Re: [dpdk-dev] [PATCH] examples/l3fwd: increase lookup burst size to 8

["Qiu, Michael" <michael.qiu@intel.com>]

Hi, Pablo

Is there any performance data for this change?

Thanks,
Michael

-----Original Message-----
From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Pablo de Lara
Sent: Thursday, July 23, 2015 9:12 AM
To: dev@dpdk.org
Subject: [dpdk-dev] [PATCH] examples/l3fwd: increase lookup burst size to 8

With the new hash implementation, the minimum lookup burst size to get good performance is 8, since its internal pipeline consists of 4 stages of 2 entries each, so to avoid duplication, burst size should be 8 or more entries.

Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
---
 examples/l3fwd/main.c | 234 ++++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 191 insertions(+), 43 deletions(-)

diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c index 45676ba..c8a0f66 100644
--- a/examples/l3fwd/main.c
+++ b/examples/l3fwd/main.c
@@ -1,7 +1,7 @@
 /*-
  *   BSD LICENSE
  *
- *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ *   Copyright(c) 2010-2015 Intel Corporation. All rights reserved.
  *   All rights reserved.
  *
  *   Redistribution and use in source and binary forms, with or without
@@ -731,26 +731,34 @@ static inline void l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid,  #if ((APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH) && \
 	(ENABLE_MULTI_BUFFER_OPTIMIZE == 1))
 
-#define MASK_ALL_PKTS    0xf
-#define EXECLUDE_1ST_PKT 0xe
-#define EXECLUDE_2ND_PKT 0xd
-#define EXECLUDE_3RD_PKT 0xb
-#define EXECLUDE_4TH_PKT 0x7
+#define MASK_ALL_PKTS    0xff
+#define EXCLUDE_1ST_PKT 0xfe
+#define EXCLUDE_2ND_PKT 0xfd
+#define EXCLUDE_3RD_PKT 0xfb
+#define EXCLUDE_4TH_PKT 0xf7
+#define EXCLUDE_5TH_PKT 0xef
+#define EXCLUDE_6TH_PKT 0xdf
+#define EXCLUDE_7TH_PKT 0xbf
+#define EXCLUDE_8TH_PKT 0x7f
 
 static inline void
-simple_ipv4_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *qconf)
+simple_ipv4_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid, struct 
+lcore_conf *qconf)
 {
-	struct ether_hdr *eth_hdr[4];
-	struct ipv4_hdr *ipv4_hdr[4];
-	uint8_t dst_port[4];
-	int32_t ret[4];
-	union ipv4_5tuple_host key[4];
-	__m128i data[4];
+	struct ether_hdr *eth_hdr[8];
+	struct ipv4_hdr *ipv4_hdr[8];
+	uint8_t dst_port[8];
+	int32_t ret[8];
+	union ipv4_5tuple_host key[8];
+	__m128i data[8];
 
 	eth_hdr[0] = rte_pktmbuf_mtod(m[0], struct ether_hdr *);
 	eth_hdr[1] = rte_pktmbuf_mtod(m[1], struct ether_hdr *);
 	eth_hdr[2] = rte_pktmbuf_mtod(m[2], struct ether_hdr *);
 	eth_hdr[3] = rte_pktmbuf_mtod(m[3], struct ether_hdr *);
+	eth_hdr[4] = rte_pktmbuf_mtod(m[4], struct ether_hdr *);
+	eth_hdr[5] = rte_pktmbuf_mtod(m[5], struct ether_hdr *);
+	eth_hdr[6] = rte_pktmbuf_mtod(m[6], struct ether_hdr *);
+	eth_hdr[7] = rte_pktmbuf_mtod(m[7], struct ether_hdr *);
 
 	/* Handle IPv4 headers.*/
 	ipv4_hdr[0] = rte_pktmbuf_mtod_offset(m[0], struct ipv4_hdr *, @@ -761,32 +769,56 @@ simple_ipv4_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 					      sizeof(struct ether_hdr));
 	ipv4_hdr[3] = rte_pktmbuf_mtod_offset(m[3], struct ipv4_hdr *,
 					      sizeof(struct ether_hdr));
+	ipv4_hdr[4] = rte_pktmbuf_mtod_offset(m[4], struct ipv4_hdr *,
+					      sizeof(struct ether_hdr));
+	ipv4_hdr[5] = rte_pktmbuf_mtod_offset(m[5], struct ipv4_hdr *,
+					      sizeof(struct ether_hdr));
+	ipv4_hdr[6] = rte_pktmbuf_mtod_offset(m[6], struct ipv4_hdr *,
+					      sizeof(struct ether_hdr));
+	ipv4_hdr[7] = rte_pktmbuf_mtod_offset(m[7], struct ipv4_hdr *,
+					      sizeof(struct ether_hdr));
 
 #ifdef DO_RFC_1812_CHECKS
 	/* Check to make sure the packet is valid (RFC1812) */
 	uint8_t valid_mask = MASK_ALL_PKTS;
 	if (is_valid_ipv4_pkt(ipv4_hdr[0], m[0]->pkt_len) < 0) {
 		rte_pktmbuf_free(m[0]);
-		valid_mask &= EXECLUDE_1ST_PKT;
+		valid_mask &= EXCLUDE_1ST_PKT;
 	}
 	if (is_valid_ipv4_pkt(ipv4_hdr[1], m[1]->pkt_len) < 0) {
 		rte_pktmbuf_free(m[1]);
-		valid_mask &= EXECLUDE_2ND_PKT;
+		valid_mask &= EXCLUDE_2ND_PKT;
 	}
 	if (is_valid_ipv4_pkt(ipv4_hdr[2], m[2]->pkt_len) < 0) {
 		rte_pktmbuf_free(m[2]);
-		valid_mask &= EXECLUDE_3RD_PKT;
+		valid_mask &= EXCLUDE_3RD_PKT;
 	}
 	if (is_valid_ipv4_pkt(ipv4_hdr[3], m[3]->pkt_len) < 0) {
 		rte_pktmbuf_free(m[3]);
-		valid_mask &= EXECLUDE_4TH_PKT;
+		valid_mask &= EXCLUDE_4TH_PKT;
+	}
+	if (is_valid_ipv4_pkt(ipv4_hdr[4], m[4]->pkt_len) < 0) {
+		rte_pktmbuf_free(m[4]);
+		valid_mask &= EXCLUDE_5TH_PKT;
+	}
+	if (is_valid_ipv4_pkt(ipv4_hdr[5], m[5]->pkt_len) < 0) {
+		rte_pktmbuf_free(m[5]);
+		valid_mask &= EXCLUDE_6TH_PKT;
+	}
+	if (is_valid_ipv4_pkt(ipv4_hdr[6], m[6]->pkt_len) < 0) {
+		rte_pktmbuf_free(m[6]);
+		valid_mask &= EXCLUDE_7TH_PKT;
+	}
+	if (is_valid_ipv4_pkt(ipv4_hdr[7], m[7]->pkt_len) < 0) {
+		rte_pktmbuf_free(m[7]);
+		valid_mask &= EXCLUDE_8TH_PKT;
 	}
 	if (unlikely(valid_mask != MASK_ALL_PKTS)) {
 		if (valid_mask == 0){
 			return;
 		} else {
 			uint8_t i = 0;
-			for (i = 0; i < 4; i++) {
+			for (i = 0; i < 8; i++) {
 				if ((0x1 << i) & valid_mask) {
 					l3fwd_simple_forward(m[i], portid, qconf);
 				}
@@ -796,22 +828,52 @@ simple_ipv4_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 	}
 #endif // End of #ifdef DO_RFC_1812_CHECKS
 
-	data[0] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[0], __m128i *, sizeof(struct ether_hdr) + offsetof(struct ipv4_hdr, time_to_live)));
-	data[1] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[1], __m128i *, sizeof(struct ether_hdr) + offsetof(struct ipv4_hdr, time_to_live)));
-	data[2] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[2], __m128i *, sizeof(struct ether_hdr) + offsetof(struct ipv4_hdr, time_to_live)));
-	data[3] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[3], __m128i *, sizeof(struct ether_hdr) + offsetof(struct ipv4_hdr, time_to_live)));
+	data[0] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[0], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[1] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[1], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[2] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[2], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[3] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[3], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[4] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[4], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[5] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[5], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[6] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[6], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
+	data[7] = _mm_loadu_si128(rte_pktmbuf_mtod_offset(m[7], __m128i *,
+					sizeof(struct ether_hdr) +
+					offsetof(struct ipv4_hdr, time_to_live)));
 
 	key[0].xmm = _mm_and_si128(data[0], mask0);
 	key[1].xmm = _mm_and_si128(data[1], mask0);
 	key[2].xmm = _mm_and_si128(data[2], mask0);
 	key[3].xmm = _mm_and_si128(data[3], mask0);
+	key[4].xmm = _mm_and_si128(data[4], mask0);
+	key[5].xmm = _mm_and_si128(data[5], mask0);
+	key[6].xmm = _mm_and_si128(data[6], mask0);
+	key[7].xmm = _mm_and_si128(data[7], mask0);
+
+	const void *key_array[8] = {&key[0], &key[1], &key[2], &key[3],
+				&key[4], &key[5], &key[6], &key[7]};
 
-	const void *key_array[4] = {&key[0], &key[1], &key[2],&key[3]};
-	rte_hash_lookup_multi(qconf->ipv4_lookup_struct, &key_array[0], 4, ret);
+	rte_hash_lookup_multi(qconf->ipv4_lookup_struct, &key_array[0], 8, 
+ret);
 	dst_port[0] = (uint8_t) ((ret[0] < 0) ? portid : ipv4_l3fwd_out_if[ret[0]]);
 	dst_port[1] = (uint8_t) ((ret[1] < 0) ? portid : ipv4_l3fwd_out_if[ret[1]]);
 	dst_port[2] = (uint8_t) ((ret[2] < 0) ? portid : ipv4_l3fwd_out_if[ret[2]]);
 	dst_port[3] = (uint8_t) ((ret[3] < 0) ? portid : ipv4_l3fwd_out_if[ret[3]]);
+	dst_port[4] = (uint8_t) ((ret[4] < 0) ? portid : ipv4_l3fwd_out_if[ret[4]]);
+	dst_port[5] = (uint8_t) ((ret[5] < 0) ? portid : ipv4_l3fwd_out_if[ret[5]]);
+	dst_port[6] = (uint8_t) ((ret[6] < 0) ? portid : ipv4_l3fwd_out_if[ret[6]]);
+	dst_port[7] = (uint8_t) ((ret[7] < 0) ? portid : 
+ipv4_l3fwd_out_if[ret[7]]);
 
 	if (dst_port[0] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[0]) == 0)
 		dst_port[0] = portid;
@@ -821,6 +883,14 @@ simple_ipv4_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 		dst_port[2] = portid;
 	if (dst_port[3] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[3]) == 0)
 		dst_port[3] = portid;
+	if (dst_port[4] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[4]) == 0)
+		dst_port[4] = portid;
+	if (dst_port[5] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[5]) == 0)
+		dst_port[5] = portid;
+	if (dst_port[6] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[6]) == 0)
+		dst_port[6] = portid;
+	if (dst_port[7] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[7]) == 0)
+		dst_port[7] = portid;
 
 #ifdef DO_RFC_1812_CHECKS
 	/* Update time to live and header checksum */ @@ -832,6 +902,14 @@ simple_ipv4_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 	++(ipv4_hdr[1]->hdr_checksum);
 	++(ipv4_hdr[2]->hdr_checksum);
 	++(ipv4_hdr[3]->hdr_checksum);
+	--(ipv4_hdr[4]->time_to_live);
+	--(ipv4_hdr[5]->time_to_live);
+	--(ipv4_hdr[6]->time_to_live);
+	--(ipv4_hdr[7]->time_to_live);
+	++(ipv4_hdr[4]->hdr_checksum);
+	++(ipv4_hdr[5]->hdr_checksum);
+	++(ipv4_hdr[6]->hdr_checksum);
+	++(ipv4_hdr[7]->hdr_checksum);
 #endif
 
 	/* dst addr */
@@ -839,17 +917,29 @@ simple_ipv4_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 	*(uint64_t *)&eth_hdr[1]->d_addr = dest_eth_addr[dst_port[1]];
 	*(uint64_t *)&eth_hdr[2]->d_addr = dest_eth_addr[dst_port[2]];
 	*(uint64_t *)&eth_hdr[3]->d_addr = dest_eth_addr[dst_port[3]];
+	*(uint64_t *)&eth_hdr[4]->d_addr = dest_eth_addr[dst_port[4]];
+	*(uint64_t *)&eth_hdr[5]->d_addr = dest_eth_addr[dst_port[5]];
+	*(uint64_t *)&eth_hdr[6]->d_addr = dest_eth_addr[dst_port[6]];
+	*(uint64_t *)&eth_hdr[7]->d_addr = dest_eth_addr[dst_port[7]];
 
 	/* src addr */
 	ether_addr_copy(&ports_eth_addr[dst_port[0]], &eth_hdr[0]->s_addr);
 	ether_addr_copy(&ports_eth_addr[dst_port[1]], &eth_hdr[1]->s_addr);
 	ether_addr_copy(&ports_eth_addr[dst_port[2]], &eth_hdr[2]->s_addr);
 	ether_addr_copy(&ports_eth_addr[dst_port[3]], &eth_hdr[3]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[4]], &eth_hdr[4]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[5]], &eth_hdr[5]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[6]], &eth_hdr[6]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[7]], &eth_hdr[7]->s_addr);
 
 	send_single_packet(m[0], (uint8_t)dst_port[0]);
 	send_single_packet(m[1], (uint8_t)dst_port[1]);
 	send_single_packet(m[2], (uint8_t)dst_port[2]);
 	send_single_packet(m[3], (uint8_t)dst_port[3]);
+	send_single_packet(m[4], (uint8_t)dst_port[4]);
+	send_single_packet(m[5], (uint8_t)dst_port[5]);
+	send_single_packet(m[6], (uint8_t)dst_port[6]);
+	send_single_packet(m[7], (uint8_t)dst_port[7]);
 
 }
 
@@ -866,18 +956,22 @@ static inline void get_ipv6_5tuple(struct rte_mbuf* m0, __m128i mask0, __m128i m  }
 
 static inline void
-simple_ipv6_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *qconf)
+simple_ipv6_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid, struct 
+lcore_conf *qconf)
 {
-	struct ether_hdr *eth_hdr[4];
-	__attribute__((unused)) struct ipv6_hdr *ipv6_hdr[4];
-	uint8_t dst_port[4];
-	int32_t ret[4];
-	union ipv6_5tuple_host key[4];
+	struct ether_hdr *eth_hdr[8];
+	__attribute__((unused)) struct ipv6_hdr *ipv6_hdr[8];
+	uint8_t dst_port[8];
+	int32_t ret[8];
+	union ipv6_5tuple_host key[8];
 
 	eth_hdr[0] = rte_pktmbuf_mtod(m[0], struct ether_hdr *);
 	eth_hdr[1] = rte_pktmbuf_mtod(m[1], struct ether_hdr *);
 	eth_hdr[2] = rte_pktmbuf_mtod(m[2], struct ether_hdr *);
 	eth_hdr[3] = rte_pktmbuf_mtod(m[3], struct ether_hdr *);
+	eth_hdr[4] = rte_pktmbuf_mtod(m[4], struct ether_hdr *);
+	eth_hdr[5] = rte_pktmbuf_mtod(m[5], struct ether_hdr *);
+	eth_hdr[6] = rte_pktmbuf_mtod(m[6], struct ether_hdr *);
+	eth_hdr[7] = rte_pktmbuf_mtod(m[7], struct ether_hdr *);
 
 	/* Handle IPv6 headers.*/
 	ipv6_hdr[0] = rte_pktmbuf_mtod_offset(m[0], struct ipv6_hdr *, @@ -888,18 +982,36 @@ simple_ipv6_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 					      sizeof(struct ether_hdr));
 	ipv6_hdr[3] = rte_pktmbuf_mtod_offset(m[3], struct ipv6_hdr *,
 					      sizeof(struct ether_hdr));
+	ipv6_hdr[4] = rte_pktmbuf_mtod_offset(m[4], struct ipv6_hdr *,
+					      sizeof(struct ether_hdr));
+	ipv6_hdr[5] = rte_pktmbuf_mtod_offset(m[5], struct ipv6_hdr *,
+					      sizeof(struct ether_hdr));
+	ipv6_hdr[6] = rte_pktmbuf_mtod_offset(m[6], struct ipv6_hdr *,
+					      sizeof(struct ether_hdr));
+	ipv6_hdr[7] = rte_pktmbuf_mtod_offset(m[7], struct ipv6_hdr *,
+					      sizeof(struct ether_hdr));
 
 	get_ipv6_5tuple(m[0], mask1, mask2, &key[0]);
 	get_ipv6_5tuple(m[1], mask1, mask2, &key[1]);
 	get_ipv6_5tuple(m[2], mask1, mask2, &key[2]);
 	get_ipv6_5tuple(m[3], mask1, mask2, &key[3]);
+	get_ipv6_5tuple(m[4], mask1, mask2, &key[4]);
+	get_ipv6_5tuple(m[5], mask1, mask2, &key[5]);
+	get_ipv6_5tuple(m[6], mask1, mask2, &key[6]);
+	get_ipv6_5tuple(m[7], mask1, mask2, &key[7]);
+
+	const void *key_array[8] = {&key[0], &key[1], &key[2], &key[3],
+				&key[4], &key[5], &key[6], &key[7]};
 
-	const void *key_array[4] = {&key[0], &key[1], &key[2],&key[3]};
 	rte_hash_lookup_multi(qconf->ipv6_lookup_struct, &key_array[0], 4, ret);
-	dst_port[0] = (uint8_t) ((ret[0] < 0)? portid:ipv6_l3fwd_out_if[ret[0]]);
-	dst_port[1] = (uint8_t) ((ret[1] < 0)? portid:ipv6_l3fwd_out_if[ret[1]]);
-	dst_port[2] = (uint8_t) ((ret[2] < 0)? portid:ipv6_l3fwd_out_if[ret[2]]);
-	dst_port[3] = (uint8_t) ((ret[3] < 0)? portid:ipv6_l3fwd_out_if[ret[3]]);
+	dst_port[0] = (uint8_t) ((ret[0] < 0) ? portid:ipv6_l3fwd_out_if[ret[0]]);
+	dst_port[1] = (uint8_t) ((ret[1] < 0) ? portid:ipv6_l3fwd_out_if[ret[1]]);
+	dst_port[2] = (uint8_t) ((ret[2] < 0) ? portid:ipv6_l3fwd_out_if[ret[2]]);
+	dst_port[3] = (uint8_t) ((ret[3] < 0) ? portid:ipv6_l3fwd_out_if[ret[3]]);
+	dst_port[4] = (uint8_t) ((ret[4] < 0) ? portid:ipv6_l3fwd_out_if[ret[4]]);
+	dst_port[5] = (uint8_t) ((ret[5] < 0) ? portid:ipv6_l3fwd_out_if[ret[5]]);
+	dst_port[6] = (uint8_t) ((ret[6] < 0) ? portid:ipv6_l3fwd_out_if[ret[6]]);
+	dst_port[7] = (uint8_t) ((ret[7] < 0) ? 
+portid:ipv6_l3fwd_out_if[ret[7]]);
 
 	if (dst_port[0] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[0]) == 0)
 		dst_port[0] = portid;
@@ -909,23 +1021,43 @@ simple_ipv6_fwd_4pkts(struct rte_mbuf* m[4], uint8_t portid, struct lcore_conf *
 		dst_port[2] = portid;
 	if (dst_port[3] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[3]) == 0)
 		dst_port[3] = portid;
+	if (dst_port[4] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[4]) == 0)
+		dst_port[4] = portid;
+	if (dst_port[5] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[5]) == 0)
+		dst_port[5] = portid;
+	if (dst_port[6] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[6]) == 0)
+		dst_port[6] = portid;
+	if (dst_port[7] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[7]) == 0)
+		dst_port[7] = portid;
 
 	/* dst addr */
 	*(uint64_t *)&eth_hdr[0]->d_addr = dest_eth_addr[dst_port[0]];
 	*(uint64_t *)&eth_hdr[1]->d_addr = dest_eth_addr[dst_port[1]];
 	*(uint64_t *)&eth_hdr[2]->d_addr = dest_eth_addr[dst_port[2]];
 	*(uint64_t *)&eth_hdr[3]->d_addr = dest_eth_addr[dst_port[3]];
+	*(uint64_t *)&eth_hdr[4]->d_addr = dest_eth_addr[dst_port[4]];
+	*(uint64_t *)&eth_hdr[5]->d_addr = dest_eth_addr[dst_port[5]];
+	*(uint64_t *)&eth_hdr[6]->d_addr = dest_eth_addr[dst_port[6]];
+	*(uint64_t *)&eth_hdr[7]->d_addr = dest_eth_addr[dst_port[7]];
 
 	/* src addr */
 	ether_addr_copy(&ports_eth_addr[dst_port[0]], &eth_hdr[0]->s_addr);
 	ether_addr_copy(&ports_eth_addr[dst_port[1]], &eth_hdr[1]->s_addr);
 	ether_addr_copy(&ports_eth_addr[dst_port[2]], &eth_hdr[2]->s_addr);
 	ether_addr_copy(&ports_eth_addr[dst_port[3]], &eth_hdr[3]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[4]], &eth_hdr[4]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[5]], &eth_hdr[5]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[6]], &eth_hdr[6]->s_addr);
+	ether_addr_copy(&ports_eth_addr[dst_port[7]], &eth_hdr[7]->s_addr);
 
 	send_single_packet(m[0], (uint8_t)dst_port[0]);
 	send_single_packet(m[1], (uint8_t)dst_port[1]);
 	send_single_packet(m[2], (uint8_t)dst_port[2]);
 	send_single_packet(m[3], (uint8_t)dst_port[3]);
+	send_single_packet(m[4], (uint8_t)dst_port[4]);
+	send_single_packet(m[5], (uint8_t)dst_port[5]);
+	send_single_packet(m[6], (uint8_t)dst_port[6]);
+	send_single_packet(m[7], (uint8_t)dst_port[7]);
 
 }
 #endif /* APP_LOOKUP_METHOD */
@@ -1548,19 +1680,23 @@ main_loop(__attribute__((unused)) void *dummy)  #if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH)
 			{
 				/*
-				 * Send nb_rx - nb_rx%4 packets
-				 * in groups of 4.
+				 * Send nb_rx - nb_rx%8 packets
+				 * in groups of 8.
 				 */
-				int32_t n = RTE_ALIGN_FLOOR(nb_rx, 4);
-				for (j = 0; j < n ; j+=4) {
+				int32_t n = RTE_ALIGN_FLOOR(nb_rx, 8);
+				for (j = 0; j < n; j+=8) {
 #ifdef RTE_NEXT_ABI
 					uint32_t pkt_type =
 						pkts_burst[j]->packet_type &
 						pkts_burst[j+1]->packet_type &
 						pkts_burst[j+2]->packet_type &
-						pkts_burst[j+3]->packet_type;
+						pkts_burst[j+3]->packet_type &
+						pkts_burst[j+4]->packet_type &
+						pkts_burst[j+5]->packet_type &
+						pkts_burst[j+6]->packet_type &
+						pkts_burst[j+7]->packet_type;
 					if (pkt_type & RTE_PTYPE_L3_IPV4) {
-						simple_ipv4_fwd_4pkts(
+						simple_ipv4_fwd_8pkts(
 						&pkts_burst[j], portid, qconf);
 					} else if (pkt_type &
 						RTE_PTYPE_L3_IPV6) {
@@ -1568,9 +1704,13 @@ main_loop(__attribute__((unused)) void *dummy)
 					uint32_t ol_flag = pkts_burst[j]->ol_flags
 							& pkts_burst[j+1]->ol_flags
 							& pkts_burst[j+2]->ol_flags
-							& pkts_burst[j+3]->ol_flags;
+							& pkts_burst[j+3]->ol_flags
+							& pkts_burst[j+4]->ol_flags
+							& pkts_burst[j+5]->ol_flags
+							& pkts_burst[j+6]->ol_flags
+							& pkts_burst[j+7]->ol_flags;
 					if (ol_flag & PKT_RX_IPV4_HDR ) {
-						simple_ipv4_fwd_4pkts(&pkts_burst[j],
+						simple_ipv8_fwd_4pkts(&pkts_burst[j],
 									portid, qconf);
 					} else if (ol_flag & PKT_RX_IPV6_HDR) {  #endif /* RTE_NEXT_ABI */ @@ -1585,6 +1725,14 @@ main_loop(__attribute__((unused)) void *dummy)
 									portid, qconf);
 						l3fwd_simple_forward(pkts_burst[j+3],
 									portid, qconf);
+						l3fwd_simple_forward(pkts_burst[j+4],
+									portid, qconf);
+						l3fwd_simple_forward(pkts_burst[j+5],
+									portid, qconf);
+						l3fwd_simple_forward(pkts_burst[j+6],
+									portid, qconf);
+						l3fwd_simple_forward(pkts_burst[j+7],
+									portid, qconf);
 					}
 				}
 				for (; j < nb_rx ; j++) {
--
2.4.2