6lowpan: handling 6lowpan fragmentation via inet_frag api

This patch drops the current way of 6lowpan fragmentation on receiving
side and replace it with a implementation which use the inet_frag api.
The old fragmentation handling has some race conditions and isn't
rfc4944 compatible. Also adding support to match fragments on
destination address, source address, tag value and datagram_size
which is missing in the current implementation.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/net/netns/ieee802154_6lowpan.h

@@ -7,7 +7,16 @@
 #ifndef __NETNS_IEEE802154_6LOWPAN_H__
 #define __NETNS_IEEE802154_6LOWPAN_H__
 
+struct netns_sysctl_lowpan {
+#ifdef CONFIG_SYSCTL
+	struct ctl_table_header *frags_hdr;
+#endif
+};
+
 struct netns_ieee802154_lowpan {
+	struct netns_sysctl_lowpan sysctl;
+	struct netns_frags	frags;
+	u16			max_dsize;
 };
 
 #endif

net/ieee802154/6lowpan_rtnl.c

@@ -54,6 +54,7 @@
 #include <net/ieee802154_netdev.h>
 #include <net/ipv6.h>
 
+#include "reassembly.h"
 #include "6lowpan.h"
 
 static LIST_HEAD(lowpan_devices);
@@ -70,18 +71,6 @@ struct lowpan_dev_record {
 	struct list_head list;
 };
 
-struct lowpan_fragment {
-	struct sk_buff		*skb;		/* skb to be assembled */
-	u16			length;		/* length to be assemled */
-	u32			bytes_rcv;	/* bytes received */
-	u16			tag;		/* current fragment tag */
-	struct timer_list	timer;		/* assembling timer */
-	struct list_head	list;		/* fragments list */
-};
-
-static LIST_HEAD(lowpan_fragments);
-static DEFINE_SPINLOCK(flist_lock);
-
 static inline struct
 lowpan_dev_info *lowpan_dev_info(const struct net_device *dev)
 {
@@ -179,69 +168,6 @@ static int lowpan_give_skb_to_devices(struct sk_buff *skb,
 	return stat;
 }
 
-static void lowpan_fragment_timer_expired(unsigned long entry_addr)
-{
-	struct lowpan_fragment *entry = (struct lowpan_fragment *)entry_addr;
-
-	pr_debug("timer expired for frame with tag %d\n", entry->tag);
-
-	list_del(&entry->list);
-	dev_kfree_skb(entry->skb);
-	kfree(entry);
-}
-
-static struct lowpan_fragment *
-lowpan_alloc_new_frame(struct sk_buff *skb, u16 len, u16 tag)
-{
-	struct lowpan_fragment *frame;
-
-	frame = kzalloc(sizeof(struct lowpan_fragment),
-			GFP_ATOMIC);
-	if (!frame)
-		goto frame_err;
-
-	INIT_LIST_HEAD(&frame->list);
-
-	frame->length = len;
-	frame->tag = tag;
-
-	/* allocate buffer for frame assembling */
-	frame->skb = netdev_alloc_skb_ip_align(skb->dev, frame->length +
-					       sizeof(struct ipv6hdr));
-
-	if (!frame->skb)
-		goto skb_err;
-
-	frame->skb->priority = skb->priority;
-
-	/* reserve headroom for uncompressed ipv6 header */
-	skb_reserve(frame->skb, sizeof(struct ipv6hdr));
-	skb_put(frame->skb, frame->length);
-
-	/* copy the first control block to keep a
-	 * trace of the link-layer addresses in case
-	 * of a link-local compressed address
-	 */
-	memcpy(frame->skb->cb, skb->cb, sizeof(skb->cb));
-
-	init_timer(&frame->timer);
-	/* time out is the same as for ipv6 - 60 sec */
-	frame->timer.expires = jiffies + LOWPAN_FRAG_TIMEOUT;
-	frame->timer.data = (unsigned long)frame;
-	frame->timer.function = lowpan_fragment_timer_expired;
-
-	add_timer(&frame->timer);
-
-	list_add_tail(&frame->list, &lowpan_fragments);
-
-	return frame;
-
-skb_err:
-	kfree(frame);
-frame_err:
-	return NULL;
-}
-
 static int process_data(struct sk_buff *skb)
 {
 	u8 iphc0, iphc1;
@@ -255,94 +181,6 @@ static int process_data(struct sk_buff *skb)
 	if (lowpan_fetch_skb_u8(skb, &iphc0))
 		goto drop;
 
-	/* fragments assembling */
-	switch (iphc0 & LOWPAN_DISPATCH_MASK) {
-	case LOWPAN_DISPATCH_FRAG1:
-	case LOWPAN_DISPATCH_FRAGN:
-	{
-		struct lowpan_fragment *frame;
-		/* slen stores the rightmost 8 bits of the 11 bits length */
-		u8 slen, offset = 0;
-		u16 len, tag;
-		bool found = false;
-
-		if (lowpan_fetch_skb_u8(skb, &slen) || /* frame length */
-		    lowpan_fetch_skb_u16(skb, &tag))  /* fragment tag */
-			goto drop;
-
-		/* adds the 3 MSB to the 8 LSB to retrieve the 11 bits length */
-		len = ((iphc0 & 7) << 8) | slen;
-
-		if ((iphc0 & LOWPAN_DISPATCH_MASK) == LOWPAN_DISPATCH_FRAG1) {
-			pr_debug("%s received a FRAG1 packet (tag: %d, "
-				 "size of the entire IP packet: %d)",
-				 __func__, tag, len);
-		} else { /* FRAGN */
-			if (lowpan_fetch_skb_u8(skb, &offset))
-				goto unlock_and_drop;
-			pr_debug("%s received a FRAGN packet (tag: %d, "
-				 "size of the entire IP packet: %d, "
-				 "offset: %d)", __func__, tag, len, offset * 8);
-		}
-
-		/*
-		 * check if frame assembling with the same tag is
-		 * already in progress
-		 */
-		spin_lock_bh(&flist_lock);
-
-		list_for_each_entry(frame, &lowpan_fragments, list)
-			if (frame->tag == tag) {
-				found = true;
-				break;
-			}
-
-		/* alloc new frame structure */
-		if (!found) {
-			pr_debug("%s first fragment received for tag %d, "
-				 "begin packet reassembly", __func__, tag);
-			frame = lowpan_alloc_new_frame(skb, len, tag);
-			if (!frame)
-				goto unlock_and_drop;
-		}
-
-		/* if payload fits buffer, copy it */
-		if (likely((offset * 8 + skb->len) <= frame->length))
-			skb_copy_to_linear_data_offset(frame->skb, offset * 8,
-							skb->data, skb->len);
-		else
-			goto unlock_and_drop;
-
-		frame->bytes_rcv += skb->len;
-
-		/* frame assembling complete */
-		if ((frame->bytes_rcv == frame->length) &&
-		     frame->timer.expires > jiffies) {
-			/* if timer haven't expired - first of all delete it */
-			del_timer_sync(&frame->timer);
-			list_del(&frame->list);
-			spin_unlock_bh(&flist_lock);
-
-			pr_debug("%s successfully reassembled fragment "
-				 "(tag %d)", __func__, tag);
-
-			dev_kfree_skb(skb);
-			skb = frame->skb;
-			kfree(frame);
-
-			if (lowpan_fetch_skb_u8(skb, &iphc0))
-				goto drop;
-
-			break;
-		}
-		spin_unlock_bh(&flist_lock);
-
-		return kfree_skb(skb), 0;
-	}
-	default:
-		break;
-	}
-
 	if (lowpan_fetch_skb_u8(skb, &iphc1))
 		goto drop;
 
@@ -355,8 +193,6 @@ static int process_data(struct sk_buff *skb)
 				IEEE802154_ADDR_LEN, iphc0, iphc1,
 				lowpan_give_skb_to_devices);
 
-unlock_and_drop:
-	spin_unlock_bh(&flist_lock);
 drop:
 	kfree_skb(skb);
 	return -EINVAL;
@@ -603,44 +439,53 @@ static int lowpan_rcv(struct sk_buff *skb, struct net_device *dev,
 	struct packet_type *pt, struct net_device *orig_dev)
 {
 	struct sk_buff *local_skb;
+	int ret;
 
 	if (!netif_running(dev))
-		goto drop;
+		goto drop_skb;
 
 	if (dev->type != ARPHRD_IEEE802154)
-		goto drop;
+		goto drop_skb;
+
+	local_skb = skb_clone(skb, GFP_ATOMIC);
+	if (!local_skb)
+		goto drop_skb;
+
+	kfree_skb(skb);
 
 	/* check that it's our buffer */
 	if (skb->data[0] == LOWPAN_DISPATCH_IPV6) {
-		/* Copy the packet so that the IPv6 header is
-		 * properly aligned.
-		 */
-		local_skb = skb_copy_expand(skb, NET_SKB_PAD - 1,
-					    skb_tailroom(skb), GFP_ATOMIC);
-		if (!local_skb)
-			goto drop;
-
 		local_skb->protocol = htons(ETH_P_IPV6);
 		local_skb->pkt_type = PACKET_HOST;
 
 		/* Pull off the 1-byte of 6lowpan header. */
 		skb_pull(local_skb, 1);
 
-		lowpan_give_skb_to_devices(local_skb, NULL);
-
-		kfree_skb(local_skb);
-		kfree_skb(skb);
+		ret = lowpan_give_skb_to_devices(local_skb, NULL);
+		if (ret == NET_RX_DROP)
+			goto drop;
 	} else {
 		switch (skb->data[0] & 0xe0) {
 		case LOWPAN_DISPATCH_IPHC:	/* ipv6 datagram */
-		case LOWPAN_DISPATCH_FRAG1:	/* first fragment header */
-		case LOWPAN_DISPATCH_FRAGN:	/* next fragments headers */
-			local_skb = skb_clone(skb, GFP_ATOMIC);
-			if (!local_skb)
+			ret = process_data(local_skb);
+			if (ret == NET_RX_DROP)
 				goto drop;
-			process_data(local_skb);
-
-			kfree_skb(skb);
+			break;
+		case LOWPAN_DISPATCH_FRAG1:	/* first fragment header */
+			ret = lowpan_frag_rcv(local_skb, LOWPAN_DISPATCH_FRAG1);
+			if (ret == 1) {
+				ret = process_data(local_skb);
+				if (ret == NET_RX_DROP)
+					goto drop;
+			}
+			break;
+		case LOWPAN_DISPATCH_FRAGN:	/* next fragments headers */
+			ret = lowpan_frag_rcv(local_skb, LOWPAN_DISPATCH_FRAGN);
+			if (ret == 1) {
+				ret = process_data(local_skb);
+				if (ret == NET_RX_DROP)
+					goto drop;
+			}
 			break;
 		default:
 			break;
@@ -648,9 +493,9 @@ static int lowpan_rcv(struct sk_buff *skb, struct net_device *dev,
 	}
 
 	return NET_RX_SUCCESS;
-
-drop:
+drop_skb:
 	kfree_skb(skb);
+drop:
 	return NET_RX_DROP;
 }
 
@@ -778,43 +623,40 @@ static int __init lowpan_init_module(void)
 {
 	int err = 0;
 
-	err = lowpan_netlink_init();
+	err = lowpan_net_frag_init();
 	if (err < 0)
 		goto out;
 
+	err = lowpan_netlink_init();
+	if (err < 0)
+		goto out_frag;
+
 	dev_add_pack(&lowpan_packet_type);
 
 	err = register_netdevice_notifier(&lowpan_dev_notifier);
-	if (err < 0) {
-		dev_remove_pack(&lowpan_packet_type);
-		lowpan_netlink_fini();
-	}
+	if (err < 0)
+		goto out_pack;
+
+	return 0;
+
+out_pack:
+	dev_remove_pack(&lowpan_packet_type);
+	lowpan_netlink_fini();
+out_frag:
+	lowpan_net_frag_exit();
 out:
 	return err;
 }
 
 static void __exit lowpan_cleanup_module(void)
 {
-	struct lowpan_fragment *frame, *tframe;
-
 	lowpan_netlink_fini();
 
 	dev_remove_pack(&lowpan_packet_type);
 
-	unregister_netdevice_notifier(&lowpan_dev_notifier);
+	lowpan_net_frag_exit();
 
-	/* Now 6lowpan packet_type is removed, so no new fragments are
-	 * expected on RX, therefore that's the time to clean incomplete
-	 * fragments.
-	 */
-	spin_lock_bh(&flist_lock);
-	list_for_each_entry_safe(frame, tframe, &lowpan_fragments, list) {
-		del_timer_sync(&frame->timer);
-		list_del(&frame->list);
-		dev_kfree_skb(frame->skb);
-		kfree(frame);
-	}
-	spin_unlock_bh(&flist_lock);
+	unregister_netdevice_notifier(&lowpan_dev_notifier);
 }
 
 module_init(lowpan_init_module);

net/ieee802154/Makefile

@@ -2,6 +2,6 @@ obj-$(CONFIG_IEEE802154) += ieee802154.o af_802154.o
 obj-$(CONFIG_IEEE802154_6LOWPAN) += 6lowpan.o
 obj-$(CONFIG_6LOWPAN_IPHC) += 6lowpan_iphc.o
 
-6lowpan-y := 6lowpan_rtnl.o
+6lowpan-y := 6lowpan_rtnl.o reassembly.o
 ieee802154-y := netlink.o nl-mac.o nl-phy.o nl_policy.o wpan-class.o
 af_802154-y := af_ieee802154.o raw.o dgram.o

net/ieee802154/reassembly.h

@@ -0,0 +1,66 @@
+#ifndef __IEEE802154_6LOWPAN_REASSEMBLY_H__
+#define __IEEE802154_6LOWPAN_REASSEMBLY_H__
+
+#include <net/inet_frag.h>
+
+struct lowpan_create_arg {
+	__be16 tag;
+	u16 d_size;
+	const struct ieee802154_addr *src;
+	const struct ieee802154_addr *dst;
+};
+
+/* Equivalent of ipv4 struct ip
+ */
+struct lowpan_frag_queue {
+	struct inet_frag_queue	q;
+
+	__be16			tag;
+	u16			d_size;
+	struct ieee802154_addr	saddr;
+	struct ieee802154_addr	daddr;
+};
+
+static inline u32 ieee802154_addr_hash(const struct ieee802154_addr *a)
+{
+	switch (a->addr_type) {
+	case IEEE802154_ADDR_LONG:
+		return (__force u32)((((u32 *)a->hwaddr))[0] ^
+				      ((u32 *)(a->hwaddr))[1]);
+	case IEEE802154_ADDR_SHORT:
+		return (__force u32)(a->short_addr);
+	default:
+		return 0;
+	}
+}
+
+static inline bool ieee802154_addr_addr_equal(const struct ieee802154_addr *a1,
+				   const struct ieee802154_addr *a2)
+{
+	if (a1->pan_id != a2->pan_id)
+		return false;
+
+	if (a1->addr_type != a2->addr_type)
+		return false;
+
+	switch (a1->addr_type) {
+	case IEEE802154_ADDR_LONG:
+		if (memcmp(a1->hwaddr, a2->hwaddr, IEEE802154_ADDR_LEN))
+			return false;
+		break;
+	case IEEE802154_ADDR_SHORT:
+		if (a1->short_addr != a2->short_addr)
+			return false;
+		break;
+	default:
+		return false;
+	}
+
+	return true;
+}
+
+int lowpan_frag_rcv(struct sk_buff *skb, const u8 frag_type);
+void lowpan_net_frag_exit(void);
+int lowpan_net_frag_init(void);
+
+#endif /* __IEEE802154_6LOWPAN_REASSEMBLY_H__ */