f3e095e7e 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1494,6 +1494,29 @@ static int addrconf_ifid_infiniband(u8 *eui, struct net_device *dev) return 0; } +int __ipv6_isatap_ifid(u8 *eui, __be32 addr) +{ + eui[0] = (ipv4_is_zeronet(addr) || ipv4_is_private_10(addr) || + ipv4_is_loopback(addr) || ipv4_is_linklocal_169(addr) || + ipv4_is_private_172(addr) || ipv4_is_test_192(addr) || + ipv4_is_anycast_6to4(addr) || ipv4_is_private_192(addr) || + ipv4_is_test_198(addr) || ipv4_is_multicast(addr) || + ipv4_is_lbcast(addr)) ? 0x00 : 0x02; + eui[1] = 0; + eui[2] = 0x5E; + eui[3] = 0xFE; + memcpy(eui + 4, &addr, 4); + return 0; +} +EXPORT_SYMBOL(__ipv6_isatap_ifid); + +static int addrconf_ifid_sit(u8 *eui, struct net_device *dev) +{ + if (dev->priv_flags & IFF_ISATAP) + return __ipv6_isatap_ifid(eui, *(__be32 *)dev->dev_addr); + return -1; +} + static int ipv6_generate_eui64(u8 *eui, struct net_device *dev) { switch (dev->type) { @@ -1506,8 +1529,7 @@ static int ipv6_generate_eui64(u8 *eui, struct net_device *dev) case ARPHRD_INFINIBAND: return addrconf_ifid_infiniband(eui, dev); case ARPHRD_SIT: - if (dev->priv_flags & IFF_ISATAP) - return ipv6_isatap_eui64(eui, *(__be32 *)dev->dev_addr); + return addrconf_ifid_sit(eui, dev); } return -1; } -- cgit v1.2.3 From 9acd9f3ae92d0dc0ca7504fb48c1040e8bbc39fe Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 10 Apr 2008 15:42:10 +0900 Subject: [IPV6]: Make address arguments const. - net/ipv6/addrconf.c: ipv6_get_ifaddr(), ipv6_dev_get_saddr() - net/ipv6/mcast.c: ipv6_sock_mc_join(), ipv6_sock_mc_drop(), inet6_mc_check(), ipv6_dev_mc_inc(), __ipv6_dev_mc_dec(), ipv6_dev_mc_dec(), ipv6_chk_mcast_addr() - net/ipv6/route.c: rt6_lookup(), icmp6_dst_alloc() - net/ipv6/ip6_output.c: ip6_nd_hdr() - net/ipv6/ndisc.c: ndisc_send_ns(), ndisc_send_rs(), ndisc_send_redirect(), ndisc_get_neigh(), __ndisc_send() Signed-off-by: YOSHIFUJI Hideaki --- include/net/addrconf.h | 28 +++++++++++++++------------- include/net/ip6_route.h | 6 +++--- include/net/ipv6.h | 4 ++-- include/net/ndisc.h | 14 +++++++------- net/ipv6/addrconf.c | 6 +++--- net/ipv6/ip6_output.c | 2 +- net/ipv6/mcast.c | 20 +++++++++----------- net/ipv6/ndisc.c | 22 ++++++++++++---------- net/ipv6/route.c | 6 +++--- 9 files changed, 55 insertions(+), 53 deletions(-) (limited to 'include') diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 8317c1bcd86c..92af23d66eb9 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -76,12 +76,12 @@ extern int ipv6_chk_prefix(struct in6_addr *addr, struct net_device *dev); extern struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, - struct in6_addr *addr, + const struct in6_addr *addr, struct net_device *dev, int strict); extern int ipv6_dev_get_saddr(struct net_device *dev, - struct in6_addr *daddr, + const struct in6_addr *daddr, unsigned int srcprefs, struct in6_addr *saddr); extern int ipv6_get_lladdr(struct net_device *dev, @@ -105,25 +105,27 @@ extern u32 ipv6_addr_label(const struct in6_addr *addr, /* * multicast prototypes (mcast.c) */ -extern int ipv6_sock_mc_join(struct sock *sk, int ifindex, - struct in6_addr *addr); -extern int ipv6_sock_mc_drop(struct sock *sk, int ifindex, - struct in6_addr *addr); +extern int ipv6_sock_mc_join(struct sock *sk, int ifindex, + const struct in6_addr *addr); +extern int ipv6_sock_mc_drop(struct sock *sk, int ifindex, + const struct in6_addr *addr); extern void ipv6_sock_mc_close(struct sock *sk); -extern int inet6_mc_check(struct sock *sk, struct in6_addr *mc_addr, - struct in6_addr *src_addr); +extern int inet6_mc_check(struct sock *sk, + const struct in6_addr *mc_addr, + const struct in6_addr *src_addr); -extern int ipv6_dev_mc_inc(struct net_device *dev, struct in6_addr *addr); -extern int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr); -extern int ipv6_dev_mc_dec(struct net_device *dev, struct in6_addr *addr); +extern int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr); +extern int __ipv6_dev_mc_dec(struct inet6_dev *idev, const struct in6_addr *addr); +extern int ipv6_dev_mc_dec(struct net_device *dev, const struct in6_addr *addr); extern void ipv6_mc_up(struct inet6_dev *idev); extern void ipv6_mc_down(struct inet6_dev *idev); extern void ipv6_mc_init_dev(struct inet6_dev *idev); extern void ipv6_mc_destroy_dev(struct inet6_dev *idev); extern void addrconf_dad_failure(struct inet6_ifaddr *ifp); -extern int ipv6_chk_mcast_addr(struct net_device *dev, struct in6_addr *group, - struct in6_addr *src_addr); +extern int ipv6_chk_mcast_addr(struct net_device *dev, + const struct in6_addr *group, + const struct in6_addr *src_addr); extern int ipv6_is_mld(struct sk_buff *skb, int nexthdr); extern void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len); diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 9080076ce0e5..9313491e3dad 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -61,13 +61,13 @@ extern int ip6_ins_rt(struct rt6_info *); extern int ip6_del_rt(struct rt6_info *); extern struct rt6_info *rt6_lookup(struct net *net, - struct in6_addr *daddr, - struct in6_addr *saddr, + const struct in6_addr *daddr, + const struct in6_addr *saddr, int oif, int flags); extern struct dst_entry *icmp6_dst_alloc(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *addr); + const struct in6_addr *addr); extern int icmp6_dst_gc(int *more); extern void fib6_force_start_gc(struct net *net); diff --git a/include/net/ipv6.h b/include/net/ipv6.h index a0c285b6311e..49c48983019f 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -451,8 +451,8 @@ extern int ip6_xmit(struct sock *sk, extern int ip6_nd_hdr(struct sock *sk, struct sk_buff *skb, struct net_device *dev, - struct in6_addr *saddr, - struct in6_addr *daddr, + const struct in6_addr *saddr, + const struct in6_addr *daddr, int proto, int len); extern int ip6_find_1stfragopt(struct sk_buff *skb, u8 **nexthdr); diff --git a/include/net/ndisc.h b/include/net/ndisc.h index 16424236fe2f..9c451ff2f4f4 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -94,17 +94,17 @@ extern int ndisc_rcv(struct sk_buff *skb); extern void ndisc_send_ns(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *solicit, - struct in6_addr *daddr, - struct in6_addr *saddr); + const struct in6_addr *solicit, + const struct in6_addr *daddr, + const struct in6_addr *saddr); extern void ndisc_send_rs(struct net_device *dev, - struct in6_addr *saddr, - struct in6_addr *daddr); + const struct in6_addr *saddr, + const struct in6_addr *daddr); extern void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, - struct in6_addr *target); + const struct in6_addr *target); extern int ndisc_mc_map(struct in6_addr *addr, char *buf, struct net_device *dev, int dir); @@ -134,7 +134,7 @@ extern int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, extern void inet6_ifinfo_notify(int event, struct inet6_dev *idev); -static inline struct neighbour * ndisc_get_neigh(struct net_device *dev, struct in6_addr *addr) +static inline struct neighbour * ndisc_get_neigh(struct net_device *dev, const struct in6_addr *addr) { if (dev) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index d15f3e095e7e..4048c2b73b0b 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -940,7 +940,7 @@ struct ipv6_saddr_score { }; struct ipv6_saddr_dst { - struct in6_addr *addr; + const struct in6_addr *addr; int ifindex; int scope; int label; @@ -1074,7 +1074,7 @@ out: } int ipv6_dev_get_saddr(struct net_device *dst_dev, - struct in6_addr *daddr, unsigned int prefs, + const struct in6_addr *daddr, unsigned int prefs, struct in6_addr *saddr) { struct ipv6_saddr_score scores[2], @@ -1309,7 +1309,7 @@ int ipv6_chk_prefix(struct in6_addr *addr, struct net_device *dev) EXPORT_SYMBOL(ipv6_chk_prefix); -struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, struct in6_addr *addr, +struct inet6_ifaddr *ipv6_get_ifaddr(struct net *net, const struct in6_addr *addr, struct net_device *dev, int strict) { struct inet6_ifaddr * ifp; diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index c0dbe549cc42..0af2e055f883 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -286,7 +286,7 @@ EXPORT_SYMBOL(ip6_xmit); */ int ip6_nd_hdr(struct sock *sk, struct sk_buff *skb, struct net_device *dev, - struct in6_addr *saddr, struct in6_addr *daddr, + const struct in6_addr *saddr, const struct in6_addr *daddr, int proto, int len) { struct ipv6_pinfo *np = inet6_sk(sk); diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 2e6a53f3cc38..0a0132a1c443 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -127,8 +127,6 @@ static struct in6_addr mld2_all_mcr = MLD2_ALL_MCR_INIT; /* Big mc list lock for all the sockets */ static DEFINE_RWLOCK(ipv6_sk_mc_lock); -int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr); - static void igmp6_join_group(struct ifmcaddr6 *ma); static void igmp6_leave_group(struct ifmcaddr6 *ma); static void igmp6_timer_handler(unsigned long data); @@ -177,7 +175,7 @@ int sysctl_mld_max_msf __read_mostly = IPV6_MLD_MAX_MSF; * socket join on multicast group */ -int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr) +int ipv6_sock_mc_join(struct sock *sk, int ifindex, const struct in6_addr *addr) { struct net_device *dev = NULL; struct ipv6_mc_socklist *mc_lst; @@ -252,7 +250,7 @@ int ipv6_sock_mc_join(struct sock *sk, int ifindex, struct in6_addr *addr) /* * socket leave on multicast group */ -int ipv6_sock_mc_drop(struct sock *sk, int ifindex, struct in6_addr *addr) +int ipv6_sock_mc_drop(struct sock *sk, int ifindex, const struct in6_addr *addr) { struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6_mc_socklist *mc_lst, **lnk; @@ -664,8 +662,8 @@ done: return err; } -int inet6_mc_check(struct sock *sk, struct in6_addr *mc_addr, - struct in6_addr *src_addr) +int inet6_mc_check(struct sock *sk, const struct in6_addr *mc_addr, + const struct in6_addr *src_addr) { struct ipv6_pinfo *np = inet6_sk(sk); struct ipv6_mc_socklist *mc; @@ -871,7 +869,7 @@ static void mld_clear_delrec(struct inet6_dev *idev) /* * device multicast group inc (add if not found) */ -int ipv6_dev_mc_inc(struct net_device *dev, struct in6_addr *addr) +int ipv6_dev_mc_inc(struct net_device *dev, const struct in6_addr *addr) { struct ifmcaddr6 *mc; struct inet6_dev *idev; @@ -942,7 +940,7 @@ int ipv6_dev_mc_inc(struct net_device *dev, struct in6_addr *addr) /* * device multicast group del */ -int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr) +int __ipv6_dev_mc_dec(struct inet6_dev *idev, const struct in6_addr *addr) { struct ifmcaddr6 *ma, **map; @@ -967,7 +965,7 @@ int __ipv6_dev_mc_dec(struct inet6_dev *idev, struct in6_addr *addr) return -ENOENT; } -int ipv6_dev_mc_dec(struct net_device *dev, struct in6_addr *addr) +int ipv6_dev_mc_dec(struct net_device *dev, const struct in6_addr *addr) { struct inet6_dev *idev = in6_dev_get(dev); int err; @@ -1012,8 +1010,8 @@ int ipv6_is_mld(struct sk_buff *skb, int nexthdr) /* * check if the interface/address pair is valid */ -int ipv6_chk_mcast_addr(struct net_device *dev, struct in6_addr *group, - struct in6_addr *src_addr) +int ipv6_chk_mcast_addr(struct net_device *dev, const struct in6_addr *group, + const struct in6_addr *src_addr) { struct inet6_dev *idev; struct ifmcaddr6 *mc; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index b3295d82fece..5b9ad5e2f56d 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -442,8 +442,9 @@ static void pndisc_destructor(struct pneigh_entry *n) */ static void __ndisc_send(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *daddr, struct in6_addr *saddr, - struct icmp6hdr *icmp6h, struct in6_addr *target, + const struct in6_addr *daddr, + const struct in6_addr *saddr, + struct icmp6hdr *icmp6h, const struct in6_addr *target, int llinfo) { struct flowi fl; @@ -529,12 +530,13 @@ static void __ndisc_send(struct net_device *dev, } static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *daddr, struct in6_addr *solicited_addr, - int router, int solicited, int override, int inc_opt) + const struct in6_addr *daddr, + const struct in6_addr *solicited_addr, + int router, int solicited, int override, int inc_opt) { struct in6_addr tmpaddr; struct inet6_ifaddr *ifp; - struct in6_addr *src_addr; + const struct in6_addr *src_addr; struct icmp6hdr icmp6h = { .icmp6_type = NDISC_NEIGHBOUR_ADVERTISEMENT, }; @@ -564,8 +566,8 @@ static void ndisc_send_na(struct net_device *dev, struct neighbour *neigh, } void ndisc_send_ns(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *solicit, - struct in6_addr *daddr, struct in6_addr *saddr) + const struct in6_addr *solicit, + const struct in6_addr *daddr, const struct in6_addr *saddr) { struct in6_addr addr_buf; struct icmp6hdr icmp6h = { @@ -584,8 +586,8 @@ void ndisc_send_ns(struct net_device *dev, struct neighbour *neigh, !ipv6_addr_any(saddr) ? ND_OPT_SOURCE_LL_ADDR : 0); } -void ndisc_send_rs(struct net_device *dev, struct in6_addr *saddr, - struct in6_addr *daddr) +void ndisc_send_rs(struct net_device *dev, const struct in6_addr *saddr, + const struct in6_addr *daddr) { struct icmp6hdr icmp6h = { .icmp6_type = NDISC_ROUTER_SOLICITATION, @@ -1447,7 +1449,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) } void ndisc_send_redirect(struct sk_buff *skb, struct neighbour *neigh, - struct in6_addr *target) + const struct in6_addr *target) { struct net_device *dev = skb->dev; struct net *net = dev_net(dev); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 269b76093288..6293cb91ed1d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -556,8 +556,8 @@ out: } -struct rt6_info *rt6_lookup(struct net *net, struct in6_addr *daddr, - struct in6_addr *saddr, int oif, int strict) +struct rt6_info *rt6_lookup(struct net *net, const struct in6_addr *daddr, + const struct in6_addr *saddr, int oif, int strict) { struct flowi fl = { .oif = oif, @@ -925,7 +925,7 @@ static DEFINE_SPINLOCK(icmp6_dst_lock); struct dst_entry *icmp6_dst_alloc(struct net_device *dev, struct neighbour *neigh, - struct in6_addr *addr) + const struct in6_addr *addr) { struct rt6_info *rt; struct inet6_dev *idev = in6_dev_get(dev); -- cgit v1.2.3 From f3ee4010e84452aa133e5163e6cfabc52b194e94 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 10 Apr 2008 15:42:11 +0900 Subject: [IPV6]: Define constants for link-local multicast addresses. - Define link-local all-node / all-router multicast addresses. - Remove ipv6_addr_all_nodes() and ipv6_addr_all_routers(). Signed-off-by: YOSHIFUJI Hideaki --- include/linux/in6.h | 8 ++++++++ include/net/addrconf.h | 11 ----------- net/ipv6/addrconf.c | 25 +++++++------------------ net/ipv6/mcast.c | 23 ++++++++--------------- net/ipv6/ndisc.c | 5 +---- 5 files changed, 24 insertions(+), 48 deletions(-) (limited to 'include') diff --git a/include/linux/in6.h b/include/linux/in6.h index e6aa8de2b939..bc492048c349 100644 --- a/include/linux/in6.h +++ b/include/linux/in6.h @@ -48,6 +48,14 @@ extern const struct in6_addr in6addr_any; #define IN6ADDR_ANY_INIT { { { 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 } } } extern const struct in6_addr in6addr_loopback; #define IN6ADDR_LOOPBACK_INIT { { { 0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1 } } } +#ifdef __KERNEL__ +extern const struct in6_addr in6addr_linklocal_allnodes; +#define IN6ADDR_LINKLOCAL_ALLNODES_INIT \ + { { { 0xff,2,0,0,0,0,0,0,0,0,0,0,0,0,0,1 } } } +extern const struct in6_addr in6addr_linklocal_allrouters; +#define IN6ADDR_LINKLOCAL_ALLROUTERS_INIT \ + { { { 0xff,2,0,0,0,0,0,0,0,0,0,0,0,0,0,2 } } } +#endif struct sockaddr_in6 { unsigned short int sin6_family; /* AF_INET6 */ diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 92af23d66eb9..0a2f0372df31 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -205,17 +205,6 @@ static inline void addrconf_addr_solict_mult(const struct in6_addr *addr, htonl(0xFF000000) | addr->s6_addr32[3]); } - -static inline void ipv6_addr_all_nodes(struct in6_addr *addr) -{ - ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x1)); -} - -static inline void ipv6_addr_all_routers(struct in6_addr *addr) -{ - ipv6_addr_set(addr, htonl(0xFF020000), 0, 0, htonl(0x2)); -} - static inline int ipv6_addr_is_multicast(const struct in6_addr *addr) { return (addr->s6_addr32[0] & htonl(0xFF000000)) == htonl(0xFF000000); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 4048c2b73b0b..7df04d294924 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -222,6 +222,8 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = { /* IPv6 Wildcard Address and Loopback Address defined by RFC2553 */ const struct in6_addr in6addr_any = IN6ADDR_ANY_INIT; const struct in6_addr in6addr_loopback = IN6ADDR_LOOPBACK_INIT; +const struct in6_addr in6addr_linklocal_allnodes = IN6ADDR_LINKLOCAL_ALLNODES_INIT; +const struct in6_addr in6addr_linklocal_allrouters = IN6ADDR_LINKLOCAL_ALLROUTERS_INIT; /* Check if a valid qdisc is available */ static inline int addrconf_qdisc_ok(struct net_device *dev) @@ -321,7 +323,6 @@ EXPORT_SYMBOL(in6_dev_finish_destroy); static struct inet6_dev * ipv6_add_dev(struct net_device *dev) { struct inet6_dev *ndev; - struct in6_addr maddr; ASSERT_RTNL(); @@ -406,8 +407,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) rcu_assign_pointer(dev->ip6_ptr, ndev); /* Join all-node multicast group */ - ipv6_addr_all_nodes(&maddr); - ipv6_dev_mc_inc(dev, &maddr); + ipv6_dev_mc_inc(dev, &in6addr_linklocal_allnodes); return ndev; } @@ -433,18 +433,15 @@ static void dev_forward_change(struct inet6_dev *idev) { struct net_device *dev; struct inet6_ifaddr *ifa; - struct in6_addr addr; if (!idev) return; dev = idev->dev; if (dev && (dev->flags & IFF_MULTICAST)) { - ipv6_addr_all_routers(&addr); - if (idev->cnf.forwarding) - ipv6_dev_mc_inc(dev, &addr); + ipv6_dev_mc_inc(dev, &in6addr_linklocal_allrouters); else - ipv6_dev_mc_dec(dev, &addr); + ipv6_dev_mc_dec(dev, &in6addr_linklocal_allrouters); } for (ifa=idev->addr_list; ifa; ifa=ifa->if_next) { if (ifa->flags&IFA_F_TENTATIVE) @@ -2654,8 +2651,6 @@ static void addrconf_rs_timer(unsigned long data) spin_lock(&ifp->lock); if (ifp->probes++ < ifp->idev->cnf.rtr_solicits) { - struct in6_addr all_routers; - /* The wait after the last probe can be shorter */ addrconf_mod_timer(ifp, AC_RS, (ifp->probes == ifp->idev->cnf.rtr_solicits) ? @@ -2663,9 +2658,7 @@ static void addrconf_rs_timer(unsigned long data) ifp->idev->cnf.rtr_solicit_interval); spin_unlock(&ifp->lock); - ipv6_addr_all_routers(&all_routers); - - ndisc_send_rs(ifp->idev->dev, &ifp->addr, &all_routers); + ndisc_send_rs(ifp->idev->dev, &ifp->addr, &in6addr_linklocal_allrouters); } else { spin_unlock(&ifp->lock); /* @@ -2806,16 +2799,12 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp) ifp->idev->cnf.rtr_solicits > 0 && (dev->flags&IFF_LOOPBACK) == 0 && (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)) { - struct in6_addr all_routers; - - ipv6_addr_all_routers(&all_routers); - /* * If a host as already performed a random delay * [...] as part of DAD [...] there is no need * to delay again before sending the first RS */ - ndisc_send_rs(ifp->idev->dev, &ifp->addr, &all_routers); + ndisc_send_rs(ifp->idev->dev, &ifp->addr, &in6addr_linklocal_allrouters); spin_lock_bh(&ifp->lock); ifp->probes = 1; diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 0a0132a1c443..c2dc2e2b6c07 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1766,10 +1766,9 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) struct inet6_dev *idev; struct sk_buff *skb; struct icmp6hdr *hdr; - struct in6_addr *snd_addr; + const struct in6_addr *snd_addr; struct in6_addr *addrp; struct in6_addr addr_buf; - struct in6_addr all_routers; int err, len, payload_len, full_len; u8 ra[8] = { IPPROTO_ICMPV6, 0, IPV6_TLV_ROUTERALERT, 2, 0, 0, @@ -1780,11 +1779,10 @@ static void igmp6_send(struct in6_addr *addr, struct net_device *dev, int type) IP6_INC_STATS(__in6_dev_get(dev), IPSTATS_MIB_OUTREQUESTS); rcu_read_unlock(); - snd_addr = addr; - if (type == ICMPV6_MGM_REDUCTION) { - snd_addr = &all_routers; - ipv6_addr_all_routers(&all_routers); - } + if (type == ICMPV6_MGM_REDUCTION) + snd_addr = &in6addr_linklocal_allrouters; + else + snd_addr = addr; len = sizeof(struct icmp6hdr) + sizeof(struct in6_addr); payload_len = len + sizeof(ra); @@ -2309,24 +2307,19 @@ void ipv6_mc_init_dev(struct inet6_dev *idev) void ipv6_mc_destroy_dev(struct inet6_dev *idev) { struct ifmcaddr6 *i; - struct in6_addr maddr; /* Deactivate timers */ ipv6_mc_down(idev); /* Delete all-nodes address. */ - ipv6_addr_all_nodes(&maddr); - /* We cannot call ipv6_dev_mc_dec() directly, our caller in * addrconf.c has NULL'd out dev->ip6_ptr so in6_dev_get() will * fail. */ - __ipv6_dev_mc_dec(idev, &maddr); + __ipv6_dev_mc_dec(idev, &in6addr_linklocal_allnodes); - if (idev->cnf.forwarding) { - ipv6_addr_all_routers(&maddr); - __ipv6_dev_mc_dec(idev, &maddr); - } + if (idev->cnf.forwarding) + __ipv6_dev_mc_dec(idev, &in6addr_linklocal_allrouters); write_lock_bh(&idev->lock); while ((i = idev->mc_list) != NULL) { diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 5b9ad5e2f56d..2c74885f8355 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -818,10 +818,7 @@ static void ndisc_recv_ns(struct sk_buff *skb) is_router = !!idev->cnf.forwarding; if (dad) { - struct in6_addr maddr; - - ipv6_addr_all_nodes(&maddr); - ndisc_send_na(dev, NULL, &maddr, &msg->target, + ndisc_send_na(dev, NULL, &in6addr_linklocal_allnodes, &msg->target, is_router, 0, (ifp != NULL), 1); goto out; } -- cgit v1.2.3 From 7f1eced8b0a001c4d5a8cfa5ac7b5cbc89fedab8 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Thu, 10 Apr 2008 15:42:12 +0900 Subject: [IPV6] MIP6: Use our standard definitions for paddings. MIP6_OPT_PAD_X are actually for paddings in destination option header. Replace them with our standard IPV6_TLV_PADX. Signed-off-by: YOSHIFUJI Hideaki --- include/net/mip6.h | 3 --- net/ipv6/mip6.c | 4 ++-- 2 files changed, 2 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/net/mip6.h b/include/net/mip6.h index 63272610a24a..a83ad1982a90 100644 --- a/include/net/mip6.h +++ b/include/net/mip6.h @@ -28,9 +28,6 @@ #include #include -#define MIP6_OPT_PAD_1 0 -#define MIP6_OPT_PAD_N 1 - /* * Mobility Header */ diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c index 42403c626c27..ad1cc5bbf977 100644 --- a/net/ipv6/mip6.c +++ b/net/ipv6/mip6.c @@ -44,9 +44,9 @@ static inline void *mip6_padn(__u8 *data, __u8 padlen) if (!data) return NULL; if (padlen == 1) { - data[0] = MIP6_OPT_PAD_1; + data[0] = IPV6_TLV_PAD0; } else if (padlen > 1) { - data[0] = MIP6_OPT_PAD_N; + data[0] = IPV6_TLV_PADN; data[1] = padlen - 2; if (padlen > 2) memset(data+2, 0, data[1]); -- cgit v1.2.3 From 00447872a643787411c2c0cb1df6169dda8b0c47 Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Sat, 12 Apr 2008 19:06:42 -0700 Subject: NetLabel: Allow passing the LSM domain as a shared pointer Smack doesn't have the need to create a private copy of the LSM "domain" when setting NetLabel security attributes like SELinux, however, the current NetLabel code requires a private copy of the LSM "domain". This patches fixes that by letting the LSM determine how it wants to pass the domain value. * NETLBL_SECATTR_DOMAIN_CPY The current behavior, NetLabel assumes that the domain value is a copy and frees it when done * NETLBL_SECATTR_DOMAIN New, Smack-friendly behavior, NetLabel assumes that the domain value is a reference to a string managed by the LSM and does not free it when done Signed-off-by: Paul Moore Acked-by: James Morris Signed-off-by: David S. Miller --- include/net/netlabel.h | 14 ++++++++++---- security/selinux/ss/services.c | 2 +- security/smack/smack_lsm.c | 2 +- 3 files changed, 12 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/net/netlabel.h b/include/net/netlabel.h index 0ca67d73c7ad..5e53a85b5ca1 100644 --- a/include/net/netlabel.h +++ b/include/net/netlabel.h @@ -162,7 +162,7 @@ struct netlbl_lsm_secattr_catmap { /** * struct netlbl_lsm_secattr - NetLabel LSM security attributes - * @flags: indicate which attributes are contained in this structure + * @flags: indicate structure attributes, see NETLBL_SECATTR_* * @type: indicate the NLTYPE of the attributes * @domain: the NetLabel LSM domain * @cache: NetLabel LSM specific cache @@ -180,17 +180,22 @@ struct netlbl_lsm_secattr_catmap { * NetLabel itself when returning security attributes to the LSM. * */ +struct netlbl_lsm_secattr { + u32 flags; + /* bitmap values for 'flags' */ #define NETLBL_SECATTR_NONE 0x00000000 #define NETLBL_SECATTR_DOMAIN 0x00000001 +#define NETLBL_SECATTR_DOMAIN_CPY (NETLBL_SECATTR_DOMAIN | \ + NETLBL_SECATTR_FREE_DOMAIN) #define NETLBL_SECATTR_CACHE 0x00000002 #define NETLBL_SECATTR_MLS_LVL 0x00000004 #define NETLBL_SECATTR_MLS_CAT 0x00000008 #define NETLBL_SECATTR_SECID 0x00000010 + /* bitmap meta-values for 'flags' */ +#define NETLBL_SECATTR_FREE_DOMAIN 0x01000000 #define NETLBL_SECATTR_CACHEABLE (NETLBL_SECATTR_MLS_LVL | \ NETLBL_SECATTR_MLS_CAT | \ NETLBL_SECATTR_SECID) -struct netlbl_lsm_secattr { - u32 flags; u32 type; char *domain; struct netlbl_lsm_cache *cache; @@ -303,7 +308,8 @@ static inline void netlbl_secattr_init(struct netlbl_lsm_secattr *secattr) */ static inline void netlbl_secattr_destroy(struct netlbl_lsm_secattr *secattr) { - kfree(secattr->domain); + if (secattr->flags & NETLBL_SECATTR_FREE_DOMAIN) + kfree(secattr->domain); if (secattr->flags & NETLBL_SECATTR_CACHE) netlbl_secattr_cache_free(secattr->cache); if (secattr->flags & NETLBL_SECATTR_MLS_CAT) diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index f37418601215..47295acd09c9 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -2649,7 +2649,7 @@ int security_netlbl_sid_to_secattr(u32 sid, struct netlbl_lsm_secattr *secattr) goto netlbl_sid_to_secattr_failure; secattr->domain = kstrdup(policydb.p_type_val_to_name[ctx->type - 1], GFP_ATOMIC); - secattr->flags |= NETLBL_SECATTR_DOMAIN; + secattr->flags |= NETLBL_SECATTR_DOMAIN_CPY; mls_export_netlbl_lvl(ctx, secattr); rc = mls_export_netlbl_cat(ctx, secattr); if (rc != 0) diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 732ba27923c4..e2d6f7cd9254 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -1275,7 +1275,7 @@ static void smack_to_secattr(char *smack, struct netlbl_lsm_secattr *nlsp) switch (smack_net_nltype) { case NETLBL_NLTYPE_CIPSOV4: - nlsp->domain = kstrdup(smack, GFP_ATOMIC); + nlsp->domain = smack; nlsp->flags = NETLBL_SECATTR_DOMAIN | NETLBL_SECATTR_MLS_LVL; rc = smack_to_cipso(smack, &cipso); -- cgit v1.2.3 From 03e1ad7b5d871d4189b1da3125c2f12d1b5f7d0b Mon Sep 17 00:00:00 2001 From: Paul Moore Date: Sat, 12 Apr 2008 19:07:52 -0700 Subject: LSM: Make the Labeled IPsec hooks more stack friendly The xfrm_get_policy() and xfrm_add_pol_expire() put some rather large structs on the stack to work around the LSM API. This patch attempts to fix that problem by changing the LSM API to require only the relevant "security" pointers instead of the entire SPD entry; we do this for all of the security_xfrm_policy*() functions to keep things consistent. Signed-off-by: Paul Moore Acked-by: James Morris Signed-off-by: David S. Miller --- include/linux/security.h | 48 ++++++++++++++++++++--------------------- net/key/af_key.c | 23 ++++++++++---------- net/xfrm/xfrm_policy.c | 24 +++++++++++++-------- net/xfrm/xfrm_user.c | 33 ++++++++++++++-------------- security/dummy.c | 14 ++++++------ security/security.c | 21 +++++++++--------- security/selinux/include/xfrm.h | 13 +++++------ security/selinux/xfrm.c | 39 ++++++++++++++------------------- 8 files changed, 109 insertions(+), 106 deletions(-) (limited to 'include') diff --git a/include/linux/security.h b/include/linux/security.h index c673dfd4dffc..f5eb9ff47ac5 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -910,24 +910,24 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts) * Security hooks for XFRM operations. * * @xfrm_policy_alloc_security: - * @xp contains the xfrm_policy being added to Security Policy Database - * used by the XFRM system. + * @ctxp is a pointer to the xfrm_sec_ctx being added to Security Policy + * Database used by the XFRM system. * @sec_ctx contains the security context information being provided by * the user-level policy update program (e.g., setkey). * Allocate a security structure to the xp->security field; the security * field is initialized to NULL when the xfrm_policy is allocated. * Return 0 if operation was successful (memory to allocate, legal context) * @xfrm_policy_clone_security: - * @old contains an existing xfrm_policy in the SPD. - * @new contains a new xfrm_policy being cloned from old. - * Allocate a security structure to the new->security field - * that contains the information from the old->security field. + * @old_ctx contains an existing xfrm_sec_ctx. + * @new_ctxp contains a new xfrm_sec_ctx being cloned from old. + * Allocate a security structure in new_ctxp that contains the + * information from the old_ctx structure. * Return 0 if operation was successful (memory to allocate). * @xfrm_policy_free_security: - * @xp contains the xfrm_policy + * @ctx contains the xfrm_sec_ctx * Deallocate xp->security. * @xfrm_policy_delete_security: - * @xp contains the xfrm_policy. + * @ctx contains the xfrm_sec_ctx. * Authorize deletion of xp->security. * @xfrm_state_alloc_security: * @x contains the xfrm_state being added to the Security Association @@ -947,7 +947,7 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts) * @x contains the xfrm_state. * Authorize deletion of x->security. * @xfrm_policy_lookup: - * @xp contains the xfrm_policy for which the access control is being + * @ctx contains the xfrm_sec_ctx for which the access control is being * checked. * @fl_secid contains the flow security label that is used to authorize * access to the policy xp. @@ -1454,17 +1454,17 @@ struct security_operations { #endif /* CONFIG_SECURITY_NETWORK */ #ifdef CONFIG_SECURITY_NETWORK_XFRM - int (*xfrm_policy_alloc_security) (struct xfrm_policy *xp, + int (*xfrm_policy_alloc_security) (struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx); - int (*xfrm_policy_clone_security) (struct xfrm_policy *old, struct xfrm_policy *new); - void (*xfrm_policy_free_security) (struct xfrm_policy *xp); - int (*xfrm_policy_delete_security) (struct xfrm_policy *xp); + int (*xfrm_policy_clone_security) (struct xfrm_sec_ctx *old_ctx, struct xfrm_sec_ctx **new_ctx); + void (*xfrm_policy_free_security) (struct xfrm_sec_ctx *ctx); + int (*xfrm_policy_delete_security) (struct xfrm_sec_ctx *ctx); int (*xfrm_state_alloc_security) (struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx, u32 secid); void (*xfrm_state_free_security) (struct xfrm_state *x); int (*xfrm_state_delete_security) (struct xfrm_state *x); - int (*xfrm_policy_lookup)(struct xfrm_policy *xp, u32 fl_secid, u8 dir); + int (*xfrm_policy_lookup)(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir); int (*xfrm_state_pol_flow_match)(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); int (*xfrm_decode_session)(struct sk_buff *skb, u32 *secid, int ckall); @@ -2562,16 +2562,16 @@ static inline void security_inet_conn_established(struct sock *sk, #ifdef CONFIG_SECURITY_NETWORK_XFRM -int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx); -int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new); -void security_xfrm_policy_free(struct xfrm_policy *xp); -int security_xfrm_policy_delete(struct xfrm_policy *xp); +int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx); +int security_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx, struct xfrm_sec_ctx **new_ctxp); +void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx); +int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx); int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx); int security_xfrm_state_alloc_acquire(struct xfrm_state *x, struct xfrm_sec_ctx *polsec, u32 secid); int security_xfrm_state_delete(struct xfrm_state *x); void security_xfrm_state_free(struct xfrm_state *x); -int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); +int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir); int security_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); int security_xfrm_decode_session(struct sk_buff *skb, u32 *secid); @@ -2579,21 +2579,21 @@ void security_skb_classify_flow(struct sk_buff *skb, struct flowi *fl); #else /* CONFIG_SECURITY_NETWORK_XFRM */ -static inline int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx) +static inline int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx) { return 0; } -static inline int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) +static inline int security_xfrm_policy_clone(struct xfrm_sec_ctx *old, struct xfrm_sec_ctx **new_ctxp) { return 0; } -static inline void security_xfrm_policy_free(struct xfrm_policy *xp) +static inline void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx) { } -static inline int security_xfrm_policy_delete(struct xfrm_policy *xp) +static inline int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx) { return 0; } @@ -2619,7 +2619,7 @@ static inline int security_xfrm_state_delete(struct xfrm_state *x) return 0; } -static inline int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir) +static inline int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir) { return 0; } diff --git a/net/key/af_key.c b/net/key/af_key.c index 6db58924368a..1fb0fe42a72e 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -2292,7 +2292,7 @@ static int pfkey_spdadd(struct sock *sk, struct sk_buff *skb, struct sadb_msg *h goto out; } - err = security_xfrm_policy_alloc(xp, uctx); + err = security_xfrm_policy_alloc(&xp->security, uctx); kfree(uctx); if (err) @@ -2352,10 +2352,11 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg int err; struct sadb_address *sa; struct sadb_x_policy *pol; - struct xfrm_policy *xp, tmp; + struct xfrm_policy *xp; struct xfrm_selector sel; struct km_event c; struct sadb_x_sec_ctx *sec_ctx; + struct xfrm_sec_ctx *pol_ctx; if (!present_and_same_family(ext_hdrs[SADB_EXT_ADDRESS_SRC-1], ext_hdrs[SADB_EXT_ADDRESS_DST-1]) || @@ -2385,25 +2386,23 @@ static int pfkey_spddelete(struct sock *sk, struct sk_buff *skb, struct sadb_msg sel.dport_mask = htons(0xffff); sec_ctx = (struct sadb_x_sec_ctx *) ext_hdrs[SADB_X_EXT_SEC_CTX-1]; - memset(&tmp, 0, sizeof(struct xfrm_policy)); - if (sec_ctx != NULL) { struct xfrm_user_sec_ctx *uctx = pfkey_sadb2xfrm_user_sec_ctx(sec_ctx); if (!uctx) return -ENOMEM; - err = security_xfrm_policy_alloc(&tmp, uctx); + err = security_xfrm_policy_alloc(&pol_ctx, uctx); kfree(uctx); - if (err) return err; - } - - xp = xfrm_policy_bysel_ctx(XFRM_POLICY_TYPE_MAIN, pol->sadb_x_policy_dir-1, - &sel, tmp.security, 1, &err); - security_xfrm_policy_free(&tmp); + } else + pol_ctx = NULL; + xp = xfrm_policy_bysel_ctx(XFRM_POLICY_TYPE_MAIN, + pol->sadb_x_policy_dir - 1, &sel, pol_ctx, + 1, &err); + security_xfrm_policy_free(pol_ctx); if (xp == NULL) return -ENOENT; @@ -3298,7 +3297,7 @@ static struct xfrm_policy *pfkey_compile_policy(struct sock *sk, int opt, if ((*dir = verify_sec_ctx_len(p))) goto out; uctx = pfkey_sadb2xfrm_user_sec_ctx(sec_ctx); - *dir = security_xfrm_policy_alloc(xp, uctx); + *dir = security_xfrm_policy_alloc(&xp->security, uctx); kfree(uctx); if (*dir) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 15d73e47cc2c..ab4d0e598a2c 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -263,7 +263,7 @@ void xfrm_policy_destroy(struct xfrm_policy *policy) list_del(&policy->bytype); write_unlock_bh(&xfrm_policy_lock); - security_xfrm_policy_free(policy); + security_xfrm_policy_free(policy->security); kfree(policy); } EXPORT_SYMBOL(xfrm_policy_destroy); @@ -676,7 +676,8 @@ struct xfrm_policy *xfrm_policy_bysel_ctx(u8 type, int dir, xfrm_sec_ctx_match(ctx, pol->security)) { xfrm_pol_hold(pol); if (delete) { - *err = security_xfrm_policy_delete(pol); + *err = security_xfrm_policy_delete( + pol->security); if (*err) { write_unlock_bh(&xfrm_policy_lock); return pol; @@ -718,7 +719,8 @@ struct xfrm_policy *xfrm_policy_byid(u8 type, int dir, u32 id, int delete, if (pol->type == type && pol->index == id) { xfrm_pol_hold(pol); if (delete) { - *err = security_xfrm_policy_delete(pol); + *err = security_xfrm_policy_delete( + pol->security); if (*err) { write_unlock_bh(&xfrm_policy_lock); return pol; @@ -756,7 +758,7 @@ xfrm_policy_flush_secctx_check(u8 type, struct xfrm_audit *audit_info) &xfrm_policy_inexact[dir], bydst) { if (pol->type != type) continue; - err = security_xfrm_policy_delete(pol); + err = security_xfrm_policy_delete(pol->security); if (err) { xfrm_audit_policy_delete(pol, 0, audit_info->loginuid, @@ -770,7 +772,8 @@ xfrm_policy_flush_secctx_check(u8 type, struct xfrm_audit *audit_info) bydst) { if (pol->type != type) continue; - err = security_xfrm_policy_delete(pol); + err = security_xfrm_policy_delete( + pol->security); if (err) { xfrm_audit_policy_delete(pol, 0, audit_info->loginuid, @@ -931,7 +934,8 @@ static int xfrm_policy_match(struct xfrm_policy *pol, struct flowi *fl, match = xfrm_selector_match(sel, fl, family); if (match) - ret = security_xfrm_policy_lookup(pol, fl->secid, dir); + ret = security_xfrm_policy_lookup(pol->security, fl->secid, + dir); return ret; } @@ -1048,8 +1052,9 @@ static struct xfrm_policy *xfrm_sk_policy_lookup(struct sock *sk, int dir, struc int err = 0; if (match) { - err = security_xfrm_policy_lookup(pol, fl->secid, - policy_to_flow_dir(dir)); + err = security_xfrm_policy_lookup(pol->security, + fl->secid, + policy_to_flow_dir(dir)); if (!err) xfrm_pol_hold(pol); else if (err == -ESRCH) @@ -1138,7 +1143,8 @@ static struct xfrm_policy *clone_policy(struct xfrm_policy *old, int dir) if (newp) { newp->selector = old->selector; - if (security_xfrm_policy_clone(old, newp)) { + if (security_xfrm_policy_clone(old->security, + &newp->security)) { kfree(newp); return NULL; /* ENOMEM */ } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 5578c909fcf6..ecf9d67daef5 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -959,7 +959,7 @@ static int copy_from_user_sec_ctx(struct xfrm_policy *pol, struct nlattr **attrs return 0; uctx = nla_data(rt); - return security_xfrm_policy_alloc(pol, uctx); + return security_xfrm_policy_alloc(&pol->security, uctx); } static void copy_templates(struct xfrm_policy *xp, struct xfrm_user_tmpl *ut, @@ -1143,7 +1143,7 @@ static int xfrm_add_policy(struct sk_buff *skb, struct nlmsghdr *nlh, NETLINK_CB(skb).sid); if (err) { - security_xfrm_policy_free(xp); + security_xfrm_policy_free(xp->security); kfree(xp); return err; } @@ -1337,22 +1337,23 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, xp = xfrm_policy_byid(type, p->dir, p->index, delete, &err); else { struct nlattr *rt = attrs[XFRMA_SEC_CTX]; - struct xfrm_policy tmp; + struct xfrm_sec_ctx *ctx; err = verify_sec_ctx_len(attrs); if (err) return err; - memset(&tmp, 0, sizeof(struct xfrm_policy)); if (rt) { struct xfrm_user_sec_ctx *uctx = nla_data(rt); - if ((err = security_xfrm_policy_alloc(&tmp, uctx))) + err = security_xfrm_policy_alloc(&ctx, uctx); + if (err) return err; - } - xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security, + } else + ctx = NULL; + xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, ctx, delete, &err); - security_xfrm_policy_free(&tmp); + security_xfrm_policy_free(ctx); } if (xp == NULL) return -ENOENT; @@ -1572,26 +1573,26 @@ static int xfrm_add_pol_expire(struct sk_buff *skb, struct nlmsghdr *nlh, xp = xfrm_policy_byid(type, p->dir, p->index, 0, &err); else { struct nlattr *rt = attrs[XFRMA_SEC_CTX]; - struct xfrm_policy tmp; + struct xfrm_sec_ctx *ctx; err = verify_sec_ctx_len(attrs); if (err) return err; - memset(&tmp, 0, sizeof(struct xfrm_policy)); if (rt) { struct xfrm_user_sec_ctx *uctx = nla_data(rt); - if ((err = security_xfrm_policy_alloc(&tmp, uctx))) + err = security_xfrm_policy_alloc(&ctx, uctx); + if (err) return err; - } - xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, tmp.security, - 0, &err); - security_xfrm_policy_free(&tmp); + } else + ctx = NULL; + xp = xfrm_policy_bysel_ctx(type, p->dir, &p->sel, ctx, 0, &err); + security_xfrm_policy_free(ctx); } - if (xp == NULL) return -ENOENT; + read_lock(&xp->lock); if (xp->dead) { read_unlock(&xp->lock); diff --git a/security/dummy.c b/security/dummy.c index 78d8f92310a4..480366f9c41d 100644 --- a/security/dummy.c +++ b/security/dummy.c @@ -876,22 +876,23 @@ static inline void dummy_req_classify_flow(const struct request_sock *req, #endif /* CONFIG_SECURITY_NETWORK */ #ifdef CONFIG_SECURITY_NETWORK_XFRM -static int dummy_xfrm_policy_alloc_security(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx) +static int dummy_xfrm_policy_alloc_security(struct xfrm_sec_ctx **ctxp, + struct xfrm_user_sec_ctx *sec_ctx) { return 0; } -static inline int dummy_xfrm_policy_clone_security(struct xfrm_policy *old, struct xfrm_policy *new) +static inline int dummy_xfrm_policy_clone_security(struct xfrm_sec_ctx *old_ctx, + struct xfrm_sec_ctx **new_ctxp) { return 0; } -static void dummy_xfrm_policy_free_security(struct xfrm_policy *xp) +static void dummy_xfrm_policy_free_security(struct xfrm_sec_ctx *ctx) { } -static int dummy_xfrm_policy_delete_security(struct xfrm_policy *xp) +static int dummy_xfrm_policy_delete_security(struct xfrm_sec_ctx *ctx) { return 0; } @@ -911,7 +912,8 @@ static int dummy_xfrm_state_delete_security(struct xfrm_state *x) return 0; } -static int dummy_xfrm_policy_lookup(struct xfrm_policy *xp, u32 sk_sid, u8 dir) +static int dummy_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, + u32 sk_sid, u8 dir) { return 0; } diff --git a/security/security.c b/security/security.c index b1387a6b416d..c9ff7d18c2f4 100644 --- a/security/security.c +++ b/security/security.c @@ -1014,26 +1014,27 @@ void security_inet_conn_established(struct sock *sk, #ifdef CONFIG_SECURITY_NETWORK_XFRM -int security_xfrm_policy_alloc(struct xfrm_policy *xp, struct xfrm_user_sec_ctx *sec_ctx) +int security_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, struct xfrm_user_sec_ctx *sec_ctx) { - return security_ops->xfrm_policy_alloc_security(xp, sec_ctx); + return security_ops->xfrm_policy_alloc_security(ctxp, sec_ctx); } EXPORT_SYMBOL(security_xfrm_policy_alloc); -int security_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) +int security_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx, + struct xfrm_sec_ctx **new_ctxp) { - return security_ops->xfrm_policy_clone_security(old, new); + return security_ops->xfrm_policy_clone_security(old_ctx, new_ctxp); } -void security_xfrm_policy_free(struct xfrm_policy *xp) +void security_xfrm_policy_free(struct xfrm_sec_ctx *ctx) { - security_ops->xfrm_policy_free_security(xp); + security_ops->xfrm_policy_free_security(ctx); } EXPORT_SYMBOL(security_xfrm_policy_free); -int security_xfrm_policy_delete(struct xfrm_policy *xp) +int security_xfrm_policy_delete(struct xfrm_sec_ctx *ctx) { - return security_ops->xfrm_policy_delete_security(xp); + return security_ops->xfrm_policy_delete_security(ctx); } int security_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx) @@ -1065,9 +1066,9 @@ void security_xfrm_state_free(struct xfrm_state *x) security_ops->xfrm_state_free_security(x); } -int security_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir) +int security_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir) { - return security_ops->xfrm_policy_lookup(xp, fl_secid, dir); + return security_ops->xfrm_policy_lookup(ctx, fl_secid, dir); } int security_xfrm_state_pol_flow_match(struct xfrm_state *x, diff --git a/security/selinux/include/xfrm.h b/security/selinux/include/xfrm.h index 36b0510efa7b..289e24b39e3e 100644 --- a/security/selinux/include/xfrm.h +++ b/security/selinux/include/xfrm.h @@ -7,16 +7,17 @@ #ifndef _SELINUX_XFRM_H_ #define _SELINUX_XFRM_H_ -int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *sec_ctx); -int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new); -void selinux_xfrm_policy_free(struct xfrm_policy *xp); -int selinux_xfrm_policy_delete(struct xfrm_policy *xp); +int selinux_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, + struct xfrm_user_sec_ctx *sec_ctx); +int selinux_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx, + struct xfrm_sec_ctx **new_ctxp); +void selinux_xfrm_policy_free(struct xfrm_sec_ctx *ctx); +int selinux_xfrm_policy_delete(struct xfrm_sec_ctx *ctx); int selinux_xfrm_state_alloc(struct xfrm_state *x, struct xfrm_user_sec_ctx *sec_ctx, u32 secid); void selinux_xfrm_state_free(struct xfrm_state *x); int selinux_xfrm_state_delete(struct xfrm_state *x); -int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir); +int selinux_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir); int selinux_xfrm_state_pol_flow_match(struct xfrm_state *x, struct xfrm_policy *xp, struct flowi *fl); diff --git a/security/selinux/xfrm.c b/security/selinux/xfrm.c index 7e158205d081..874d17c83c61 100644 --- a/security/selinux/xfrm.c +++ b/security/selinux/xfrm.c @@ -77,20 +77,18 @@ static inline int selinux_authorizable_xfrm(struct xfrm_state *x) * LSM hook implementation that authorizes that a flow can use * a xfrm policy rule. */ -int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir) +int selinux_xfrm_policy_lookup(struct xfrm_sec_ctx *ctx, u32 fl_secid, u8 dir) { int rc; u32 sel_sid; - struct xfrm_sec_ctx *ctx; /* Context sid is either set to label or ANY_ASSOC */ - if ((ctx = xp->security)) { + if (ctx) { if (!selinux_authorizable_ctx(ctx)) return -EINVAL; sel_sid = ctx->ctx_sid; - } - else + } else /* * All flows should be treated as polmatch'ing an * otherwise applicable "non-labeled" policy. This @@ -103,7 +101,7 @@ int selinux_xfrm_policy_lookup(struct xfrm_policy *xp, u32 fl_secid, u8 dir) NULL); if (rc == -EACCES) - rc = -ESRCH; + return -ESRCH; return rc; } @@ -287,15 +285,14 @@ out2: * LSM hook implementation that allocs and transfers uctx spec to * xfrm_policy. */ -int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, - struct xfrm_user_sec_ctx *uctx) +int selinux_xfrm_policy_alloc(struct xfrm_sec_ctx **ctxp, + struct xfrm_user_sec_ctx *uctx) { int err; - BUG_ON(!xp); BUG_ON(!uctx); - err = selinux_xfrm_sec_ctx_alloc(&xp->security, uctx, 0); + err = selinux_xfrm_sec_ctx_alloc(ctxp, uctx, 0); if (err == 0) atomic_inc(&selinux_xfrm_refcount); @@ -307,32 +304,29 @@ int selinux_xfrm_policy_alloc(struct xfrm_policy *xp, * LSM hook implementation that copies security data structure from old to * new for policy cloning. */ -int selinux_xfrm_policy_clone(struct xfrm_policy *old, struct xfrm_policy *new) +int selinux_xfrm_policy_clone(struct xfrm_sec_ctx *old_ctx, + struct xfrm_sec_ctx **new_ctxp) { - struct xfrm_sec_ctx *old_ctx, *new_ctx; - - old_ctx = old->security; + struct xfrm_sec_ctx *new_ctx; if (old_ctx) { - new_ctx = new->security = kmalloc(sizeof(*new_ctx) + - old_ctx->ctx_len, - GFP_KERNEL); - + new_ctx = kmalloc(sizeof(*old_ctx) + old_ctx->ctx_len, + GFP_KERNEL); if (!new_ctx) return -ENOMEM; memcpy(new_ctx, old_ctx, sizeof(*new_ctx)); memcpy(new_ctx->ctx_str, old_ctx->ctx_str, new_ctx->ctx_len); + *new_ctxp = new_ctx; } return 0; } /* - * LSM hook implementation that frees xfrm_policy security information. + * LSM hook implementation that frees xfrm_sec_ctx security information. */ -void selinux_xfrm_policy_free(struct xfrm_policy *xp) +void selinux_xfrm_policy_free(struct xfrm_sec_ctx *ctx) { - struct xfrm_sec_ctx *ctx = xp->security; if (ctx) kfree(ctx); } @@ -340,10 +334,9 @@ void selinux_xfrm_policy_free(struct xfrm_policy *xp) /* * LSM hook implementation that authorizes deletion of labeled policies. */ -int selinux_xfrm_policy_delete(struct xfrm_policy *xp) +int selinux_xfrm_policy_delete(struct xfrm_sec_ctx *ctx) { struct task_security_struct *tsec = current->security; - struct xfrm_sec_ctx *ctx = xp->security; int rc = 0; if (ctx) { -- cgit v1.2.3 From a4146b1b2c6ba995db08b1a2aef5af1b17b151e6 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Sun, 13 Apr 2008 22:11:14 -0700 Subject: [TCP]: Replace struct net on tcp_iter_state with seq_net_private. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/tcp.h | 2 +- net/ipv4/tcp_ipv4.c | 11 +++++------ 2 files changed, 6 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 7b41bb962b9c..f5b61e0f01f2 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1333,7 +1333,7 @@ struct tcp_seq_afinfo { }; struct tcp_iter_state { - struct net *net; + struct seq_net_private p; sa_family_t family; enum tcp_seq_states state; struct sock *syn_wait_sk; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 52e3ae603ca9..86148cdfb21f 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1954,7 +1954,7 @@ static void *listening_get_next(struct seq_file *seq, void *cur) struct hlist_node *node; struct sock *sk = cur; struct tcp_iter_state* st = seq->private; - struct net *net = st->net; + struct net *net = seq_file_net(seq); if (!sk) { st->bucket = 0; @@ -2035,7 +2035,7 @@ static void *listening_get_idx(struct seq_file *seq, loff_t *pos) static void *established_get_first(struct seq_file *seq) { struct tcp_iter_state* st = seq->private; - struct net *net = st->net; + struct net *net = seq_file_net(seq); void *rc = NULL; for (st->bucket = 0; st->bucket < tcp_hashinfo.ehash_size; ++st->bucket) { @@ -2076,7 +2076,7 @@ static void *established_get_next(struct seq_file *seq, void *cur) struct inet_timewait_sock *tw; struct hlist_node *node; struct tcp_iter_state* st = seq->private; - struct net *net = st->net; + struct net *net = seq_file_net(seq); ++st->num; @@ -2233,7 +2233,7 @@ static int tcp_seq_open(struct inode *inode, struct file *file) s->seq_ops.next = tcp_seq_next; s->seq_ops.show = afinfo->seq_show; s->seq_ops.stop = tcp_seq_stop; - s->net = net; + s->p.net = net; rc = seq_open(file, &s->seq_ops); if (rc) @@ -2252,9 +2252,8 @@ out_kfree: static int tcp_seq_release(struct inode *inode, struct file *file) { struct seq_file *seq = file->private_data; - struct tcp_iter_state *s = seq->private; - put_net(s->net); + put_net(seq_file_net(seq)); seq_release_private(inode, file); return 0; } -- cgit v1.2.3 From 9427c4b36b8fe652df1d7c89eae678948e1f4b32 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Sun, 13 Apr 2008 22:12:13 -0700 Subject: [TCP]: Move seq_ops from tcp_iter_state to tcp_seq_afinfo. No need to create seq_operations for each instance of 'netstat'. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/tcp.h | 3 +-- net/ipv4/tcp_ipv4.c | 14 ++++++++------ net/ipv6/tcp_ipv6.c | 4 +++- 3 files changed, 12 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index f5b61e0f01f2..2c9a650412c0 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1328,8 +1328,8 @@ struct tcp_seq_afinfo { struct module *owner; char *name; sa_family_t family; - int (*seq_show) (struct seq_file *m, void *v); struct file_operations *seq_fops; + struct seq_operations seq_ops; }; struct tcp_iter_state { @@ -1338,7 +1338,6 @@ struct tcp_iter_state { enum tcp_seq_states state; struct sock *syn_wait_sk; int bucket, sbucket, num, uid; - struct seq_operations seq_ops; }; extern int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo); diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 978d9db1df34..c50dd1793643 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2229,13 +2229,9 @@ static int tcp_seq_open(struct inode *inode, struct file *file) goto out_kfree; s->family = afinfo->family; - s->seq_ops.start = tcp_seq_start; - s->seq_ops.next = tcp_seq_next; - s->seq_ops.show = afinfo->seq_show; - s->seq_ops.stop = tcp_seq_stop; s->p.net = net; - rc = seq_open(file, &s->seq_ops); + rc = seq_open(file, &afinfo->seq_ops); if (rc) goto out_put_net; seq = file->private_data; @@ -2269,6 +2265,10 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) afinfo->seq_fops->llseek = seq_lseek; afinfo->seq_fops->release = tcp_seq_release; + afinfo->seq_ops.start = tcp_seq_start; + afinfo->seq_ops.next = tcp_seq_next; + afinfo->seq_ops.stop = tcp_seq_stop; + p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops); if (p) p->data = afinfo; @@ -2414,8 +2414,10 @@ static struct tcp_seq_afinfo tcp4_seq_afinfo = { .owner = THIS_MODULE, .name = "tcp", .family = AF_INET, - .seq_show = tcp4_seq_show, .seq_fops = &tcp4_seq_fops, + .seq_ops = { + .show = tcp4_seq_show, + }, }; static int tcp4_proc_init_net(struct net *net) diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 80eab71e77ff..8bf59ee51cdb 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2125,8 +2125,10 @@ static struct tcp_seq_afinfo tcp6_seq_afinfo = { .owner = THIS_MODULE, .name = "tcp6", .family = AF_INET6, - .seq_show = tcp6_seq_show, .seq_fops = &tcp6_seq_fops, + .seq_ops = { + .show = tcp6_seq_show, + }, }; int tcp6_proc_init(struct net *net) -- cgit v1.2.3 From 68fcadd16c371d5e0698ba366f33a4f990ce83ce Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Sun, 13 Apr 2008 22:13:30 -0700 Subject: [TCP]: Place file operations directly into tcp_seq_afinfo. No need to have separate never-used variable. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/tcp.h | 2 +- net/ipv4/tcp_ipv4.c | 15 ++++++--------- net/ipv6/tcp_ipv6.c | 2 -- 3 files changed, 7 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 2c9a650412c0..93479ebf3e33 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1328,7 +1328,7 @@ struct tcp_seq_afinfo { struct module *owner; char *name; sa_family_t family; - struct file_operations *seq_fops; + struct file_operations seq_fops; struct seq_operations seq_ops; }; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 840346b390d7..3696c83aec19 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2232,17 +2232,17 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) int rc = 0; struct proc_dir_entry *p; - afinfo->seq_fops->owner = afinfo->owner; - afinfo->seq_fops->open = tcp_seq_open; - afinfo->seq_fops->read = seq_read; - afinfo->seq_fops->llseek = seq_lseek; - afinfo->seq_fops->release = seq_release_net; + afinfo->seq_fops.owner = afinfo->owner; + afinfo->seq_fops.open = tcp_seq_open; + afinfo->seq_fops.read = seq_read; + afinfo->seq_fops.llseek = seq_lseek; + afinfo->seq_fops.release = seq_release_net; afinfo->seq_ops.start = tcp_seq_start; afinfo->seq_ops.next = tcp_seq_next; afinfo->seq_ops.stop = tcp_seq_stop; - p = proc_net_fops_create(net, afinfo->name, S_IRUGO, afinfo->seq_fops); + p = proc_net_fops_create(net, afinfo->name, S_IRUGO, &afinfo->seq_fops); if (p) p->data = afinfo; else @@ -2253,7 +2253,6 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) void tcp_proc_unregister(struct net *net, struct tcp_seq_afinfo *afinfo) { proc_net_remove(net, afinfo->name); - memset(afinfo->seq_fops, 0, sizeof(*afinfo->seq_fops)); } static void get_openreq4(struct sock *sk, struct request_sock *req, @@ -2382,12 +2381,10 @@ out: return 0; } -static struct file_operations tcp4_seq_fops; static struct tcp_seq_afinfo tcp4_seq_afinfo = { .owner = THIS_MODULE, .name = "tcp", .family = AF_INET, - .seq_fops = &tcp4_seq_fops, .seq_ops = { .show = tcp4_seq_show, }, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 8bf59ee51cdb..e33a3dc7a000 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2120,12 +2120,10 @@ out: return 0; } -static struct file_operations tcp6_seq_fops; static struct tcp_seq_afinfo tcp6_seq_afinfo = { .owner = THIS_MODULE, .name = "tcp6", .family = AF_INET6, - .seq_fops = &tcp6_seq_fops, .seq_ops = { .show = tcp6_seq_show, }, -- cgit v1.2.3 From 5f4472c5a640c9671ca5becaebdfd6e651482176 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Sun, 13 Apr 2008 22:13:53 -0700 Subject: [TCP]: Remove owner from tcp_seq_afinfo. Move it to tcp_seq_afinfo->seq_fops as should be. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/tcp.h | 1 - net/ipv4/tcp_ipv4.c | 5 +++-- net/ipv6/tcp_ipv6.c | 4 +++- 3 files changed, 6 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 93479ebf3e33..58d82822414d 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1325,7 +1325,6 @@ enum tcp_seq_states { }; struct tcp_seq_afinfo { - struct module *owner; char *name; sa_family_t family; struct file_operations seq_fops; diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 3696c83aec19..02519730e0d5 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2232,7 +2232,6 @@ int tcp_proc_register(struct net *net, struct tcp_seq_afinfo *afinfo) int rc = 0; struct proc_dir_entry *p; - afinfo->seq_fops.owner = afinfo->owner; afinfo->seq_fops.open = tcp_seq_open; afinfo->seq_fops.read = seq_read; afinfo->seq_fops.llseek = seq_lseek; @@ -2382,9 +2381,11 @@ out: } static struct tcp_seq_afinfo tcp4_seq_afinfo = { - .owner = THIS_MODULE, .name = "tcp", .family = AF_INET, + .seq_fops = { + .owner = THIS_MODULE, + }, .seq_ops = { .show = tcp4_seq_show, }, diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e33a3dc7a000..231c4dddfb8c 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -2121,9 +2121,11 @@ out: } static struct tcp_seq_afinfo tcp6_seq_afinfo = { - .owner = THIS_MODULE, .name = "tcp6", .family = AF_INET6, + .seq_fops = { + .owner = THIS_MODULE, + }, .seq_ops = { .show = tcp6_seq_show, }, -- cgit v1.2.3 From 67019cc9ee3f4868c8e5e493b2873c4722306019 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Sun, 13 Apr 2008 22:28:42 -0700 Subject: [NETNS]: Add an empty netns_dccp structure on struct net. According to the overall struct net design, it will be filled with DCCP-related members. Signed-off-by: Pavel Emelyanov Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/net_namespace.h | 4 ++++ include/net/netns/dccp.h | 7 +++++++ 2 files changed, 11 insertions(+) create mode 100644 include/net/netns/dccp.h (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 0ab62ed2fdef..e2aee2689abe 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -13,6 +13,7 @@ #include #include #include +#include #include struct proc_dir_entry; @@ -54,6 +55,9 @@ struct net { #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) struct netns_ipv6 ipv6; #endif +#if defined(CONFIG_IP_DCCP) || defined(CONFIG_IP_DCCP_MODULE) + struct netns_dccp dccp; +#endif #ifdef CONFIG_NETFILTER struct netns_xt xt; #endif diff --git a/include/net/netns/dccp.h b/include/net/netns/dccp.h new file mode 100644 index 000000000000..056e6adc70b6 --- /dev/null +++ b/include/net/netns/dccp.h @@ -0,0 +1,7 @@ +#ifndef __NETNS_DCCP_H__ +#define __NETNS_DCCP_H__ + +struct netns_dccp { +}; + +#endif -- cgit v1.2.3 From 7b1cffa8c90269dc3dc721d084d1e0d742d87c31 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Sun, 13 Apr 2008 22:29:37 -0700 Subject: [NETNS][DCCPV4]: Move the dccp_v4_ctl_sk on the struct net. And replace all its usage with init_net's socket. Signed-off-by: Pavel Emelyanov Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/netns/dccp.h | 3 +++ net/dccp/ipv4.c | 20 ++++++++++---------- 2 files changed, 13 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/net/netns/dccp.h b/include/net/netns/dccp.h index 056e6adc70b6..e4f16faa898a 100644 --- a/include/net/netns/dccp.h +++ b/include/net/netns/dccp.h @@ -1,7 +1,10 @@ #ifndef __NETNS_DCCP_H__ #define __NETNS_DCCP_H__ +struct sock; + struct netns_dccp { + struct sock *v4_ctl_sk; }; #endif diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c index 10bba039c4b8..85931dce744c 100644 --- a/net/dccp/ipv4.c +++ b/net/dccp/ipv4.c @@ -32,11 +32,10 @@ #include "feat.h" /* - * This is the global socket data structure used for responding to + * The dccp_ctl_sk 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 sock *dccp_v4_ctl_sk; int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) { @@ -506,6 +505,7 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) const struct iphdr *rxiph; struct sk_buff *skb; struct dst_entry *dst; + struct sock *ctl_sk = init_net.dccp.v4_ctl_sk; /* Never send a reset in response to a reset. */ if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET) @@ -514,11 +514,11 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) if (rxskb->rtable->rt_type != RTN_LOCAL) return; - dst = dccp_v4_route_skb(dccp_v4_ctl_sk, rxskb); + dst = dccp_v4_route_skb(ctl_sk, rxskb); if (dst == NULL) return; - skb = dccp_ctl_make_reset(dccp_v4_ctl_sk, rxskb); + skb = dccp_ctl_make_reset(ctl_sk, rxskb); if (skb == NULL) goto out; @@ -527,10 +527,10 @@ static void dccp_v4_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) rxiph->daddr); skb->dst = dst_clone(dst); - bh_lock_sock(dccp_v4_ctl_sk); - err = ip_build_and_send_pkt(skb, dccp_v4_ctl_sk, + bh_lock_sock(ctl_sk); + err = ip_build_and_send_pkt(skb, ctl_sk, rxiph->daddr, rxiph->saddr, NULL); - bh_unlock_sock(dccp_v4_ctl_sk); + bh_unlock_sock(ctl_sk); if (net_xmit_eval(err) == 0) { DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS); @@ -1016,7 +1016,7 @@ static int __init dccp_v4_init(void) inet_register_protosw(&dccp_v4_protosw); - err = inet_ctl_sock_create(&dccp_v4_ctl_sk, PF_INET, + err = inet_ctl_sock_create(&init_net.dccp.v4_ctl_sk, PF_INET, SOCK_DCCP, IPPROTO_DCCP, &init_net); if (err) goto out_unregister_protosw; @@ -1027,7 +1027,7 @@ static int __init dccp_v4_init(void) out: return err; out_destroy_ctl_sock: - inet_ctl_sock_destroy(dccp_v4_ctl_sk); + inet_ctl_sock_destroy(init_net.dccp.v4_ctl_sk); out_unregister_protosw: inet_unregister_protosw(&dccp_v4_protosw); inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP); @@ -1039,7 +1039,7 @@ out_proto_unregister: static void __exit dccp_v4_exit(void) { unregister_pernet_subsys(&dccp_v4_ops); - inet_ctl_sock_destroy(dccp_v4_ctl_sk); + inet_ctl_sock_destroy(init_net.dccp.v4_ctl_sk); inet_unregister_protosw(&dccp_v4_protosw); inet_del_protocol(&dccp_v4_protocol, IPPROTO_DCCP); proto_unregister(&dccp_v4_prot); -- cgit v1.2.3 From 0204774191d3e7bc69e3ae6bbf328b635607505a Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Sun, 13 Apr 2008 22:32:25 -0700 Subject: [NETNS][DCCPV6]: Move the dccp_v6_ctl_sk on the struct net. And replace all its usage with init_net's socket. Signed-off-by: Pavel Emelyanov Acked-by: Arnaldo Carvalho de Melo Signed-off-by: David S. Miller --- include/net/netns/dccp.h | 1 + net/dccp/ipv6.c | 16 ++++++++-------- 2 files changed, 9 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/netns/dccp.h b/include/net/netns/dccp.h index e4f16faa898a..98d2a7ce1f71 100644 --- a/include/net/netns/dccp.h +++ b/include/net/netns/dccp.h @@ -5,6 +5,7 @@ struct sock; struct netns_dccp { struct sock *v4_ctl_sk; + struct sock *v6_ctl_sk; }; #endif diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 2ca52913f4be..109dab3550f2 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -33,8 +33,7 @@ #include "ipv6.h" #include "feat.h" -/* Socket used for sending RSTs and ACKs */ -static struct sock *dccp_v6_ctl_sk; +/* dccp_v6_ctl_sk is used for sending RSTs and ACKs */ static struct inet_connection_sock_af_ops dccp_ipv6_mapped; static struct inet_connection_sock_af_ops dccp_ipv6_af_ops; @@ -296,6 +295,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) struct ipv6hdr *rxip6h; struct sk_buff *skb; struct flowi fl; + struct sock *ctl_sk = init_net.dccp.v6_ctl_sk; if (dccp_hdr(rxskb)->dccph_type == DCCP_PKT_RESET) return; @@ -303,7 +303,7 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) if (!ipv6_unicast_destination(rxskb)) return; - skb = dccp_ctl_make_reset(dccp_v6_ctl_sk, rxskb); + skb = dccp_ctl_make_reset(ctl_sk, rxskb); if (skb == NULL) return; @@ -322,9 +322,9 @@ static void dccp_v6_ctl_send_reset(struct sock *sk, struct sk_buff *rxskb) security_skb_classify_flow(rxskb, &fl); /* sk = NULL, but it is safe for now. RST socket required. */ - if (!ip6_dst_lookup(dccp_v6_ctl_sk, &skb->dst, &fl)) { + if (!ip6_dst_lookup(ctl_sk, &skb->dst, &fl)) { if (xfrm_lookup(&skb->dst, &fl, NULL, 0) >= 0) { - ip6_xmit(dccp_v6_ctl_sk, skb, &fl, NULL, 0); + ip6_xmit(ctl_sk, skb, &fl, NULL, 0); DCCP_INC_STATS_BH(DCCP_MIB_OUTSEGS); DCCP_INC_STATS_BH(DCCP_MIB_OUTRSTS); return; @@ -1198,7 +1198,7 @@ static int __init dccp_v6_init(void) inet6_register_protosw(&dccp_v6_protosw); - err = inet_ctl_sock_create(&dccp_v6_ctl_sk, PF_INET6, + err = inet_ctl_sock_create(&init_net.dccp.v6_ctl_sk, PF_INET6, SOCK_DCCP, IPPROTO_DCCP, &init_net); if (err != 0) goto out_unregister_protosw; @@ -1210,7 +1210,7 @@ out: return err; out_destroy_ctl_sock: - inet_ctl_sock_destroy(dccp_v6_ctl_sk); + inet_ctl_sock_destroy(init_net.dccp.v6_ctl_sk); out_unregister_protosw: inet6_del_protocol(&dccp_v6_protocol, IPPROTO_DCCP); inet6_unregister_protosw(&dccp_v6_protosw); @@ -1222,7 +1222,7 @@ out_unregister_proto: static void __exit dccp_v6_exit(void) { unregister_pernet_subsys(&dccp_v6_ops); - inet_ctl_sock_destroy(dccp_v6_ctl_sk); + inet_ctl_sock_destroy(init_net.dccp.v6_ctl_sk); inet6_del_protocol(&dccp_v6_protocol, IPPROTO_DCCP); inet6_unregister_protosw(&dccp_v6_protosw); proto_unregister(&dccp_v6_prot); -- cgit v1.2.3 From cee8947338d46bccece54c752bf6cd4043035f05 Mon Sep 17 00:00:00 2001 From: YOSHIFUJI Hideaki Date: Sun, 13 Apr 2008 23:21:16 -0700 Subject: [IPV6] MROUTE: Do not call ipv6_find_idev() directly. Since NETDEV_REGISTER notifier chain is responsible for creating inet6_dev{}, we do not need to call ipv6_find_idev() directly here. Signed-off-by: YOSHIFUJI Hideaki Signed-off-by: David S. Miller --- include/linux/mroute6.h | 3 --- net/ipv6/addrconf.c | 2 +- net/ipv6/ip6mr.c | 5 ----- 3 files changed, 1 insertion(+), 9 deletions(-) (limited to 'include') diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h index f6469fb90840..e7989593142b 100644 --- a/include/linux/mroute6.h +++ b/include/linux/mroute6.h @@ -117,9 +117,6 @@ struct sioc_mif_req6 #include /* for struct sk_buff_head */ -struct net_device; -struct inet6_dev *ipv6_find_idev(struct net_device *dev); - #ifdef CONFIG_IPV6_MROUTE static inline int ip6_mroute_opt(int opt) { diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index e93fa62089f8..9d49ed2578d7 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -412,7 +412,7 @@ static struct inet6_dev * ipv6_add_dev(struct net_device *dev) return ndev; } -struct inet6_dev * ipv6_find_idev(struct net_device *dev) +static struct inet6_dev * ipv6_find_idev(struct net_device *dev) { struct inet6_dev *idev; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 94ede696da2a..6e2e3c957a31 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -435,7 +435,6 @@ static void reg_vif_setup(struct net_device *dev) static struct net_device *ip6mr_reg_vif(void) { struct net_device *dev; - struct inet6_dev *in_dev; dev = alloc_netdev(sizeof(struct net_device_stats), "pim6reg", reg_vif_setup); @@ -449,10 +448,6 @@ static struct net_device *ip6mr_reg_vif(void) } dev->iflink = 0; - in_dev = ipv6_find_idev(dev); - if (!in_dev) - goto failure; - if (dev_open(dev)) goto failure; -- cgit v1.2.3 From f525c06d12b72cddb085df7f6f348c3c5a39b3ce Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 14 Apr 2008 00:04:12 -0700 Subject: [SKB]: __skb_dequeue = skb_peek + __skb_unlink By rearranging the order of declarations, __skb_dequeue() is expressed in terms of * skb_peek() and * __skb_unlink(), thus in effect mirroring the analogue implementation of __skb_dequeue_tail(). Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- include/linux/skbuff.h | 47 ++++++++++++++++------------------------------- 1 file changed, 16 insertions(+), 31 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index e517701c25ba..c2116200580a 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -734,35 +734,6 @@ static inline void __skb_queue_tail(struct sk_buff_head *list, next->prev = prev->next = newsk; } - -/** - * __skb_dequeue - remove from the head of the queue - * @list: list to dequeue from - * - * Remove the head of the list. This function does not take any locks - * so must be used with appropriate locks held only. The head item is - * returned or %NULL if the list is empty. - */ -extern struct sk_buff *skb_dequeue(struct sk_buff_head *list); -static inline struct sk_buff *__skb_dequeue(struct sk_buff_head *list) -{ - struct sk_buff *next, *prev, *result; - - prev = (struct sk_buff *) list; - next = prev->next; - result = NULL; - if (next != prev) { - result = next; - next = next->next; - list->qlen--; - next->prev = prev; - prev->next = next; - result->next = result->prev = NULL; - } - return result; -} - - /* * Insert a packet on a list. */ @@ -803,8 +774,22 @@ static inline void __skb_unlink(struct sk_buff *skb, struct sk_buff_head *list) prev->next = next; } - -/* XXX: more streamlined implementation */ +/** + * __skb_dequeue - remove from the head of the queue + * @list: list to dequeue from + * + * Remove the head of the list. This function does not take any locks + * so must be used with appropriate locks held only. The head item is + * returned or %NULL if the list is empty. + */ +extern struct sk_buff *skb_dequeue(struct sk_buff_head *list); +static inline struct sk_buff *__skb_dequeue(struct sk_buff_head *list) +{ + struct sk_buff *skb = skb_peek(list); + if (skb) + __skb_unlink(skb, list); + return skb; +} /** * __skb_dequeue_tail - remove from the tail of the queue -- cgit v1.2.3 From bf299275882624b1908521ee8074df85160e9679 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 14 Apr 2008 00:04:51 -0700 Subject: [SKB]: __skb_queue_after(prev) = __skb_insert(prev, prev->next) By reordering, __skb_queue_after() is expressed in terms of __skb_insert(). Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- include/linux/skbuff.h | 34 ++++++++++++---------------------- 1 file changed, 12 insertions(+), 22 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index c2116200580a..bb107ab675fc 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -663,11 +663,21 @@ static inline void skb_queue_head_init_class(struct sk_buff_head *list, } /* - * Insert an sk_buff at the start of a list. + * Insert an sk_buff on a list. * * The "__skb_xxxx()" functions are the non-atomic ones that * can only be called with interrupts disabled. */ +extern void skb_insert(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list); +static inline void __skb_insert(struct sk_buff *newsk, + struct sk_buff *prev, struct sk_buff *next, + struct sk_buff_head *list) +{ + newsk->next = next; + newsk->prev = prev; + next->prev = prev->next = newsk; + list->qlen++; +} /** * __skb_queue_after - queue a buffer at the list head @@ -684,13 +694,7 @@ static inline void __skb_queue_after(struct sk_buff_head *list, struct sk_buff *prev, struct sk_buff *newsk) { - struct sk_buff *next; - list->qlen++; - - next = prev->next; - newsk->next = next; - newsk->prev = prev; - next->prev = prev->next = newsk; + __skb_insert(newsk, prev, prev->next, list); } /** @@ -734,20 +738,6 @@ static inline void __skb_queue_tail(struct sk_buff_head *list, next->prev = prev->next = newsk; } -/* - * Insert a packet on a list. - */ -extern void skb_insert(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list); -static inline void __skb_insert(struct sk_buff *newsk, - struct sk_buff *prev, struct sk_buff *next, - struct sk_buff_head *list) -{ - newsk->next = next; - newsk->prev = prev; - next->prev = prev->next = newsk; - list->qlen++; -} - /* * Place a packet after a given packet in a list. */ -- cgit v1.2.3 From 7de6c033367ab86f39c7723392caf73325cbf286 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 14 Apr 2008 00:05:09 -0700 Subject: [SKB]: __skb_append = __skb_queue_after This expresses __skb_append in terms of __skb_queue_after, exploiting that __skb_append(old, new, list) = __skb_queue_after(list, old, new). Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- include/linux/skbuff.h | 12 +++--------- include/net/tcp.h | 2 +- net/core/skbuff.c | 2 +- net/ipv4/tcp_input.c | 2 +- 4 files changed, 6 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index bb107ab675fc..83c851846829 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -697,6 +697,9 @@ static inline void __skb_queue_after(struct sk_buff_head *list, __skb_insert(newsk, prev, prev->next, list); } +extern void skb_append(struct sk_buff *old, struct sk_buff *newsk, + struct sk_buff_head *list); + /** * __skb_queue_head - queue a buffer at the list head * @list: list to use @@ -738,15 +741,6 @@ static inline void __skb_queue_tail(struct sk_buff_head *list, next->prev = prev->next = newsk; } -/* - * Place a packet after a given packet in a list. - */ -extern void skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list); -static inline void __skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list) -{ - __skb_insert(newsk, old, old->next, list); -} - /* * remove sk_buff from list. _Must_ be called atomically, and with * the list known.. diff --git a/include/net/tcp.h b/include/net/tcp.h index 58d82822414d..2ab350eca02e 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1247,7 +1247,7 @@ static inline void tcp_insert_write_queue_after(struct sk_buff *skb, struct sk_buff *buff, struct sock *sk) { - __skb_append(skb, buff, &sk->sk_write_queue); + __skb_queue_after(&sk->sk_write_queue, skb, buff); } /* Insert skb between prev and next on the write queue of sk. */ diff --git a/net/core/skbuff.c b/net/core/skbuff.c index e4259215607f..4cd12d99b12e 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1860,7 +1860,7 @@ void skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head unsigned long flags; spin_lock_irqsave(&list->lock, flags); - __skb_append(old, newsk, list); + __skb_queue_after(list, old, newsk); spin_unlock_irqrestore(&list->lock, flags); } diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 6e46b4c0f28c..743611956045 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -3968,7 +3968,7 @@ drop: u32 end_seq = TCP_SKB_CB(skb)->end_seq; if (seq == TCP_SKB_CB(skb1)->end_seq) { - __skb_append(skb1, skb, &tp->out_of_order_queue); + __skb_queue_after(&tp->out_of_order_queue, skb1, skb); if (!tp->rx_opt.num_sacks || tp->selective_acks[0].end_seq != seq) -- cgit v1.2.3 From f5572855ec492334d8c3ec0e0e86c31865d5cf07 Mon Sep 17 00:00:00 2001 From: Gerrit Renker Date: Mon, 14 Apr 2008 00:05:28 -0700 Subject: [SKB]: __skb_queue_tail = __skb_insert before This expresses __skb_queue_tail() in terms of __skb_insert(), using __skb_insert_before() as auxiliary function. Signed-off-by: Gerrit Renker Signed-off-by: David S. Miller --- include/linux/skbuff.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 83c851846829..11fd9f2c4093 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -700,6 +700,13 @@ static inline void __skb_queue_after(struct sk_buff_head *list, extern void skb_append(struct sk_buff *old, struct sk_buff *newsk, struct sk_buff_head *list); +static inline void __skb_queue_before(struct sk_buff_head *list, + struct sk_buff *next, + struct sk_buff *newsk) +{ + __skb_insert(newsk, next->prev, next, list); +} + /** * __skb_queue_head - queue a buffer at the list head * @list: list to use @@ -731,14 +738,7 @@ extern void skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk); static inline void __skb_queue_tail(struct sk_buff_head *list, struct sk_buff *newsk) { - struct sk_buff *prev, *next; - - list->qlen++; - next = (struct sk_buff *)list; - prev = next->prev; - newsk->next = next; - newsk->prev = prev; - next->prev = prev->next = newsk; + __skb_queue_before(list, (struct sk_buff *)list, newsk); } /* -- cgit v1.2.3 From 666953df353194bef76086fa3f126241cbac3e3a Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Mon, 14 Apr 2008 09:56:02 +0200 Subject: [NETFILTER]: ip_tables: per-netns FILTER/MANGLE/RAW tables for real Commit 9335f047fe61587ec82ff12fbb1220bcfdd32006 aka "[NETFILTER]: ip_tables: per-netns FILTER, MANGLE, RAW" added per-netns _view_ of iptables rules. They were shown to user, but ignored by filtering code. Now that it's possible to at least ping loopback, per-netns tables can affect filtering decisions. netns is taken in case of PRE_ROUTING, LOCAL_IN -- from in device, POST_ROUTING, LOCAL_OUT -- from out device, FORWARD -- from in device which should be equal to out device's netns. This code is relatively new, so BUG_ON was plugged. Wrappers were added to a) keep code the same from CONFIG_NET_NS=n users (overwhelming majority), b) consolidate code in one place -- similar changes will be done in ipv6 and arp netfilter code. Signed-off-by: Alexey Dobriyan Signed-off-by: Patrick McHardy --- include/linux/netfilter.h | 54 ++++++++++++++++++++++++++++++++++++- net/ipv4/netfilter/iptable_filter.c | 19 ++++++++++--- net/ipv4/netfilter/iptable_mangle.c | 49 ++++++++++++++++++++++++++++----- net/ipv4/netfilter/iptable_raw.c | 6 +++-- 4 files changed, 115 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 89e6c72ad295..66bc52060fd6 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -6,11 +6,13 @@ #include #include #include +#include #include #include #include #include #include +#include #endif #include @@ -76,7 +78,6 @@ extern void netfilter_init(void); #define NF_MAX_HOOKS 8 struct sk_buff; -struct net_device; typedef unsigned int nf_hookfn(unsigned int hooknum, struct sk_buff *skb, @@ -320,5 +321,56 @@ extern void (*nf_ct_destroy)(struct nf_conntrack *); static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {} #endif +static inline struct net *nf_pre_routing_net(const struct net_device *in, + const struct net_device *out) +{ +#ifdef CONFIG_NET_NS + return in->nd_net; +#else + return &init_net; +#endif +} + +static inline struct net *nf_local_in_net(const struct net_device *in, + const struct net_device *out) +{ +#ifdef CONFIG_NET_NS + return in->nd_net; +#else + return &init_net; +#endif +} + +static inline struct net *nf_forward_net(const struct net_device *in, + const struct net_device *out) +{ +#ifdef CONFIG_NET_NS + BUG_ON(in->nd_net != out->nd_net); + return in->nd_net; +#else + return &init_net; +#endif +} + +static inline struct net *nf_local_out_net(const struct net_device *in, + const struct net_device *out) +{ +#ifdef CONFIG_NET_NS + return out->nd_net; +#else + return &init_net; +#endif +} + +static inline struct net *nf_post_routing_net(const struct net_device *in, + const struct net_device *out) +{ +#ifdef CONFIG_NET_NS + return out->nd_net; +#else + return &init_net; +#endif +} + #endif /*__KERNEL__*/ #endif /*__LINUX_NETFILTER_H*/ diff --git a/net/ipv4/netfilter/iptable_filter.c b/net/ipv4/netfilter/iptable_filter.c index 69f3d7e6e96f..7fcf60adbbed 100644 --- a/net/ipv4/netfilter/iptable_filter.c +++ b/net/ipv4/netfilter/iptable_filter.c @@ -62,6 +62,17 @@ static struct xt_table packet_filter = { }; /* The work comes in here from netfilter.c. */ +static unsigned int +ipt_local_in_hook(unsigned int hook, + struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ipt_do_table(skb, hook, in, out, + nf_local_in_net(in, out)->ipv4.iptable_filter); +} + static unsigned int ipt_hook(unsigned int hook, struct sk_buff *skb, @@ -69,7 +80,8 @@ ipt_hook(unsigned int hook, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_filter); + return ipt_do_table(skb, hook, in, out, + nf_forward_net(in, out)->ipv4.iptable_filter); } static unsigned int @@ -88,12 +100,13 @@ ipt_local_out_hook(unsigned int hook, return NF_ACCEPT; } - return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_filter); + return ipt_do_table(skb, hook, in, out, + nf_local_out_net(in, out)->ipv4.iptable_filter); } static struct nf_hook_ops ipt_ops[] __read_mostly = { { - .hook = ipt_hook, + .hook = ipt_local_in_hook, .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_INET_LOCAL_IN, diff --git a/net/ipv4/netfilter/iptable_mangle.c b/net/ipv4/netfilter/iptable_mangle.c index c55a210853a7..ba827035b691 100644 --- a/net/ipv4/netfilter/iptable_mangle.c +++ b/net/ipv4/netfilter/iptable_mangle.c @@ -74,13 +74,47 @@ static struct xt_table packet_mangler = { /* The work comes in here from netfilter.c. */ static unsigned int -ipt_route_hook(unsigned int hook, +ipt_pre_routing_hook(unsigned int hook, + struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ipt_do_table(skb, hook, in, out, + nf_pre_routing_net(in, out)->ipv4.iptable_mangle); +} + +static unsigned int +ipt_post_routing_hook(unsigned int hook, + struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ipt_do_table(skb, hook, in, out, + nf_post_routing_net(in, out)->ipv4.iptable_mangle); +} + +static unsigned int +ipt_local_in_hook(unsigned int hook, + struct sk_buff *skb, + const struct net_device *in, + const struct net_device *out, + int (*okfn)(struct sk_buff *)) +{ + return ipt_do_table(skb, hook, in, out, + nf_local_in_net(in, out)->ipv4.iptable_mangle); +} + +static unsigned int +ipt_forward_hook(unsigned int hook, struct sk_buff *skb, const struct net_device *in, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_mangle); + return ipt_do_table(skb, hook, in, out, + nf_forward_net(in, out)->ipv4.iptable_mangle); } static unsigned int @@ -112,7 +146,8 @@ ipt_local_hook(unsigned int hook, daddr = iph->daddr; tos = iph->tos; - ret = ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_mangle); + ret = ipt_do_table(skb, hook, in, out, + nf_local_out_net(in, out)->ipv4.iptable_mangle); /* Reroute for ANY change. */ if (ret != NF_DROP && ret != NF_STOLEN && ret != NF_QUEUE) { iph = ip_hdr(skb); @@ -130,21 +165,21 @@ ipt_local_hook(unsigned int hook, static struct nf_hook_ops ipt_ops[] __read_mostly = { { - .hook = ipt_route_hook, + .hook = ipt_pre_routing_hook, .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_INET_PRE_ROUTING, .priority = NF_IP_PRI_MANGLE, }, { - .hook = ipt_route_hook, + .hook = ipt_local_in_hook, .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_INET_LOCAL_IN, .priority = NF_IP_PRI_MANGLE, }, { - .hook = ipt_route_hook, + .hook = ipt_forward_hook, .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_INET_FORWARD, @@ -158,7 +193,7 @@ static struct nf_hook_ops ipt_ops[] __read_mostly = { .priority = NF_IP_PRI_MANGLE, }, { - .hook = ipt_route_hook, + .hook = ipt_post_routing_hook, .owner = THIS_MODULE, .pf = PF_INET, .hooknum = NF_INET_POST_ROUTING, diff --git a/net/ipv4/netfilter/iptable_raw.c b/net/ipv4/netfilter/iptable_raw.c index e41fe8ca4e1c..4b689742d58b 100644 --- a/net/ipv4/netfilter/iptable_raw.c +++ b/net/ipv4/netfilter/iptable_raw.c @@ -52,7 +52,8 @@ ipt_hook(unsigned int hook, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_raw); + return ipt_do_table(skb, hook, in, out, + nf_pre_routing_net(in, out)->ipv4.iptable_raw); } static unsigned int @@ -70,7 +71,8 @@ ipt_local_hook(unsigned int hook, "packet.\n"); return NF_ACCEPT; } - return ipt_do_table(skb, hook, in, out, init_net.ipv4.iptable_raw); + return ipt_do_table(skb, hook, in, out, + nf_local_out_net(in, out)->ipv4.iptable_raw); } /* 'raw' is the very first table. */ -- cgit v1.2.3 From b9f61b160336da5eaaacb0cb41ebe32169e3bde5 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 09:56:04 +0200 Subject: [NETFILTER]: xt_sctp: simplify xt_sctp.h The use of xt_sctp.h flagged up -Wshadow warnings in userspace, which prompted me to look at it and clean it up. Basic operations have been directly replaced by library calls (memcpy, memset is both available in the kernel and userspace, and usually faster than a self-made loop). The is_set and is_clear functions now use a processing time shortcut, too. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter/xt_sctp.h | 84 ++++++++++++++++----------------------- 1 file changed, 35 insertions(+), 49 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/xt_sctp.h b/include/linux/netfilter/xt_sctp.h index dd5a4fd4cfd3..32000ba6ecef 100644 --- a/include/linux/netfilter/xt_sctp.h +++ b/include/linux/netfilter/xt_sctp.h @@ -37,68 +37,54 @@ struct xt_sctp_info { #define SCTP_CHUNKMAP_SET(chunkmap, type) \ do { \ - chunkmap[type / bytes(u_int32_t)] |= \ + (chunkmap)[type / bytes(u_int32_t)] |= \ 1 << (type % bytes(u_int32_t)); \ } while (0) #define SCTP_CHUNKMAP_CLEAR(chunkmap, type) \ do { \ - chunkmap[type / bytes(u_int32_t)] &= \ + (chunkmap)[type / bytes(u_int32_t)] &= \ ~(1 << (type % bytes(u_int32_t))); \ } while (0) #define SCTP_CHUNKMAP_IS_SET(chunkmap, type) \ ({ \ - (chunkmap[type / bytes (u_int32_t)] & \ + ((chunkmap)[type / bytes (u_int32_t)] & \ (1 << (type % bytes (u_int32_t)))) ? 1: 0; \ }) -#define SCTP_CHUNKMAP_RESET(chunkmap) \ - do { \ - int i; \ - for (i = 0; i < ARRAY_SIZE(chunkmap); i++) \ - chunkmap[i] = 0; \ - } while (0) - -#define SCTP_CHUNKMAP_SET_ALL(chunkmap) \ - do { \ - int i; \ - for (i = 0; i < ARRAY_SIZE(chunkmap); i++) \ - chunkmap[i] = ~0; \ - } while (0) - -#define SCTP_CHUNKMAP_COPY(destmap, srcmap) \ - do { \ - int i; \ - for (i = 0; i < ARRAY_SIZE(srcmap); i++) \ - destmap[i] = srcmap[i]; \ - } while (0) - -#define SCTP_CHUNKMAP_IS_CLEAR(chunkmap) \ -({ \ - int i; \ - int flag = 1; \ - for (i = 0; i < ARRAY_SIZE(chunkmap); i++) { \ - if (chunkmap[i]) { \ - flag = 0; \ - break; \ - } \ - } \ - flag; \ -}) - -#define SCTP_CHUNKMAP_IS_ALL_SET(chunkmap) \ -({ \ - int i; \ - int flag = 1; \ - for (i = 0; i < ARRAY_SIZE(chunkmap); i++) { \ - if (chunkmap[i] != ~0) { \ - flag = 0; \ - break; \ - } \ - } \ - flag; \ -}) +#define SCTP_CHUNKMAP_RESET(chunkmap) \ + memset((chunkmap), 0, sizeof(chunkmap)) + +#define SCTP_CHUNKMAP_SET_ALL(chunkmap) \ + memset((chunkmap), ~0U, sizeof(chunkmap)) + +#define SCTP_CHUNKMAP_COPY(destmap, srcmap) \ + memcpy((destmap), (srcmap), sizeof(srcmap)) + +#define SCTP_CHUNKMAP_IS_CLEAR(chunkmap) \ + __sctp_chunkmap_is_clear((chunkmap), ARRAY_SIZE(chunkmap)) +static inline bool +__sctp_chunkmap_is_clear(const u_int32_t *chunkmap, unsigned int n) +{ + unsigned int i; + for (i = 0; i < n; ++i) + if (chunkmap[i]) + return false; + return true; +} + +#define SCTP_CHUNKMAP_IS_ALL_SET(chunkmap) \ + __sctp_chunkmap_is_all_set((chunkmap), ARRAY_SIZE(chunkmap)) +static inline bool +__sctp_chunkmap_is_all_set(const u_int32_t *chunkmap, unsigned int n) +{ + unsigned int i; + for (i = 0; i < n; ++i) + if (chunkmap[i] != ~0U) + return false; + return true; +} #endif /* _XT_SCTP_H_ */ -- cgit v1.2.3 From 5452e425adfdfc4647b618e303f73d48f2405b0e Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:35 +0200 Subject: [NETFILTER]: annotate {arp,ip,ip6,x}tables with const Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 4 ++-- net/ipv4/netfilter/arp_tables.c | 31 ++++++++++++++++--------------- net/ipv4/netfilter/arpt_mangle.c | 2 +- net/ipv4/netfilter/ip_tables.c | 31 ++++++++++++++++--------------- net/ipv6/netfilter/ip6_tables.c | 29 +++++++++++++++-------------- net/netfilter/x_tables.c | 18 +++++++++--------- 6 files changed, 59 insertions(+), 56 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index b2c62cc618f5..2326296b6f25 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -430,13 +430,13 @@ extern int xt_compat_add_offset(int af, unsigned int offset, short delta); extern void xt_compat_flush_offsets(int af); extern short xt_compat_calc_jump(int af, unsigned int offset); -extern int xt_compat_match_offset(struct xt_match *match); +extern int xt_compat_match_offset(const struct xt_match *match); extern int xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr, unsigned int *size); extern int xt_compat_match_to_user(struct xt_entry_match *m, void __user **dstptr, unsigned int *size); -extern int xt_compat_target_offset(struct xt_target *target); +extern int xt_compat_target_offset(const struct xt_target *target); extern void xt_compat_target_from_user(struct xt_entry_target *t, void **dstptr, unsigned int *size); extern int xt_compat_target_to_user(struct xt_entry_target *t, diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 1563f29b5117..10cc442330c3 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -59,7 +59,7 @@ do { \ #endif static inline int arp_devaddr_compare(const struct arpt_devaddr_info *ap, - char *hdr_addr, int len) + const char *hdr_addr, int len) { int i, ret; @@ -80,8 +80,8 @@ static inline int arp_packet_match(const struct arphdr *arphdr, const char *outdev, const struct arpt_arp *arpinfo) { - char *arpptr = (char *)(arphdr + 1); - char *src_devaddr, *tgt_devaddr; + const char *arpptr = (char *)(arphdr + 1); + const char *src_devaddr, *tgt_devaddr; __be32 src_ipaddr, tgt_ipaddr; int i, ret; @@ -226,12 +226,12 @@ unsigned int arpt_do_table(struct sk_buff *skb, { static const char nulldevname[IFNAMSIZ]; unsigned int verdict = NF_DROP; - struct arphdr *arp; + const struct arphdr *arp; bool hotdrop = false; struct arpt_entry *e, *back; const char *indev, *outdev; void *table_base; - struct xt_table_info *private; + const struct xt_table_info *private; if (!pskb_may_pull(skb, arp_hdr_len(skb->dev))) return NF_DROP; @@ -352,7 +352,7 @@ static int mark_source_chains(struct xt_table_info *newinfo, e->counters.pcnt = pos; for (;;) { - struct arpt_standard_target *t + const struct arpt_standard_target *t = (void *)arpt_get_target(e); int visited = e->comefrom & (1 << hook); @@ -437,7 +437,7 @@ static int mark_source_chains(struct xt_table_info *newinfo, static inline int check_entry(struct arpt_entry *e, const char *name) { - struct arpt_entry_target *t; + const struct arpt_entry_target *t; if (!arp_checkentry(&e->arp)) { duprintf("arp_tables: arp check failed %p %s.\n", e, name); @@ -710,7 +710,7 @@ static inline struct xt_counters *alloc_counters(struct arpt_table *table) { unsigned int countersize; struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; /* We need atomic snapshot of counters: rest doesn't change * (other than comefrom, which userspace doesn't care @@ -737,7 +737,7 @@ static int copy_entries_to_user(unsigned int total_size, unsigned int off, num; struct arpt_entry *e; struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; int ret = 0; void *loc_cpu_entry; @@ -872,7 +872,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat) "arptable_%s", name); if (t && !IS_ERR(t)) { struct arpt_getinfo info; - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; #ifdef CONFIG_COMPAT if (compat) { @@ -927,7 +927,8 @@ static int get_entries(struct net *net, struct arpt_get_entries __user *uptr, t = xt_find_table_lock(net, NF_ARP, get.name); if (t && !IS_ERR(t)) { - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; + duprintf("t->private->number = %u\n", private->number); if (get.size == private->size) @@ -1087,11 +1088,11 @@ static int do_add_counters(struct net *net, void __user *user, unsigned int len, struct xt_counters_info tmp; struct xt_counters *paddc; unsigned int num_counters; - char *name; + const char *name; int size; void *ptmp; struct arpt_table *t; - struct xt_table_info *private; + const struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; #ifdef CONFIG_COMPAT @@ -1558,7 +1559,7 @@ static int compat_copy_entries_to_user(unsigned int total_size, void __user *userptr) { struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; void __user *pos; unsigned int size; int ret = 0; @@ -1609,7 +1610,7 @@ static int compat_get_entries(struct net *net, xt_compat_lock(NF_ARP); t = xt_find_table_lock(net, NF_ARP, get.name); if (t && !IS_ERR(t)) { - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; struct xt_table_info info; duprintf("t->private->number = %u\n", private->number); diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index 3f4222b0a803..3e732c827fc2 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -15,7 +15,7 @@ target(struct sk_buff *skb, const void *targinfo) { const struct arpt_mangle *mangle = targinfo; - struct arphdr *arp; + const struct arphdr *arp; unsigned char *arpptr; int pln, hln; diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index a819d191e1aa..aa124b50cb4a 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -296,7 +296,7 @@ static void trace_packet(struct sk_buff *skb, struct ipt_entry *e) { void *table_base; - struct ipt_entry *root; + const struct ipt_entry *root; char *hookname, *chainname, *comment; unsigned int rulenum = 0; @@ -327,7 +327,7 @@ ipt_do_table(struct sk_buff *skb, { static const char nulldevname[IFNAMSIZ] __attribute__((aligned(sizeof(long)))); u_int16_t offset; - struct iphdr *ip; + const struct iphdr *ip; u_int16_t datalen; bool hotdrop = false; /* Initializing verdict to NF_DROP keeps gcc happy. */ @@ -926,7 +926,7 @@ static struct xt_counters * alloc_counters(struct xt_table *table) { unsigned int countersize; struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; /* We need atomic snapshot of counters: rest doesn't change (other than comefrom, which userspace doesn't care @@ -953,9 +953,9 @@ copy_entries_to_user(unsigned int total_size, unsigned int off, num; struct ipt_entry *e; struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; int ret = 0; - void *loc_cpu_entry; + const void *loc_cpu_entry; counters = alloc_counters(table); if (IS_ERR(counters)) @@ -975,8 +975,8 @@ copy_entries_to_user(unsigned int total_size, /* ... then go back and fix counters and names */ for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){ unsigned int i; - struct ipt_entry_match *m; - struct ipt_entry_target *t; + const struct ipt_entry_match *m; + const struct ipt_entry_target *t; e = (struct ipt_entry *)(loc_cpu_entry + off); if (copy_to_user(userptr + off @@ -1116,7 +1116,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat) "iptable_%s", name); if (t && !IS_ERR(t)) { struct ipt_getinfo info; - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; #ifdef CONFIG_COMPAT if (compat) { @@ -1172,7 +1172,7 @@ get_entries(struct net *net, struct ipt_get_entries __user *uptr, int *len) t = xt_find_table_lock(net, AF_INET, get.name); if (t && !IS_ERR(t)) { - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; duprintf("t->private->number = %u\n", private->number); if (get.size == private->size) ret = copy_entries_to_user(private->size, @@ -1337,11 +1337,11 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, int compat struct xt_counters_info tmp; struct xt_counters *paddc; unsigned int num_counters; - char *name; + const char *name; int size; void *ptmp; struct xt_table *t; - struct xt_table_info *private; + const struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; #ifdef CONFIG_COMPAT @@ -1878,11 +1878,11 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, void __user *userptr) { struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; void __user *pos; unsigned int size; int ret = 0; - void *loc_cpu_entry; + const void *loc_cpu_entry; unsigned int i = 0; counters = alloc_counters(table); @@ -1929,7 +1929,7 @@ compat_get_entries(struct net *net, struct compat_ipt_get_entries __user *uptr, xt_compat_lock(AF_INET); t = xt_find_table_lock(net, AF_INET, get.name); if (t && !IS_ERR(t)) { - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; struct xt_table_info info; duprintf("t->private->number = %u\n", private->number); ret = compat_table_info(private, &info); @@ -2130,7 +2130,8 @@ icmp_match(const struct sk_buff *skb, unsigned int protoff, bool *hotdrop) { - struct icmphdr _icmph, *ic; + const struct icmphdr *ic; + struct icmphdr _icmph; const struct ipt_icmp *icmpinfo = matchinfo; /* Must not be a fragment. */ diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 70ef0d276cc0..782183f63366 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -325,7 +325,7 @@ static void trace_packet(struct sk_buff *skb, struct ip6t_entry *e) { void *table_base; - struct ip6t_entry *root; + const struct ip6t_entry *root; char *hookname, *chainname, *comment; unsigned int rulenum = 0; @@ -952,7 +952,7 @@ static struct xt_counters *alloc_counters(struct xt_table *table) { unsigned int countersize; struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; /* We need atomic snapshot of counters: rest doesn't change (other than comefrom, which userspace doesn't care @@ -979,9 +979,9 @@ copy_entries_to_user(unsigned int total_size, unsigned int off, num; struct ip6t_entry *e; struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; int ret = 0; - void *loc_cpu_entry; + const void *loc_cpu_entry; counters = alloc_counters(table); if (IS_ERR(counters)) @@ -1001,8 +1001,8 @@ copy_entries_to_user(unsigned int total_size, /* ... then go back and fix counters and names */ for (off = 0, num = 0; off < total_size; off += e->next_offset, num++){ unsigned int i; - struct ip6t_entry_match *m; - struct ip6t_entry_target *t; + const struct ip6t_entry_match *m; + const struct ip6t_entry_target *t; e = (struct ip6t_entry *)(loc_cpu_entry + off); if (copy_to_user(userptr + off @@ -1142,7 +1142,7 @@ static int get_info(struct net *net, void __user *user, int *len, int compat) "ip6table_%s", name); if (t && !IS_ERR(t)) { struct ip6t_getinfo info; - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; #ifdef CONFIG_COMPAT if (compat) { @@ -1225,7 +1225,7 @@ __do_replace(struct net *net, const char *name, unsigned int valid_hooks, struct xt_table *t; struct xt_table_info *oldinfo; struct xt_counters *counters; - void *loc_cpu_old_entry; + const void *loc_cpu_old_entry; ret = 0; counters = vmalloc_node(num_counters * sizeof(struct xt_counters), @@ -1369,9 +1369,9 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, int size; void *ptmp; struct xt_table *t; - struct xt_table_info *private; + const struct xt_table_info *private; int ret = 0; - void *loc_cpu_entry; + const void *loc_cpu_entry; #ifdef CONFIG_COMPAT struct compat_xt_counters_info compat_tmp; @@ -1905,11 +1905,11 @@ compat_copy_entries_to_user(unsigned int total_size, struct xt_table *table, void __user *userptr) { struct xt_counters *counters; - struct xt_table_info *private = table->private; + const struct xt_table_info *private = table->private; void __user *pos; unsigned int size; int ret = 0; - void *loc_cpu_entry; + const void *loc_cpu_entry; unsigned int i = 0; counters = alloc_counters(table); @@ -1956,7 +1956,7 @@ compat_get_entries(struct net *net, struct compat_ip6t_get_entries __user *uptr, xt_compat_lock(AF_INET6); t = xt_find_table_lock(net, AF_INET6, get.name); if (t && !IS_ERR(t)) { - struct xt_table_info *private = t->private; + const struct xt_table_info *private = t->private; struct xt_table_info info; duprintf("t->private->number = %u\n", private->number); ret = compat_table_info(private, &info); @@ -2155,7 +2155,8 @@ icmp6_match(const struct sk_buff *skb, unsigned int protoff, bool *hotdrop) { - struct icmp6hdr _icmph, *ic; + const struct icmp6hdr *ic; + struct icmp6hdr _icmph; const struct ip6t_icmp *icmpinfo = matchinfo; /* Must not be a fragment. */ diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index 0bd95680a494..f52f7f810ac4 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -58,7 +58,7 @@ static struct xt_af *xt; #define duprintf(format, args...) #endif -static const char *xt_prefix[NPROTO] = { +static const char *const xt_prefix[NPROTO] = { [AF_INET] = "ip", [AF_INET6] = "ip6", [NF_ARP] = "arp", @@ -248,7 +248,7 @@ EXPORT_SYMBOL_GPL(xt_request_find_target); static int match_revfn(int af, const char *name, u8 revision, int *bestp) { - struct xt_match *m; + const struct xt_match *m; int have_rev = 0; list_for_each_entry(m, &xt[af].match, list) { @@ -264,7 +264,7 @@ static int match_revfn(int af, const char *name, u8 revision, int *bestp) static int target_revfn(int af, const char *name, u8 revision, int *bestp) { - struct xt_target *t; + const struct xt_target *t; int have_rev = 0; list_for_each_entry(t, &xt[af].target, list) { @@ -385,7 +385,7 @@ short xt_compat_calc_jump(int af, unsigned int offset) } EXPORT_SYMBOL_GPL(xt_compat_calc_jump); -int xt_compat_match_offset(struct xt_match *match) +int xt_compat_match_offset(const struct xt_match *match) { u_int16_t csize = match->compatsize ? : match->matchsize; return XT_ALIGN(match->matchsize) - COMPAT_XT_ALIGN(csize); @@ -395,7 +395,7 @@ EXPORT_SYMBOL_GPL(xt_compat_match_offset); int xt_compat_match_from_user(struct xt_entry_match *m, void **dstptr, unsigned int *size) { - struct xt_match *match = m->u.kernel.match; + const struct xt_match *match = m->u.kernel.match; struct compat_xt_entry_match *cm = (struct compat_xt_entry_match *)m; int pad, off = xt_compat_match_offset(match); u_int16_t msize = cm->u.user.match_size; @@ -422,7 +422,7 @@ EXPORT_SYMBOL_GPL(xt_compat_match_from_user); int xt_compat_match_to_user(struct xt_entry_match *m, void __user **dstptr, unsigned int *size) { - struct xt_match *match = m->u.kernel.match; + const struct xt_match *match = m->u.kernel.match; struct compat_xt_entry_match __user *cm = *dstptr; int off = xt_compat_match_offset(match); u_int16_t msize = m->u.user.match_size - off; @@ -479,7 +479,7 @@ int xt_check_target(const struct xt_target *target, unsigned short family, EXPORT_SYMBOL_GPL(xt_check_target); #ifdef CONFIG_COMPAT -int xt_compat_target_offset(struct xt_target *target) +int xt_compat_target_offset(const struct xt_target *target) { u_int16_t csize = target->compatsize ? : target->targetsize; return XT_ALIGN(target->targetsize) - COMPAT_XT_ALIGN(csize); @@ -489,7 +489,7 @@ EXPORT_SYMBOL_GPL(xt_compat_target_offset); void xt_compat_target_from_user(struct xt_entry_target *t, void **dstptr, unsigned int *size) { - struct xt_target *target = t->u.kernel.target; + const struct xt_target *target = t->u.kernel.target; struct compat_xt_entry_target *ct = (struct compat_xt_entry_target *)t; int pad, off = xt_compat_target_offset(target); u_int16_t tsize = ct->u.user.target_size; @@ -515,7 +515,7 @@ EXPORT_SYMBOL_GPL(xt_compat_target_from_user); int xt_compat_target_to_user(struct xt_entry_target *t, void __user **dstptr, unsigned int *size) { - struct xt_target *target = t->u.kernel.target; + const struct xt_target *target = t->u.kernel.target; struct compat_xt_entry_target __user *ct = *dstptr; int off = xt_compat_target_offset(target); u_int16_t tsize = t->u.user.target_size - off; -- cgit v1.2.3 From 4abff0775d5e4feb20b21371e1c63a1b30fc2140 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:43 +0200 Subject: [NETFILTER]: remove arpt_table indirection macro Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter_arp/arp_tables.h | 11 +++++------ net/ipv4/netfilter/arp_tables.c | 27 +++++++++++++-------------- net/ipv4/netfilter/arptable_filter.c | 2 +- 3 files changed, 19 insertions(+), 21 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index db223ca92c8b..102c4134a713 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -24,7 +24,6 @@ #define ARPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN #define arpt_target xt_target -#define arpt_table xt_table #define ARPT_DEV_ADDR_LEN_MAX 16 @@ -271,15 +270,15 @@ struct arpt_error xt_register_target(tgt); }) #define arpt_unregister_target(tgt) xt_unregister_target(tgt) -extern struct arpt_table *arpt_register_table(struct net *net, - struct arpt_table *table, - const struct arpt_replace *repl); -extern void arpt_unregister_table(struct arpt_table *table); +extern struct xt_table *arpt_register_table(struct net *net, + struct xt_table *table, + const struct arpt_replace *repl); +extern void arpt_unregister_table(struct xt_table *table); extern unsigned int arpt_do_table(struct sk_buff *skb, unsigned int hook, const struct net_device *in, const struct net_device *out, - struct arpt_table *table); + struct xt_table *table); #define ARPT_ALIGN(s) XT_ALIGN(s) diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 10cc442330c3..34c42c831b18 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -222,7 +222,7 @@ unsigned int arpt_do_table(struct sk_buff *skb, unsigned int hook, const struct net_device *in, const struct net_device *out, - struct arpt_table *table) + struct xt_table *table) { static const char nulldevname[IFNAMSIZ]; unsigned int verdict = NF_DROP; @@ -706,7 +706,7 @@ static void get_counters(const struct xt_table_info *t, } } -static inline struct xt_counters *alloc_counters(struct arpt_table *table) +static inline struct xt_counters *alloc_counters(struct xt_table *table) { unsigned int countersize; struct xt_counters *counters; @@ -731,13 +731,13 @@ static inline struct xt_counters *alloc_counters(struct arpt_table *table) } static int copy_entries_to_user(unsigned int total_size, - struct arpt_table *table, + struct xt_table *table, void __user *userptr) { unsigned int off, num; struct arpt_entry *e; struct xt_counters *counters; - const struct xt_table_info *private = table->private; + struct xt_table_info *private = table->private; int ret = 0; void *loc_cpu_entry; @@ -851,7 +851,7 @@ static int compat_table_info(const struct xt_table_info *info, static int get_info(struct net *net, void __user *user, int *len, int compat) { char name[ARPT_TABLE_MAXNAMELEN]; - struct arpt_table *t; + struct xt_table *t; int ret; if (*len != sizeof(struct arpt_getinfo)) { @@ -911,7 +911,7 @@ static int get_entries(struct net *net, struct arpt_get_entries __user *uptr, { int ret; struct arpt_get_entries get; - struct arpt_table *t; + struct xt_table *t; if (*len < sizeof(get)) { duprintf("get_entries: %u < %Zu\n", *len, sizeof(get)); @@ -954,7 +954,7 @@ static int __do_replace(struct net *net, const char *name, void __user *counters_ptr) { int ret; - struct arpt_table *t; + struct xt_table *t; struct xt_table_info *oldinfo; struct xt_counters *counters; void *loc_cpu_old_entry; @@ -1091,7 +1091,7 @@ static int do_add_counters(struct net *net, void __user *user, unsigned int len, const char *name; int size; void *ptmp; - struct arpt_table *t; + struct xt_table *t; const struct xt_table_info *private; int ret = 0; void *loc_cpu_entry; @@ -1555,7 +1555,7 @@ out: } static int compat_copy_entries_to_user(unsigned int total_size, - struct arpt_table *table, + struct xt_table *table, void __user *userptr) { struct xt_counters *counters; @@ -1593,7 +1593,7 @@ static int compat_get_entries(struct net *net, { int ret; struct compat_arpt_get_entries get; - struct arpt_table *t; + struct xt_table *t; if (*len < sizeof(get)) { duprintf("compat_get_entries: %u < %zu\n", *len, sizeof(get)); @@ -1723,9 +1723,8 @@ static int do_arpt_get_ctl(struct sock *sk, int cmd, void __user *user, int *len return ret; } -struct arpt_table *arpt_register_table(struct net *net, - struct arpt_table *table, - const struct arpt_replace *repl) +struct xt_table *arpt_register_table(struct net *net, struct xt_table *table, + const struct arpt_replace *repl) { int ret; struct xt_table_info *newinfo; @@ -1767,7 +1766,7 @@ out: return ERR_PTR(ret); } -void arpt_unregister_table(struct arpt_table *table) +void arpt_unregister_table(struct xt_table *table) { struct xt_table_info *private; void *loc_cpu_entry; diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c index 629e4951a9b1..9f6526c87757 100644 --- a/net/ipv4/netfilter/arptable_filter.c +++ b/net/ipv4/netfilter/arptable_filter.c @@ -45,7 +45,7 @@ static struct .term = ARPT_ERROR_INIT, }; -static struct arpt_table packet_filter = { +static struct xt_table packet_filter = { .name = "filter", .valid_hooks = FILTER_VALID_HOOKS, .lock = __RW_LOCK_UNLOCKED(packet_filter.lock), -- cgit v1.2.3 From 95eea855af69bfd54a7b73546190e76046ca2e07 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:43 +0200 Subject: [NETFILTER]: remove arpt_target indirection macro Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter_arp/arp_tables.h | 1 - net/ipv4/netfilter/arp_tables.c | 8 ++++---- net/ipv4/netfilter/arpt_mangle.c | 2 +- 3 files changed, 5 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index 102c4134a713..782b83e5bdb9 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -23,7 +23,6 @@ #define ARPT_FUNCTION_MAXNAMELEN XT_FUNCTION_MAXNAMELEN #define ARPT_TABLE_MAXNAMELEN XT_TABLE_MAXNAMELEN -#define arpt_target xt_target #define ARPT_DEV_ADDR_LEN_MAX 16 diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 34c42c831b18..d55f3b42eba5 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -457,7 +457,7 @@ static inline int check_entry(struct arpt_entry *e, const char *name) static inline int check_target(struct arpt_entry *e, const char *name) { struct arpt_entry_target *t; - struct arpt_target *target; + struct xt_target *target; int ret; t = arpt_get_target(e); @@ -480,7 +480,7 @@ find_check_entry(struct arpt_entry *e, const char *name, unsigned int size, unsigned int *i) { struct arpt_entry_target *t; - struct arpt_target *target; + struct xt_target *target; int ret; ret = check_entry(e, name); @@ -1784,7 +1784,7 @@ void arpt_unregister_table(struct xt_table *table) } /* The built-in targets: standard (NULL) and error. */ -static struct arpt_target arpt_standard_target __read_mostly = { +static struct xt_target arpt_standard_target __read_mostly = { .name = ARPT_STANDARD_TARGET, .targetsize = sizeof(int), .family = NF_ARP, @@ -1795,7 +1795,7 @@ static struct arpt_target arpt_standard_target __read_mostly = { #endif }; -static struct arpt_target arpt_error_target __read_mostly = { +static struct xt_target arpt_error_target __read_mostly = { .name = ARPT_ERROR_TARGET, .target = arpt_error, .targetsize = ARPT_FUNCTION_MAXNAMELEN, diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index 3e732c827fc2..f9c102ab891b 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -73,7 +73,7 @@ checkentry(const char *tablename, const void *e, const struct xt_target *target, return true; } -static struct arpt_target arpt_mangle_reg __read_mostly = { +static struct xt_target arpt_mangle_reg __read_mostly = { .name = "mangle", .target = target, .targetsize = sizeof(struct arpt_mangle), -- cgit v1.2.3 From 3bb0362d2f53fa54a17b88c96b43fc093e47699b Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:44 +0200 Subject: [NETFILTER]: remove arpt_(un)register_target indirection macros Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter_arp/arp_tables.h | 5 ----- net/ipv4/netfilter/arpt_mangle.c | 8 +++----- 2 files changed, 3 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_arp/arp_tables.h b/include/linux/netfilter_arp/arp_tables.h index 782b83e5bdb9..dd9c97f2d436 100644 --- a/include/linux/netfilter_arp/arp_tables.h +++ b/include/linux/netfilter_arp/arp_tables.h @@ -264,11 +264,6 @@ struct arpt_error .target.errorname = "ERROR", \ } -#define arpt_register_target(tgt) \ -({ (tgt)->family = NF_ARP; \ - xt_register_target(tgt); }) -#define arpt_unregister_target(tgt) xt_unregister_target(tgt) - extern struct xt_table *arpt_register_table(struct net *net, struct xt_table *table, const struct arpt_replace *repl); diff --git a/net/ipv4/netfilter/arpt_mangle.c b/net/ipv4/netfilter/arpt_mangle.c index f9c102ab891b..a385959d2655 100644 --- a/net/ipv4/netfilter/arpt_mangle.c +++ b/net/ipv4/netfilter/arpt_mangle.c @@ -75,6 +75,7 @@ checkentry(const char *tablename, const void *e, const struct xt_target *target, static struct xt_target arpt_mangle_reg __read_mostly = { .name = "mangle", + .family = NF_ARP, .target = target, .targetsize = sizeof(struct arpt_mangle), .checkentry = checkentry, @@ -83,15 +84,12 @@ static struct xt_target arpt_mangle_reg __read_mostly = { static int __init arpt_mangle_init(void) { - if (arpt_register_target(&arpt_mangle_reg)) - return -EINVAL; - - return 0; + return xt_register_target(&arpt_mangle_reg); } static void __exit arpt_mangle_fini(void) { - arpt_unregister_target(&arpt_mangle_reg); + xt_unregister_target(&arpt_mangle_reg); } module_init(arpt_mangle_init); -- cgit v1.2.3 From 937e0dfd87a8b7946a17161664500fba93eb13fd Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 20 Mar 2008 15:15:47 +0100 Subject: [NETFILTER]: nf_nat: add helpers for common NAT protocol operations Add generic ->in_range and ->unique_tuple ops to avoid duplicating them again and again for future NAT modules and save a few bytes of text: net/ipv4/netfilter/nf_nat_proto_tcp.c: tcp_in_range | -62 (removed) tcp_unique_tuple | -259 # 271 -> 12, # inlines: 1 -> 0, size inlines: 7 -> 0 2 functions changed, 321 bytes removed net/ipv4/netfilter/nf_nat_proto_udp.c: udp_in_range | -62 (removed) udp_unique_tuple | -259 # 271 -> 12, # inlines: 1 -> 0, size inlines: 7 -> 0 2 functions changed, 321 bytes removed net/ipv4/netfilter/nf_nat_proto_gre.c: gre_in_range | -62 (removed) 1 function changed, 62 bytes removed vmlinux: 5 functions changed, 704 bytes removed Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_nat_protocol.h | 11 +++++ net/ipv4/netfilter/Makefile | 2 +- net/ipv4/netfilter/nf_nat_proto_common.c | 85 ++++++++++++++++++++++++++++++++ net/ipv4/netfilter/nf_nat_proto_gre.c | 20 +------- net/ipv4/netfilter/nf_nat_proto_tcp.c | 65 ++---------------------- net/ipv4/netfilter/nf_nat_proto_udp.c | 64 ++---------------------- 6 files changed, 106 insertions(+), 141 deletions(-) create mode 100644 net/ipv4/netfilter/nf_nat_proto_common.c (limited to 'include') diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h index 4aa0edbb5b96..fa06f6d0de54 100644 --- a/include/net/netfilter/nf_nat_protocol.h +++ b/include/net/netfilter/nf_nat_protocol.h @@ -62,6 +62,17 @@ extern int init_protocols(void) __init; extern void cleanup_protocols(void); extern const struct nf_nat_protocol *find_nat_proto(u_int16_t protonum); +extern int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max); + +extern int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct, + u_int16_t *rover); + extern int nf_nat_port_range_to_nlattr(struct sk_buff *skb, const struct nf_nat_range *range); extern int nf_nat_port_nlattr_to_range(struct nlattr *tb[], diff --git a/net/ipv4/netfilter/Makefile b/net/ipv4/netfilter/Makefile index 0c7dc78a62e9..e73d0eb9994a 100644 --- a/net/ipv4/netfilter/Makefile +++ b/net/ipv4/netfilter/Makefile @@ -10,7 +10,7 @@ nf_conntrack_ipv4-objs += nf_conntrack_l3proto_ipv4_compat.o endif endif -nf_nat-objs := nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o +nf_nat-objs := nf_nat_core.o nf_nat_helper.o nf_nat_proto_unknown.o nf_nat_proto_common.o nf_nat_proto_tcp.o nf_nat_proto_udp.o nf_nat_proto_icmp.o iptable_nat-objs := nf_nat_rule.o nf_nat_standalone.o # connection tracking diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c new file mode 100644 index 000000000000..a124213fb9da --- /dev/null +++ b/net/ipv4/netfilter/nf_nat_proto_common.c @@ -0,0 +1,85 @@ +/* (C) 1999-2001 Paul `Rusty' Russell + * (C) 2002-2006 Netfilter Core Team + * (C) 2008 Patrick McHardy + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include + +#include +#include +#include +#include +#include + +int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) +{ + __be16 port; + + if (maniptype == IP_NAT_MANIP_SRC) + port = tuple->src.u.all; + else + port = tuple->dst.u.all; + + return ntohs(port) >= ntohs(min->all) && + ntohs(port) <= ntohs(max->all); +} +EXPORT_SYMBOL_GPL(nf_nat_proto_in_range); + +int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct, + u_int16_t *rover) +{ + unsigned int range_size, min, i; + __be16 *portptr; + + if (maniptype == IP_NAT_MANIP_SRC) + portptr = &tuple->src.u.all; + else + portptr = &tuple->dst.u.all; + + /* If no range specified... */ + if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { + /* If it's dst rewrite, can't change port */ + if (maniptype == IP_NAT_MANIP_DST) + return 0; + + if (ntohs(*portptr) < 1024) { + /* Loose convention: >> 512 is credential passing */ + if (ntohs(*portptr) < 512) { + min = 1; + range_size = 511 - min + 1; + } else { + min = 600; + range_size = 1023 - min + 1; + } + } else { + min = 1024; + range_size = 65535 - 1024 + 1; + } + } else { + min = ntohs(range->min.all); + range_size = ntohs(range->max.all) - min + 1; + } + + if (range->flags & IP_NAT_RANGE_PROTO_RANDOM) + *rover = net_random(); + + for (i = 0; i < range_size; i++, (*rover)++) { + *portptr = htons(min + *rover % range_size); + if (!nf_nat_used_tuple(tuple, ct)) + return 1; + } + return 0; +} +EXPORT_SYMBOL_GPL(nf_nat_proto_unique_tuple); diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c index a1e4da16da2e..87af63d9e692 100644 --- a/net/ipv4/netfilter/nf_nat_proto_gre.c +++ b/net/ipv4/netfilter/nf_nat_proto_gre.c @@ -36,24 +36,6 @@ MODULE_LICENSE("GPL"); MODULE_AUTHOR("Harald Welte "); MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE"); -/* is key in given range between min and max */ -static int -gre_in_range(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max) -{ - __be16 key; - - if (maniptype == IP_NAT_MANIP_SRC) - key = tuple->src.u.gre.key; - else - key = tuple->dst.u.gre.key; - - return ntohs(key) >= ntohs(min->gre.key) && - ntohs(key) <= ntohs(max->gre.key); -} - /* generate unique tuple ... */ static int gre_unique_tuple(struct nf_conntrack_tuple *tuple, @@ -140,7 +122,7 @@ static const struct nf_nat_protocol gre = { .protonum = IPPROTO_GRE, .me = THIS_MODULE, .manip_pkt = gre_manip_pkt, - .in_range = gre_in_range, + .in_range = nf_nat_proto_in_range, .unique_tuple = gre_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) .range_to_nlattr = nf_nat_port_range_to_nlattr, diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c index ffd5d1589eca..f8c498fc24fd 100644 --- a/net/ipv4/netfilter/nf_nat_proto_tcp.c +++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c @@ -8,7 +8,6 @@ #include #include -#include #include #include @@ -19,22 +18,7 @@ #include #include -static int -tcp_in_range(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max) -{ - __be16 port; - - if (maniptype == IP_NAT_MANIP_SRC) - port = tuple->src.u.tcp.port; - else - port = tuple->dst.u.tcp.port; - - return ntohs(port) >= ntohs(min->tcp.port) && - ntohs(port) <= ntohs(max->tcp.port); -} +static u_int16_t tcp_port_rover; static int tcp_unique_tuple(struct nf_conntrack_tuple *tuple, @@ -42,49 +26,8 @@ tcp_unique_tuple(struct nf_conntrack_tuple *tuple, enum nf_nat_manip_type maniptype, const struct nf_conn *ct) { - static u_int16_t port; - __be16 *portptr; - unsigned int range_size, min, i; - - if (maniptype == IP_NAT_MANIP_SRC) - portptr = &tuple->src.u.tcp.port; - else - portptr = &tuple->dst.u.tcp.port; - - /* If no range specified... */ - if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { - /* If it's dst rewrite, can't change port */ - if (maniptype == IP_NAT_MANIP_DST) - return 0; - - /* Map privileged onto privileged. */ - if (ntohs(*portptr) < 1024) { - /* Loose convention: >> 512 is credential passing */ - if (ntohs(*portptr)<512) { - min = 1; - range_size = 511 - min + 1; - } else { - min = 600; - range_size = 1023 - min + 1; - } - } else { - min = 1024; - range_size = 65535 - 1024 + 1; - } - } else { - min = ntohs(range->min.tcp.port); - range_size = ntohs(range->max.tcp.port) - min + 1; - } - - if (range->flags & IP_NAT_RANGE_PROTO_RANDOM) - port = net_random(); - - for (i = 0; i < range_size; i++, port++) { - *portptr = htons(min + port % range_size); - if (!nf_nat_used_tuple(tuple, ct)) - return 1; - } - return 0; + return nf_nat_proto_unique_tuple(tuple, range, maniptype, ct, + &tcp_port_rover); } static int @@ -142,7 +85,7 @@ const struct nf_nat_protocol nf_nat_protocol_tcp = { .protonum = IPPROTO_TCP, .me = THIS_MODULE, .manip_pkt = tcp_manip_pkt, - .in_range = tcp_in_range, + .in_range = nf_nat_proto_in_range, .unique_tuple = tcp_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) .range_to_nlattr = nf_nat_port_range_to_nlattr, diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c index 4b8f49910ff2..a182f5ac3177 100644 --- a/net/ipv4/netfilter/nf_nat_proto_udp.c +++ b/net/ipv4/netfilter/nf_nat_proto_udp.c @@ -8,7 +8,6 @@ #include #include -#include #include #include @@ -18,22 +17,7 @@ #include #include -static int -udp_in_range(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max) -{ - __be16 port; - - if (maniptype == IP_NAT_MANIP_SRC) - port = tuple->src.u.udp.port; - else - port = tuple->dst.u.udp.port; - - return ntohs(port) >= ntohs(min->udp.port) && - ntohs(port) <= ntohs(max->udp.port); -} +static u_int16_t udp_port_rover; static int udp_unique_tuple(struct nf_conntrack_tuple *tuple, @@ -41,48 +25,8 @@ udp_unique_tuple(struct nf_conntrack_tuple *tuple, enum nf_nat_manip_type maniptype, const struct nf_conn *ct) { - static u_int16_t port; - __be16 *portptr; - unsigned int range_size, min, i; - - if (maniptype == IP_NAT_MANIP_SRC) - portptr = &tuple->src.u.udp.port; - else - portptr = &tuple->dst.u.udp.port; - - /* If no range specified... */ - if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { - /* If it's dst rewrite, can't change port */ - if (maniptype == IP_NAT_MANIP_DST) - return 0; - - if (ntohs(*portptr) < 1024) { - /* Loose convention: >> 512 is credential passing */ - if (ntohs(*portptr)<512) { - min = 1; - range_size = 511 - min + 1; - } else { - min = 600; - range_size = 1023 - min + 1; - } - } else { - min = 1024; - range_size = 65535 - 1024 + 1; - } - } else { - min = ntohs(range->min.udp.port); - range_size = ntohs(range->max.udp.port) - min + 1; - } - - if (range->flags & IP_NAT_RANGE_PROTO_RANDOM) - port = net_random(); - - for (i = 0; i < range_size; i++, port++) { - *portptr = htons(min + port % range_size); - if (!nf_nat_used_tuple(tuple, ct)) - return 1; - } - return 0; + return nf_nat_proto_unique_tuple(tuple, range, maniptype, ct, + &udp_port_rover); } static int @@ -132,7 +76,7 @@ const struct nf_nat_protocol nf_nat_protocol_udp = { .protonum = IPPROTO_UDP, .me = THIS_MODULE, .manip_pkt = udp_manip_pkt, - .in_range = udp_in_range, + .in_range = nf_nat_proto_in_range, .unique_tuple = udp_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) .range_to_nlattr = nf_nat_port_range_to_nlattr, -- cgit v1.2.3 From 535b57c7c1524125444aa1b874332f6ff1608ef5 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 14 Apr 2008 11:15:47 +0200 Subject: [NETFILTER]: nf_nat: move NAT ctnetlink helpers to nf_nat_proto_common Move to nf_nat_proto_common and rename to nf_nat_proto_... since they're also used by protocols that don't have port numbers. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_nat_protocol.h | 8 +++---- net/ipv4/netfilter/nf_nat_core.c | 40 -------------------------------- net/ipv4/netfilter/nf_nat_proto_common.c | 38 ++++++++++++++++++++++++++++++ net/ipv4/netfilter/nf_nat_proto_gre.c | 4 ++-- net/ipv4/netfilter/nf_nat_proto_icmp.c | 4 ++-- net/ipv4/netfilter/nf_nat_proto_tcp.c | 4 ++-- net/ipv4/netfilter/nf_nat_proto_udp.c | 4 ++-- 7 files changed, 50 insertions(+), 52 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h index fa06f6d0de54..8ce227624eac 100644 --- a/include/net/netfilter/nf_nat_protocol.h +++ b/include/net/netfilter/nf_nat_protocol.h @@ -73,9 +73,9 @@ extern int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_conn *ct, u_int16_t *rover); -extern int nf_nat_port_range_to_nlattr(struct sk_buff *skb, - const struct nf_nat_range *range); -extern int nf_nat_port_nlattr_to_range(struct nlattr *tb[], - struct nf_nat_range *range); +extern int nf_nat_proto_range_to_nlattr(struct sk_buff *skb, + const struct nf_nat_range *range); +extern int nf_nat_proto_nlattr_to_range(struct nlattr *tb[], + struct nf_nat_range *range); #endif /*_NF_NAT_PROTO_H*/ diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c index 9c8aa8df2d5e..9320c7ac5729 100644 --- a/net/ipv4/netfilter/nf_nat_core.c +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -544,46 +544,6 @@ void nf_nat_protocol_unregister(const struct nf_nat_protocol *proto) } EXPORT_SYMBOL(nf_nat_protocol_unregister); -#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) -int -nf_nat_port_range_to_nlattr(struct sk_buff *skb, - const struct nf_nat_range *range) -{ - NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MIN, range->min.tcp.port); - NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MAX, range->max.tcp.port); - - return 0; - -nla_put_failure: - return -1; -} -EXPORT_SYMBOL_GPL(nf_nat_port_nlattr_to_range); - -int -nf_nat_port_nlattr_to_range(struct nlattr *tb[], struct nf_nat_range *range) -{ - int ret = 0; - - /* we have to return whether we actually parsed something or not */ - - if (tb[CTA_PROTONAT_PORT_MIN]) { - ret = 1; - range->min.tcp.port = nla_get_be16(tb[CTA_PROTONAT_PORT_MIN]); - } - - if (!tb[CTA_PROTONAT_PORT_MAX]) { - if (ret) - range->max.tcp.port = range->min.tcp.port; - } else { - ret = 1; - range->max.tcp.port = nla_get_be16(tb[CTA_PROTONAT_PORT_MAX]); - } - - return ret; -} -EXPORT_SYMBOL_GPL(nf_nat_port_range_to_nlattr); -#endif - /* Noone using conntrack by the time this called. */ static void nf_nat_cleanup_conntrack(struct nf_conn *ct) { diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c index 871ab0eb325d..ef4dc3988925 100644 --- a/net/ipv4/netfilter/nf_nat_proto_common.c +++ b/net/ipv4/netfilter/nf_nat_proto_common.c @@ -88,3 +88,41 @@ int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, return 0; } EXPORT_SYMBOL_GPL(nf_nat_proto_unique_tuple); + +#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) +int nf_nat_proto_range_to_nlattr(struct sk_buff *skb, + const struct nf_nat_range *range) +{ + NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MIN, range->min.all); + NLA_PUT_BE16(skb, CTA_PROTONAT_PORT_MAX, range->max.all); + return 0; + +nla_put_failure: + return -1; +} +EXPORT_SYMBOL_GPL(nf_nat_proto_nlattr_to_range); + +int nf_nat_proto_nlattr_to_range(struct nlattr *tb[], + struct nf_nat_range *range) +{ + int ret = 0; + + /* we have to return whether we actually parsed something or not */ + + if (tb[CTA_PROTONAT_PORT_MIN]) { + ret = 1; + range->min.all = nla_get_be16(tb[CTA_PROTONAT_PORT_MIN]); + } + + if (!tb[CTA_PROTONAT_PORT_MAX]) { + if (ret) + range->max.all = range->min.all; + } else { + ret = 1; + range->max.all = nla_get_be16(tb[CTA_PROTONAT_PORT_MAX]); + } + + return ret; +} +EXPORT_SYMBOL_GPL(nf_nat_proto_range_to_nlattr); +#endif diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c index 87af63d9e692..71b0935ee1c8 100644 --- a/net/ipv4/netfilter/nf_nat_proto_gre.c +++ b/net/ipv4/netfilter/nf_nat_proto_gre.c @@ -125,8 +125,8 @@ static const struct nf_nat_protocol gre = { .in_range = nf_nat_proto_in_range, .unique_tuple = gre_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) - .range_to_nlattr = nf_nat_port_range_to_nlattr, - .nlattr_to_range = nf_nat_port_nlattr_to_range, + .range_to_nlattr = nf_nat_proto_range_to_nlattr, + .nlattr_to_range = nf_nat_proto_nlattr_to_range, #endif }; diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c index 03a02969aa57..ca601f84c4dc 100644 --- a/net/ipv4/netfilter/nf_nat_proto_icmp.c +++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c @@ -79,7 +79,7 @@ const struct nf_nat_protocol nf_nat_protocol_icmp = { .in_range = icmp_in_range, .unique_tuple = icmp_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) - .range_to_nlattr = nf_nat_port_range_to_nlattr, - .nlattr_to_range = nf_nat_port_nlattr_to_range, + .range_to_nlattr = nf_nat_proto_range_to_nlattr, + .nlattr_to_range = nf_nat_proto_nlattr_to_range, #endif }; diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c index f8c498fc24fd..1d73a11f55d9 100644 --- a/net/ipv4/netfilter/nf_nat_proto_tcp.c +++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c @@ -88,7 +88,7 @@ const struct nf_nat_protocol nf_nat_protocol_tcp = { .in_range = nf_nat_proto_in_range, .unique_tuple = tcp_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) - .range_to_nlattr = nf_nat_port_range_to_nlattr, - .nlattr_to_range = nf_nat_port_nlattr_to_range, + .range_to_nlattr = nf_nat_proto_range_to_nlattr, + .nlattr_to_range = nf_nat_proto_nlattr_to_range, #endif }; diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c index a182f5ac3177..f36ce552a161 100644 --- a/net/ipv4/netfilter/nf_nat_proto_udp.c +++ b/net/ipv4/netfilter/nf_nat_proto_udp.c @@ -79,7 +79,7 @@ const struct nf_nat_protocol nf_nat_protocol_udp = { .in_range = nf_nat_proto_in_range, .unique_tuple = udp_unique_tuple, #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) - .range_to_nlattr = nf_nat_port_range_to_nlattr, - .nlattr_to_range = nf_nat_port_nlattr_to_range, + .range_to_nlattr = nf_nat_proto_range_to_nlattr, + .nlattr_to_range = nf_nat_proto_nlattr_to_range, #endif }; -- cgit v1.2.3 From 2d2d84c40e19a7fce51ba1f124ecde105104192d Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 20 Mar 2008 15:15:49 +0100 Subject: [NETFILTER]: nf_nat: remove unused name from struct nf_nat_protocol Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_nat_protocol.h | 3 --- net/ipv4/netfilter/nf_nat_proto_gre.c | 1 - net/ipv4/netfilter/nf_nat_proto_icmp.c | 1 - net/ipv4/netfilter/nf_nat_proto_tcp.c | 1 - net/ipv4/netfilter/nf_nat_proto_udp.c | 1 - net/ipv4/netfilter/nf_nat_proto_unknown.c | 1 - 6 files changed, 8 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h index 8ce227624eac..fba94a2028d5 100644 --- a/include/net/netfilter/nf_nat_protocol.h +++ b/include/net/netfilter/nf_nat_protocol.h @@ -8,9 +8,6 @@ struct nf_nat_range; struct nf_nat_protocol { - /* Protocol name */ - const char *name; - /* Protocol number. */ unsigned int protonum; diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c index 71b0935ee1c8..84bb7854137a 100644 --- a/net/ipv4/netfilter/nf_nat_proto_gre.c +++ b/net/ipv4/netfilter/nf_nat_proto_gre.c @@ -118,7 +118,6 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, } static const struct nf_nat_protocol gre = { - .name = "GRE", .protonum = IPPROTO_GRE, .me = THIS_MODULE, .manip_pkt = gre_manip_pkt, diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c index ca601f84c4dc..ab3a0ec2a2d1 100644 --- a/net/ipv4/netfilter/nf_nat_proto_icmp.c +++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c @@ -72,7 +72,6 @@ icmp_manip_pkt(struct sk_buff *skb, } const struct nf_nat_protocol nf_nat_protocol_icmp = { - .name = "ICMP", .protonum = IPPROTO_ICMP, .me = THIS_MODULE, .manip_pkt = icmp_manip_pkt, diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c index 1d73a11f55d9..5d4c8a0e89c0 100644 --- a/net/ipv4/netfilter/nf_nat_proto_tcp.c +++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c @@ -81,7 +81,6 @@ tcp_manip_pkt(struct sk_buff *skb, } const struct nf_nat_protocol nf_nat_protocol_tcp = { - .name = "TCP", .protonum = IPPROTO_TCP, .me = THIS_MODULE, .manip_pkt = tcp_manip_pkt, diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c index f36ce552a161..74a7e7b63465 100644 --- a/net/ipv4/netfilter/nf_nat_proto_udp.c +++ b/net/ipv4/netfilter/nf_nat_proto_udp.c @@ -72,7 +72,6 @@ udp_manip_pkt(struct sk_buff *skb, } const struct nf_nat_protocol nf_nat_protocol_udp = { - .name = "UDP", .protonum = IPPROTO_UDP, .me = THIS_MODULE, .manip_pkt = udp_manip_pkt, diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c index a26efeb073cb..cda21ff0e4cf 100644 --- a/net/ipv4/netfilter/nf_nat_proto_unknown.c +++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c @@ -46,7 +46,6 @@ unknown_manip_pkt(struct sk_buff *skb, } const struct nf_nat_protocol nf_nat_unknown_protocol = { - .name = "unknown", /* .me isn't set: getting a ref to this cannot fail. */ .manip_pkt = unknown_manip_pkt, .in_range = unknown_in_range, -- cgit v1.2.3 From d63a650736f566a1f9e9434725d2089597c0d2cc Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 20 Mar 2008 15:15:53 +0100 Subject: [NETFILTER]: Add partial checksum validation helper Move the UDP-Lite conntrack checksum validation to a generic helper similar to nf_checksum() and make it fall back to nf_checksum() in case the full packet is to be checksummed and hardware checksums are available. This is to be used by DCCP conntrack, which also needs to verify partial checksums. Signed-off-by: Patrick McHardy --- include/linux/netfilter.h | 22 ++++++++++++++++ net/ipv4/netfilter.c | 37 +++++++++++++++++++++----- net/ipv6/netfilter.c | 42 +++++++++++++++++++++++++----- net/netfilter/nf_conntrack_proto_udplite.c | 33 +++++------------------ 4 files changed, 94 insertions(+), 40 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h index 66bc52060fd6..e4c66593b5c6 100644 --- a/include/linux/netfilter.h +++ b/include/linux/netfilter.h @@ -234,6 +234,11 @@ struct nf_afinfo { unsigned short family; __sum16 (*checksum)(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, u_int8_t protocol); + __sum16 (*checksum_partial)(struct sk_buff *skb, + unsigned int hook, + unsigned int dataoff, + unsigned int len, + u_int8_t protocol); int (*route)(struct dst_entry **dst, struct flowi *fl); void (*saveroute)(const struct sk_buff *skb, struct nf_queue_entry *entry); @@ -263,6 +268,23 @@ nf_checksum(struct sk_buff *skb, unsigned int hook, unsigned int dataoff, return csum; } +static inline __sum16 +nf_checksum_partial(struct sk_buff *skb, unsigned int hook, + unsigned int dataoff, unsigned int len, + u_int8_t protocol, unsigned short family) +{ + const struct nf_afinfo *afinfo; + __sum16 csum = 0; + + rcu_read_lock(); + afinfo = nf_get_afinfo(family); + if (afinfo) + csum = afinfo->checksum_partial(skb, hook, dataoff, len, + protocol); + rcu_read_unlock(); + return csum; +} + extern int nf_register_afinfo(const struct nf_afinfo *afinfo); extern void nf_unregister_afinfo(const struct nf_afinfo *afinfo); diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c index 9a904c6c0dc8..f8edacdf991d 100644 --- a/net/ipv4/netfilter.c +++ b/net/ipv4/netfilter.c @@ -182,21 +182,44 @@ __sum16 nf_ip_checksum(struct sk_buff *skb, unsigned int hook, } return csum; } - EXPORT_SYMBOL(nf_ip_checksum); +static __sum16 nf_ip_checksum_partial(struct sk_buff *skb, unsigned int hook, + unsigned int dataoff, unsigned int len, + u_int8_t protocol) +{ + const struct iphdr *iph = ip_hdr(skb); + __sum16 csum = 0; + + switch (skb->ip_summed) { + case CHECKSUM_COMPLETE: + if (len == skb->len - dataoff) + return nf_ip_checksum(skb, hook, dataoff, protocol); + /* fall through */ + case CHECKSUM_NONE: + skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, protocol, + skb->len - dataoff, 0); + skb->ip_summed = CHECKSUM_NONE; + csum = __skb_checksum_complete_head(skb, dataoff + len); + if (!csum) + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + return csum; +} + static int nf_ip_route(struct dst_entry **dst, struct flowi *fl) { return ip_route_output_key(&init_net, (struct rtable **)dst, fl); } static const struct nf_afinfo nf_ip_afinfo = { - .family = AF_INET, - .checksum = nf_ip_checksum, - .route = nf_ip_route, - .saveroute = nf_ip_saveroute, - .reroute = nf_ip_reroute, - .route_key_size = sizeof(struct ip_rt_info), + .family = AF_INET, + .checksum = nf_ip_checksum, + .checksum_partial = nf_ip_checksum_partial, + .route = nf_ip_route, + .saveroute = nf_ip_saveroute, + .reroute = nf_ip_reroute, + .route_key_size = sizeof(struct ip_rt_info), }; static int ipv4_netfilter_init(void) diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index aed51bcc66b4..8c6c5e71f210 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -121,16 +121,44 @@ __sum16 nf_ip6_checksum(struct sk_buff *skb, unsigned int hook, } return csum; } - EXPORT_SYMBOL(nf_ip6_checksum); +static __sum16 nf_ip6_checksum_partial(struct sk_buff *skb, unsigned int hook, + unsigned int dataoff, unsigned int len, + u_int8_t protocol) +{ + struct ipv6hdr *ip6h = ipv6_hdr(skb); + __wsum hsum; + __sum16 csum = 0; + + switch (skb->ip_summed) { + case CHECKSUM_COMPLETE: + if (len == skb->len - dataoff) + return nf_ip6_checksum(skb, hook, dataoff, protocol); + /* fall through */ + case CHECKSUM_NONE: + hsum = skb_checksum(skb, 0, dataoff, 0); + skb->csum = ~csum_unfold(csum_ipv6_magic(&ip6h->saddr, + &ip6h->daddr, + skb->len - dataoff, + protocol, + csum_sub(0, hsum))); + skb->ip_summed = CHECKSUM_NONE; + csum = __skb_checksum_complete_head(skb, dataoff + len); + if (!csum) + skb->ip_summed = CHECKSUM_UNNECESSARY; + } + return csum; +}; + static const struct nf_afinfo nf_ip6_afinfo = { - .family = AF_INET6, - .checksum = nf_ip6_checksum, - .route = nf_ip6_route, - .saveroute = nf_ip6_saveroute, - .reroute = nf_ip6_reroute, - .route_key_size = sizeof(struct ip6_rt_info), + .family = AF_INET6, + .checksum = nf_ip6_checksum, + .checksum_partial = nf_ip6_checksum_partial, + .route = nf_ip6_route, + .saveroute = nf_ip6_saveroute, + .reroute = nf_ip6_reroute, + .route_key_size = sizeof(struct ip6_rt_info), }; int __init ipv6_netfilter_init(void) diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c index 9dd03c7aeac6..c3eaee6afffd 100644 --- a/net/netfilter/nf_conntrack_proto_udplite.c +++ b/net/netfilter/nf_conntrack_proto_udplite.c @@ -127,32 +127,13 @@ static int udplite_error(struct sk_buff *skb, unsigned int dataoff, } /* Checksum invalid? Ignore. */ - if (nf_conntrack_checksum && !skb_csum_unnecessary(skb) && - hooknum == NF_INET_PRE_ROUTING) { - if (pf == PF_INET) { - struct iphdr *iph = ip_hdr(skb); - - skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr, - udplen, IPPROTO_UDPLITE, 0); - } else { - struct ipv6hdr *ipv6h = ipv6_hdr(skb); - __wsum hsum = skb_checksum(skb, 0, dataoff, 0); - - skb->csum = ~csum_unfold( - csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr, - udplen, IPPROTO_UDPLITE, - csum_sub(0, hsum))); - } - - skb->ip_summed = CHECKSUM_NONE; - if (__skb_checksum_complete_head(skb, dataoff + cscov)) { - if (LOG_INVALID(IPPROTO_UDPLITE)) - nf_log_packet(pf, 0, skb, NULL, NULL, NULL, - "nf_ct_udplite: bad UDPLite " - "checksum "); - return -NF_ACCEPT; - } - skb->ip_summed = CHECKSUM_UNNECESSARY; + if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING && + nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_UDP, + pf)) { + if (LOG_INVALID(IPPROTO_UDPLITE)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_udplite: bad UDPLite checksum "); + return -NF_ACCEPT; } return NF_ACCEPT; -- cgit v1.2.3 From 2bc780499aa33311ec0f3e42624dfaa7be0ade5e Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 20 Mar 2008 15:15:55 +0100 Subject: [NETFILTER]: nf_conntrack: add DCCP protocol support Add DCCP conntrack helper. Thanks to Gerrit Renker for review and testing. Signed-off-by: Patrick McHardy --- include/linux/netfilter/nf_conntrack_dccp.h | 40 ++ include/linux/netfilter/nfnetlink_conntrack.h | 8 + include/net/netfilter/nf_conntrack.h | 2 + include/net/netfilter/nf_conntrack_tuple.h | 6 + net/netfilter/Kconfig | 10 + net/netfilter/Makefile | 1 + net/netfilter/nf_conntrack_proto_dccp.c | 816 ++++++++++++++++++++++++++ 7 files changed, 883 insertions(+) create mode 100644 include/linux/netfilter/nf_conntrack_dccp.h create mode 100644 net/netfilter/nf_conntrack_proto_dccp.c (limited to 'include') diff --git a/include/linux/netfilter/nf_conntrack_dccp.h b/include/linux/netfilter/nf_conntrack_dccp.h new file mode 100644 index 000000000000..40dcc82058d1 --- /dev/null +++ b/include/linux/netfilter/nf_conntrack_dccp.h @@ -0,0 +1,40 @@ +#ifndef _NF_CONNTRACK_DCCP_H +#define _NF_CONNTRACK_DCCP_H + +/* Exposed to userspace over nfnetlink */ +enum ct_dccp_states { + CT_DCCP_NONE, + CT_DCCP_REQUEST, + CT_DCCP_RESPOND, + CT_DCCP_PARTOPEN, + CT_DCCP_OPEN, + CT_DCCP_CLOSEREQ, + CT_DCCP_CLOSING, + CT_DCCP_TIMEWAIT, + CT_DCCP_IGNORE, + CT_DCCP_INVALID, + __CT_DCCP_MAX +}; +#define CT_DCCP_MAX (__CT_DCCP_MAX - 1) + +enum ct_dccp_roles { + CT_DCCP_ROLE_CLIENT, + CT_DCCP_ROLE_SERVER, + __CT_DCCP_ROLE_MAX +}; +#define CT_DCCP_ROLE_MAX (__CT_DCCP_ROLE_MAX - 1) + +#ifdef __KERNEL__ +#include + +struct nf_ct_dccp { + u_int8_t role[IP_CT_DIR_MAX]; + u_int8_t state; + u_int8_t last_pkt; + u_int8_t last_dir; + u_int64_t handshake_seq; +}; + +#endif /* __KERNEL__ */ + +#endif /* _NF_CONNTRACK_DCCP_H */ diff --git a/include/linux/netfilter/nfnetlink_conntrack.h b/include/linux/netfilter/nfnetlink_conntrack.h index e3e1533aba2d..0a383ac083cb 100644 --- a/include/linux/netfilter/nfnetlink_conntrack.h +++ b/include/linux/netfilter/nfnetlink_conntrack.h @@ -80,6 +80,7 @@ enum ctattr_l4proto { enum ctattr_protoinfo { CTA_PROTOINFO_UNSPEC, CTA_PROTOINFO_TCP, + CTA_PROTOINFO_DCCP, __CTA_PROTOINFO_MAX }; #define CTA_PROTOINFO_MAX (__CTA_PROTOINFO_MAX - 1) @@ -95,6 +96,13 @@ enum ctattr_protoinfo_tcp { }; #define CTA_PROTOINFO_TCP_MAX (__CTA_PROTOINFO_TCP_MAX - 1) +enum ctattr_protoinfo_dccp { + CTA_PROTOINFO_DCCP_UNSPEC, + CTA_PROTOINFO_DCCP_STATE, + __CTA_PROTOINFO_DCCP_MAX, +}; +#define CTA_PROTOINFO_DCCP_MAX (__CTA_PROTOINFO_DCCP_MAX - 1) + enum ctattr_counters { CTA_COUNTERS_UNSPEC, CTA_COUNTERS_PACKETS, /* old 64bit counters */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index a3567a7a6d67..bb9fc852e973 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -20,6 +20,7 @@ #include #include +#include #include #include #include @@ -30,6 +31,7 @@ /* per conntrack: protocol private data */ union nf_conntrack_proto { /* insert conntrack proto private data here */ + struct nf_ct_dccp dccp; struct ip_ct_sctp sctp; struct ip_ct_tcp tcp; struct ip_ct_icmp icmp; diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index 168c91754d89..bdeec3461384 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -39,6 +39,9 @@ union nf_conntrack_man_proto struct { __be16 id; } icmp; + struct { + __be16 port; + } dccp; struct { __be16 port; } sctp; @@ -77,6 +80,9 @@ struct nf_conntrack_tuple struct { u_int8_t type, code; } icmp; + struct { + __be16 port; + } dccp; struct { __be16 port; } sctp; diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index daf5b881064d..c1fc0f1a641c 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -86,6 +86,16 @@ config NF_CONNTRACK_EVENTS If unsure, say `N'. +config NF_CT_PROTO_DCCP + tristate 'DCCP protocol connection tracking support (EXPERIMENTAL)' + depends on EXPERIMENTAL && NF_CONNTRACK + depends on NETFILTER_ADVANCED + help + With this option enabled, the layer 3 independent connection + tracking code will be able to do state tracking on DCCP connections. + + If unsure, say 'N'. + config NF_CT_PROTO_GRE tristate depends on NF_CONNTRACK diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index ea7508387f95..5c4b183f6422 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -13,6 +13,7 @@ obj-$(CONFIG_NETFILTER_NETLINK_LOG) += nfnetlink_log.o obj-$(CONFIG_NF_CONNTRACK) += nf_conntrack.o # SCTP protocol connection tracking +obj-$(CONFIG_NF_CT_PROTO_DCCP) += nf_conntrack_proto_dccp.o obj-$(CONFIG_NF_CT_PROTO_GRE) += nf_conntrack_proto_gre.o obj-$(CONFIG_NF_CT_PROTO_SCTP) += nf_conntrack_proto_sctp.o obj-$(CONFIG_NF_CT_PROTO_UDPLITE) += nf_conntrack_proto_udplite.o diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c new file mode 100644 index 000000000000..db88c5bcc5fd --- /dev/null +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -0,0 +1,816 @@ +/* + * DCCP connection tracking protocol helper + * + * Copyright (c) 2005, 2006, 2008 Patrick McHardy + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +static DEFINE_RWLOCK(dccp_lock); + +static int nf_ct_dccp_loose __read_mostly = 1; + +/* Timeouts are based on values from RFC4340: + * + * - REQUEST: + * + * 8.1.2. Client Request + * + * A client MAY give up on its DCCP-Requests after some time + * (3 minutes, for example). + * + * - RESPOND: + * + * 8.1.3. Server Response + * + * It MAY also leave the RESPOND state for CLOSED after a timeout of + * not less than 4MSL (8 minutes); + * + * - PARTOPEN: + * + * 8.1.5. Handshake Completion + * + * If the client remains in PARTOPEN for more than 4MSL (8 minutes), + * it SHOULD reset the connection with Reset Code 2, "Aborted". + * + * - OPEN: + * + * The DCCP timestamp overflows after 11.9 hours. If the connection + * stays idle this long the sequence number won't be recognized + * as valid anymore. + * + * - CLOSEREQ/CLOSING: + * + * 8.3. Termination + * + * The retransmission timer should initially be set to go off in two + * round-trip times and should back off to not less than once every + * 64 seconds ... + * + * - TIMEWAIT: + * + * 4.3. States + * + * A server or client socket remains in this state for 2MSL (4 minutes) + * after the connection has been town down, ... + */ + +#define DCCP_MSL (2 * 60 * HZ) + +static unsigned int dccp_timeout[CT_DCCP_MAX + 1] __read_mostly = { + [CT_DCCP_REQUEST] = 2 * DCCP_MSL, + [CT_DCCP_RESPOND] = 4 * DCCP_MSL, + [CT_DCCP_PARTOPEN] = 4 * DCCP_MSL, + [CT_DCCP_OPEN] = 12 * 3600 * HZ, + [CT_DCCP_CLOSEREQ] = 64 * HZ, + [CT_DCCP_CLOSING] = 64 * HZ, + [CT_DCCP_TIMEWAIT] = 2 * DCCP_MSL, +}; + +static const char * const dccp_state_names[] = { + [CT_DCCP_NONE] = "NONE", + [CT_DCCP_REQUEST] = "REQUEST", + [CT_DCCP_RESPOND] = "RESPOND", + [CT_DCCP_PARTOPEN] = "PARTOPEN", + [CT_DCCP_OPEN] = "OPEN", + [CT_DCCP_CLOSEREQ] = "CLOSEREQ", + [CT_DCCP_CLOSING] = "CLOSING", + [CT_DCCP_TIMEWAIT] = "TIMEWAIT", + [CT_DCCP_IGNORE] = "IGNORE", + [CT_DCCP_INVALID] = "INVALID", +}; + +#define sNO CT_DCCP_NONE +#define sRQ CT_DCCP_REQUEST +#define sRS CT_DCCP_RESPOND +#define sPO CT_DCCP_PARTOPEN +#define sOP CT_DCCP_OPEN +#define sCR CT_DCCP_CLOSEREQ +#define sCG CT_DCCP_CLOSING +#define sTW CT_DCCP_TIMEWAIT +#define sIG CT_DCCP_IGNORE +#define sIV CT_DCCP_INVALID + +/* + * DCCP state transistion table + * + * The assumption is the same as for TCP tracking: + * + * We are the man in the middle. All the packets go through us but might + * get lost in transit to the destination. It is assumed that the destination + * can't receive segments we haven't seen. + * + * The following states exist: + * + * NONE: Initial state, expecting Request + * REQUEST: Request seen, waiting for Response from server + * RESPOND: Response from server seen, waiting for Ack from client + * PARTOPEN: Ack after Response seen, waiting for packet other than Response, + * Reset or Sync from server + * OPEN: Packet other than Response, Reset or Sync seen + * CLOSEREQ: CloseReq from server seen, expecting Close from client + * CLOSING: Close seen, expecting Reset + * TIMEWAIT: Reset seen + * IGNORE: Not determinable whether packet is valid + * + * Some states exist only on one side of the connection: REQUEST, RESPOND, + * PARTOPEN, CLOSEREQ. For the other side these states are equivalent to + * the one it was in before. + * + * Packets are marked as ignored (sIG) if we don't know if they're valid + * (for example a reincarnation of a connection we didn't notice is dead + * already) and the server may send back a connection closing Reset or a + * Response. They're also used for Sync/SyncAck packets, which we don't + * care about. + */ +static const u_int8_t +dccp_state_table[CT_DCCP_ROLE_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = { + [CT_DCCP_ROLE_CLIENT] = { + [DCCP_PKT_REQUEST] = { + /* + * sNO -> sRQ Regular Request + * sRQ -> sRQ Retransmitted Request or reincarnation + * sRS -> sRS Retransmitted Request (apparently Response + * got lost after we saw it) or reincarnation + * sPO -> sIG Ignore, conntrack might be out of sync + * sOP -> sIG Ignore, conntrack might be out of sync + * sCR -> sIG Ignore, conntrack might be out of sync + * sCG -> sIG Ignore, conntrack might be out of sync + * sTW -> sRQ Reincarnation + * + * sNO, sRQ, sRS, sPO. sOP, sCR, sCG, sTW, */ + sRQ, sRQ, sRS, sIG, sIG, sIG, sIG, sRQ, + }, + [DCCP_PKT_RESPONSE] = { + /* + * sNO -> sIV Invalid + * sRQ -> sIG Ignore, might be response to ignored Request + * sRS -> sIG Ignore, might be response to ignored Request + * sPO -> sIG Ignore, might be response to ignored Request + * sOP -> sIG Ignore, might be response to ignored Request + * sCR -> sIG Ignore, might be response to ignored Request + * sCG -> sIG Ignore, might be response to ignored Request + * sTW -> sIV Invalid, reincarnation in reverse direction + * goes through sRQ + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIG, sIG, sIG, sIG, sIG, sIG, sIV, + }, + [DCCP_PKT_ACK] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sPO Ack for Response, move to PARTOPEN (8.1.5.) + * sPO -> sPO Retransmitted Ack for Response, remain in PARTOPEN + * sOP -> sOP Regular ACK, remain in OPEN + * sCR -> sCR Ack in CLOSEREQ MAY be processed (8.3.) + * sCG -> sCG Ack in CLOSING MAY be processed (8.3.) + * sTW -> sIV + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sPO, sPO, sOP, sCR, sCG, sIV + }, + [DCCP_PKT_DATA] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sIV MUST use DataAck in PARTOPEN state (8.1.5.) + * sOP -> sOP Regular Data packet + * sCR -> sCR Data in CLOSEREQ MAY be processed (8.3.) + * sCG -> sCG Data in CLOSING MAY be processed (8.3.) + * sTW -> sIV + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sIV, sOP, sCR, sCG, sIV, + }, + [DCCP_PKT_DATAACK] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sPO Ack for Response, move to PARTOPEN (8.1.5.) + * sPO -> sPO Remain in PARTOPEN state + * sOP -> sOP Regular DataAck packet in OPEN state + * sCR -> sCR DataAck in CLOSEREQ MAY be processed (8.3.) + * sCG -> sCG DataAck in CLOSING MAY be processed (8.3.) + * sTW -> sIV + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sPO, sPO, sOP, sCR, sCG, sIV + }, + [DCCP_PKT_CLOSEREQ] = { + /* + * CLOSEREQ may only be sent by the server. + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sIV, sIV, sIV, sIV, sIV + }, + [DCCP_PKT_CLOSE] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sCG Client-initiated close + * sOP -> sCG Client-initiated close + * sCR -> sCG Close in response to CloseReq (8.3.) + * sCG -> sCG Retransmit + * sTW -> sIV Late retransmit, already in TIME_WAIT + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sCG, sCG, sCG, sIV, sIV + }, + [DCCP_PKT_RESET] = { + /* + * sNO -> sIV No connection + * sRQ -> sTW Sync received or timeout, SHOULD send Reset (8.1.1.) + * sRS -> sTW Response received without Request + * sPO -> sTW Timeout, SHOULD send Reset (8.1.5.) + * sOP -> sTW Connection reset + * sCR -> sTW Connection reset + * sCG -> sTW Connection reset + * sTW -> sIG Ignore (don't refresh timer) + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sTW, sTW, sTW, sTW, sTW, sTW, sIG + }, + [DCCP_PKT_SYNC] = { + /* + * We currently ignore Sync packets + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG, + }, + [DCCP_PKT_SYNCACK] = { + /* + * We currently ignore SyncAck packets + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG, + }, + }, + [CT_DCCP_ROLE_SERVER] = { + [DCCP_PKT_REQUEST] = { + /* + * sNO -> sIV Invalid + * sRQ -> sIG Ignore, conntrack might be out of sync + * sRS -> sIG Ignore, conntrack might be out of sync + * sPO -> sIG Ignore, conntrack might be out of sync + * sOP -> sIG Ignore, conntrack might be out of sync + * sCR -> sIG Ignore, conntrack might be out of sync + * sCG -> sIG Ignore, conntrack might be out of sync + * sTW -> sRQ Reincarnation, must reverse roles + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIG, sIG, sIG, sIG, sIG, sIG, sRQ + }, + [DCCP_PKT_RESPONSE] = { + /* + * sNO -> sIV Response without Request + * sRQ -> sRS Response to clients Request + * sRS -> sRS Retransmitted Response (8.1.3. SHOULD NOT) + * sPO -> sIG Response to an ignored Request or late retransmit + * sOP -> sIG Ignore, might be response to ignored Request + * sCR -> sIG Ignore, might be response to ignored Request + * sCG -> sIG Ignore, might be response to ignored Request + * sTW -> sIV Invalid, Request from client in sTW moves to sRQ + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sRS, sRS, sIG, sIG, sIG, sIG, sIV + }, + [DCCP_PKT_ACK] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sOP Enter OPEN state (8.1.5.) + * sOP -> sOP Regular Ack in OPEN state + * sCR -> sIV Waiting for Close from client + * sCG -> sCG Ack in CLOSING MAY be processed (8.3.) + * sTW -> sIV + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sOP, sOP, sIV, sCG, sIV + }, + [DCCP_PKT_DATA] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sOP Enter OPEN state (8.1.5.) + * sOP -> sOP Regular Data packet in OPEN state + * sCR -> sIV Waiting for Close from client + * sCG -> sCG Data in CLOSING MAY be processed (8.3.) + * sTW -> sIV + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sOP, sOP, sIV, sCG, sIV + }, + [DCCP_PKT_DATAACK] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sOP Enter OPEN state (8.1.5.) + * sOP -> sOP Regular DataAck in OPEN state + * sCR -> sIV Waiting for Close from client + * sCG -> sCG Data in CLOSING MAY be processed (8.3.) + * sTW -> sIV + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sOP, sOP, sIV, sCG, sIV + }, + [DCCP_PKT_CLOSEREQ] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sOP -> sCR Move directly to CLOSEREQ (8.1.5.) + * sOP -> sCR CloseReq in OPEN state + * sCR -> sCR Retransmit + * sCG -> sCR Simultaneous close, client sends another Close + * sTW -> sIV Already closed + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sCR, sCR, sCR, sCR, sIV + }, + [DCCP_PKT_CLOSE] = { + /* + * sNO -> sIV No connection + * sRQ -> sIV No connection + * sRS -> sIV No connection + * sPO -> sOP -> sCG Move direcly to CLOSING + * sOP -> sCG Move to CLOSING + * sCR -> sIV Close after CloseReq is invalid + * sCG -> sCG Retransmit + * sTW -> sIV Already closed + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIV, sIV, sIV, sCG, sCG, sIV, sCG, sIV + }, + [DCCP_PKT_RESET] = { + /* + * sNO -> sIV No connection + * sRQ -> sTW Reset in response to Request + * sRS -> sTW Timeout, SHOULD send Reset (8.1.3.) + * sPO -> sTW Timeout, SHOULD send Reset (8.1.3.) + * sOP -> sTW + * sCR -> sTW + * sCG -> sTW + * sTW -> sIG Ignore (don't refresh timer) + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW, sTW */ + sIV, sTW, sTW, sTW, sTW, sTW, sTW, sTW, sIG + }, + [DCCP_PKT_SYNC] = { + /* + * We currently ignore Sync packets + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG, + }, + [DCCP_PKT_SYNCACK] = { + /* + * We currently ignore SyncAck packets + * + * sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */ + sIG, sIG, sIG, sIG, sIG, sIG, sIG, sIG, + }, + }, +}; + +static int dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple) +{ + struct dccp_hdr _hdr, *dh; + + dh = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); + if (dh == NULL) + return 0; + + tuple->src.u.dccp.port = dh->dccph_sport; + tuple->dst.u.dccp.port = dh->dccph_dport; + return 1; +} + +static int dccp_invert_tuple(struct nf_conntrack_tuple *inv, + const struct nf_conntrack_tuple *tuple) +{ + inv->src.u.dccp.port = tuple->dst.u.dccp.port; + inv->dst.u.dccp.port = tuple->src.u.dccp.port; + return 1; +} + +static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) +{ + int pf = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + struct dccp_hdr _dh, *dh; + const char *msg; + u_int8_t state; + + dh = skb_header_pointer(skb, dataoff, sizeof(_dh), &dh); + BUG_ON(dh == NULL); + + state = dccp_state_table[CT_DCCP_ROLE_CLIENT][dh->dccph_type][CT_DCCP_NONE]; + switch (state) { + default: + if (nf_ct_dccp_loose == 0) { + msg = "nf_ct_dccp: not picking up existing connection "; + goto out_invalid; + } + case CT_DCCP_REQUEST: + break; + case CT_DCCP_INVALID: + msg = "nf_ct_dccp: invalid state transition "; + goto out_invalid; + } + + ct->proto.dccp.role[IP_CT_DIR_ORIGINAL] = CT_DCCP_ROLE_CLIENT; + ct->proto.dccp.role[IP_CT_DIR_REPLY] = CT_DCCP_ROLE_SERVER; + ct->proto.dccp.state = CT_DCCP_NONE; + return 1; + +out_invalid: + if (LOG_INVALID(IPPROTO_DCCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg); + return 0; +} + +static u64 dccp_ack_seq(const struct dccp_hdr *dh) +{ + const struct dccp_hdr_ack_bits *dhack; + + dhack = (void *)dh + __dccp_basic_hdr_len(dh); + return ((u64)ntohs(dhack->dccph_ack_nr_high) << 32) + + ntohl(dhack->dccph_ack_nr_low); +} + +static int dccp_packet(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff, enum ip_conntrack_info ctinfo, + int pf, unsigned int hooknum) +{ + enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); + struct dccp_hdr _dh, *dh; + u_int8_t type, old_state, new_state; + enum ct_dccp_roles role; + + dh = skb_header_pointer(skb, dataoff, sizeof(_dh), &dh); + BUG_ON(dh == NULL); + type = dh->dccph_type; + + if (type == DCCP_PKT_RESET && + !test_bit(IPS_SEEN_REPLY_BIT, &ct->status)) { + /* Tear down connection immediately if only reply is a RESET */ + if (del_timer(&ct->timeout)) + ct->timeout.function((unsigned long)ct); + return NF_ACCEPT; + } + + write_lock_bh(&dccp_lock); + + role = ct->proto.dccp.role[dir]; + old_state = ct->proto.dccp.state; + new_state = dccp_state_table[role][type][old_state]; + + switch (new_state) { + case CT_DCCP_REQUEST: + if (old_state == CT_DCCP_TIMEWAIT && + role == CT_DCCP_ROLE_SERVER) { + /* Reincarnation in the reverse direction: reopen and + * reverse client/server roles. */ + ct->proto.dccp.role[dir] = CT_DCCP_ROLE_CLIENT; + ct->proto.dccp.role[!dir] = CT_DCCP_ROLE_SERVER; + } + break; + case CT_DCCP_RESPOND: + if (old_state == CT_DCCP_REQUEST) + ct->proto.dccp.handshake_seq = dccp_hdr_seq(dh); + break; + case CT_DCCP_PARTOPEN: + if (old_state == CT_DCCP_RESPOND && + type == DCCP_PKT_ACK && + dccp_ack_seq(dh) == ct->proto.dccp.handshake_seq) + set_bit(IPS_ASSURED_BIT, &ct->status); + break; + case CT_DCCP_IGNORE: + /* + * Connection tracking might be out of sync, so we ignore + * packets that might establish a new connection and resync + * if the server responds with a valid Response. + */ + if (ct->proto.dccp.last_dir == !dir && + ct->proto.dccp.last_pkt == DCCP_PKT_REQUEST && + type == DCCP_PKT_RESPONSE) { + ct->proto.dccp.role[!dir] = CT_DCCP_ROLE_CLIENT; + ct->proto.dccp.role[dir] = CT_DCCP_ROLE_SERVER; + ct->proto.dccp.handshake_seq = dccp_hdr_seq(dh); + new_state = CT_DCCP_RESPOND; + break; + } + ct->proto.dccp.last_dir = dir; + ct->proto.dccp.last_pkt = type; + + write_unlock_bh(&dccp_lock); + if (LOG_INVALID(IPPROTO_DCCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_dccp: invalid packet ignored "); + return NF_ACCEPT; + case CT_DCCP_INVALID: + write_unlock_bh(&dccp_lock); + if (LOG_INVALID(IPPROTO_DCCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, + "nf_ct_dccp: invalid state transition "); + return -NF_ACCEPT; + } + + ct->proto.dccp.last_dir = dir; + ct->proto.dccp.last_pkt = type; + ct->proto.dccp.state = new_state; + write_unlock_bh(&dccp_lock); + nf_ct_refresh_acct(ct, ctinfo, skb, dccp_timeout[new_state]); + + return NF_ACCEPT; +} + +static int dccp_error(struct sk_buff *skb, unsigned int dataoff, + enum ip_conntrack_info *ctinfo, int pf, + unsigned int hooknum) +{ + struct dccp_hdr _dh, *dh; + unsigned int dccp_len = skb->len - dataoff; + unsigned int cscov; + const char *msg; + + dh = skb_header_pointer(skb, dataoff, sizeof(_dh), &dh); + if (dh == NULL) { + msg = "nf_ct_dccp: short packet "; + goto out_invalid; + } + + if (dh->dccph_doff * 4 < sizeof(struct dccp_hdr) || + dh->dccph_doff * 4 > dccp_len) { + msg = "nf_ct_dccp: truncated/malformed packet "; + goto out_invalid; + } + + cscov = dccp_len; + if (dh->dccph_cscov) { + cscov = (dh->dccph_cscov - 1) * 4; + if (cscov > dccp_len) { + msg = "nf_ct_dccp: bad checksum coverage "; + goto out_invalid; + } + } + + if (nf_conntrack_checksum && hooknum == NF_INET_PRE_ROUTING && + nf_checksum_partial(skb, hooknum, dataoff, cscov, IPPROTO_DCCP, + pf)) { + msg = "nf_ct_dccp: bad checksum "; + goto out_invalid; + } + + if (dh->dccph_type >= DCCP_PKT_INVALID) { + msg = "nf_ct_dccp: reserved packet type "; + goto out_invalid; + } + + return NF_ACCEPT; + +out_invalid: + if (LOG_INVALID(IPPROTO_DCCP)) + nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg); + return -NF_ACCEPT; +} + +static int dccp_print_tuple(struct seq_file *s, + const struct nf_conntrack_tuple *tuple) +{ + return seq_printf(s, "sport=%hu dport=%hu ", + ntohs(tuple->src.u.dccp.port), + ntohs(tuple->dst.u.dccp.port)); +} + +static int dccp_print_conntrack(struct seq_file *s, const struct nf_conn *ct) +{ + return seq_printf(s, "%s ", dccp_state_names[ct->proto.dccp.state]); +} + +#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) +static int dccp_to_nlattr(struct sk_buff *skb, struct nlattr *nla, + const struct nf_conn *ct) +{ + struct nlattr *nest_parms; + + read_lock_bh(&dccp_lock); + nest_parms = nla_nest_start(skb, CTA_PROTOINFO_DCCP | NLA_F_NESTED); + if (!nest_parms) + goto nla_put_failure; + NLA_PUT_U8(skb, CTA_PROTOINFO_DCCP_STATE, ct->proto.dccp.state); + nla_nest_end(skb, nest_parms); + read_unlock_bh(&dccp_lock); + return 0; + +nla_put_failure: + read_unlock_bh(&dccp_lock); + return -1; +} + +static const struct nla_policy dccp_nla_policy[CTA_PROTOINFO_DCCP_MAX + 1] = { + [CTA_PROTOINFO_DCCP_STATE] = { .type = NLA_U8 }, +}; + +static int nlattr_to_dccp(struct nlattr *cda[], struct nf_conn *ct) +{ + struct nlattr *attr = cda[CTA_PROTOINFO_DCCP]; + struct nlattr *tb[CTA_PROTOINFO_DCCP_MAX + 1]; + int err; + + if (!attr) + return 0; + + err = nla_parse_nested(tb, CTA_PROTOINFO_DCCP_MAX, attr, + dccp_nla_policy); + if (err < 0) + return err; + + if (!tb[CTA_PROTOINFO_DCCP_STATE] || + nla_get_u8(tb[CTA_PROTOINFO_DCCP_STATE]) >= CT_DCCP_IGNORE) + return -EINVAL; + + write_lock_bh(&dccp_lock); + ct->proto.dccp.state = nla_get_u8(tb[CTA_PROTOINFO_DCCP_STATE]); + write_unlock_bh(&dccp_lock); + return 0; +} +#endif + +#ifdef CONFIG_SYSCTL +static unsigned int dccp_sysctl_table_users; +static struct ctl_table_header *dccp_sysctl_header; +static ctl_table dccp_sysctl_table[] = { + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_request", + .data = &dccp_timeout[CT_DCCP_REQUEST], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_respond", + .data = &dccp_timeout[CT_DCCP_RESPOND], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_partopen", + .data = &dccp_timeout[CT_DCCP_PARTOPEN], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_open", + .data = &dccp_timeout[CT_DCCP_OPEN], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_closereq", + .data = &dccp_timeout[CT_DCCP_CLOSEREQ], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_closing", + .data = &dccp_timeout[CT_DCCP_CLOSING], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_timeout_timewait", + .data = &dccp_timeout[CT_DCCP_TIMEWAIT], + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_jiffies, + }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "nf_conntrack_dccp_loose", + .data = &nf_ct_dccp_loose, + .maxlen = sizeof(nf_ct_dccp_loose), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .ctl_name = 0, + } +}; +#endif /* CONFIG_SYSCTL */ + +static struct nf_conntrack_l4proto dccp_proto4 __read_mostly = { + .l3proto = AF_INET, + .l4proto = IPPROTO_DCCP, + .name = "dccp", + .pkt_to_tuple = dccp_pkt_to_tuple, + .invert_tuple = dccp_invert_tuple, + .new = dccp_new, + .packet = dccp_packet, + .error = dccp_error, + .print_tuple = dccp_print_tuple, + .print_conntrack = dccp_print_conntrack, +#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) + .to_nlattr = dccp_to_nlattr, + .from_nlattr = nlattr_to_dccp, + .tuple_to_nlattr = nf_ct_port_tuple_to_nlattr, + .nlattr_to_tuple = nf_ct_port_nlattr_to_tuple, + .nla_policy = nf_ct_port_nla_policy, +#endif +#ifdef CONFIG_SYSCTL + .ctl_table_users = &dccp_sysctl_table_users, + .ctl_table_header = &dccp_sysctl_header, + .ctl_table = dccp_sysctl_table, +#endif +}; + +static struct nf_conntrack_l4proto dccp_proto6 __read_mostly = { + .l3proto = AF_INET6, + .l4proto = IPPROTO_DCCP, + .name = "dccp", + .pkt_to_tuple = dccp_pkt_to_tuple, + .invert_tuple = dccp_invert_tuple, + .new = dccp_new, + .packet = dccp_packet, + .error = dccp_error, + .print_tuple = dccp_print_tuple, + .print_conntrack = dccp_print_conntrack, +#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) + .to_nlattr = dccp_to_nlattr, + .from_nlattr = nlattr_to_dccp, + .tuple_to_nlattr = nf_ct_port_tuple_to_nlattr, + .nlattr_to_tuple = nf_ct_port_nlattr_to_tuple, + .nla_policy = nf_ct_port_nla_policy, +#endif +#ifdef CONFIG_SYSCTL + .ctl_table_users = &dccp_sysctl_table_users, + .ctl_table_header = &dccp_sysctl_header, + .ctl_table = dccp_sysctl_table, +#endif +}; + +static int __init nf_conntrack_proto_dccp_init(void) +{ + int err; + + err = nf_conntrack_l4proto_register(&dccp_proto4); + if (err < 0) + goto err1; + + err = nf_conntrack_l4proto_register(&dccp_proto6); + if (err < 0) + goto err2; + return 0; + +err2: + nf_conntrack_l4proto_unregister(&dccp_proto4); +err1: + return err; +} + +static void __exit nf_conntrack_proto_dccp_fini(void) +{ + nf_conntrack_l4proto_unregister(&dccp_proto6); + nf_conntrack_l4proto_unregister(&dccp_proto4); +} + +module_init(nf_conntrack_proto_dccp_init); +module_exit(nf_conntrack_proto_dccp_fini); + +MODULE_AUTHOR("Patrick McHardy "); +MODULE_DESCRIPTION("DCCP connection tracking protocol helper"); +MODULE_LICENSE("GPL"); -- cgit v1.2.3 From 8c87238b726e543f8af4bdb4296020a328df4744 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 14 Apr 2008 11:15:51 +0200 Subject: [NETFILTER]: nf_nat: don't add NAT extension for confirmed conntracks Adding extensions to confirmed conntracks is not allowed to avoid races on reallocation. Don't setup NAT for confirmed conntracks in case NAT module is loaded late. The has one side-effect, the connections existing before the NAT module was loaded won't enter the bysource hash. The only case where this actually makes a difference is in case of SNAT to a multirange where the IP before NAT is also part of the range. Since old connections don't enter the bysource hash the first new connection from the IP will have a new address selected. This shouldn't matter at all. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_nat_rule.h | 3 --- net/ipv4/netfilter/nf_nat_rule.c | 19 ------------------- net/ipv4/netfilter/nf_nat_standalone.c | 8 ++++---- 3 files changed, 4 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_nat_rule.h b/include/net/netfilter/nf_nat_rule.h index 75d1825031d7..e4a18ae361c6 100644 --- a/include/net/netfilter/nf_nat_rule.h +++ b/include/net/netfilter/nf_nat_rule.h @@ -14,7 +14,4 @@ extern int nf_nat_rule_find(struct sk_buff *skb, extern unsigned int alloc_null_binding(struct nf_conn *ct, unsigned int hooknum); - -extern unsigned int -alloc_null_binding_confirmed(struct nf_conn *ct, unsigned int hooknum); #endif /* _NF_NAT_RULE_H */ diff --git a/net/ipv4/netfilter/nf_nat_rule.c b/net/ipv4/netfilter/nf_nat_rule.c index ebe0c7903ae9..e8b4d0d4439e 100644 --- a/net/ipv4/netfilter/nf_nat_rule.c +++ b/net/ipv4/netfilter/nf_nat_rule.c @@ -188,25 +188,6 @@ alloc_null_binding(struct nf_conn *ct, unsigned int hooknum) return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum)); } -unsigned int -alloc_null_binding_confirmed(struct nf_conn *ct, unsigned int hooknum) -{ - __be32 ip - = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC - ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u3.ip - : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u3.ip); - __be16 all - = (HOOK2MANIP(hooknum) == IP_NAT_MANIP_SRC - ? ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.u.all - : ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.u.all); - struct nf_nat_range range - = { IP_NAT_RANGE_MAP_IPS, ip, ip, { all }, { all } }; - - pr_debug("Allocating NULL binding for confirmed %p (%u.%u.%u.%u)\n", - ct, NIPQUAD(ip)); - return nf_nat_setup_info(ct, &range, HOOK2MANIP(hooknum)); -} - int nf_nat_rule_find(struct sk_buff *skb, unsigned int hooknum, const struct net_device *in, diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c index c362f672755a..a366b5865b9c 100644 --- a/net/ipv4/netfilter/nf_nat_standalone.c +++ b/net/ipv4/netfilter/nf_nat_standalone.c @@ -102,6 +102,9 @@ nf_nat_fn(unsigned int hooknum, nat = nfct_nat(ct); if (!nat) { + /* NAT module was loaded late. */ + if (nf_ct_is_confirmed(ct)) + return NF_ACCEPT; nat = nf_ct_ext_add(ct, NF_CT_EXT_NAT, GFP_ATOMIC); if (nat == NULL) { pr_debug("failed to add NAT extension\n"); @@ -127,10 +130,7 @@ nf_nat_fn(unsigned int hooknum, if (!nf_nat_initialized(ct, maniptype)) { unsigned int ret; - if (unlikely(nf_ct_is_confirmed(ct))) - /* NAT module was loaded late */ - ret = alloc_null_binding_confirmed(ct, hooknum); - else if (hooknum == NF_INET_LOCAL_IN) + if (hooknum == NF_INET_LOCAL_IN) /* LOCAL_IN hook doesn't have a chain! */ ret = alloc_null_binding(ct, hooknum); else -- cgit v1.2.3 From 55871d04793d9c069ee277b1e98794b88d92ed80 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 14 Apr 2008 11:15:51 +0200 Subject: [NETFILTER]: nf_conntrack_extend: warn on confirmed conntracks New extensions may only be added to unconfirmed conntracks to avoid races when reallocating the storage. Also change NF_CT_ASSERT to use WARN_ON to get backtraces. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack.h | 9 +-------- net/netfilter/nf_conntrack_extend.c | 3 +++ 2 files changed, 4 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index bb9fc852e973..4a0496aa32d5 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -65,14 +65,7 @@ union nf_conntrack_help { #include #ifdef CONFIG_NETFILTER_DEBUG -#define NF_CT_ASSERT(x) \ -do { \ - if (!(x)) \ - /* Wooah! I'm tripping my conntrack in a frenzy of \ - netplay... */ \ - printk("NF_CT_ASSERT: %s:%i(%s)\n", \ - __FILE__, __LINE__, __FUNCTION__); \ -} while(0) +#define NF_CT_ASSERT(x) WARN_ON(!(x)) #else #define NF_CT_ASSERT(x) #endif diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c index 2bd9963b5b3e..bcc19fa4ed1e 100644 --- a/net/netfilter/nf_conntrack_extend.c +++ b/net/netfilter/nf_conntrack_extend.c @@ -71,6 +71,9 @@ void *__nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp) int i, newlen, newoff; struct nf_ct_ext_type *t; + /* Conntrack must not be confirmed to avoid races on reallocation. */ + NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); + if (!ct->ext) return nf_ct_ext_create(&ct->ext, id, gfp); -- cgit v1.2.3 From dd13b010368f85dfa59364ba87bfe8ae930b2832 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 14 Apr 2008 11:15:52 +0200 Subject: [NETFILTER]: nf_nat: kill helper and seq_adjust hooks Connection tracking helpers (specifically FTP) need to be called before NAT sequence numbers adjustments are performed to be able to compare them against previously seen ones. We've introduced two new hooks around 2.6.11 to maintain this ordering when NAT modules were changed to get called from conntrack helpers directly. The cost of netfilter hooks is quite high and sequence number adjustments are only rarely needed however. Add a RCU-protected sequence number adjustment function pointer and call it from IPv4 conntrack after calling the helper. Signed-off-by: Patrick McHardy --- include/linux/netfilter_ipv4.h | 2 - include/net/netfilter/nf_nat_helper.h | 3 ++ net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 56 +++++++++++++------------- net/ipv4/netfilter/nf_nat_core.c | 5 +++ net/ipv4/netfilter/nf_nat_helper.c | 1 - net/ipv4/netfilter/nf_nat_standalone.c | 35 ---------------- 6 files changed, 35 insertions(+), 67 deletions(-) (limited to 'include') diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h index 9a10092e358c..650318b0c405 100644 --- a/include/linux/netfilter_ipv4.h +++ b/include/linux/netfilter_ipv4.h @@ -62,8 +62,6 @@ enum nf_ip_hook_priorities { NF_IP_PRI_FILTER = 0, NF_IP_PRI_NAT_SRC = 100, NF_IP_PRI_SELINUX_LAST = 225, - NF_IP_PRI_CONNTRACK_HELPER = INT_MAX - 2, - NF_IP_PRI_NAT_SEQ_ADJUST = INT_MAX - 1, NF_IP_PRI_CONNTRACK_CONFIRM = INT_MAX, NF_IP_PRI_LAST = INT_MAX, }; diff --git a/include/net/netfilter/nf_nat_helper.h b/include/net/netfilter/nf_nat_helper.h index 58dd22687949..237a961f40e1 100644 --- a/include/net/netfilter/nf_nat_helper.h +++ b/include/net/netfilter/nf_nat_helper.h @@ -24,6 +24,9 @@ extern int nf_nat_mangle_udp_packet(struct sk_buff *skb, extern int nf_nat_seq_adjust(struct sk_buff *skb, struct nf_conn *ct, enum ip_conntrack_info ctinfo); +extern int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo); /* Setup NAT on this expected conntrack so it follows master, but goes * to port ct->master->saved_proto. */ diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index a65b845c5f15..41e79613eb0a 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -23,6 +23,12 @@ #include #include #include +#include + +int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb, + struct nf_conn *ct, + enum ip_conntrack_info ctinfo); +EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook); static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, struct nf_conntrack_tuple *tuple) @@ -100,36 +106,42 @@ static unsigned int ipv4_confirm(unsigned int hooknum, const struct net_device *in, const struct net_device *out, int (*okfn)(struct sk_buff *)) -{ - /* We've seen it coming out the other side: confirm it */ - return nf_conntrack_confirm(skb); -} - -static unsigned int ipv4_conntrack_help(unsigned int hooknum, - struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) { struct nf_conn *ct; enum ip_conntrack_info ctinfo; const struct nf_conn_help *help; const struct nf_conntrack_helper *helper; + unsigned int ret; /* This is where we call the helper: as the packet goes out. */ ct = nf_ct_get(skb, &ctinfo); if (!ct || ctinfo == IP_CT_RELATED + IP_CT_IS_REPLY) - return NF_ACCEPT; + goto out; help = nfct_help(ct); if (!help) - return NF_ACCEPT; + goto out; + /* rcu_read_lock()ed by nf_hook_slow */ helper = rcu_dereference(help->helper); if (!helper) - return NF_ACCEPT; - return helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb), - ct, ctinfo); + goto out; + + ret = helper->help(skb, skb_network_offset(skb) + ip_hdrlen(skb), + ct, ctinfo); + if (ret != NF_ACCEPT) + return ret; + + if (test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) { + typeof(nf_nat_seq_adjust_hook) seq_adjust; + + seq_adjust = rcu_dereference(nf_nat_seq_adjust_hook); + if (!seq_adjust || !seq_adjust(skb, ct, ctinfo)) + return NF_DROP; + } +out: + /* We've seen it coming out the other side: confirm it */ + return nf_conntrack_confirm(skb); } static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, @@ -210,20 +222,6 @@ static struct nf_hook_ops ipv4_conntrack_ops[] __read_mostly = { .hooknum = NF_INET_LOCAL_OUT, .priority = NF_IP_PRI_CONNTRACK, }, - { - .hook = ipv4_conntrack_help, - .owner = THIS_MODULE, - .pf = PF_INET, - .hooknum = NF_INET_POST_ROUTING, - .priority = NF_IP_PRI_CONNTRACK_HELPER, - }, - { - .hook = ipv4_conntrack_help, - .owner = THIS_MODULE, - .pf = PF_INET, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP_PRI_CONNTRACK_HELPER, - }, { .hook = ipv4_confirm, .owner = THIS_MODULE, diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c index 9320c7ac5729..25c3efe4207e 100644 --- a/net/ipv4/netfilter/nf_nat_core.c +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -618,6 +618,9 @@ static int __init nf_nat_init(void) nf_conntrack_untracked.status |= IPS_NAT_DONE_MASK; l3proto = nf_ct_l3proto_find_get((u_int16_t)AF_INET); + + BUG_ON(nf_nat_seq_adjust_hook != NULL); + rcu_assign_pointer(nf_nat_seq_adjust_hook, nf_nat_seq_adjust); return 0; cleanup_extend: @@ -644,6 +647,8 @@ static void __exit nf_nat_cleanup(void) nf_ct_free_hashtable(bysource, nf_nat_vmalloced, nf_nat_htable_size); nf_ct_l3proto_put(l3proto); nf_ct_extend_unregister(&nat_extend); + rcu_assign_pointer(nf_nat_seq_adjust_hook, NULL); + synchronize_net(); } MODULE_LICENSE("GPL"); diff --git a/net/ipv4/netfilter/nf_nat_helper.c b/net/ipv4/netfilter/nf_nat_helper.c index 2fca727aa8ba..11976ea29884 100644 --- a/net/ipv4/netfilter/nf_nat_helper.c +++ b/net/ipv4/netfilter/nf_nat_helper.c @@ -416,7 +416,6 @@ nf_nat_seq_adjust(struct sk_buff *skb, return 1; } -EXPORT_SYMBOL(nf_nat_seq_adjust); /* Setup NAT on this expected conntrack so it follows master. */ /* If we fail to get a free NAT slot, we'll get dropped on confirm */ diff --git a/net/ipv4/netfilter/nf_nat_standalone.c b/net/ipv4/netfilter/nf_nat_standalone.c index a366b5865b9c..b7dd695691a0 100644 --- a/net/ipv4/netfilter/nf_nat_standalone.c +++ b/net/ipv4/netfilter/nf_nat_standalone.c @@ -245,25 +245,6 @@ nf_nat_local_fn(unsigned int hooknum, return ret; } -static unsigned int -nf_nat_adjust(unsigned int hooknum, - struct sk_buff *skb, - const struct net_device *in, - const struct net_device *out, - int (*okfn)(struct sk_buff *)) -{ - struct nf_conn *ct; - enum ip_conntrack_info ctinfo; - - ct = nf_ct_get(skb, &ctinfo); - if (ct && test_bit(IPS_SEQ_ADJUST_BIT, &ct->status)) { - pr_debug("nf_nat_standalone: adjusting sequence number\n"); - if (!nf_nat_seq_adjust(skb, ct, ctinfo)) - return NF_DROP; - } - return NF_ACCEPT; -} - /* We must be after connection tracking and before packet filtering. */ static struct nf_hook_ops nf_nat_ops[] __read_mostly = { @@ -283,14 +264,6 @@ static struct nf_hook_ops nf_nat_ops[] __read_mostly = { .hooknum = NF_INET_POST_ROUTING, .priority = NF_IP_PRI_NAT_SRC, }, - /* After conntrack, adjust sequence number */ - { - .hook = nf_nat_adjust, - .owner = THIS_MODULE, - .pf = PF_INET, - .hooknum = NF_INET_POST_ROUTING, - .priority = NF_IP_PRI_NAT_SEQ_ADJUST, - }, /* Before packet filtering, change destination */ { .hook = nf_nat_local_fn, @@ -307,14 +280,6 @@ static struct nf_hook_ops nf_nat_ops[] __read_mostly = { .hooknum = NF_INET_LOCAL_IN, .priority = NF_IP_PRI_NAT_SRC, }, - /* After conntrack, adjust sequence number */ - { - .hook = nf_nat_adjust, - .owner = THIS_MODULE, - .pf = PF_INET, - .hooknum = NF_INET_LOCAL_IN, - .priority = NF_IP_PRI_NAT_SEQ_ADJUST, - }, }; static int __init nf_nat_standalone_init(void) -- cgit v1.2.3 From 5e8fbe2ac8a3f1e34e7004c5750ef59bf9304f82 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Mon, 14 Apr 2008 11:15:52 +0200 Subject: [NETFILTER]: nf_conntrack: add tuplehash l3num/protonum accessors Add accessors for l3num and protonum and get rid of some overly long expressions. Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack.h | 10 ++++++ .../netfilter/nf_conntrack_l3proto_ipv4_compat.c | 13 +++----- net/netfilter/nf_conntrack_amanda.c | 4 +-- net/netfilter/nf_conntrack_core.c | 3 +- net/netfilter/nf_conntrack_ftp.c | 4 +-- net/netfilter/nf_conntrack_h323_main.c | 39 ++++++++-------------- net/netfilter/nf_conntrack_netlink.c | 19 ++++------- net/netfilter/nf_conntrack_pptp.c | 4 +-- net/netfilter/nf_conntrack_proto.c | 9 ++--- net/netfilter/nf_conntrack_proto_dccp.c | 3 +- net/netfilter/nf_conntrack_sane.c | 3 +- net/netfilter/nf_conntrack_sip.c | 20 +++++------ net/netfilter/nf_conntrack_standalone.c | 15 +++------ net/netfilter/nf_conntrack_tftp.c | 4 +-- net/netfilter/xt_connlimit.c | 4 +-- net/netfilter/xt_conntrack.c | 4 +-- 16 files changed, 64 insertions(+), 94 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 4a0496aa32d5..26e6a6e2b5a2 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -140,6 +140,16 @@ nf_ct_tuplehash_to_ctrack(const struct nf_conntrack_tuple_hash *hash) tuplehash[hash->tuple.dst.dir]); } +static inline u_int16_t nf_ct_l3num(const struct nf_conn *ct) +{ + return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; +} + +static inline u_int8_t nf_ct_protonum(const struct nf_conn *ct) +{ + return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum; +} + /* get master conntrack via master expectation */ #define master_ct(conntr) (conntr->master) diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c index e60b885d2dcd..40a46d482490 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4_compat.c @@ -106,21 +106,16 @@ static int ct_seq_show(struct seq_file *s, void *v) /* we only want to print DIR_ORIGINAL */ if (NF_CT_DIRECTION(hash)) return 0; - if (ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num != AF_INET) + if (nf_ct_l3num(ct) != AF_INET) return 0; - l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.src.l3num); + l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct)); NF_CT_ASSERT(l3proto); - l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.src.l3num, - ct->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.dst.protonum); + l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); NF_CT_ASSERT(l4proto); if (seq_printf(s, "%-8s %u %ld ", - l4proto->name, - ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum, + l4proto->name, nf_ct_protonum(ct), timer_pending(&ct->timeout) ? (long)(ct->timeout.expires - jiffies)/HZ : 0) != 0) return -ENOSPC; diff --git a/net/netfilter/nf_conntrack_amanda.c b/net/netfilter/nf_conntrack_amanda.c index ddfac99cbe63..38aedeeaf4e1 100644 --- a/net/netfilter/nf_conntrack_amanda.c +++ b/net/netfilter/nf_conntrack_amanda.c @@ -91,7 +91,6 @@ static int amanda_help(struct sk_buff *skb, char pbuf[sizeof("65535")], *tmp; u_int16_t len; __be16 port; - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int ret = NF_ACCEPT; typeof(nf_nat_amanda_hook) nf_nat_amanda; @@ -148,7 +147,8 @@ static int amanda_help(struct sk_buff *skb, goto out; } tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + nf_ct_l3num(ct), &tuple->src.u3, &tuple->dst.u3, IPPROTO_TCP, NULL, &port); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index b77eb56a87e3..21ab0c3846ac 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -194,8 +194,7 @@ destroy_conntrack(struct nf_conntrack *nfct) * destroy_conntrack() MUST NOT be called with a write lock * to nf_conntrack_lock!!! -HW */ rcu_read_lock(); - l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_REPLY].tuple.src.l3num, - ct->tuplehash[IP_CT_DIR_REPLY].tuple.dst.protonum); + l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto && l4proto->destroy) l4proto->destroy(ct); diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c index 87ca39b353bb..bb20672fe036 100644 --- a/net/netfilter/nf_conntrack_ftp.c +++ b/net/netfilter/nf_conntrack_ftp.c @@ -406,7 +406,7 @@ static int help(struct sk_buff *skb, /* Initialize IP/IPv6 addr to expected address (it's not mentioned in EPSV responses) */ - cmd.l3num = ct->tuplehash[dir].tuple.src.l3num; + cmd.l3num = nf_ct_l3num(ct); memcpy(cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all, sizeof(cmd.u3.all)); @@ -453,7 +453,7 @@ static int help(struct sk_buff *skb, daddr = &ct->tuplehash[!dir].tuple.dst.u3; /* Update the ftp info */ - if ((cmd.l3num == ct->tuplehash[dir].tuple.src.l3num) && + if ((cmd.l3num == nf_ct_l3num(ct)) && memcmp(&cmd.u3.all, &ct->tuplehash[dir].tuple.src.u3.all, sizeof(cmd.u3.all))) { /* Enrico Scholz's passive FTP to partially RNAT'd ftp diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index 505052d495cf..c3f87094de43 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -218,7 +218,6 @@ static int get_h245_addr(struct nf_conn *ct, const unsigned char *data, union nf_inet_addr *addr, __be16 *port) { const unsigned char *p; - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int len; if (taddr->choice != eH245_TransportAddress_unicastAddress) @@ -226,13 +225,13 @@ static int get_h245_addr(struct nf_conn *ct, const unsigned char *data, switch (taddr->unicastAddress.choice) { case eUnicastAddress_iPAddress: - if (family != AF_INET) + if (nf_ct_l3num(ct) != AF_INET) return 0; p = data + taddr->unicastAddress.iPAddress.network; len = 4; break; case eUnicastAddress_iP6Address: - if (family != AF_INET6) + if (nf_ct_l3num(ct) != AF_INET6) return 0; p = data + taddr->unicastAddress.iP6Address.network; len = 16; @@ -277,8 +276,7 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, /* Create expect for RTP */ if ((rtp_exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(rtp_exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(rtp_exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_UDP, NULL, &rtp_port); @@ -288,8 +286,7 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, nf_ct_expect_put(rtp_exp); return -1; } - nf_ct_expect_init(rtcp_exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(rtcp_exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_UDP, NULL, &rtcp_port); @@ -346,8 +343,7 @@ static int expect_t120(struct sk_buff *skb, /* Create expect for T.120 connections */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); @@ -634,18 +630,17 @@ int get_h225_addr(struct nf_conn *ct, unsigned char *data, union nf_inet_addr *addr, __be16 *port) { const unsigned char *p; - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int len; switch (taddr->choice) { case eTransportAddress_ipAddress: - if (family != AF_INET) + if (nf_ct_l3num(ct) != AF_INET) return 0; p = data + taddr->ipAddress.ip; len = 4; break; case eTransportAddress_ip6Address: - if (family != AF_INET6) + if (nf_ct_l3num(ct) != AF_INET6) return 0; p = data + taddr->ip6Address.ip; len = 16; @@ -683,8 +678,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct, /* Create expect for h245 connection */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); @@ -792,7 +786,7 @@ static int expect_callforwarding(struct sk_buff *skb, * we don't need to track the second call */ if (callforward_filter && callforward_do_filter(&addr, &ct->tuplehash[!dir].tuple.src.u3, - ct->tuplehash[!dir].tuple.src.l3num)) { + nf_ct_l3num(ct))) { pr_debug("nf_ct_q931: Call Forwarding not tracked\n"); return 0; } @@ -800,8 +794,7 @@ static int expect_callforwarding(struct sk_buff *skb, /* Create expect for the second call leg */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->helper = nf_conntrack_helper_q931; @@ -1272,8 +1265,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct, /* Create expect for Q.931 */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), gkrouted_only ? /* only accept calls from GK? */ &ct->tuplehash[!dir].tuple.src.u3 : NULL, &ct->tuplehash[!dir].tuple.dst.u3, @@ -1344,8 +1336,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct, /* Need new expect */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_UDP, NULL, &port); exp->helper = nf_conntrack_helper_ras; @@ -1549,8 +1540,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct, /* Need new expect */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; @@ -1603,8 +1593,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct, /* Need new expect for call signal */ if ((exp = nf_ct_expect_alloc(ct)) == NULL) return -1; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[!dir].tuple.src.l3num, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 79d5ecde0ddc..16774ecd1c4e 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -145,10 +145,11 @@ nla_put_failure: static inline int ctnetlink_dump_protoinfo(struct sk_buff *skb, const struct nf_conn *ct) { - struct nf_conntrack_l4proto *l4proto = nf_ct_l4proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); + struct nf_conntrack_l4proto *l4proto; struct nlattr *nest_proto; int ret; + l4proto = nf_ct_l4proto_find_get(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (!l4proto->to_nlattr) { nf_ct_l4proto_put(l4proto); return 0; @@ -368,8 +369,7 @@ ctnetlink_fill_info(struct sk_buff *skb, u32 pid, u32 seq, nfmsg = NLMSG_DATA(nlh); nlh->nlmsg_flags = (nowait && pid) ? NLM_F_MULTI : 0; - nfmsg->nfgen_family = - ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + nfmsg->nfgen_family = nf_ct_l3num(ct); nfmsg->version = NFNETLINK_V0; nfmsg->res_id = 0; @@ -454,7 +454,7 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, nfmsg = NLMSG_DATA(nlh); nlh->nlmsg_flags = flags; - nfmsg->nfgen_family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; + nfmsg->nfgen_family = nf_ct_l3num(ct); nfmsg->version = NFNETLINK_V0; nfmsg->res_id = 0; @@ -535,8 +535,6 @@ static int ctnetlink_done(struct netlink_callback *cb) return 0; } -#define L3PROTO(ct) (ct)->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num - static int ctnetlink_dump_table(struct sk_buff *skb, struct netlink_callback *cb) { @@ -558,7 +556,7 @@ restart: /* Dump entries of a given L3 protocol number. * If it is not specified, ie. l3proto == 0, * then dump everything. */ - if (l3proto && L3PROTO(ct) != l3proto) + if (l3proto && nf_ct_l3num(ct) != l3proto) continue; if (cb->args[1]) { if (ct != last) @@ -704,7 +702,7 @@ static int nfnetlink_parse_nat_proto(struct nlattr *attr, if (err < 0) return err; - npt = nf_nat_proto_find_get(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum); + npt = nf_nat_proto_find_get(nf_ct_protonum(ct)); if (npt->nlattr_to_range) err = npt->nlattr_to_range(tb, range); nf_nat_proto_put(npt); @@ -1001,14 +999,11 @@ ctnetlink_change_protoinfo(struct nf_conn *ct, struct nlattr *cda[]) { struct nlattr *tb[CTA_PROTOINFO_MAX+1], *attr = cda[CTA_PROTOINFO]; struct nf_conntrack_l4proto *l4proto; - u_int16_t npt = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum; - u_int16_t l3num = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int err = 0; nla_parse_nested(tb, CTA_PROTOINFO_MAX, attr, NULL); - l4proto = nf_ct_l4proto_find_get(l3num, npt); - + l4proto = nf_ct_l4proto_find_get(nf_ct_l3num(ct), nf_ct_protonum(ct)); if (l4proto->from_nlattr) err = l4proto->from_nlattr(tb, ct); nf_ct_l4proto_put(l4proto); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 8fd83470d1b3..4793cc078789 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -209,7 +209,7 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) /* original direction, PNS->PAC */ dir = IP_CT_DIR_ORIGINAL; nf_ct_expect_init(exp_orig, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[dir].tuple.src.l3num, + nf_ct_l3num(ct), &ct->tuplehash[dir].tuple.src.u3, &ct->tuplehash[dir].tuple.dst.u3, IPPROTO_GRE, &peer_callid, &callid); @@ -218,7 +218,7 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) /* reply direction, PAC->PNS */ dir = IP_CT_DIR_REPLY; nf_ct_expect_init(exp_reply, NF_CT_EXPECT_CLASS_DEFAULT, - ct->tuplehash[dir].tuple.src.l3num, + nf_ct_l3num(ct), &ct->tuplehash[dir].tuple.src.u3, &ct->tuplehash[dir].tuple.dst.u3, IPPROTO_GRE, &callid, &peer_callid); diff --git a/net/netfilter/nf_conntrack_proto.c b/net/netfilter/nf_conntrack_proto.c index 8595b5946acf..a49fc932629b 100644 --- a/net/netfilter/nf_conntrack_proto.c +++ b/net/netfilter/nf_conntrack_proto.c @@ -146,18 +146,15 @@ EXPORT_SYMBOL_GPL(nf_ct_l3proto_module_put); static int kill_l3proto(struct nf_conn *i, void *data) { - return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == - ((struct nf_conntrack_l3proto *)data)->l3proto); + return nf_ct_l3num(i) == ((struct nf_conntrack_l3proto *)data)->l3proto; } static int kill_l4proto(struct nf_conn *i, void *data) { struct nf_conntrack_l4proto *l4proto; l4proto = (struct nf_conntrack_l4proto *)data; - return (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum == - l4proto->l4proto) && - (i->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num == - l4proto->l3proto); + return nf_ct_protonum(i) == l4proto->l4proto && + nf_ct_l3num(i) == l4proto->l3proto; } static int nf_ct_l3proto_register_sysctl(struct nf_conntrack_l3proto *l3proto) diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index db88c5bcc5fd..9376dcd394bd 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -418,7 +418,6 @@ static int dccp_invert_tuple(struct nf_conntrack_tuple *inv, static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff) { - int pf = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; struct dccp_hdr _dh, *dh; const char *msg; u_int8_t state; @@ -447,7 +446,7 @@ static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb, out_invalid: if (LOG_INVALID(IPPROTO_DCCP)) - nf_log_packet(pf, 0, skb, NULL, NULL, NULL, msg); + nf_log_packet(nf_ct_l3num(ct), 0, skb, NULL, NULL, NULL, msg); return 0; } diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c index 7542e25eede3..c3d5e84dcc9b 100644 --- a/net/netfilter/nf_conntrack_sane.c +++ b/net/netfilter/nf_conntrack_sane.c @@ -72,7 +72,6 @@ static int help(struct sk_buff *skb, struct nf_conntrack_tuple *tuple; struct sane_request *req; struct sane_reply_net_start *reply; - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; ct_sane_info = &nfct_help(ct)->help.ct_sane_info; /* Until there's been traffic both ways, don't look in packets. */ @@ -143,7 +142,7 @@ static int help(struct sk_buff *skb, } tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &tuple->src.u3, &tuple->dst.u3, IPPROTO_TCP, NULL, &reply->port); diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index f3915f8724f9..65b3ba57a3b7 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -142,11 +142,10 @@ static int parse_addr(const struct nf_conn *ct, const char *cp, const char *limit) { const char *end; - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; int ret = 0; memset(addr, 0, sizeof(*addr)); - switch (family) { + switch (nf_ct_l3num(ct)) { case AF_INET: ret = in4_pton(cp, limit - cp, (u8 *)&addr->ip, -1, &end); break; @@ -740,7 +739,6 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); union nf_inet_addr *saddr; struct nf_conntrack_tuple tuple; - int family = ct->tuplehash[!dir].tuple.src.l3num; int direct_rtp = 0, skip_expect = 0, ret = NF_DROP; u_int16_t base_port; __be16 rtp_port, rtcp_port; @@ -770,7 +768,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, memset(&tuple, 0, sizeof(tuple)); if (saddr) tuple.src.u3 = *saddr; - tuple.src.l3num = family; + tuple.src.l3num = nf_ct_l3num(ct); tuple.dst.protonum = IPPROTO_UDP; tuple.dst.u3 = *daddr; tuple.dst.u.udp.port = port; @@ -815,13 +813,13 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, rtp_exp = nf_ct_expect_alloc(ct); if (rtp_exp == NULL) goto err1; - nf_ct_expect_init(rtp_exp, class, family, saddr, daddr, + nf_ct_expect_init(rtp_exp, class, nf_ct_l3num(ct), saddr, daddr, IPPROTO_UDP, NULL, &rtp_port); rtcp_exp = nf_ct_expect_alloc(ct); if (rtcp_exp == NULL) goto err2; - nf_ct_expect_init(rtcp_exp, class, family, saddr, daddr, + nf_ct_expect_init(rtcp_exp, class, nf_ct_l3num(ct), saddr, daddr, IPPROTO_UDP, NULL, &rtcp_port); nf_nat_sdp_media = rcu_dereference(nf_nat_sdp_media_hook); @@ -871,7 +869,6 @@ static int process_sdp(struct sk_buff *skb, { enum ip_conntrack_info ctinfo; struct nf_conn *ct = nf_ct_get(skb, &ctinfo); - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; unsigned int matchoff, matchlen; unsigned int mediaoff, medialen; unsigned int sdpoff; @@ -886,8 +883,8 @@ static int process_sdp(struct sk_buff *skb, typeof(nf_nat_sdp_session_hook) nf_nat_sdp_session; nf_nat_sdp_addr = rcu_dereference(nf_nat_sdp_addr_hook); - c_hdr = family == AF_INET ? SDP_HDR_CONNECTION_IP4 : - SDP_HDR_CONNECTION_IP6; + c_hdr = nf_ct_l3num(ct) == AF_INET ? SDP_HDR_CONNECTION_IP4 : + SDP_HDR_CONNECTION_IP6; /* Find beginning of session description */ if (ct_sip_get_sdp_header(ct, *dptr, 0, *datalen, @@ -1034,7 +1031,6 @@ static int process_register_request(struct sk_buff *skb, struct nf_conn *ct = nf_ct_get(skb, &ctinfo); struct nf_conn_help *help = nfct_help(ct); enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo); - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; unsigned int matchoff, matchlen; struct nf_conntrack_expect *exp; union nf_inet_addr *saddr, daddr; @@ -1089,8 +1085,8 @@ static int process_register_request(struct sk_buff *skb, if (sip_direct_signalling) saddr = &ct->tuplehash[!dir].tuple.src.u3; - nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, family, saddr, &daddr, - IPPROTO_UDP, NULL, &port); + nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, nf_ct_l3num(ct), + saddr, &daddr, IPPROTO_UDP, NULL, &port); exp->timeout.expires = sip_timeout * HZ; exp->helper = nfct_help(ct)->helper; exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE; diff --git a/net/netfilter/nf_conntrack_standalone.c b/net/netfilter/nf_conntrack_standalone.c index 01d1f7e178f3..b59871f6bdda 100644 --- a/net/netfilter/nf_conntrack_standalone.c +++ b/net/netfilter/nf_conntrack_standalone.c @@ -127,21 +127,14 @@ static int ct_seq_show(struct seq_file *s, void *v) if (NF_CT_DIRECTION(hash)) return 0; - l3proto = __nf_ct_l3proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.src.l3num); - + l3proto = __nf_ct_l3proto_find(nf_ct_l3num(ct)); NF_CT_ASSERT(l3proto); - l4proto = __nf_ct_l4proto_find(ct->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.src.l3num, - ct->tuplehash[IP_CT_DIR_ORIGINAL] - .tuple.dst.protonum); + l4proto = __nf_ct_l4proto_find(nf_ct_l3num(ct), nf_ct_protonum(ct)); NF_CT_ASSERT(l4proto); if (seq_printf(s, "%-8s %u %-8s %u %ld ", - l3proto->name, - ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num, - l4proto->name, - ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum, + l3proto->name, nf_ct_l3num(ct), + l4proto->name, nf_ct_protonum(ct), timer_pending(&ct->timeout) ? (long)(ct->timeout.expires - jiffies)/HZ : 0) != 0) return -ENOSPC; diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c index a28341b30f21..ea5ff49d77bc 100644 --- a/net/netfilter/nf_conntrack_tftp.c +++ b/net/netfilter/nf_conntrack_tftp.c @@ -44,7 +44,6 @@ static int tftp_help(struct sk_buff *skb, struct nf_conntrack_expect *exp; struct nf_conntrack_tuple *tuple; unsigned int ret = NF_ACCEPT; - int family = ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.l3num; typeof(nf_nat_tftp_hook) nf_nat_tftp; tfh = skb_header_pointer(skb, protoff + sizeof(struct udphdr), @@ -63,7 +62,8 @@ static int tftp_help(struct sk_buff *skb, if (exp == NULL) return NF_DROP; tuple = &ct->tuplehash[IP_CT_DIR_REPLY].tuple; - nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, family, + nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, + nf_ct_l3num(ct), &tuple->src.u3, &tuple->dst.u3, IPPROTO_UDP, NULL, &tuple->dst.u.udp.port); diff --git a/net/netfilter/xt_connlimit.c b/net/netfilter/xt_connlimit.c index 0ca9fe9da203..2e89a00df92c 100644 --- a/net/netfilter/xt_connlimit.c +++ b/net/netfilter/xt_connlimit.c @@ -72,9 +72,7 @@ connlimit_iphash6(const union nf_inet_addr *addr, static inline bool already_closed(const struct nf_conn *conn) { - u_int16_t proto = conn->tuplehash[0].tuple.dst.protonum; - - if (proto == IPPROTO_TCP) + if (nf_ct_protonum(conn) == IPPROTO_TCP) return conn->proto.tcp.state == TCP_CONNTRACK_TIME_WAIT; else return 0; diff --git a/net/netfilter/xt_conntrack.c b/net/netfilter/xt_conntrack.c index 0c50b2894055..d61412f58ef7 100644 --- a/net/netfilter/xt_conntrack.c +++ b/net/netfilter/xt_conntrack.c @@ -65,7 +65,7 @@ conntrack_mt_v0(const struct sk_buff *skb, const struct net_device *in, } if (sinfo->flags & XT_CONNTRACK_PROTO && - FWINV(ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum != + FWINV(nf_ct_protonum(ct) != sinfo->tuple[IP_CT_DIR_ORIGINAL].dst.protonum, XT_CONNTRACK_PROTO)) return false; @@ -174,7 +174,7 @@ ct_proto_port_check(const struct xt_conntrack_mtinfo1 *info, tuple = &ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple; if ((info->match_flags & XT_CONNTRACK_PROTO) && - (tuple->dst.protonum == info->l4proto) ^ + (nf_ct_protonum(ct) == info->l4proto) ^ !(info->invert_flags & XT_CONNTRACK_PROTO)) return false; -- cgit v1.2.3 From 9dbae7917899d78a094aceeb5062cd76efa89052 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:52 +0200 Subject: [NETFILTER]: Remove unused callbacks in nf_conntrack_l3proto These functions are never called. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l3proto.h | 11 ----------- 1 file changed, 11 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index b886e3ae6cad..db0825803b16 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -42,17 +42,6 @@ struct nf_conntrack_l3proto int (*print_tuple)(struct seq_file *s, const struct nf_conntrack_tuple *); - /* Returns verdict for packet, or -1 for invalid. */ - int (*packet)(struct nf_conn *ct, - const struct sk_buff *skb, - enum ip_conntrack_info ctinfo); - - /* - * Called when a new connection for this protocol found; - * returns TRUE if it's OK. If so, packet() called next. - */ - int (*new)(struct nf_conn *ct, const struct sk_buff *skb); - /* * Called before tracking. * *dataoff: offset of protocol header (TCP, UDP,...) in skb -- cgit v1.2.3 From 8ce8439a31f723f3aa28adf27fe8797a5678dde1 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:52 +0200 Subject: [NETFILTER]: nf_conntrack: use bool type in struct nf_conntrack_l3proto Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l3proto.h | 8 ++++---- net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 14 +++++++------- net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c | 14 +++++++------- net/netfilter/nf_conntrack_l3proto_generic.c | 12 ++++++------ 4 files changed, 24 insertions(+), 24 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l3proto.h b/include/net/netfilter/nf_conntrack_l3proto.h index db0825803b16..0378676c3dd8 100644 --- a/include/net/netfilter/nf_conntrack_l3proto.h +++ b/include/net/netfilter/nf_conntrack_l3proto.h @@ -28,15 +28,15 @@ struct nf_conntrack_l3proto * Try to fill in the third arg: nhoff is offset of l3 proto * hdr. Return true if possible. */ - int (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int nhoff, - struct nf_conntrack_tuple *tuple); + bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple); /* * Invert the per-proto part of the tuple: ie. turn xmit into reply. * Some packets can't be inverted: return 0 in that case. */ - int (*invert_tuple)(struct nf_conntrack_tuple *inverse, - const struct nf_conntrack_tuple *orig); + bool (*invert_tuple)(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); /* Print out the per-protocol part of the tuple. */ int (*print_tuple)(struct seq_file *s, diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c index 41e79613eb0a..cacb9cb27dab 100644 --- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c +++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c @@ -30,29 +30,29 @@ int (*nf_nat_seq_adjust_hook)(struct sk_buff *skb, enum ip_conntrack_info ctinfo); EXPORT_SYMBOL_GPL(nf_nat_seq_adjust_hook); -static int ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, - struct nf_conntrack_tuple *tuple) +static bool ipv4_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) { const __be32 *ap; __be32 _addrs[2]; ap = skb_header_pointer(skb, nhoff + offsetof(struct iphdr, saddr), sizeof(u_int32_t) * 2, _addrs); if (ap == NULL) - return 0; + return false; tuple->src.u3.ip = ap[0]; tuple->dst.u3.ip = ap[1]; - return 1; + return true; } -static int ipv4_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool ipv4_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->src.u3.ip = orig->dst.u3.ip; tuple->dst.u3.ip = orig->src.u3.ip; - return 1; + return true; } static int ipv4_print_tuple(struct seq_file *s, diff --git a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c index 3717bdf34f6e..85050c072abd 100644 --- a/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c +++ b/net/ipv6/netfilter/nf_conntrack_l3proto_ipv6.c @@ -27,8 +27,8 @@ #include #include -static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, - struct nf_conntrack_tuple *tuple) +static bool ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) { const u_int32_t *ap; u_int32_t _addrs[8]; @@ -36,21 +36,21 @@ static int ipv6_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, ap = skb_header_pointer(skb, nhoff + offsetof(struct ipv6hdr, saddr), sizeof(_addrs), _addrs); if (ap == NULL) - return 0; + return false; memcpy(tuple->src.u3.ip6, ap, sizeof(tuple->src.u3.ip6)); memcpy(tuple->dst.u3.ip6, ap + 4, sizeof(tuple->dst.u3.ip6)); - return 1; + return true; } -static int ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool ipv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { memcpy(tuple->src.u3.ip6, orig->dst.u3.ip6, sizeof(tuple->src.u3.ip6)); memcpy(tuple->dst.u3.ip6, orig->src.u3.ip6, sizeof(tuple->dst.u3.ip6)); - return 1; + return true; } static int ipv6_print_tuple(struct seq_file *s, diff --git a/net/netfilter/nf_conntrack_l3proto_generic.c b/net/netfilter/nf_conntrack_l3proto_generic.c index 8e914e5ffea8..e7eb807fe07d 100644 --- a/net/netfilter/nf_conntrack_l3proto_generic.c +++ b/net/netfilter/nf_conntrack_l3proto_generic.c @@ -31,22 +31,22 @@ #include #include -static int generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, - struct nf_conntrack_tuple *tuple) +static bool generic_pkt_to_tuple(const struct sk_buff *skb, unsigned int nhoff, + struct nf_conntrack_tuple *tuple) { memset(&tuple->src.u3, 0, sizeof(tuple->src.u3)); memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3)); - return 1; + return true; } -static int generic_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool generic_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { memset(&tuple->src.u3, 0, sizeof(tuple->src.u3)); memset(&tuple->dst.u3, 0, sizeof(tuple->dst.u3)); - return 1; + return true; } static int generic_print_tuple(struct seq_file *s, -- cgit v1.2.3 From 09f263cd39751cada63dec2dccc71e67c00bc38c Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:53 +0200 Subject: [NETFILTER]: nf_conntrack: use bool type in struct nf_conntrack_l4proto Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_l4proto.h | 13 +++---- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 25 ++++++------- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 27 +++++++------ net/netfilter/nf_conntrack_proto_dccp.c | 22 +++++------ net/netfilter/nf_conntrack_proto_generic.c | 20 +++++----- net/netfilter/nf_conntrack_proto_gre.c | 25 ++++++------- net/netfilter/nf_conntrack_proto_sctp.c | 33 ++++++++-------- net/netfilter/nf_conntrack_proto_tcp.c | 52 +++++++++++++------------- net/netfilter/nf_conntrack_proto_udp.c | 18 ++++----- net/netfilter/nf_conntrack_proto_udplite.c | 22 +++++------ 10 files changed, 125 insertions(+), 132 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h index efc16eccddb1..723df9d1cc35 100644 --- a/include/net/netfilter/nf_conntrack_l4proto.h +++ b/include/net/netfilter/nf_conntrack_l4proto.h @@ -25,15 +25,14 @@ struct nf_conntrack_l4proto /* Try to fill in the third arg: dataoff is offset past network protocol hdr. Return true if possible. */ - int (*pkt_to_tuple)(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple); + bool (*pkt_to_tuple)(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple); /* Invert the per-proto part of the tuple: ie. turn xmit into reply. * Some packets can't be inverted: return 0 in that case. */ - int (*invert_tuple)(struct nf_conntrack_tuple *inverse, - const struct nf_conntrack_tuple *orig); + bool (*invert_tuple)(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); /* Returns verdict for packet, or -1 for invalid. */ int (*packet)(struct nf_conn *ct, @@ -45,8 +44,8 @@ struct nf_conntrack_l4proto /* Called when a new connection for this protocol found; * returns TRUE if it's OK. If so, packet() called next. */ - int (*new)(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff); + bool (*new)(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff); /* Called when a conntrack entry is destroyed */ void (*destroy)(struct nf_conn *ct); diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 6873fddb3529..193a845fe7f8 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -22,22 +22,21 @@ static unsigned long nf_ct_icmp_timeout __read_mostly = 30*HZ; -static int icmp_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool icmp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { const struct icmphdr *hp; struct icmphdr _hdr; hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hp == NULL) - return 0; + return false; tuple->dst.u.icmp.type = hp->type; tuple->src.u.icmp.id = hp->un.echo.id; tuple->dst.u.icmp.code = hp->code; - return 1; + return true; } /* Add 1; spaces filled with 0. */ @@ -52,17 +51,17 @@ static const u_int8_t invmap[] = { [ICMP_ADDRESSREPLY] = ICMP_ADDRESS + 1 }; -static int icmp_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool icmp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { if (orig->dst.u.icmp.type >= sizeof(invmap) || !invmap[orig->dst.u.icmp.type]) - return 0; + return false; tuple->src.u.icmp.id = orig->src.u.icmp.id; tuple->dst.u.icmp.type = invmap[orig->dst.u.icmp.type] - 1; tuple->dst.u.icmp.code = orig->dst.u.icmp.code; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -101,8 +100,8 @@ static int icmp_packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int icmp_new(struct nf_conn *ct, - const struct sk_buff *skb, unsigned int dataoff) +static bool icmp_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { static const u_int8_t valid_new[] = { [ICMP_ECHO] = 1, @@ -117,10 +116,10 @@ static int icmp_new(struct nf_conn *ct, pr_debug("icmp: can't create new conn with type %u\n", ct->tuplehash[0].tuple.dst.u.icmp.type); NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple); - return 0; + return false; } atomic_set(&ct->proto.icmp.count, 0); - return 1; + return true; } /* Returns conntrack if it dealt with ICMP, and filled in skb fields */ diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index 0897d0f4c4a2..9ad40e0e17fc 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -28,21 +28,21 @@ static unsigned long nf_ct_icmpv6_timeout __read_mostly = 30*HZ; -static int icmpv6_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool icmpv6_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { const struct icmp6hdr *hp; struct icmp6hdr _hdr; hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hp == NULL) - return 0; + return false; tuple->dst.u.icmp.type = hp->icmp6_type; tuple->src.u.icmp.id = hp->icmp6_identifier; tuple->dst.u.icmp.code = hp->icmp6_code; - return 1; + return true; } /* Add 1; spaces filled with 0. */ @@ -53,17 +53,17 @@ static const u_int8_t invmap[] = { [ICMPV6_NI_REPLY - 128] = ICMPV6_NI_REPLY +1 }; -static int icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { int type = orig->dst.u.icmp.type - 128; if (type < 0 || type >= sizeof(invmap) || !invmap[type]) - return 0; + return false; tuple->src.u.icmp.id = orig->src.u.icmp.id; tuple->dst.u.icmp.type = invmap[type] - 1; tuple->dst.u.icmp.code = orig->dst.u.icmp.code; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -102,9 +102,8 @@ static int icmpv6_packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int icmpv6_new(struct nf_conn *ct, - const struct sk_buff *skb, - unsigned int dataoff) +static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { static const u_int8_t valid_new[] = { [ICMPV6_ECHO_REQUEST - 128] = 1, @@ -117,10 +116,10 @@ static int icmpv6_new(struct nf_conn *ct, pr_debug("icmpv6: can't create new conn with type %u\n", type + 128); NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple); - return 0; + return false; } atomic_set(&ct->proto.icmp.count, 0); - return 1; + return true; } static int diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c index 9376dcd394bd..afb4a1861d2c 100644 --- a/net/netfilter/nf_conntrack_proto_dccp.c +++ b/net/netfilter/nf_conntrack_proto_dccp.c @@ -393,30 +393,30 @@ dccp_state_table[CT_DCCP_ROLE_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = }, }; -static int dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool dccp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { struct dccp_hdr _hdr, *dh; dh = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (dh == NULL) - return 0; + return false; tuple->src.u.dccp.port = dh->dccph_sport; tuple->dst.u.dccp.port = dh->dccph_dport; - return 1; + return true; } -static int dccp_invert_tuple(struct nf_conntrack_tuple *inv, - const struct nf_conntrack_tuple *tuple) +static bool dccp_invert_tuple(struct nf_conntrack_tuple *inv, + const struct nf_conntrack_tuple *tuple) { inv->src.u.dccp.port = tuple->dst.u.dccp.port; inv->dst.u.dccp.port = tuple->src.u.dccp.port; - return 1; + return true; } -static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff) +static bool dccp_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { struct dccp_hdr _dh, *dh; const char *msg; @@ -442,12 +442,12 @@ static int dccp_new(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.dccp.role[IP_CT_DIR_ORIGINAL] = CT_DCCP_ROLE_CLIENT; ct->proto.dccp.role[IP_CT_DIR_REPLY] = CT_DCCP_ROLE_SERVER; ct->proto.dccp.state = CT_DCCP_NONE; - return 1; + return true; out_invalid: if (LOG_INVALID(IPPROTO_DCCP)) nf_log_packet(nf_ct_l3num(ct), 0, skb, NULL, NULL, NULL, msg); - return 0; + return false; } static u64 dccp_ack_seq(const struct dccp_hdr *dh) diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c index 55458915575f..e31b0e7bd0b1 100644 --- a/net/netfilter/nf_conntrack_proto_generic.c +++ b/net/netfilter/nf_conntrack_proto_generic.c @@ -14,23 +14,23 @@ static unsigned int nf_ct_generic_timeout __read_mostly = 600*HZ; -static int generic_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool generic_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { tuple->src.u.all = 0; tuple->dst.u.all = 0; - return 1; + return true; } -static int generic_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool generic_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->src.u.all = 0; tuple->dst.u.all = 0; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -53,10 +53,10 @@ static int packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int new(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff) +static bool new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { - return 1; + return true; } #ifdef CONFIG_SYSCTL diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index e10024a1b666..7d37a2ea67b2 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -148,18 +148,17 @@ EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_destroy); /* PUBLIC CONNTRACK PROTO HELPER FUNCTIONS */ /* invert gre part of tuple */ -static int gre_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool gre_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->dst.u.gre.key = orig->src.u.gre.key; tuple->src.u.gre.key = orig->dst.u.gre.key; - return 1; + return true; } /* gre hdr info to tuple */ -static int gre_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool gre_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { const struct gre_hdr_pptp *pgrehdr; struct gre_hdr_pptp _pgrehdr; @@ -173,24 +172,24 @@ static int gre_pkt_to_tuple(const struct sk_buff *skb, /* try to behave like "nf_conntrack_proto_generic" */ tuple->src.u.all = 0; tuple->dst.u.all = 0; - return 1; + return true; } /* PPTP header is variable length, only need up to the call_id field */ pgrehdr = skb_header_pointer(skb, dataoff, 8, &_pgrehdr); if (!pgrehdr) - return 1; + return true; if (ntohs(grehdr->protocol) != GRE_PROTOCOL_PPTP) { pr_debug("GRE_VERSION_PPTP but unknown proto\n"); - return 0; + return false; } tuple->dst.u.gre.key = pgrehdr->call_id; srckey = gre_keymap_lookup(tuple); tuple->src.u.gre.key = srckey; - return 1; + return true; } /* print gre part of tuple */ @@ -235,8 +234,8 @@ static int gre_packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int gre_new(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff) +static bool gre_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { pr_debug(": "); NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); @@ -246,7 +245,7 @@ static int gre_new(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.gre.stream_timeout = GRE_STREAM_TIMEOUT; ct->proto.gre.timeout = GRE_TIMEOUT; - return 1; + return true; } /* Called when a conntrack entry has already been removed from the hashes diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c index f9a08370dbb3..2d47351f70dc 100644 --- a/net/netfilter/nf_conntrack_proto_sctp.c +++ b/net/netfilter/nf_conntrack_proto_sctp.c @@ -130,28 +130,27 @@ static const u8 sctp_conntracks[2][9][SCTP_CONNTRACK_MAX] = { } }; -static int sctp_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool sctp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { sctp_sctphdr_t _hdr, *hp; /* Actually only need first 8 bytes. */ hp = skb_header_pointer(skb, dataoff, 8, &_hdr); if (hp == NULL) - return 0; + return false; tuple->src.u.sctp.port = hp->source; tuple->dst.u.sctp.port = hp->dest; - return 1; + return true; } -static int sctp_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool sctp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->src.u.sctp.port = orig->dst.u.sctp.port; tuple->dst.u.sctp.port = orig->src.u.sctp.port; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -390,8 +389,8 @@ out: } /* Called when a new connection for this protocol found. */ -static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff) +static bool sctp_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { enum sctp_conntrack new_state; sctp_sctphdr_t _sctph, *sh; @@ -401,16 +400,16 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb, sh = skb_header_pointer(skb, dataoff, sizeof(_sctph), &_sctph); if (sh == NULL) - return 0; + return false; if (do_basic_checks(ct, skb, dataoff, map) != 0) - return 0; + return false; /* If an OOTB packet has any of these chunks discard (Sec 8.4) */ if (test_bit(SCTP_CID_ABORT, map) || test_bit(SCTP_CID_SHUTDOWN_COMPLETE, map) || test_bit(SCTP_CID_COOKIE_ACK, map)) - return 0; + return false; new_state = SCTP_CONNTRACK_MAX; for_each_sctp_chunk (skb, sch, _sch, offset, dataoff, count) { @@ -422,7 +421,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb, if (new_state == SCTP_CONNTRACK_NONE || new_state == SCTP_CONNTRACK_MAX) { pr_debug("nf_conntrack_sctp: invalid new deleting.\n"); - return 0; + return false; } /* Copy the vtag into the state info */ @@ -433,7 +432,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb, ih = skb_header_pointer(skb, offset + sizeof(sctp_chunkhdr_t), sizeof(_inithdr), &_inithdr); if (ih == NULL) - return 0; + return false; pr_debug("Setting vtag %x for new conn\n", ih->init_tag); @@ -442,7 +441,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb, ih->init_tag; } else { /* Sec 8.5.1 (A) */ - return 0; + return false; } } /* If it is a shutdown ack OOTB packet, we expect a return @@ -456,7 +455,7 @@ static int sctp_new(struct nf_conn *ct, const struct sk_buff *skb, ct->proto.sctp.state = new_state; } - return 1; + return true; } #ifdef CONFIG_SYSCTL diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 57831c75fa9f..73a8b32db7be 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -257,9 +257,8 @@ static const u8 tcp_conntracks[2][6][TCP_CONNTRACK_MAX] = { } }; -static int tcp_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool tcp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { const struct tcphdr *hp; struct tcphdr _hdr; @@ -267,20 +266,20 @@ static int tcp_pkt_to_tuple(const struct sk_buff *skb, /* Actually only need first 8 bytes. */ hp = skb_header_pointer(skb, dataoff, 8, &_hdr); if (hp == NULL) - return 0; + return false; tuple->src.u.tcp.port = hp->source; tuple->dst.u.tcp.port = hp->dest; - return 1; + return true; } -static int tcp_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool tcp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->src.u.tcp.port = orig->dst.u.tcp.port; tuple->dst.u.tcp.port = orig->src.u.tcp.port; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -478,20 +477,20 @@ static void tcp_sack(const struct sk_buff *skb, unsigned int dataoff, } } -static int tcp_in_window(const struct nf_conn *ct, - struct ip_ct_tcp *state, - enum ip_conntrack_dir dir, - unsigned int index, - const struct sk_buff *skb, - unsigned int dataoff, - const struct tcphdr *tcph, - int pf) +static bool tcp_in_window(const struct nf_conn *ct, + struct ip_ct_tcp *state, + enum ip_conntrack_dir dir, + unsigned int index, + const struct sk_buff *skb, + unsigned int dataoff, + const struct tcphdr *tcph, + int pf) { struct ip_ct_tcp_state *sender = &state->seen[dir]; struct ip_ct_tcp_state *receiver = &state->seen[!dir]; const struct nf_conntrack_tuple *tuple = &ct->tuplehash[dir].tuple; __u32 seq, ack, sack, end, win, swin; - int res; + bool res; /* * Get the required data from the packet. @@ -657,12 +656,12 @@ static int tcp_in_window(const struct nf_conn *ct, state->retrans = 0; } } - res = 1; + res = true; } else { - res = 0; + res = false; if (sender->flags & IP_CT_TCP_FLAG_BE_LIBERAL || nf_ct_tcp_be_liberal) - res = 1; + res = true; if (!res && LOG_INVALID(IPPROTO_TCP)) nf_log_packet(pf, 0, skb, NULL, NULL, NULL, "nf_ct_tcp: %s ", @@ -676,7 +675,7 @@ static int tcp_in_window(const struct nf_conn *ct, : "SEQ is over the upper bound (over the window of the receiver)"); } - pr_debug("tcp_in_window: res=%i sender end=%u maxend=%u maxwin=%u " + pr_debug("tcp_in_window: res=%u sender end=%u maxend=%u maxwin=%u " "receiver end=%u maxend=%u maxwin=%u\n", res, sender->td_end, sender->td_maxend, sender->td_maxwin, receiver->td_end, receiver->td_maxend, receiver->td_maxwin); @@ -982,9 +981,8 @@ static int tcp_packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int tcp_new(struct nf_conn *ct, - const struct sk_buff *skb, - unsigned int dataoff) +static bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { enum tcp_conntrack new_state; const struct tcphdr *th; @@ -1003,7 +1001,7 @@ static int tcp_new(struct nf_conn *ct, /* Invalid: delete conntrack */ if (new_state >= TCP_CONNTRACK_MAX) { pr_debug("nf_ct_tcp: invalid new deleting.\n"); - return 0; + return false; } if (new_state == TCP_CONNTRACK_SYN_SENT) { @@ -1021,7 +1019,7 @@ static int tcp_new(struct nf_conn *ct, ct->proto.tcp.seen[1].flags = 0; } else if (nf_ct_tcp_loose == 0) { /* Don't try to pick up connections. */ - return 0; + return false; } else { /* * We are in the middle of a connection, @@ -1061,7 +1059,7 @@ static int tcp_new(struct nf_conn *ct, sender->td_scale, receiver->td_end, receiver->td_maxend, receiver->td_maxwin, receiver->td_scale); - return 1; + return true; } #if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c index b8a35cc06416..8b21762e65de 100644 --- a/net/netfilter/nf_conntrack_proto_udp.c +++ b/net/netfilter/nf_conntrack_proto_udp.c @@ -26,7 +26,7 @@ static unsigned int nf_ct_udp_timeout __read_mostly = 30*HZ; static unsigned int nf_ct_udp_timeout_stream __read_mostly = 180*HZ; -static int udp_pkt_to_tuple(const struct sk_buff *skb, +static bool udp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff, struct nf_conntrack_tuple *tuple) { @@ -36,20 +36,20 @@ static int udp_pkt_to_tuple(const struct sk_buff *skb, /* Actually only need first 8 bytes. */ hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hp == NULL) - return 0; + return false; tuple->src.u.udp.port = hp->source; tuple->dst.u.udp.port = hp->dest; - return 1; + return true; } -static int udp_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool udp_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->src.u.udp.port = orig->dst.u.udp.port; tuple->dst.u.udp.port = orig->src.u.udp.port; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -83,10 +83,10 @@ static int udp_packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int udp_new(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff) +static bool udp_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { - return 1; + return true; } static int udp_error(struct sk_buff *skb, unsigned int dataoff, diff --git a/net/netfilter/nf_conntrack_proto_udplite.c b/net/netfilter/nf_conntrack_proto_udplite.c index c3eaee6afffd..1fa62f3c24f1 100644 --- a/net/netfilter/nf_conntrack_proto_udplite.c +++ b/net/netfilter/nf_conntrack_proto_udplite.c @@ -27,28 +27,28 @@ static unsigned int nf_ct_udplite_timeout __read_mostly = 30*HZ; static unsigned int nf_ct_udplite_timeout_stream __read_mostly = 180*HZ; -static int udplite_pkt_to_tuple(const struct sk_buff *skb, - unsigned int dataoff, - struct nf_conntrack_tuple *tuple) +static bool udplite_pkt_to_tuple(const struct sk_buff *skb, + unsigned int dataoff, + struct nf_conntrack_tuple *tuple) { const struct udphdr *hp; struct udphdr _hdr; hp = skb_header_pointer(skb, dataoff, sizeof(_hdr), &_hdr); if (hp == NULL) - return 0; + return false; tuple->src.u.udp.port = hp->source; tuple->dst.u.udp.port = hp->dest; - return 1; + return true; } -static int udplite_invert_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple *orig) +static bool udplite_invert_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple *orig) { tuple->src.u.udp.port = orig->dst.u.udp.port; tuple->dst.u.udp.port = orig->src.u.udp.port; - return 1; + return true; } /* Print out the per-protocol part of the tuple. */ @@ -83,10 +83,10 @@ static int udplite_packet(struct nf_conn *ct, } /* Called when a new connection for this protocol found. */ -static int udplite_new(struct nf_conn *ct, const struct sk_buff *skb, - unsigned int dataoff) +static bool udplite_new(struct nf_conn *ct, const struct sk_buff *skb, + unsigned int dataoff) { - return 1; + return true; } static int udplite_error(struct sk_buff *skb, unsigned int dataoff, -- cgit v1.2.3 From 5f2b4c9006fc667c4614f0b079efab3721f68316 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:53 +0200 Subject: [NETFILTER]: nf_conntrack: use bool type in struct nf_conntrack_tuple.h Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack.h | 11 ++++----- include/net/netfilter/nf_conntrack_core.h | 4 +-- include/net/netfilter/nf_conntrack_tuple.h | 39 ++++++++++++++++-------------- net/netfilter/nf_conntrack_core.c | 22 ++++++++--------- 4 files changed, 38 insertions(+), 38 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index 26e6a6e2b5a2..2dbd6c015b94 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -194,12 +194,11 @@ extern void nf_conntrack_hash_insert(struct nf_conn *ct); extern void nf_conntrack_flush(void); -extern int nf_ct_get_tuplepr(const struct sk_buff *skb, - unsigned int nhoff, - u_int16_t l3num, - struct nf_conntrack_tuple *tuple); -extern int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, - const struct nf_conntrack_tuple *orig); +extern bool nf_ct_get_tuplepr(const struct sk_buff *skb, + unsigned int nhoff, u_int16_t l3num, + struct nf_conntrack_tuple *tuple); +extern bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig); extern void __nf_ct_refresh_acct(struct nf_conn *ct, enum ip_conntrack_info ctinfo, diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 9ee26469c759..a81771210934 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -30,7 +30,7 @@ extern void nf_conntrack_cleanup(void); extern int nf_conntrack_proto_init(void); extern void nf_conntrack_proto_fini(void); -extern int +extern bool nf_ct_get_tuple(const struct sk_buff *skb, unsigned int nhoff, unsigned int dataoff, @@ -40,7 +40,7 @@ nf_ct_get_tuple(const struct sk_buff *skb, const struct nf_conntrack_l3proto *l3proto, const struct nf_conntrack_l4proto *l4proto); -extern int +extern bool nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_l3proto *l3proto, diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index bdeec3461384..2722b13ecd6c 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -166,61 +166,64 @@ struct nf_conntrack_tuple_hash #endif /* __KERNEL__ */ -static inline int __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1, - const struct nf_conntrack_tuple *t2) +static inline bool __nf_ct_tuple_src_equal(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2) { return (nf_inet_addr_cmp(&t1->src.u3, &t2->src.u3) && t1->src.u.all == t2->src.u.all && t1->src.l3num == t2->src.l3num); } -static inline int __nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1, - const struct nf_conntrack_tuple *t2) +static inline bool __nf_ct_tuple_dst_equal(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2) { return (nf_inet_addr_cmp(&t1->dst.u3, &t2->dst.u3) && t1->dst.u.all == t2->dst.u.all && t1->dst.protonum == t2->dst.protonum); } -static inline int nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1, - const struct nf_conntrack_tuple *t2) +static inline bool nf_ct_tuple_equal(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2) { return __nf_ct_tuple_src_equal(t1, t2) && __nf_ct_tuple_dst_equal(t1, t2); } -static inline int nf_ct_tuple_mask_equal(const struct nf_conntrack_tuple_mask *m1, - const struct nf_conntrack_tuple_mask *m2) +static inline bool +nf_ct_tuple_mask_equal(const struct nf_conntrack_tuple_mask *m1, + const struct nf_conntrack_tuple_mask *m2) { return (nf_inet_addr_cmp(&m1->src.u3, &m2->src.u3) && m1->src.u.all == m2->src.u.all); } -static inline int nf_ct_tuple_src_mask_cmp(const struct nf_conntrack_tuple *t1, - const struct nf_conntrack_tuple *t2, - const struct nf_conntrack_tuple_mask *mask) +static inline bool +nf_ct_tuple_src_mask_cmp(const struct nf_conntrack_tuple *t1, + const struct nf_conntrack_tuple *t2, + const struct nf_conntrack_tuple_mask *mask) { int count; for (count = 0; count < NF_CT_TUPLE_L3SIZE; count++) { if ((t1->src.u3.all[count] ^ t2->src.u3.all[count]) & mask->src.u3.all[count]) - return 0; + return false; } if ((t1->src.u.all ^ t2->src.u.all) & mask->src.u.all) - return 0; + return false; if (t1->src.l3num != t2->src.l3num || t1->dst.protonum != t2->dst.protonum) - return 0; + return false; - return 1; + return true; } -static inline int nf_ct_tuple_mask_cmp(const struct nf_conntrack_tuple *t, - const struct nf_conntrack_tuple *tuple, - const struct nf_conntrack_tuple_mask *mask) +static inline bool +nf_ct_tuple_mask_cmp(const struct nf_conntrack_tuple *t, + const struct nf_conntrack_tuple *tuple, + const struct nf_conntrack_tuple_mask *mask) { return nf_ct_tuple_src_mask_cmp(t, tuple, mask) && __nf_ct_tuple_dst_equal(t, tuple); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 21ab0c3846ac..a3fe9db412d3 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -94,7 +94,7 @@ static inline u_int32_t hash_conntrack(const struct nf_conntrack_tuple *tuple) nf_conntrack_hash_rnd); } -int +bool nf_ct_get_tuple(const struct sk_buff *skb, unsigned int nhoff, unsigned int dataoff, @@ -108,7 +108,7 @@ nf_ct_get_tuple(const struct sk_buff *skb, tuple->src.l3num = l3num; if (l3proto->pkt_to_tuple(skb, nhoff, tuple) == 0) - return 0; + return false; tuple->dst.protonum = protonum; tuple->dst.dir = IP_CT_DIR_ORIGINAL; @@ -117,10 +117,8 @@ nf_ct_get_tuple(const struct sk_buff *skb, } EXPORT_SYMBOL_GPL(nf_ct_get_tuple); -int nf_ct_get_tuplepr(const struct sk_buff *skb, - unsigned int nhoff, - u_int16_t l3num, - struct nf_conntrack_tuple *tuple) +bool nf_ct_get_tuplepr(const struct sk_buff *skb, unsigned int nhoff, + u_int16_t l3num, struct nf_conntrack_tuple *tuple) { struct nf_conntrack_l3proto *l3proto; struct nf_conntrack_l4proto *l4proto; @@ -134,7 +132,7 @@ int nf_ct_get_tuplepr(const struct sk_buff *skb, ret = l3proto->get_l4proto(skb, nhoff, &protoff, &protonum); if (ret != NF_ACCEPT) { rcu_read_unlock(); - return 0; + return false; } l4proto = __nf_ct_l4proto_find(l3num, protonum); @@ -147,7 +145,7 @@ int nf_ct_get_tuplepr(const struct sk_buff *skb, } EXPORT_SYMBOL_GPL(nf_ct_get_tuplepr); -int +bool nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, const struct nf_conntrack_tuple *orig, const struct nf_conntrack_l3proto *l3proto, @@ -157,7 +155,7 @@ nf_ct_invert_tuple(struct nf_conntrack_tuple *inverse, inverse->src.l3num = orig->src.l3num; if (l3proto->invert_tuple(inverse, orig) == 0) - return 0; + return false; inverse->dst.dir = !orig->dst.dir; @@ -738,10 +736,10 @@ nf_conntrack_in(int pf, unsigned int hooknum, struct sk_buff *skb) } EXPORT_SYMBOL_GPL(nf_conntrack_in); -int nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, - const struct nf_conntrack_tuple *orig) +bool nf_ct_invert_tuplepr(struct nf_conntrack_tuple *inverse, + const struct nf_conntrack_tuple *orig) { - int ret; + bool ret; rcu_read_lock(); ret = nf_ct_invert_tuple(inverse, orig, -- cgit v1.2.3 From f2ea825f483d5d78754ae813b6db63f8b74e9343 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:53 +0200 Subject: [NETFILTER]: nf_nat: use bool type in nf_nat_proto Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_nat_protocol.h | 42 +++++++++++++++---------------- net/ipv4/netfilter/nf_nat_core.c | 8 +++--- net/ipv4/netfilter/nf_nat_proto_common.c | 24 +++++++++--------- net/ipv4/netfilter/nf_nat_proto_dccp.c | 10 ++++---- net/ipv4/netfilter/nf_nat_proto_gre.c | 18 ++++++------- net/ipv4/netfilter/nf_nat_proto_icmp.c | 14 +++++------ net/ipv4/netfilter/nf_nat_proto_sctp.c | 8 +++--- net/ipv4/netfilter/nf_nat_proto_tcp.c | 10 ++++---- net/ipv4/netfilter/nf_nat_proto_udp.c | 8 +++--- net/ipv4/netfilter/nf_nat_proto_udplite.c | 8 +++--- net/ipv4/netfilter/nf_nat_proto_unknown.c | 24 +++++++++--------- 11 files changed, 87 insertions(+), 87 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_nat_protocol.h b/include/net/netfilter/nf_nat_protocol.h index fba94a2028d5..f3662c4394ef 100644 --- a/include/net/netfilter/nf_nat_protocol.h +++ b/include/net/netfilter/nf_nat_protocol.h @@ -15,25 +15,25 @@ struct nf_nat_protocol /* Translate a packet to the target according to manip type. Return true if succeeded. */ - int (*manip_pkt)(struct sk_buff *skb, - unsigned int iphdroff, - const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype); + bool (*manip_pkt)(struct sk_buff *skb, + unsigned int iphdroff, + const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype); /* Is the manipable part of the tuple between min and max incl? */ - int (*in_range)(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max); + bool (*in_range)(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max); /* Alter the per-proto part of the tuple (depending on maniptype), to give a unique tuple in the given range if possible; return false if not. Per-protocol part of tuple is initialized to the incoming packet. */ - int (*unique_tuple)(struct nf_conntrack_tuple *tuple, - const struct nf_nat_range *range, - enum nf_nat_manip_type maniptype, - const struct nf_conn *ct); + bool (*unique_tuple)(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct); int (*range_to_nlattr)(struct sk_buff *skb, const struct nf_nat_range *range); @@ -59,16 +59,16 @@ extern int init_protocols(void) __init; extern void cleanup_protocols(void); extern const struct nf_nat_protocol *find_nat_proto(u_int16_t protonum); -extern int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max); +extern bool nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max); -extern int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_nat_range *range, - enum nf_nat_manip_type maniptype, - const struct nf_conn *ct, - u_int16_t *rover); +extern bool nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct, + u_int16_t *rover); extern int nf_nat_proto_range_to_nlattr(struct sk_buff *skb, const struct nf_nat_range *range); diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c index 25c3efe4207e..07a2fbc59622 100644 --- a/net/ipv4/netfilter/nf_nat_core.c +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -349,7 +349,7 @@ nf_nat_setup_info(struct nf_conn *ct, EXPORT_SYMBOL(nf_nat_setup_info); /* Returns true if succeeded. */ -static int +static bool manip_pkt(u_int16_t proto, struct sk_buff *skb, unsigned int iphdroff, @@ -360,7 +360,7 @@ manip_pkt(u_int16_t proto, const struct nf_nat_protocol *p; if (!skb_make_writable(skb, iphdroff + sizeof(*iph))) - return 0; + return false; iph = (void *)skb->data + iphdroff; @@ -369,7 +369,7 @@ manip_pkt(u_int16_t proto, /* rcu_read_lock()ed by nf_hook_slow */ p = __nf_nat_proto_find(proto); if (!p->manip_pkt(skb, iphdroff, target, maniptype)) - return 0; + return false; iph = (void *)skb->data + iphdroff; @@ -380,7 +380,7 @@ manip_pkt(u_int16_t proto, csum_replace4(&iph->check, iph->daddr, target->dst.u3.ip); iph->daddr = target->dst.u3.ip; } - return 1; + return true; } /* Do packet manipulations according to nf_nat_setup_info. */ diff --git a/net/ipv4/netfilter/nf_nat_proto_common.c b/net/ipv4/netfilter/nf_nat_proto_common.c index 4904b86265e1..91537f11273f 100644 --- a/net/ipv4/netfilter/nf_nat_proto_common.c +++ b/net/ipv4/netfilter/nf_nat_proto_common.c @@ -17,10 +17,10 @@ #include #include -int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type maniptype, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max) +bool nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type maniptype, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) { __be16 port; @@ -34,11 +34,11 @@ int nf_nat_proto_in_range(const struct nf_conntrack_tuple *tuple, } EXPORT_SYMBOL_GPL(nf_nat_proto_in_range); -int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_nat_range *range, - enum nf_nat_manip_type maniptype, - const struct nf_conn *ct, - u_int16_t *rover) +bool nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct, + u_int16_t *rover) { unsigned int range_size, min, i; __be16 *portptr; @@ -53,7 +53,7 @@ int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, if (!(range->flags & IP_NAT_RANGE_PROTO_SPECIFIED)) { /* If it's dst rewrite, can't change port */ if (maniptype == IP_NAT_MANIP_DST) - return 0; + return false; if (ntohs(*portptr) < 1024) { /* Loose convention: >> 512 is credential passing */ @@ -83,9 +83,9 @@ int nf_nat_proto_unique_tuple(struct nf_conntrack_tuple *tuple, continue; if (!(range->flags & IP_NAT_RANGE_PROTO_RANDOM)) *rover = off; - return 1; + return true; } - return 0; + return false; } EXPORT_SYMBOL_GPL(nf_nat_proto_unique_tuple); diff --git a/net/ipv4/netfilter/nf_nat_proto_dccp.c b/net/ipv4/netfilter/nf_nat_proto_dccp.c index 12b51b38442e..f78eb26e9a20 100644 --- a/net/ipv4/netfilter/nf_nat_proto_dccp.c +++ b/net/ipv4/netfilter/nf_nat_proto_dccp.c @@ -22,7 +22,7 @@ static u_int16_t dccp_port_rover; -static int +static bool dccp_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -32,7 +32,7 @@ dccp_unique_tuple(struct nf_conntrack_tuple *tuple, &dccp_port_rover); } -static int +static bool dccp_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, @@ -49,7 +49,7 @@ dccp_manip_pkt(struct sk_buff *skb, hdrsize = sizeof(struct dccp_hdr); if (!skb_make_writable(skb, hdroff + hdrsize)) - return 0; + return false; iph = (struct iphdr *)(skb->data + iphdroff); hdr = (struct dccp_hdr *)(skb->data + hdroff); @@ -70,12 +70,12 @@ dccp_manip_pkt(struct sk_buff *skb, *portptr = newport; if (hdrsize < sizeof(*hdr)) - return 1; + return true; inet_proto_csum_replace4(&hdr->dccph_checksum, skb, oldip, newip, 1); inet_proto_csum_replace2(&hdr->dccph_checksum, skb, oldport, newport, 0); - return 1; + return true; } static const struct nf_nat_protocol nf_nat_protocol_dccp = { diff --git a/net/ipv4/netfilter/nf_nat_proto_gre.c b/net/ipv4/netfilter/nf_nat_proto_gre.c index 84bb7854137a..4c4af5a6d6c8 100644 --- a/net/ipv4/netfilter/nf_nat_proto_gre.c +++ b/net/ipv4/netfilter/nf_nat_proto_gre.c @@ -37,7 +37,7 @@ MODULE_AUTHOR("Harald Welte "); MODULE_DESCRIPTION("Netfilter NAT protocol helper module for GRE"); /* generate unique tuple ... */ -static int +static bool gre_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -50,7 +50,7 @@ gre_unique_tuple(struct nf_conntrack_tuple *tuple, /* If there is no master conntrack we are not PPTP, do not change tuples */ if (!ct->master) - return 0; + return false; if (maniptype == IP_NAT_MANIP_SRC) keyptr = &tuple->src.u.gre.key; @@ -71,15 +71,15 @@ gre_unique_tuple(struct nf_conntrack_tuple *tuple, for (i = 0; i < range_size; i++, key++) { *keyptr = htons(min + key % range_size); if (!nf_nat_used_tuple(tuple, ct)) - return 1; + return true; } pr_debug("%p: no NAT mapping\n", ct); - return 0; + return false; } /* manipulate a GRE packet according to maniptype */ -static int +static bool gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, enum nf_nat_manip_type maniptype) @@ -92,7 +92,7 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, /* pgreh includes two optional 32bit fields which are not required * to be there. That's where the magic '8' comes from */ if (!skb_make_writable(skb, hdroff + sizeof(*pgreh) - 8)) - return 0; + return false; greh = (void *)skb->data + hdroff; pgreh = (struct gre_hdr_pptp *)greh; @@ -100,7 +100,7 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, /* we only have destination manip of a packet, since 'source key' * is not present in the packet itself */ if (maniptype != IP_NAT_MANIP_DST) - return 1; + return true; switch (greh->version) { case GRE_VERSION_1701: /* We do not currently NAT any GREv0 packets. @@ -112,9 +112,9 @@ gre_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, break; default: pr_debug("can't nat unknown GRE version\n"); - return 0; + return false; } - return 1; + return true; } static const struct nf_nat_protocol gre = { diff --git a/net/ipv4/netfilter/nf_nat_proto_icmp.c b/net/ipv4/netfilter/nf_nat_proto_icmp.c index ab3a0ec2a2d1..19a8b0b07d8e 100644 --- a/net/ipv4/netfilter/nf_nat_proto_icmp.c +++ b/net/ipv4/netfilter/nf_nat_proto_icmp.c @@ -17,7 +17,7 @@ #include #include -static int +static bool icmp_in_range(const struct nf_conntrack_tuple *tuple, enum nf_nat_manip_type maniptype, const union nf_conntrack_man_proto *min, @@ -27,7 +27,7 @@ icmp_in_range(const struct nf_conntrack_tuple *tuple, ntohs(tuple->src.u.icmp.id) <= ntohs(max->icmp.id); } -static int +static bool icmp_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -46,12 +46,12 @@ icmp_unique_tuple(struct nf_conntrack_tuple *tuple, tuple->src.u.icmp.id = htons(ntohs(range->min.icmp.id) + (id % range_size)); if (!nf_nat_used_tuple(tuple, ct)) - return 1; + return true; } - return 0; + return false; } -static int +static bool icmp_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, @@ -62,13 +62,13 @@ icmp_manip_pkt(struct sk_buff *skb, unsigned int hdroff = iphdroff + iph->ihl*4; if (!skb_make_writable(skb, hdroff + sizeof(*hdr))) - return 0; + return false; hdr = (struct icmphdr *)(skb->data + hdroff); inet_proto_csum_replace2(&hdr->checksum, skb, hdr->un.echo.id, tuple->src.u.icmp.id, 0); hdr->un.echo.id = tuple->src.u.icmp.id; - return 1; + return true; } const struct nf_nat_protocol nf_nat_protocol_icmp = { diff --git a/net/ipv4/netfilter/nf_nat_proto_sctp.c b/net/ipv4/netfilter/nf_nat_proto_sctp.c index 3d3faa9d5f6d..82e4c0e286b8 100644 --- a/net/ipv4/netfilter/nf_nat_proto_sctp.c +++ b/net/ipv4/netfilter/nf_nat_proto_sctp.c @@ -16,7 +16,7 @@ static u_int16_t nf_sctp_port_rover; -static int +static bool sctp_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -26,7 +26,7 @@ sctp_unique_tuple(struct nf_conntrack_tuple *tuple, &nf_sctp_port_rover); } -static int +static bool sctp_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, @@ -39,7 +39,7 @@ sctp_manip_pkt(struct sk_buff *skb, u32 crc32; if (!skb_make_writable(skb, hdroff + sizeof(*hdr))) - return 0; + return false; iph = (struct iphdr *)(skb->data + iphdroff); hdr = (struct sctphdr *)(skb->data + hdroff); @@ -63,7 +63,7 @@ sctp_manip_pkt(struct sk_buff *skb, crc32 = sctp_end_cksum(crc32); hdr->checksum = htonl(crc32); - return 1; + return true; } static const struct nf_nat_protocol nf_nat_protocol_sctp = { diff --git a/net/ipv4/netfilter/nf_nat_proto_tcp.c b/net/ipv4/netfilter/nf_nat_proto_tcp.c index 5d4c8a0e89c0..399e2cfa263b 100644 --- a/net/ipv4/netfilter/nf_nat_proto_tcp.c +++ b/net/ipv4/netfilter/nf_nat_proto_tcp.c @@ -20,7 +20,7 @@ static u_int16_t tcp_port_rover; -static int +static bool tcp_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -30,7 +30,7 @@ tcp_unique_tuple(struct nf_conntrack_tuple *tuple, &tcp_port_rover); } -static int +static bool tcp_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, @@ -50,7 +50,7 @@ tcp_manip_pkt(struct sk_buff *skb, hdrsize = sizeof(struct tcphdr); if (!skb_make_writable(skb, hdroff + hdrsize)) - return 0; + return false; iph = (struct iphdr *)(skb->data + iphdroff); hdr = (struct tcphdr *)(skb->data + hdroff); @@ -73,11 +73,11 @@ tcp_manip_pkt(struct sk_buff *skb, *portptr = newport; if (hdrsize < sizeof(*hdr)) - return 1; + return true; inet_proto_csum_replace4(&hdr->check, skb, oldip, newip, 1); inet_proto_csum_replace2(&hdr->check, skb, oldport, newport, 0); - return 1; + return true; } const struct nf_nat_protocol nf_nat_protocol_tcp = { diff --git a/net/ipv4/netfilter/nf_nat_proto_udp.c b/net/ipv4/netfilter/nf_nat_proto_udp.c index 74a7e7b63465..9e61c79492e4 100644 --- a/net/ipv4/netfilter/nf_nat_proto_udp.c +++ b/net/ipv4/netfilter/nf_nat_proto_udp.c @@ -19,7 +19,7 @@ static u_int16_t udp_port_rover; -static int +static bool udp_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -29,7 +29,7 @@ udp_unique_tuple(struct nf_conntrack_tuple *tuple, &udp_port_rover); } -static int +static bool udp_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, @@ -42,7 +42,7 @@ udp_manip_pkt(struct sk_buff *skb, __be16 *portptr, newport; if (!skb_make_writable(skb, hdroff + sizeof(*hdr))) - return 0; + return false; iph = (struct iphdr *)(skb->data + iphdroff); hdr = (struct udphdr *)(skb->data + hdroff); @@ -68,7 +68,7 @@ udp_manip_pkt(struct sk_buff *skb, hdr->check = CSUM_MANGLED_0; } *portptr = newport; - return 1; + return true; } const struct nf_nat_protocol nf_nat_protocol_udp = { diff --git a/net/ipv4/netfilter/nf_nat_proto_udplite.c b/net/ipv4/netfilter/nf_nat_proto_udplite.c index b29346d0e7ab..440a229bbd87 100644 --- a/net/ipv4/netfilter/nf_nat_proto_udplite.c +++ b/net/ipv4/netfilter/nf_nat_proto_udplite.c @@ -18,7 +18,7 @@ static u_int16_t udplite_port_rover; -static int +static bool udplite_unique_tuple(struct nf_conntrack_tuple *tuple, const struct nf_nat_range *range, enum nf_nat_manip_type maniptype, @@ -28,7 +28,7 @@ udplite_unique_tuple(struct nf_conntrack_tuple *tuple, &udplite_port_rover); } -static int +static bool udplite_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, @@ -41,7 +41,7 @@ udplite_manip_pkt(struct sk_buff *skb, __be16 *portptr, newport; if (!skb_make_writable(skb, hdroff + sizeof(*hdr))) - return 0; + return false; iph = (struct iphdr *)(skb->data + iphdroff); hdr = (struct udphdr *)(skb->data + hdroff); @@ -66,7 +66,7 @@ udplite_manip_pkt(struct sk_buff *skb, hdr->check = CSUM_MANGLED_0; *portptr = newport; - return 1; + return true; } static const struct nf_nat_protocol nf_nat_protocol_udplite = { diff --git a/net/ipv4/netfilter/nf_nat_proto_unknown.c b/net/ipv4/netfilter/nf_nat_proto_unknown.c index cda21ff0e4cf..14381c62acea 100644 --- a/net/ipv4/netfilter/nf_nat_proto_unknown.c +++ b/net/ipv4/netfilter/nf_nat_proto_unknown.c @@ -18,31 +18,31 @@ #include #include -static int unknown_in_range(const struct nf_conntrack_tuple *tuple, - enum nf_nat_manip_type manip_type, - const union nf_conntrack_man_proto *min, - const union nf_conntrack_man_proto *max) +static bool unknown_in_range(const struct nf_conntrack_tuple *tuple, + enum nf_nat_manip_type manip_type, + const union nf_conntrack_man_proto *min, + const union nf_conntrack_man_proto *max) { - return 1; + return true; } -static int unknown_unique_tuple(struct nf_conntrack_tuple *tuple, - const struct nf_nat_range *range, - enum nf_nat_manip_type maniptype, - const struct nf_conn *ct) +static bool unknown_unique_tuple(struct nf_conntrack_tuple *tuple, + const struct nf_nat_range *range, + enum nf_nat_manip_type maniptype, + const struct nf_conn *ct) { /* Sorry: we can't help you; if it's not unique, we can't frob anything. */ - return 0; + return false; } -static int +static bool unknown_manip_pkt(struct sk_buff *skb, unsigned int iphdroff, const struct nf_conntrack_tuple *tuple, enum nf_nat_manip_type maniptype) { - return 1; + return true; } const struct nf_nat_protocol nf_nat_unknown_protocol = { -- cgit v1.2.3 From 3c9fba656a185cf56872a325e5594d9b4d4168ec Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 14 Apr 2008 11:15:54 +0200 Subject: [NETFILTER]: nf_conntrack: replace NF_CT_DUMP_TUPLE macro indrection by function call Directly call IPv4 and IPv6 variants where the address family is easily known. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/net/netfilter/nf_conntrack_tuple.h | 2 -- net/ipv4/netfilter/ipt_CLUSTERIP.c | 2 +- net/ipv4/netfilter/nf_conntrack_proto_icmp.c | 2 +- net/ipv4/netfilter/nf_nat_pptp.c | 2 +- net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c | 2 +- net/netfilter/nf_conntrack_core.c | 2 +- net/netfilter/nf_conntrack_h323_main.c | 26 +++++++++++++------------- net/netfilter/nf_conntrack_pptp.c | 4 ++-- net/netfilter/nf_conntrack_proto_gre.c | 6 +++--- net/netfilter/nf_conntrack_proto_tcp.c | 6 +++--- net/netfilter/nf_conntrack_sane.c | 2 +- net/netfilter/nf_conntrack_tftp.c | 6 +++--- 12 files changed, 30 insertions(+), 32 deletions(-) (limited to 'include') diff --git a/include/net/netfilter/nf_conntrack_tuple.h b/include/net/netfilter/nf_conntrack_tuple.h index 2722b13ecd6c..1bb7087833d3 100644 --- a/include/net/netfilter/nf_conntrack_tuple.h +++ b/include/net/netfilter/nf_conntrack_tuple.h @@ -151,8 +151,6 @@ static inline void nf_ct_dump_tuple(const struct nf_conntrack_tuple *t) } } -#define NF_CT_DUMP_TUPLE(tp) nf_ct_dump_tuple(tp) - /* If we're the first tuple, it's the original dir. */ #define NF_CT_DIRECTION(h) \ ((enum ip_conntrack_dir)(h)->tuple.dst.dir) diff --git a/net/ipv4/netfilter/ipt_CLUSTERIP.c b/net/ipv4/netfilter/ipt_CLUSTERIP.c index 2510d4fcdb54..c1f970cb7221 100644 --- a/net/ipv4/netfilter/ipt_CLUSTERIP.c +++ b/net/ipv4/netfilter/ipt_CLUSTERIP.c @@ -331,7 +331,7 @@ clusterip_tg(struct sk_buff *skb, const struct net_device *in, } #ifdef DEBUG - NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + nf_ct_dump_tuple_ip(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); #endif pr_debug("hash=%u ct_hash=%u ", hash, ct->mark); if (!clusterip_responsible(cipinfo->config, hash)) { diff --git a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c index 193a845fe7f8..78ab19accace 100644 --- a/net/ipv4/netfilter/nf_conntrack_proto_icmp.c +++ b/net/ipv4/netfilter/nf_conntrack_proto_icmp.c @@ -115,7 +115,7 @@ static bool icmp_new(struct nf_conn *ct, const struct sk_buff *skb, /* Can't create a new ICMP `conn' with this. */ pr_debug("icmp: can't create new conn with type %u\n", ct->tuplehash[0].tuple.dst.u.icmp.type); - NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple); + nf_ct_dump_tuple_ip(&ct->tuplehash[0].tuple); return false; } atomic_set(&ct->proto.icmp.count, 0); diff --git a/net/ipv4/netfilter/nf_nat_pptp.c b/net/ipv4/netfilter/nf_nat_pptp.c index 3a1e6d6afc0a..da3d91a5ef5c 100644 --- a/net/ipv4/netfilter/nf_nat_pptp.c +++ b/net/ipv4/netfilter/nf_nat_pptp.c @@ -72,7 +72,7 @@ static void pptp_nat_expected(struct nf_conn *ct, } pr_debug("trying to unexpect other dir: "); - NF_CT_DUMP_TUPLE(&t); + nf_ct_dump_tuple_ip(&t); other_exp = nf_ct_expect_find_get(&t); if (other_exp) { nf_ct_unexpect_related(other_exp); diff --git a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c index 9ad40e0e17fc..ee713b03e9ec 100644 --- a/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c +++ b/net/ipv6/netfilter/nf_conntrack_proto_icmpv6.c @@ -115,7 +115,7 @@ static bool icmpv6_new(struct nf_conn *ct, const struct sk_buff *skb, /* Can't create a new ICMPv6 `conn' with this. */ pr_debug("icmpv6: can't create new conn with type %u\n", type + 128); - NF_CT_DUMP_TUPLE(&ct->tuplehash[0].tuple); + nf_ct_dump_tuple_ipv6(&ct->tuplehash[0].tuple); return false; } atomic_set(&ct->proto.icmp.count, 0); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index a3fe9db412d3..351237399e27 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -763,7 +763,7 @@ void nf_conntrack_alter_reply(struct nf_conn *ct, NF_CT_ASSERT(!nf_ct_is_confirmed(ct)); pr_debug("Altering reply tuple of %p to ", ct); - NF_CT_DUMP_TUPLE(newreply); + nf_ct_dump_tuple(newreply); ct->tuplehash[IP_CT_DIR_REPLY].tuple = *newreply; if (ct->master || (help && help->expecting != 0)) diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index c3f87094de43..95da1a24aab7 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -303,9 +303,9 @@ static int expect_rtp_rtcp(struct sk_buff *skb, struct nf_conn *ct, if (nf_ct_expect_related(rtp_exp) == 0) { if (nf_ct_expect_related(rtcp_exp) == 0) { pr_debug("nf_ct_h323: expect RTP "); - NF_CT_DUMP_TUPLE(&rtp_exp->tuple); + nf_ct_dump_tuple(&rtp_exp->tuple); pr_debug("nf_ct_h323: expect RTCP "); - NF_CT_DUMP_TUPLE(&rtcp_exp->tuple); + nf_ct_dump_tuple(&rtcp_exp->tuple); } else { nf_ct_unexpect_related(rtp_exp); ret = -1; @@ -360,7 +360,7 @@ static int expect_t120(struct sk_buff *skb, } else { /* Conntrack only */ if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_h323: expect T.120 "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); } else ret = -1; } @@ -582,7 +582,7 @@ static int h245_help(struct sk_buff *skb, unsigned int protoff, while (get_tpkt_data(skb, protoff, ct, ctinfo, &data, &datalen, &dataoff)) { pr_debug("nf_ct_h245: TPKT len=%d ", datalen); - NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + nf_ct_dump_tuple(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); /* Decode H.245 signal */ ret = DecodeMultimediaSystemControlMessage(data, datalen, @@ -695,7 +695,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct, } else { /* Conntrack only */ if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_q931: expect H.245 "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); } else ret = -1; } @@ -810,7 +810,7 @@ static int expect_callforwarding(struct sk_buff *skb, } else { /* Conntrack only */ if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_q931: expect Call Forwarding "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); } else ret = -1; } @@ -1130,7 +1130,7 @@ static int q931_help(struct sk_buff *skb, unsigned int protoff, while (get_tpkt_data(skb, protoff, ct, ctinfo, &data, &datalen, &dataoff)) { pr_debug("nf_ct_q931: TPKT len=%d ", datalen); - NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + nf_ct_dump_tuple(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); /* Decode Q.931 signal */ ret = DecodeQ931(data, datalen, &q931); @@ -1279,7 +1279,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct, } else { /* Conntrack only */ if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); /* Save port for looking up expect in processing RCF */ info->sig_port[dir] = port; @@ -1343,7 +1343,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct, if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_ras: expect RAS "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); } else ret = -1; @@ -1427,7 +1427,7 @@ static int process_rcf(struct sk_buff *skb, struct nf_conn *ct, pr_debug("nf_ct_ras: set Q.931 expect " "timeout to %u seconds for", info->timeout); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); set_expect_timeout(exp, info->timeout); } spin_unlock_bh(&nf_conntrack_lock); @@ -1548,7 +1548,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct, if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); } else ret = -1; @@ -1601,7 +1601,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct, if (nf_ct_expect_related(exp) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); } else ret = -1; @@ -1705,7 +1705,7 @@ static int ras_help(struct sk_buff *skb, unsigned int protoff, if (data == NULL) goto accept; pr_debug("nf_ct_ras: RAS message len=%d ", datalen); - NF_CT_DUMP_TUPLE(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); + nf_ct_dump_tuple(&ct->tuplehash[CTINFO2DIR(ctinfo)].tuple); /* Decode RAS message */ ret = DecodeRasMessage(data, datalen, &ras); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 4793cc078789..97e54b0e43a3 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -119,7 +119,7 @@ static void pptp_expectfn(struct nf_conn *ct, /* obviously this tuple inversion only works until you do NAT */ nf_ct_invert_tuplepr(&inv_t, &exp->tuple); pr_debug("trying to unexpect other dir: "); - NF_CT_DUMP_TUPLE(&inv_t); + nf_ct_dump_tuple(&inv_t); exp_other = nf_ct_expect_find_get(&inv_t); if (exp_other) { @@ -141,7 +141,7 @@ static int destroy_sibling_or_exp(const struct nf_conntrack_tuple *t) struct nf_conn *sibling; pr_debug("trying to timeout ct or exp for tuple "); - NF_CT_DUMP_TUPLE(t); + nf_ct_dump_tuple(t); h = nf_conntrack_find_get(t); if (h) { diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 7d37a2ea67b2..654a4f7f12c6 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -82,7 +82,7 @@ static __be16 gre_keymap_lookup(struct nf_conntrack_tuple *t) read_unlock_bh(&nf_ct_gre_lock); pr_debug("lookup src key 0x%x for ", key); - NF_CT_DUMP_TUPLE(t); + nf_ct_dump_tuple(t); return key; } @@ -113,7 +113,7 @@ int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, *kmp = km; pr_debug("adding new entry %p: ", km); - NF_CT_DUMP_TUPLE(&km->tuple); + nf_ct_dump_tuple(&km->tuple); write_lock_bh(&nf_ct_gre_lock); list_add_tail(&km->list, &gre_keymap_list); @@ -238,7 +238,7 @@ static bool gre_new(struct nf_conn *ct, const struct sk_buff *skb, unsigned int dataoff) { pr_debug(": "); - NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + nf_ct_dump_tuple(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); /* initialize to sane value. Ideally a conntrack helper * (e.g. in case of pptp) is increasing them */ diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 73a8b32db7be..ba94004fe323 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -505,7 +505,7 @@ static bool tcp_in_window(const struct nf_conn *ct, pr_debug("tcp_in_window: START\n"); pr_debug("tcp_in_window: "); - NF_CT_DUMP_TUPLE(tuple); + nf_ct_dump_tuple(tuple); pr_debug("seq=%u ack=%u sack=%u win=%u end=%u\n", seq, ack, sack, win, end); pr_debug("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i " @@ -592,7 +592,7 @@ static bool tcp_in_window(const struct nf_conn *ct, seq = end = sender->td_end; pr_debug("tcp_in_window: "); - NF_CT_DUMP_TUPLE(tuple); + nf_ct_dump_tuple(tuple); pr_debug("seq=%u ack=%u sack =%u win=%u end=%u\n", seq, ack, sack, win, end); pr_debug("tcp_in_window: sender end=%u maxend=%u maxwin=%u scale=%i " @@ -936,7 +936,7 @@ static int tcp_packet(struct nf_conn *ct, ct->proto.tcp.last_dir = dir; pr_debug("tcp_conntracks: "); - NF_CT_DUMP_TUPLE(tuple); + nf_ct_dump_tuple(tuple); pr_debug("syn=%i ack=%i fin=%i rst=%i old=%i new=%i\n", (th->syn ? 1 : 0), (th->ack ? 1 : 0), (th->fin ? 1 : 0), (th->rst ? 1 : 0), diff --git a/net/netfilter/nf_conntrack_sane.c b/net/netfilter/nf_conntrack_sane.c index c3d5e84dcc9b..a94294b2b23c 100644 --- a/net/netfilter/nf_conntrack_sane.c +++ b/net/netfilter/nf_conntrack_sane.c @@ -147,7 +147,7 @@ static int help(struct sk_buff *skb, IPPROTO_TCP, NULL, &reply->port); pr_debug("nf_ct_sane: expect: "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); /* Can't expect this? Best to drop packet now. */ if (nf_ct_expect_related(exp) != 0) diff --git a/net/netfilter/nf_conntrack_tftp.c b/net/netfilter/nf_conntrack_tftp.c index ea5ff49d77bc..f57f6e7a71ee 100644 --- a/net/netfilter/nf_conntrack_tftp.c +++ b/net/netfilter/nf_conntrack_tftp.c @@ -55,8 +55,8 @@ static int tftp_help(struct sk_buff *skb, case TFTP_OPCODE_READ: case TFTP_OPCODE_WRITE: /* RRQ and WRQ works the same way */ - NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); - NF_CT_DUMP_TUPLE(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); + nf_ct_dump_tuple(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple); + nf_ct_dump_tuple(&ct->tuplehash[IP_CT_DIR_REPLY].tuple); exp = nf_ct_expect_alloc(ct); if (exp == NULL) @@ -68,7 +68,7 @@ static int tftp_help(struct sk_buff *skb, IPPROTO_UDP, NULL, &tuple->dst.u.udp.port); pr_debug("expect: "); - NF_CT_DUMP_TUPLE(&exp->tuple); + nf_ct_dump_tuple(&exp->tuple); nf_nat_tftp = rcu_dereference(nf_nat_tftp_hook); if (nf_nat_tftp && ct->status & IPS_NAT_MASK) -- cgit v1.2.3 From e7bfd0a1a6c8f82977253dab19be9d9979c1ec1b Mon Sep 17 00:00:00 2001 From: Peter Warasin Date: Mon, 14 Apr 2008 11:15:54 +0200 Subject: [NETFILTER]: bridge: add ebt_nflog watcher This patch adds the ebtables nflog watcher to the kernel in order to allow ebtables log through the nfnetlink_log backend. Signed-off-by: Peter Warasin Signed-off-by: Patrick McHardy --- include/linux/netfilter_bridge/ebt_nflog.h | 21 +++++++++ net/bridge/netfilter/Kconfig | 14 ++++++ net/bridge/netfilter/Makefile | 1 + net/bridge/netfilter/ebt_nflog.c | 74 ++++++++++++++++++++++++++++++ 4 files changed, 110 insertions(+) create mode 100644 include/linux/netfilter_bridge/ebt_nflog.h create mode 100644 net/bridge/netfilter/ebt_nflog.c (limited to 'include') diff --git a/include/linux/netfilter_bridge/ebt_nflog.h b/include/linux/netfilter_bridge/ebt_nflog.h new file mode 100644 index 000000000000..052817849b83 --- /dev/null +++ b/include/linux/netfilter_bridge/ebt_nflog.h @@ -0,0 +1,21 @@ +#ifndef __LINUX_BRIDGE_EBT_NFLOG_H +#define __LINUX_BRIDGE_EBT_NFLOG_H + +#define EBT_NFLOG_MASK 0x0 + +#define EBT_NFLOG_PREFIX_SIZE 64 +#define EBT_NFLOG_WATCHER "nflog" + +#define EBT_NFLOG_DEFAULT_GROUP 0x1 +#define EBT_NFLOG_DEFAULT_THRESHOLD 1 + +struct ebt_nflog_info { + u_int32_t len; + u_int16_t group; + u_int16_t threshold; + u_int16_t flags; + u_int16_t pad; + char prefix[EBT_NFLOG_PREFIX_SIZE]; +}; + +#endif /* __LINUX_BRIDGE_EBT_NFLOG_H */ diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig index 4a3e2bf892c7..7beeefa0f9c0 100644 --- a/net/bridge/netfilter/Kconfig +++ b/net/bridge/netfilter/Kconfig @@ -212,4 +212,18 @@ config BRIDGE_EBT_ULOG To compile it as a module, choose M here. If unsure, say N. +config BRIDGE_EBT_NFLOG + tristate "ebt: nflog support" + depends on BRIDGE_NF_EBTABLES + help + This option enables the nflog watcher, which allows to LOG + messages through the netfilter logging API, which can use + either the old LOG target, the old ULOG target or nfnetlink_log + as backend. + + This option adds the ulog watcher, that you can use in any rule + in any ebtables table. + + To compile it as a module, choose M here. If unsure, say N. + endmenu diff --git a/net/bridge/netfilter/Makefile b/net/bridge/netfilter/Makefile index 905087e0d485..83715d73a503 100644 --- a/net/bridge/netfilter/Makefile +++ b/net/bridge/netfilter/Makefile @@ -30,3 +30,4 @@ obj-$(CONFIG_BRIDGE_EBT_SNAT) += ebt_snat.o # watchers obj-$(CONFIG_BRIDGE_EBT_LOG) += ebt_log.o obj-$(CONFIG_BRIDGE_EBT_ULOG) += ebt_ulog.o +obj-$(CONFIG_BRIDGE_EBT_NFLOG) += ebt_nflog.o diff --git a/net/bridge/netfilter/ebt_nflog.c b/net/bridge/netfilter/ebt_nflog.c new file mode 100644 index 000000000000..8e799aa9e560 --- /dev/null +++ b/net/bridge/netfilter/ebt_nflog.c @@ -0,0 +1,74 @@ +/* + * ebt_nflog + * + * Author: + * Peter Warasin + * + * February, 2008 + * + * Based on: + * xt_NFLOG.c, (C) 2006 by Patrick McHardy + * ebt_ulog.c, (C) 2004 by Bart De Schuymer + * + */ + +#include +#include +#include +#include +#include + +static void ebt_nflog(const struct sk_buff *skb, + unsigned int hooknr, + const struct net_device *in, + const struct net_device *out, + const void *data, unsigned int datalen) +{ + struct ebt_nflog_info *info = (struct ebt_nflog_info *)data; + struct nf_loginfo li; + + li.type = NF_LOG_TYPE_ULOG; + li.u.ulog.copy_len = info->len; + li.u.ulog.group = info->group; + li.u.ulog.qthreshold = info->threshold; + + nf_log_packet(PF_BRIDGE, hooknr, skb, in, out, &li, "%s", info->prefix); +} + +static int ebt_nflog_check(const char *tablename, + unsigned int hookmask, + const struct ebt_entry *e, + void *data, unsigned int datalen) +{ + struct ebt_nflog_info *info = (struct ebt_nflog_info *)data; + + if (datalen != EBT_ALIGN(sizeof(struct ebt_nflog_info))) + return -EINVAL; + if (info->flags & ~EBT_NFLOG_MASK) + return -EINVAL; + info->prefix[EBT_NFLOG_PREFIX_SIZE - 1] = '\0'; + return 0; +} + +static struct ebt_watcher nflog __read_mostly = { + .name = EBT_NFLOG_WATCHER, + .watcher = ebt_nflog, + .check = ebt_nflog_check, + .me = THIS_MODULE, +}; + +static int __init ebt_nflog_init(void) +{ + return ebt_register_watcher(&nflog); +} + +static void __exit ebt_nflog_fini(void) +{ + ebt_unregister_watcher(&nflog); +} + +module_init(ebt_nflog_init); +module_exit(ebt_nflog_fini); +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Peter Warasin "); +MODULE_DESCRIPTION("ebtables NFLOG netfilter logging module"); -- cgit v1.2.3 From 0c3141e910eaaa0b617e2f26c69b266d1cd1f035 Mon Sep 17 00:00:00 2001 From: Allan Stephens Date: Tue, 15 Apr 2008 00:22:02 -0700 Subject: [TIPC]: Overhaul of socket locking logic This patch modifies TIPC's socket code to follow the same approach used by other protocols. This change eliminates the need for a mutex in the TIPC-specific portion of the socket protocol data structure -- in its place, the standard Linux socket backlog queue and associated locking routines are utilized. These changes fix a long-standing receive queue bug on SMP systems, and also enable individual read and write threads to utilize a socket without unnecessarily interfering with each other. Signed-off-by: Allan Stephens Signed-off-by: David S. Miller --- include/net/tipc/tipc_port.h | 6 + net/tipc/port.c | 31 +- net/tipc/socket.c | 1002 ++++++++++++++++++++++++------------------ 3 files changed, 608 insertions(+), 431 deletions(-) (limited to 'include') diff --git a/include/net/tipc/tipc_port.h b/include/net/tipc/tipc_port.h index c9b36b77a0b9..11105bcc4457 100644 --- a/include/net/tipc/tipc_port.h +++ b/include/net/tipc/tipc_port.h @@ -96,6 +96,12 @@ struct tipc_port *tipc_get_port(const u32 ref); void *tipc_get_handle(const u32 ref); +/* + * The following routines require that the port be locked on entry + */ + +int tipc_disconnect_port(struct tipc_port *tp_ptr); + #endif diff --git a/net/tipc/port.c b/net/tipc/port.c index e2646a96935d..2f5806410c64 100644 --- a/net/tipc/port.c +++ b/net/tipc/port.c @@ -1240,6 +1240,28 @@ exit: return res; } +/** + * tipc_disconnect_port - disconnect port from peer + * + * Port must be locked. + */ + +int tipc_disconnect_port(struct tipc_port *tp_ptr) +{ + int res; + + if (tp_ptr->connected) { + tp_ptr->connected = 0; + /* let timer expire on it's own to avoid deadlock! */ + tipc_nodesub_unsubscribe( + &((struct port *)tp_ptr)->subscription); + res = TIPC_OK; + } else { + res = -ENOTCONN; + } + return res; +} + /* * tipc_disconnect(): Disconnect port form peer. * This is a node local operation. @@ -1248,17 +1270,12 @@ exit: int tipc_disconnect(u32 ref) { struct port *p_ptr; - int res = -ENOTCONN; + int res; p_ptr = tipc_port_lock(ref); if (!p_ptr) return -EINVAL; - if (p_ptr->publ.connected) { - p_ptr->publ.connected = 0; - /* let timer expire on it's own to avoid deadlock! */ - tipc_nodesub_unsubscribe(&p_ptr->subscription); - res = TIPC_OK; - } + res = tipc_disconnect_port((struct tipc_port *)p_ptr); tipc_port_unlock(p_ptr); return res; } diff --git a/net/tipc/socket.c b/net/tipc/socket.c index df1960384364..05853159536a 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -43,7 +43,6 @@ #include #include #include -#include #include #include #include @@ -64,11 +63,12 @@ struct tipc_sock { struct sock sk; struct tipc_port *p; - struct mutex lock; }; -#define tipc_sk(sk) ((struct tipc_sock*)sk) +#define tipc_sk(sk) ((struct tipc_sock *)(sk)) +#define tipc_sk_port(sk) ((struct tipc_port *)(tipc_sk(sk)->p)) +static int backlog_rcv(struct sock *sk, struct sk_buff *skb); static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf); static void wakeupdispatch(struct tipc_port *tport); @@ -82,55 +82,115 @@ static int sockets_enabled = 0; static atomic_t tipc_queue_size = ATOMIC_INIT(0); - /* - * sock_lock(): Lock a port/socket pair. lock_sock() can - * not be used here, since the same lock must protect ports - * with non-socket interfaces. - * See net.c for description of locking policy. + * Revised TIPC socket locking policy: + * + * Most socket operations take the standard socket lock when they start + * and hold it until they finish (or until they need to sleep). Acquiring + * this lock grants the owner exclusive access to the fields of the socket + * data structures, with the exception of the backlog queue. A few socket + * operations can be done without taking the socket lock because they only + * read socket information that never changes during the life of the socket. + * + * Socket operations may acquire the lock for the associated TIPC port if they + * need to perform an operation on the port. If any routine needs to acquire + * both the socket lock and the port lock it must take the socket lock first + * to avoid the risk of deadlock. + * + * The dispatcher handling incoming messages cannot grab the socket lock in + * the standard fashion, since invoked it runs at the BH level and cannot block. + * Instead, it checks to see if the socket lock is currently owned by someone, + * and either handles the message itself or adds it to the socket's backlog + * queue; in the latter case the queued message is processed once the process + * owning the socket lock releases it. + * + * NOTE: Releasing the socket lock while an operation is sleeping overcomes + * the problem of a blocked socket operation preventing any other operations + * from occurring. However, applications must be careful if they have + * multiple threads trying to send (or receive) on the same socket, as these + * operations might interfere with each other. For example, doing a connect + * and a receive at the same time might allow the receive to consume the + * ACK message meant for the connect. While additional work could be done + * to try and overcome this, it doesn't seem to be worthwhile at the present. + * + * NOTE: Releasing the socket lock while an operation is sleeping also ensures + * that another operation that must be performed in a non-blocking manner is + * not delayed for very long because the lock has already been taken. + * + * NOTE: This code assumes that certain fields of a port/socket pair are + * constant over its lifetime; such fields can be examined without taking + * the socket lock and/or port lock, and do not need to be re-read even + * after resuming processing after waiting. These fields include: + * - socket type + * - pointer to socket sk structure (aka tipc_sock structure) + * - pointer to port structure + * - port reference + */ + +/** + * advance_rx_queue - discard first buffer in socket receive queue + * + * Caller must hold socket lock */ -static void sock_lock(struct tipc_sock* tsock) + +static void advance_rx_queue(struct sock *sk) { - spin_lock_bh(tsock->p->lock); + buf_discard(__skb_dequeue(&sk->sk_receive_queue)); + atomic_dec(&tipc_queue_size); } -/* - * sock_unlock(): Unlock a port/socket pair +/** + * discard_rx_queue - discard all buffers in socket receive queue + * + * Caller must hold socket lock */ -static void sock_unlock(struct tipc_sock* tsock) + +static void discard_rx_queue(struct sock *sk) { - spin_unlock_bh(tsock->p->lock); + struct sk_buff *buf; + + while ((buf = __skb_dequeue(&sk->sk_receive_queue))) { + atomic_dec(&tipc_queue_size); + buf_discard(buf); + } } /** - * advance_queue - discard first buffer in queue - * @tsock: TIPC socket + * reject_rx_queue - reject all buffers in socket receive queue + * + * Caller must hold socket lock */ -static void advance_queue(struct tipc_sock *tsock) +static void reject_rx_queue(struct sock *sk) { - sock_lock(tsock); - buf_discard(skb_dequeue(&tsock->sk.sk_receive_queue)); - sock_unlock(tsock); - atomic_dec(&tipc_queue_size); + struct sk_buff *buf; + + while ((buf = __skb_dequeue(&sk->sk_receive_queue))) { + tipc_reject_msg(buf, TIPC_ERR_NO_PORT); + atomic_dec(&tipc_queue_size); + } } /** * tipc_create - create a TIPC socket + * @net: network namespace (must be default network) * @sock: pre-allocated socket structure * @protocol: protocol indicator (must be 0) * - * This routine creates and attaches a 'struct sock' to the 'struct socket', - * then create and attaches a TIPC port to the 'struct sock' part. + * This routine creates additional data structures used by the TIPC socket, + * initializes them, and links them together. * * Returns 0 on success, errno otherwise */ + static int tipc_create(struct net *net, struct socket *sock, int protocol) { - struct tipc_sock *tsock; - struct tipc_port *port; + const struct proto_ops *ops; + socket_state state; struct sock *sk; - u32 ref; + u32 portref; + + /* Validate arguments */ if (net != &init_net) return -EAFNOSUPPORT; @@ -138,53 +198,56 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol) if (unlikely(protocol != 0)) return -EPROTONOSUPPORT; - ref = tipc_createport_raw(NULL, &dispatch, &wakeupdispatch, TIPC_LOW_IMPORTANCE); - if (unlikely(!ref)) - return -ENOMEM; - - sock->state = SS_UNCONNECTED; - switch (sock->type) { case SOCK_STREAM: - sock->ops = &stream_ops; + ops = &stream_ops; + state = SS_UNCONNECTED; break; case SOCK_SEQPACKET: - sock->ops = &packet_ops; + ops = &packet_ops; + state = SS_UNCONNECTED; break; case SOCK_DGRAM: - tipc_set_portunreliable(ref, 1); - /* fall through */ case SOCK_RDM: - tipc_set_portunreturnable(ref, 1); - sock->ops = &msg_ops; - sock->state = SS_READY; + ops = &msg_ops; + state = SS_READY; break; default: - tipc_deleteport(ref); return -EPROTOTYPE; } + /* Allocate socket's protocol area */ + sk = sk_alloc(net, AF_TIPC, GFP_KERNEL, &tipc_proto); - if (!sk) { - tipc_deleteport(ref); + if (sk == NULL) return -ENOMEM; - } - sock_init_data(sock, sk); - sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT); + /* Allocate TIPC port for socket to use */ - tsock = tipc_sk(sk); - port = tipc_get_port(ref); + portref = tipc_createport_raw(sk, &dispatch, &wakeupdispatch, + TIPC_LOW_IMPORTANCE); + if (unlikely(portref == 0)) { + sk_free(sk); + return -ENOMEM; + } - tsock->p = port; - port->usr_handle = tsock; + /* Finish initializing socket data structures */ - mutex_init(&tsock->lock); + sock->ops = ops; + sock->state = state; - dbg("sock_create: %x\n",tsock); + sock_init_data(sock, sk); + sk->sk_rcvtimeo = msecs_to_jiffies(CONN_TIMEOUT_DEFAULT); + sk->sk_backlog_rcv = backlog_rcv; + tipc_sk(sk)->p = tipc_get_port(portref); - atomic_inc(&tipc_user_count); + if (sock->state == SS_READY) { + tipc_set_portunreturnable(portref, 1); + if (sock->type == SOCK_DGRAM) + tipc_set_portunreliable(portref, 1); + } + atomic_inc(&tipc_user_count); return 0; } @@ -207,52 +270,62 @@ static int tipc_create(struct net *net, struct socket *sock, int protocol) static int release(struct socket *sock) { - struct tipc_sock *tsock = tipc_sk(sock->sk); struct sock *sk = sock->sk; - int res = TIPC_OK; + struct tipc_port *tport; struct sk_buff *buf; + int res; - dbg("sock_delete: %x\n",tsock); - if (!tsock) - return 0; - mutex_lock(&tsock->lock); - if (!sock->sk) { - mutex_unlock(&tsock->lock); + /* + * Exit if socket isn't fully initialized (occurs when a failed accept() + * releases a pre-allocated child socket that was never used) + */ + + if (sk == NULL) return 0; - } - /* Reject unreceived messages, unless no longer connected */ + tport = tipc_sk_port(sk); + lock_sock(sk); + + /* + * Reject all unreceived messages, except on an active connection + * (which disconnects locally & sends a 'FIN+' to peer) + */ while (sock->state != SS_DISCONNECTING) { - sock_lock(tsock); - buf = skb_dequeue(&sk->sk_receive_queue); - if (!buf) - tsock->p->usr_handle = NULL; - sock_unlock(tsock); - if (!buf) + buf = __skb_dequeue(&sk->sk_receive_queue); + if (buf == NULL) break; + atomic_dec(&tipc_queue_size); if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) buf_discard(buf); - else + else { + if ((sock->state == SS_CONNECTING) || + (sock->state == SS_CONNECTED)) { + sock->state = SS_DISCONNECTING; + tipc_disconnect(tport->ref); + } tipc_reject_msg(buf, TIPC_ERR_NO_PORT); - atomic_dec(&tipc_queue_size); + } } - /* Delete TIPC port */ + /* + * Delete TIPC port; this ensures no more messages are queued + * (also disconnects an active connection & sends a 'FIN-' to peer) + */ - res = tipc_deleteport(tsock->p->ref); - sock->sk = NULL; + res = tipc_deleteport(tport->ref); - /* Discard any remaining messages */ + /* Discard any remaining (connection-based) messages in receive queue */ - while ((buf = skb_dequeue(&sk->sk_receive_queue))) { - buf_discard(buf); - atomic_dec(&tipc_queue_size); - } + discard_rx_queue(sk); - mutex_unlock(&tsock->lock); + /* Reject any messages that accumulated in backlog queue */ + + sock->state = SS_DISCONNECTING; + release_sock(sk); sock_put(sk); + sock->sk = NULL; atomic_dec(&tipc_user_count); return res; @@ -269,47 +342,32 @@ static int release(struct socket *sock) * (i.e. a socket address length of 0) unbinds all names from the socket. * * Returns 0 on success, errno otherwise + * + * NOTE: This routine doesn't need to take the socket lock since it doesn't + * access any non-constant socket information. */ static int bind(struct socket *sock, struct sockaddr *uaddr, int uaddr_len) { - struct tipc_sock *tsock = tipc_sk(sock->sk); struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; - int res; + u32 portref = tipc_sk_port(sock->sk)->ref; - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; + if (unlikely(!uaddr_len)) + return tipc_withdraw(portref, 0, NULL); - if (unlikely(!uaddr_len)) { - res = tipc_withdraw(tsock->p->ref, 0, NULL); - goto exit; - } - - if (uaddr_len < sizeof(struct sockaddr_tipc)) { - res = -EINVAL; - goto exit; - } + if (uaddr_len < sizeof(struct sockaddr_tipc)) + return -EINVAL; + if (addr->family != AF_TIPC) + return -EAFNOSUPPORT; - if (addr->family != AF_TIPC) { - res = -EAFNOSUPPORT; - goto exit; - } if (addr->addrtype == TIPC_ADDR_NAME) addr->addr.nameseq.upper = addr->addr.nameseq.lower; - else if (addr->addrtype != TIPC_ADDR_NAMESEQ) { - res = -EAFNOSUPPORT; - goto exit; - } + else if (addr->addrtype != TIPC_ADDR_NAMESEQ) + return -EAFNOSUPPORT; - if (addr->scope > 0) - res = tipc_publish(tsock->p->ref, addr->scope, - &addr->addr.nameseq); - else - res = tipc_withdraw(tsock->p->ref, -addr->scope, - &addr->addr.nameseq); -exit: - mutex_unlock(&tsock->lock); - return res; + return (addr->scope > 0) ? + tipc_publish(portref, addr->scope, &addr->addr.nameseq) : + tipc_withdraw(portref, -addr->scope, &addr->addr.nameseq); } /** @@ -320,30 +378,33 @@ exit: * @peer: 0 to obtain socket name, 1 to obtain peer socket name * * Returns 0 on success, errno otherwise + * + * NOTE: This routine doesn't need to take the socket lock since it doesn't + * access any non-constant socket information. */ static int get_name(struct socket *sock, struct sockaddr *uaddr, int *uaddr_len, int peer) { - struct tipc_sock *tsock = tipc_sk(sock->sk); struct sockaddr_tipc *addr = (struct sockaddr_tipc *)uaddr; + u32 portref = tipc_sk_port(sock->sk)->ref; u32 res; - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; + if (peer) { + res = tipc_peer(portref, &addr->addr.id); + if (res) + return res; + } else { + tipc_ownidentity(portref, &addr->addr.id); + } *uaddr_len = sizeof(*addr); addr->addrtype = TIPC_ADDR_ID; addr->family = AF_TIPC; addr->scope = 0; - if (peer) - res = tipc_peer(tsock->p->ref, &addr->addr.id); - else - res = tipc_ownidentity(tsock->p->ref, &addr->addr.id); addr->addr.name.domain = 0; - mutex_unlock(&tsock->lock); - return res; + return 0; } /** @@ -414,7 +475,6 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) return 0; if (likely(dest->addr.name.name.type == TIPC_TOP_SRV)) return 0; - if (likely(dest->addr.name.name.type != TIPC_CFG_SRV)) return -EACCES; @@ -428,7 +488,7 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) /** * send_msg - send message in connectionless manner - * @iocb: (unused) + * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send * @total_len: length of message @@ -444,9 +504,9 @@ static int dest_name_check(struct sockaddr_tipc *dest, struct msghdr *m) static int send_msg(struct kiocb *iocb, struct socket *sock, struct msghdr *m, size_t total_len) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; - struct sk_buff *buf; int needs_conn; int res = -EINVAL; @@ -456,48 +516,46 @@ static int send_msg(struct kiocb *iocb, struct socket *sock, (dest->family != AF_TIPC))) return -EINVAL; + if (iocb) + lock_sock(sk); + needs_conn = (sock->state != SS_READY); if (unlikely(needs_conn)) { - if (sock->state == SS_LISTENING) - return -EPIPE; - if (sock->state != SS_UNCONNECTED) - return -EISCONN; - if ((tsock->p->published) || - ((sock->type == SOCK_STREAM) && (total_len != 0))) - return -EOPNOTSUPP; + if (sock->state == SS_LISTENING) { + res = -EPIPE; + goto exit; + } + if (sock->state != SS_UNCONNECTED) { + res = -EISCONN; + goto exit; + } + if ((tport->published) || + ((sock->type == SOCK_STREAM) && (total_len != 0))) { + res = -EOPNOTSUPP; + goto exit; + } if (dest->addrtype == TIPC_ADDR_NAME) { - tsock->p->conn_type = dest->addr.name.name.type; - tsock->p->conn_instance = dest->addr.name.name.instance; + tport->conn_type = dest->addr.name.name.type; + tport->conn_instance = dest->addr.name.name.instance; } - } - - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; - - if (needs_conn) { /* Abort any pending connection attempts (very unlikely) */ - while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { - tipc_reject_msg(buf, TIPC_ERR_NO_PORT); - atomic_dec(&tipc_queue_size); - } - - sock->state = SS_CONNECTING; + reject_rx_queue(sk); } do { if (dest->addrtype == TIPC_ADDR_NAME) { if ((res = dest_name_check(dest, m))) - goto exit; - res = tipc_send2name(tsock->p->ref, + break; + res = tipc_send2name(tport->ref, &dest->addr.name.name, dest->addr.name.domain, m->msg_iovlen, m->msg_iov); } else if (dest->addrtype == TIPC_ADDR_ID) { - res = tipc_send2port(tsock->p->ref, + res = tipc_send2port(tport->ref, &dest->addr.id, m->msg_iovlen, m->msg_iov); @@ -505,36 +563,43 @@ static int send_msg(struct kiocb *iocb, struct socket *sock, else if (dest->addrtype == TIPC_ADDR_MCAST) { if (needs_conn) { res = -EOPNOTSUPP; - goto exit; + break; } if ((res = dest_name_check(dest, m))) - goto exit; - res = tipc_multicast(tsock->p->ref, + break; + res = tipc_multicast(tport->ref, &dest->addr.nameseq, 0, m->msg_iovlen, m->msg_iov); } if (likely(res != -ELINKCONG)) { -exit: - mutex_unlock(&tsock->lock); - return res; + if (needs_conn && (res >= 0)) { + sock->state = SS_CONNECTING; + } + break; } if (m->msg_flags & MSG_DONTWAIT) { res = -EWOULDBLOCK; - goto exit; - } - if (wait_event_interruptible(*sock->sk->sk_sleep, - !tsock->p->congested)) { - res = -ERESTARTSYS; - goto exit; + break; } + release_sock(sk); + res = wait_event_interruptible(*sk->sk_sleep, + !tport->congested); + lock_sock(sk); + if (res) + break; } while (1); + +exit: + if (iocb) + release_sock(sk); + return res; } /** * send_packet - send a connection-oriented message - * @iocb: (unused) + * @iocb: if NULL, indicates that socket lock is already held * @sock: socket structure * @m: message to send * @total_len: length of message @@ -547,7 +612,8 @@ exit: static int send_packet(struct kiocb *iocb, struct socket *sock, struct msghdr *m, size_t total_len) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); struct sockaddr_tipc *dest = (struct sockaddr_tipc *)m->msg_name; int res; @@ -556,9 +622,8 @@ static int send_packet(struct kiocb *iocb, struct socket *sock, if (unlikely(dest)) return send_msg(iocb, sock, m, total_len); - if (mutex_lock_interruptible(&tsock->lock)) { - return -ERESTARTSYS; - } + if (iocb) + lock_sock(sk); do { if (unlikely(sock->state != SS_CONNECTED)) { @@ -566,25 +631,28 @@ static int send_packet(struct kiocb *iocb, struct socket *sock, res = -EPIPE; else res = -ENOTCONN; - goto exit; + break; } - res = tipc_send(tsock->p->ref, m->msg_iovlen, m->msg_iov); + res = tipc_send(tport->ref, m->msg_iovlen, m->msg_iov); if (likely(res != -ELINKCONG)) { -exit: - mutex_unlock(&tsock->lock); - return res; + break; } if (m->msg_flags & MSG_DONTWAIT) { res = -EWOULDBLOCK; - goto exit; - } - if (wait_event_interruptible(*sock->sk->sk_sleep, - !tsock->p->congested)) { - res = -ERESTARTSYS; - goto exit; + break; } + release_sock(sk); + res = wait_event_interruptible(*sk->sk_sleep, + (!tport->congested || !tport->connected)); + lock_sock(sk); + if (res) + break; } while (1); + + if (iocb) + release_sock(sk); + return res; } /** @@ -600,11 +668,11 @@ exit: * or errno if no data sent */ - static int send_stream(struct kiocb *iocb, struct socket *sock, struct msghdr *m, size_t total_len) { - struct tipc_port *tport; + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); struct msghdr my_msg; struct iovec my_iov; struct iovec *curr_iov; @@ -616,19 +684,27 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, int bytes_sent; int res; + lock_sock(sk); + /* Handle special cases where there is no connection */ if (unlikely(sock->state != SS_CONNECTED)) { - if (sock->state == SS_UNCONNECTED) - return send_packet(iocb, sock, m, total_len); - else if (sock->state == SS_DISCONNECTING) - return -EPIPE; - else - return -ENOTCONN; + if (sock->state == SS_UNCONNECTED) { + res = send_packet(NULL, sock, m, total_len); + goto exit; + } else if (sock->state == SS_DISCONNECTING) { + res = -EPIPE; + goto exit; + } else { + res = -ENOTCONN; + goto exit; + } } - if (unlikely(m->msg_name)) - return -EISCONN; + if (unlikely(m->msg_name)) { + res = -EISCONN; + goto exit; + } /* * Send each iovec entry using one or more messages @@ -646,7 +722,6 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, my_msg.msg_name = NULL; bytes_sent = 0; - tport = tipc_sk(sock->sk)->p; hdr_size = msg_hdr_sz(&tport->phdr); while (curr_iovlen--) { @@ -661,10 +736,10 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, bytes_to_send = curr_left; my_iov.iov_base = curr_start; my_iov.iov_len = bytes_to_send; - if ((res = send_packet(iocb, sock, &my_msg, 0)) < 0) { - if (bytes_sent != 0) + if ((res = send_packet(NULL, sock, &my_msg, 0)) < 0) { + if (bytes_sent) res = bytes_sent; - return res; + goto exit; } curr_left -= bytes_to_send; curr_start += bytes_to_send; @@ -673,22 +748,23 @@ static int send_stream(struct kiocb *iocb, struct socket *sock, curr_iov++; } - - return bytes_sent; + res = bytes_sent; +exit: + release_sock(sk); + return res; } /** * auto_connect - complete connection setup to a remote port * @sock: socket structure - * @tsock: TIPC-specific socket structure * @msg: peer's response message * * Returns 0 on success, errno otherwise */ -static int auto_connect(struct socket *sock, struct tipc_sock *tsock, - struct tipc_msg *msg) +static int auto_connect(struct socket *sock, struct tipc_msg *msg) { + struct tipc_port *tport = tipc_sk_port(sock->sk); struct tipc_portid peer; if (msg_errcode(msg)) { @@ -698,8 +774,8 @@ static int auto_connect(struct socket *sock, struct tipc_sock *tsock, peer.ref = msg_origport(msg); peer.node = msg_orignode(msg); - tipc_connect2port(tsock->p->ref, &peer); - tipc_set_portimportance(tsock->p->ref, msg_importance(msg)); + tipc_connect2port(tport->ref, &peer); + tipc_set_portimportance(tport->ref, msg_importance(msg)); sock->state = SS_CONNECTED; return 0; } @@ -812,62 +888,54 @@ static int anc_data_recv(struct msghdr *m, struct tipc_msg *msg, static int recv_msg(struct kiocb *iocb, struct socket *sock, struct msghdr *m, size_t buf_len, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); struct sk_buff *buf; struct tipc_msg *msg; - unsigned int q_len; unsigned int sz; u32 err; int res; - /* Currently doesn't support receiving into multiple iovec entries */ + /* Catch invalid receive requests */ if (m->msg_iovlen != 1) - return -EOPNOTSUPP; - - /* Catch invalid receive attempts */ + return -EOPNOTSUPP; /* Don't do multiple iovec entries yet */ if (unlikely(!buf_len)) return -EINVAL; - if (sock->type == SOCK_SEQPACKET) { - if (unlikely(sock->state == SS_UNCONNECTED)) - return -ENOTCONN; - if (unlikely((sock->state == SS_DISCONNECTING) && - (skb_queue_len(&sock->sk->sk_receive_queue) == 0))) - return -ENOTCONN; - } + lock_sock(sk); - /* Look for a message in receive queue; wait if necessary */ - - if (unlikely(mutex_lock_interruptible(&tsock->lock))) - return -ERESTARTSYS; - -restart: - if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && - (flags & MSG_DONTWAIT))) { - res = -EWOULDBLOCK; + if (unlikely(sock->state == SS_UNCONNECTED)) { + res = -ENOTCONN; goto exit; } - if ((res = wait_event_interruptible( - *sock->sk->sk_sleep, - ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || - (sock->state == SS_DISCONNECTING))) )) { - goto exit; - } +restart: - /* Catch attempt to receive on an already terminated connection */ - /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + /* Look for a message in receive queue; wait if necessary */ - if (!q_len) { - res = -ENOTCONN; - goto exit; + while (skb_queue_empty(&sk->sk_receive_queue)) { + if (sock->state == SS_DISCONNECTING) { + res = -ENOTCONN; + goto exit; + } + if (flags & MSG_DONTWAIT) { + res = -EWOULDBLOCK; + goto exit; + } + release_sock(sk); + res = wait_event_interruptible(*sk->sk_sleep, + (!skb_queue_empty(&sk->sk_receive_queue) || + (sock->state == SS_DISCONNECTING))); + lock_sock(sk); + if (res) + goto exit; } - /* Get access to first message in receive queue */ + /* Look at first message in receive queue */ - buf = skb_peek(&sock->sk->sk_receive_queue); + buf = skb_peek(&sk->sk_receive_queue); msg = buf_msg(buf); sz = msg_data_sz(msg); err = msg_errcode(msg); @@ -875,14 +943,15 @@ restart: /* Complete connection setup for an implied connect */ if (unlikely(sock->state == SS_CONNECTING)) { - if ((res = auto_connect(sock, tsock, msg))) + res = auto_connect(sock, msg); + if (res) goto exit; } /* Discard an empty non-errored message & try again */ if ((!sz) && (!err)) { - advance_queue(tsock); + advance_rx_queue(sk); goto restart; } @@ -892,7 +961,8 @@ restart: /* Capture ancillary data (optional) */ - if ((res = anc_data_recv(m, msg, tsock->p))) + res = anc_data_recv(m, msg, tport); + if (res) goto exit; /* Capture message data (if valid) & compute return value (always) */ @@ -920,12 +990,12 @@ restart: if (likely(!(flags & MSG_PEEK))) { if ((sock->state != SS_READY) && - (++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); - advance_queue(tsock); + (++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(tport->ref, tport->conn_unacked); + advance_rx_queue(sk); } exit: - mutex_unlock(&tsock->lock); + release_sock(sk); return res; } @@ -945,10 +1015,10 @@ exit: static int recv_stream(struct kiocb *iocb, struct socket *sock, struct msghdr *m, size_t buf_len, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); struct sk_buff *buf; struct tipc_msg *msg; - unsigned int q_len; unsigned int sz; int sz_to_copy; int sz_copied = 0; @@ -956,54 +1026,49 @@ static int recv_stream(struct kiocb *iocb, struct socket *sock, char __user *crs = m->msg_iov->iov_base; unsigned char *buf_crs; u32 err; - int res; + int res = 0; - /* Currently doesn't support receiving into multiple iovec entries */ + /* Catch invalid receive attempts */ if (m->msg_iovlen != 1) - return -EOPNOTSUPP; - - /* Catch invalid receive attempts */ + return -EOPNOTSUPP; /* Don't do multiple iovec entries yet */ if (unlikely(!buf_len)) return -EINVAL; - if (unlikely(sock->state == SS_DISCONNECTING)) { - if (skb_queue_len(&sock->sk->sk_receive_queue) == 0) - return -ENOTCONN; - } else if (unlikely(sock->state != SS_CONNECTED)) - return -ENOTCONN; - - /* Look for a message in receive queue; wait if necessary */ - - if (unlikely(mutex_lock_interruptible(&tsock->lock))) - return -ERESTARTSYS; + lock_sock(sk); -restart: - if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && - (flags & MSG_DONTWAIT))) { - res = -EWOULDBLOCK; + if (unlikely((sock->state == SS_UNCONNECTED) || + (sock->state == SS_CONNECTING))) { + res = -ENOTCONN; goto exit; } - if ((res = wait_event_interruptible( - *sock->sk->sk_sleep, - ((q_len = skb_queue_len(&sock->sk->sk_receive_queue)) || - (sock->state == SS_DISCONNECTING))) )) { - goto exit; - } +restart: - /* Catch attempt to receive on an already terminated connection */ - /* [THIS CHECK MAY OVERLAP WITH AN EARLIER CHECK] */ + /* Look for a message in receive queue; wait if necessary */ - if (!q_len) { - res = -ENOTCONN; - goto exit; + while (skb_queue_empty(&sk->sk_receive_queue)) { + if (sock->state == SS_DISCONNECTING) { + res = -ENOTCONN; + goto exit; + } + if (flags & MSG_DONTWAIT) { + res = -EWOULDBLOCK; + goto exit; + } + release_sock(sk); + res = wait_event_interruptible(*sk->sk_sleep, + (!skb_queue_empty(&sk->sk_receive_queue) || + (sock->state == SS_DISCONNECTING))); + lock_sock(sk); + if (res) + goto exit; } - /* Get access to first message in receive queue */ + /* Look at first message in receive queue */ - buf = skb_peek(&sock->sk->sk_receive_queue); + buf = skb_peek(&sk->sk_receive_queue); msg = buf_msg(buf); sz = msg_data_sz(msg); err = msg_errcode(msg); @@ -1011,7 +1076,7 @@ restart: /* Discard an empty non-errored message & try again */ if ((!sz) && (!err)) { - advance_queue(tsock); + advance_rx_queue(sk); goto restart; } @@ -1019,7 +1084,8 @@ restart: if (sz_copied == 0) { set_orig_addr(m, msg); - if ((res = anc_data_recv(m, msg, tsock->p))) + res = anc_data_recv(m, msg, tport); + if (res) goto exit; } @@ -1057,9 +1123,9 @@ restart: /* Consume received message (optional) */ if (likely(!(flags & MSG_PEEK))) { - if (unlikely(++tsock->p->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) - tipc_acknowledge(tsock->p->ref, tsock->p->conn_unacked); - advance_queue(tsock); + if (unlikely(++tport->conn_unacked >= TIPC_FLOW_CONTROL_WIN)) + tipc_acknowledge(tport->ref, tport->conn_unacked); + advance_rx_queue(sk); } /* Loop around if more data is required */ @@ -1074,7 +1140,7 @@ restart: goto restart; exit: - mutex_unlock(&tsock->lock); + release_sock(sk); return sz_copied ? sz_copied : res; } @@ -1108,37 +1174,24 @@ static int rx_queue_full(struct tipc_msg *msg, u32 queue_size, u32 base) } /** - * async_disconnect - wrapper function used to disconnect port - * @portref: TIPC port reference (passed as pointer-sized value) - */ - -static void async_disconnect(unsigned long portref) -{ - tipc_disconnect((u32)portref); -} - -/** - * dispatch - handle arriving message - * @tport: TIPC port that received message + * filter_rcv - validate incoming message + * @sk: socket * @buf: message * - * Called with port locked. Must not take socket lock to avoid deadlock risk. + * Enqueues message on receive queue if acceptable; optionally handles + * disconnect indication for a connected socket. + * + * Called with socket lock already taken; port lock may also be taken. * * Returns TIPC error status code (TIPC_OK if message is not to be rejected) */ -static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) +static u32 filter_rcv(struct sock *sk, struct sk_buff *buf) { + struct socket *sock = sk->sk_socket; struct tipc_msg *msg = buf_msg(buf); - struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; - struct socket *sock; u32 recv_q_len; - /* Reject message if socket is closing */ - - if (!tsock) - return TIPC_ERR_NO_PORT; - /* Reject message if it is wrong sort of message for socket */ /* @@ -1146,7 +1199,7 @@ static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) * "NO PORT" ISN'T REALLY THE RIGHT ERROR CODE, AND THERE MAY * BE SECURITY IMPLICATIONS INHERENT IN REJECTING INVALID TRAFFIC */ - sock = tsock->sk.sk_socket; + if (sock->state == SS_READY) { if (msg_connected(msg)) { msg_dbg(msg, "dispatch filter 1\n"); @@ -1194,45 +1247,98 @@ static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) if (rx_queue_full(msg, recv_q_len, OVERLOAD_LIMIT_BASE)) return TIPC_ERR_OVERLOAD; } - recv_q_len = skb_queue_len(&tsock->sk.sk_receive_queue); + recv_q_len = skb_queue_len(&sk->sk_receive_queue); if (unlikely(recv_q_len >= (OVERLOAD_LIMIT_BASE / 2))) { if (rx_queue_full(msg, recv_q_len, OVERLOAD_LIMIT_BASE / 2)) return TIPC_ERR_OVERLOAD; } + /* Enqueue message (finally!) */ + + msg_dbg(msg, "handle = msg_data(msg); + atomic_inc(&tipc_queue_size); + __skb_queue_tail(&sk->sk_receive_queue, buf); + /* Initiate connection termination for an incoming 'FIN' */ if (unlikely(msg_errcode(msg) && (sock->state == SS_CONNECTED))) { sock->state = SS_DISCONNECTING; - /* Note: Use signal since port lock is already taken! */ - tipc_k_signal((Handler)async_disconnect, tport->ref); + tipc_disconnect_port(tipc_sk_port(sk)); } - /* Enqueue message (finally!) */ + if (waitqueue_active(sk->sk_sleep)) + wake_up_interruptible(sk->sk_sleep); + return TIPC_OK; +} - msg_dbg(msg,"handle = msg_data(msg); - atomic_inc(&tipc_queue_size); - skb_queue_tail(&sock->sk->sk_receive_queue, buf); +/** + * backlog_rcv - handle incoming message from backlog queue + * @sk: socket + * @buf: message + * + * Caller must hold socket lock, but not port lock. + * + * Returns 0 + */ - if (waitqueue_active(sock->sk->sk_sleep)) - wake_up_interruptible(sock->sk->sk_sleep); - return TIPC_OK; +static int backlog_rcv(struct sock *sk, struct sk_buff *buf) +{ + u32 res; + + res = filter_rcv(sk, buf); + if (res) + tipc_reject_msg(buf, res); + return 0; +} + +/** + * dispatch - handle incoming message + * @tport: TIPC port that received message + * @buf: message + * + * Called with port lock already taken. + * + * Returns TIPC error status code (TIPC_OK if message is not to be rejected) + */ + +static u32 dispatch(struct tipc_port *tport, struct sk_buff *buf) +{ + struct sock *sk = (struct sock *)tport->usr_handle; + u32 res; + + /* + * Process message if socket is unlocked; otherwise add to backlog queue + * + * This code is based on sk_receive_skb(), but must be distinct from it + * since a TIPC-specific filter/reject mechanism is utilized + */ + + bh_lock_sock(sk); + if (!sock_owned_by_user(sk)) { + res = filter_rcv(sk, buf); + } else { + sk_add_backlog(sk, buf); + res = TIPC_OK; + } + bh_unlock_sock(sk); + + return res; } /** * wakeupdispatch - wake up port after congestion * @tport: port to wakeup * - * Called with port lock on. + * Called with port lock already taken. */ static void wakeupdispatch(struct tipc_port *tport) { - struct tipc_sock *tsock = (struct tipc_sock *)tport->usr_handle; + struct sock *sk = (struct sock *)tport->usr_handle; - if (waitqueue_active(tsock->sk.sk_sleep)) - wake_up_interruptible(tsock->sk.sk_sleep); + if (waitqueue_active(sk->sk_sleep)) + wake_up_interruptible(sk->sk_sleep); } /** @@ -1240,7 +1346,7 @@ static void wakeupdispatch(struct tipc_port *tport) * @sock: socket structure * @dest: socket address for destination port * @destlen: size of socket address data structure - * @flags: (unused) + * @flags: file-related flags associated with socket * * Returns 0 on success, errno otherwise */ @@ -1248,31 +1354,43 @@ static void wakeupdispatch(struct tipc_port *tport) static int connect(struct socket *sock, struct sockaddr *dest, int destlen, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; struct sockaddr_tipc *dst = (struct sockaddr_tipc *)dest; struct msghdr m = {NULL,}; struct sk_buff *buf; struct tipc_msg *msg; int res; + lock_sock(sk); + /* For now, TIPC does not allow use of connect() with DGRAM/RDM types */ - if (sock->state == SS_READY) - return -EOPNOTSUPP; + if (sock->state == SS_READY) { + res = -EOPNOTSUPP; + goto exit; + } /* For now, TIPC does not support the non-blocking form of connect() */ - if (flags & O_NONBLOCK) - return -EWOULDBLOCK; + if (flags & O_NONBLOCK) { + res = -EWOULDBLOCK; + goto exit; + } /* Issue Posix-compliant error code if socket is in the wrong state */ - if (sock->state == SS_LISTENING) - return -EOPNOTSUPP; - if (sock->state == SS_CONNECTING) - return -EALREADY; - if (sock->state != SS_UNCONNECTED) - return -EISCONN; + if (sock->state == SS_LISTENING) { + res = -EOPNOTSUPP; + goto exit; + } + if (sock->state == SS_CONNECTING) { + res = -EALREADY; + goto exit; + } + if (sock->state != SS_UNCONNECTED) { + res = -EISCONN; + goto exit; + } /* * Reject connection attempt using multicast address @@ -1281,8 +1399,14 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, * so there's no need to do it here */ - if (dst->addrtype == TIPC_ADDR_MCAST) - return -EINVAL; + if (dst->addrtype == TIPC_ADDR_MCAST) { + res = -EINVAL; + goto exit; + } + + /* Reject any messages already in receive queue (very unlikely) */ + + reject_rx_queue(sk); /* Send a 'SYN-' to destination */ @@ -1290,25 +1414,33 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, m.msg_namelen = destlen; res = send_msg(NULL, sock, &m, 0); if (res < 0) { - sock->state = SS_DISCONNECTING; - return res; + goto exit; } - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; + /* Wait until an 'ACK' or 'RST' arrives, or a timeout occurs */ - /* Wait for destination's 'ACK' response */ + release_sock(sk); + res = wait_event_interruptible_timeout(*sk->sk_sleep, + (!skb_queue_empty(&sk->sk_receive_queue) || + (sock->state != SS_CONNECTING)), + sk->sk_rcvtimeo); + lock_sock(sk); - res = wait_event_interruptible_timeout(*sock->sk->sk_sleep, - skb_queue_len(&sock->sk->sk_receive_queue), - sock->sk->sk_rcvtimeo); - buf = skb_peek(&sock->sk->sk_receive_queue); if (res > 0) { - msg = buf_msg(buf); - res = auto_connect(sock, tsock, msg); - if (!res) { - if (!msg_data_sz(msg)) - advance_queue(tsock); + buf = skb_peek(&sk->sk_receive_queue); + if (buf != NULL) { + msg = buf_msg(buf); + res = auto_connect(sock, msg); + if (!res) { + if (!msg_data_sz(msg)) + advance_rx_queue(sk); + } + } else { + if (sock->state == SS_CONNECTED) { + res = -EISCONN; + } else { + res = -ECONNREFUSED; + } } } else { if (res == 0) @@ -1318,7 +1450,8 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, sock->state = SS_DISCONNECTING; } - mutex_unlock(&tsock->lock); +exit: + release_sock(sk); return res; } @@ -1332,14 +1465,22 @@ static int connect(struct socket *sock, struct sockaddr *dest, int destlen, static int listen(struct socket *sock, int len) { - /* REQUIRES SOCKET LOCKING OF SOME SORT? */ + struct sock *sk = sock->sk; + int res; + + lock_sock(sk); if (sock->state == SS_READY) - return -EOPNOTSUPP; - if (sock->state != SS_UNCONNECTED) - return -EINVAL; - sock->state = SS_LISTENING; - return 0; + res = -EOPNOTSUPP; + else if (sock->state != SS_UNCONNECTED) + res = -EINVAL; + else { + sock->state = SS_LISTENING; + res = 0; + } + + release_sock(sk); + return res; } /** @@ -1351,50 +1492,69 @@ static int listen(struct socket *sock, int len) * Returns 0 on success, errno otherwise */ -static int accept(struct socket *sock, struct socket *newsock, int flags) +static int accept(struct socket *sock, struct socket *new_sock, int flags) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; struct sk_buff *buf; - int res = -EFAULT; - - if (sock->state == SS_READY) - return -EOPNOTSUPP; - if (sock->state != SS_LISTENING) - return -EINVAL; - - if (unlikely((skb_queue_len(&sock->sk->sk_receive_queue) == 0) && - (flags & O_NONBLOCK))) - return -EWOULDBLOCK; + int res; - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; + lock_sock(sk); - if (wait_event_interruptible(*sock->sk->sk_sleep, - skb_queue_len(&sock->sk->sk_receive_queue))) { - res = -ERESTARTSYS; + if (sock->state == SS_READY) { + res = -EOPNOTSUPP; + goto exit; + } + if (sock->state != SS_LISTENING) { + res = -EINVAL; goto exit; } - buf = skb_peek(&sock->sk->sk_receive_queue); - res = tipc_create(sock_net(sock->sk), newsock, 0); + while (skb_queue_empty(&sk->sk_receive_queue)) { + if (flags & O_NONBLOCK) { + res = -EWOULDBLOCK; + goto exit; + } + release_sock(sk); + res = wait_event_interruptible(*sk->sk_sleep, + (!skb_queue_empty(&sk->sk_receive_queue))); + lock_sock(sk); + if (res) + goto exit; + } + + buf = skb_peek(&sk->sk_receive_queue); + + res = tipc_create(sock_net(sock->sk), new_sock, 0); if (!res) { - struct tipc_sock *new_tsock = tipc_sk(newsock->sk); + struct sock *new_sk = new_sock->sk; + struct tipc_port *new_tport = tipc_sk_port(new_sk); + u32 new_ref = new_tport->ref; struct tipc_portid id; struct tipc_msg *msg = buf_msg(buf); - u32 new_ref = new_tsock->p->ref; + + lock_sock(new_sk); + + /* + * Reject any stray messages received by new socket + * before the socket lock was taken (very, very unlikely) + */ + + reject_rx_queue(new_sk); + + /* Connect new socket to it's peer */ id.ref = msg_origport(msg); id.node = msg_orignode(msg); tipc_connect2port(new_ref, &id); - newsock->state = SS_CONNECTED; + new_sock->state = SS_CONNECTED; tipc_set_portimportance(new_ref, msg_importance(msg)); if (msg_named(msg)) { - new_tsock->p->conn_type = msg_nametype(msg); - new_tsock->p->conn_instance = msg_nameinst(msg); + new_tport->conn_type = msg_nametype(msg); + new_tport->conn_instance = msg_nameinst(msg); } - /* + /* * Respond to 'SYN-' by discarding it & returning 'ACK'-. * Respond to 'SYN+' by queuing it on new socket. */ @@ -1403,17 +1563,16 @@ static int accept(struct socket *sock, struct socket *newsock, int flags) if (!msg_data_sz(msg)) { struct msghdr m = {NULL,}; - send_packet(NULL, newsock, &m, 0); - advance_queue(tsock); + advance_rx_queue(sk); + send_packet(NULL, new_sock, &m, 0); } else { - sock_lock(tsock); - skb_dequeue(&sock->sk->sk_receive_queue); - sock_unlock(tsock); - skb_queue_head(&newsock->sk->sk_receive_queue, buf); + __skb_dequeue(&sk->sk_receive_queue); + __skb_queue_head(&new_sk->sk_receive_queue, buf); } + release_sock(new_sk); } exit: - mutex_unlock(&tsock->lock); + release_sock(sk); return res; } @@ -1429,54 +1588,46 @@ exit: static int shutdown(struct socket *sock, int how) { - struct tipc_sock* tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); struct sk_buff *buf; int res; if (how != SHUT_RDWR) return -EINVAL; - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; - - sock_lock(tsock); + lock_sock(sk); switch (sock->state) { + case SS_CONNECTING: case SS_CONNECTED: - /* Send 'FIN+' or 'FIN-' message to peer */ - - sock_unlock(tsock); + /* Disconnect and send a 'FIN+' or 'FIN-' message to peer */ restart: - if ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { + buf = __skb_dequeue(&sk->sk_receive_queue); + if (buf) { atomic_dec(&tipc_queue_size); if (TIPC_SKB_CB(buf)->handle != msg_data(buf_msg(buf))) { buf_discard(buf); goto restart; } + tipc_disconnect(tport->ref); tipc_reject_msg(buf, TIPC_CONN_SHUTDOWN); + } else { + tipc_shutdown(tport->ref); } - else { - tipc_shutdown(tsock->p->ref); - } - sock_lock(tsock); + + sock->state = SS_DISCONNECTING; /* fall through */ case SS_DISCONNECTING: - /* Discard any unreceived messages */ - - while ((buf = skb_dequeue(&sock->sk->sk_receive_queue))) { - atomic_dec(&tipc_queue_size); - buf_discard(buf); - } - tsock->p->conn_unacked = 0; + /* Discard any unreceived messages; wake up sleeping tasks */ - /* fall through */ - - case SS_CONNECTING: - sock->state = SS_DISCONNECTING; + discard_rx_queue(sk); + if (waitqueue_active(sk->sk_sleep)) + wake_up_interruptible(sk->sk_sleep); res = 0; break; @@ -1484,9 +1635,7 @@ restart: res = -ENOTCONN; } - sock_unlock(tsock); - - mutex_unlock(&tsock->lock); + release_sock(sk); return res; } @@ -1507,7 +1656,8 @@ restart: static int setsockopt(struct socket *sock, int lvl, int opt, char __user *ov, int ol) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); u32 value; int res; @@ -1520,30 +1670,31 @@ static int setsockopt(struct socket *sock, if ((res = get_user(value, (u32 __user *)ov))) return res; - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; + lock_sock(sk); switch (opt) { case TIPC_IMPORTANCE: - res = tipc_set_portimportance(tsock->p->ref, value); + res = tipc_set_portimportance(tport->ref, value); break; case TIPC_SRC_DROPPABLE: if (sock->type != SOCK_STREAM) - res = tipc_set_portunreliable(tsock->p->ref, value); + res = tipc_set_portunreliable(tport->ref, value); else res = -ENOPROTOOPT; break; case TIPC_DEST_DROPPABLE: - res = tipc_set_portunreturnable(tsock->p->ref, value); + res = tipc_set_portunreturnable(tport->ref, value); break; case TIPC_CONN_TIMEOUT: - sock->sk->sk_rcvtimeo = msecs_to_jiffies(value); + sk->sk_rcvtimeo = msecs_to_jiffies(value); + /* no need to set "res", since already 0 at this point */ break; default: res = -EINVAL; } - mutex_unlock(&tsock->lock); + release_sock(sk); + return res; } @@ -1564,7 +1715,8 @@ static int setsockopt(struct socket *sock, static int getsockopt(struct socket *sock, int lvl, int opt, char __user *ov, int __user *ol) { - struct tipc_sock *tsock = tipc_sk(sock->sk); + struct sock *sk = sock->sk; + struct tipc_port *tport = tipc_sk_port(sk); int len; u32 value; int res; @@ -1576,26 +1728,28 @@ static int getsockopt(struct socket *sock, if ((res = get_user(len, ol))) return res; - if (mutex_lock_interruptible(&tsock->lock)) - return -ERESTARTSYS; + lock_sock(sk); switch (opt) { case TIPC_IMPORTANCE: - res = tipc_portimportance(tsock->p->ref, &value); + res = tipc_portimportance(tport->ref, &value); break; case TIPC_SRC_DROPPABLE: - res = tipc_portunreliable(tsock->p->ref, &value); + res = tipc_portunreliable(tport->ref, &value); break; case TIPC_DEST_DROPPABLE: - res = tipc_portunreturnable(tsock->p->ref, &value); + res = tipc_portunreturnable(tport->ref, &value); break; case TIPC_CONN_TIMEOUT: - value = jiffies_to_msecs(sock->sk->sk_rcvtimeo); + value = jiffies_to_msecs(sk->sk_rcvtimeo); + /* no need to set "res", since already 0 at this point */ break; default: res = -EINVAL; } + release_sock(sk); + if (res) { /* "get" failed */ } @@ -1609,7 +1763,6 @@ static int getsockopt(struct socket *sock, res = put_user(sizeof(value), ol); } - mutex_unlock(&tsock->lock); return res; } @@ -1722,6 +1875,7 @@ int tipc_socket_init(void) /** * tipc_socket_stop - stop TIPC socket interface */ + void tipc_socket_stop(void) { if (!sockets_enabled) -- cgit v1.2.3 From 99971e70fdc1862e120f3319fc0a4dba8c728acf Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 15 Apr 2008 00:27:58 -0700 Subject: [WANPIPE]: Forgotten bits of Sangoma drivers removal. Robert P. J. Day spotted that my removal of the Sangoma drivers missed a few bits. Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- Documentation/networking/00-INDEX | 2 - Documentation/networking/wan-router.txt | 621 -------------------------------- include/linux/Kbuild | 1 - include/linux/if_wanpipe.h | 124 ------- 4 files changed, 748 deletions(-) delete mode 100644 Documentation/networking/wan-router.txt delete mode 100644 include/linux/if_wanpipe.h (limited to 'include') diff --git a/Documentation/networking/00-INDEX b/Documentation/networking/00-INDEX index c485ee028bd9..1634c6dcecae 100644 --- a/Documentation/networking/00-INDEX +++ b/Documentation/networking/00-INDEX @@ -100,8 +100,6 @@ tuntap.txt - TUN/TAP device driver, allowing user space Rx/Tx of packets. vortex.txt - info on using 3Com Vortex (3c590, 3c592, 3c595, 3c597) Ethernet cards. -wan-router.txt - - WAN router documentation wavelan.txt - AT&T GIS (nee NCR) WaveLAN card: An Ethernet-like radio transceiver x25.txt diff --git a/Documentation/networking/wan-router.txt b/Documentation/networking/wan-router.txt deleted file mode 100644 index bc2ab419a74a..000000000000 --- a/Documentation/networking/wan-router.txt +++ /dev/null @@ -1,621 +0,0 @@ ------------------------------------------------------------------------------- -Linux WAN Router Utilities Package ------------------------------------------------------------------------------- -Version 2.2.1 -Mar 28, 2001 -Author: Nenad Corbic -Copyright (c) 1995-2001 Sangoma Technologies Inc. ------------------------------------------------------------------------------- - -INTRODUCTION - -Wide Area Networks (WANs) are used to interconnect Local Area Networks (LANs) -and/or stand-alone hosts over vast distances with data transfer rates -significantly higher than those achievable with commonly used dial-up -connections. - -Usually an external device called `WAN router' sitting on your local network -or connected to your machine's serial port provides physical connection to -WAN. Although router's job may be as simple as taking your local network -traffic, converting it to WAN format and piping it through the WAN link, these -devices are notoriously expensive, with prices as much as 2 - 5 times higher -then the price of a typical PC box. - -Alternatively, considering robustness and multitasking capabilities of Linux, -an internal router can be built (most routers use some sort of stripped down -Unix-like operating system anyway). With a number of relatively inexpensive WAN -interface cards available on the market, a perfectly usable router can be -built for less than half a price of an external router. Yet a Linux box -acting as a router can still be used for other purposes, such as fire-walling, -running FTP, WWW or DNS server, etc. - -This kernel module introduces the notion of a WAN Link Driver (WLD) to Linux -operating system and provides generic hardware-independent services for such -drivers. Why can existing Linux network device interface not be used for -this purpose? Well, it can. However, there are a few key differences between -a typical network interface (e.g. Ethernet) and a WAN link. - -Many WAN protocols, such as X.25 and frame relay, allow for multiple logical -connections (known as `virtual circuits' in X.25 terminology) over a single -physical link. Each such virtual circuit may (and almost always does) lead -to a different geographical location and, therefore, different network. As a -result, it is the virtual circuit, not the physical link, that represents a -route and, therefore, a network interface in Linux terms. - -To further complicate things, virtual circuits are usually volatile in nature -(excluding so called `permanent' virtual circuits or PVCs). With almost no -time required to set up and tear down a virtual circuit, it is highly desirable -to implement on-demand connections in order to minimize network charges. So -unlike a typical network driver, the WAN driver must be able to handle multiple -network interfaces and cope as multiple virtual circuits come into existence -and go away dynamically. - -Last, but not least, WAN configuration is much more complex than that of say -Ethernet and may well amount to several dozens of parameters. Some of them -are "link-wide" while others are virtual circuit-specific. The same holds -true for WAN statistics which is by far more extensive and extremely useful -when troubleshooting WAN connections. Extending the ifconfig utility to suit -these needs may be possible, but does not seem quite reasonable. Therefore, a -WAN configuration utility and corresponding application programmer's interface -is needed for this purpose. - -Most of these problems are taken care of by this module. Its goal is to -provide a user with more-or-less standard look and feel for all WAN devices and -assist a WAN device driver writer by providing common services, such as: - - o User-level interface via /proc file system - o Centralized configuration - o Device management (setup, shutdown, etc.) - o Network interface management (dynamic creation/destruction) - o Protocol encapsulation/decapsulation - -To ba able to use the Linux WAN Router you will also need a WAN Tools package -available from - - ftp.sangoma.com/pub/linux/current_wanpipe/wanpipe-X.Y.Z.tgz - -where vX.Y.Z represent the wanpipe version number. - -For technical questions and/or comments please e-mail to ncorbic@sangoma.com. -For general inquiries please contact Sangoma Technologies Inc. by - - Hotline: 1-800-388-2475 (USA and Canada, toll free) - Phone: (905) 474-1990 ext: 106 - Fax: (905) 474-9223 - E-mail: dm@sangoma.com (David Mandelstam) - WWW: http://www.sangoma.com - - -INSTALLATION - -Please read the WanpipeForLinux.pdf manual on how to -install the WANPIPE tools and drivers properly. - - -After installing wanpipe package: /usr/local/wanrouter/doc. -On the ftp.sangoma.com : /linux/current_wanpipe/doc - - -COPYRIGHT AND LICENSING INFORMATION - -This program 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. - -This program 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 -this program; if not, write to the Free Software Foundation, Inc., 675 Mass -Ave, Cambridge, MA 02139, USA. - - - -ACKNOWLEDGEMENTS - -This product is based on the WANPIPE(tm) Multiprotocol WAN Router developed -by Sangoma Technologies Inc. for Linux 2.0.x and 2.2.x. Success of the WANPIPE -together with the next major release of Linux kernel in summer 1996 commanded -adequate changes to the WANPIPE code to take full advantage of new Linux -features. - -Instead of continuing developing proprietary interface tied to Sangoma WAN -cards, we decided to separate all hardware-independent code into a separate -module and defined two levels of interfaces - one for user-level applications -and another for kernel-level WAN drivers. WANPIPE is now implemented as a -WAN driver compliant with the WAN Link Driver interface. Also a general -purpose WAN configuration utility and a set of shell scripts was developed to -support WAN router at the user level. - -Many useful ideas concerning hardware-independent interface implementation -were given by Mike McLagan and his implementation -of the Frame Relay router and drivers for Sangoma cards (dlci/sdla). - -With the new implementation of the APIs being incorporated into the WANPIPE, -a special thank goes to Alan Cox in providing insight into BSD sockets. - -Special thanks to all the WANPIPE users who performed field-testing, reported -bugs and made valuable comments and suggestions that help us to improve this -product. - - - -NEW IN THIS RELEASE - - o Updated the WANCFG utility - Calls the pppconfig to configure the PPPD - for async connections. - - o Added the PPPCONFIG utility - Used to configure the PPPD daemon for the - WANPIPE Async PPP and standard serial port. - The wancfg calls the pppconfig to configure - the pppd. - - o Fixed the PCI autodetect feature. - The SLOT 0 was used as an autodetect option - however, some high end PC's slot numbers start - from 0. - - o This release has been tested with the new backupd - daemon release. - - -PRODUCT COMPONENTS AND RELATED FILES - -/etc: (or user defined) - wanpipe1.conf default router configuration file - -/lib/modules/X.Y.Z/misc: - wanrouter.o router kernel loadable module - af_wanpipe.o wanpipe api socket module - -/lib/modules/X.Y.Z/net: - sdladrv.o Sangoma SDLA support module - wanpipe.o Sangoma WANPIPE(tm) driver module - -/proc/net/wanrouter - Config reads current router configuration - Status reads current router status - {name} reads WAN driver statistics - -/usr/sbin: - wanrouter wanrouter start-up script - wanconfig wanrouter configuration utility - sdladump WANPIPE adapter memory dump utility - fpipemon Monitor for Frame Relay - cpipemon Monitor for Cisco HDLC - ppipemon Monitor for PPP - xpipemon Monitor for X25 - wpkbdmon WANPIPE keyboard led monitor/debugger - -/usr/local/wanrouter: - README this file - COPYING GNU General Public License - Setup installation script - Filelist distribution definition file - wanrouter.rc meta-configuration file - (used by the Setup and wanrouter script) - -/usr/local/wanrouter/doc: - wanpipeForLinux.pdf WAN Router User's Manual - -/usr/local/wanrouter/patches: - wanrouter-v2213.gz patch for Linux kernels 2.2.11 up to 2.2.13. - wanrouter-v2214.gz patch for Linux kernel 2.2.14. - wanrouter-v2215.gz patch for Linux kernels 2.2.15 to 2.2.17. - wanrouter-v2218.gz patch for Linux kernels 2.2.18 and up. - wanrouter-v240.gz patch for Linux kernel 2.4.0. - wanrouter-v242.gz patch for Linux kernel 2.4.2 and up. - wanrouter-v2034.gz patch for Linux kernel 2.0.34 - wanrouter-v2036.gz patch for Linux kernel 2.0.36 and up. - -/usr/local/wanrouter/patches/kdrivers: - Sources of the latest WANPIPE device drivers. - These are used to UPGRADE the linux kernel to the newest - version if the kernel source has already been patched with - WANPIPE drivers. - -/usr/local/wanrouter/samples: - interface sample interface configuration file - wanpipe1.cpri CHDLC primary port - wanpipe2.csec CHDLC secondary port - wanpipe1.fr Frame Relay protocol - wanpipe1.ppp PPP protocol ) - wanpipe1.asy CHDLC ASYNC protocol - wanpipe1.x25 X25 protocol - wanpipe1.stty Sync TTY driver (Used by Kernel PPPD daemon) - wanpipe1.atty Async TTY driver (Used by Kernel PPPD daemon) - wanrouter.rc sample meta-configuration file - -/usr/local/wanrouter/util: - * wan-tools utilities source code - -/usr/local/wanrouter/api/x25: - * x25 api sample programs. -/usr/local/wanrouter/api/chdlc: - * chdlc api sample programs. -/usr/local/wanrouter/api/fr: - * fr api sample programs. -/usr/local/wanrouter/config/wancfg: - wancfg WANPIPE GUI configuration program. - Creates wanpipe#.conf files. -/usr/local/wanrouter/config/cfgft1: - cfgft1 GUI CSU/DSU configuration program. - -/usr/include/linux: - wanrouter.h router API definitions - wanpipe.h WANPIPE API definitions - sdladrv.h SDLA support module API definitions - sdlasfm.h SDLA firmware module definitions - if_wanpipe.h WANPIPE Socket definitions - sdlapci.h WANPIPE PCI definitions - - -/usr/src/linux/net/wanrouter: - * wanrouter source code - -/var/log: - wanrouter wanrouter start-up log (created by the Setup script) - -/var/lock: (or /var/lock/subsys for RedHat) - wanrouter wanrouter lock file (created by the Setup script) - -/usr/local/wanrouter/firmware: - fr514.sfm Frame relay firmware for Sangoma S508/S514 card - cdual514.sfm Dual Port Cisco HDLC firmware for Sangoma S508/S514 card - ppp514.sfm PPP Firmware for Sangoma S508 and S514 cards - x25_508.sfm X25 Firmware for Sangoma S508 card. - - -REVISION HISTORY - -1.0.0 December 31, 1996 Initial version - -1.0.1 January 30, 1997 Status and statistics can be read via /proc - filesystem entries. - -1.0.2 April 30, 1997 Added UDP management via monitors. - -1.0.3 June 3, 1997 UDP management for multiple boards using Frame - Relay and PPP - Enabled continuous transmission of Configure - Request Packet for PPP (for 508 only) - Connection Timeout for PPP changed from 900 to 0 - Flow Control Problem fixed for Frame Relay - -1.0.4 July 10, 1997 S508/FT1 monitoring capability in fpipemon and - ppipemon utilities. - Configurable TTL for UDP packets. - Multicast and Broadcast IP source addresses are - silently discarded. - -1.0.5 July 28, 1997 Configurable T391,T392,N391,N392,N393 for Frame - Relay in router.conf. - Configurable Memory Address through router.conf - for Frame Relay, PPP and X.25. (commenting this - out enables auto-detection). - Fixed freeing up received buffers using kfree() - for Frame Relay and X.25. - Protect sdla_peek() by calling save_flags(), - cli() and restore_flags(). - Changed number of Trace elements from 32 to 20 - Added DLCI specific data monitoring in FPIPEMON. -2.0.0 Nov 07, 1997 Implemented protection of RACE conditions by - critical flags for FRAME RELAY and PPP. - DLCI List interrupt mode implemented. - IPX support in FRAME RELAY and PPP. - IPX Server Support (MARS) - More driver specific stats included in FPIPEMON - and PIPEMON. - -2.0.1 Nov 28, 1997 Bug Fixes for version 2.0.0. - Protection of "enable_irq()" while - "disable_irq()" has been enabled from any other - routine (for Frame Relay, PPP and X25). - Added additional Stats for Fpipemon and Ppipemon - Improved Load Sharing for multiple boards - -2.0.2 Dec 09, 1997 Support for PAP and CHAP for ppp has been - implemented. - -2.0.3 Aug 15, 1998 New release supporting Cisco HDLC, CIR for Frame - relay, Dynamic IP assignment for PPP and Inverse - Arp support for Frame-relay. Man Pages are - included for better support and a new utility - for configuring FT1 cards. - -2.0.4 Dec 09, 1998 Dual Port support for Cisco HDLC. - Support for HDLC (LAPB) API. - Supports BiSync Streaming code for S502E - and S503 cards. - Support for Streaming HDLC API. - Provides a BSD socket interface for - creating applications using BiSync - streaming. - -2.0.5 Aug 04, 1999 CHDLC initialization bug fix. - PPP interrupt driven driver: - Fix to the PPP line hangup problem. - New PPP firmware - Added comments to the startup SYSTEM ERROR messages - Xpipemon debugging application for the X25 protocol - New USER_MANUAL.txt - Fixed the odd boundary 4byte writes to the board. - BiSync Streaming code has been taken out. - Available as a patch. - Streaming HDLC API has been taken out. - Available as a patch. - -2.0.6 Aug 17, 1999 Increased debugging in statup scripts - Fixed installation bugs from 2.0.5 - Kernel patch works for both 2.2.10 and 2.2.11 kernels. - There is no functional difference between the two packages - -2.0.7 Aug 26, 1999 o Merged X25API code into WANPIPE. - o Fixed a memory leak for X25API - o Updated the X25API code for 2.2.X kernels. - o Improved NEM handling. - -2.1.0 Oct 25, 1999 o New code for S514 PCI Card - o New CHDLC and Frame Relay drivers - o PPP and X25 are not supported in this release - -2.1.1 Nov 30, 1999 o PPP support for S514 PCI Cards - -2.1.3 Apr 06, 2000 o Socket based x25api - o Socket based chdlc api - o Socket based fr api - o Dual Port Receive only CHDLC support. - o Asynchronous CHDLC support (Secondary Port) - o cfgft1 GUI csu/dsu configurator - o wancfg GUI configuration file - configurator. - o Architectural directory changes. - -beta-2.1.4 Jul 2000 o Dynamic interface configuration: - Network interfaces reflect the state - of protocol layer. If the protocol becomes - disconnected, driver will bring down - the interface. Once the protocol reconnects - the interface will be brought up. - - Note: This option is turned off by default. - - o Dynamic wanrouter setup using 'wanconfig': - wanconfig utility can be used to - shutdown,restart,start or reconfigure - a virtual circuit dynamically. - - Frame Relay: Each DLCI can be: - created,stopped,restarted and reconfigured - dynamically using wanconfig. - - ex: wanconfig card wanpipe1 dev wp1_fr16 up - - o Wanrouter startup via command line arguments: - wanconfig also supports wanrouter startup via command line - arguments. Thus, there is no need to create a wanpipe#.conf - configuration file. - - o Socket based x25api update/bug fixes. - Added support for LCN numbers greater than 255. - Option to pass up modem messages. - Provided a PCI IRQ check, so a single S514 - card is guaranteed to have a non-sharing interrupt. - - o Fixes to the wancfg utility. - o New FT1 debugging support via *pipemon utilities. - o Frame Relay ARP support Enabled. - -beta3-2.1.4 Jul 2000 o X25 M_BIT Problem fix. - o Added the Multi-Port PPP - Updated utilities for the Multi-Port PPP. - -2.1.4 Aut 2000 - o In X25API: - Maximum packet an application can send - to the driver has been extended to 4096 bytes. - - Fixed the x25 startup bug. Enable - communications only after all interfaces - come up. HIGH SVC/PVC is used to calculate - the number of channels. - Enable protocol only after all interfaces - are enabled. - - o Added an extra state to the FT1 config, kernel module. - o Updated the pipemon debuggers. - - o Blocked the Multi-Port PPP from running on kernels - 2.2.16 or greater, due to syncppp kernel module - change. - -beta1-2.1.5 Nov 15 2000 - o Fixed the MultiPort PPP Support for kernels 2.2.16 and above. - 2.2.X kernels only - - o Secured the driver UDP debugging calls - - All illegal network debugging calls are reported to - the log. - - Defined a set of allowed commands, all other denied. - - o Cpipemon - - Added set FT1 commands to the cpipemon. Thus CSU/DSU - configuration can be performed using cpipemon. - All systems that cannot run cfgft1 GUI utility should - use cpipemon to configure the on board CSU/DSU. - - - o Keyboard Led Monitor/Debugger - - A new utility /usr/sbin/wpkbdmon uses keyboard leds - to convey operational statistic information of the - Sangoma WANPIPE cards. - NUM_LOCK = Line State (On=connected, Off=disconnected) - CAPS_LOCK = Tx data (On=transmitting, Off=no tx data) - SCROLL_LOCK = Rx data (On=receiving, Off=no rx data - - o Hardware probe on module load and dynamic device allocation - - During WANPIPE module load, all Sangoma cards are probed - and found information is printed in the /var/log/messages. - - If no cards are found, the module load fails. - - Appropriate number of devices are dynamically loaded - based on the number of Sangoma cards found. - - Note: The kernel configuration option - CONFIG_WANPIPE_CARDS has been taken out. - - o Fixed the Frame Relay and Chdlc network interfaces so they are - compatible with libpcap libraries. Meaning, tcpdump, snort, - ethereal, and all other packet sniffers and debuggers work on - all WANPIPE network interfaces. - - Set the network interface encoding type to ARPHRD_PPP. - This tell the sniffers that data obtained from the - network interface is in pure IP format. - Fix for 2.2.X kernels only. - - o True interface encoding option for Frame Relay and CHDLC - - The above fix sets the network interface encoding - type to ARPHRD_PPP, however some customers use - the encoding interface type to determine the - protocol running. Therefore, the TURE ENCODING - option will set the interface type back to the - original value. - - NOTE: If this option is used with Frame Relay and CHDLC - libpcap library support will be broken. - i.e. tcpdump will not work. - Fix for 2.2.x Kernels only. - - o Ethernet Bridgind over Frame Relay - - The Frame Relay bridging has been developed by - Kristian Hoffmann and Mark Wells. - - The Linux kernel bridge is used to send ethernet - data over the frame relay links. - For 2.2.X Kernels only. - - o Added extensive 2.0.X support. Most new features of - 2.1.5 for protocols Frame Relay, PPP and CHDLC are - supported under 2.0.X kernels. - -beta1-2.2.0 Dec 30 2000 - o Updated drivers for 2.4.X kernels. - o Updated drivers for SMP support. - o X25API is now able to share PCI interrupts. - o Took out a general polling routine that was used - only by X25API. - o Added appropriate locks to the dynamic reconfiguration - code. - o Fixed a bug in the keyboard debug monitor. - -beta2-2.2.0 Jan 8 2001 - o Patches for 2.4.0 kernel - o Patches for 2.2.18 kernel - o Minor updates to PPP and CHLDC drivers. - Note: No functional difference. - -beta3-2.2.9 Jan 10 2001 - o I missed the 2.2.18 kernel patches in beta2-2.2.0 - release. They are included in this release. - -Stable Release -2.2.0 Feb 01 2001 - o Bug fix in wancfg GUI configurator. - The edit function didn't work properly. - - -bata1-2.2.1 Feb 09 2001 - o WANPIPE TTY Driver emulation. - Two modes of operation Sync and Async. - Sync: Using the PPPD daemon, kernel SyncPPP layer - and the Wanpipe sync TTY driver: a PPP protocol - connection can be established via Sangoma adapter, over - a T1 leased line. - - The 2.4.0 kernel PPP layer supports MULTILINK - protocol, that can be used to bundle any number of Sangoma - adapters (T1 lines) into one, under a single IP address. - Thus, efficiently obtaining multiple T1 throughput. - - NOTE: The remote side must also implement MULTILINK PPP - protocol. - - Async:Using the PPPD daemon, kernel AsyncPPP layer - and the WANPIPE async TTY driver: a PPP protocol - connection can be established via Sangoma adapter and - a modem, over a telephone line. - - Thus, the WANPIPE async TTY driver simulates a serial - TTY driver that would normally be used to interface the - MODEM to the linux kernel. - - o WANPIPE PPP Backup Utility - This utility will monitor the state of the PPP T1 line. - In case of failure, a dial up connection will be established - via pppd daemon, ether via a serial tty driver (serial port), - or a WANPIPE async TTY driver (in case serial port is unavailable). - - Furthermore, while in dial up mode, the primary PPP T1 link - will be monitored for signs of life. - - If the PPP T1 link comes back to life, the dial up connection - will be shutdown and T1 line re-established. - - - o New Setup installation script. - Option to UPGRADE device drivers if the kernel source has - already been patched with WANPIPE. - - Option to COMPILE WANPIPE modules against the currently - running kernel, thus no need for manual kernel and module - re-compilation. - - o Updates and Bug Fixes to wancfg utility. - -bata2-2.2.1 Feb 20 2001 - - o Bug fixes to the CHDLC device drivers. - The driver had compilation problems under kernels - 2.2.14 or lower. - - o Bug fixes to the Setup installation script. - The device drivers compilation options didn't work - properly. - - o Update to the wpbackupd daemon. - Optimized the cross-over times, between the primary - link and the backup dialup. - -beta3-2.2.1 Mar 02 2001 - o Patches for 2.4.2 kernel. - - o Bug fixes to util/ make files. - o Bug fixes to the Setup installation script. - - o Took out the backupd support and made it into - as separate package. - -beta4-2.2.1 Mar 12 2001 - - o Fix to the Frame Relay Device driver. - IPSAC sends a packet of zero length - header to the frame relay driver. The - driver tries to push its own 2 byte header - into the packet, which causes the driver to - crash. - - o Fix the WANPIPE re-configuration code. - Bug was found by trying to run the cfgft1 while the - interface was already running. - - o Updates to cfgft1. - Writes a wanpipe#.cfgft1 configuration file - once the CSU/DSU is configured. This file can - holds the current CSU/DSU configuration. - - - ->>>>>> END OF README <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< - - diff --git a/include/linux/Kbuild b/include/linux/Kbuild index e56b739d8e23..b3d9ccde0c27 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -231,7 +231,6 @@ unifdef-y += if_pppol2tp.h unifdef-y += if_pppox.h unifdef-y += if_tr.h unifdef-y += if_vlan.h -unifdef-y += if_wanpipe.h unifdef-y += igmp.h unifdef-y += inet_diag.h unifdef-y += in.h diff --git a/include/linux/if_wanpipe.h b/include/linux/if_wanpipe.h deleted file mode 100644 index e594ca6069e5..000000000000 --- a/include/linux/if_wanpipe.h +++ /dev/null @@ -1,124 +0,0 @@ -/***************************************************************************** -* if_wanpipe.h Header file for the Sangoma AF_WANPIPE Socket -* -* Author: Nenad Corbic -* -* Copyright: (c) 2000 Sangoma Technologies Inc. -* -* This program 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 of the License, or (at your option) any later version. -* ============================================================================ -* -* Jan 28, 2000 Nenad Corbic Initial Version -* -*****************************************************************************/ - -#ifndef __LINUX_IF_WAN_PACKET_H -#define __LINUX_IF_WAN_PACKET_H - -struct wan_sockaddr_ll -{ - unsigned short sll_family; - unsigned short sll_protocol; - int sll_ifindex; - unsigned short sll_hatype; - unsigned char sll_pkttype; - unsigned char sll_halen; - unsigned char sll_addr[8]; - unsigned char sll_device[14]; - unsigned char sll_card[14]; -}; - -typedef struct -{ - unsigned char free; - unsigned char state_sk; - int rcvbuf; - int sndbuf; - int rmem; - int wmem; - int sk_count; - unsigned char bound; - char name[14]; - unsigned char d_state; - unsigned char svc; - unsigned short lcn; - unsigned char mbox; - unsigned char cmd_busy; - unsigned char command; - unsigned poll; - unsigned poll_cnt; - int rblock; -} wan_debug_hdr_t; - -#define MAX_NUM_DEBUG 10 -#define X25_PROT 0x16 -#define PVC_PROT 0x17 - -typedef struct -{ - wan_debug_hdr_t debug[MAX_NUM_DEBUG]; -}wan_debug_t; - -#define SIOC_WANPIPE_GET_CALL_DATA (SIOCPROTOPRIVATE + 0) -#define SIOC_WANPIPE_SET_CALL_DATA (SIOCPROTOPRIVATE + 1) -#define SIOC_WANPIPE_ACCEPT_CALL (SIOCPROTOPRIVATE + 2) -#define SIOC_WANPIPE_CLEAR_CALL (SIOCPROTOPRIVATE + 3) -#define SIOC_WANPIPE_RESET_CALL (SIOCPROTOPRIVATE + 4) -#define SIOC_WANPIPE_DEBUG (SIOCPROTOPRIVATE + 5) -#define SIOC_WANPIPE_SET_NONBLOCK (SIOCPROTOPRIVATE + 6) -#define SIOC_WANPIPE_CHECK_TX (SIOCPROTOPRIVATE + 7) -#define SIOC_WANPIPE_SOCK_STATE (SIOCPROTOPRIVATE + 8) - -/* Packet types */ - -#define WAN_PACKET_HOST 0 /* To us */ -#define WAN_PACKET_BROADCAST 1 /* To all */ -#define WAN_PACKET_MULTICAST 2 /* To group */ -#define WAN_PACKET_OTHERHOST 3 /* To someone else */ -#define WAN_PACKET_OUTGOING 4 /* Outgoing of any type */ -/* These ones are invisible by user level */ -#define WAN_PACKET_LOOPBACK 5 /* MC/BRD frame looped back */ -#define WAN_PACKET_FASTROUTE 6 /* Fastrouted frame */ - - -/* X25 specific */ -#define WAN_PACKET_DATA 7 -#define WAN_PACKET_CMD 8 -#define WAN_PACKET_ASYNC 9 -#define WAN_PACKET_ERR 10 - -/* Packet socket options */ - -#define WAN_PACKET_ADD_MEMBERSHIP 1 -#define WAN_PACKET_DROP_MEMBERSHIP 2 - -#define WAN_PACKET_MR_MULTICAST 0 -#define WAN_PACKET_MR_PROMISC 1 -#define WAN_PACKET_MR_ALLMULTI 2 - -#ifdef __KERNEL__ - -/* Private wanpipe socket structures. */ -struct wanpipe_opt -{ - void *mbox; /* Mail box */ - void *card; /* Card bouded to */ - struct net_device *dev; /* Bounded device */ - unsigned short lcn; /* Binded LCN */ - unsigned char svc; /* 0=pvc, 1=svc */ - unsigned char timer; /* flag for delayed transmit*/ - struct timer_list tx_timer; - unsigned poll_cnt; - unsigned char force; /* Used to force sock release */ - atomic_t packet_sent; - unsigned short num; -}; - -#define wp_sk(__sk) ((struct wanpipe_opt *)(__sk)->sk_protinfo) - -#endif - -#endif -- cgit v1.2.3 From 7ef3abd2104232a35f259dad6a213310edc7c9fe Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 15 Apr 2008 00:29:24 -0700 Subject: [IRDA]: Remove irlan_eth_send_gratuitous_arp() Even kernel 2.2.26 (sic) already contains the #undef CONFIG_IRLAN_SEND_GRATUITOUS_ARP with the comment "but for some reason the machine crashes if you use DHCP". Either someone finally looks into this or it's simply time to remove this dead code. Reported-by: Robert P. J. Day Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- include/net/irda/irlan_eth.h | 1 - net/irda/irlan/irlan_common.c | 10 ---------- net/irda/irlan/irlan_eth.c | 33 --------------------------------- 3 files changed, 44 deletions(-) (limited to 'include') diff --git a/include/net/irda/irlan_eth.h b/include/net/irda/irlan_eth.h index 0062347600b9..de5c81691f33 100644 --- a/include/net/irda/irlan_eth.h +++ b/include/net/irda/irlan_eth.h @@ -29,5 +29,4 @@ struct net_device *alloc_irlandev(const char *name); int irlan_eth_receive(void *instance, void *sap, struct sk_buff *skb); void irlan_eth_flow_indication( void *instance, void *sap, LOCAL_FLOW flow); -void irlan_eth_send_gratuitous_arp(struct net_device *dev); #endif diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c index 75bb6a9dcaaa..9a1cd87e7142 100644 --- a/net/irda/irlan/irlan_common.c +++ b/net/irda/irlan/irlan_common.c @@ -54,13 +54,6 @@ #include -/* - * Send gratuitous ARP when connected to a new AP or not. May be a clever - * thing to do, but for some reason the machine crashes if you use DHCP. So - * lets not use it by default. - */ -#undef CONFIG_IRLAN_SEND_GRATUITOUS_ARP - /* extern char sysctl_devname[]; */ /* @@ -393,9 +386,6 @@ static void irlan_connect_confirm(void *instance, void *sap, /* Ready to transfer Ethernet frames */ netif_start_queue(self->dev); self->disconnect_reason = 0; /* Clear reason */ -#ifdef CONFIG_IRLAN_SEND_GRATUITOUS_ARP - irlan_eth_send_gratuitous_arp(&self->dev); -#endif wake_up_interruptible(&self->open_wait); } diff --git a/net/irda/irlan/irlan_eth.c b/net/irda/irlan/irlan_eth.c index 7a6b14ab1e7f..05112be99569 100644 --- a/net/irda/irlan/irlan_eth.c +++ b/net/irda/irlan/irlan_eth.c @@ -288,39 +288,6 @@ void irlan_eth_flow_indication(void *instance, void *sap, LOCAL_FLOW flow) } } -/* - * Function irlan_etc_send_gratuitous_arp (dev) - * - * Send gratuitous ARP to announce that we have changed - * hardware address, so that all peers updates their ARP tables - */ -void irlan_eth_send_gratuitous_arp(struct net_device *dev) -{ -#ifdef CONFIG_INET - struct in_device *in_dev; - - /* - * When we get a new MAC address do a gratuitous ARP. This - * is useful if we have changed access points on the same - * subnet. - */ - IRDA_DEBUG(4, "IrLAN: Sending gratuitous ARP\n"); - rcu_read_lock(); - in_dev = __in_dev_get_rcu(dev); - if (in_dev == NULL) - goto out; - if (in_dev->ifa_list) - - arp_send(ARPOP_REQUEST, ETH_P_ARP, - in_dev->ifa_list->ifa_address, - dev, - in_dev->ifa_list->ifa_address, - NULL, dev->dev_addr, NULL); -out: - rcu_read_unlock(); -#endif /* CONFIG_INET */ -} - /* * Function set_multicast_list (dev) * -- cgit v1.2.3 From 31efdf0530b6351b0658d35a602a0f2d6bc2ed6f Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 15 Apr 2008 00:30:16 -0700 Subject: [ISDN] include/linux/isdn.h: remove dead code This patch remove the usage of a nonexisting kconfig variable. Reported-by: Robert P. J. Day Signed-off-by: Adrian Bunk Signed-off-by: David S. Miller --- include/linux/isdn.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include') diff --git a/include/linux/isdn.h b/include/linux/isdn.h index 9cb2855bb170..44cd663c53b6 100644 --- a/include/linux/isdn.h +++ b/include/linux/isdn.h @@ -16,14 +16,8 @@ #include -#ifdef CONFIG_COBALT_MICRO_SERVER -/* Save memory */ -#define ISDN_MAX_DRIVERS 2 -#define ISDN_MAX_CHANNELS 8 -#else #define ISDN_MAX_DRIVERS 32 #define ISDN_MAX_CHANNELS 64 -#endif /* New ioctl-codes */ #define IIOCNETAIF _IO('I',1) -- cgit v1.2.3 From c93cf61fd1d5378134f9b06703f7078067542e00 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Tue, 15 Apr 2008 00:35:23 -0700 Subject: [NETNS]: The net-subsys IDs generator. To make some per-net generic pointers, we need some way to address them, i.e. - IDs. This is simple IDA-based IDs generator for pernet subsystems. Addressing questions about potential checkpoint/restart problems: these IDs are "lite-offsets" within the net structure and are by no means supposed to be exported to the userspace. Since it will be used in the nearest future by devices only (tun, vlan, tunnels, bridge, etc), I make it resemble the functionality of register_pernet_device(). The new ids is stored in the *id pointer _before_ calling the init callback to make this id available in this callback. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/net_namespace.h | 2 ++ net/core/net_namespace.c | 38 ++++++++++++++++++++++++++++++++++++++ 2 files changed, 40 insertions(+) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index e2aee2689abe..72fad1a0956b 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -185,6 +185,8 @@ extern int register_pernet_subsys(struct pernet_operations *); extern void unregister_pernet_subsys(struct pernet_operations *); extern int register_pernet_device(struct pernet_operations *); extern void unregister_pernet_device(struct pernet_operations *); +extern int register_pernet_gen_device(int *id, struct pernet_operations *); +extern void unregister_pernet_gen_device(int id, struct pernet_operations *); struct ctl_path; struct ctl_table; diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 7b660834a4c2..2197d51aef3b 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -5,6 +5,7 @@ #include #include #include +#include #include /* @@ -253,6 +254,8 @@ static void unregister_pernet_operations(struct pernet_operations *ops) } #endif +static DEFINE_IDA(net_generic_ids); + /** * register_pernet_subsys - register a network namespace subsystem * @ops: pernet operations structure for the subsystem @@ -330,6 +333,30 @@ int register_pernet_device(struct pernet_operations *ops) } EXPORT_SYMBOL_GPL(register_pernet_device); +int register_pernet_gen_device(int *id, struct pernet_operations *ops) +{ + int error; + mutex_lock(&net_mutex); +again: + error = ida_get_new_above(&net_generic_ids, 1, id); + if (error) { + if (error == -EAGAIN) { + ida_pre_get(&net_generic_ids, GFP_KERNEL); + goto again; + } + goto out; + } + error = register_pernet_operations(&pernet_list, ops); + if (error) + ida_remove(&net_generic_ids, *id); + else if (first_device == &pernet_list) + first_device = &ops->list; +out: + mutex_unlock(&net_mutex); + return error; +} +EXPORT_SYMBOL_GPL(register_pernet_gen_device); + /** * unregister_pernet_device - unregister a network namespace netdevice * @ops: pernet operations structure to manipulate @@ -348,3 +375,14 @@ void unregister_pernet_device(struct pernet_operations *ops) mutex_unlock(&net_mutex); } EXPORT_SYMBOL_GPL(unregister_pernet_device); + +void unregister_pernet_gen_device(int id, struct pernet_operations *ops) +{ + mutex_lock(&net_mutex); + if (&ops->list == first_device) + first_device = first_device->next; + unregister_pernet_operations(ops); + ida_remove(&net_generic_ids, id); + mutex_unlock(&net_mutex); +} +EXPORT_SYMBOL_GPL(unregister_pernet_gen_device); -- cgit v1.2.3 From dec827d174d7f76c457238800183ca864a639365 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Tue, 15 Apr 2008 00:36:08 -0700 Subject: [NETNS]: The generic per-net pointers. Add the elastic array of void * pointer to the struct net. The access rules are simple: 1. register the ops with register_pernet_gen_device to get the id of your private pointer 2. call net_assign_generic() to put the private data on the struct net (most preferably this should be done in the ->init callback of the ops registered) 3. do not store any private reference on the net_generic array; 4. do not change this pointer while the net is alive; 5. use the net_generic() to get the pointer. When adding a new pointer, I copy the old array, replace it with a new one and schedule the old for kfree after an RCU grace period. Since the net_generic explores the net->gen array inside rcu read section and once set the net->gen->ptr[x] pointer never changes, this grants us a safe access to generic pointers. Quoting Paul: "... RCU is protecting -only- the net_generic structure that net_generic() is traversing, and the [pointer] returned by net_generic() is protected by a reference counter in the upper-level struct net." Signed-off-by: Pavel Emelyanov Acked-by: Paul E. McKenney Signed-off-by: David S. Miller --- include/net/net_namespace.h | 2 ++ include/net/netns/generic.h | 49 +++++++++++++++++++++++++++++++++++ net/core/net_namespace.c | 62 +++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 113 insertions(+) create mode 100644 include/net/netns/generic.h (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index 72fad1a0956b..f880b0f9f107 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -20,6 +20,7 @@ struct proc_dir_entry; struct net_device; struct sock; struct ctl_table_header; +struct net_generic; struct net { atomic_t count; /* To decided when the network @@ -61,6 +62,7 @@ struct net { #ifdef CONFIG_NETFILTER struct netns_xt xt; #endif + struct net_generic *gen; }; diff --git a/include/net/netns/generic.h b/include/net/netns/generic.h new file mode 100644 index 000000000000..0c04fd2a700b --- /dev/null +++ b/include/net/netns/generic.h @@ -0,0 +1,49 @@ +/* + * generic net pointers + */ + +#ifndef __NET_GENERIC_H__ +#define __NET_GENERIC_H__ + +#include + +/* + * Generic net pointers are to be used by modules to put some private + * stuff on the struct net without explicit struct net modification + * + * The rules are simple: + * 1. register the ops with register_pernet_gen_device to get the id + * of your private pointer; + * 2. call net_assign_generic() to put the private data on the struct + * net (most preferably this should be done in the ->init callback + * of the ops registered); + * 3. do not change this pointer while the net is alive; + * 4. do not try to have any private reference on the net_generic object. + * + * After accomplishing all of the above, the private pointer can be + * accessed with the net_generic() call. + */ + +struct net_generic { + unsigned int len; + struct rcu_head rcu; + + void *ptr[0]; +}; + +static inline void *net_generic(struct net *net, int id) +{ + struct net_generic *ng; + void *ptr; + + rcu_read_lock(); + ng = rcu_dereference(net->gen); + BUG_ON(id == 0 || id > ng->len); + ptr = ng->ptr[id - 1]; + rcu_read_unlock(); + + return ptr; +} + +extern int net_assign_generic(struct net *net, int id, void *data); +#endif diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 2197d51aef3b..763674e1e593 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -7,6 +7,7 @@ #include #include #include +#include /* * Our network namespace constructor/destructor lists @@ -21,6 +22,8 @@ LIST_HEAD(net_namespace_list); struct net init_net; EXPORT_SYMBOL(init_net); +#define INITIAL_NET_GEN_PTRS 13 /* +1 for len +2 for rcu_head */ + /* * setup_net runs the initializers for the network namespace object. */ @@ -29,10 +32,21 @@ static __net_init int setup_net(struct net *net) /* Must be called with net_mutex held */ struct pernet_operations *ops; int error; + struct net_generic *ng; atomic_set(&net->count, 1); atomic_set(&net->use_count, 0); + error = -ENOMEM; + ng = kzalloc(sizeof(struct net_generic) + + INITIAL_NET_GEN_PTRS * sizeof(void *), GFP_KERNEL); + if (ng == NULL) + goto out; + + ng->len = INITIAL_NET_GEN_PTRS; + INIT_RCU_HEAD(&ng->rcu); + rcu_assign_pointer(net->gen, ng); + error = 0; list_for_each_entry(ops, &pernet_list, list) { if (ops->init) { @@ -54,6 +68,7 @@ out_undo: } rcu_barrier(); + kfree(ng); goto out; } @@ -386,3 +401,50 @@ void unregister_pernet_gen_device(int id, struct pernet_operations *ops) mutex_unlock(&net_mutex); } EXPORT_SYMBOL_GPL(unregister_pernet_gen_device); + +static void net_generic_release(struct rcu_head *rcu) +{ + struct net_generic *ng; + + ng = container_of(rcu, struct net_generic, rcu); + kfree(ng); +} + +int net_assign_generic(struct net *net, int id, void *data) +{ + struct net_generic *ng, *old_ng; + + BUG_ON(!mutex_is_locked(&net_mutex)); + BUG_ON(id == 0); + + ng = old_ng = net->gen; + if (old_ng->len >= id) + goto assign; + + ng = kzalloc(sizeof(struct net_generic) + + id * sizeof(void *), GFP_KERNEL); + if (ng == NULL) + return -ENOMEM; + + /* + * Some synchronisation notes: + * + * The net_generic explores the net->gen array inside rcu + * read section. Besides once set the net->gen->ptr[x] + * pointer never changes (see rules in netns/generic.h). + * + * That said, we simply duplicate this array and schedule + * the old copy for kfree after a grace period. + */ + + ng->len = id; + INIT_RCU_HEAD(&ng->rcu); + memcpy(&ng->ptr, &old_ng->ptr, old_ng->len); + + rcu_assign_pointer(net->gen, ng); + call_rcu(&old_ng->rcu, net_generic_release); +assign: + ng->ptr[id - 1] = data; + return 0; +} +EXPORT_SYMBOL_GPL(net_assign_generic); -- cgit v1.2.3 From 669f87baab90183e13b95480aecf8d7bac92ca3c Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 16 Apr 2008 00:46:52 -0700 Subject: [RTNL]: Introduce the rtnl_kill_links helper. This one is responsible for calling ->dellink on each net device found in net to help with vlan net_exit hook in the nearest future. Signed-off-by: Pavel Emelyanov Acked-by: Patrick McHardy Signed-off-by: David S. Miller --- include/net/rtnetlink.h | 1 + net/core/rtnetlink.c | 29 +++++++++++++++++++++-------- 2 files changed, 22 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h index 793863e09c69..3c1895e54b7f 100644 --- a/include/net/rtnetlink.h +++ b/include/net/rtnetlink.h @@ -74,6 +74,7 @@ struct rtnl_link_ops { extern int __rtnl_link_register(struct rtnl_link_ops *ops); extern void __rtnl_link_unregister(struct rtnl_link_ops *ops); +extern void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops); extern int rtnl_link_register(struct rtnl_link_ops *ops); extern void rtnl_link_unregister(struct rtnl_link_ops *ops); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index edc6dbfe48f2..bc39e417694a 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -269,6 +269,26 @@ int rtnl_link_register(struct rtnl_link_ops *ops) EXPORT_SYMBOL_GPL(rtnl_link_register); +static void __rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops) +{ + struct net_device *dev; +restart: + for_each_netdev(net, dev) { + if (dev->rtnl_link_ops == ops) { + ops->dellink(dev); + goto restart; + } + } +} + +void rtnl_kill_links(struct net *net, struct rtnl_link_ops *ops) +{ + rtnl_lock(); + __rtnl_kill_links(net, ops); + rtnl_unlock(); +} +EXPORT_SYMBOL_GPL(rtnl_kill_links); + /** * __rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink. * @ops: struct rtnl_link_ops * to unregister @@ -277,17 +297,10 @@ EXPORT_SYMBOL_GPL(rtnl_link_register); */ void __rtnl_link_unregister(struct rtnl_link_ops *ops) { - struct net_device *dev; struct net *net; for_each_net(net) { -restart: - for_each_netdev(net, dev) { - if (dev->rtnl_link_ops == ops) { - ops->dellink(dev); - goto restart; - } - } + __rtnl_kill_links(net, ops); } list_del(&ops->list); } -- cgit v1.2.3 From a9fde2607895667823e9d1172fc193087125ef68 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Wed, 16 Apr 2008 00:48:04 -0700 Subject: [VLAN]: Tag vlan_group_device with net device, not ifindex. Currently vlan group is searched using one key - the ifindex. We'll have to lookup the vlan_group by two keys - ifindex and net. Turning the vlan_group lookup key to struct net_device pointer will make this process easier. Besides, this will eliminate one more place in the networking, that assumes that indexes are unique in the kernel. Signed-off-by: Pavel Emelyanov Acked-by: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/if_vlan.h | 4 +++- net/8021q/vlan.c | 22 +++++++++++----------- 2 files changed, 14 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h index edd55af7ebd6..15ace02b7b24 100644 --- a/include/linux/if_vlan.h +++ b/include/linux/if_vlan.h @@ -81,7 +81,9 @@ extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *)); #define VLAN_GROUP_ARRAY_PART_LEN (VLAN_GROUP_ARRAY_LEN/VLAN_GROUP_ARRAY_SPLIT_PARTS) struct vlan_group { - int real_dev_ifindex; /* The ifindex of the ethernet(like) device the vlan is attached to. */ + struct net_device *real_dev; /* The ethernet(like) device + * the vlan is attached to. + */ unsigned int nr_vlans; struct hlist_node hlist; /* linked list */ struct net_device **vlan_devices_arrays[VLAN_GROUP_ARRAY_SPLIT_PARTS]; diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index 5975ec3be7f3..cf8d810a130d 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -65,14 +65,14 @@ static inline unsigned int vlan_grp_hashfn(unsigned int idx) } /* Must be invoked with RCU read lock (no preempt) */ -static struct vlan_group *__vlan_find_group(int real_dev_ifindex) +static struct vlan_group *__vlan_find_group(struct net_device *real_dev) { struct vlan_group *grp; struct hlist_node *n; - int hash = vlan_grp_hashfn(real_dev_ifindex); + int hash = vlan_grp_hashfn(real_dev->ifindex); hlist_for_each_entry_rcu(grp, n, &vlan_group_hash[hash], hlist) { - if (grp->real_dev_ifindex == real_dev_ifindex) + if (grp->real_dev == real_dev) return grp; } @@ -86,7 +86,7 @@ static struct vlan_group *__vlan_find_group(int real_dev_ifindex) struct net_device *__find_vlan_dev(struct net_device *real_dev, unsigned short VID) { - struct vlan_group *grp = __vlan_find_group(real_dev->ifindex); + struct vlan_group *grp = __vlan_find_group(real_dev); if (grp) return vlan_group_get_device(grp, VID); @@ -103,7 +103,7 @@ static void vlan_group_free(struct vlan_group *grp) kfree(grp); } -static struct vlan_group *vlan_group_alloc(int ifindex) +static struct vlan_group *vlan_group_alloc(struct net_device *real_dev) { struct vlan_group *grp; @@ -111,9 +111,9 @@ static struct vlan_group *vlan_group_alloc(int ifindex) if (!grp) return NULL; - grp->real_dev_ifindex = ifindex; + grp->real_dev = real_dev; hlist_add_head_rcu(&grp->hlist, - &vlan_group_hash[vlan_grp_hashfn(ifindex)]); + &vlan_group_hash[vlan_grp_hashfn(real_dev->ifindex)]); return grp; } @@ -151,7 +151,7 @@ void unregister_vlan_dev(struct net_device *dev) ASSERT_RTNL(); - grp = __vlan_find_group(real_dev->ifindex); + grp = __vlan_find_group(real_dev); BUG_ON(!grp); vlan_proc_rem_dev(dev); @@ -246,9 +246,9 @@ int register_vlan_dev(struct net_device *dev) struct vlan_group *grp, *ngrp = NULL; int err; - grp = __vlan_find_group(real_dev->ifindex); + grp = __vlan_find_group(real_dev); if (!grp) { - ngrp = grp = vlan_group_alloc(real_dev->ifindex); + ngrp = grp = vlan_group_alloc(real_dev); if (!grp) return -ENOBUFS; } @@ -412,7 +412,7 @@ static int vlan_device_event(struct notifier_block *unused, unsigned long event, goto out; } - grp = __vlan_find_group(dev->ifindex); + grp = __vlan_find_group(dev); if (!grp) goto out; -- cgit v1.2.3 From 5d1e4468a7705db7c1415a65fd16f07113afc1b2 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 16 Apr 2008 01:58:04 -0700 Subject: [NETNS]: Make netns refconting debug like a socket one. Make release_net/hold_net noop for performance-hungry people. This is a debug staff and should be used in the debug mode only. Add check for net != NULL in hold/release calls. This will be required later on. [ Added minor simplifications suggested by Brian Haley. -DaveM ] Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/net_namespace.h | 40 ++++++++++++++++++++++++---------------- net/core/net_namespace.c | 4 ++++ 2 files changed, 28 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index f880b0f9f107..aa540e6be502 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -26,9 +26,11 @@ struct net { atomic_t count; /* To decided when the network * namespace should be freed. */ +#ifdef NETNS_REFCNT_DEBUG atomic_t use_count; /* To track references we * destroy on demand */ +#endif struct list_head list; /* list of network namespaces */ struct work_struct work; /* work struct for freeing */ @@ -117,17 +119,6 @@ static inline void put_net(struct net *net) __put_net(net); } -static inline struct net *hold_net(struct net *net) -{ - atomic_inc(&net->use_count); - return net; -} - -static inline void release_net(struct net *net) -{ - atomic_dec(&net->use_count); -} - static inline int net_eq(const struct net *net1, const struct net *net2) { @@ -143,27 +134,44 @@ static inline void put_net(struct net *net) { } +static inline struct net *maybe_get_net(struct net *net) +{ + return net; +} + +static inline +int net_eq(const struct net *net1, const struct net *net2) +{ + return 1; +} +#endif + + +#ifdef NETNS_REFCNT_DEBUG static inline struct net *hold_net(struct net *net) { + if (net) + atomic_inc(&net->use_count); return net; } static inline void release_net(struct net *net) { + if (net) + atomic_dec(&net->use_count); } - -static inline struct net *maybe_get_net(struct net *net) +#else +static inline struct net *hold_net(struct net *net) { return net; } -static inline -int net_eq(const struct net *net1, const struct net *net2) +static inline void release_net(struct net *net) { - return 1; } #endif + #define for_each_net(VAR) \ list_for_each_entry(VAR, &net_namespace_list, list) diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 763674e1e593..72b4c184dd84 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -35,7 +35,9 @@ static __net_init int setup_net(struct net *net) struct net_generic *ng; atomic_set(&net->count, 1); +#ifdef NETNS_REFCNT_DEBUG atomic_set(&net->use_count, 0); +#endif error = -ENOMEM; ng = kzalloc(sizeof(struct net_generic) + @@ -86,11 +88,13 @@ static void net_free(struct net *net) if (!net) return; +#ifdef NETNS_REFCNT_DEBUG if (unlikely(atomic_read(&net->use_count) != 0)) { printk(KERN_EMERG "network namespace not free! Usage: %d\n", atomic_read(&net->use_count)); return; } +#endif kmem_cache_free(net_cachep, net); } -- cgit v1.2.3 From 65a18ec58e5e6186103f62f720acea94dfb26f4e Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 16 Apr 2008 01:59:46 -0700 Subject: [NETNS]: Add netns refcnt debug for kernel sockets. Protocol control sockets and netlink kernel sockets should not prevent the namespace stop request. They are initialized and disposed in a special way by sk_change_net/sk_release_kernel. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/sock.h | 2 +- net/core/sock.c | 1 + 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/sock.h b/include/net/sock.h index 09255eae93e9..dc42b44c2aa1 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1314,7 +1314,7 @@ void sock_net_set(struct sock *sk, struct net *net) static inline void sk_change_net(struct sock *sk, struct net *net) { put_net(sock_net(sk)); - sock_net_set(sk, net); + sock_net_set(sk, hold_net(net)); } extern void sock_enable_timestamp(struct sock *sk); diff --git a/net/core/sock.c b/net/core/sock.c index c0ecbdcf75d8..54c836a2216b 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1001,6 +1001,7 @@ void sk_release_kernel(struct sock *sk) sock_hold(sk); sock_release(sk->sk_socket); + release_net(sock_net(sk)); sock_net_set(sk, get_net(&init_net)); sock_put(sk); } -- cgit v1.2.3 From 3661a910836a509be65afc3c1e512d900e1280f9 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 16 Apr 2008 02:01:56 -0700 Subject: [NETNS]: Add netns refcnt debug to fib rules. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/net/fib_rules.h | 1 + net/core/fib_rules.c | 5 +++-- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index 34349f9f4331..a5c6ccc5bb19 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -87,6 +87,7 @@ static inline void fib_rule_get(struct fib_rule *rule) static inline void fib_rule_put_rcu(struct rcu_head *head) { struct fib_rule *rule = container_of(head, struct fib_rule, rcu); + release_net(rule->fr_net); kfree(rule); } diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 540c07283e31..e3e9ab0f74e3 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -29,7 +29,7 @@ int fib_default_rule_add(struct fib_rules_ops *ops, r->pref = pref; r->table = table; r->flags = flags; - r->fr_net = ops->fro_net; + r->fr_net = hold_net(ops->fro_net); /* The lock is not required here, the list in unreacheable * at the moment this function is called */ @@ -243,7 +243,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) err = -ENOMEM; goto errout; } - rule->fr_net = net; + rule->fr_net = hold_net(net); if (tb[FRA_PRIORITY]) rule->pref = nla_get_u32(tb[FRA_PRIORITY]); @@ -344,6 +344,7 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) return 0; errout_free: + release_net(rule->fr_net); kfree(rule); errout: rules_ops_put(ops); -- cgit v1.2.3 From f3005d7f4abe03ad41af33b1548602cd086d86a2 Mon Sep 17 00:00:00 2001 From: "Denis V. Lunev" Date: Wed, 16 Apr 2008 02:02:18 -0700 Subject: [NETNS]: Add netns refcnt debug for network devices. dev_set_net is called for - just allocated devices - devices moving from one namespace to another release_net has proper check inside to distinguish these cases. Signed-off-by: Denis V. Lunev Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 ++- net/core/dev.c | 2 ++ 2 files changed, 4 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 8b17ed40dea2..7c1d4466583b 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -758,7 +758,8 @@ static inline void dev_net_set(struct net_device *dev, struct net *net) { #ifdef CONFIG_NET_NS - dev->nd_net = net; + release_net(dev->nd_net); + dev->nd_net = hold_net(net); #endif } diff --git a/net/core/dev.c b/net/core/dev.c index 7aa01125287e..77530e9a34fc 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -4042,6 +4042,8 @@ EXPORT_SYMBOL(alloc_netdev_mq); */ void free_netdev(struct net_device *dev) { + release_net(dev_net(dev)); + /* Compatibility with error handling in drivers */ if (dev->reg_state == NETREG_UNINITIALIZED) { kfree((char *)dev - dev->padded); -- cgit v1.2.3 From dd9e0dda66ba38a2ddd1405ac279894260dc5c36 Mon Sep 17 00:00:00 2001 From: John Heffner Date: Tue, 15 Apr 2008 15:26:39 -0700 Subject: [TCP]: Increase the max_burst threshold from 3 to tp->reordering. This change is necessary to allow cwnd to grow during persistent reordering. Cwnd moderation is applied when in the disorder state and an ack that fills the hole comes in. If the hole was greater than 3 packets, but less than tp->reordering, cwnd will shrink when it should not have. Signed-off-by: John Heffner Signed-off-by: David S. Miller --- include/net/tcp.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 2c14edf7b076..633147cb6bbc 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -787,11 +787,14 @@ extern void tcp_enter_cwr(struct sock *sk, const int set_ssthresh); extern __u32 tcp_init_cwnd(struct tcp_sock *tp, struct dst_entry *dst); /* Slow start with delack produces 3 packets of burst, so that - * it is safe "de facto". + * it is safe "de facto". This will be the default - same as + * the default reordering threshold - but if reordering increases, + * we must be able to allow cwnd to burst at least this much in order + * to not pull it back when holes are filled. */ static __inline__ __u32 tcp_max_burst(const struct tcp_sock *tp) { - return 3; + return tp->reordering; } /* Returns end sequence number of the receiver's advertised window */ -- cgit v1.2.3 From d18ef29f34eb33099d387a327abe139f3915a829 Mon Sep 17 00:00:00 2001 From: Reinette Chatre Date: Wed, 9 Apr 2008 16:56:15 -0700 Subject: mac80211: no BSS changes to driver from beacons processed during scanning MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There is no need to send BSS changes to driver from beacons processed during scanning. We are more interested in beacons from an AP with which we are associated - these will still be used to send updates to driver as the beacons are received without scanning. This change·removes the requirement that bss_info_changed needs to be atomic. The beacons received during scanning are processed from a tasklet, but if we do not call bss_info_changed for these beacons there is no need for it to be atomic. This function (bss_info_changed) is called either from workqueue or ioctl in all other instances. Signed-off-by: Reinette Chatre Acked-by: Tomas Winkler Acked-by: Johannes Berg Signed-off-by: John W. Linville --- include/net/mac80211.h | 3 +-- net/mac80211/mlme.c | 16 +++++++++++----- 2 files changed, 12 insertions(+), 7 deletions(-) (limited to 'include') diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 079e7bd86c90..4a80d74975e8 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -1020,8 +1020,7 @@ enum ieee80211_ampdu_mlme_action { * level driver (e.g. assoc/disassoc status, erp parameters). * This function should not be used if no BSS has been set, unless * for association indication. The @changed parameter indicates which - * of the bss parameters has changed when a call is made. This callback - * has to be atomic. + * of the bss parameters has changed when a call is made. * * @configure_filter: Configure the device's RX filter. * See the section "Frame filtering" for more information. diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 83e8b497e6db..e3f2cb086588 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2797,6 +2797,17 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev, ieee802_11_parse_elems(mgmt->u.beacon.variable, len - baselen, &elems); + if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) { + ieee80211_sta_wmm_params(dev, ifsta, elems.wmm_param, + elems.wmm_param_len); + } + + /* Do not send changes to driver if we are scanning. This removes + * requirement that driver's bss_info_changed function needs to be + * atomic. */ + if (local->sta_sw_scanning || local->sta_hw_scanning) + return; + if (elems.erp_info && elems.erp_info_len >= 1) changed |= ieee80211_handle_erp_ie(sdata, elems.erp_info[0]); else { @@ -2816,11 +2827,6 @@ static void ieee80211_rx_mgmt_beacon(struct net_device *dev, &bss_info); } - if (elems.wmm_param && (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) { - ieee80211_sta_wmm_params(dev, ifsta, elems.wmm_param, - elems.wmm_param_len); - } - ieee80211_bss_info_change_notify(sdata, changed); } -- cgit v1.2.3 From 9d9326d3bc0ea9a8bbe40bf3e5e66c7b9858caa0 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Wed, 9 Apr 2008 19:38:13 -0500 Subject: phy: Change mii_bus id field to a string Having the id field be an int was making more complex bus topologies excessively difficult. For now, just convert it to a string, and change all instances of "bus->id = val" to snprintf(id, MII_BUS_ID_LEN, "%x", val). Signed-off-by: Andy Fleming Signed-off-by: Jeff Garzik --- arch/powerpc/platforms/82xx/ep8248e.c | 2 +- arch/powerpc/platforms/pasemi/gpio_mdio.c | 2 +- arch/powerpc/sysdev/fsl_soc.c | 5 +++-- drivers/net/au1000_eth.c | 6 +++--- drivers/net/bfin_mac.c | 2 +- drivers/net/cpmac.c | 5 ++--- drivers/net/fec_mpc52xx.c | 2 +- drivers/net/fec_mpc52xx_phy.c | 2 +- drivers/net/fs_enet/fs_enet-main.c | 4 ++-- drivers/net/fs_enet/mii-bitbang.c | 4 ++-- drivers/net/fs_enet/mii-fec.c | 4 ++-- drivers/net/gianfar_mii.c | 2 +- drivers/net/macb.c | 2 +- drivers/net/pasemi_mac.c | 2 +- drivers/net/phy/fixed.c | 2 +- drivers/net/sb1250-mac.c | 2 +- drivers/net/ucc_geth.c | 2 +- drivers/net/ucc_geth.h | 2 +- drivers/net/ucc_geth_mii.c | 2 +- include/linux/fsl_devices.h | 2 +- include/linux/phy.h | 12 ++++++++---- 21 files changed, 36 insertions(+), 32 deletions(-) (limited to 'include') diff --git a/arch/powerpc/platforms/82xx/ep8248e.c b/arch/powerpc/platforms/82xx/ep8248e.c index ba93d8ae9b0c..d5770fdf7f09 100644 --- a/arch/powerpc/platforms/82xx/ep8248e.c +++ b/arch/powerpc/platforms/82xx/ep8248e.c @@ -138,7 +138,7 @@ static int __devinit ep8248e_mdio_probe(struct of_device *ofdev, bus->name = "ep8248e-mdio-bitbang"; bus->dev = &ofdev->dev; - bus->id = res.start; + snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start); return mdiobus_register(bus); } diff --git a/arch/powerpc/platforms/pasemi/gpio_mdio.c b/arch/powerpc/platforms/pasemi/gpio_mdio.c index b46542990cf8..ab6955412ba4 100644 --- a/arch/powerpc/platforms/pasemi/gpio_mdio.c +++ b/arch/powerpc/platforms/pasemi/gpio_mdio.c @@ -241,7 +241,7 @@ static int __devinit gpio_mdio_probe(struct of_device *ofdev, new_bus->reset = &gpio_mdio_reset; prop = of_get_property(np, "reg", NULL); - new_bus->id = *prop; + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", *prop); new_bus->priv = priv; new_bus->phy_mask = 0; diff --git a/arch/powerpc/sysdev/fsl_soc.c b/arch/powerpc/sysdev/fsl_soc.c index 2c5388ce902a..3581416905ea 100644 --- a/arch/powerpc/sysdev/fsl_soc.c +++ b/arch/powerpc/sysdev/fsl_soc.c @@ -341,7 +341,7 @@ static int __init gfar_of_init(void) goto unreg; } - gfar_data.bus_id = 0; + snprintf(gfar_data.bus_id, MII_BUS_ID_SIZE, "0"); gfar_data.phy_id = fixed_link[0]; } else { phy = of_find_node_by_phandle(*ph); @@ -362,7 +362,8 @@ static int __init gfar_of_init(void) } gfar_data.phy_id = *id; - gfar_data.bus_id = res.start; + snprintf(gfar_data.bus_id, MII_BUS_ID_SIZE, "%x", + res.start); of_node_put(phy); of_node_put(mdio); diff --git a/drivers/net/au1000_eth.c b/drivers/net/au1000_eth.c index 504b7ce2747d..3634b5fd7919 100644 --- a/drivers/net/au1000_eth.c +++ b/drivers/net/au1000_eth.c @@ -701,7 +701,7 @@ static struct net_device * au1000_probe(int port_num) aup->mii_bus.write = mdiobus_write; aup->mii_bus.reset = mdiobus_reset; aup->mii_bus.name = "au1000_eth_mii"; - aup->mii_bus.id = aup->mac_id; + snprintf(aup->mii_bus.id, MII_BUS_ID_SIZE, "%x", aup->mac_id); aup->mii_bus.irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL); for(i = 0; i < PHY_MAX_ADDR; ++i) aup->mii_bus.irq[i] = PHY_POLL; @@ -709,11 +709,11 @@ static struct net_device * au1000_probe(int port_num) /* if known, set corresponding PHY IRQs */ #if defined(AU1XXX_PHY_STATIC_CONFIG) # if defined(AU1XXX_PHY0_IRQ) - if (AU1XXX_PHY0_BUSID == aup->mii_bus.id) + if (AU1XXX_PHY0_BUSID == aup->mac_id) aup->mii_bus.irq[AU1XXX_PHY0_ADDR] = AU1XXX_PHY0_IRQ; # endif # if defined(AU1XXX_PHY1_IRQ) - if (AU1XXX_PHY1_BUSID == aup->mii_bus.id) + if (AU1XXX_PHY1_BUSID == aup->mac_id) aup->mii_bus.irq[AU1XXX_PHY1_ADDR] = AU1XXX_PHY1_IRQ; # endif #endif diff --git a/drivers/net/bfin_mac.c b/drivers/net/bfin_mac.c index 26b2dd5016cd..717dcc1aa1e9 100644 --- a/drivers/net/bfin_mac.c +++ b/drivers/net/bfin_mac.c @@ -969,7 +969,7 @@ static int __init bf537mac_probe(struct net_device *dev) lp->mii_bus.write = mdiobus_write; lp->mii_bus.reset = mdiobus_reset; lp->mii_bus.name = "bfin_mac_mdio"; - lp->mii_bus.id = 0; + snprintf(lp->mii_bus.id, MII_BUS_ID_SIZE, "0"); lp->mii_bus.irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL); for (i = 0; i < PHY_MAX_ADDR; ++i) lp->mii_bus.irq[i] = PHY_POLL; diff --git a/drivers/net/cpmac.c b/drivers/net/cpmac.c index c85194f2cd2d..9da7ff437031 100644 --- a/drivers/net/cpmac.c +++ b/drivers/net/cpmac.c @@ -987,7 +987,7 @@ static int external_switch; static int __devinit cpmac_probe(struct platform_device *pdev) { int rc, phy_id, i; - int mdio_bus_id = cpmac_mii.id; + char *mdio_bus_id = "0"; struct resource *mem; struct cpmac_priv *priv; struct net_device *dev; @@ -1008,8 +1008,6 @@ static int __devinit cpmac_probe(struct platform_device *pdev) if (external_switch || dumb_switch) { struct fixed_phy_status status = {}; - mdio_bus_id = 0; - /* * FIXME: this should be in the platform code! * Since there is not platform code at all (that is, @@ -1143,6 +1141,7 @@ int __devinit cpmac_init(void) } cpmac_mii.phy_mask = ~(mask | 0x80000000); + snprintf(cpmac_mii.id, MII_BUS_ID_SIZE, "0"); res = mdiobus_register(&cpmac_mii); if (res) diff --git a/drivers/net/fec_mpc52xx.c b/drivers/net/fec_mpc52xx.c index 58b71e60204e..43b5f30743c2 100644 --- a/drivers/net/fec_mpc52xx.c +++ b/drivers/net/fec_mpc52xx.c @@ -198,7 +198,7 @@ static int mpc52xx_fec_init_phy(struct net_device *dev) struct phy_device *phydev; char phy_id[BUS_ID_SIZE]; - snprintf(phy_id, BUS_ID_SIZE, PHY_ID_FMT, + snprintf(phy_id, BUS_ID_SIZE, "%x:%02x", (unsigned int)dev->base_addr, priv->phy_addr); priv->link = PHY_DOWN; diff --git a/drivers/net/fec_mpc52xx_phy.c b/drivers/net/fec_mpc52xx_phy.c index 6a3ac4ea97e9..956836fc5ec0 100644 --- a/drivers/net/fec_mpc52xx_phy.c +++ b/drivers/net/fec_mpc52xx_phy.c @@ -124,7 +124,7 @@ static int mpc52xx_fec_mdio_probe(struct of_device *of, const struct of_device_i goto out_free; } - bus->id = res.start; + snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start); bus->priv = priv; bus->dev = dev; diff --git a/drivers/net/fs_enet/fs_enet-main.c b/drivers/net/fs_enet/fs_enet-main.c index 940e2041ba38..67b4b0728fce 100644 --- a/drivers/net/fs_enet/fs_enet-main.c +++ b/drivers/net/fs_enet/fs_enet-main.c @@ -1178,7 +1178,7 @@ static int __devinit find_phy(struct device_node *np, data = of_get_property(np, "fixed-link", NULL); if (data) { - snprintf(fpi->bus_id, 16, PHY_ID_FMT, 0, *data); + snprintf(fpi->bus_id, 16, "%x:%02x", 0, *data); return 0; } @@ -1202,7 +1202,7 @@ static int __devinit find_phy(struct device_node *np, if (!data || len != 4) goto out_put_mdio; - snprintf(fpi->bus_id, 16, PHY_ID_FMT, res.start, *data); + snprintf(fpi->bus_id, 16, "%x:%02x", res.start, *data); out_put_mdio: of_node_put(mdionode); diff --git a/drivers/net/fs_enet/mii-bitbang.c b/drivers/net/fs_enet/mii-bitbang.c index b8e4a736a130..1620030cd33c 100644 --- a/drivers/net/fs_enet/mii-bitbang.c +++ b/drivers/net/fs_enet/mii-bitbang.c @@ -130,7 +130,7 @@ static int __devinit fs_mii_bitbang_init(struct mii_bus *bus, * we get is an int, and the odds of multiple bitbang mdio buses * is low enough that it's not worth going too crazy. */ - bus->id = res.start; + snprintf(bus->id, MII_BUS_ID_SIZE, "%x", res.start); data = of_get_property(np, "fsl,mdio-pin", &len); if (!data || len != 4) @@ -307,7 +307,7 @@ static int __devinit fs_enet_mdio_probe(struct device *dev) return -ENOMEM; new_bus->name = "BB MII Bus", - new_bus->id = pdev->id; + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id); new_bus->phy_mask = ~0x9; pdata = (struct fs_mii_bb_platform_info *)pdev->dev.platform_data; diff --git a/drivers/net/fs_enet/mii-fec.c b/drivers/net/fs_enet/mii-fec.c index a89cf15090b8..ba75efc9f5b5 100644 --- a/drivers/net/fs_enet/mii-fec.c +++ b/drivers/net/fs_enet/mii-fec.c @@ -196,7 +196,7 @@ static int __devinit fs_enet_mdio_probe(struct of_device *ofdev, if (ret) return ret; - new_bus->id = res.start; + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start); fec->fecp = ioremap(res.start, res.end - res.start + 1); if (!fec->fecp) @@ -309,7 +309,7 @@ static int __devinit fs_enet_fec_mdio_probe(struct device *dev) new_bus->read = &fs_enet_fec_mii_read, new_bus->write = &fs_enet_fec_mii_write, new_bus->reset = &fs_enet_fec_mii_reset, - new_bus->id = pdev->id; + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id); pdata = (struct fs_mii_fec_platform_info *)pdev->dev.platform_data; diff --git a/drivers/net/gianfar_mii.c b/drivers/net/gianfar_mii.c index 24327629bf03..b8898927236a 100644 --- a/drivers/net/gianfar_mii.c +++ b/drivers/net/gianfar_mii.c @@ -173,7 +173,7 @@ int gfar_mdio_probe(struct device *dev) new_bus->read = &gfar_mdio_read, new_bus->write = &gfar_mdio_write, new_bus->reset = &gfar_mdio_reset, - new_bus->id = pdev->id; + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", pdev->id); pdata = (struct gianfar_mdio_data *)pdev->dev.platform_data; diff --git a/drivers/net/macb.c b/drivers/net/macb.c index 489c7c3b90d9..d513bb8a4902 100644 --- a/drivers/net/macb.c +++ b/drivers/net/macb.c @@ -246,7 +246,7 @@ static int macb_mii_init(struct macb *bp) bp->mii_bus.read = &macb_mdio_read; bp->mii_bus.write = &macb_mdio_write; bp->mii_bus.reset = &macb_mdio_reset; - bp->mii_bus.id = bp->pdev->id; + snprintf(bp->mii_bus.id, MII_BUS_ID_SIZE, "%x", bp->pdev->id); bp->mii_bus.priv = bp; bp->mii_bus.dev = &bp->dev->dev; pdata = bp->pdev->dev.platform_data; diff --git a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c index 2e39e0285d8f..bcd7f9814ed8 100644 --- a/drivers/net/pasemi_mac.c +++ b/drivers/net/pasemi_mac.c @@ -1012,7 +1012,7 @@ static int pasemi_mac_phy_init(struct net_device *dev) goto err; phy_id = *prop; - snprintf(mac->phy_id, BUS_ID_SIZE, PHY_ID_FMT, (int)r.start, phy_id); + snprintf(mac->phy_id, BUS_ID_SIZE, "%x:%02x", (int)r.start, phy_id); of_node_put(phy_dn); diff --git a/drivers/net/phy/fixed.c b/drivers/net/phy/fixed.c index ca9b040f9ad9..4e07956a483b 100644 --- a/drivers/net/phy/fixed.c +++ b/drivers/net/phy/fixed.c @@ -213,7 +213,7 @@ static int __init fixed_mdio_bus_init(void) goto err_pdev; } - fmb->mii_bus.id = 0; + snprintf(fmb->mii_bus.id, MII_BUS_ID_SIZE, "0"); fmb->mii_bus.name = "Fixed MDIO Bus"; fmb->mii_bus.dev = &pdev->dev; fmb->mii_bus.read = &fixed_mdio_read; diff --git a/drivers/net/sb1250-mac.c b/drivers/net/sb1250-mac.c index 7b53d658e337..888b7dec9866 100644 --- a/drivers/net/sb1250-mac.c +++ b/drivers/net/sb1250-mac.c @@ -2374,7 +2374,7 @@ static int sbmac_init(struct platform_device *pldev, long long base) dev->name, base, print_mac(mac, eaddr)); sc->mii_bus.name = sbmac_mdio_string; - sc->mii_bus.id = idx; + snprintf(sc->mii_bus.id, MII_BUS_ID_SIZE, "%x", idx); sc->mii_bus.priv = sc; sc->mii_bus.read = sbmac_mii_read; sc->mii_bus.write = sbmac_mii_write; diff --git a/drivers/net/ucc_geth.c b/drivers/net/ucc_geth.c index 0ee4c168e4c0..29a4d650e8a8 100644 --- a/drivers/net/ucc_geth.c +++ b/drivers/net/ucc_geth.c @@ -3954,7 +3954,7 @@ static int ucc_geth_probe(struct of_device* ofdev, const struct of_device_id *ma if (err) return -1; - ug_info->mdio_bus = res.start; + snprintf(ug_info->mdio_bus, MII_BUS_ID_SIZE, "%x", res.start); } /* get the phy interface type, or default to MII */ diff --git a/drivers/net/ucc_geth.h b/drivers/net/ucc_geth.h index 4fb95b3af948..9f8b7580a3a4 100644 --- a/drivers/net/ucc_geth.h +++ b/drivers/net/ucc_geth.h @@ -1156,7 +1156,7 @@ struct ucc_geth_info { u16 pausePeriod; u16 extensionField; u8 phy_address; - u32 mdio_bus; + char mdio_bus[MII_BUS_ID_SIZE]; u8 weightfactor[NUM_TX_QUEUES]; u8 interruptcoalescingmaxvalue[NUM_RX_QUEUES]; u8 l2qt[UCC_GETH_VLAN_PRIORITY_MAX]; diff --git a/drivers/net/ucc_geth_mii.c b/drivers/net/ucc_geth_mii.c index c69e654d539f..e4d3f330bac3 100644 --- a/drivers/net/ucc_geth_mii.c +++ b/drivers/net/ucc_geth_mii.c @@ -157,7 +157,7 @@ static int uec_mdio_probe(struct of_device *ofdev, const struct of_device_id *ma if (err) goto reg_map_fail; - new_bus->id = res.start; + snprintf(new_bus->id, MII_BUS_ID_SIZE, "%x", res.start); new_bus->irq = kmalloc(32 * sizeof(int), GFP_KERNEL); diff --git a/include/linux/fsl_devices.h b/include/linux/fsl_devices.h index 1831b196c70a..2cad5c67397e 100644 --- a/include/linux/fsl_devices.h +++ b/include/linux/fsl_devices.h @@ -50,7 +50,7 @@ struct gianfar_platform_data { u32 device_flags; /* board specific information */ u32 board_flags; - u32 bus_id; + char bus_id[MII_BUS_ID_SIZE]; u32 phy_id; u8 mac_addr[6]; phy_interface_t interface; diff --git a/include/linux/phy.h b/include/linux/phy.h index 5e43ae751412..6509f377bb10 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -63,8 +63,6 @@ typedef enum { PHY_INTERFACE_MODE_RTBI } phy_interface_t; -#define MII_BUS_MAX 4 - #define PHY_INIT_TIMEOUT 100000 #define PHY_STATE_TIME 1 @@ -74,13 +72,19 @@ typedef enum { #define PHY_MAX_ADDR 32 /* Used when trying to connect to a specific phy (mii bus id:phy device id) */ -#define PHY_ID_FMT "%x:%02x" +#define PHY_ID_FMT "%s:%02x" + +/* + * Need to be a little smaller than phydev->dev.bus_id to leave room + * for the ":%02x" + */ +#define MII_BUS_ID_SIZE (BUS_ID_SIZE - 3) /* The Bus class for PHYs. Devices which provide access to * PHYs should register using this structure */ struct mii_bus { const char *name; - int id; + char id[MII_BUS_ID_SIZE]; void *priv; int (*read)(struct mii_bus *bus, int phy_id, int regnum); int (*write)(struct mii_bus *bus, int phy_id, int regnum, u16 val); -- cgit v1.2.3 From c5e38a949bfa11d10f73927fbf4fe66b73bc3001 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Wed, 9 Apr 2008 19:38:27 -0500 Subject: phy: Clean up header style Multi-line comments weren't all CodingStyle compliant Signed-off-by: Andy Fleming Signed-off-by: Jeff Garzik --- include/linux/phy.h | 48 ++++++++++++++++++++++++++++++++---------------- 1 file changed, 32 insertions(+), 16 deletions(-) (limited to 'include') diff --git a/include/linux/phy.h b/include/linux/phy.h index 6509f377bb10..2d838448415c 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -39,7 +39,8 @@ SUPPORTED_1000baseT_Half | \ SUPPORTED_1000baseT_Full) -/* Set phydev->irq to PHY_POLL if interrupts are not supported, +/* + * Set phydev->irq to PHY_POLL if interrupts are not supported, * or not desired for this PHY. Set to PHY_IGNORE_INTERRUPT if * the attached driver handles the interrupt */ @@ -80,8 +81,10 @@ typedef enum { */ #define MII_BUS_ID_SIZE (BUS_ID_SIZE - 3) -/* The Bus class for PHYs. Devices which provide access to - * PHYs should register using this structure */ +/* + * The Bus class for PHYs. Devices which provide access to + * PHYs should register using this structure + */ struct mii_bus { const char *name; char id[MII_BUS_ID_SIZE]; @@ -90,8 +93,10 @@ struct mii_bus { int (*write)(struct mii_bus *bus, int phy_id, int regnum, u16 val); int (*reset)(struct mii_bus *bus); - /* A lock to ensure that only one thing can read/write - * the MDIO bus at a time */ + /* + * A lock to ensure that only one thing can read/write + * the MDIO bus at a time + */ struct mutex mdio_lock; struct device *dev; @@ -102,8 +107,10 @@ struct mii_bus { /* Phy addresses to be ignored when probing */ u32 phy_mask; - /* Pointer to an array of interrupts, each PHY's - * interrupt at the index matching its address */ + /* + * Pointer to an array of interrupts, each PHY's + * interrupt at the index matching its address + */ int *irq; }; @@ -255,7 +262,8 @@ struct phy_device { /* Bus address of the PHY (0-32) */ int addr; - /* forced speed & duplex (no autoneg) + /* + * forced speed & duplex (no autoneg) * partner speed & duplex & pause (autoneg) */ int speed; @@ -278,8 +286,10 @@ struct phy_device { int link_timeout; - /* Interrupt number for this PHY - * -1 means no interrupt */ + /* + * Interrupt number for this PHY + * -1 means no interrupt + */ int irq; /* private data pointer */ @@ -329,22 +339,28 @@ struct phy_driver { u32 features; u32 flags; - /* Called to initialize the PHY, - * including after a reset */ + /* + * Called to initialize the PHY, + * including after a reset + */ int (*config_init)(struct phy_device *phydev); - /* Called during discovery. Used to set - * up device-specific structures, if any */ + /* + * Called during discovery. Used to set + * up device-specific structures, if any + */ int (*probe)(struct phy_device *phydev); /* PHY Power Management */ int (*suspend)(struct phy_device *phydev); int (*resume)(struct phy_device *phydev); - /* Configures the advertisement and resets + /* + * Configures the advertisement and resets * autonegotiation if phydev->autoneg is on, * forces the speed to the current settings in phydev - * if phydev->autoneg is off */ + * if phydev->autoneg is off + */ int (*config_aneg)(struct phy_device *phydev); /* Determines the negotiated speed and duplex */ -- cgit v1.2.3 From cac1f3c8a80f3fc0b4489d1d3ba29214677ffab2 Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Tue, 15 Apr 2008 12:49:21 -0400 Subject: phylib: factor out get_phy_id from within get_phy_device We were already doing what amounts to a get_phy_id from within get_phy_device, and rather than duplicate this for the TBIPA probing, we might as well just factor it out and make it available instead. Signed-off-by: Paul Gortmaker Acked-by: Andy Fleming Signed-off-by: Jeff Garzik --- drivers/net/phy/phy_device.c | 38 +++++++++++++++++++++++++++++--------- include/linux/phy.h | 1 + 2 files changed, 30 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index f4c4fd85425f..8b1121b02f98 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -86,35 +86,55 @@ struct phy_device* phy_device_create(struct mii_bus *bus, int addr, int phy_id) EXPORT_SYMBOL(phy_device_create); /** - * get_phy_device - reads the specified PHY device and returns its @phy_device struct + * get_phy_id - reads the specified addr for its ID. * @bus: the target MII bus * @addr: PHY address on the MII bus + * @phy_id: where to store the ID retrieved. * * Description: Reads the ID registers of the PHY at @addr on the - * @bus, then allocates and returns the phy_device to represent it. + * @bus, stores it in @phy_id and returns zero on success. */ -struct phy_device * get_phy_device(struct mii_bus *bus, int addr) +int get_phy_id(struct mii_bus *bus, int addr, u32 *phy_id) { int phy_reg; - u32 phy_id; - struct phy_device *dev = NULL; /* Grab the bits from PHYIR1, and put them * in the upper half */ phy_reg = bus->read(bus, addr, MII_PHYSID1); if (phy_reg < 0) - return ERR_PTR(phy_reg); + return -EIO; - phy_id = (phy_reg & 0xffff) << 16; + *phy_id = (phy_reg & 0xffff) << 16; /* Grab the bits from PHYIR2, and put them in the lower half */ phy_reg = bus->read(bus, addr, MII_PHYSID2); if (phy_reg < 0) - return ERR_PTR(phy_reg); + return -EIO; + + *phy_id |= (phy_reg & 0xffff); + + return 0; +} + +/** + * get_phy_device - reads the specified PHY device and returns its @phy_device struct + * @bus: the target MII bus + * @addr: PHY address on the MII bus + * + * Description: Reads the ID registers of the PHY at @addr on the + * @bus, then allocates and returns the phy_device to represent it. + */ +struct phy_device * get_phy_device(struct mii_bus *bus, int addr) +{ + struct phy_device *dev = NULL; + u32 phy_id; + int r; - phy_id |= (phy_reg & 0xffff); + r = get_phy_id(bus, addr, &phy_id); + if (r) + return ERR_PTR(r); /* If the phy_id is all Fs, there is no device there */ if (0xffffffff == phy_id) diff --git a/include/linux/phy.h b/include/linux/phy.h index 2d838448415c..779cbcd65f62 100644 --- a/include/linux/phy.h +++ b/include/linux/phy.h @@ -381,6 +381,7 @@ struct phy_driver { int phy_read(struct phy_device *phydev, u16 regnum); int phy_write(struct phy_device *phydev, u16 regnum, u16 val); +int get_phy_id(struct mii_bus *bus, int addr, u32 *phy_id); struct phy_device* get_phy_device(struct mii_bus *bus, int addr); int phy_clear_interrupt(struct phy_device *phydev); int phy_config_interrupt(struct phy_device *phydev, u32 interrupts); -- cgit v1.2.3 From e56d8b8a2ee5fb7f63ceba58e1c0fb3c844888a4 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Thu, 17 Apr 2008 23:17:34 -0700 Subject: [INET]: Drop the inet_inherit_port() call. As I can see from the code, two places (tcp_v6_syn_recv_sock and dccp_v6_request_recv_sock) that call this one already run with BHs disabled, so it's safe to call __inet_inherit_port there. Besides (in case I missed smth with code review) the calltrace tcp_v6_syn_recv_sock `- tcp_v4_syn_recv_sock `- __inet_inherit_port and the similar for DCCP are valid, but assumes BHs to be disabled. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/inet_hashtables.h | 7 ------- net/dccp/ipv6.c | 2 +- net/ipv6/tcp_ipv6.c | 2 +- 3 files changed, 2 insertions(+), 9 deletions(-) (limited to 'include') diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 5525227c5e92..5ec91d88b517 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -235,13 +235,6 @@ static inline void __inet_inherit_port(struct sock *sk, struct sock *child) spin_unlock(&head->lock); } -static inline void inet_inherit_port(struct sock *sk, struct sock *child) -{ - local_bh_disable(); - __inet_inherit_port(sk, child); - local_bh_enable(); -} - extern void inet_put_port(struct sock *sk); extern void inet_listen_wlock(struct inet_hashinfo *hashinfo); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index f8371c7e5e80..9b1129bb7ece 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -625,7 +625,7 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk, newinet->daddr = newinet->saddr = newinet->rcv_saddr = LOOPBACK4_IPV6; __inet6_hash(newsk); - inet_inherit_port(sk, newsk); + __inet_inherit_port(sk, newsk); return newsk; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 231c4dddfb8c..715965f0fac0 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -1534,7 +1534,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, #endif __inet6_hash(newsk); - inet_inherit_port(sk, newsk); + __inet_inherit_port(sk, newsk); return newsk; -- cgit v1.2.3 From 53083773dcbd3c80477e2ace143e361e1e806745 Mon Sep 17 00:00:00 2001 From: Pavel Emelyanov Date: Thu, 17 Apr 2008 23:18:15 -0700 Subject: [INET]: Uninline the __inet_inherit_port call. This deblats ~200 bytes when ipv6 and dccp are 'y'. Besides, this will ease compilation issues for patches I'm working on to make inet hash tables more scalable wrt net namespaces. Signed-off-by: Pavel Emelyanov Signed-off-by: David S. Miller --- include/net/inet_hashtables.h | 14 +------------- net/ipv4/inet_hashtables.c | 16 ++++++++++++++++ 2 files changed, 17 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 5ec91d88b517..735b926a3497 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -221,19 +221,7 @@ static inline int inet_sk_listen_hashfn(const struct sock *sk) } /* Caller must disable local BH processing. */ -static inline void __inet_inherit_port(struct sock *sk, struct sock *child) -{ - struct inet_hashinfo *table = sk->sk_prot->h.hashinfo; - const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size); - struct inet_bind_hashbucket *head = &table->bhash[bhash]; - struct inet_bind_bucket *tb; - - spin_lock(&head->lock); - tb = inet_csk(sk)->icsk_bind_hash; - sk_add_bind_node(child, &tb->owners); - inet_csk(child)->icsk_bind_hash = tb; - spin_unlock(&head->lock); -} +extern void __inet_inherit_port(struct sock *sk, struct sock *child); extern void inet_put_port(struct sock *sk); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 1612184ecc6f..2023d37b2708 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -92,6 +92,22 @@ void inet_put_port(struct sock *sk) EXPORT_SYMBOL(inet_put_port); +void __inet_inherit_port(struct sock *sk, struct sock *child) +{ + struct inet_hashinfo *table = sk->sk_prot->h.hashinfo; + const int bhash = inet_bhashfn(inet_sk(child)->num, table->bhash_size); + struct inet_bind_hashbucket *head = &table->bhash[bhash]; + struct inet_bind_bucket *tb; + + spin_lock(&head->lock); + tb = inet_csk(sk)->icsk_bind_hash; + sk_add_bind_node(child, &tb->owners); + inet_csk(child)->icsk_bind_hash = tb; + spin_unlock(&head->lock); +} + +EXPORT_SYMBOL_GPL(__inet_inherit_port); + /* * This lock without WQ_FLAG_EXCLUSIVE is good on UP and it can be very bad on SMP. * Look, when several writers sleep and reader wakes them up, all but one -- cgit v1.2.3