1 files changed, 249 insertions, 329 deletions

diff --git a/net/netfilter/xt_hashlimit.c b/net/netfilter/xt_hashlimit.c
index 215a64835de..a3910fc2122 100644
--- a/net/netfilter/xt_hashlimit.c
+++ b/net/netfilter/xt_hashlimit.c
@@ -3,10 +3,12 @@
  *	separately for each hashbucket (sourceip/sourceport/dstip/dstport)
  *
  *	(C) 2003-2004 by Harald Welte <laforge@netfilter.org>
+ *	(C) 2006-2012 Patrick McHardy <kaber@trash.net>
  *	Copyright © CC Computer Consultants GmbH, 2007 - 2008
  *
  * Development of this code was funded by Astaro AG, http://www.astaro.com/
  */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 #include <linux/module.h>
 #include <linux/spinlock.h>
 #include <linux/random.h>
@@ -20,7 +22,7 @@
 #include <linux/mm.h>
 #include <linux/in.h>
 #include <linux/ip.h>
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 #include <linux/ipv6.h>
 #include <net/ipv6.h>
 #endif
@@ -36,7 +38,7 @@
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Harald Welte <laforge@netfilter.org>");
-MODULE_AUTHOR("Jan Engelhardt <jengelh@computergmbh.de>");
+MODULE_AUTHOR("Jan Engelhardt <jengelh@medozas.de>");
 MODULE_DESCRIPTION("Xtables: per hash-bucket rate-limit match");
 MODULE_ALIAS("ipt_hashlimit");
 MODULE_ALIAS("ip6t_hashlimit");
@@ -63,7 +65,7 @@ struct dsthash_dst {
 			__be32 src;
 			__be32 dst;
 		} ip;
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 		struct {
 			__be32 src[4];
 			__be32 dst[4];
@@ -80,12 +82,14 @@ struct dsthash_ent {
 	struct dsthash_dst dst;
 
 	/* modified structure members in the end */
+	spinlock_t lock;
 	unsigned long expires;		/* precalculated expiry time */
 	struct {
 		unsigned long prev;	/* last modification */
 		u_int32_t credit;
 		u_int32_t credit_cap, cost;
 	} rateinfo;
+	struct rcu_head rcu;
 };
 
 struct xt_hashlimit_htable {
@@ -104,6 +108,7 @@ struct xt_hashlimit_htable {
 
 	/* seq_file stuff */
 	struct proc_dir_entry *pde;
+	const char *name;
 	struct net *net;
 
 	struct hlist_head hash[0];	/* hashtable itself */
@@ -138,13 +143,14 @@ dsthash_find(const struct xt_hashlimit_htable *ht,
 	     const struct dsthash_dst *dst)
 {
 	struct dsthash_ent *ent;
-	struct hlist_node *pos;
 	u_int32_t hash = hash_dst(ht, dst);
 
 	if (!hlist_empty(&ht->hash[hash])) {
-		hlist_for_each_entry(ent, pos, &ht->hash[hash], node)
-			if (dst_cmp(ent, dst))
+		hlist_for_each_entry_rcu(ent, &ht->hash[hash], node)
+			if (dst_cmp(ent, dst)) {
+				spin_lock(&ent->lock);
 				return ent;
+			}
 	}
 	return NULL;
 }
@@ -152,121 +158,63 @@ dsthash_find(const struct xt_hashlimit_htable *ht,
 /* allocate dsthash_ent, initialize dst, put in htable and lock it */
 static struct dsthash_ent *
 dsthash_alloc_init(struct xt_hashlimit_htable *ht,
-		   const struct dsthash_dst *dst)
+		   const struct dsthash_dst *dst, bool *race)
 {
 	struct dsthash_ent *ent;
 
+	spin_lock(&ht->lock);
+
+	/* Two or more packets may race to create the same entry in the
+	 * hashtable, double check if this packet lost race.
+	 */
+	ent = dsthash_find(ht, dst);
+	if (ent != NULL) {
+		spin_unlock(&ht->lock);
+		*race = true;
+		return ent;
+	}
+
 	/* initialize hash with random val at the time we allocate
 	 * the first hashtable entry */
-	if (!ht->rnd_initialized) {
+	if (unlikely(!ht->rnd_initialized)) {
 		get_random_bytes(&ht->rnd, sizeof(ht->rnd));
 		ht->rnd_initialized = true;
 	}
 
 	if (ht->cfg.max && ht->count >= ht->cfg.max) {
 		/* FIXME: do something. question is what.. */
-		if (net_ratelimit())
-			printk(KERN_WARNING
-				"xt_hashlimit: max count of %u reached\n",
-				ht->cfg.max);
-		return NULL;
+		net_err_ratelimited("max count of %u reached\n", ht->cfg.max);
+		ent = NULL;
+	} else
+		ent = kmem_cache_alloc(hashlimit_cachep, GFP_ATOMIC);
+	if (ent) {
+		memcpy(&ent->dst, dst, sizeof(ent->dst));
+		spin_lock_init(&ent->lock);
+
+		spin_lock(&ent->lock);
+		hlist_add_head_rcu(&ent->node, &ht->hash[hash_dst(ht, dst)]);
+		ht->count++;
 	}
+	spin_unlock(&ht->lock);
+	return ent;
+}
 
-	ent = kmem_cache_alloc(hashlimit_cachep, GFP_ATOMIC);
-	if (!ent) {
-		if (net_ratelimit())
-			printk(KERN_ERR
-				"xt_hashlimit: can't allocate dsthash_ent\n");
-		return NULL;
-	}
-	memcpy(&ent->dst, dst, sizeof(ent->dst));
+static void dsthash_free_rcu(struct rcu_head *head)
+{
+	struct dsthash_ent *ent = container_of(head, struct dsthash_ent, rcu);
 
-	hlist_add_head(&ent->node, &ht->hash[hash_dst(ht, dst)]);
-	ht->count++;
-	return ent;
+	kmem_cache_free(hashlimit_cachep, ent);
 }
 
 static inline void
 dsthash_free(struct xt_hashlimit_htable *ht, struct dsthash_ent *ent)
 {
-	hlist_del(&ent->node);
-	kmem_cache_free(hashlimit_cachep, ent);
+	hlist_del_rcu(&ent->node);
+	call_rcu_bh(&ent->rcu, dsthash_free_rcu);
 	ht->count--;
 }
 static void htable_gc(unsigned long htlong);
 
-static int htable_create_v0(struct net *net, struct xt_hashlimit_info *minfo, u_int8_t family)
-{
-	struct hashlimit_net *hashlimit_net = hashlimit_pernet(net);
-	struct xt_hashlimit_htable *hinfo;
-	unsigned int size;
-	unsigned int i;
-
-	if (minfo->cfg.size)
-		size = minfo->cfg.size;
-	else {
-		size = ((totalram_pages << PAGE_SHIFT) / 16384) /
-		       sizeof(struct list_head);
-		if (totalram_pages > (1024 * 1024 * 1024 / PAGE_SIZE))
-			size = 8192;
-		if (size < 16)
-			size = 16;
-	}
-	/* FIXME: don't use vmalloc() here or anywhere else -HW */
-	hinfo = vmalloc(sizeof(struct xt_hashlimit_htable) +
-			sizeof(struct list_head) * size);
-	if (!hinfo) {
-		printk(KERN_ERR "xt_hashlimit: unable to create hashtable\n");
-		return -1;
-	}
-	minfo->hinfo = hinfo;
-
-	/* copy match config into hashtable config */
-	hinfo->cfg.mode        = minfo->cfg.mode;
-	hinfo->cfg.avg         = minfo->cfg.avg;
-	hinfo->cfg.burst       = minfo->cfg.burst;
-	hinfo->cfg.max         = minfo->cfg.max;
-	hinfo->cfg.gc_interval = minfo->cfg.gc_interval;
-	hinfo->cfg.expire      = minfo->cfg.expire;
-
-	if (family == NFPROTO_IPV4)
-		hinfo->cfg.srcmask = hinfo->cfg.dstmask = 32;
-	else
-		hinfo->cfg.srcmask = hinfo->cfg.dstmask = 128;
-
-	hinfo->cfg.size = size;
-	if (!hinfo->cfg.max)
-		hinfo->cfg.max = 8 * hinfo->cfg.size;
-	else if (hinfo->cfg.max < hinfo->cfg.size)
-		hinfo->cfg.max = hinfo->cfg.size;
-
-	for (i = 0; i < hinfo->cfg.size; i++)
-		INIT_HLIST_HEAD(&hinfo->hash[i]);
-
-	hinfo->use = 1;
-	hinfo->count = 0;
-	hinfo->family = family;
-	hinfo->rnd_initialized = false;
-	spin_lock_init(&hinfo->lock);
-	hinfo->pde = proc_create_data(minfo->name, 0,
-		(family == NFPROTO_IPV4) ?
-		hashlimit_net->ipt_hashlimit : hashlimit_net->ip6t_hashlimit,
-		&dl_file_ops, hinfo);
-	if (!hinfo->pde) {
-		vfree(hinfo);
-		return -1;
-	}
-	hinfo->net = net;
-
-	setup_timer(&hinfo->timer, htable_gc, (unsigned long )hinfo);
-	hinfo->timer.expires = jiffies + msecs_to_jiffies(hinfo->cfg.gc_interval);
-	add_timer(&hinfo->timer);
-
-	hlist_add_head(&hinfo->node, &hashlimit_net->htables);
-
-	return 0;
-}
-
 static int htable_create(struct net *net, struct xt_hashlimit_mtinfo1 *minfo,
 			 u_int8_t family)
 {
@@ -288,10 +236,8 @@ static int htable_create(struct net *net, struct xt_hashlimit_mtinfo1 *minfo,
 	/* FIXME: don't use vmalloc() here or anywhere else -HW */
 	hinfo = vmalloc(sizeof(struct xt_hashlimit_htable) +
 	                sizeof(struct list_head) * size);
-	if (hinfo == NULL) {
-		printk(KERN_ERR "xt_hashlimit: unable to create hashtable\n");
-		return -1;
-	}
+	if (hinfo == NULL)
+		return -ENOMEM;
 	minfo->hinfo = hinfo;
 
 	/* copy match config into hashtable config */
@@ -309,6 +255,11 @@ static int htable_create(struct net *net, struct xt_hashlimit_mtinfo1 *minfo,
 	hinfo->count = 0;
 	hinfo->family = family;
 	hinfo->rnd_initialized = false;
+	hinfo->name = kstrdup(minfo->name, GFP_KERNEL);
+	if (!hinfo->name) {
+		vfree(hinfo);
+		return -ENOMEM;
+	}
 	spin_lock_init(&hinfo->lock);
 
 	hinfo->pde = proc_create_data(minfo->name, 0,
@@ -316,8 +267,9 @@ static int htable_create(struct net *net, struct xt_hashlimit_mtinfo1 *minfo,
 		hashlimit_net->ipt_hashlimit : hashlimit_net->ip6t_hashlimit,
 		&dl_file_ops, hinfo);
 	if (hinfo->pde == NULL) {
+		kfree(hinfo->name);
 		vfree(hinfo);
-		return -1;
+		return -ENOMEM;
 	}
 	hinfo->net = net;
 
@@ -352,8 +304,8 @@ static void htable_selective_cleanup(struct xt_hashlimit_htable *ht,
 	spin_lock_bh(&ht->lock);
 	for (i = 0; i < ht->cfg.size; i++) {
 		struct dsthash_ent *dh;
-		struct hlist_node *pos, *n;
-		hlist_for_each_entry_safe(dh, pos, n, &ht->hash[i], node) {
+		struct hlist_node *n;
+		hlist_for_each_entry_safe(dh, n, &ht->hash[i], node) {
 			if ((*select)(ht, dh))
 				dsthash_free(ht, dh);
 		}
@@ -373,19 +325,26 @@ static void htable_gc(unsigned long htlong)
 	add_timer(&ht->timer);
 }
 
-static void htable_destroy(struct xt_hashlimit_htable *hinfo)
+static void htable_remove_proc_entry(struct xt_hashlimit_htable *hinfo)
 {
 	struct hashlimit_net *hashlimit_net = hashlimit_pernet(hinfo->net);
 	struct proc_dir_entry *parent;
 
-	del_timer_sync(&hinfo->timer);
-
 	if (hinfo->family == NFPROTO_IPV4)
 		parent = hashlimit_net->ipt_hashlimit;
 	else
 		parent = hashlimit_net->ip6t_hashlimit;
-	remove_proc_entry(hinfo->pde->name, parent);
+
+	if (parent != NULL)
+		remove_proc_entry(hinfo->name, parent);
+}
+
+static void htable_destroy(struct xt_hashlimit_htable *hinfo)
+{
+	del_timer_sync(&hinfo->timer);
+	htable_remove_proc_entry(hinfo);
 	htable_selective_cleanup(hinfo, select_all);
+	kfree(hinfo->name);
 	vfree(hinfo);
 }
 
@@ -395,10 +354,9 @@ static struct xt_hashlimit_htable *htable_find_get(struct net *net,
 {
 	struct hashlimit_net *hashlimit_net = hashlimit_pernet(net);
 	struct xt_hashlimit_htable *hinfo;
-	struct hlist_node *pos;
 
-	hlist_for_each_entry(hinfo, pos, &hashlimit_net->htables, node) {
-		if (!strcmp(name, hinfo->pde->name) &&
+	hlist_for_each_entry(hinfo, &hashlimit_net->htables, node) {
+		if (!strcmp(name, hinfo->name) &&
 		    hinfo->family == family) {
 			hinfo->use++;
 			return hinfo;
@@ -453,9 +411,20 @@ static void htable_put(struct xt_hashlimit_htable *hinfo)
 
 #define CREDITS_PER_JIFFY POW2_BELOW32(MAX_CPJ)
 
+/* in byte mode, the lowest possible rate is one packet/second.
+ * credit_cap is used as a counter that tells us how many times we can
+ * refill the "credits available" counter when it becomes empty.
+ */
+#define MAX_CPJ_BYTES (0xFFFFFFFF / HZ)
+#define CREDITS_PER_JIFFY_BYTES POW2_BELOW32(MAX_CPJ_BYTES)
+
+static u32 xt_hashlimit_len_to_chunks(u32 len)
+{
+	return (len >> XT_HASHLIMIT_BYTE_SHIFT) + 1;
+}
+
 /* Precision saver. */
-static inline u_int32_t
-user2credits(u_int32_t user)
+static u32 user2credits(u32 user)
 {
 	/* If multiplying would overflow... */
 	if (user > 0xFFFFFFFF / (HZ*CREDITS_PER_JIFFY))
@@ -465,12 +434,53 @@ user2credits(u_int32_t user)
 	return (user * HZ * CREDITS_PER_JIFFY) / XT_HASHLIMIT_SCALE;
 }
 
-static inline void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now)
+static u32 user2credits_byte(u32 user)
+{
+	u64 us = user;
+	us *= HZ * CREDITS_PER_JIFFY_BYTES;
+	return (u32) (us >> 32);
+}
+
+static void rateinfo_recalc(struct dsthash_ent *dh, unsigned long now, u32 mode)
 {
-	dh->rateinfo.credit += (now - dh->rateinfo.prev) * CREDITS_PER_JIFFY;
-	if (dh->rateinfo.credit > dh->rateinfo.credit_cap)
-		dh->rateinfo.credit = dh->rateinfo.credit_cap;
+	unsigned long delta = now - dh->rateinfo.prev;
+	u32 cap;
+
+	if (delta == 0)
+		return;
+
 	dh->rateinfo.prev = now;
+
+	if (mode & XT_HASHLIMIT_BYTES) {
+		u32 tmp = dh->rateinfo.credit;
+		dh->rateinfo.credit += CREDITS_PER_JIFFY_BYTES * delta;
+		cap = CREDITS_PER_JIFFY_BYTES * HZ;
+		if (tmp >= dh->rateinfo.credit) {/* overflow */
+			dh->rateinfo.credit = cap;
+			return;
+		}
+	} else {
+		dh->rateinfo.credit += delta * CREDITS_PER_JIFFY;
+		cap = dh->rateinfo.credit_cap;
+	}
+	if (dh->rateinfo.credit > cap)
+		dh->rateinfo.credit = cap;
+}
+
+static void rateinfo_init(struct dsthash_ent *dh,
+			  struct xt_hashlimit_htable *hinfo)
+{
+	dh->rateinfo.prev = jiffies;
+	if (hinfo->cfg.mode & XT_HASHLIMIT_BYTES) {
+		dh->rateinfo.credit = CREDITS_PER_JIFFY_BYTES * HZ;
+		dh->rateinfo.cost = user2credits_byte(hinfo->cfg.avg);
+		dh->rateinfo.credit_cap = hinfo->cfg.burst;
+	} else {
+		dh->rateinfo.credit = user2credits(hinfo->cfg.avg *
+						   hinfo->cfg.burst);
+		dh->rateinfo.cost = user2credits(hinfo->cfg.avg);
+		dh->rateinfo.credit_cap = dh->rateinfo.credit;
+	}
 }
 
 static inline __be32 maskl(__be32 a, unsigned int l)
@@ -478,7 +488,7 @@ static inline __be32 maskl(__be32 a, unsigned int l)
 	return l ? htonl(ntohl(a) & ~0 << (32 - l)) : 0;
 }
 
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 static void hashlimit_ipv6_mask(__be32 *i, unsigned int p)
 {
 	switch (p) {
@@ -510,6 +520,7 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo,
 {
 	__be16 _ports[2], *ports;
 	u8 nexthdr;
+	int poff;
 
 	memset(dst, 0, sizeof(*dst));
 
@@ -527,8 +538,11 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo,
 			return 0;
 		nexthdr = ip_hdr(skb)->protocol;
 		break;
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 	case NFPROTO_IPV6:
+	{
+		__be16 frag_off;
+
 		if (hinfo->cfg.mode & XT_HASHLIMIT_HASH_DIP) {
 			memcpy(&dst->ip6.dst, &ipv6_hdr(skb)->daddr,
 			       sizeof(dst->ip6.dst));
@@ -544,29 +558,24 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo,
 		      (XT_HASHLIMIT_HASH_DPT | XT_HASHLIMIT_HASH_SPT)))
 			return 0;
 		nexthdr = ipv6_hdr(skb)->nexthdr;
-		protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr);
+		protoff = ipv6_skip_exthdr(skb, sizeof(struct ipv6hdr), &nexthdr, &frag_off);
 		if ((int)protoff < 0)
 			return -1;
 		break;
+	}
 #endif
 	default:
 		BUG();
 		return 0;
 	}
 
-	switch (nexthdr) {
-	case IPPROTO_TCP:
-	case IPPROTO_UDP:
-	case IPPROTO_UDPLITE:
-	case IPPROTO_SCTP:
-	case IPPROTO_DCCP:
-		ports = skb_header_pointer(skb, protoff, sizeof(_ports),
+	poff = proto_ports_offset(nexthdr);
+	if (poff >= 0) {
+		ports = skb_header_pointer(skb, protoff + poff, sizeof(_ports),
 					   &_ports);
-		break;
-	default:
+	} else {
 		_ports[0] = _ports[1] = 0;
 		ports = _ports;
-		break;
 	}
 	if (!ports)
 		return -1;
@@ -577,184 +586,128 @@ hashlimit_init_dst(const struct xt_hashlimit_htable *hinfo,
 	return 0;
 }
 
-static bool
-hashlimit_mt_v0(const struct sk_buff *skb, const struct xt_match_param *par)
+static u32 hashlimit_byte_cost(unsigned int len, struct dsthash_ent *dh)
 {
-	const struct xt_hashlimit_info *r = par->matchinfo;
-	struct xt_hashlimit_htable *hinfo = r->hinfo;
-	unsigned long now = jiffies;
-	struct dsthash_ent *dh;
-	struct dsthash_dst dst;
-
-	if (hashlimit_init_dst(hinfo, &dst, skb, par->thoff) < 0)
-		goto hotdrop;
+	u64 tmp = xt_hashlimit_len_to_chunks(len);
+	tmp = tmp * dh->rateinfo.cost;
 
-	spin_lock_bh(&hinfo->lock);
-	dh = dsthash_find(hinfo, &dst);
-	if (!dh) {
-		dh = dsthash_alloc_init(hinfo, &dst);
-		if (!dh) {
-			spin_unlock_bh(&hinfo->lock);
-			goto hotdrop;
-		}
+	if (unlikely(tmp > CREDITS_PER_JIFFY_BYTES * HZ))
+		tmp = CREDITS_PER_JIFFY_BYTES * HZ;
 
-		dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire);
-		dh->rateinfo.prev = jiffies;
-		dh->rateinfo.credit = user2credits(hinfo->cfg.avg *
-						   hinfo->cfg.burst);
-		dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg *
-						       hinfo->cfg.burst);
-		dh->rateinfo.cost = user2credits(hinfo->cfg.avg);
-	} else {
-		/* update expiration timeout */
-		dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire);
-		rateinfo_recalc(dh, now);
-	}
-
-	if (dh->rateinfo.credit >= dh->rateinfo.cost) {
-		/* We're underlimit. */
-		dh->rateinfo.credit -= dh->rateinfo.cost;
-		spin_unlock_bh(&hinfo->lock);
-		return true;
+	if (dh->rateinfo.credit < tmp && dh->rateinfo.credit_cap) {
+		dh->rateinfo.credit_cap--;
+		dh->rateinfo.credit = CREDITS_PER_JIFFY_BYTES * HZ;
 	}
-
-	spin_unlock_bh(&hinfo->lock);
-
-	/* default case: we're overlimit, thus don't match */
-	return false;
-
-hotdrop:
-	*par->hotdrop = true;
-	return false;
+	return (u32) tmp;
 }
 
 static bool
-hashlimit_mt(const struct sk_buff *skb, const struct xt_match_param *par)
+hashlimit_mt(const struct sk_buff *skb, struct xt_action_param *par)
 {
 	const struct xt_hashlimit_mtinfo1 *info = par->matchinfo;
 	struct xt_hashlimit_htable *hinfo = info->hinfo;
 	unsigned long now = jiffies;
 	struct dsthash_ent *dh;
 	struct dsthash_dst dst;
+	bool race = false;
+	u32 cost;
 
 	if (hashlimit_init_dst(hinfo, &dst, skb, par->thoff) < 0)
 		goto hotdrop;
 
-	spin_lock_bh(&hinfo->lock);
+	rcu_read_lock_bh();
 	dh = dsthash_find(hinfo, &dst);
 	if (dh == NULL) {
-		dh = dsthash_alloc_init(hinfo, &dst);
+		dh = dsthash_alloc_init(hinfo, &dst, &race);
 		if (dh == NULL) {
-			spin_unlock_bh(&hinfo->lock);
+			rcu_read_unlock_bh();
 			goto hotdrop;
+		} else if (race) {
+			/* Already got an entry, update expiration timeout */
+			dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire);
+			rateinfo_recalc(dh, now, hinfo->cfg.mode);
+		} else {
+			dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire);
+			rateinfo_init(dh, hinfo);
 		}
-
-		dh->expires = jiffies + msecs_to_jiffies(hinfo->cfg.expire);
-		dh->rateinfo.prev = jiffies;
-		dh->rateinfo.credit = user2credits(hinfo->cfg.avg *
-		                      hinfo->cfg.burst);
-		dh->rateinfo.credit_cap = user2credits(hinfo->cfg.avg *
-		                          hinfo->cfg.burst);
-		dh->rateinfo.cost = user2credits(hinfo->cfg.avg);
 	} else {
 		/* update expiration timeout */
 		dh->expires = now + msecs_to_jiffies(hinfo->cfg.expire);
-		rateinfo_recalc(dh, now);
+		rateinfo_recalc(dh, now, hinfo->cfg.mode);
 	}
 
-	if (dh->rateinfo.credit >= dh->rateinfo.cost) {
+	if (info->cfg.mode & XT_HASHLIMIT_BYTES)
+		cost = hashlimit_byte_cost(skb->len, dh);
+	else
+		cost = dh->rateinfo.cost;
+
+	if (dh->rateinfo.credit >= cost) {
 		/* below the limit */
-		dh->rateinfo.credit -= dh->rateinfo.cost;
-		spin_unlock_bh(&hinfo->lock);
+		dh->rateinfo.credit -= cost;
+		spin_unlock(&dh->lock);
+		rcu_read_unlock_bh();
 		return !(info->cfg.mode & XT_HASHLIMIT_INVERT);
 	}
 
-	spin_unlock_bh(&hinfo->lock);
+	spin_unlock(&dh->lock);
+	rcu_read_unlock_bh();
 	/* default match is underlimit - so over the limit, we need to invert */
 	return info->cfg.mode & XT_HASHLIMIT_INVERT;
 
  hotdrop:
-	*par->hotdrop = true;
+	par->hotdrop = true;
 	return false;
 }
 
-static bool hashlimit_mt_check_v0(const struct xt_mtchk_param *par)
-{
-	struct net *net = par->net;
-	struct xt_hashlimit_info *r = par->matchinfo;
-
-	/* Check for overflow. */
-	if (r->cfg.burst == 0 ||
-	    user2credits(r->cfg.avg * r->cfg.burst) < user2credits(r->cfg.avg)) {
-		printk(KERN_ERR "xt_hashlimit: overflow, try lower: %u/%u\n",
-		       r->cfg.avg, r->cfg.burst);
-		return false;
-	}
-	if (r->cfg.mode == 0 ||
-	    r->cfg.mode > (XT_HASHLIMIT_HASH_DPT |
-			   XT_HASHLIMIT_HASH_DIP |
-			   XT_HASHLIMIT_HASH_SIP |
-			   XT_HASHLIMIT_HASH_SPT))
-		return false;
-	if (!r->cfg.gc_interval)
-		return false;
-	if (!r->cfg.expire)
-		return false;
-	if (r->name[sizeof(r->name) - 1] != '\0')
-		return false;
-
-	mutex_lock(&hashlimit_mutex);
-	r->hinfo = htable_find_get(net, r->name, par->match->family);
-	if (!r->hinfo && htable_create_v0(net, r, par->match->family) != 0) {
-		mutex_unlock(&hashlimit_mutex);
-		return false;
-	}
-	mutex_unlock(&hashlimit_mutex);
-
-	return true;
-}
-
-static bool hashlimit_mt_check(const struct xt_mtchk_param *par)
+static int hashlimit_mt_check(const struct xt_mtchk_param *par)
 {
 	struct net *net = par->net;
 	struct xt_hashlimit_mtinfo1 *info = par->matchinfo;
+	int ret;
 
-	/* Check for overflow. */
-	if (info->cfg.burst == 0 ||
-	    user2credits(info->cfg.avg * info->cfg.burst) <
-	    user2credits(info->cfg.avg)) {
-		printk(KERN_ERR "xt_hashlimit: overflow, try lower: %u/%u\n",
-		       info->cfg.avg, info->cfg.burst);
-		return false;
-	}
 	if (info->cfg.gc_interval == 0 || info->cfg.expire == 0)
-		return false;
+		return -EINVAL;
 	if (info->name[sizeof(info->name)-1] != '\0')
-		return false;
-	if (par->match->family == NFPROTO_IPV4) {
+		return -EINVAL;
+	if (par->family == NFPROTO_IPV4) {
 		if (info->cfg.srcmask > 32 || info->cfg.dstmask > 32)
-			return false;
+			return -EINVAL;
 	} else {
 		if (info->cfg.srcmask > 128 || info->cfg.dstmask > 128)
-			return false;
+			return -EINVAL;
 	}
 
-	mutex_lock(&hashlimit_mutex);
-	info->hinfo = htable_find_get(net, info->name, par->match->family);
-	if (!info->hinfo && htable_create(net, info, par->match->family) != 0) {
-		mutex_unlock(&hashlimit_mutex);
-		return false;
+	if (info->cfg.mode & ~XT_HASHLIMIT_ALL) {
+		pr_info("Unknown mode mask %X, kernel too old?\n",
+						info->cfg.mode);
+		return -EINVAL;
 	}
-	mutex_unlock(&hashlimit_mutex);
-	return true;
-}
 
-static void
-hashlimit_mt_destroy_v0(const struct xt_mtdtor_param *par)
-{
-	const struct xt_hashlimit_info *r = par->matchinfo;
+	/* Check for overflow. */
+	if (info->cfg.mode & XT_HASHLIMIT_BYTES) {
+		if (user2credits_byte(info->cfg.avg) == 0) {
+			pr_info("overflow, rate too high: %u\n", info->cfg.avg);
+			return -EINVAL;
+		}
+	} else if (info->cfg.burst == 0 ||
+		    user2credits(info->cfg.avg * info->cfg.burst) <
+		    user2credits(info->cfg.avg)) {
+			pr_info("overflow, try lower: %u/%u\n",
+				info->cfg.avg, info->cfg.burst);
+			return -ERANGE;
+	}
 
-	htable_put(r->hinfo);
+	mutex_lock(&hashlimit_mutex);
+	info->hinfo = htable_find_get(net, info->name, par->family);
+	if (info->hinfo == NULL) {
+		ret = htable_create(net, info, par->family);
+		if (ret < 0) {
+			mutex_unlock(&hashlimit_mutex);
+			return ret;
+		}
+	}
+	mutex_unlock(&hashlimit_mutex);
+	return 0;
 }
 
 static void hashlimit_mt_destroy(const struct xt_mtdtor_param *par)
@@ -764,47 +717,8 @@ static void hashlimit_mt_destroy(const struct xt_mtdtor_param *par)
 	htable_put(info->hinfo);
 }
 
-#ifdef CONFIG_COMPAT
-struct compat_xt_hashlimit_info {
-	char name[IFNAMSIZ];
-	struct hashlimit_cfg cfg;
-	compat_uptr_t hinfo;
-	compat_uptr_t master;
-};
-
-static void hashlimit_mt_compat_from_user(void *dst, const void *src)
-{
-	int off = offsetof(struct compat_xt_hashlimit_info, hinfo);
-
-	memcpy(dst, src, off);
-	memset(dst + off, 0, sizeof(struct compat_xt_hashlimit_info) - off);
-}
-
-static int hashlimit_mt_compat_to_user(void __user *dst, const void *src)
-{
-	int off = offsetof(struct compat_xt_hashlimit_info, hinfo);
-
-	return copy_to_user(dst, src, off) ? -EFAULT : 0;
-}
-#endif
-
 static struct xt_match hashlimit_mt_reg[] __read_mostly = {
 	{
-		.name		= "hashlimit",
-		.revision	= 0,
-		.family		= NFPROTO_IPV4,
-		.match		= hashlimit_mt_v0,
-		.matchsize	= sizeof(struct xt_hashlimit_info),
-#ifdef CONFIG_COMPAT
-		.compatsize	= sizeof(struct compat_xt_hashlimit_info),
-		.compat_from_user = hashlimit_mt_compat_from_user,
-		.compat_to_user	= hashlimit_mt_compat_to_user,
-#endif
-		.checkentry	= hashlimit_mt_check_v0,
-		.destroy	= hashlimit_mt_destroy_v0,
-		.me		= THIS_MODULE
-	},
-	{
 		.name           = "hashlimit",
 		.revision       = 1,
 		.family         = NFPROTO_IPV4,
@@ -814,21 +728,7 @@ static struct xt_match hashlimit_mt_reg[] __read_mostly = {
 		.destroy        = hashlimit_mt_destroy,
 		.me             = THIS_MODULE,
 	},
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
-	{
-		.name		= "hashlimit",
-		.family		= NFPROTO_IPV6,
-		.match		= hashlimit_mt_v0,
-		.matchsize	= sizeof(struct xt_hashlimit_info),
-#ifdef CONFIG_COMPAT
-		.compatsize	= sizeof(struct compat_xt_hashlimit_info),
-		.compat_from_user = hashlimit_mt_compat_from_user,
-		.compat_to_user	= hashlimit_mt_compat_to_user,
-#endif
-		.checkentry	= hashlimit_mt_check_v0,
-		.destroy	= hashlimit_mt_destroy_v0,
-		.me		= THIS_MODULE
-	},
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 	{
 		.name           = "hashlimit",
 		.revision       = 1,
@@ -888,12 +788,16 @@ static void dl_seq_stop(struct seq_file *s, void *v)
 static int dl_seq_real_show(struct dsthash_ent *ent, u_int8_t family,
 				   struct seq_file *s)
 {
+	int res;
+	const struct xt_hashlimit_htable *ht = s->private;
+
+	spin_lock(&ent->lock);
 	/* recalculate to show accurate numbers */
-	rateinfo_recalc(ent, jiffies);
+	rateinfo_recalc(ent, jiffies, ht->cfg.mode);
 
 	switch (family) {
 	case NFPROTO_IPV4:
-		return seq_printf(s, "%ld %pI4:%u->%pI4:%u %u %u %u\n",
+		res = seq_printf(s, "%ld %pI4:%u->%pI4:%u %u %u %u\n",
 				 (long)(ent->expires - jiffies)/HZ,
 				 &ent->dst.ip.src,
 				 ntohs(ent->dst.src_port),
@@ -901,9 +805,10 @@ static int dl_seq_real_show(struct dsthash_ent *ent, u_int8_t family,
 				 ntohs(ent->dst.dst_port),
 				 ent->rateinfo.credit, ent->rateinfo.credit_cap,
 				 ent->rateinfo.cost);
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+		break;
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 	case NFPROTO_IPV6:
-		return seq_printf(s, "%ld %pI6:%u->%pI6:%u %u %u %u\n",
+		res = seq_printf(s, "%ld %pI6:%u->%pI6:%u %u %u %u\n",
 				 (long)(ent->expires - jiffies)/HZ,
 				 &ent->dst.ip6.src,
 				 ntohs(ent->dst.src_port),
@@ -911,11 +816,14 @@ static int dl_seq_real_show(struct dsthash_ent *ent, u_int8_t family,
 				 ntohs(ent->dst.dst_port),
 				 ent->rateinfo.credit, ent->rateinfo.credit_cap,
 				 ent->rateinfo.cost);
+		break;
 #endif
 	default:
 		BUG();
-		return 0;
+		res = 0;
 	}
+	spin_unlock(&ent->lock);
+	return res;
 }
 
 static int dl_seq_show(struct seq_file *s, void *v)
@@ -923,10 +831,9 @@ static int dl_seq_show(struct seq_file *s, void *v)
 	struct xt_hashlimit_htable *htable = s->private;
 	unsigned int *bucket = (unsigned int *)v;
 	struct dsthash_ent *ent;
-	struct hlist_node *pos;
 
 	if (!hlist_empty(&htable->hash[*bucket])) {
-		hlist_for_each_entry(ent, pos, &htable->hash[*bucket], node)
+		hlist_for_each_entry(ent, &htable->hash[*bucket], node)
 			if (dl_seq_real_show(ent, htable->family, s))
 				return -1;
 	}
@@ -946,7 +853,7 @@ static int dl_proc_open(struct inode *inode, struct file *file)
 
 	if (!ret) {
 		struct seq_file *sf = file->private_data;
-		sf->private = PDE(inode)->data;
+		sf->private = PDE_DATA(inode);
 	}
 	return ret;
 }
@@ -966,10 +873,10 @@ static int __net_init hashlimit_proc_net_init(struct net *net)
 	hashlimit_net->ipt_hashlimit = proc_mkdir("ipt_hashlimit", net->proc_net);
 	if (!hashlimit_net->ipt_hashlimit)
 		return -ENOMEM;
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
 	hashlimit_net->ip6t_hashlimit = proc_mkdir("ip6t_hashlimit", net->proc_net);
 	if (!hashlimit_net->ip6t_hashlimit) {
-		proc_net_remove(net, "ipt_hashlimit");
+		remove_proc_entry("ipt_hashlimit", net->proc_net);
 		return -ENOMEM;
 	}
 #endif
@@ -978,9 +885,23 @@ static int __net_init hashlimit_proc_net_init(struct net *net)
 
 static void __net_exit hashlimit_proc_net_exit(struct net *net)
 {
-	proc_net_remove(net, "ipt_hashlimit");
-#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
-	proc_net_remove(net, "ip6t_hashlimit");
+	struct xt_hashlimit_htable *hinfo;
+	struct hashlimit_net *hashlimit_net = hashlimit_pernet(net);
+
+	/* hashlimit_net_exit() is called before hashlimit_mt_destroy().
+	 * Make sure that the parent ipt_hashlimit and ip6t_hashlimit proc
+	 * entries is empty before trying to remove it.
+	 */
+	mutex_lock(&hashlimit_mutex);
+	hlist_for_each_entry(hinfo, &hashlimit_net->htables, node)
+		htable_remove_proc_entry(hinfo);
+	hashlimit_net->ipt_hashlimit = NULL;
+	hashlimit_net->ip6t_hashlimit = NULL;
+	mutex_unlock(&hashlimit_mutex);
+
+	remove_proc_entry("ipt_hashlimit", net->proc_net);
+#if IS_ENABLED(CONFIG_IP6_NF_IPTABLES)
+	remove_proc_entry("ip6t_hashlimit", net->proc_net);
 #endif
 }
 
@@ -994,9 +915,6 @@ static int __net_init hashlimit_net_init(struct net *net)
 
 static void __net_exit hashlimit_net_exit(struct net *net)
 {
-	struct hashlimit_net *hashlimit_net = hashlimit_pernet(net);
-
-	BUG_ON(!hlist_empty(&hashlimit_net->htables));
 	hashlimit_proc_net_exit(net);
 }
 
@@ -1024,7 +942,7 @@ static int __init hashlimit_mt_init(void)
 					    sizeof(struct dsthash_ent), 0, 0,
 					    NULL);
 	if (!hashlimit_cachep) {
-		printk(KERN_ERR "xt_hashlimit: unable to create slab cache\n");
+		pr_warning("unable to create slab cache\n");
 		goto err2;
 	}
 	return 0;
@@ -1039,9 +957,11 @@ err1:
 
 static void __exit hashlimit_mt_exit(void)
 {
-	kmem_cache_destroy(hashlimit_cachep);
 	xt_unregister_matches(hashlimit_mt_reg, ARRAY_SIZE(hashlimit_mt_reg));
 	unregister_pernet_subsys(&hashlimit_net_ops);
+
+	rcu_barrier_bh();
+	kmem_cache_destroy(hashlimit_cachep);
 }
 
 module_init(hashlimit_mt_init);