1 files changed, 103 insertions, 96 deletions

diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 486ecb0aeb87..f7012479713b 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -76,23 +76,27 @@ static u8 ip4_frag_ecn(u8 tos)
 static struct inet_frags ip4_frags;
 
 static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb,
-			 struct sk_buff *prev_tail, struct net_device *dev);
+			 struct sk_buff *prev_tail, struct net_device *dev,
+			 int *refs);
 
 
 static void ip4_frag_init(struct inet_frag_queue *q, const void *a)
 {
 	struct ipq *qp = container_of(q, struct ipq, q);
-	struct netns_ipv4 *ipv4 = container_of(q->net, struct netns_ipv4,
-					       frags);
-	struct net *net = container_of(ipv4, struct net, ipv4);
-
 	const struct frag_v4_compare_key *key = a;
+	struct net *net = q->fqdir->net;
+	struct inet_peer *p = NULL;
 
 	q->key.v4 = *key;
 	qp->ecn = 0;
-	qp->peer = q->net->max_dist ?
-		inet_getpeer_v4(net->ipv4.peers, key->saddr, key->vif, 1) :
-		NULL;
+	if (q->fqdir->max_dist) {
+		rcu_read_lock();
+		p = inet_getpeer_v4(net->ipv4.peers, key->saddr, key->vif);
+		if (p && !refcount_inc_not_zero(&p->refcnt))
+			p = NULL;
+		rcu_read_unlock();
+	}
+	qp->peer = p;
 }
 
 static void ip4_frag_free(struct inet_frag_queue *q)
@@ -104,22 +108,6 @@ static void ip4_frag_free(struct inet_frag_queue *q)
 		inet_putpeer(qp->peer);
 }
 
-
-/* Destruction primitives. */
-
-static void ipq_put(struct ipq *ipq)
-{
-	inet_frag_put(&ipq->q);
-}
-
-/* Kill ipq entry. It is not destroyed immediately,
- * because caller (and someone more) holds reference count.
- */
-static void ipq_kill(struct ipq *ipq)
-{
-	inet_frag_kill(&ipq->q);
-}
-
 static bool frag_expire_skip_icmp(u32 user)
 {
 	return user == IP_DEFRAG_AF_PACKET ||
@@ -134,23 +122,30 @@ static bool frag_expire_skip_icmp(u32 user)
  */
 static void ip_expire(struct timer_list *t)
 {
-	struct inet_frag_queue *frag = from_timer(frag, t, timer);
+	enum skb_drop_reason reason = SKB_DROP_REASON_FRAG_REASM_TIMEOUT;
+	struct inet_frag_queue *frag = timer_container_of(frag, t, timer);
 	const struct iphdr *iph;
 	struct sk_buff *head = NULL;
 	struct net *net;
 	struct ipq *qp;
-	int err;
+	int refs = 1;
 
 	qp = container_of(frag, struct ipq, q);
-	net = container_of(qp->q.net, struct net, ipv4.frags);
+	net = qp->q.fqdir->net;
 
 	rcu_read_lock();
+
+	/* Paired with WRITE_ONCE() in fqdir_pre_exit(). */
+	if (READ_ONCE(qp->q.fqdir->dead))
+		goto out_rcu_unlock;
+
 	spin_lock(&qp->q.lock);
 
 	if (qp->q.flags & INET_FRAG_COMPLETE)
 		goto out;
 
-	ipq_kill(qp);
+	qp->q.flags |= INET_FRAG_DROP;
+	inet_frag_kill(&qp->q, &refs);
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMFAILS);
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMTIMEOUT);
 
@@ -171,14 +166,15 @@ static void ip_expire(struct timer_list *t)
 
 	/* skb has no dst, perform route lookup again */
 	iph = ip_hdr(head);
-	err = ip_route_input_noref(head, iph->daddr, iph->saddr,
-					   iph->tos, head->dev);
-	if (err)
+	reason = ip_route_input_noref(head, iph->daddr, iph->saddr,
+				      ip4h_dscp(iph), head->dev);
+	if (reason)
 		goto out;
 
 	/* Only an end host needs to send an ICMP
 	 * "Fragment Reassembly Timeout" message, per RFC792.
 	 */
+	reason = SKB_DROP_REASON_FRAG_REASM_TIMEOUT;
 	if (frag_expire_skip_icmp(qp->q.key.v4.user) &&
 	    (skb_rtable(head)->rt_type != RTN_LOCAL))
 		goto out;
@@ -191,8 +187,8 @@ out:
 	spin_unlock(&qp->q.lock);
 out_rcu_unlock:
 	rcu_read_unlock();
-	kfree_skb(head);
-	ipq_put(qp);
+	kfree_skb_reason(head, reason);
+	inet_frag_putn(&qp->q, refs);
 }
 
 /* Find the correct entry in the "incomplete datagrams" queue for
@@ -211,7 +207,7 @@ static struct ipq *ip_find(struct net *net, struct iphdr *iph,
 	};
 	struct inet_frag_queue *q;
 
-	q = inet_frag_find(&net->ipv4.frags, &key);
+	q = inet_frag_find(net->ipv4.fqdir, &key);
 	if (!q)
 		return NULL;
 
@@ -222,7 +218,7 @@ static struct ipq *ip_find(struct net *net, struct iphdr *iph,
 static int ip_frag_too_far(struct ipq *qp)
 {
 	struct inet_peer *peer = qp->peer;
-	unsigned int max = qp->q.net->max_dist;
+	unsigned int max = qp->q.fqdir->max_dist;
 	unsigned int start, end;
 
 	int rc;
@@ -236,12 +232,8 @@ static int ip_frag_too_far(struct ipq *qp)
 
 	rc = qp->q.fragments_tail && (end - start) > max;
 
-	if (rc) {
-		struct net *net;
-
-		net = container_of(qp->q.net, struct net, ipv4.frags);
-		__IP_INC_STATS(net, IPSTATS_MIB_REASMFAILS);
-	}
+	if (rc)
+		__IP_INC_STATS(qp->q.fqdir->net, IPSTATS_MIB_REASMFAILS);
 
 	return rc;
 }
@@ -250,18 +242,18 @@ static int ip_frag_reinit(struct ipq *qp)
 {
 	unsigned int sum_truesize = 0;
 
-	if (!mod_timer(&qp->q.timer, jiffies + qp->q.net->timeout)) {
+	if (!mod_timer(&qp->q.timer, jiffies + qp->q.fqdir->timeout)) {
 		refcount_inc(&qp->q.refcnt);
 		return -ETIMEDOUT;
 	}
 
-	sum_truesize = inet_frag_rbtree_purge(&qp->q.rb_fragments);
-	sub_frag_mem_limit(qp->q.net, sum_truesize);
+	sum_truesize = inet_frag_rbtree_purge(&qp->q.rb_fragments,
+					      SKB_DROP_REASON_FRAG_TOO_FAR);
+	sub_frag_mem_limit(qp->q.fqdir, sum_truesize);
 
 	qp->q.flags = 0;
 	qp->q.len = 0;
 	qp->q.meat = 0;
-	qp->q.fragments = NULL;
 	qp->q.rb_fragments = RB_ROOT;
 	qp->q.fragments_tail = NULL;
 	qp->q.last_run_head = NULL;
@@ -272,23 +264,27 @@ static int ip_frag_reinit(struct ipq *qp)
 }
 
 /* Add new segment to existing queue. */
-static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
+static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb, int *refs)
 {
-	struct net *net = container_of(qp->q.net, struct net, ipv4.frags);
+	struct net *net = qp->q.fqdir->net;
 	int ihl, end, flags, offset;
 	struct sk_buff *prev_tail;
 	struct net_device *dev;
 	unsigned int fragsize;
 	int err = -ENOENT;
+	SKB_DR(reason);
 	u8 ecn;
 
-	if (qp->q.flags & INET_FRAG_COMPLETE)
+	/* If reassembly is already done, @skb must be a duplicate frag. */
+	if (qp->q.flags & INET_FRAG_COMPLETE) {
+		SKB_DR_SET(reason, DUP_FRAG);
 		goto err;
+	}
 
 	if (!(IPCB(skb)->flags & IPSKB_FRAG_COMPLETE) &&
 	    unlikely(ip_frag_too_far(qp)) &&
 	    unlikely(err = ip_frag_reinit(qp))) {
-		ipq_kill(qp);
+		inet_frag_kill(&qp->q, refs);
 		goto err;
 	}
 
@@ -351,9 +347,10 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 		qp->iif = dev->ifindex;
 
 	qp->q.stamp = skb->tstamp;
+	qp->q.tstamp_type = skb->tstamp_type;
 	qp->q.meat += skb->len;
 	qp->ecn |= ecn;
-	add_frag_mem_limit(qp->q.net, skb->truesize);
+	add_frag_mem_limit(qp->q.fqdir, skb->truesize);
 	if (offset == 0)
 		qp->q.flags |= INET_FRAG_FIRST_IN;
 
@@ -371,42 +368,50 @@ static int ip_frag_queue(struct ipq *qp, struct sk_buff *skb)
 		unsigned long orefdst = skb->_skb_refdst;
 
 		skb->_skb_refdst = 0UL;
-		err = ip_frag_reasm(qp, skb, prev_tail, dev);
+		err = ip_frag_reasm(qp, skb, prev_tail, dev, refs);
 		skb->_skb_refdst = orefdst;
 		if (err)
-			inet_frag_kill(&qp->q);
+			inet_frag_kill(&qp->q, refs);
 		return err;
 	}
 
 	skb_dst_drop(skb);
+	skb_orphan(skb);
 	return -EINPROGRESS;
 
 insert_error:
 	if (err == IPFRAG_DUP) {
-		kfree_skb(skb);
-		return -EINVAL;
+		SKB_DR_SET(reason, DUP_FRAG);
+		err = -EINVAL;
+		goto err;
 	}
 	err = -EINVAL;
 	__IP_INC_STATS(net, IPSTATS_MIB_REASM_OVERLAPS);
 discard_qp:
-	inet_frag_kill(&qp->q);
+	inet_frag_kill(&qp->q, refs);
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMFAILS);
 err:
-	kfree_skb(skb);
+	kfree_skb_reason(skb, reason);
 	return err;
 }
 
+static bool ip_frag_coalesce_ok(const struct ipq *qp)
+{
+	return qp->q.key.v4.user == IP_DEFRAG_LOCAL_DELIVER;
+}
+
 /* Build a new IP datagram from all its fragments. */
 static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb,
-			 struct sk_buff *prev_tail, struct net_device *dev)
+			 struct sk_buff *prev_tail, struct net_device *dev,
+			 int *refs)
 {
-	struct net *net = container_of(qp->q.net, struct net, ipv4.frags);
+	struct net *net = qp->q.fqdir->net;
 	struct iphdr *iph;
 	void *reasm_data;
 	int len, err;
 	u8 ecn;
 
-	ipq_kill(qp);
+	inet_frag_kill(&qp->q, refs);
 
 	ecn = ip_frag_ecn_table[qp->ecn];
 	if (unlikely(ecn == 0xff)) {
@@ -424,7 +429,8 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb,
 	if (len > 65535)
 		goto out_oversize;
 
-	inet_frag_reasm_finish(&qp->q, skb, reasm_data);
+	inet_frag_reasm_finish(&qp->q, skb, reasm_data,
+			       ip_frag_coalesce_ok(qp));
 
 	skb->dev = dev;
 	IPCB(skb)->frag_max_size = max(qp->max_df_size, qp->q.max_size);
@@ -451,7 +457,6 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *skb,
 	ip_send_check(iph);
 
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMOKS);
-	qp->q.fragments = NULL;
 	qp->q.rb_fragments = RB_ROOT;
 	qp->q.fragments_tail = NULL;
 	qp->q.last_run_head = NULL;
@@ -471,26 +476,30 @@ out_fail:
 /* Process an incoming IP datagram fragment. */
 int ip_defrag(struct net *net, struct sk_buff *skb, u32 user)
 {
-	struct net_device *dev = skb->dev ? : skb_dst(skb)->dev;
-	int vif = l3mdev_master_ifindex_rcu(dev);
+	struct net_device *dev;
 	struct ipq *qp;
+	int vif;
 
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMREQDS);
-	skb_orphan(skb);
 
 	/* Lookup (or create) queue header */
+	rcu_read_lock();
+	dev = skb->dev ? : skb_dst_dev_rcu(skb);
+	vif = l3mdev_master_ifindex_rcu(dev);
 	qp = ip_find(net, ip_hdr(skb), user, vif);
 	if (qp) {
-		int ret;
+		int ret, refs = 0;
 
 		spin_lock(&qp->q.lock);
 
-		ret = ip_frag_queue(qp, skb);
+		ret = ip_frag_queue(qp, skb, &refs);
 
 		spin_unlock(&qp->q.lock);
-		ipq_put(qp);
+		rcu_read_unlock();
+		inet_frag_putn(&qp->q, refs);
 		return ret;
 	}
+	rcu_read_unlock();
 
 	__IP_INC_STATS(net, IPSTATS_MIB_REASMFAILS);
 	kfree_skb(skb);
@@ -546,36 +555,29 @@ static int dist_min;
 static struct ctl_table ip4_frags_ns_ctl_table[] = {
 	{
 		.procname	= "ipfrag_high_thresh",
-		.data		= &init_net.ipv4.frags.high_thresh,
 		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
 		.proc_handler	= proc_doulongvec_minmax,
-		.extra1		= &init_net.ipv4.frags.low_thresh
 	},
 	{
 		.procname	= "ipfrag_low_thresh",
-		.data		= &init_net.ipv4.frags.low_thresh,
 		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
 		.proc_handler	= proc_doulongvec_minmax,
-		.extra2		= &init_net.ipv4.frags.high_thresh
 	},
 	{
 		.procname	= "ipfrag_time",
-		.data		= &init_net.ipv4.frags.timeout,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
 	{
 		.procname	= "ipfrag_max_dist",
-		.data		= &init_net.ipv4.frags.max_dist,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= &dist_min,
 	},
-	{ }
 };
 
 /* secret interval has been deprecated */
@@ -588,7 +590,6 @@ static struct ctl_table ip4_frags_ctl_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec_jiffies,
 	},
-	{ }
 };
 
 static int __net_init ip4_frags_ns_ctl_register(struct net *net)
@@ -602,15 +603,16 @@ static int __net_init ip4_frags_ns_ctl_register(struct net *net)
 		if (!table)
 			goto err_alloc;
 
-		table[0].data = &net->ipv4.frags.high_thresh;
-		table[0].extra1 = &net->ipv4.frags.low_thresh;
-		table[1].data = &net->ipv4.frags.low_thresh;
-		table[1].extra2 = &net->ipv4.frags.high_thresh;
-		table[2].data = &net->ipv4.frags.timeout;
-		table[3].data = &net->ipv4.frags.max_dist;
 	}
-
-	hdr = register_net_sysctl(net, "net/ipv4", table);
+	table[0].data	= &net->ipv4.fqdir->high_thresh;
+	table[0].extra1	= &net->ipv4.fqdir->low_thresh;
+	table[1].data	= &net->ipv4.fqdir->low_thresh;
+	table[1].extra2	= &net->ipv4.fqdir->high_thresh;
+	table[2].data	= &net->ipv4.fqdir->timeout;
+	table[3].data	= &net->ipv4.fqdir->max_dist;
+
+	hdr = register_net_sysctl_sz(net, "net/ipv4", table,
+				     ARRAY_SIZE(ip4_frags_ns_ctl_table));
 	if (!hdr)
 		goto err_reg;
 
@@ -626,7 +628,7 @@ err_alloc:
 
 static void __net_exit ip4_frags_ns_ctl_unregister(struct net *net)
 {
-	struct ctl_table *table;
+	const struct ctl_table *table;
 
 	table = net->ipv4.frags_hdr->ctl_table_arg;
 	unregister_net_sysctl_table(net->ipv4.frags_hdr);
@@ -656,6 +658,9 @@ static int __net_init ipv4_frags_init_net(struct net *net)
 {
 	int res;
 
+	res = fqdir_init(&net->ipv4.fqdir, &ip4_frags, net);
+	if (res < 0)
+		return res;
 	/* Fragment cache limits.
 	 *
 	 * The fragment memory accounting code, (tries to) account for
@@ -670,36 +675,38 @@ static int __net_init ipv4_frags_init_net(struct net *net)
 	 * we will prune down to 3MB, making room for approx 8 big 64K
 	 * fragments 8x128k.
 	 */
-	net->ipv4.frags.high_thresh = 4 * 1024 * 1024;
-	net->ipv4.frags.low_thresh  = 3 * 1024 * 1024;
+	net->ipv4.fqdir->high_thresh = 4 * 1024 * 1024;
+	net->ipv4.fqdir->low_thresh  = 3 * 1024 * 1024;
 	/*
 	 * Important NOTE! Fragment queue must be destroyed before MSL expires.
 	 * RFC791 is wrong proposing to prolongate timer each fragment arrival
 	 * by TTL.
 	 */
-	net->ipv4.frags.timeout = IP_FRAG_TIME;
+	net->ipv4.fqdir->timeout = IP_FRAG_TIME;
 
-	net->ipv4.frags.max_dist = 64;
-	net->ipv4.frags.f = &ip4_frags;
+	net->ipv4.fqdir->max_dist = 64;
 
-	res = inet_frags_init_net(&net->ipv4.frags);
-	if (res < 0)
-		return res;
 	res = ip4_frags_ns_ctl_register(net);
 	if (res < 0)
-		inet_frags_exit_net(&net->ipv4.frags);
+		fqdir_exit(net->ipv4.fqdir);
 	return res;
 }
 
+static void __net_exit ipv4_frags_pre_exit_net(struct net *net)
+{
+	fqdir_pre_exit(net->ipv4.fqdir);
+}
+
 static void __net_exit ipv4_frags_exit_net(struct net *net)
 {
 	ip4_frags_ns_ctl_unregister(net);
-	inet_frags_exit_net(&net->ipv4.frags);
+	fqdir_exit(net->ipv4.fqdir);
 }
 
 static struct pernet_operations ip4_frags_ops = {
-	.init = ipv4_frags_init_net,
-	.exit = ipv4_frags_exit_net,
+	.init		= ipv4_frags_init_net,
+	.pre_exit	= ipv4_frags_pre_exit_net,
+	.exit		= ipv4_frags_exit_net,
 };