Improve performance in IPv6 generator

Change-Id: I3abc92652402162d15249e345446debdeb778176
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>

diff --git a/VNFs/DPPD-PROX/handle_swap.c b/VNFs/DPPD-PROX/handle_swap.c
index e097bc0..094199e 100644 (file)
--- a/VNFs/DPPD-PROX/handle_swap.c
+++ b/VNFs/DPPD-PROX/handle_swap.c
@@ -61,7 +61,7 @@ static void write_src_and_dst_mac(struct task_swap *task, struct rte_mbuf *mbuf)
                rte_memcpy(hdr, task->src_dst_mac, sizeof(task->src_dst_mac));
        } else {
                hdr = rte_pktmbuf_mtod(mbuf, prox_rte_ether_hdr *);
-               if (likely((task->runtime_flags & TASK_ARG_SRC_MAC_SET) == 0)) {
+               if (unlikely((task->runtime_flags & TASK_ARG_SRC_MAC_SET) == 0)) {
                        /* dst mac will be used as src mac */
                        prox_rte_ether_addr_copy(&hdr->d_addr, &mac);
                }
@@ -71,7 +71,7 @@ static void write_src_and_dst_mac(struct task_swap *task, struct rte_mbuf *mbuf)
                else
                        prox_rte_ether_addr_copy(&hdr->s_addr, &hdr->d_addr);
 
-               if (unlikely(task->runtime_flags & TASK_ARG_SRC_MAC_SET)) {
+               if (likely(task->runtime_flags & TASK_ARG_SRC_MAC_SET)) {
                        prox_rte_ether_addr_copy((prox_rte_ether_addr *)&task->src_dst_mac[6], &hdr->s_addr);
                } else {
                        prox_rte_ether_addr_copy(&mac, &hdr->s_addr);

diff --git a/VNFs/DPPD-PROX/packet_utils.c b/VNFs/DPPD-PROX/packet_utils.c
index 7be978d..bb94299 100644 (file)
--- a/VNFs/DPPD-PROX/packet_utils.c
+++ b/VNFs/DPPD-PROX/packet_utils.c
@@ -46,7 +46,7 @@ static inline int find_ip(struct ether_hdr_arp *pkt, uint16_t len, uint32_t *ip_
 
        *vlan = 0;
        // Unstack VLAN tags
-       while (((ether_type == ETYPE_8021ad) || (ether_type == ETYPE_VLAN)) && (l2_len + sizeof(prox_rte_vlan_hdr) < len)) {
+       while (((ether_type == ETYPE_VLAN) || (ether_type == ETYPE_8021ad)) && (l2_len + sizeof(prox_rte_vlan_hdr) < len)) {
                vlan_hdr = (prox_rte_vlan_hdr *)((uint8_t *)pkt + l2_len);
                l2_len +=4;
                ether_type = vlan_hdr->eth_proto;
@@ -100,40 +100,23 @@ static inline void find_vlan(struct ether_hdr_arp *pkt, uint16_t len, uint16_t *
 
 static inline struct ipv6_addr *find_ip6(prox_rte_ether_hdr *pkt, uint16_t len, struct ipv6_addr *ip_dst, uint16_t *vlan)
 {
-       prox_rte_vlan_hdr *vlan_hdr;
-       prox_rte_ipv6_hdr *ip;
        uint16_t ether_type = pkt->ether_type;
        uint16_t l2_len = sizeof(prox_rte_ether_hdr);
-
        *vlan = 0;
-       // Unstack VLAN tags
-       while (((ether_type == ETYPE_8021ad) || (ether_type == ETYPE_VLAN)) && (l2_len + sizeof(prox_rte_vlan_hdr) < len)) {
-               vlan_hdr = (prox_rte_vlan_hdr *)((uint8_t *)pkt + l2_len);
-               l2_len +=4;
-               ether_type = vlan_hdr->eth_proto;
-               *vlan = rte_be_to_cpu_16(vlan_hdr->vlan_tci & 0xFF0F);  // Store VLAN, or CVLAN if QinQ
-       }
 
-       switch (ether_type) {
-       case ETYPE_MPLSU:
-       case ETYPE_MPLSM:
-               // In case of MPLS, next hop MAC is based on MPLS, not destination IP
-               l2_len = 0;
-               break;
-       case ETYPE_IPv4:
-       case ETYPE_EoGRE:
-       case ETYPE_ARP:
-               l2_len = 0;
-               break;
-       case ETYPE_IPv6:
-               break;
-       default:
-               l2_len = 0;
-               plog_warn("Unsupported packet type %x - CRC might be wrong\n", ether_type);
-               break;
+       if ((ether_type == ETYPE_VLAN) || (ether_type == ETYPE_8021ad)) {
+               prox_rte_vlan_hdr *vlan_hdr = (prox_rte_vlan_hdr *)((uint8_t *)pkt + l2_len);
+               ether_type = vlan_hdr->eth_proto;
+               l2_len +=4;
+               *vlan = rte_be_to_cpu_16(vlan_hdr->vlan_tci & 0xFF0F);
+               if (ether_type == ETYPE_VLAN) {
+                       vlan_hdr = (prox_rte_vlan_hdr *)(vlan_hdr + 1);
+                       ether_type = vlan_hdr->eth_proto;
+                       l2_len +=4;
+                       *vlan = rte_be_to_cpu_16(vlan_hdr->vlan_tci & 0xFF0F);
+               }
        }
-
-       if (l2_len && (l2_len + sizeof(prox_rte_ipv6_hdr) <= len)) {
+       if ((ether_type == ETYPE_IPv6) && (l2_len + sizeof(prox_rte_ipv6_hdr) <= len)) {
                prox_rte_ipv6_hdr *ip = (prox_rte_ipv6_hdr *)((uint8_t *)pkt + l2_len);
                // TODO: implement LPM => replace ip_dst by next hop IP DST
                memcpy(ip_dst, &ip->dst_addr, sizeof(struct ipv6_addr));
@@ -148,7 +131,7 @@ void send_unsollicited_neighbour_advertisement(struct task_base *tbase)
        uint8_t out = 0, port_id = tbase->l3.reachable_port_id;
        struct rte_mbuf *mbuf = NULL;
 
-       if (memcmp(&tbase->l3.local_ipv6, &null_addr, 16) != 0) {
+       if (*(__int128 *)(&tbase->l3.local_ipv6) != 0) {
                ret = rte_mempool_get(tbase->l3.arp_nd_pool, (void **)&mbuf);
                if (likely(ret == 0)) {
                        mbuf->port = port_id;
@@ -160,7 +143,7 @@ void send_unsollicited_neighbour_advertisement(struct task_base *tbase)
                        return;
                }
        }
-       if (memcmp(&tbase->l3.global_ipv6, &null_addr, 16) != 0) {
+       if (*(__int128 *)(&tbase->l3.global_ipv6) != 0) {
                ret = rte_mempool_get(tbase->l3.arp_nd_pool, (void **)&mbuf);
                if (likely(ret == 0)) {
                        mbuf->port = port_id;
@@ -365,14 +348,13 @@ int write_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, uint32_t *ip_d
        return DROP_MBUF;
 }
 
-int write_ip6_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, struct ipv6_addr *ip_dst, uint16_t *vlan)
+int write_ip6_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, struct ipv6_addr *ip_dst, uint16_t *vlan, uint64_t tsc)
 {
        const uint64_t hz = rte_get_tsc_hz();
        prox_rte_ether_hdr *packet = rte_pktmbuf_mtod(mbuf, prox_rte_ether_hdr *);
        prox_rte_ether_addr *mac = &packet->d_addr;
        struct ipv6_addr *used_ip_src;
 
-       uint64_t tsc = rte_rdtsc();
        uint16_t len = rte_pktmbuf_pkt_len(mbuf);
 
        struct ipv6_addr *pkt_src_ip6;
@@ -383,33 +365,33 @@ int write_ip6_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, struct ipv
        struct l3_base *l3 = &(tbase->l3);
 
        // Configure source IP
-       if (memcmp(&l3->local_ipv6, ip_dst, 8) == 0) {
+       if (*(uint64_t *)(&l3->local_ipv6) == *(uint64_t *)ip_dst) {
                // Same prefix as local -> use local
                used_ip_src = &l3->local_ipv6;
-       } else if (memcmp(&l3->global_ipv6 , ip_dst, 8) == 0) {
+       } else if (*(uint64_t *)(&l3->global_ipv6) == *(uint64_t *)ip_dst) {
                // Same prefix as global -> use global
                used_ip_src = &l3->global_ipv6;
-       } else if (memcmp(&l3->gw.ip6 , &null_addr, sizeof(struct ipv6_addr)) != 0) {
+       } else if (*(__int128 *)(&l3->gw.ip6) != 0) {
                used_ip_src = &l3->global_ipv6;
                memcpy(ip_dst, &l3->gw.ip6, sizeof(struct ipv6_addr));
-       } else if (memcmp(&l3->global_ipv6 , &null_addr, sizeof(struct ipv6_addr)) != 0) {
+       } else if (*(__int128 *)(&l3->global_ipv6) != 0) {
                // Global IP is defined -> use it
                used_ip_src = &l3->global_ipv6;
        } else {
                plog_info("Error as trying to send a packet to "IPv6_BYTES_FMT" using "IPv6_BYTES_FMT" (local)\n", IPv6_BYTES(ip_dst->bytes), IPv6_BYTES(l3->local_ipv6.bytes));
                return DROP_MBUF;
        }
-       memcpy(pkt_src_ip6, used_ip_src, sizeof(struct ipv6_addr));
+       rte_memcpy(pkt_src_ip6, used_ip_src, sizeof(struct ipv6_addr));
 
        // Configure dst mac
        if (likely(l3->n_pkts < 4)) {
                for (unsigned int idx = 0; idx < l3->n_pkts; idx++) {
-                       if (memcmp(ip_dst, &l3->optimized_arp_table[idx].ip6, sizeof(struct ipv6_addr)) == 0) {
+                       if (*(__int128 *)ip_dst == *(__int128 *)(&l3->optimized_arp_table[idx].ip6)) {
                                 // IP address already in table
                                if ((tsc < l3->optimized_arp_table[idx].arp_ndp_retransmit_timeout) && (tsc < l3->optimized_arp_table[idx].reachable_timeout)) {
                                        // MAC address was recently updated in table, use it
                                        // plog_dbg("Valid MAC address found => send packet\n");
-                                       memcpy(mac, &l3->optimized_arp_table[idx].mac, sizeof(prox_rte_ether_addr));
+                                       rte_memcpy(mac, &l3->optimized_arp_table[idx].mac, sizeof(prox_rte_ether_addr));
                                        return SEND_MBUF;
                                } else if (tsc > l3->optimized_arp_table[idx].arp_ndp_retransmit_timeout) {
                                        // NDP not sent since a long time, send NDP

diff --git a/VNFs/DPPD-PROX/packet_utils.h b/VNFs/DPPD-PROX/packet_utils.h
index ce54a89..ef15cd2 100644 (file)
--- a/VNFs/DPPD-PROX/packet_utils.h
+++ b/VNFs/DPPD-PROX/packet_utils.h
@@ -79,7 +79,7 @@ struct l3_base {
 void task_init_l3(struct task_base *tbase, struct task_args *targ);
 void task_start_l3(struct task_base *tbase, struct task_args *targ);
 int write_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, uint32_t *ip_dst, uint16_t *vlan, uint64_t **time, uint64_t tsc);
-int write_ip6_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, struct ipv6_addr *ip_dst, uint16_t *vlan);
+int write_ip6_dst_mac(struct task_base *tbase, struct rte_mbuf *mbuf, struct ipv6_addr *ip_dst, uint16_t *vlan, uint64_t tsc);
 void task_set_gateway_ip(struct task_base *tbase, uint32_t ip);
 void task_set_local_ip(struct task_base *tbase, uint32_t ip);
 void handle_ctrl_plane_pkts(struct task_base *tbase, struct rte_mbuf **mbufs, uint16_t n_pkts);

diff --git a/VNFs/DPPD-PROX/tx_pkt.c b/VNFs/DPPD-PROX/tx_pkt.c
index d93c5fb..551056a 100644 (file)
--- a/VNFs/DPPD-PROX/tx_pkt.c
+++ b/VNFs/DPPD-PROX/tx_pkt.c
@@ -66,11 +66,12 @@ int tx_pkt_ndp(struct task_base *tbase, struct rte_mbuf **mbufs, uint16_t n_pkts
        const struct port_queue *port_queue = &tbase->tx_params_hw.tx_port_queue[0];
        struct rte_mbuf *mbuf = NULL;       // used when one need to send both an ARP and a mbuf
        uint16_t vlan;
+       uint64_t tsc = rte_rdtsc();
 
        for (int j = 0; j < n_pkts; j++) {
                if ((out) && (out[j] >= OUT_HANDLED))
                        continue;
-               if (unlikely((rc = write_ip6_dst_mac(tbase, mbufs[j], &ip_dst, &vlan)) != SEND_MBUF)) {
+               if (unlikely((rc = write_ip6_dst_mac(tbase, mbufs[j], &ip_dst, &vlan, tsc)) != SEND_MBUF)) {
                        if (j - first) {
                                ret = tbase->aux->tx_pkt_l2(tbase, mbufs + first, j - first, out);
                                ok += ret;