[PATCH net-next 0/6] 6lowpan: reimplementation of fragmentation handling

[PATCH net-next 0/6] 6lowpan: reimplementation of fragmentation handling

[Alexander Aring]

Hi,

this patch series reimplementation the fragmentation handling of 6lowpan
accroding to rfc4944 [1].

The first big note is, that the current fragmentation behaviour isn't rfc
complaint. The main issue is a wrong datagram_size value which needs to be:
datagram_size = ipv6_payload + ipv6 header + (maybe compressed transport header,
                currently only udp is supported)

but the current datagram_size value is calculated as:
datagram_size = ipv6_payload

Fragmentation work in a linux<->linux communication only.


Why reimplementation?

I reimplemted the reassembly side only. The current behaviour is to allocate a
skb with the reassembled size and hold all fragments in a list, protected by a
spinlock. After we received all fragments (detected by the sum of all fragments,
it begins to place all fragments into the allocated skb).

This reassembly implementation has some race condition. Additional I make it more
rfc complaint. The current implementation match on the tag value inside the frag
header only, but rfc4944 says we need to match on dst addr(mac), src addr(mac),
tag value, datagram_size value. [2]

The new reassembly handling use the inet_frag api (I mean the callback interface
of ipv6 and ipv4 reassembly). I looked into ipv6 and wanted to see how ipv6 is
dealing with reassembly, so I based my code on this implementation.



On the sending side to generate the fragments I improved the current code to use
the nearest 8 divided payload. (We can do that, because the mac layer has a
dynamic size, so it depends on mac_header how big we can do the payload).

Of course I fix also the reassembly/sending side to be rfc complaint now.

Regards
Alexander Aring

[1] http://tools.ietf.org/html/rfc4944
[2] http://tools.ietf.org/html/rfc4944#section-5.3

Alexander Aring (6):
  6lowpan: add frag information struct
  6lowpan: fix fragmentation on sending side
  6lowpan: move 6lowpan.c to 6lowpan_rtnl.c
  6lowpan: fix some checkpatch issues
  net: ns: add ieee802154_6lowpan namespace
  6lowpan: handling 6lowpan fragmentation via inet_frag api

 include/net/ieee802154_netdev.h              |   7 +
 include/net/net_namespace.h                  |   4 +
 include/net/netns/ieee802154_6lowpan.h       |  22 ++
 net/ieee802154/6lowpan.h                     | 112 ++++++
 net/ieee802154/{6lowpan.c => 6lowpan_rtnl.c} | 327 +++++-----------
 net/ieee802154/Makefile                      |   1 +
 net/ieee802154/reassembly.c                  | 565 +++++++++++++++++++++++++++
 net/ieee802154/reassembly.h                  |  66 ++++
 8 files changed, 869 insertions(+), 235 deletions(-)
 create mode 100644 include/net/netns/ieee802154_6lowpan.h
 rename net/ieee802154/{6lowpan.c => 6lowpan_rtnl.c} (70%)
 create mode 100644 net/ieee802154/reassembly.c
 create mode 100644 net/ieee802154/reassembly.h

-- 
1.9.0

[PATCH net-next 1/6] 6lowpan: add frag information struct

[Alexander Aring]

This patch adds a 6lowpan fragmentation struct into cb of skb which
is necessary to hold fragmentation information.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Tested-by: Martin Townsend <martin.townsend@xsilon.com>
---
 include/net/ieee802154_netdev.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/include/net/ieee802154_netdev.h b/include/net/ieee802154_netdev.h
index 8196d5d..97b2e34 100644
--- a/include/net/ieee802154_netdev.h
+++ b/include/net/ieee802154_netdev.h
@@ -29,6 +29,12 @@
 
 #include <net/af_ieee802154.h>
 
+struct ieee802154_frag_info {
+	__be16 d_tag;
+	u16 d_size;
+	u8 d_offset;
+};
+
 /*
  * A control block of skb passed between the ARPHRD_IEEE802154 device
  * and other stack parts.
@@ -39,6 +45,7 @@ struct ieee802154_mac_cb {
 	struct ieee802154_addr da;
 	u8 flags;
 	u8 seq;
+	struct ieee802154_frag_info frag_info;
 };
 
 static inline struct ieee802154_mac_cb *mac_cb(struct sk_buff *skb)
-- 
1.9.0

[PATCH net-next 2/6] 6lowpan: fix fragmentation on sending side

[Alexander Aring]

This patch fix the fragmentation on sending side according to rfc4944.

Also add improvement to use the full payload of a PDU which calculate
the nearest divided to 8 payload length for the fragmentation datagram
size attribute.

The main issue is that the datagram size of fragmentation header use the
ipv6 payload length, but rfc4944 says it's the ipv6 payload length inclusive
network header size (and transport header size if compressed).

Signed-off-by: Alexander Aring <alex.aring-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Tested-by: Martin Townsend <martin.townsend-I0Qfutc96jjQT0dZR+AlfA@public.gmane.org>
---
 net/ieee802154/6lowpan.c | 44 ++++++++++++++++++++++++++++++++++----------
 1 file changed, 34 insertions(+), 10 deletions(-)

diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c
index 8edfea5..0fe0173 100644
--- a/net/ieee802154/6lowpan.c
+++ b/net/ieee802154/6lowpan.c
@@ -106,6 +106,7 @@ static int lowpan_header_create(struct sk_buff *skb,
 			   unsigned short type, const void *_daddr,
 			   const void *_saddr, unsigned int len)
 {
+	struct ipv6hdr *hdr;
 	const u8 *saddr = _saddr;
 	const u8 *daddr = _daddr;
 	struct ieee802154_addr sa, da;
@@ -116,12 +117,20 @@ static int lowpan_header_create(struct sk_buff *skb,
 	if (type != ETH_P_IPV6)
 		return 0;
 
+	hdr = ipv6_hdr(skb);
+
 	if (!saddr)
 		saddr = dev->dev_addr;
 
 	raw_dump_inline(__func__, "saddr", (unsigned char *)saddr, 8);
 	raw_dump_inline(__func__, "daddr", (unsigned char *)daddr, 8);
 
+	/* Before replace ipv6hdr to 6lowpan header we save the dgram_size */
+	mac_cb(skb)->frag_info.d_size = len;
+	mac_cb(skb)->frag_info.d_offset = sizeof(struct ipv6hdr);
+	if (hdr->nexthdr == UIP_PROTO_UDP)
+		mac_cb(skb)->frag_info.d_offset += sizeof(struct udphdr);
+
 	lowpan_header_compress(skb, dev, type, daddr, saddr, len);
 
 	/*
@@ -422,44 +431,58 @@ lowpan_fragment_xmit(struct sk_buff *skb, u8 *head,
 static int
 lowpan_skb_fragmentation(struct sk_buff *skb, struct net_device *dev)
 {
-	int  err, header_length, payload_length, tag, offset = 0;
+	int err, header_length, payload_length, tag, dgram_size,
+	    dgram_offset, lowpan_size, frag_plen, offset = 0;
 	u8 head[5];
 
 	header_length = skb->mac_len;
 	payload_length = skb->len - header_length;
 	tag = lowpan_dev_info(dev)->fragment_tag++;
+	lowpan_size = skb_network_header_len(skb);
+	dgram_size = mac_cb(skb)->frag_info.d_size;
 
 	/* first fragment header */
-	head[0] = LOWPAN_DISPATCH_FRAG1 | ((payload_length >> 8) & 0x7);
-	head[1] = payload_length & 0xff;
+	head[0] = LOWPAN_DISPATCH_FRAG1 | ((dgram_size >> 8) & 0x7);
+	head[1] = dgram_size & 0xff;
 	head[2] = tag >> 8;
 	head[3] = tag & 0xff;
 
-	err = lowpan_fragment_xmit(skb, head, header_length, LOWPAN_FRAG_SIZE,
-				   0, LOWPAN_DISPATCH_FRAG1);
+	/* calc the nearest payload length(divided to 8) for first fragment
+	 * which fits into a IEEE802154_MTU
+	 */
+	frag_plen = round_down(IEEE802154_MTU - header_length -
+			       LOWPAN_FRAG1_HEAD_SIZE - lowpan_size -
+			       IEEE802154_MFR_SIZE, 8);
 
+	err = lowpan_fragment_xmit(skb, head, header_length,
+				   frag_plen + lowpan_size, 0,
+				   LOWPAN_DISPATCH_FRAG1);
 	if (err) {
 		pr_debug("%s unable to send FRAG1 packet (tag: %d)",
 			 __func__, tag);
 		goto exit;
 	}
 
-	offset = LOWPAN_FRAG_SIZE;
+	offset = lowpan_size + frag_plen;
+	dgram_offset = mac_cb(skb)->frag_info.d_offset + frag_plen;
 
 	/* next fragment header */
 	head[0] &= ~LOWPAN_DISPATCH_FRAG1;
 	head[0] |= LOWPAN_DISPATCH_FRAGN;
 
+	frag_plen = round_down(IEEE802154_MTU - header_length -
+			       LOWPAN_FRAGN_HEAD_SIZE - IEEE802154_MFR_SIZE, 8);
+
 	while (payload_length - offset > 0) {
-		int len = LOWPAN_FRAG_SIZE;
+		int len = frag_plen;
 
-		head[4] = offset / 8;
+		head[4] = dgram_offset >> 3;
 
 		if (payload_length - offset < len)
 			len = payload_length - offset;
 
-		err = lowpan_fragment_xmit(skb, head, header_length,
-					   len, offset, LOWPAN_DISPATCH_FRAGN);
+		err = lowpan_fragment_xmit(skb, head, header_length, len,
+					   offset, LOWPAN_DISPATCH_FRAGN);
 		if (err) {
 			pr_debug("%s unable to send a subsequent FRAGN packet "
 				 "(tag: %d, offset: %d", __func__, tag, offset);
@@ -467,6 +490,7 @@ lowpan_skb_fragmentation(struct sk_buff *skb, struct net_device *dev)
 		}
 
 		offset += len;
+		dgram_offset += len;
 	}
 
 exit:
-- 
1.9.0


------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk

[PATCH net-next 3/6] 6lowpan: move 6lowpan.c to 6lowpan_rtnl.c

[Alexander Aring]

We have a 6lowpan.c file and 6lowpan.ko file. To avoid confusing we
should move 6lowpan.c to 6lowpan.ko. Then we can support multiple source
files for 6lowpan module.

Signed-off-by: Alexander Aring <alex.aring@gmail.com>
Tested-by: Martin Townsend <martin.townsend@xsilon.com>
---
 net/ieee802154/{6lowpan.c => 6lowpan_rtnl.c} | 0
 net/ieee802154/Makefile                      | 1 +
 2 files changed, 1 insertion(+)
 rename net/ieee802154/{6lowpan.c => 6lowpan_rtnl.c} (100%)

diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan_rtnl.c
similarity index 100%
rename from net/ieee802154/6lowpan.c
rename to net/ieee802154/6lowpan_rtnl.c
diff --git a/net/ieee802154/Makefile b/net/ieee802154/Makefile
index e8f0588..3d08adf 100644
--- a/net/ieee802154/Makefile
+++ b/net/ieee802154/Makefile
@@ -2,5 +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
 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
-- 
1.9.0

[PATCH net-next 4/6] 6lowpan: fix some checkpatch issues

[Alexander Aring]

Detected with:

./scripts/checkpatch.pl --strict -f net/ieee802154/6lowpan_rtnl.c

Signed-off-by: Alexander Aring <alex.aring-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Tested-by: Martin Townsend <martin.townsend-I0Qfutc96jjQT0dZR+AlfA@public.gmane.org>
---
 net/ieee802154/6lowpan_rtnl.c | 27 +++++++++------------------
 1 file changed, 9 insertions(+), 18 deletions(-)

diff --git a/net/ieee802154/6lowpan_rtnl.c b/net/ieee802154/6lowpan_rtnl.c
index 0fe0173..b543488 100644
--- a/net/ieee802154/6lowpan_rtnl.c
+++ b/net/ieee802154/6lowpan_rtnl.c
@@ -1,10 +1,8 @@
-/*
- * Copyright 2011, Siemens AG
+/* Copyright 2011, Siemens AG
  * written by Alexander Smirnov <alex.bluesman.smirnov-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
  */
 
-/*
- * Based on patches from Jon Smirl <jonsmirl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
+/* Based on patches from Jon Smirl <jonsmirl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
  * Copyright (c) 2011 Jon Smirl <jonsmirl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
  *
  * This program is free software; you can redistribute it and/or modify
@@ -15,10 +13,6 @@
  * 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.
  */
 
 /* Jon's code is based on 6lowpan implementation for Contiki which is:
@@ -133,13 +127,11 @@ static int lowpan_header_create(struct sk_buff *skb,
 
 	lowpan_header_compress(skb, dev, type, daddr, saddr, len);
 
-	/*
-	 * NOTE1: I'm still unsure about the fact that compression and WPAN
+	/* NOTE1: I'm still unsure about the fact that compression and WPAN
 	 * header are created here and not later in the xmit. So wait for
 	 * an opinion of net maintainers.
 	 */
-	/*
-	 * NOTE2: to be absolutely correct, we must derive PANid information
+	/* NOTE2: to be absolutely correct, we must derive PANid information
 	 * from MAC subif of the 'dev' and 'real_dev' network devices, but
 	 * this isn't implemented in mainline yet, so currently we assign 0xff
 	 */
@@ -154,8 +146,7 @@ static int lowpan_header_create(struct sk_buff *skb,
 	/* intra-PAN communications */
 	da.pan_id = ieee802154_mlme_ops(dev)->get_pan_id(dev);
 
-	/*
-	 * if the destination address is the broadcast address, use the
+	/* if the destination address is the broadcast address, use the
 	 * corresponding short address
 	 */
 	if (lowpan_is_addr_broadcast(daddr)) {
@@ -395,7 +386,7 @@ static int lowpan_set_address(struct net_device *dev, void *p)
 
 static int
 lowpan_fragment_xmit(struct sk_buff *skb, u8 *head,
-			int mlen, int plen, int offset, int type)
+		     int mlen, int plen, int offset, int type)
 {
 	struct sk_buff *frag;
 	int hlen;
@@ -484,8 +475,8 @@ lowpan_skb_fragmentation(struct sk_buff *skb, struct net_device *dev)
 		err = lowpan_fragment_xmit(skb, head, header_length, len,
 					   offset, LOWPAN_DISPATCH_FRAGN);
 		if (err) {
-			pr_debug("%s unable to send a subsequent FRAGN packet "
-				 "(tag: %d, offset: %d", __func__, tag, offset);
+			pr_debug("%s unable to send a FRAGN packet.", __func__);
+			pr_debug("tag: %d, offset: %d", tag, offset);
 			goto exit;
 		}
 
@@ -692,7 +683,7 @@ static int lowpan_newlink(struct net *src_net, struct net_device *dev,
 	lowpan_dev_info(dev)->fragment_tag = 0;
 	mutex_init(&lowpan_dev_info(dev)->dev_list_mtx);
 
-	entry = kzalloc(sizeof(struct lowpan_dev_record), GFP_KERNEL);
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (!entry) {
 		dev_put(real_dev);
 		lowpan_dev_info(dev)->real_dev = NULL;
-- 
1.9.0


------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk

[PATCH net-next 5/6] net: ns: add ieee802154_6lowpan namespace

[Alexander Aring]

This patch adds necessary ieee802154 6lowpan namespace to provide the
inet_frag information. This is a initial support for handling 6lowpan
fragmentation with the inet_frag api.

Signed-off-by: Alexander Aring <alex.aring-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Tested-by: Martin Townsend <martin.townsend-I0Qfutc96jjQT0dZR+AlfA@public.gmane.org>
---
 include/net/net_namespace.h            |  4 ++++
 include/net/netns/ieee802154_6lowpan.h | 13 +++++++++++++
 2 files changed, 17 insertions(+)
 create mode 100644 include/net/netns/ieee802154_6lowpan.h

diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 991dcd9..79387f7 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -15,6 +15,7 @@
 #include <net/netns/packet.h>
 #include <net/netns/ipv4.h>
 #include <net/netns/ipv6.h>
+#include <net/netns/ieee802154_6lowpan.h>
 #include <net/netns/sctp.h>
 #include <net/netns/dccp.h>
 #include <net/netns/netfilter.h>
@@ -90,6 +91,9 @@ struct net {
 #if IS_ENABLED(CONFIG_IPV6)
 	struct netns_ipv6	ipv6;
 #endif
+#if IS_ENABLED(CONFIG_IEEE802154_6LOWPAN)
+	struct netns_ieee802154_lowpan	ieee802154_lowpan;
+#endif
 #if defined(CONFIG_IP_SCTP) || defined(CONFIG_IP_SCTP_MODULE)
 	struct netns_sctp	sctp;
 #endif
diff --git a/include/net/netns/ieee802154_6lowpan.h b/include/net/netns/ieee802154_6lowpan.h
new file mode 100644
index 0000000..88110b7
--- /dev/null
+++ b/include/net/netns/ieee802154_6lowpan.h
@@ -0,0 +1,13 @@
+/*
+ * ieee802154 6lowpan in net namespaces
+ */
+
+#include <net/inet_frag.h>
+
+#ifndef __NETNS_IEEE802154_6LOWPAN_H__
+#define __NETNS_IEEE802154_6LOWPAN_H__
+
+struct netns_ieee802154_lowpan {
+};
+
+#endif
-- 
1.9.0


------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk

[PATCH net-next 6/6] 6lowpan: handling 6lowpan fragmentation via inet_frag api

[Alexander Aring]

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 and source address which is missing in the current
implementation.

This patch add also a lookup function for uncompressed 6LoWPAN header
size. This is needed to estimate the real size of fragmented packet.

Signed-off-by: Alexander Aring <alex.aring-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Tested-by: Martin Townsend <martin.townsend-I0Qfutc96jjQT0dZR+AlfA@public.gmane.org>
---
 include/net/netns/ieee802154_6lowpan.h |   9 +
 net/ieee802154/6lowpan.h               | 112 +++++++
 net/ieee802154/6lowpan_rtnl.c          | 256 +++------------
 net/ieee802154/Makefile                |   2 +-
 net/ieee802154/reassembly.c            | 565 +++++++++++++++++++++++++++++++++
 net/ieee802154/reassembly.h            |  66 ++++
 6 files changed, 802 insertions(+), 208 deletions(-)
 create mode 100644 net/ieee802154/reassembly.c
 create mode 100644 net/ieee802154/reassembly.h

diff --git a/include/net/netns/ieee802154_6lowpan.h b/include/net/netns/ieee802154_6lowpan.h
index 88110b7..079030c 100644
--- a/include/net/netns/ieee802154_6lowpan.h
+++ b/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
diff --git a/net/ieee802154/6lowpan.h b/net/ieee802154/6lowpan.h
index 2b835db..feafa8e 100644
--- a/net/ieee802154/6lowpan.h
+++ b/net/ieee802154/6lowpan.h
@@ -306,6 +306,118 @@ static inline void lowpan_push_hc_data(u8 **hc_ptr, const void *data,
 	*hc_ptr += len;
 }
 
+static inline size_t lowpan_addr_mode_size(const u8 addr_mode)
+{
+	u16 ret = 0;
+
+	switch (addr_mode) {
+	case LOWPAN_IPHC_ADDR_00:
+		ret += 16;
+		break;
+	case LOWPAN_IPHC_ADDR_01:
+		ret += 8;
+		break;
+	case LOWPAN_IPHC_ADDR_02:
+		ret += 2;
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+static inline size_t lowpan_next_hdr_size(const u8 h_enc, u16 *uncomp_header)
+{
+	u16 ret = 1;
+
+	if ((h_enc & LOWPAN_NHC_UDP_MASK) == LOWPAN_NHC_UDP_ID) {
+		*uncomp_header += sizeof(struct udphdr);
+
+		switch (h_enc & LOWPAN_NHC_UDP_CS_P_11) {
+		case LOWPAN_NHC_UDP_CS_P_00:
+			ret += 4;
+			break;
+		case LOWPAN_NHC_UDP_CS_P_01:
+		case LOWPAN_NHC_UDP_CS_P_10:
+			ret += 3;
+			break;
+		case LOWPAN_NHC_UDP_CS_P_11:
+			ret++;
+			break;
+		default:
+			break;
+		}
+
+		if (!(h_enc & LOWPAN_NHC_UDP_CS_C))
+			ret += 2;
+	}
+
+	return ret;
+}
+
+static inline size_t lowpan_header_delta_size(const struct sk_buff *skb)
+{
+	u16 ret = 2, uncomp_header = sizeof(struct ipv6hdr);
+	u8 iphc0, iphc1, h_enc;
+
+	iphc0 = skb_network_header(skb)[0];
+	iphc1 = skb_network_header(skb)[1];
+
+	switch ((iphc0 & LOWPAN_IPHC_TF) >> 3) {
+	case 0:
+		ret += 4;
+		break;
+	case 1:
+		ret += 3;
+		break;
+	case 2:
+		ret++;
+		break;
+	default:
+		break;
+	}
+
+	if (!(iphc0 & LOWPAN_IPHC_NH_C))
+		ret++;
+
+	if (!(iphc0 & 0x03))
+		ret++;
+
+	ret += lowpan_addr_mode_size((iphc1 & LOWPAN_IPHC_SAM) >>
+				     LOWPAN_IPHC_SAM_BIT);
+
+	if (iphc1 & LOWPAN_IPHC_M) {
+		switch ((iphc1 & LOWPAN_IPHC_DAM_11) >>
+			LOWPAN_IPHC_DAM_BIT) {
+		case LOWPAN_IPHC_DAM_00:
+			ret += 16;
+			break;
+		case LOWPAN_IPHC_DAM_01:
+			ret += 6;
+			break;
+		case LOWPAN_IPHC_DAM_10:
+			ret += 4;
+			break;
+		case LOWPAN_IPHC_DAM_11:
+			ret++;
+			break;
+		default:
+			break;
+		}
+	} else {
+		ret += lowpan_addr_mode_size((iphc1 & LOWPAN_IPHC_DAM_11) >>
+					     LOWPAN_IPHC_DAM_BIT);
+	}
+
+	if (iphc0 & LOWPAN_IPHC_NH_C) {
+		h_enc = skb_network_header(skb)[ret];
+		ret += lowpan_next_hdr_size(h_enc, &uncomp_header);
+	}
+
+	return uncomp_header - ret;
+}
+
 typedef int (*skb_delivery_cb)(struct sk_buff *skb, struct net_device *dev);
 
 int lowpan_process_data(struct sk_buff *skb, struct net_device *dev,
diff --git a/net/ieee802154/6lowpan_rtnl.c b/net/ieee802154/6lowpan_rtnl.c
index b543488..7e4fd93 100644
--- a/net/ieee802154/6lowpan_rtnl.c
+++ b/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)
 {
@@ -188,69 +177,6 @@ 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;
@@ -264,94 +190,6 @@ 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;
 
@@ -364,8 +202,6 @@ 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;
@@ -609,44 +445,53 @@ 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 */
+			ret = process_data(local_skb);
+			if (ret == NET_RX_DROP)
+				goto drop;
+			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 */
-			local_skb = skb_clone(skb, GFP_ATOMIC);
-			if (!local_skb)
-				goto drop;
-			process_data(local_skb);
-
-			kfree_skb(skb);
+			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;
@@ -654,9 +499,9 @@ 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;
 }
 
@@ -784,43 +629,40 @@ 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);
diff --git a/net/ieee802154/Makefile b/net/ieee802154/Makefile
index 3d08adf..b113fc4 100644
--- a/net/ieee802154/Makefile
+++ b/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
diff --git a/net/ieee802154/reassembly.c b/net/ieee802154/reassembly.c
new file mode 100644
index 0000000..3b49b97
--- /dev/null
+++ b/net/ieee802154/reassembly.c
@@ -0,0 +1,565 @@
+/*	6LoWPAN fragment reassembly
+ *
+ *
+ *	Authors:
+ *	Alexander Aring		<aar-bIcnvbaLZ9MEGnE8C9+IrQ@public.gmane.org>
+ *
+ *	Based on: net/ipv6/reassembly.c
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ */
+
+#define pr_fmt(fmt) "6LoWPAN: " fmt
+
+#include <linux/net.h>
+#include <linux/list.h>
+#include <linux/netdevice.h>
+#include <linux/random.h>
+#include <linux/jhash.h>
+#include <linux/skbuff.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+
+#include <net/ieee802154_netdev.h>
+#include <net/ipv6.h>
+#include <net/inet_frag.h>
+
+#include "6lowpan.h"
+#include "reassembly.h"
+
+static struct inet_frags lowpan_frags;
+
+static int lowpan_frag_reasm(struct lowpan_frag_queue *fq,
+			     struct sk_buff *prev, struct net_device *dev);
+
+static unsigned int lowpan_hash_frag(__be16 tag, __be16 d_size,
+				     const struct ieee802154_addr *saddr,
+				     const struct ieee802154_addr *daddr)
+{
+	u32 c;
+
+	net_get_random_once(&lowpan_frags.rnd, sizeof(lowpan_frags.rnd));
+	c = jhash_3words(ieee802154_addr_hash(saddr),
+			 ieee802154_addr_hash(daddr),
+			 (__force u32)(tag + (d_size << 16)),
+			 lowpan_frags.rnd);
+
+	return c & (INETFRAGS_HASHSZ - 1);
+}
+
+static unsigned int lowpan_hashfn(struct inet_frag_queue *q)
+{
+	struct lowpan_frag_queue *fq;
+
+	fq = container_of(q, struct lowpan_frag_queue, q);
+	return lowpan_hash_frag(fq->tag, fq->d_size, &fq->saddr, &fq->daddr);
+}
+
+bool lowpan_frag_match(struct inet_frag_queue *q, void *a)
+{
+	struct lowpan_frag_queue *fq;
+	struct lowpan_create_arg *arg = a;
+
+	fq = container_of(q, struct lowpan_frag_queue, q);
+	return	fq->tag == arg->tag && fq->d_size == arg->d_size &&
+		ieee802154_addr_addr_equal(&fq->saddr, arg->src) &&
+		ieee802154_addr_addr_equal(&fq->daddr, arg->dst);
+}
+EXPORT_SYMBOL(lowpan_frag_match);
+
+void lowpan_frag_init(struct inet_frag_queue *q, void *a)
+{
+	struct lowpan_frag_queue *fq;
+	struct lowpan_create_arg *arg = a;
+
+	fq = container_of(q, struct lowpan_frag_queue, q);
+
+	fq->tag = arg->tag;
+	fq->d_size = arg->d_size;
+	fq->saddr = *arg->src;
+	fq->daddr = *arg->dst;
+}
+EXPORT_SYMBOL(lowpan_frag_init);
+
+void lowpan_expire_frag_queue(struct frag_queue *fq, struct inet_frags *frags)
+{
+	spin_lock(&fq->q.lock);
+
+	if (fq->q.last_in & INET_FRAG_COMPLETE)
+		goto out;
+
+	inet_frag_kill(&fq->q, frags);
+out:
+	spin_unlock(&fq->q.lock);
+	inet_frag_put(&fq->q, frags);
+}
+EXPORT_SYMBOL(lowpan_expire_frag_queue);
+
+static void lowpan_frag_expire(unsigned long data)
+{
+	struct frag_queue *fq;
+	struct net *net;
+
+	fq = container_of((struct inet_frag_queue *)data, struct frag_queue, q);
+	net = container_of(fq->q.net, struct net, ieee802154_lowpan.frags);
+
+	lowpan_expire_frag_queue(fq, &lowpan_frags);
+}
+
+static inline struct lowpan_frag_queue *
+fq_find(struct net *net, const struct ieee802154_frag_info *frag_info,
+	const struct ieee802154_addr *src, const struct ieee802154_addr *dst)
+{
+	struct inet_frag_queue *q;
+	struct lowpan_create_arg arg;
+	unsigned int hash;
+
+	arg.tag = frag_info->d_tag;
+	arg.d_size = frag_info->d_size;
+	arg.src = src;
+	arg.dst = dst;
+
+	read_lock(&lowpan_frags.lock);
+	hash = lowpan_hash_frag(frag_info->d_tag, frag_info->d_size, src, dst);
+
+	q = inet_frag_find(&net->ieee802154_lowpan.frags,
+			   &lowpan_frags, &arg, hash);
+	if (IS_ERR_OR_NULL(q)) {
+		inet_frag_maybe_warn_overflow(q, pr_fmt());
+		return NULL;
+	}
+	return container_of(q, struct lowpan_frag_queue, q);
+}
+
+static int lowpan_frag_queue(struct lowpan_frag_queue *fq,
+			     struct sk_buff *skb, const u8 frag_type,
+			     const struct ieee802154_frag_info *frag_info)
+{
+	struct sk_buff *prev, *next;
+	struct net_device *dev;
+	int end, offset;
+
+	if (fq->q.last_in & INET_FRAG_COMPLETE)
+		goto err;
+
+	offset = mac_cb(skb)->frag_info.d_offset << 3;
+	end = mac_cb(skb)->frag_info.d_size;
+
+	/* Is this the final fragment? */
+	if (offset + skb->len == end) {
+		/* If we already have some bits beyond end
+		 * or have different end, the segment is corrupted.
+		 */
+		if (end < fq->q.len ||
+		    ((fq->q.last_in & INET_FRAG_LAST_IN) && end != fq->q.len))
+			goto err;
+		fq->q.last_in |= INET_FRAG_LAST_IN;
+		fq->q.len = end;
+	} else {
+		if (end > fq->q.len) {
+			/* Some bits beyond end -> corruption. */
+			if (fq->q.last_in & INET_FRAG_LAST_IN)
+				goto err;
+			fq->q.len = end;
+		}
+	}
+
+	/* Find out which fragments are in front and at the back of us
+	 * in the chain of fragments so far.  We must know where to put
+	 * this fragment, right?
+	 */
+	prev = fq->q.fragments_tail;
+	if (!prev || mac_cb(prev)->frag_info.d_offset <
+		     mac_cb(skb)->frag_info.d_offset) {
+		next = NULL;
+		goto found;
+	}
+	prev = NULL;
+	for (next = fq->q.fragments; next != NULL; next = next->next) {
+		if (mac_cb(next)->frag_info.d_offset >=
+		    mac_cb(skb)->frag_info.d_offset)
+			break;	/* bingo! */
+		prev = next;
+	}
+
+found:
+	/* Insert this fragment in the chain of fragments. */
+	skb->next = next;
+	if (!next)
+		fq->q.fragments_tail = skb;
+	if (prev)
+		prev->next = skb;
+	else
+		fq->q.fragments = skb;
+
+	dev = skb->dev;
+	if (dev)
+		skb->dev = NULL;
+
+	fq->q.stamp = skb->tstamp;
+	if (frag_type == LOWPAN_DISPATCH_FRAG1) {
+		/* Calculate uncomp. 6lowpan header to estimate full size */
+		fq->q.meat += skb->len + lowpan_header_delta_size(skb);
+		fq->q.last_in |= INET_FRAG_FIRST_IN;
+	} else {
+		fq->q.meat += skb->len;
+	}
+	add_frag_mem_limit(&fq->q, skb->truesize);
+
+	if (fq->q.last_in == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) &&
+	    fq->q.meat == fq->q.len) {
+		int res;
+		unsigned long orefdst = skb->_skb_refdst;
+
+		skb->_skb_refdst = 0UL;
+		res = lowpan_frag_reasm(fq, prev, dev);
+		skb->_skb_refdst = orefdst;
+		return res;
+	}
+
+	inet_frag_lru_move(&fq->q);
+	return -1;
+err:
+	kfree_skb(skb);
+	return -1;
+}
+
+/*	Check if this packet is complete.
+ *	Returns NULL on failure by any reason, and pointer
+ *	to current nexthdr field in reassembled frame.
+ *
+ *	It is called with locked fq, and caller must check that
+ *	queue is eligible for reassembly i.e. it is not COMPLETE,
+ *	the last and the first frames arrived and all the bits are here.
+ */
+static int lowpan_frag_reasm(struct lowpan_frag_queue *fq, struct sk_buff *prev,
+			     struct net_device *dev)
+{
+	struct sk_buff *fp, *head = fq->q.fragments;
+	int sum_truesize;
+
+	inet_frag_kill(&fq->q, &lowpan_frags);
+
+	/* Make the one we just received the head. */
+	if (prev) {
+		head = prev->next;
+		fp = skb_clone(head, GFP_ATOMIC);
+
+		if (!fp)
+			goto out_oom;
+
+		fp->next = head->next;
+		if (!fp->next)
+			fq->q.fragments_tail = fp;
+		prev->next = fp;
+
+		skb_morph(head, fq->q.fragments);
+		head->next = fq->q.fragments->next;
+
+		consume_skb(fq->q.fragments);
+		fq->q.fragments = head;
+	}
+
+	/* Head of list must not be cloned. */
+	if (skb_unclone(head, GFP_ATOMIC))
+		goto out_oom;
+
+	/* If the first fragment is fragmented itself, we split
+	 * it to two chunks: the first with data and paged part
+	 * and the second, holding only fragments.
+	 */
+	if (skb_has_frag_list(head)) {
+		struct sk_buff *clone;
+		int i, plen = 0;
+
+		clone = alloc_skb(0, GFP_ATOMIC);
+		if (!clone)
+			goto out_oom;
+		clone->next = head->next;
+		head->next = clone;
+		skb_shinfo(clone)->frag_list = skb_shinfo(head)->frag_list;
+		skb_frag_list_init(head);
+		for (i = 0; i < skb_shinfo(head)->nr_frags; i++)
+			plen += skb_frag_size(&skb_shinfo(head)->frags[i]);
+		clone->len = head->data_len - plen;
+		clone->data_len = clone->len;
+		head->data_len -= clone->len;
+		head->len -= clone->len;
+		add_frag_mem_limit(&fq->q, clone->truesize);
+	}
+
+	WARN_ON(head == NULL);
+
+	sum_truesize = head->truesize;
+	for (fp = head->next; fp;) {
+		bool headstolen;
+		int delta;
+		struct sk_buff *next = fp->next;
+
+		sum_truesize += fp->truesize;
+		if (skb_try_coalesce(head, fp, &headstolen, &delta)) {
+			kfree_skb_partial(fp, headstolen);
+		} else {
+			if (!skb_shinfo(head)->frag_list)
+				skb_shinfo(head)->frag_list = fp;
+			head->data_len += fp->len;
+			head->len += fp->len;
+			head->truesize += fp->truesize;
+		}
+		fp = next;
+	}
+	sub_frag_mem_limit(&fq->q, sum_truesize);
+
+	head->next = NULL;
+	head->dev = dev;
+	head->tstamp = fq->q.stamp;
+
+	fq->q.fragments = NULL;
+	fq->q.fragments_tail = NULL;
+
+	return 1;
+out_oom:
+	net_dbg_ratelimited("lowpan_frag_reasm: no memory for reassembly\n");
+	return -1;
+}
+
+static int lowpan_get_frag_info(struct sk_buff *skb, const u8 frag_type,
+				struct ieee802154_frag_info *frag_info)
+{
+	bool fail;
+	u8 pattern = 0, low = 0;
+
+	fail = lowpan_fetch_skb(skb, &pattern, 1);
+	fail |= lowpan_fetch_skb(skb, &low, 1);
+	frag_info->d_size = (pattern & 7) << 8 | low;
+	fail |= lowpan_fetch_skb(skb, &frag_info->d_tag, 2);
+
+	if (frag_type == LOWPAN_DISPATCH_FRAGN)
+		fail |= lowpan_fetch_skb(skb, &frag_info->d_offset, 1);
+	else
+		frag_info->d_offset = 0;
+
+	if (unlikely(fail))
+		return -EIO;
+
+	skb_reset_network_header(skb);
+
+	return 0;
+}
+
+int lowpan_frag_rcv(struct sk_buff *skb, const u8 frag_type)
+{
+	struct lowpan_frag_queue *fq;
+	struct net *net = dev_net(skb->dev);
+	struct ieee802154_frag_info *frag_info = &mac_cb(skb)->frag_info;
+	int err;
+
+	err = lowpan_get_frag_info(skb, frag_type, frag_info);
+	if (err < 0)
+		goto err;
+
+	if (frag_info->d_size > net->ieee802154_lowpan.max_dsize)
+		goto err;
+
+	inet_frag_evictor(&net->ieee802154_lowpan.frags, &lowpan_frags, false);
+
+	fq = fq_find(net, frag_info, &mac_cb(skb)->sa, &mac_cb(skb)->da);
+	if (fq != NULL) {
+		int ret;
+		spin_lock(&fq->q.lock);
+		ret = lowpan_frag_queue(fq, skb, frag_type, frag_info);
+		spin_unlock(&fq->q.lock);
+
+		inet_frag_put(&fq->q, &lowpan_frags);
+		return ret;
+	}
+
+err:
+	kfree_skb(skb);
+	return -1;
+}
+EXPORT_SYMBOL(lowpan_frag_rcv);
+
+#ifdef CONFIG_SYSCTL
+static struct ctl_table lowpan_frags_ns_ctl_table[] = {
+	{
+		.procname	= "6lowpanfrag_high_thresh",
+		.data		= &init_net.ieee802154_lowpan.frags.high_thresh,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{
+		.procname	= "6lowpanfrag_low_thresh",
+		.data		= &init_net.ieee802154_lowpan.frags.low_thresh,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{
+		.procname	= "6lowpanfrag_time",
+		.data		= &init_net.ieee802154_lowpan.frags.timeout,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{
+		.procname	= "6lowpanfrag_max_datagram_size",
+		.data		= &init_net.ieee802154_lowpan.max_dsize,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec
+	},
+	{ }
+};
+
+static struct ctl_table lowpan_frags_ctl_table[] = {
+	{
+		.procname	= "6lowpanfrag_secret_interval",
+		.data		= &lowpan_frags.secret_interval,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_jiffies,
+	},
+	{ }
+};
+
+static int __net_init lowpan_frags_ns_sysctl_register(struct net *net)
+{
+	struct ctl_table *table;
+	struct ctl_table_header *hdr;
+
+	table = lowpan_frags_ns_ctl_table;
+	if (!net_eq(net, &init_net)) {
+		table = kmemdup(table, sizeof(lowpan_frags_ns_ctl_table),
+				GFP_KERNEL);
+		if (table == NULL)
+			goto err_alloc;
+
+		table[0].data = &net->ieee802154_lowpan.frags.high_thresh;
+		table[1].data = &net->ieee802154_lowpan.frags.low_thresh;
+		table[2].data = &net->ieee802154_lowpan.frags.timeout;
+		table[2].data = &net->ieee802154_lowpan.max_dsize;
+
+		/* Don't export sysctls to unprivileged users */
+		if (net->user_ns != &init_user_ns)
+			table[0].procname = NULL;
+	}
+
+	hdr = register_net_sysctl(net, "net/ieee802154/6lowpan", table);
+	if (hdr == NULL)
+		goto err_reg;
+
+	net->ieee802154_lowpan.sysctl.frags_hdr = hdr;
+	return 0;
+
+err_reg:
+	if (!net_eq(net, &init_net))
+		kfree(table);
+err_alloc:
+	return -ENOMEM;
+}
+
+static void __net_exit lowpan_frags_ns_sysctl_unregister(struct net *net)
+{
+	struct ctl_table *table;
+
+	table = net->ieee802154_lowpan.sysctl.frags_hdr->ctl_table_arg;
+	unregister_net_sysctl_table(net->ieee802154_lowpan.sysctl.frags_hdr);
+	if (!net_eq(net, &init_net))
+		kfree(table);
+}
+
+static struct ctl_table_header *lowpan_ctl_header;
+
+static int lowpan_frags_sysctl_register(void)
+{
+	lowpan_ctl_header = register_net_sysctl(&init_net,
+						"net/ieee802154/6lowpan",
+						lowpan_frags_ctl_table);
+	return lowpan_ctl_header == NULL ? -ENOMEM : 0;
+}
+
+static void lowpan_frags_sysctl_unregister(void)
+{
+	unregister_net_sysctl_table(lowpan_ctl_header);
+}
+#else
+static inline int lowpan_frags_ns_sysctl_register(struct net *net)
+{
+	return 0;
+}
+
+static inline void lowpan_frags_ns_sysctl_unregister(struct net *net)
+{
+}
+
+static inline int lowpan_frags_sysctl_register(void)
+{
+	return 0;
+}
+
+static inline void lowpan_frags_sysctl_unregister(void)
+{
+}
+#endif
+
+static int __net_init lowpan_frags_init_net(struct net *net)
+{
+	net->ieee802154_lowpan.frags.high_thresh = IPV6_FRAG_HIGH_THRESH;
+	net->ieee802154_lowpan.frags.low_thresh = IPV6_FRAG_LOW_THRESH;
+	net->ieee802154_lowpan.frags.timeout = IPV6_FRAG_TIMEOUT;
+	net->ieee802154_lowpan.max_dsize = 0xFFFF;
+
+	inet_frags_init_net(&net->ieee802154_lowpan.frags);
+
+	return lowpan_frags_ns_sysctl_register(net);
+}
+
+static void __net_exit lowpan_frags_exit_net(struct net *net)
+{
+	lowpan_frags_ns_sysctl_unregister(net);
+	inet_frags_exit_net(&net->ieee802154_lowpan.frags, &lowpan_frags);
+}
+
+static struct pernet_operations lowpan_frags_ops = {
+	.init = lowpan_frags_init_net,
+	.exit = lowpan_frags_exit_net,
+};
+
+int __init lowpan_net_frag_init(void)
+{
+	int ret;
+
+	ret = lowpan_frags_sysctl_register();
+	if (ret)
+		goto out;
+
+	ret = register_pernet_subsys(&lowpan_frags_ops);
+	if (ret)
+		goto err_pernet;
+
+	lowpan_frags.hashfn = lowpan_hashfn;
+	lowpan_frags.constructor = lowpan_frag_init;
+	lowpan_frags.destructor = NULL;
+	lowpan_frags.skb_free = NULL;
+	lowpan_frags.qsize = sizeof(struct frag_queue);
+	lowpan_frags.match = lowpan_frag_match;
+	lowpan_frags.frag_expire = lowpan_frag_expire;
+	lowpan_frags.secret_interval = 10 * 60 * HZ;
+	inet_frags_init(&lowpan_frags);
+err_pernet:
+	lowpan_frags_sysctl_unregister();
+out:
+	return ret;
+}
+
+void lowpan_net_frag_exit(void)
+{
+	inet_frags_fini(&lowpan_frags);
+	lowpan_frags_sysctl_unregister();
+	unregister_pernet_subsys(&lowpan_frags_ops);
+}
diff --git a/net/ieee802154/reassembly.h b/net/ieee802154/reassembly.h
new file mode 100644
index 0000000..055518b
--- /dev/null
+++ b/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__ */
-- 
1.9.0


------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk

Re: [PATCH net-next 4/6] 6lowpan: fix some checkpatch issues

[Joe Perches]

On Fri, 2014-02-21 at 16:15 +0100, Alexander Aring wrote:
> diff --git a/net/ieee802154/6lowpan_rtnl.c b/net/ieee802154/6lowpan_rtnl.c
[]
> @@ -484,8 +475,8 @@ lowpan_skb_fragmentation(struct sk_buff *skb, struct net_device *dev)
>  		err = lowpan_fragment_xmit(skb, head, header_length, len,
>  					   offset, LOWPAN_DISPATCH_FRAGN);
>  		if (err) {
> -			pr_debug("%s unable to send a subsequent FRAGN packet "
> -				 "(tag: %d, offset: %d", __func__, tag, offset);
> +			pr_debug("%s unable to send a FRAGN packet.", __func__);
> +			pr_debug("tag: %d, offset: %d", tag, offset);

No need for 2 pr_debug statements.

Better to coalesce the format pieces into a single string
adding the close parenthesis and a terminating newline.

			pr_debug("%s: unable to send a subsequent FRAGN packet "(tag: %d, offset: %d)\n",
				 __func__, tag, offset);

Re: [PATCH net-next 4/6] 6lowpan: fix some checkpatch issues

[Alexander Aring]

Hi,

On Fri, Feb 21, 2014 at 08:58:57AM -0800, Joe Perches wrote:
> On Fri, 2014-02-21 at 16:15 +0100, Alexander Aring wrote:
> > diff --git a/net/ieee802154/6lowpan_rtnl.c b/net/ieee802154/6lowpan_rtnl.c
> []
> > @@ -484,8 +475,8 @@ lowpan_skb_fragmentation(struct sk_buff *skb, struct net_device *dev)
> >  		err = lowpan_fragment_xmit(skb, head, header_length, len,
> >  					   offset, LOWPAN_DISPATCH_FRAGN);
> >  		if (err) {
> > -			pr_debug("%s unable to send a subsequent FRAGN packet "
> > -				 "(tag: %d, offset: %d", __func__, tag, offset);
> > +			pr_debug("%s unable to send a FRAGN packet.", __func__);
> > +			pr_debug("tag: %d, offset: %d", tag, offset);
> 
> No need for 2 pr_debug statements.
> 
> Better to coalesce the format pieces into a single string
> adding the close parenthesis and a terminating newline.
> 
> 			pr_debug("%s: unable to send a subsequent FRAGN packet "(tag: %d, offset: %d)\n",
> 				 __func__, tag, offset);
ah ok, thanks I will fix it.

- Alex

Re: [PATCH net-next 3/6] 6lowpan: move 6lowpan.c to 6lowpan_rtnl.c

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 1377 bytes --]

On Fri, 2014-02-21 at 16:15 +0100, Alexander Aring wrote:
> We have a 6lowpan.c file and 6lowpan.ko file. To avoid confusing we
> should move 6lowpan.c to 6lowpan.ko. Then we can support multiple source

You put the wrong filename here ^

Ben.

> files for 6lowpan module.
> 
> Signed-off-by: Alexander Aring <alex.aring@gmail.com>
> Tested-by: Martin Townsend <martin.townsend@xsilon.com>
> ---
>  net/ieee802154/{6lowpan.c => 6lowpan_rtnl.c} | 0
>  net/ieee802154/Makefile                      | 1 +
>  2 files changed, 1 insertion(+)
>  rename net/ieee802154/{6lowpan.c => 6lowpan_rtnl.c} (100%)
> 
> diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan_rtnl.c
> similarity index 100%
> rename from net/ieee802154/6lowpan.c
> rename to net/ieee802154/6lowpan_rtnl.c
> diff --git a/net/ieee802154/Makefile b/net/ieee802154/Makefile
> index e8f0588..3d08adf 100644
> --- a/net/ieee802154/Makefile
> +++ b/net/ieee802154/Makefile
> @@ -2,5 +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
>  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

-- 
Ben Hutchings
I haven't lost my mind; it's backed up on tape somewhere.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 811 bytes --]

Re: [PATCH net-next 3/6] 6lowpan: move 6lowpan.c to 6lowpan_rtnl.c

[Alexander Aring]

Hi Ben,

thanks for your reply.

On Sat, Feb 22, 2014 at 01:40:52PM +0000, Ben Hutchings wrote:
> On Fri, 2014-02-21 at 16:15 +0100, Alexander Aring wrote:
> > We have a 6lowpan.c file and 6lowpan.ko file. To avoid confusing we
> > should move 6lowpan.c to 6lowpan.ko. Then we can support multiple source
> 
> You put the wrong filename here ^
> 
mhh, I meant here:

Currently we have only one-file supporting like:

obj-$(CONFIG_IEEE802154_6LOWPAN) += 6lowpan.o


It's generate a 6lowpan.ko from a 6lowpan.c file.

Now we need some multiple-file supporting and I change it to:

obj-$(CONFIG_IEEE802154_6LOWPAN) += 6lowpan.o
6lowpan-y := 6lowpan_rtnl.o reassembly.o


to generate a still the 6lowpan.ko module with object files of
6lowpan_rtnl.o (which contains mostly some rtnl callbacks implementations
now) and reassembly.o file.

It's to support multiple files for a module only.


I can't do something like:

obj-$(CONFIG_IEEE802154_6LOWPAN) += 6lowpan.o
6lowpan-y := 6lowpan.o reassembly.o

because I have the same name for 6lowpan.c and 6lowpan.ko.

Is there maybe a better solution for this?



btw.

I also detected:

obj-$(CONFIG_6LOWPAN_IPHC) += 6lowpan_iphc.o

which is a seperate module. (if you selected it as module) but this
makes no sense, we should built it in 6lowpan.ko like:

6lowpan-y := 6lowpan_rtnl.o 6lowpan_iphc.o reassembly.o

aswell.


- Alex

------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk

Re: [PATCH net-next 3/6] 6lowpan: move 6lowpan.c to 6lowpan_rtnl.c

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 685 bytes --]

On Sat, 2014-02-22 at 15:03 +0100, Alexander Aring wrote:
> Hi Ben,
> 
> thanks for your reply.
> 
> On Sat, Feb 22, 2014 at 01:40:52PM +0000, Ben Hutchings wrote:
> > On Fri, 2014-02-21 at 16:15 +0100, Alexander Aring wrote:
> > > We have a 6lowpan.c file and 6lowpan.ko file. To avoid confusing we
> > > should move 6lowpan.c to 6lowpan.ko. Then we can support multiple source
> > 
> > You put the wrong filename here ^
> > 
> mhh, I meant here:
[...]

I know what you mean, but the second '6lowpan.ko' in the description
should surely be '6lowpan_rtnl.c' like in the subject.

Ben.

-- 
Ben Hutchings
I haven't lost my mind; it's backed up on tape somewhere.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 811 bytes --]

Re: [PATCH net-next 3/6] 6lowpan: move 6lowpan.c to 6lowpan_rtnl.c

[Alexander Aring]

On Sat, Feb 22, 2014 at 02:07:15PM +0000, Ben Hutchings wrote:
> On Sat, 2014-02-22 at 15:03 +0100, Alexander Aring wrote:
> > Hi Ben,
> > 
> > thanks for your reply.
> > 
> > On Sat, Feb 22, 2014 at 01:40:52PM +0000, Ben Hutchings wrote:
> > > On Fri, 2014-02-21 at 16:15 +0100, Alexander Aring wrote:
> > > > We have a 6lowpan.c file and 6lowpan.ko file. To avoid confusing we
> > > > should move 6lowpan.c to 6lowpan.ko. Then we can support multiple source
> > > 
> > > You put the wrong filename here ^
> > > 
> > mhh, I meant here:
> [...]
> 
> I know what you mean, but the second '6lowpan.ko' in the description
> should surely be '6lowpan_rtnl.c' like in the subject.
> 
oh yes. Thanks. :-)

- Alex

------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121054471&iu=/4140/ostg.clktrk