From 4a2f965ca5a4e2593744bf75425d85e0e8ff814a Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 18 Feb 2009 16:29:44 +0100 Subject: netfilter: x_tables: change elements in x_tables Change to proper type on private pointer rather than anonymous void. Keep active elements on same cache line. Signed-off-by: Stephen Hemminger Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'include/linux/netfilter') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index c7ee8744d26b..9fac88fc0e72 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -349,9 +349,6 @@ struct xt_table { struct list_head list; - /* A unique name... */ - const char name[XT_TABLE_MAXNAMELEN]; - /* What hooks you will enter on */ unsigned int valid_hooks; @@ -359,13 +356,15 @@ struct xt_table rwlock_t lock; /* Man behind the curtain... */ - //struct ip6t_table_info *private; - void *private; + struct xt_table_info *private; /* Set this to THIS_MODULE if you are a module, otherwise NULL */ struct module *me; u_int8_t af; /* address/protocol family */ + + /* A unique name... */ + const char name[XT_TABLE_MAXNAMELEN]; }; #include -- cgit v1.2.3 From 784544739a25c30637397ace5489eeb6e15d7d49 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Fri, 20 Feb 2009 10:35:32 +0100 Subject: netfilter: iptables: lock free counters The reader/writer lock in ip_tables is acquired in the critical path of processing packets and is one of the reasons just loading iptables can cause a 20% performance loss. The rwlock serves two functions: 1) it prevents changes to table state (xt_replace) while table is in use. This is now handled by doing rcu on the xt_table. When table is replaced, the new table(s) are put in and the old one table(s) are freed after RCU period. 2) it provides synchronization when accesing the counter values. This is now handled by swapping in new table_info entries for each cpu then summing the old values, and putting the result back onto one cpu. On a busy system it may cause sampling to occur at different times on each cpu, but no packet/byte counts are lost in the process. Signed-off-by: Stephen Hemminger Sucessfully tested on my dual quad core machine too, but iptables only (no ipv6 here) BTW, my new "tbench 8" result is 2450 MB/s, (it was 2150 MB/s not so long ago) Acked-by: Eric Dumazet Signed-off-by: Patrick McHardy --- include/linux/netfilter/x_tables.h | 6 +- net/ipv4/netfilter/arp_tables.c | 115 ++++++++++++++++++++++++++--------- net/ipv4/netfilter/ip_tables.c | 120 +++++++++++++++++++++++++++---------- net/ipv6/netfilter/ip6_tables.c | 119 +++++++++++++++++++++++++----------- net/netfilter/x_tables.c | 26 ++++++-- 5 files changed, 284 insertions(+), 102 deletions(-) (limited to 'include/linux/netfilter') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index 9fac88fc0e72..e8e08d036752 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -353,7 +353,7 @@ struct xt_table unsigned int valid_hooks; /* Lock for the curtain */ - rwlock_t lock; + struct mutex lock; /* Man behind the curtain... */ struct xt_table_info *private; @@ -385,7 +385,7 @@ struct xt_table_info /* ipt_entry tables: one per CPU */ /* Note : this field MUST be the last one, see XT_TABLE_INFO_SZ */ - char *entries[1]; + void *entries[1]; }; #define XT_TABLE_INFO_SZ (offsetof(struct xt_table_info, entries) \ @@ -432,6 +432,8 @@ extern void xt_proto_fini(struct net *net, u_int8_t af); extern struct xt_table_info *xt_alloc_table_info(unsigned int size); extern void xt_free_table_info(struct xt_table_info *info); +extern void xt_table_entry_swap_rcu(struct xt_table_info *old, + struct xt_table_info *new); #ifdef CONFIG_COMPAT #include diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index b5db46342614..64a7c6ce0b98 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -261,9 +261,10 @@ unsigned int arpt_do_table(struct sk_buff *skb, indev = in ? in->name : nulldevname; outdev = out ? out->name : nulldevname; - read_lock_bh(&table->lock); - private = table->private; - table_base = (void *)private->entries[smp_processor_id()]; + rcu_read_lock(); + private = rcu_dereference(table->private); + table_base = rcu_dereference(private->entries[smp_processor_id()]); + e = get_entry(table_base, private->hook_entry[hook]); back = get_entry(table_base, private->underflow[hook]); @@ -335,7 +336,8 @@ unsigned int arpt_do_table(struct sk_buff *skb, e = (void *)e + e->next_offset; } } while (!hotdrop); - read_unlock_bh(&table->lock); + + rcu_read_unlock(); if (hotdrop) return NF_DROP; @@ -738,11 +740,65 @@ static void get_counters(const struct xt_table_info *t, } } -static inline struct xt_counters *alloc_counters(struct xt_table *table) + +/* We're lazy, and add to the first CPU; overflow works its fey magic + * and everything is OK. */ +static int +add_counter_to_entry(struct arpt_entry *e, + const struct xt_counters addme[], + unsigned int *i) +{ + ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt); + + (*i)++; + return 0; +} + +/* Take values from counters and add them back onto the current cpu */ +static void put_counters(struct xt_table_info *t, + const struct xt_counters counters[]) +{ + unsigned int i, cpu; + + local_bh_disable(); + cpu = smp_processor_id(); + i = 0; + ARPT_ENTRY_ITERATE(t->entries[cpu], + t->size, + add_counter_to_entry, + counters, + &i); + local_bh_enable(); +} + +static inline int +zero_entry_counter(struct arpt_entry *e, void *arg) +{ + e->counters.bcnt = 0; + e->counters.pcnt = 0; + return 0; +} + +static void +clone_counters(struct xt_table_info *newinfo, const struct xt_table_info *info) +{ + unsigned int cpu; + const void *loc_cpu_entry = info->entries[raw_smp_processor_id()]; + + memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); + for_each_possible_cpu(cpu) { + memcpy(newinfo->entries[cpu], loc_cpu_entry, info->size); + ARPT_ENTRY_ITERATE(newinfo->entries[cpu], newinfo->size, + zero_entry_counter, NULL); + } +} + +static struct xt_counters *alloc_counters(struct xt_table *table) { unsigned int countersize; struct xt_counters *counters; - const struct xt_table_info *private = table->private; + struct xt_table_info *private = table->private; + struct xt_table_info *info; /* We need atomic snapshot of counters: rest doesn't change * (other than comefrom, which userspace doesn't care @@ -752,14 +808,30 @@ static inline struct xt_counters *alloc_counters(struct xt_table *table) counters = vmalloc_node(countersize, numa_node_id()); if (counters == NULL) - return ERR_PTR(-ENOMEM); + goto nomem; + + info = xt_alloc_table_info(private->size); + if (!info) + goto free_counters; - /* First, sum counters... */ - write_lock_bh(&table->lock); - get_counters(private, counters); - write_unlock_bh(&table->lock); + clone_counters(info, private); + + mutex_lock(&table->lock); + xt_table_entry_swap_rcu(private, info); + synchronize_net(); /* Wait until smoke has cleared */ + + get_counters(info, counters); + put_counters(private, counters); + mutex_unlock(&table->lock); + + xt_free_table_info(info); return counters; + + free_counters: + vfree(counters); + nomem: + return ERR_PTR(-ENOMEM); } static int copy_entries_to_user(unsigned int total_size, @@ -1099,20 +1171,6 @@ static int do_replace(struct net *net, void __user *user, unsigned int len) return ret; } -/* We're lazy, and add to the first CPU; overflow works its fey magic - * and everything is OK. - */ -static inline int add_counter_to_entry(struct arpt_entry *e, - const struct xt_counters addme[], - unsigned int *i) -{ - - ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt); - - (*i)++; - return 0; -} - static int do_add_counters(struct net *net, void __user *user, unsigned int len, int compat) { @@ -1172,13 +1230,14 @@ static int do_add_counters(struct net *net, void __user *user, unsigned int len, goto free; } - write_lock_bh(&t->lock); + mutex_lock(&t->lock); private = t->private; if (private->number != num_counters) { ret = -EINVAL; goto unlock_up_free; } + preempt_disable(); i = 0; /* Choose the copy that is on our node */ loc_cpu_entry = private->entries[smp_processor_id()]; @@ -1187,8 +1246,10 @@ static int do_add_counters(struct net *net, void __user *user, unsigned int len, add_counter_to_entry, paddc, &i); + preempt_enable(); unlock_up_free: - write_unlock_bh(&t->lock); + mutex_unlock(&t->lock); + xt_table_unlock(t); module_put(t->me); free: diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index ef8b6ca068b2..08cde5bd70a5 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -347,10 +347,12 @@ ipt_do_table(struct sk_buff *skb, mtpar.family = tgpar.family = NFPROTO_IPV4; tgpar.hooknum = hook; - read_lock_bh(&table->lock); IP_NF_ASSERT(table->valid_hooks & (1 << hook)); - private = table->private; - table_base = (void *)private->entries[smp_processor_id()]; + + rcu_read_lock(); + private = rcu_dereference(table->private); + table_base = rcu_dereference(private->entries[smp_processor_id()]); + e = get_entry(table_base, private->hook_entry[hook]); /* For return from builtin chain */ @@ -445,7 +447,7 @@ ipt_do_table(struct sk_buff *skb, } } while (!hotdrop); - read_unlock_bh(&table->lock); + rcu_read_unlock(); #ifdef DEBUG_ALLOW_ALL return NF_ACCEPT; @@ -924,13 +926,68 @@ get_counters(const struct xt_table_info *t, counters, &i); } + +} + +/* We're lazy, and add to the first CPU; overflow works its fey magic + * and everything is OK. */ +static int +add_counter_to_entry(struct ipt_entry *e, + const struct xt_counters addme[], + unsigned int *i) +{ + ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt); + + (*i)++; + return 0; +} + +/* Take values from counters and add them back onto the current cpu */ +static void put_counters(struct xt_table_info *t, + const struct xt_counters counters[]) +{ + unsigned int i, cpu; + + local_bh_disable(); + cpu = smp_processor_id(); + i = 0; + IPT_ENTRY_ITERATE(t->entries[cpu], + t->size, + add_counter_to_entry, + counters, + &i); + local_bh_enable(); +} + + +static inline int +zero_entry_counter(struct ipt_entry *e, void *arg) +{ + e->counters.bcnt = 0; + e->counters.pcnt = 0; + return 0; +} + +static void +clone_counters(struct xt_table_info *newinfo, const struct xt_table_info *info) +{ + unsigned int cpu; + const void *loc_cpu_entry = info->entries[raw_smp_processor_id()]; + + memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); + for_each_possible_cpu(cpu) { + memcpy(newinfo->entries[cpu], loc_cpu_entry, info->size); + IPT_ENTRY_ITERATE(newinfo->entries[cpu], newinfo->size, + zero_entry_counter, NULL); + } } static struct xt_counters * alloc_counters(struct xt_table *table) { unsigned int countersize; struct xt_counters *counters; - const struct xt_table_info *private = table->private; + struct xt_table_info *private = table->private; + struct xt_table_info *info; /* We need atomic snapshot of counters: rest doesn't change (other than comefrom, which userspace doesn't care @@ -939,14 +996,30 @@ static struct xt_counters * alloc_counters(struct xt_table *table) counters = vmalloc_node(countersize, numa_node_id()); if (counters == NULL) - return ERR_PTR(-ENOMEM); + goto nomem; - /* First, sum counters... */ - write_lock_bh(&table->lock); - get_counters(private, counters); - write_unlock_bh(&table->lock); + info = xt_alloc_table_info(private->size); + if (!info) + goto free_counters; + + clone_counters(info, private); + + mutex_lock(&table->lock); + xt_table_entry_swap_rcu(private, info); + synchronize_net(); /* Wait until smoke has cleared */ + + get_counters(info, counters); + put_counters(private, counters); + mutex_unlock(&table->lock); + + xt_free_table_info(info); return counters; + + free_counters: + vfree(counters); + nomem: + return ERR_PTR(-ENOMEM); } static int @@ -1312,27 +1385,6 @@ do_replace(struct net *net, void __user *user, unsigned int len) return ret; } -/* We're lazy, and add to the first CPU; overflow works its fey magic - * and everything is OK. */ -static int -add_counter_to_entry(struct ipt_entry *e, - const struct xt_counters addme[], - unsigned int *i) -{ -#if 0 - duprintf("add_counter: Entry %u %lu/%lu + %lu/%lu\n", - *i, - (long unsigned int)e->counters.pcnt, - (long unsigned int)e->counters.bcnt, - (long unsigned int)addme[*i].pcnt, - (long unsigned int)addme[*i].bcnt); -#endif - - ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt); - - (*i)++; - return 0; -} static int do_add_counters(struct net *net, void __user *user, unsigned int len, int compat) @@ -1393,13 +1445,14 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, int compat goto free; } - write_lock_bh(&t->lock); + mutex_lock(&t->lock); private = t->private; if (private->number != num_counters) { ret = -EINVAL; goto unlock_up_free; } + preempt_disable(); i = 0; /* Choose the copy that is on our node */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; @@ -1408,8 +1461,9 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, int compat add_counter_to_entry, paddc, &i); + preempt_enable(); unlock_up_free: - write_unlock_bh(&t->lock); + mutex_unlock(&t->lock); xt_table_unlock(t); module_put(t->me); free: diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index d64594b6c061..34af7bb8df5f 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -382,10 +382,12 @@ ip6t_do_table(struct sk_buff *skb, mtpar.family = tgpar.family = NFPROTO_IPV6; tgpar.hooknum = hook; - read_lock_bh(&table->lock); IP_NF_ASSERT(table->valid_hooks & (1 << hook)); - private = table->private; - table_base = (void *)private->entries[smp_processor_id()]; + + rcu_read_lock(); + private = rcu_dereference(table->private); + table_base = rcu_dereference(private->entries[smp_processor_id()]); + e = get_entry(table_base, private->hook_entry[hook]); /* For return from builtin chain */ @@ -483,7 +485,7 @@ ip6t_do_table(struct sk_buff *skb, #ifdef CONFIG_NETFILTER_DEBUG ((struct ip6t_entry *)table_base)->comefrom = NETFILTER_LINK_POISON; #endif - read_unlock_bh(&table->lock); + rcu_read_unlock(); #ifdef DEBUG_ALLOW_ALL return NF_ACCEPT; @@ -964,11 +966,64 @@ get_counters(const struct xt_table_info *t, } } +/* We're lazy, and add to the first CPU; overflow works its fey magic + * and everything is OK. */ +static int +add_counter_to_entry(struct ip6t_entry *e, + const struct xt_counters addme[], + unsigned int *i) +{ + ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt); + + (*i)++; + return 0; +} + +/* Take values from counters and add them back onto the current cpu */ +static void put_counters(struct xt_table_info *t, + const struct xt_counters counters[]) +{ + unsigned int i, cpu; + + local_bh_disable(); + cpu = smp_processor_id(); + i = 0; + IP6T_ENTRY_ITERATE(t->entries[cpu], + t->size, + add_counter_to_entry, + counters, + &i); + local_bh_enable(); +} + +static inline int +zero_entry_counter(struct ip6t_entry *e, void *arg) +{ + e->counters.bcnt = 0; + e->counters.pcnt = 0; + return 0; +} + +static void +clone_counters(struct xt_table_info *newinfo, const struct xt_table_info *info) +{ + unsigned int cpu; + const void *loc_cpu_entry = info->entries[raw_smp_processor_id()]; + + memcpy(newinfo, info, offsetof(struct xt_table_info, entries)); + for_each_possible_cpu(cpu) { + memcpy(newinfo->entries[cpu], loc_cpu_entry, info->size); + IP6T_ENTRY_ITERATE(newinfo->entries[cpu], newinfo->size, + zero_entry_counter, NULL); + } +} + static struct xt_counters *alloc_counters(struct xt_table *table) { unsigned int countersize; struct xt_counters *counters; - const struct xt_table_info *private = table->private; + struct xt_table_info *private = table->private; + struct xt_table_info *info; /* We need atomic snapshot of counters: rest doesn't change (other than comefrom, which userspace doesn't care @@ -977,14 +1032,28 @@ static struct xt_counters *alloc_counters(struct xt_table *table) counters = vmalloc_node(countersize, numa_node_id()); if (counters == NULL) - return ERR_PTR(-ENOMEM); + goto nomem; + + info = xt_alloc_table_info(private->size); + if (!info) + goto free_counters; + + clone_counters(info, private); + + mutex_lock(&table->lock); + xt_table_entry_swap_rcu(private, info); + synchronize_net(); /* Wait until smoke has cleared */ + + get_counters(info, counters); + put_counters(private, counters); + mutex_unlock(&table->lock); - /* First, sum counters... */ - write_lock_bh(&table->lock); - get_counters(private, counters); - write_unlock_bh(&table->lock); + xt_free_table_info(info); - return counters; + free_counters: + vfree(counters); + nomem: + return ERR_PTR(-ENOMEM); } static int @@ -1351,28 +1420,6 @@ do_replace(struct net *net, void __user *user, unsigned int len) return ret; } -/* We're lazy, and add to the first CPU; overflow works its fey magic - * and everything is OK. */ -static inline int -add_counter_to_entry(struct ip6t_entry *e, - const struct xt_counters addme[], - unsigned int *i) -{ -#if 0 - duprintf("add_counter: Entry %u %lu/%lu + %lu/%lu\n", - *i, - (long unsigned int)e->counters.pcnt, - (long unsigned int)e->counters.bcnt, - (long unsigned int)addme[*i].pcnt, - (long unsigned int)addme[*i].bcnt); -#endif - - ADD_COUNTER(e->counters, addme[*i].bcnt, addme[*i].pcnt); - - (*i)++; - return 0; -} - static int do_add_counters(struct net *net, void __user *user, unsigned int len, int compat) @@ -1433,13 +1480,14 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, goto free; } - write_lock_bh(&t->lock); + mutex_lock(&t->lock); private = t->private; if (private->number != num_counters) { ret = -EINVAL; goto unlock_up_free; } + preempt_disable(); i = 0; /* Choose the copy that is on our node */ loc_cpu_entry = private->entries[raw_smp_processor_id()]; @@ -1448,8 +1496,9 @@ do_add_counters(struct net *net, void __user *user, unsigned int len, add_counter_to_entry, paddc, &i); + preempt_enable(); unlock_up_free: - write_unlock_bh(&t->lock); + mutex_unlock(&t->lock); xt_table_unlock(t); module_put(t->me); free: diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index bfbf521f6ea5..bfcac92d5563 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -625,6 +625,20 @@ void xt_free_table_info(struct xt_table_info *info) } EXPORT_SYMBOL(xt_free_table_info); +void xt_table_entry_swap_rcu(struct xt_table_info *oldinfo, + struct xt_table_info *newinfo) +{ + unsigned int cpu; + + for_each_possible_cpu(cpu) { + void *p = oldinfo->entries[cpu]; + rcu_assign_pointer(oldinfo->entries[cpu], newinfo->entries[cpu]); + newinfo->entries[cpu] = p; + } + +} +EXPORT_SYMBOL_GPL(xt_table_entry_swap_rcu); + /* Find table by name, grabs mutex & ref. Returns ERR_PTR() on error. */ struct xt_table *xt_find_table_lock(struct net *net, u_int8_t af, const char *name) @@ -671,21 +685,22 @@ xt_replace_table(struct xt_table *table, struct xt_table_info *oldinfo, *private; /* Do the substitution. */ - write_lock_bh(&table->lock); + mutex_lock(&table->lock); private = table->private; /* Check inside lock: is the old number correct? */ if (num_counters != private->number) { duprintf("num_counters != table->private->number (%u/%u)\n", num_counters, private->number); - write_unlock_bh(&table->lock); + mutex_unlock(&table->lock); *error = -EAGAIN; return NULL; } oldinfo = private; - table->private = newinfo; + rcu_assign_pointer(table->private, newinfo); newinfo->initial_entries = oldinfo->initial_entries; - write_unlock_bh(&table->lock); + mutex_unlock(&table->lock); + synchronize_net(); return oldinfo; } EXPORT_SYMBOL_GPL(xt_replace_table); @@ -719,7 +734,8 @@ struct xt_table *xt_register_table(struct net *net, struct xt_table *table, /* Simplifies replace_table code. */ table->private = bootstrap; - rwlock_init(&table->lock); + mutex_init(&table->lock); + if (!xt_replace_table(table, 0, newinfo, &ret)) goto unlock; -- cgit v1.2.3 From 268cb38e1802db560c73167e643f14a3dcb4b07c Mon Sep 17 00:00:00 2001 From: Adam Nielsen Date: Fri, 20 Feb 2009 10:55:14 +0100 Subject: netfilter: x_tables: add LED trigger target Kernel module providing implementation of LED netfilter target. Each instance of the target appears as a led-trigger device, which can be associated with one or more LEDs in /sys/class/leds/ Signed-off-by: Adam Nielsen Acked-by: Richard Purdie Signed-off-by: Patrick McHardy --- drivers/leds/Kconfig | 3 + include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_LED.h | 13 ++++ net/netfilter/Kconfig | 24 ++++++ net/netfilter/Makefile | 1 + net/netfilter/xt_LED.c | 161 +++++++++++++++++++++++++++++++++++++++ 6 files changed, 203 insertions(+) create mode 100644 include/linux/netfilter/xt_LED.h create mode 100644 net/netfilter/xt_LED.c (limited to 'include/linux/netfilter') diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig index 742713611bc5..556aeca0d860 100644 --- a/drivers/leds/Kconfig +++ b/drivers/leds/Kconfig @@ -223,4 +223,7 @@ config LEDS_TRIGGER_DEFAULT_ON This allows LEDs to be initialised in the ON state. If unsure, say Y. +comment "iptables trigger is under Netfilter config (LED target)" + depends on LEDS_TRIGGERS + endif # NEW_LEDS diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index 5a8af875bce2..deeaee5c83f2 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -7,6 +7,7 @@ header-y += xt_CLASSIFY.h header-y += xt_CONNMARK.h header-y += xt_CONNSECMARK.h header-y += xt_DSCP.h +header-y += xt_LED.h header-y += xt_MARK.h header-y += xt_NFLOG.h header-y += xt_NFQUEUE.h diff --git a/include/linux/netfilter/xt_LED.h b/include/linux/netfilter/xt_LED.h new file mode 100644 index 000000000000..4c91a0d770d0 --- /dev/null +++ b/include/linux/netfilter/xt_LED.h @@ -0,0 +1,13 @@ +#ifndef _XT_LED_H +#define _XT_LED_H + +struct xt_led_info { + char id[27]; /* Unique ID for this trigger in the LED class */ + __u8 always_blink; /* Blink even if the LED is already on */ + __u32 delay; /* Delay until LED is switched off after trigger */ + + /* Kernel data used in the module */ + void *internal_data __attribute__((aligned(8))); +}; + +#endif /* _XT_LED_H */ diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 0eb98b4fbf44..cdbaaff6d0d6 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -372,6 +372,30 @@ config NETFILTER_XT_TARGET_HL since you can easily create immortal packets that loop forever on the network. +config NETFILTER_XT_TARGET_LED + tristate '"LED" target support' + depends on LEDS_CLASS + depends on NETFILTER_ADVANCED + help + This option adds a `LED' target, which allows you to blink LEDs in + response to particular packets passing through your machine. + + This can be used to turn a spare LED into a network activity LED, + which only flashes in response to FTP transfers, for example. Or + you could have an LED which lights up for a minute or two every time + somebody connects to your machine via SSH. + + You will need support for the "led" class to make this work. + + To create an LED trigger for incoming SSH traffic: + iptables -A INPUT -p tcp --dport 22 -j LED --led-trigger-id ssh --led-delay 1000 + + Then attach the new trigger to an LED on your system: + echo netfilter-ssh > /sys/class/leds//trigger + + For more information on the LEDs available on your system, see + Documentation/leds-class.txt + config NETFILTER_XT_TARGET_MARK tristate '"MARK" target support' default m if NETFILTER_ADVANCED=n diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index da73ed25701c..7a9b8397573a 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -46,6 +46,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_CONNMARK) += xt_CONNMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_CONNSECMARK) += xt_CONNSECMARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_DSCP) += xt_DSCP.o obj-$(CONFIG_NETFILTER_XT_TARGET_HL) += xt_HL.o +obj-$(CONFIG_NETFILTER_XT_TARGET_LED) += xt_LED.o obj-$(CONFIG_NETFILTER_XT_TARGET_MARK) += xt_MARK.o obj-$(CONFIG_NETFILTER_XT_TARGET_NFLOG) += xt_NFLOG.o obj-$(CONFIG_NETFILTER_XT_TARGET_NFQUEUE) += xt_NFQUEUE.o diff --git a/net/netfilter/xt_LED.c b/net/netfilter/xt_LED.c new file mode 100644 index 000000000000..8ff7843bb921 --- /dev/null +++ b/net/netfilter/xt_LED.c @@ -0,0 +1,161 @@ +/* + * xt_LED.c - netfilter target to make LEDs blink upon packet matches + * + * Copyright (C) 2008 Adam Nielsen + * + * 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; version 2 of the License. + * + * 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 Street, Fifth Floor, Boston, MA + * 02110-1301 USA. + * + */ + +#include +#include +#include +#include +#include + +#include + +MODULE_LICENSE("GPL"); +MODULE_AUTHOR("Adam Nielsen "); +MODULE_DESCRIPTION("Xtables: trigger LED devices on packet match"); + +/* + * This is declared in here (the kernel module) only, to avoid having these + * dependencies in userspace code. This is what xt_led_info.internal_data + * points to. + */ +struct xt_led_info_internal { + struct led_trigger netfilter_led_trigger; + struct timer_list timer; +}; + +static unsigned int +led_tg(struct sk_buff *skb, const struct xt_target_param *par) +{ + const struct xt_led_info *ledinfo = par->targinfo; + struct xt_led_info_internal *ledinternal = ledinfo->internal_data; + + /* + * If "always blink" is enabled, and there's still some time until the + * LED will switch off, briefly switch it off now. + */ + if ((ledinfo->delay > 0) && ledinfo->always_blink && + timer_pending(&ledinternal->timer)) + led_trigger_event(&ledinternal->netfilter_led_trigger,LED_OFF); + + led_trigger_event(&ledinternal->netfilter_led_trigger, LED_FULL); + + /* If there's a positive delay, start/update the timer */ + if (ledinfo->delay > 0) { + mod_timer(&ledinternal->timer, + jiffies + msecs_to_jiffies(ledinfo->delay)); + + /* Otherwise if there was no delay given, blink as fast as possible */ + } else if (ledinfo->delay == 0) { + led_trigger_event(&ledinternal->netfilter_led_trigger, LED_OFF); + } + + /* else the delay is negative, which means switch on and stay on */ + + return XT_CONTINUE; +} + +static void led_timeout_callback(unsigned long data) +{ + struct xt_led_info *ledinfo = (struct xt_led_info *)data; + struct xt_led_info_internal *ledinternal = ledinfo->internal_data; + + led_trigger_event(&ledinternal->netfilter_led_trigger, LED_OFF); +} + +static bool led_tg_check(const struct xt_tgchk_param *par) +{ + struct xt_led_info *ledinfo = par->targinfo; + struct xt_led_info_internal *ledinternal; + int err; + + if (ledinfo->id[0] == '\0') { + printk(KERN_ERR KBUILD_MODNAME ": No 'id' parameter given.\n"); + return false; + } + + ledinternal = kzalloc(sizeof(struct xt_led_info_internal), GFP_KERNEL); + if (!ledinternal) { + printk(KERN_CRIT KBUILD_MODNAME ": out of memory\n"); + return false; + } + + ledinternal->netfilter_led_trigger.name = ledinfo->id; + + err = led_trigger_register(&ledinternal->netfilter_led_trigger); + if (err) { + printk(KERN_CRIT KBUILD_MODNAME + ": led_trigger_register() failed\n"); + if (err == -EEXIST) + printk(KERN_ERR KBUILD_MODNAME + ": Trigger name is already in use.\n"); + goto exit_alloc; + } + + /* See if we need to set up a timer */ + if (ledinfo->delay > 0) + setup_timer(&ledinternal->timer, led_timeout_callback, + (unsigned long)ledinfo); + + ledinfo->internal_data = ledinternal; + + return true; + +exit_alloc: + kfree(ledinternal); + + return false; +} + +static void led_tg_destroy(const struct xt_tgdtor_param *par) +{ + const struct xt_led_info *ledinfo = par->targinfo; + struct xt_led_info_internal *ledinternal = ledinfo->internal_data; + + if (ledinfo->delay > 0) + del_timer_sync(&ledinternal->timer); + + led_trigger_unregister(&ledinternal->netfilter_led_trigger); + kfree(ledinternal); +} + +static struct xt_target led_tg_reg __read_mostly = { + .name = "LED", + .revision = 0, + .family = NFPROTO_UNSPEC, + .target = led_tg, + .targetsize = XT_ALIGN(sizeof(struct xt_led_info)), + .checkentry = led_tg_check, + .destroy = led_tg_destroy, + .me = THIS_MODULE, +}; + +static int __init led_tg_init(void) +{ + return xt_register_target(&led_tg_reg); +} + +static void __exit led_tg_exit(void) +{ + xt_unregister_target(&led_tg_reg); +} + +module_init(led_tg_init); +module_exit(led_tg_exit); -- cgit v1.2.3 From d060ffc1840e37100628f520e66600c5ae483b44 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Tue, 24 Feb 2009 15:23:58 +0100 Subject: netfilter: install missing headers iptables imports headers from (the unifdefed headers of a) kernel tree, but some headers happened to not be installed. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 6 ++++++ include/linux/netfilter_ipv6/Kbuild | 1 + 2 files changed, 7 insertions(+) (limited to 'include/linux/netfilter') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index deeaee5c83f2..947b47d7f6c0 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -14,8 +14,11 @@ header-y += xt_NFQUEUE.h header-y += xt_RATEEST.h header-y += xt_SECMARK.h header-y += xt_TCPMSS.h +header-y += xt_TCPOPTSTRIP.h +header-y += xt_TPROXY.h header-y += xt_comment.h header-y += xt_connbytes.h +header-y += xt_connlimit.h header-y += xt_connmark.h header-y += xt_conntrack.h header-y += xt_dccp.h @@ -31,6 +34,7 @@ header-y += xt_mark.h header-y += xt_multiport.h header-y += xt_owner.h header-y += xt_pkttype.h +header-y += xt_quota.h header-y += xt_rateest.h header-y += xt_realm.h header-y += xt_recent.h @@ -40,6 +44,8 @@ header-y += xt_statistic.h header-y += xt_string.h header-y += xt_tcpmss.h header-y += xt_tcpudp.h +header-y += xt_time.h +header-y += xt_u32.h unifdef-y += nf_conntrack_common.h unifdef-y += nf_conntrack_ftp.h diff --git a/include/linux/netfilter_ipv6/Kbuild b/include/linux/netfilter_ipv6/Kbuild index 8887a5fcd1d0..aca4bd1f6d7c 100644 --- a/include/linux/netfilter_ipv6/Kbuild +++ b/include/linux/netfilter_ipv6/Kbuild @@ -11,6 +11,7 @@ header-y += ip6t_length.h header-y += ip6t_limit.h header-y += ip6t_mac.h header-y += ip6t_mark.h +header-y += ip6t_mh.h header-y += ip6t_multiport.h header-y += ip6t_opts.h header-y += ip6t_owner.h -- cgit v1.2.3 From acc738fec03bdaa5b77340c32a82fbfedaaabef0 Mon Sep 17 00:00:00 2001 From: Jan Engelhardt Date: Mon, 16 Mar 2009 15:35:29 +0100 Subject: netfilter: xtables: avoid pointer to self Commit 784544739a25c30637397ace5489eeb6e15d7d49 (netfilter: iptables: lock free counters) broke a number of modules whose rule data referenced itself. A reallocation would not reestablish the correct references, so it is best to use a separate struct that does not fall under RCU. Signed-off-by: Jan Engelhardt Signed-off-by: Patrick McHardy --- include/linux/netfilter/xt_limit.h | 9 ++++---- include/linux/netfilter/xt_quota.h | 4 +++- include/linux/netfilter/xt_statistic.h | 7 +++--- net/netfilter/xt_limit.c | 40 ++++++++++++++++++++++++---------- net/netfilter/xt_quota.c | 31 ++++++++++++++++++++------ net/netfilter/xt_statistic.c | 28 +++++++++++++++++++----- 6 files changed, 88 insertions(+), 31 deletions(-) (limited to 'include/linux/netfilter') diff --git a/include/linux/netfilter/xt_limit.h b/include/linux/netfilter/xt_limit.h index b3ce65375ecb..fda222c7953b 100644 --- a/include/linux/netfilter/xt_limit.h +++ b/include/linux/netfilter/xt_limit.h @@ -4,6 +4,8 @@ /* timings are in milliseconds. */ #define XT_LIMIT_SCALE 10000 +struct xt_limit_priv; + /* 1/10,000 sec period => max of 10,000/sec. Min rate is then 429490 seconds, or one every 59 hours. */ struct xt_rateinfo { @@ -11,11 +13,10 @@ struct xt_rateinfo { u_int32_t burst; /* Period multiplier for upper limit. */ /* Used internally by the kernel */ - unsigned long prev; - u_int32_t credit; + unsigned long prev; /* moved to xt_limit_priv */ + u_int32_t credit; /* moved to xt_limit_priv */ u_int32_t credit_cap, cost; - /* Ugly, ugly fucker. */ - struct xt_rateinfo *master; + struct xt_limit_priv *master; }; #endif /*_XT_RATE_H*/ diff --git a/include/linux/netfilter/xt_quota.h b/include/linux/netfilter/xt_quota.h index 4c8368d781e5..8dc89dfc1361 100644 --- a/include/linux/netfilter/xt_quota.h +++ b/include/linux/netfilter/xt_quota.h @@ -6,13 +6,15 @@ enum xt_quota_flags { }; #define XT_QUOTA_MASK 0x1 +struct xt_quota_priv; + struct xt_quota_info { u_int32_t flags; u_int32_t pad; /* Used internally by the kernel */ aligned_u64 quota; - struct xt_quota_info *master; + struct xt_quota_priv *master; }; #endif /* _XT_QUOTA_H */ diff --git a/include/linux/netfilter/xt_statistic.h b/include/linux/netfilter/xt_statistic.h index 3d38bc975048..8f521ab49ef7 100644 --- a/include/linux/netfilter/xt_statistic.h +++ b/include/linux/netfilter/xt_statistic.h @@ -13,6 +13,8 @@ enum xt_statistic_flags { }; #define XT_STATISTIC_MASK 0x1 +struct xt_statistic_priv; + struct xt_statistic_info { u_int16_t mode; u_int16_t flags; @@ -23,11 +25,10 @@ struct xt_statistic_info { struct { u_int32_t every; u_int32_t packet; - /* Used internally by the kernel */ - u_int32_t count; + u_int32_t count; /* unused */ } nth; } u; - struct xt_statistic_info *master __attribute__((aligned(8))); + struct xt_statistic_priv *master __attribute__((aligned(8))); }; #endif /* _XT_STATISTIC_H */ diff --git a/net/netfilter/xt_limit.c b/net/netfilter/xt_limit.c index c908d69a5595..2e8089ecd0af 100644 --- a/net/netfilter/xt_limit.c +++ b/net/netfilter/xt_limit.c @@ -14,6 +14,11 @@ #include #include +struct xt_limit_priv { + unsigned long prev; + uint32_t credit; +}; + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Herve Eychenne "); MODULE_DESCRIPTION("Xtables: rate-limit match"); @@ -60,18 +65,18 @@ static DEFINE_SPINLOCK(limit_lock); static bool limit_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - struct xt_rateinfo *r = - ((const struct xt_rateinfo *)par->matchinfo)->master; + const struct xt_rateinfo *r = par->matchinfo; + struct xt_limit_priv *priv = r->master; unsigned long now = jiffies; spin_lock_bh(&limit_lock); - r->credit += (now - xchg(&r->prev, now)) * CREDITS_PER_JIFFY; - if (r->credit > r->credit_cap) - r->credit = r->credit_cap; + priv->credit += (now - xchg(&priv->prev, now)) * CREDITS_PER_JIFFY; + if (priv->credit > r->credit_cap) + priv->credit = r->credit_cap; - if (r->credit >= r->cost) { + if (priv->credit >= r->cost) { /* We're not limited. */ - r->credit -= r->cost; + priv->credit -= r->cost; spin_unlock_bh(&limit_lock); return true; } @@ -95,6 +100,7 @@ user2credits(u_int32_t user) static bool limit_mt_check(const struct xt_mtchk_param *par) { struct xt_rateinfo *r = par->matchinfo; + struct xt_limit_priv *priv; /* Check for overflow. */ if (r->burst == 0 @@ -104,19 +110,30 @@ static bool limit_mt_check(const struct xt_mtchk_param *par) return false; } - /* For SMP, we only want to use one set of counters. */ - r->master = r; + priv = kmalloc(sizeof(*priv), GFP_KERNEL); + if (priv == NULL) + return -ENOMEM; + + /* For SMP, we only want to use one set of state. */ + r->master = priv; if (r->cost == 0) { /* User avg in seconds * XT_LIMIT_SCALE: convert to jiffies * 128. */ - r->prev = jiffies; - r->credit = user2credits(r->avg * r->burst); /* Credits full. */ + priv->prev = jiffies; + priv->credit = user2credits(r->avg * r->burst); /* Credits full. */ r->credit_cap = user2credits(r->avg * r->burst); /* Credits full. */ r->cost = user2credits(r->avg); } return true; } +static void limit_mt_destroy(const struct xt_mtdtor_param *par) +{ + const struct xt_rateinfo *info = par->matchinfo; + + kfree(info->master); +} + #ifdef CONFIG_COMPAT struct compat_xt_rateinfo { u_int32_t avg; @@ -167,6 +184,7 @@ static struct xt_match limit_mt_reg __read_mostly = { .family = NFPROTO_UNSPEC, .match = limit_mt, .checkentry = limit_mt_check, + .destroy = limit_mt_destroy, .matchsize = sizeof(struct xt_rateinfo), #ifdef CONFIG_COMPAT .compatsize = sizeof(struct compat_xt_rateinfo), diff --git a/net/netfilter/xt_quota.c b/net/netfilter/xt_quota.c index c84fce5e0f3e..01dd07b764ec 100644 --- a/net/netfilter/xt_quota.c +++ b/net/netfilter/xt_quota.c @@ -9,6 +9,10 @@ #include #include +struct xt_quota_priv { + uint64_t quota; +}; + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Sam Johnston "); MODULE_DESCRIPTION("Xtables: countdown quota match"); @@ -20,18 +24,20 @@ static DEFINE_SPINLOCK(quota_lock); static bool quota_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - struct xt_quota_info *q = - ((const struct xt_quota_info *)par->matchinfo)->master; + struct xt_quota_info *q = (void *)par->matchinfo; + struct xt_quota_priv *priv = q->master; bool ret = q->flags & XT_QUOTA_INVERT; spin_lock_bh("a_lock); - if (q->quota >= skb->len) { - q->quota -= skb->len; + if (priv->quota >= skb->len) { + priv->quota -= skb->len; ret = !ret; } else { /* we do not allow even small packets from now on */ - q->quota = 0; + priv->quota = 0; } + /* Copy quota back to matchinfo so that iptables can display it */ + q->quota = priv->quota; spin_unlock_bh("a_lock); return ret; @@ -43,17 +49,28 @@ static bool quota_mt_check(const struct xt_mtchk_param *par) if (q->flags & ~XT_QUOTA_MASK) return false; - /* For SMP, we only want to use one set of counters. */ - q->master = q; + + q->master = kmalloc(sizeof(*q->master), GFP_KERNEL); + if (q->master == NULL) + return -ENOMEM; + return true; } +static void quota_mt_destroy(const struct xt_mtdtor_param *par) +{ + const struct xt_quota_info *q = par->matchinfo; + + kfree(q->master); +} + static struct xt_match quota_mt_reg __read_mostly = { .name = "quota", .revision = 0, .family = NFPROTO_UNSPEC, .match = quota_mt, .checkentry = quota_mt_check, + .destroy = quota_mt_destroy, .matchsize = sizeof(struct xt_quota_info), .me = THIS_MODULE, }; diff --git a/net/netfilter/xt_statistic.c b/net/netfilter/xt_statistic.c index 0d75141139d5..d8c0f8f1a78e 100644 --- a/net/netfilter/xt_statistic.c +++ b/net/netfilter/xt_statistic.c @@ -16,6 +16,10 @@ #include #include +struct xt_statistic_priv { + uint32_t count; +}; + MODULE_LICENSE("GPL"); MODULE_AUTHOR("Patrick McHardy "); MODULE_DESCRIPTION("Xtables: statistics-based matching (\"Nth\", random)"); @@ -27,7 +31,7 @@ static DEFINE_SPINLOCK(nth_lock); static bool statistic_mt(const struct sk_buff *skb, const struct xt_match_param *par) { - struct xt_statistic_info *info = (void *)par->matchinfo; + const struct xt_statistic_info *info = par->matchinfo; bool ret = info->flags & XT_STATISTIC_INVERT; switch (info->mode) { @@ -36,10 +40,9 @@ statistic_mt(const struct sk_buff *skb, const struct xt_match_param *par) ret = !ret; break; case XT_STATISTIC_MODE_NTH: - info = info->master; spin_lock_bh(&nth_lock); - if (info->u.nth.count++ == info->u.nth.every) { - info->u.nth.count = 0; + if (info->master->count++ == info->u.nth.every) { + info->master->count = 0; ret = !ret; } spin_unlock_bh(&nth_lock); @@ -56,16 +59,31 @@ static bool statistic_mt_check(const struct xt_mtchk_param *par) if (info->mode > XT_STATISTIC_MODE_MAX || info->flags & ~XT_STATISTIC_MASK) return false; - info->master = info; + + info->master = kzalloc(sizeof(*info->master), GFP_KERNEL); + if (info->master == NULL) { + printk(KERN_ERR KBUILD_MODNAME ": Out of memory\n"); + return false; + } + info->master->count = info->u.nth.count; + return true; } +static void statistic_mt_destroy(const struct xt_mtdtor_param *par) +{ + const struct xt_statistic_info *info = par->matchinfo; + + kfree(info->master); +} + static struct xt_match xt_statistic_mt_reg __read_mostly = { .name = "statistic", .revision = 0, .family = NFPROTO_UNSPEC, .match = statistic_mt, .checkentry = statistic_mt_check, + .destroy = statistic_mt_destroy, .matchsize = sizeof(struct xt_statistic_info), .me = THIS_MODULE, }; -- cgit v1.2.3 From 0269ea4937343536ec7e85649932bc8c9686ea78 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 16 Mar 2009 17:10:36 +0100 Subject: netfilter: xtables: add cluster match This patch adds the iptables cluster match. This match can be used to deploy gateway and back-end load-sharing clusters. The cluster can be composed of 32 nodes maximum (although I have only tested this with two nodes, so I cannot tell what is the real scalability limit of this solution in terms of cluster nodes). Assuming that all the nodes see all packets (see below for an example on how to do that if your switch does not allow this), the cluster match decides if this node has to handle a packet given: (jhash(source IP) % total_nodes) & node_mask For related connections, the master conntrack is used. The following is an example of its use to deploy a gateway cluster composed of two nodes (where this is the node 1): iptables -I PREROUTING -t mangle -i eth1 -m cluster \ --cluster-total-nodes 2 --cluster-local-node 1 \ --cluster-proc-name eth1 -j MARK --set-mark 0xffff iptables -A PREROUTING -t mangle -i eth1 \ -m mark ! --mark 0xffff -j DROP iptables -A PREROUTING -t mangle -i eth2 -m cluster \ --cluster-total-nodes 2 --cluster-local-node 1 \ --cluster-proc-name eth2 -j MARK --set-mark 0xffff iptables -A PREROUTING -t mangle -i eth2 \ -m mark ! --mark 0xffff -j DROP And the following commands to make all nodes see the same packets: ip maddr add 01:00:5e:00:01:01 dev eth1 ip maddr add 01:00:5e:00:01:02 dev eth2 arptables -I OUTPUT -o eth1 --h-length 6 \ -j mangle --mangle-mac-s 01:00:5e:00:01:01 arptables -I INPUT -i eth1 --h-length 6 \ --destination-mac 01:00:5e:00:01:01 \ -j mangle --mangle-mac-d 00:zz:yy:xx:5a:27 arptables -I OUTPUT -o eth2 --h-length 6 \ -j mangle --mangle-mac-s 01:00:5e:00:01:02 arptables -I INPUT -i eth2 --h-length 6 \ --destination-mac 01:00:5e:00:01:02 \ -j mangle --mangle-mac-d 00:zz:yy:xx:5a:27 In the case of TCP connections, pickup facility has to be disabled to avoid marking TCP ACK packets coming in the reply direction as valid. echo 0 > /proc/sys/net/netfilter/nf_conntrack_tcp_loose BTW, some final notes: * This match mangles the skbuff pkt_type in case that it detects PACKET_MULTICAST for a non-multicast address. This may be done in a PKTTYPE target for this sole purpose. * This match supersedes the CLUSTERIP target. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- include/linux/netfilter/Kbuild | 1 + include/linux/netfilter/xt_cluster.h | 15 ++++ net/netfilter/Kconfig | 16 ++++ net/netfilter/Makefile | 1 + net/netfilter/xt_cluster.c | 164 +++++++++++++++++++++++++++++++++++ 5 files changed, 197 insertions(+) create mode 100644 include/linux/netfilter/xt_cluster.h create mode 100644 net/netfilter/xt_cluster.c (limited to 'include/linux/netfilter') diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild index 947b47d7f6c0..af9d2fb97212 100644 --- a/include/linux/netfilter/Kbuild +++ b/include/linux/netfilter/Kbuild @@ -21,6 +21,7 @@ header-y += xt_connbytes.h header-y += xt_connlimit.h header-y += xt_connmark.h header-y += xt_conntrack.h +header-y += xt_cluster.h header-y += xt_dccp.h header-y += xt_dscp.h header-y += xt_esp.h diff --git a/include/linux/netfilter/xt_cluster.h b/include/linux/netfilter/xt_cluster.h new file mode 100644 index 000000000000..5e0a0d07b526 --- /dev/null +++ b/include/linux/netfilter/xt_cluster.h @@ -0,0 +1,15 @@ +#ifndef _XT_CLUSTER_MATCH_H +#define _XT_CLUSTER_MATCH_H + +enum xt_cluster_flags { + XT_CLUSTER_F_INV = (1 << 0) +}; + +struct xt_cluster_match_info { + u_int32_t total_nodes; + u_int32_t node_mask; + u_int32_t hash_seed; + u_int32_t flags; +}; + +#endif /* _XT_CLUSTER_MATCH_H */ diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index cdbaaff6d0d6..2562d05dbaf5 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -527,6 +527,22 @@ config NETFILTER_XT_TARGET_TCPOPTSTRIP This option adds a "TCPOPTSTRIP" target, which allows you to strip TCP options from TCP packets. +config NETFILTER_XT_MATCH_CLUSTER + tristate '"cluster" match support' + depends on NF_CONNTRACK + depends on NETFILTER_ADVANCED + ---help--- + This option allows you to build work-load-sharing clusters of + network servers/stateful firewalls without having a dedicated + load-balancing router/server/switch. Basically, this match returns + true when the packet must be handled by this cluster node. Thus, + all nodes see all packets and this match decides which node handles + what packets. The work-load sharing algorithm is based on source + address hashing. + + If you say Y or M here, try `iptables -m cluster --help` for + more information. + config NETFILTER_XT_MATCH_COMMENT tristate '"comment" match support' depends on NETFILTER_ADVANCED diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile index 7a9b8397573a..6282060fbda9 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -59,6 +59,7 @@ obj-$(CONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP) += xt_TCPOPTSTRIP.o obj-$(CONFIG_NETFILTER_XT_TARGET_TRACE) += xt_TRACE.o # matches +obj-$(CONFIG_NETFILTER_XT_MATCH_CLUSTER) += xt_cluster.o obj-$(CONFIG_NETFILTER_XT_MATCH_COMMENT) += xt_comment.o obj-$(CONFIG_NETFILTER_XT_MATCH_CONNBYTES) += xt_connbytes.o obj-$(CONFIG_NETFILTER_XT_MATCH_CONNLIMIT) += xt_connlimit.o diff --git a/net/netfilter/xt_cluster.c b/net/netfilter/xt_cluster.c new file mode 100644 index 000000000000..ad5bd890e4e8 --- /dev/null +++ b/net/netfilter/xt_cluster.c @@ -0,0 +1,164 @@ +/* + * (C) 2008-2009 Pablo Neira Ayuso + * + * 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 + +static inline u_int32_t nf_ct_orig_ipv4_src(const struct nf_conn *ct) +{ + return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip; +} + +static inline const void *nf_ct_orig_ipv6_src(const struct nf_conn *ct) +{ + return ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3.ip6; +} + +static inline u_int32_t +xt_cluster_hash_ipv4(u_int32_t ip, const struct xt_cluster_match_info *info) +{ + return jhash_1word(ip, info->hash_seed); +} + +static inline u_int32_t +xt_cluster_hash_ipv6(const void *ip, const struct xt_cluster_match_info *info) +{ + return jhash2(ip, NF_CT_TUPLE_L3SIZE / sizeof(__u32), info->hash_seed); +} + +static inline u_int32_t +xt_cluster_hash(const struct nf_conn *ct, + const struct xt_cluster_match_info *info) +{ + u_int32_t hash = 0; + + switch(nf_ct_l3num(ct)) { + case AF_INET: + hash = xt_cluster_hash_ipv4(nf_ct_orig_ipv4_src(ct), info); + break; + case AF_INET6: + hash = xt_cluster_hash_ipv6(nf_ct_orig_ipv6_src(ct), info); + break; + default: + WARN_ON(1); + break; + } + return (((u64)hash * info->total_nodes) >> 32); +} + +static inline bool +xt_cluster_is_multicast_addr(const struct sk_buff *skb, u_int8_t family) +{ + bool is_multicast = false; + + switch(family) { + case NFPROTO_IPV4: + is_multicast = ipv4_is_multicast(ip_hdr(skb)->daddr); + break; + case NFPROTO_IPV6: + is_multicast = ipv6_addr_type(&ipv6_hdr(skb)->daddr) & + IPV6_ADDR_MULTICAST; + break; + default: + WARN_ON(1); + break; + } + return is_multicast; +} + +static bool +xt_cluster_mt(const struct sk_buff *skb, const struct xt_match_param *par) +{ + struct sk_buff *pskb = (struct sk_buff *)skb; + const struct xt_cluster_match_info *info = par->matchinfo; + const struct nf_conn *ct; + enum ip_conntrack_info ctinfo; + unsigned long hash; + + /* This match assumes that all nodes see the same packets. This can be + * achieved if the switch that connects the cluster nodes support some + * sort of 'port mirroring'. However, if your switch does not support + * this, your cluster nodes can reply ARP request using a multicast MAC + * address. Thus, your switch will flood the same packets to the + * cluster nodes with the same multicast MAC address. Using a multicast + * link address is a RFC 1812 (section 3.3.2) violation, but this works + * fine in practise. + * + * Unfortunately, if you use the multicast MAC address, the link layer + * sets skbuff's pkt_type to PACKET_MULTICAST, which is not accepted + * by TCP and others for packets coming to this node. For that reason, + * this match mangles skbuff's pkt_type if it detects a packet + * addressed to a unicast address but using PACKET_MULTICAST. Yes, I + * know, matches should not alter packets, but we are doing this here + * because we would need to add a PKTTYPE target for this sole purpose. + */ + if (!xt_cluster_is_multicast_addr(skb, par->family) && + skb->pkt_type == PACKET_MULTICAST) { + pskb->pkt_type = PACKET_HOST; + } + + ct = nf_ct_get(skb, &ctinfo); + if (ct == NULL) + return false; + + if (ct == &nf_conntrack_untracked) + return false; + + if (ct->master) + hash = xt_cluster_hash(ct->master, info); + else + hash = xt_cluster_hash(ct, info); + + return !!((1 << hash) & info->node_mask) ^ + !!(info->flags & XT_CLUSTER_F_INV); +} + +static bool xt_cluster_mt_checkentry(const struct xt_mtchk_param *par) +{ + struct xt_cluster_match_info *info = par->matchinfo; + + if (info->node_mask >= (1 << info->total_nodes)) { + printk(KERN_ERR "xt_cluster: this node mask cannot be " + "higher than the total number of nodes\n"); + return false; + } + return true; +} + +static struct xt_match xt_cluster_match __read_mostly = { + .name = "cluster", + .family = NFPROTO_UNSPEC, + .match = xt_cluster_mt, + .checkentry = xt_cluster_mt_checkentry, + .matchsize = sizeof(struct xt_cluster_match_info), + .me = THIS_MODULE, +}; + +static int __init xt_cluster_mt_init(void) +{ + return xt_register_match(&xt_cluster_match); +} + +static void __exit xt_cluster_mt_fini(void) +{ + xt_unregister_match(&xt_cluster_match); +} + +MODULE_AUTHOR("Pablo Neira Ayuso "); +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("Xtables: hash-based cluster match"); +MODULE_ALIAS("ipt_cluster"); +MODULE_ALIAS("ip6t_cluster"); +module_init(xt_cluster_mt_init); +module_exit(xt_cluster_mt_fini); -- cgit v1.2.3 From dd5b6ce6fd465eab90357711c8e8124dc3a31ff0 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Mon, 23 Mar 2009 13:21:06 +0100 Subject: nefilter: nfnetlink: add nfnetlink_set_err and use it in ctnetlink This patch adds nfnetlink_set_err() to propagate the error to netlink broadcast listener in case of memory allocation errors in the message building. Signed-off-by: Pablo Neira Ayuso Signed-off-by: Patrick McHardy --- include/linux/netfilter/nfnetlink.h | 1 + net/netfilter/nf_conntrack_netlink.c | 2 ++ net/netfilter/nfnetlink.c | 6 ++++++ net/netlink/af_netlink.c | 1 + 4 files changed, 10 insertions(+) (limited to 'include/linux/netfilter') diff --git a/include/linux/netfilter/nfnetlink.h b/include/linux/netfilter/nfnetlink.h index 7d8e0455ccac..135e5cfe68a2 100644 --- a/include/linux/netfilter/nfnetlink.h +++ b/include/linux/netfilter/nfnetlink.h @@ -76,6 +76,7 @@ extern int nfnetlink_subsys_unregister(const struct nfnetlink_subsystem *n); extern int nfnetlink_has_listeners(unsigned int group); extern int nfnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo); +extern void nfnetlink_set_err(u32 pid, u32 group, int error); extern int nfnetlink_unicast(struct sk_buff *skb, u_int32_t pid, int flags); extern void nfnl_lock(void); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index d1fe9d15ac5c..1b75c9efb0eb 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -518,6 +518,7 @@ static int ctnetlink_conntrack_event(struct notifier_block *this, nla_put_failure: rcu_read_unlock(); nlmsg_failure: + nfnetlink_set_err(0, group, -ENOBUFS); kfree_skb(skb); return NOTIFY_DONE; } @@ -1514,6 +1515,7 @@ static int ctnetlink_expect_event(struct notifier_block *this, nla_put_failure: rcu_read_unlock(); nlmsg_failure: + nfnetlink_set_err(0, 0, -ENOBUFS); kfree_skb(skb); return NOTIFY_DONE; } diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index 9c0ba17a1ddb..2785d66a7e38 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -113,6 +113,12 @@ int nfnetlink_send(struct sk_buff *skb, u32 pid, unsigned group, int echo) } EXPORT_SYMBOL_GPL(nfnetlink_send); +void nfnetlink_set_err(u32 pid, u32 group, int error) +{ + netlink_set_err(nfnl, pid, group, error); +} +EXPORT_SYMBOL_GPL(nfnetlink_set_err); + int nfnetlink_unicast(struct sk_buff *skb, u_int32_t pid, int flags) { return netlink_unicast(nfnl, skb, pid, flags); diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 6ee69c27f806..5b33879c6422 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1106,6 +1106,7 @@ void netlink_set_err(struct sock *ssk, u32 pid, u32 group, int code) read_unlock(&nl_table_lock); } +EXPORT_SYMBOL(netlink_set_err); /* must be called with netlink table grabbed */ static void netlink_update_socket_mc(struct netlink_sock *nlk, -- cgit v1.2.3