[dpdk-dev] [PATCH 12/13] ip_frag: add support for IPv6 reassembly

[Anatoly Burakov <anatoly.burakov@intel.com>]

Mostly a copy-paste of IPv4, with a few caveats.

Only supported packets are those in which fragment extension header is
just after the IPv6 header.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
---
 lib/librte_ip_frag/Makefile              |   1 +
 lib/librte_ip_frag/ip_frag_common.h      |  25 +++-
 lib/librte_ip_frag/ip_frag_internal.c    | 172 +++++++++++++++++-------
 lib/librte_ip_frag/rte_ip_frag.h         |  51 +++++++-
 lib/librte_ip_frag/rte_ipv4_reassembly.c |   4 +-
 lib/librte_ip_frag/rte_ipv6_reassembly.c | 218 +++++++++++++++++++++++++++++++
 6 files changed, 421 insertions(+), 50 deletions(-)
 create mode 100644 lib/librte_ip_frag/rte_ipv6_reassembly.c

diff --git a/lib/librte_ip_frag/Makefile b/lib/librte_ip_frag/Makefile
index 13a4f9f..29aa36f 100644
--- a/lib/librte_ip_frag/Makefile
+++ b/lib/librte_ip_frag/Makefile
@@ -41,6 +41,7 @@ CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR)
 SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ipv4_fragmentation.c
 SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ipv4_reassembly.c
 SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ipv6_fragmentation.c
+SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ipv6_reassembly.c
 SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += rte_ip_frag_common.c
 SRCS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += ip_frag_internal.c
 
diff --git a/lib/librte_ip_frag/ip_frag_common.h b/lib/librte_ip_frag/ip_frag_common.h
index 3e588a0..ac5cd61 100644
--- a/lib/librte_ip_frag/ip_frag_common.h
+++ b/lib/librte_ip_frag/ip_frag_common.h
@@ -51,9 +51,17 @@ if (!(exp))	{							\
 #define RTE_IP_FRAG_ASSERT(exp)	do { } while(0)
 #endif /* IP_FRAG_DEBUG */
 
+#define IPV4_KEYLEN 1
+#define IPV6_KEYLEN 4
+
 /* helper macros */
 #define	IP_FRAG_MBUF2DR(dr, mb)	((dr)->row[(dr)->cnt++] = (mb))
 
+#define IPv6_KEY_BYTES(key) \
+	(key)[0], (key)[1], (key)[2], (key)[3]
+#define IPv6_KEY_BYTES_FMT \
+	"%08" PRIx64 "%08" PRIx64 "%08" PRIx64 "%08" PRIx64
+
 /* internal functions declarations */
 struct rte_mbuf * ip_frag_process(struct rte_ip_frag_pkt *fp,
 		struct rte_ip_frag_death_row *dr, struct rte_mbuf *mb,
@@ -69,6 +77,7 @@ struct rte_ip_frag_pkt * ip_frag_lookup(struct rte_ip_frag_tbl *tbl,
 
 /* these functions need to be declared here as ip_frag_process relies on them */
 struct rte_mbuf * ipv4_frag_reassemble(const struct rte_ip_frag_pkt *fp);
+struct rte_mbuf * ipv6_frag_reassemble(const struct rte_ip_frag_pkt *fp);
 
 
 
@@ -80,8 +89,10 @@ struct rte_mbuf * ipv4_frag_reassemble(const struct rte_ip_frag_pkt *fp);
 static inline int
 ip_frag_key_is_empty(const struct ip_frag_key * key)
 {
-	if (key->src_dst != 0)
-		return 0;
+	uint32_t i;
+	for (i = 0; i < key->key_len; i++)
+		if (key->src_dst[i] != 0)
+			return 0;
 	return 1;
 }
 
@@ -89,14 +100,20 @@ ip_frag_key_is_empty(const struct ip_frag_key * key)
 static inline void
 ip_frag_key_invalidate(struct ip_frag_key * key)
 {
-	key->src_dst = 0;
+	uint32_t i;
+	for (i = 0; i < key->key_len; i++)
+		key->src_dst[i] = 0;
 }
 
 /* compare two keys */
 static inline int
 ip_frag_key_cmp(const struct ip_frag_key * k1, const struct ip_frag_key * k2)
 {
-	return k1->src_dst ^ k2->src_dst;
+	uint32_t i, val;
+	val = k1->id ^ k2->id;
+	for (i = 0; i < k1->key_len; i++)
+		val |= k1->src_dst[i] ^ k2->src_dst[i];
+	return val;
 }
 
 /*
diff --git a/lib/librte_ip_frag/ip_frag_internal.c b/lib/librte_ip_frag/ip_frag_internal.c
index 2f5a4b8..5d35037 100644
--- a/lib/librte_ip_frag/ip_frag_internal.c
+++ b/lib/librte_ip_frag/ip_frag_internal.c
@@ -110,6 +110,35 @@ ipv4_frag_hash(const struct ip_frag_key *key, uint32_t *v1, uint32_t *v2)
 	*v2 = (v << 7) + (v >> 14);
 }
 
+static inline void
+ipv6_frag_hash(const struct ip_frag_key *key, uint32_t *v1, uint32_t *v2)
+{
+	uint32_t v;
+	const uint32_t *p;
+
+	p = (const uint32_t *) &key->src_dst;
+
+#ifdef RTE_MACHINE_CPUFLAG_SSE4_2
+	v = rte_hash_crc_4byte(p[0], PRIME_VALUE);
+	v = rte_hash_crc_4byte(p[1], v);
+	v = rte_hash_crc_4byte(p[2], v);
+	v = rte_hash_crc_4byte(p[3], v);
+	v = rte_hash_crc_4byte(p[4], v);
+	v = rte_hash_crc_4byte(p[5], v);
+	v = rte_hash_crc_4byte(p[6], v);
+	v = rte_hash_crc_4byte(p[7], v);
+	v = rte_hash_crc_4byte(key->id, v);
+#else
+
+	v = rte_jhash_3words(p[0], p[1], p[2], PRIME_VALUE);
+	v = rte_jhash_3words(p[3], p[4], p[5], v);
+	v = rte_jhash_3words(p[6], p[7], key->id, v);
+#endif /* RTE_MACHINE_CPUFLAG_SSE4_2 */
+
+	*v1 =  v;
+	*v2 = (v << 7) + (v >> 14);
+}
+
 struct rte_mbuf *
 ip_frag_process(struct rte_ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
 	struct rte_mbuf *mb, uint16_t ofs, uint16_t len, uint16_t more_frags)
@@ -142,18 +171,32 @@ ip_frag_process(struct rte_ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
 	if (idx >= sizeof (fp->frags) / sizeof (fp->frags[0])) {
 
 		/* report an error. */
-		IP_FRAG_LOG(DEBUG, "%s:%d invalid fragmented packet:\n"
-			"ipv4_frag_pkt: %p, key: <%" PRIx64 ", %#x>, "
-			"total_size: %u, frag_size: %u, last_idx: %u\n"
-			"first fragment: ofs: %u, len: %u\n"
-			"last fragment: ofs: %u, len: %u\n\n",
-			__func__, __LINE__,
-			fp, fp->key.src_dst[0], fp->key.id,
-			fp->total_size, fp->frag_size, fp->last_idx,
-			fp->frags[IP_FIRST_FRAG_IDX].ofs,
-			fp->frags[IP_FIRST_FRAG_IDX].len,
-			fp->frags[IP_LAST_FRAG_IDX].ofs,
-			fp->frags[IP_LAST_FRAG_IDX].len);
+		if (fp->key.key_len == IPV4_KEYLEN)
+			IP_FRAG_LOG(DEBUG, "%s:%d invalid fragmented packet:\n"
+				"ipv4_frag_pkt: %p, key: <%" PRIx64 ", %#x>, "
+				"total_size: %u, frag_size: %u, last_idx: %u\n"
+				"first fragment: ofs: %u, len: %u\n"
+				"last fragment: ofs: %u, len: %u\n\n",
+				__func__, __LINE__,
+				fp, fp->key.src_dst[0], fp->key.id,
+				fp->total_size, fp->frag_size, fp->last_idx,
+				fp->frags[IP_FIRST_FRAG_IDX].ofs,
+				fp->frags[IP_FIRST_FRAG_IDX].len,
+				fp->frags[IP_LAST_FRAG_IDX].ofs,
+				fp->frags[IP_LAST_FRAG_IDX].len);
+		else
+			IP_FRAG_LOG(DEBUG, "%s:%d invalid fragmented packet:\n"
+				"ipv4_frag_pkt: %p, key: <" IPv6_KEY_BYTES_FMT ", %#x>, "
+				"total_size: %u, frag_size: %u, last_idx: %u\n"
+				"first fragment: ofs: %u, len: %u\n"
+				"last fragment: ofs: %u, len: %u\n\n",
+				__func__, __LINE__,
+				fp, IPv6_KEY_BYTES(fp->key.src_dst), fp->key.id,
+				fp->total_size, fp->frag_size, fp->last_idx,
+				fp->frags[IP_FIRST_FRAG_IDX].ofs,
+				fp->frags[IP_FIRST_FRAG_IDX].len,
+				fp->frags[IP_LAST_FRAG_IDX].ofs,
+				fp->frags[IP_LAST_FRAG_IDX].len);
 
 		/* free all fragments, invalidate the entry. */
 		ip_frag_free(fp, dr);
@@ -175,25 +218,43 @@ ip_frag_process(struct rte_ip_frag_pkt *fp, struct rte_ip_frag_death_row *dr,
 
 	/* if we collected all fragments, then try to reassemble. */
 	} else if (fp->frag_size == fp->total_size &&
-			fp->frags[IP_FIRST_FRAG_IDX].mb != NULL)
-		mb = ipv4_frag_reassemble(fp);
+			fp->frags[IP_FIRST_FRAG_IDX].mb != NULL) {
+		if (fp->key.key_len == IPV4_KEYLEN)
+			mb = ipv4_frag_reassemble(fp);
+		else
+			mb = ipv6_frag_reassemble(fp);
+	}
 
 	/* errorenous set of fragments. */
 	if (mb == NULL) {
 
 		/* report an error. */
-		IP_FRAG_LOG(DEBUG, "%s:%d invalid fragmented packet:\n"
-			"ipv4_frag_pkt: %p, key: <%" PRIx64 ", %#x>, "
-			"total_size: %u, frag_size: %u, last_idx: %u\n"
-			"first fragment: ofs: %u, len: %u\n"
-			"last fragment: ofs: %u, len: %u\n\n",
-			__func__, __LINE__,
-			fp, fp->key.src_dst[0], fp->key.id,
-			fp->total_size, fp->frag_size, fp->last_idx,
-			fp->frags[IP_FIRST_FRAG_IDX].ofs,
-			fp->frags[IP_FIRST_FRAG_IDX].len,
-			fp->frags[IP_LAST_FRAG_IDX].ofs,
-			fp->frags[IP_LAST_FRAG_IDX].len);
+		if (fp->key.key_len == IPV4_KEYLEN)
+			IP_FRAG_LOG(DEBUG, "%s:%d invalid fragmented packet:\n"
+				"ipv4_frag_pkt: %p, key: <%" PRIx64 ", %#x>, "
+				"total_size: %u, frag_size: %u, last_idx: %u\n"
+				"first fragment: ofs: %u, len: %u\n"
+				"last fragment: ofs: %u, len: %u\n\n",
+				__func__, __LINE__,
+				fp, fp->key.src_dst[0], fp->key.id,
+				fp->total_size, fp->frag_size, fp->last_idx,
+				fp->frags[IP_FIRST_FRAG_IDX].ofs,
+				fp->frags[IP_FIRST_FRAG_IDX].len,
+				fp->frags[IP_LAST_FRAG_IDX].ofs,
+				fp->frags[IP_LAST_FRAG_IDX].len);
+		else
+			IP_FRAG_LOG(DEBUG, "%s:%d invalid fragmented packet:\n"
+				"ipv4_frag_pkt: %p, key: <" IPv6_KEY_BYTES_FMT ", %#x>, "
+				"total_size: %u, frag_size: %u, last_idx: %u\n"
+				"first fragment: ofs: %u, len: %u\n"
+				"last fragment: ofs: %u, len: %u\n\n",
+				__func__, __LINE__,
+				fp, IPv6_KEY_BYTES(fp->key.src_dst), fp->key.id,
+				fp->total_size, fp->frag_size, fp->last_idx,
+				fp->frags[IP_FIRST_FRAG_IDX].ofs,
+				fp->frags[IP_FIRST_FRAG_IDX].len,
+				fp->frags[IP_LAST_FRAG_IDX].ofs,
+				fp->frags[IP_LAST_FRAG_IDX].len);
 
 		/* free associated resources. */
 		ip_frag_free(fp, dr);
@@ -291,21 +352,34 @@ ip_frag_lookup(struct rte_ip_frag_tbl *tbl,
 	if (tbl->last != NULL && ip_frag_key_cmp(&tbl->last->key, key) == 0)
 		return (tbl->last);
 
-	ipv4_frag_hash(key, &sig1, &sig2);
+	/* different hashing methods for IPv4 and IPv6 */
+	if (key->key_len == 1)
+		ipv4_frag_hash(key, &sig1, &sig2);
+	else
+		ipv6_frag_hash(key, &sig1, &sig2);
 
 	p1 = IP_FRAG_TBL_POS(tbl, sig1);
 	p2 = IP_FRAG_TBL_POS(tbl, sig2);
 
 	for (i = 0; i != assoc; i++) {
-
-		IP_FRAG_LOG(DEBUG, "%s:%d:\n"
-				"tbl: %p, max_entries: %u, use_entries: %u\n"
-				"ipv6_frag_pkt line0: %p, index: %u from %u\n"
-		"key: <%" PRIx64 ", %#x>, start: %" PRIu64 "\n",
-				__func__, __LINE__,
-				tbl, tbl->max_entries, tbl->use_entries,
-				p1, i, assoc,
-		p1[i].key.src_dst[0], p1[i].key.id, p1[i].start);
+		if (p1->key.key_len == IPV4_KEYLEN)
+			IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+					"tbl: %p, max_entries: %u, use_entries: %u\n"
+					"ipv6_frag_pkt line0: %p, index: %u from %u\n"
+			"key: <%" PRIx64 ", %#x>, start: %" PRIu64 "\n",
+					__func__, __LINE__,
+					tbl, tbl->max_entries, tbl->use_entries,
+					p1, i, assoc,
+			p1[i].key.src_dst[0], p1[i].key.id, p1[i].start);
+		else
+			IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+					"tbl: %p, max_entries: %u, use_entries: %u\n"
+					"ipv6_frag_pkt line0: %p, index: %u from %u\n"
+			"key: <" IPv6_KEY_BYTES_FMT ", %#x>, start: %" PRIu64 "\n",
+					__func__, __LINE__,
+					tbl, tbl->max_entries, tbl->use_entries,
+					p1, i, assoc,
+			IPv6_KEY_BYTES(p1[i].key.src_dst), p1[i].key.id, p1[i].start);
 
 		if (ip_frag_key_cmp(&p1[i].key, key) == 0)
 			return (p1 + i);
@@ -314,14 +388,24 @@ ip_frag_lookup(struct rte_ip_frag_tbl *tbl,
 		else if (max_cycles + p1[i].start < tms)
 			old = (old == NULL) ? (p1 + i) : old;
 
-		IP_FRAG_LOG(DEBUG, "%s:%d:\n"
-				"tbl: %p, max_entries: %u, use_entries: %u\n"
-				"ipv6_frag_pkt line1: %p, index: %u from %u\n"
-		"key: <%" PRIx64 ", %#x>, start: %" PRIu64 "\n",
-				__func__, __LINE__,
-				tbl, tbl->max_entries, tbl->use_entries,
-				p2, i, assoc,
-		p2[i].key.src_dst[0], p2[i].key.id, p2[i].start);
+		if (p2->key.key_len == IPV4_KEYLEN)
+			IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+					"tbl: %p, max_entries: %u, use_entries: %u\n"
+					"ipv6_frag_pkt line1: %p, index: %u from %u\n"
+			"key: <%" PRIx64 ", %#x>, start: %" PRIu64 "\n",
+					__func__, __LINE__,
+					tbl, tbl->max_entries, tbl->use_entries,
+					p2, i, assoc,
+			p2[i].key.src_dst[0], p2[i].key.id, p2[i].start);
+		else
+			IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+					"tbl: %p, max_entries: %u, use_entries: %u\n"
+					"ipv6_frag_pkt line1: %p, index: %u from %u\n"
+			"key: <" IPv6_KEY_BYTES_FMT ", %#x>, start: %" PRIu64 "\n",
+					__func__, __LINE__,
+					tbl, tbl->max_entries, tbl->use_entries,
+					p2, i, assoc,
+			IPv6_KEY_BYTES(p2[i].key.src_dst), p2[i].key.id, p2[i].start);
 
 		if (ip_frag_key_cmp(&p2[i].key, key) == 0)
 			return (p2 + i);
diff --git a/lib/librte_ip_frag/rte_ip_frag.h b/lib/librte_ip_frag/rte_ip_frag.h
index 4a4b5c3..8546c3a 100644
--- a/lib/librte_ip_frag/rte_ip_frag.h
+++ b/lib/librte_ip_frag/rte_ip_frag.h
@@ -65,8 +65,9 @@ struct ip_frag {
 
 /** @internal <src addr, dst_addr, id> to uniquely indetify fragmented datagram. */
 struct ip_frag_key {
-	uint64_t src_dst;      /**< src address */
+	uint64_t src_dst[4];      /**< src address, first 8 bytes used for IPv4 */
 	uint32_t id;           /**< dst address */
+	uint32_t key_len;      /**< src/dst key length */
 };
 
 /*
@@ -200,6 +201,54 @@ rte_ipv6_fragment_packet(struct rte_mbuf *pkt_in,
 		struct rte_mempool *pool_direct,
 		struct rte_mempool *pool_indirect);
 
+
+/*
+ * This function implements reassembly of fragmented IPv6 packets.
+ * Incoming mbuf should have its l2_len/l3_len fields setup correctly.
+ *
+ * @param tbl
+ *   Table where to lookup/add the fragmented packet.
+ * @param dr
+ *   Death row to free buffers to
+ * @param mb
+ *   Incoming mbuf with IPv6 fragment.
+ * @param tms
+ *   Fragment arrival timestamp.
+ * @param ip_hdr
+ *   Pointer to the IPv6 header.
+ * @param frag_hdr
+ *   Pointer to the IPv6 fragment extension header.
+ * @return
+ *   Pointer to mbuf for reassembled packet, or NULL if:
+ *   - an error occured.
+ *   - not all fragments of the packet are collected yet.
+ */
+struct rte_mbuf * rte_ipv6_frag_reassemble_packet(struct rte_ip_frag_tbl *tbl,
+		struct rte_ip_frag_death_row *dr,
+		struct rte_mbuf *mb, uint64_t tms, struct ipv6_hdr *ip_hdr,
+		struct ipv6_extension_fragment *frag_hdr);
+
+/*
+ * Return a pointer to the packet's fragment header, if found.
+ * It only looks at the extension header that's right after the fixed IPv6
+ * header, and doesn't follow the whole chain of extension headers.
+ *
+ * @param hdr
+ *   Pointer to the IPv6 header.
+ * @return
+ *   Pointer to the IPv6 fragment extension header, or NULL if it's not
+ *   present.
+ */
+static inline struct ipv6_extension_fragment *
+rte_ipv6_frag_get_ipv6_fragment_header(struct ipv6_hdr *hdr)
+{
+	if (hdr->proto == IPPROTO_FRAGMENT) {
+		return (struct ipv6_extension_fragment *) ++hdr;
+	}
+	else
+		return NULL;
+}
+
 /**
  * IPv4 fragmentation.
  *
diff --git a/lib/librte_ip_frag/rte_ipv4_reassembly.c b/lib/librte_ip_frag/rte_ipv4_reassembly.c
index 483fb95..cc9a9c8 100644
--- a/lib/librte_ip_frag/rte_ipv4_reassembly.c
+++ b/lib/librte_ip_frag/rte_ipv4_reassembly.c
@@ -138,8 +138,10 @@ rte_ipv4_frag_reassemble_packet(struct rte_ip_frag_tbl *tbl,
 	ip_flag = (uint16_t)(flag_offset & IPV4_HDR_MF_FLAG);
 
 	psd = (uint64_t *)&ip_hdr->src_addr;
-	key.src_dst = *psd;
+	/* use first 8 bytes only */
+	key.src_dst[0] = psd[0];
 	key.id = ip_hdr->packet_id;
+	key.key_len = IPV4_KEYLEN;
 
 	ip_ofs *= IPV4_HDR_OFFSET_UNITS;
 	ip_len = (uint16_t)(rte_be_to_cpu_16(ip_hdr->total_length) -
diff --git a/lib/librte_ip_frag/rte_ipv6_reassembly.c b/lib/librte_ip_frag/rte_ipv6_reassembly.c
new file mode 100644
index 0000000..0cc7f93
--- /dev/null
+++ b/lib/librte_ip_frag/rte_ipv6_reassembly.c
@@ -0,0 +1,218 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <stddef.h>
+#include <stdint.h>
+
+#include <rte_malloc.h>
+#include <rte_memcpy.h>
+#include <rte_byteorder.h>
+#include <rte_mbuf.h>
+#include <rte_ip.h>
+
+#include "rte_ip_frag.h"
+#include "ip_frag_common.h"
+
+/**
+ * @file
+ * IPv6 reassemble
+ *
+ * Implementation of IPv6 reassembly.
+ *
+ */
+
+/*
+ * Reassemble fragments into one packet.
+ */
+struct rte_mbuf *
+ipv6_frag_reassemble(const struct rte_ip_frag_pkt *fp)
+{
+	struct ipv6_hdr *ip_hdr;
+	struct ipv6_extension_fragment * frag_hdr;
+	struct rte_mbuf *m, *prev;
+	uint32_t i, n, ofs, first_len;
+	uint32_t last_len, move_len, payload_len;
+
+	first_len = fp->frags[IP_FIRST_FRAG_IDX].len;
+	n = fp->last_idx - 1;
+
+	/*start from the last fragment. */
+	m = fp->frags[IP_LAST_FRAG_IDX].mb;
+	ofs = fp->frags[IP_LAST_FRAG_IDX].ofs;
+	last_len = fp->frags[IP_LAST_FRAG_IDX].len;
+
+	payload_len = ofs + last_len;
+
+	while (ofs != first_len) {
+
+		prev = m;
+
+		for (i = n; i != IP_FIRST_FRAG_IDX && ofs != first_len; i--) {
+
+			/* previous fragment found. */
+			if(fp->frags[i].ofs + fp->frags[i].len == ofs) {
+
+				ip_frag_chain(fp->frags[i].mb, m);
+
+				/* update our last fragment and offset. */
+				m = fp->frags[i].mb;
+				ofs = fp->frags[i].ofs;
+			}
+		}
+
+		/* error - hole in the packet. */
+		if (m == prev) {
+			return (NULL);
+		}
+	}
+
+	/* chain with the first fragment. */
+	ip_frag_chain(fp->frags[IP_FIRST_FRAG_IDX].mb, m);
+	m = fp->frags[IP_FIRST_FRAG_IDX].mb;
+
+	/* update mbuf fields for reassembled packet. */
+	m->ol_flags |= PKT_TX_IP_CKSUM;
+
+	/* update ipv6 header for the reassembled datagram */
+	ip_hdr = (struct ipv6_hdr *) (rte_pktmbuf_mtod(m, uint8_t *) +
+								  m->pkt.vlan_macip.f.l2_len);
+
+	ip_hdr->payload_len = rte_cpu_to_be_16(payload_len);
+
+	/*
+	 * remove fragmentation header. note that per RFC2460, we need to update
+	 * the last non-fragmentable header with the "next header" field to contain
+	 * type of the first fragmentable header, but we currently don't support
+	 * other headers, so we assume there are no other headers and thus update
+	 * the main IPv6 header instead.
+	 */
+	move_len = m->pkt.vlan_macip.f.l2_len + m->pkt.vlan_macip.f.l3_len -
+			sizeof(*frag_hdr);
+	frag_hdr = (struct ipv6_extension_fragment *) (ip_hdr + 1);
+	ip_hdr->proto = frag_hdr->next_header;
+
+	memmove(rte_pktmbuf_mtod(m, char*) + sizeof(*frag_hdr),
+			rte_pktmbuf_mtod(m, char*), move_len);
+
+	rte_pktmbuf_adj(m, sizeof(*frag_hdr));
+
+	return (m);
+}
+
+/*
+ * Process new mbuf with fragment of IPV6 datagram.
+ * Incoming mbuf should have its l2_len/l3_len fields setup correctly.
+ * @param tbl
+ *   Table where to lookup/add the fragmented packet.
+ * @param mb
+ *   Incoming mbuf with IPV6 fragment.
+ * @param tms
+ *   Fragment arrival timestamp.
+ * @param ip_hdr
+ *   Pointer to the IPV6 header.
+ * @param frag_hdr
+ *   Pointer to the IPV6 fragment extension header.
+ * @return
+ *   Pointer to mbuf for reassembled packet, or NULL if:
+ *   - an error occured.
+ *   - not all fragments of the packet are collected yet.
+ */
+#define MORE_FRAGS(x) (((x) & 0x100) >> 8)
+#define FRAG_OFFSET(x) (rte_cpu_to_be_16(x) >> 3)
+struct rte_mbuf *
+rte_ipv6_frag_reassemble_packet(struct rte_ip_frag_tbl *tbl,
+		struct rte_ip_frag_death_row *dr, struct rte_mbuf *mb, uint64_t tms,
+		struct ipv6_hdr *ip_hdr, struct ipv6_extension_fragment *frag_hdr)
+{
+	struct rte_ip_frag_pkt *fp;
+	struct ip_frag_key key;
+	uint16_t ip_len, ip_ofs;
+
+	rte_memcpy(&key.src_dst[0], ip_hdr->src_addr, 16);
+	rte_memcpy(&key.src_dst[2], ip_hdr->dst_addr, 16);
+
+	key.id = frag_hdr->id;
+	key.key_len = IPV6_KEYLEN;
+
+	ip_ofs = FRAG_OFFSET(frag_hdr->frag_data) * 8;
+
+	/*
+	 * as per RFC2460, payload length contains all extension headers as well.
+	 * since we don't support anything but frag headers, this is what we remove
+	 * from the payload len.
+	 */
+	ip_len = rte_be_to_cpu_16(ip_hdr->payload_len) - sizeof(*frag_hdr);
+
+	IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+		"mbuf: %p, tms: %" PRIu64
+		", key: <" IPv6_KEY_BYTES_FMT ", %#x>, ofs: %u, len: %u, flags: %#x\n"
+		"tbl: %p, max_cycles: %" PRIu64 ", entry_mask: %#x, "
+		"max_entries: %u, use_entries: %u\n\n",
+		__func__, __LINE__,
+		mb, tms, IPv6_KEY_BYTES(key.src_dst), key.id, ip_ofs, ip_len, frag_hdr->more_frags,
+		tbl, tbl->max_cycles, tbl->entry_mask, tbl->max_entries,
+		tbl->use_entries);
+
+	/* try to find/add entry into the fragment's table. */
+	if ((fp = ip_frag_find(tbl, dr, &key, tms)) == NULL) {
+		IP_FRAG_MBUF2DR(dr, mb);
+		return (NULL);
+	}
+
+	IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+		"tbl: %p, max_entries: %u, use_entries: %u\n"
+		"ipv6_frag_pkt: %p, key: <" IPv6_KEY_BYTES_FMT ", %#x>, start: %" PRIu64
+		", total_size: %u, frag_size: %u, last_idx: %u\n\n",
+		__func__, __LINE__,
+		tbl, tbl->max_entries, tbl->use_entries,
+		fp, IPv6_KEY_BYTES(fp->key.src_dst), fp->key.id, fp->start,
+		fp->total_size, fp->frag_size, fp->last_idx);
+
+
+	/* process the fragmented packet. */
+	mb = ip_frag_process(fp, dr, mb, ip_ofs, ip_len,
+			MORE_FRAGS(frag_hdr->frag_data));
+	ip_frag_inuse(tbl, fp);
+
+	IP_FRAG_LOG(DEBUG, "%s:%d:\n"
+		"mbuf: %p\n"
+		"tbl: %p, max_entries: %u, use_entries: %u\n"
+		"ipv6_frag_pkt: %p, key: <" IPv6_KEY_BYTES_FMT ", %#x>, start: %" PRIu64
+		", total_size: %u, frag_size: %u, last_idx: %u\n\n",
+		__func__, __LINE__, mb,
+		tbl, tbl->max_entries, tbl->use_entries,
+		fp, IPv6_KEY_BYTES(fp->key.src_dst), fp->key.id, fp->start,
+		fp->total_size, fp->frag_size, fp->last_idx);
+
+	return (mb);
+}
-- 
1.8.1.4