#include <errno.h>
#include <getopt.h>
-#include <tmmintrin.h>
#include <rte_common.h>
+#include <rte_vect.h>
#include <rte_byteorder.h>
#include <rte_log.h>
#include <rte_memory.h>
#include <rte_memcpy.h>
#include <rte_memzone.h>
-#include <rte_tailq.h>
#include <rte_eal.h>
#include <rte_per_lcore.h>
#include <rte_launch.h>
#include <rte_udp.h>
#include <rte_string_fns.h>
-#include "main.h"
-
#define APP_LOOKUP_EXACT_MATCH 0
#define APP_LOOKUP_LPM 1
#define DO_RFC_1812_CHECKS
#define APP_LOOKUP_METHOD APP_LOOKUP_LPM
#endif
+/*
+ * When set to zero, simple forwaring path is eanbled.
+ * When set to one, optimized forwarding path is enabled.
+ * Note that LPM optimisation path uses SSE4.1 instructions.
+ */
+#if ((APP_LOOKUP_METHOD == APP_LOOKUP_LPM) && !defined(__SSE4_1__))
+#define ENABLE_MULTI_BUFFER_OPTIMIZE 0
+#else
#define ENABLE_MULTI_BUFFER_OPTIMIZE 1
+#endif
#if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH)
#include <rte_hash.h>
#define MEMPOOL_CACHE_SIZE 256
-#define MBUF_SIZE (2048 + sizeof(struct rte_mbuf) + RTE_PKTMBUF_HEADROOM)
+#define MBUF_DATA_SIZE (2048 + RTE_PKTMBUF_HEADROOM)
/*
* This expression is used to calculate the number of mbufs needed depending on user input, taking
nb_lcores*MEMPOOL_CACHE_SIZE), \
(unsigned)8192)
-/*
- * RX and TX Prefetch, Host, and Write-back threshold values should be
- * carefully set for optimal performance. Consult the network
- * controller's datasheet and supporting DPDK documentation for guidance
- * on how these parameters should be set.
- */
-#define RX_PTHRESH 8 /**< Default values of RX prefetch threshold reg. */
-#define RX_HTHRESH 8 /**< Default values of RX host threshold reg. */
-#define RX_WTHRESH 4 /**< Default values of RX write-back threshold reg. */
+#define MAX_PKT_BURST 32
+#define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */
/*
- * These default values are optimized for use with the Intel(R) 82599 10 GbE
- * Controller and the DPDK ixgbe PMD. Consider using other values for other
- * network controllers and/or network drivers.
+ * Try to avoid TX buffering if we have at least MAX_TX_BURST packets to send.
*/
-#define TX_PTHRESH 36 /**< Default values of TX prefetch threshold reg. */
-#define TX_HTHRESH 0 /**< Default values of TX host threshold reg. */
-#define TX_WTHRESH 0 /**< Default values of TX write-back threshold reg. */
-
-#define MAX_PKT_BURST 32
-#define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */
+#define MAX_TX_BURST (MAX_PKT_BURST / 2)
#define NB_SOCKETS 8
/* Configure how many packets ahead to prefetch, when reading packets */
#define PREFETCH_OFFSET 3
+/* Used to mark destination port as 'invalid'. */
+#define BAD_PORT ((uint16_t)-1)
+
+#define FWDSTEP 4
+
/*
* Configurable number of RX/TX ring descriptors
*/
/* ethernet addresses of ports */
static struct ether_addr ports_eth_addr[RTE_MAX_ETHPORTS];
+static __m128i val_eth[RTE_MAX_ETHPORTS];
+
+/* replace first 12B of the ethernet header. */
+#define MASK_ETH 0x3f
+
/* mask of enabled ports */
static uint32_t enabled_port_mask = 0;
static int promiscuous_on = 0; /**< Ports set in promiscuous mode off by default. */
.rx_adv_conf = {
.rss_conf = {
.rss_key = NULL,
- .rss_hf = ETH_RSS_IPV4 | ETH_RSS_IPV6,
+ .rss_hf = ETH_RSS_IP,
},
},
.txmode = {
},
};
-static const struct rte_eth_rxconf rx_conf = {
- .rx_thresh = {
- .pthresh = RX_PTHRESH,
- .hthresh = RX_HTHRESH,
- .wthresh = RX_WTHRESH,
- },
- .rx_free_thresh = 32,
-};
-
-static struct rte_eth_txconf tx_conf = {
- .tx_thresh = {
- .pthresh = TX_PTHRESH,
- .hthresh = TX_HTHRESH,
- .wthresh = TX_WTHRESH,
- },
- .tx_free_thresh = 0, /* Use PMD default values */
- .tx_rs_thresh = 0, /* Use PMD default values */
- .txq_flags = (ETH_TXQ_FLAGS_NOMULTSEGS |
- ETH_TXQ_FLAGS_NOVLANOFFL |
- ETH_TXQ_FLAGS_NOXSUMSCTP |
- ETH_TXQ_FLAGS_NOXSUMUDP |
- ETH_TXQ_FLAGS_NOXSUMTCP)
-
-};
-
static struct rte_mempool * pktmbuf_pool[NB_SOCKETS];
#if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH)
return 0;
}
+static inline __attribute__((always_inline)) void
+send_packetsx4(struct lcore_conf *qconf, uint8_t port,
+ struct rte_mbuf *m[], uint32_t num)
+{
+ uint32_t len, j, n;
+
+ len = qconf->tx_mbufs[port].len;
+
+ /*
+ * If TX buffer for that queue is empty, and we have enough packets,
+ * then send them straightway.
+ */
+ if (num >= MAX_TX_BURST && len == 0) {
+ n = rte_eth_tx_burst(port, qconf->tx_queue_id[port], m, num);
+ if (unlikely(n < num)) {
+ do {
+ rte_pktmbuf_free(m[n]);
+ } while (++n < num);
+ }
+ return;
+ }
+
+ /*
+ * Put packets into TX buffer for that queue.
+ */
+
+ n = len + num;
+ n = (n > MAX_PKT_BURST) ? MAX_PKT_BURST - len : num;
+
+ j = 0;
+ switch (n % FWDSTEP) {
+ while (j < n) {
+ case 0:
+ qconf->tx_mbufs[port].m_table[len + j] = m[j];
+ j++;
+ case 3:
+ qconf->tx_mbufs[port].m_table[len + j] = m[j];
+ j++;
+ case 2:
+ qconf->tx_mbufs[port].m_table[len + j] = m[j];
+ j++;
+ case 1:
+ qconf->tx_mbufs[port].m_table[len + j] = m[j];
+ j++;
+ }
+ }
+
+ len += n;
+
+ /* enough pkts to be sent */
+ if (unlikely(len == MAX_PKT_BURST)) {
+
+ send_burst(qconf, MAX_PKT_BURST, port);
+
+ /* copy rest of the packets into the TX buffer. */
+ len = num - n;
+ j = 0;
+ switch (len % FWDSTEP) {
+ while (j < len) {
+ case 0:
+ qconf->tx_mbufs[port].m_table[j] = m[n + j];
+ j++;
+ case 3:
+ qconf->tx_mbufs[port].m_table[j] = m[n + j];
+ j++;
+ case 2:
+ qconf->tx_mbufs[port].m_table[j] = m[n + j];
+ j++;
+ case 1:
+ qconf->tx_mbufs[port].m_table[j] = m[n + j];
+ j++;
+ }
+ }
+ }
+
+ qconf->tx_mbufs[port].len = len;
+}
+
#ifdef DO_RFC_1812_CHECKS
static inline int
is_valid_ipv4_pkt(struct ipv4_hdr *pkt, uint32_t link_len)
#endif
#if (APP_LOOKUP_METHOD == APP_LOOKUP_LPM)
+
static inline uint8_t
get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, lookup_struct_t * ipv4_l3fwd_lookup_struct)
{
uint8_t next_hop;
return (uint8_t) ((rte_lpm_lookup(ipv4_l3fwd_lookup_struct,
- rte_be_to_cpu_32(((struct ipv4_hdr*)ipv4_hdr)->dst_addr), &next_hop) == 0)?
- next_hop : portid);
+ rte_be_to_cpu_32(((struct ipv4_hdr *)ipv4_hdr)->dst_addr),
+ &next_hop) == 0) ? next_hop : portid);
}
static inline uint8_t
}
#endif
-#if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH) & (ENABLE_MULTI_BUFFER_OPTIMIZE == 1)
-static inline void l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid, struct lcore_conf *qconf);
+static inline void l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid,
+ struct lcore_conf *qconf) __attribute__((unused));
+
+#if ((APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH) && \
+ (ENABLE_MULTI_BUFFER_OPTIMIZE == 1))
#define MASK_ALL_PKTS 0xf
#define EXECLUDE_1ST_PKT 0xe
#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.pkt_len) < 0) {
+ if (is_valid_ipv4_pkt(ipv4_hdr[0], m[0]->pkt_len) < 0) {
rte_pktmbuf_free(m[0]);
valid_mask &= EXECLUDE_1ST_PKT;
}
- if (is_valid_ipv4_pkt(ipv4_hdr[1], m[1]->pkt.pkt_len) < 0) {
+ if (is_valid_ipv4_pkt(ipv4_hdr[1], m[1]->pkt_len) < 0) {
rte_pktmbuf_free(m[1]);
valid_mask &= EXECLUDE_2ND_PKT;
}
- if (is_valid_ipv4_pkt(ipv4_hdr[2], m[2]->pkt.pkt_len) < 0) {
+ if (is_valid_ipv4_pkt(ipv4_hdr[2], m[2]->pkt_len) < 0) {
rte_pktmbuf_free(m[2]);
valid_mask &= EXECLUDE_3RD_PKT;
}
- if (is_valid_ipv4_pkt(ipv4_hdr[3], m[3]->pkt.pkt_len) < 0) {
+ if (is_valid_ipv4_pkt(ipv4_hdr[3], m[3]->pkt_len) < 0) {
rte_pktmbuf_free(m[3]);
valid_mask &= EXECLUDE_4TH_PKT;
}
send_single_packet(m[3], (uint8_t)dst_port[3]);
}
-#endif // End of #if(APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH)&(ENABLE_MULTI_BUFFER_OPTIMIZE == 1)
+#endif /* APP_LOOKUP_METHOD */
static inline __attribute__((always_inline)) void
l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid, struct lcore_conf *qconf)
#ifdef DO_RFC_1812_CHECKS
/* Check to make sure the packet is valid (RFC1812) */
- if (is_valid_ipv4_pkt(ipv4_hdr, m->pkt.pkt_len) < 0) {
+ if (is_valid_ipv4_pkt(ipv4_hdr, m->pkt_len) < 0) {
rte_pktmbuf_free(m);
return;
}
#endif
- dst_port = get_ipv4_dst_port(ipv4_hdr, portid, qconf->ipv4_lookup_struct);
- if (dst_port >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port) == 0)
+ dst_port = get_ipv4_dst_port(ipv4_hdr, portid,
+ qconf->ipv4_lookup_struct);
+ if (dst_port >= RTE_MAX_ETHPORTS ||
+ (enabled_port_mask & 1 << dst_port) == 0)
dst_port = portid;
/* 02:00:00:00:00:xx */
d_addr_bytes = ð_hdr->d_addr.addr_bytes[0];
- *((uint64_t *)d_addr_bytes) = 0x000000000002 + ((uint64_t)dst_port << 40);
+ *((uint64_t *)d_addr_bytes) = ETHER_LOCAL_ADMIN_ADDR +
+ ((uint64_t)dst_port << 40);
#ifdef DO_RFC_1812_CHECKS
/* Update time to live and header checksum */
/* 02:00:00:00:00:xx */
d_addr_bytes = ð_hdr->d_addr.addr_bytes[0];
- *((uint64_t *)d_addr_bytes) = 0x000000000002 + ((uint64_t)dst_port << 40);
+ *((uint64_t *)d_addr_bytes) = ETHER_LOCAL_ADMIN_ADDR +
+ ((uint64_t)dst_port << 40);
/* src addr */
ether_addr_copy(&ports_eth_addr[dst_port], ð_hdr->s_addr);
}
+#ifdef DO_RFC_1812_CHECKS
+
+#define IPV4_MIN_VER_IHL 0x45
+#define IPV4_MAX_VER_IHL 0x4f
+#define IPV4_MAX_VER_IHL_DIFF (IPV4_MAX_VER_IHL - IPV4_MIN_VER_IHL)
+
+/* Minimum value of IPV4 total length (20B) in network byte order. */
+#define IPV4_MIN_LEN_BE (sizeof(struct ipv4_hdr) << 8)
+
+/*
+ * From http://www.rfc-editor.org/rfc/rfc1812.txt section 5.2.2:
+ * - The IP version number must be 4.
+ * - The IP header length field must be large enough to hold the
+ * minimum length legal IP datagram (20 bytes = 5 words).
+ * - The IP total length field must be large enough to hold the IP
+ * datagram header, whose length is specified in the IP header length
+ * field.
+ * If we encounter invalid IPV4 packet, then set destination port for it
+ * to BAD_PORT value.
+ */
+static inline __attribute__((always_inline)) void
+rfc1812_process(struct ipv4_hdr *ipv4_hdr, uint16_t *dp, uint32_t flags)
+{
+ uint8_t ihl;
+
+ if ((flags & PKT_RX_IPV4_HDR) != 0) {
+
+ ihl = ipv4_hdr->version_ihl - IPV4_MIN_VER_IHL;
+
+ ipv4_hdr->time_to_live--;
+ ipv4_hdr->hdr_checksum++;
+
+ if (ihl > IPV4_MAX_VER_IHL_DIFF ||
+ ((uint8_t)ipv4_hdr->total_length == 0 &&
+ ipv4_hdr->total_length < IPV4_MIN_LEN_BE)) {
+ dp[0] = BAD_PORT;
+ }
+ }
+}
+
+#else
+#define rfc1812_process(mb, dp) do { } while (0)
+#endif /* DO_RFC_1812_CHECKS */
+
+
+#if ((APP_LOOKUP_METHOD == APP_LOOKUP_LPM) && \
+ (ENABLE_MULTI_BUFFER_OPTIMIZE == 1))
+
+static inline __attribute__((always_inline)) uint16_t
+get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt,
+ uint32_t dst_ipv4, uint8_t portid)
+{
+ uint8_t next_hop;
+ struct ipv6_hdr *ipv6_hdr;
+ struct ether_hdr *eth_hdr;
+
+ if (pkt->ol_flags & PKT_RX_IPV4_HDR) {
+ if (rte_lpm_lookup(qconf->ipv4_lookup_struct, dst_ipv4,
+ &next_hop) != 0)
+ next_hop = portid;
+ } else if (pkt->ol_flags & PKT_RX_IPV6_HDR) {
+ eth_hdr = rte_pktmbuf_mtod(pkt, struct ether_hdr *);
+ ipv6_hdr = (struct ipv6_hdr *)(eth_hdr + 1);
+ if (rte_lpm6_lookup(qconf->ipv6_lookup_struct,
+ ipv6_hdr->dst_addr, &next_hop) != 0)
+ next_hop = portid;
+ } else {
+ next_hop = portid;
+ }
+
+ return next_hop;
+}
+
+static inline void
+process_packet(struct lcore_conf *qconf, struct rte_mbuf *pkt,
+ uint16_t *dst_port, uint8_t portid)
+{
+ struct ether_hdr *eth_hdr;
+ struct ipv4_hdr *ipv4_hdr;
+ uint32_t dst_ipv4;
+ uint16_t dp;
+ __m128i te, ve;
+
+ eth_hdr = rte_pktmbuf_mtod(pkt, struct ether_hdr *);
+ ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1);
+
+ dst_ipv4 = ipv4_hdr->dst_addr;
+ dst_ipv4 = rte_be_to_cpu_32(dst_ipv4);
+ dp = get_dst_port(qconf, pkt, dst_ipv4, portid);
+
+ te = _mm_load_si128((__m128i *)eth_hdr);
+ ve = val_eth[dp];
+
+ dst_port[0] = dp;
+ rfc1812_process(ipv4_hdr, dst_port, pkt->ol_flags);
+
+ te = _mm_blend_epi16(te, ve, MASK_ETH);
+ _mm_store_si128((__m128i *)eth_hdr, te);
+}
+
+/*
+ * Read ol_flags and destination IPV4 addresses from 4 mbufs.
+ */
+static inline void
+processx4_step1(struct rte_mbuf *pkt[FWDSTEP], __m128i *dip, uint32_t *flag)
+{
+ struct ipv4_hdr *ipv4_hdr;
+ struct ether_hdr *eth_hdr;
+ uint32_t x0, x1, x2, x3;
+
+ eth_hdr = rte_pktmbuf_mtod(pkt[0], struct ether_hdr *);
+ ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1);
+ x0 = ipv4_hdr->dst_addr;
+ flag[0] = pkt[0]->ol_flags & PKT_RX_IPV4_HDR;
+
+ eth_hdr = rte_pktmbuf_mtod(pkt[1], struct ether_hdr *);
+ ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1);
+ x1 = ipv4_hdr->dst_addr;
+ flag[0] &= pkt[1]->ol_flags;
+
+ eth_hdr = rte_pktmbuf_mtod(pkt[2], struct ether_hdr *);
+ ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1);
+ x2 = ipv4_hdr->dst_addr;
+ flag[0] &= pkt[2]->ol_flags;
+
+ eth_hdr = rte_pktmbuf_mtod(pkt[3], struct ether_hdr *);
+ ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1);
+ x3 = ipv4_hdr->dst_addr;
+ flag[0] &= pkt[3]->ol_flags;
+
+ dip[0] = _mm_set_epi32(x3, x2, x1, x0);
+}
+
+/*
+ * Lookup into LPM for destination port.
+ * If lookup fails, use incoming port (portid) as destination port.
+ */
+static inline void
+processx4_step2(const struct lcore_conf *qconf, __m128i dip, uint32_t flag,
+ uint8_t portid, struct rte_mbuf *pkt[FWDSTEP], uint16_t dprt[FWDSTEP])
+{
+ rte_xmm_t dst;
+ const __m128i bswap_mask = _mm_set_epi8(12, 13, 14, 15, 8, 9, 10, 11,
+ 4, 5, 6, 7, 0, 1, 2, 3);
+
+ /* Byte swap 4 IPV4 addresses. */
+ dip = _mm_shuffle_epi8(dip, bswap_mask);
+
+ /* if all 4 packets are IPV4. */
+ if (likely(flag != 0)) {
+ rte_lpm_lookupx4(qconf->ipv4_lookup_struct, dip, dprt, portid);
+ } else {
+ dst.x = dip;
+ dprt[0] = get_dst_port(qconf, pkt[0], dst.u32[0], portid);
+ dprt[1] = get_dst_port(qconf, pkt[1], dst.u32[1], portid);
+ dprt[2] = get_dst_port(qconf, pkt[2], dst.u32[2], portid);
+ dprt[3] = get_dst_port(qconf, pkt[3], dst.u32[3], portid);
+ }
+}
+
+/*
+ * Update source and destination MAC addresses in the ethernet header.
+ * Perform RFC1812 checks and updates for IPV4 packets.
+ */
+static inline void
+processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP])
+{
+ __m128i te[FWDSTEP];
+ __m128i ve[FWDSTEP];
+ __m128i *p[FWDSTEP];
+
+ p[0] = (rte_pktmbuf_mtod(pkt[0], __m128i *));
+ p[1] = (rte_pktmbuf_mtod(pkt[1], __m128i *));
+ p[2] = (rte_pktmbuf_mtod(pkt[2], __m128i *));
+ p[3] = (rte_pktmbuf_mtod(pkt[3], __m128i *));
+
+ ve[0] = val_eth[dst_port[0]];
+ te[0] = _mm_load_si128(p[0]);
+
+ ve[1] = val_eth[dst_port[1]];
+ te[1] = _mm_load_si128(p[1]);
+
+ ve[2] = val_eth[dst_port[2]];
+ te[2] = _mm_load_si128(p[2]);
+
+ ve[3] = val_eth[dst_port[3]];
+ te[3] = _mm_load_si128(p[3]);
+
+ /* Update first 12 bytes, keep rest bytes intact. */
+ te[0] = _mm_blend_epi16(te[0], ve[0], MASK_ETH);
+ te[1] = _mm_blend_epi16(te[1], ve[1], MASK_ETH);
+ te[2] = _mm_blend_epi16(te[2], ve[2], MASK_ETH);
+ te[3] = _mm_blend_epi16(te[3], ve[3], MASK_ETH);
+
+ _mm_store_si128(p[0], te[0]);
+ _mm_store_si128(p[1], te[1]);
+ _mm_store_si128(p[2], te[2]);
+ _mm_store_si128(p[3], te[3]);
+
+ rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[0] + 1),
+ &dst_port[0], pkt[0]->ol_flags);
+ rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[1] + 1),
+ &dst_port[1], pkt[1]->ol_flags);
+ rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[2] + 1),
+ &dst_port[2], pkt[2]->ol_flags);
+ rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[3] + 1),
+ &dst_port[3], pkt[3]->ol_flags);
+}
+
+/*
+ * We group consecutive packets with the same destionation port into one burst.
+ * To avoid extra latency this is done together with some other packet
+ * processing, but after we made a final decision about packet's destination.
+ * To do this we maintain:
+ * pnum - array of number of consecutive packets with the same dest port for
+ * each packet in the input burst.
+ * lp - pointer to the last updated element in the pnum.
+ * dlp - dest port value lp corresponds to.
+ */
+
+#define GRPSZ (1 << FWDSTEP)
+#define GRPMSK (GRPSZ - 1)
+
+#define GROUP_PORT_STEP(dlp, dcp, lp, pn, idx) do { \
+ if (likely((dlp) == (dcp)[(idx)])) { \
+ (lp)[0]++; \
+ } else { \
+ (dlp) = (dcp)[idx]; \
+ (lp) = (pn) + (idx); \
+ (lp)[0] = 1; \
+ } \
+} while (0)
+
+/*
+ * Group consecutive packets with the same destination port in bursts of 4.
+ * Suppose we have array of destionation ports:
+ * dst_port[] = {a, b, c, d,, e, ... }
+ * dp1 should contain: <a, b, c, d>, dp2: <b, c, d, e>.
+ * We doing 4 comparisions at once and the result is 4 bit mask.
+ * This mask is used as an index into prebuild array of pnum values.
+ */
+static inline uint16_t *
+port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, __m128i dp1, __m128i dp2)
+{
+ static const struct {
+ uint64_t pnum; /* prebuild 4 values for pnum[]. */
+ int32_t idx; /* index for new last updated elemnet. */
+ uint16_t lpv; /* add value to the last updated element. */
+ } gptbl[GRPSZ] = {
+ {
+ /* 0: a != b, b != c, c != d, d != e */
+ .pnum = UINT64_C(0x0001000100010001),
+ .idx = 4,
+ .lpv = 0,
+ },
+ {
+ /* 1: a == b, b != c, c != d, d != e */
+ .pnum = UINT64_C(0x0001000100010002),
+ .idx = 4,
+ .lpv = 1,
+ },
+ {
+ /* 2: a != b, b == c, c != d, d != e */
+ .pnum = UINT64_C(0x0001000100020001),
+ .idx = 4,
+ .lpv = 0,
+ },
+ {
+ /* 3: a == b, b == c, c != d, d != e */
+ .pnum = UINT64_C(0x0001000100020003),
+ .idx = 4,
+ .lpv = 2,
+ },
+ {
+ /* 4: a != b, b != c, c == d, d != e */
+ .pnum = UINT64_C(0x0001000200010001),
+ .idx = 4,
+ .lpv = 0,
+ },
+ {
+ /* 5: a == b, b != c, c == d, d != e */
+ .pnum = UINT64_C(0x0001000200010002),
+ .idx = 4,
+ .lpv = 1,
+ },
+ {
+ /* 6: a != b, b == c, c == d, d != e */
+ .pnum = UINT64_C(0x0001000200030001),
+ .idx = 4,
+ .lpv = 0,
+ },
+ {
+ /* 7: a == b, b == c, c == d, d != e */
+ .pnum = UINT64_C(0x0001000200030004),
+ .idx = 4,
+ .lpv = 3,
+ },
+ {
+ /* 8: a != b, b != c, c != d, d == e */
+ .pnum = UINT64_C(0x0002000100010001),
+ .idx = 3,
+ .lpv = 0,
+ },
+ {
+ /* 9: a == b, b != c, c != d, d == e */
+ .pnum = UINT64_C(0x0002000100010002),
+ .idx = 3,
+ .lpv = 1,
+ },
+ {
+ /* 0xa: a != b, b == c, c != d, d == e */
+ .pnum = UINT64_C(0x0002000100020001),
+ .idx = 3,
+ .lpv = 0,
+ },
+ {
+ /* 0xb: a == b, b == c, c != d, d == e */
+ .pnum = UINT64_C(0x0002000100020003),
+ .idx = 3,
+ .lpv = 2,
+ },
+ {
+ /* 0xc: a != b, b != c, c == d, d == e */
+ .pnum = UINT64_C(0x0002000300010001),
+ .idx = 2,
+ .lpv = 0,
+ },
+ {
+ /* 0xd: a == b, b != c, c == d, d == e */
+ .pnum = UINT64_C(0x0002000300010002),
+ .idx = 2,
+ .lpv = 1,
+ },
+ {
+ /* 0xe: a != b, b == c, c == d, d == e */
+ .pnum = UINT64_C(0x0002000300040001),
+ .idx = 1,
+ .lpv = 0,
+ },
+ {
+ /* 0xf: a == b, b == c, c == d, d == e */
+ .pnum = UINT64_C(0x0002000300040005),
+ .idx = 0,
+ .lpv = 4,
+ },
+ };
+
+ union {
+ uint16_t u16[FWDSTEP + 1];
+ uint64_t u64;
+ } *pnum = (void *)pn;
+
+ int32_t v;
+
+ dp1 = _mm_cmpeq_epi16(dp1, dp2);
+ dp1 = _mm_unpacklo_epi16(dp1, dp1);
+ v = _mm_movemask_ps((__m128)dp1);
+
+ /* update last port counter. */
+ lp[0] += gptbl[v].lpv;
+
+ /* if dest port value has changed. */
+ if (v != GRPMSK) {
+ lp = pnum->u16 + gptbl[v].idx;
+ lp[0] = 1;
+ pnum->u64 = gptbl[v].pnum;
+ }
+
+ return lp;
+}
+
+#endif /* APP_LOOKUP_METHOD */
+
/* main processing loop */
static int
main_loop(__attribute__((unused)) void *dummy)
int i, j, nb_rx;
uint8_t portid, queueid;
struct lcore_conf *qconf;
- const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US;
+ const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) /
+ US_PER_S * BURST_TX_DRAIN_US;
+
+#if ((APP_LOOKUP_METHOD == APP_LOOKUP_LPM) && \
+ (ENABLE_MULTI_BUFFER_OPTIMIZE == 1))
+ int32_t k;
+ uint16_t dlp;
+ uint16_t *lp;
+ uint16_t dst_port[MAX_PKT_BURST];
+ __m128i dip[MAX_PKT_BURST / FWDSTEP];
+ uint32_t flag[MAX_PKT_BURST / FWDSTEP];
+ uint16_t pnum[MAX_PKT_BURST + 1];
+#endif
prev_tsc = 0;
for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
if (qconf->tx_mbufs[portid].len == 0)
continue;
- send_burst(&lcore_conf[lcore_id],
+ send_burst(qconf,
qconf->tx_mbufs[portid].len,
portid);
qconf->tx_mbufs[portid].len = 0;
for (i = 0; i < qconf->n_rx_queue; ++i) {
portid = qconf->rx_queue_list[i].port_id;
queueid = qconf->rx_queue_list[i].queue_id;
- nb_rx = rte_eth_rx_burst(portid, queueid, pkts_burst, MAX_PKT_BURST);
-#if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH) & (ENABLE_MULTI_BUFFER_OPTIMIZE == 1)
+ nb_rx = rte_eth_rx_burst(portid, queueid, pkts_burst,
+ MAX_PKT_BURST);
+ if (nb_rx == 0)
+ continue;
+
+#if (ENABLE_MULTI_BUFFER_OPTIMIZE == 1)
+#if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH)
{
- /* Send nb_rx - nb_rx%4 packets in groups of 4.*/
+ /*
+ * Send nb_rx - nb_rx%4 packets
+ * in groups of 4.
+ */
int32_t n = RTE_ALIGN_FLOOR(nb_rx, 4);
for (j = 0; j < n ; j+=4) {
uint32_t ol_flag = pkts_burst[j]->ol_flags
portid, qconf);
}
}
-#else
+#elif (APP_LOOKUP_METHOD == APP_LOOKUP_LPM)
+
+ k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP);
+ for (j = 0; j != k; j += FWDSTEP) {
+ processx4_step1(&pkts_burst[j],
+ &dip[j / FWDSTEP],
+ &flag[j / FWDSTEP]);
+ }
+
+ k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP);
+ for (j = 0; j != k; j += FWDSTEP) {
+ processx4_step2(qconf, dip[j / FWDSTEP],
+ flag[j / FWDSTEP], portid,
+ &pkts_burst[j], &dst_port[j]);
+ }
+
+ /*
+ * Finish packet processing and group consecutive
+ * packets with the same destination port.
+ */
+ k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP);
+ if (k != 0) {
+ __m128i dp1, dp2;
+
+ lp = pnum;
+ lp[0] = 1;
+
+ processx4_step3(pkts_burst, dst_port);
+
+ /* dp1: <d[0], d[1], d[2], d[3], ... > */
+ dp1 = _mm_loadu_si128((__m128i *)dst_port);
+
+ for (j = FWDSTEP; j != k; j += FWDSTEP) {
+ processx4_step3(&pkts_burst[j],
+ &dst_port[j]);
+
+ /*
+ * dp2:
+ * <d[j-3], d[j-2], d[j-1], d[j], ... >
+ */
+ dp2 = _mm_loadu_si128((__m128i *)
+ &dst_port[j - FWDSTEP + 1]);
+ lp = port_groupx4(&pnum[j - FWDSTEP],
+ lp, dp1, dp2);
+
+ /*
+ * dp1:
+ * <d[j], d[j+1], d[j+2], d[j+3], ... >
+ */
+ dp1 = _mm_srli_si128(dp2,
+ (FWDSTEP - 1) *
+ sizeof(dst_port[0]));
+ }
+
+ /*
+ * dp2: <d[j-3], d[j-2], d[j-1], d[j-1], ... >
+ */
+ dp2 = _mm_shufflelo_epi16(dp1, 0xf9);
+ lp = port_groupx4(&pnum[j - FWDSTEP], lp,
+ dp1, dp2);
+
+ /*
+ * remove values added by the last repeated
+ * dst port.
+ */
+ lp[0]--;
+ dlp = dst_port[j - 1];
+ } else {
+ /* set dlp and lp to the never used values. */
+ dlp = BAD_PORT - 1;
+ lp = pnum + MAX_PKT_BURST;
+ }
+
+ /* Process up to last 3 packets one by one. */
+ switch (nb_rx % FWDSTEP) {
+ case 3:
+ process_packet(qconf, pkts_burst[j],
+ dst_port + j, portid);
+ GROUP_PORT_STEP(dlp, dst_port, lp, pnum, j);
+ j++;
+ case 2:
+ process_packet(qconf, pkts_burst[j],
+ dst_port + j, portid);
+ GROUP_PORT_STEP(dlp, dst_port, lp, pnum, j);
+ j++;
+ case 1:
+ process_packet(qconf, pkts_burst[j],
+ dst_port + j, portid);
+ GROUP_PORT_STEP(dlp, dst_port, lp, pnum, j);
+ j++;
+ }
+
+ /*
+ * Send packets out, through destination port.
+ * Consecuteve pacekts with the same destination port
+ * are already grouped together.
+ * If destination port for the packet equals BAD_PORT,
+ * then free the packet without sending it out.
+ */
+ for (j = 0; j < nb_rx; j += k) {
+
+ int32_t m;
+ uint16_t pn;
+
+ pn = dst_port[j];
+ k = pnum[j];
+
+ if (likely(pn != BAD_PORT)) {
+ send_packetsx4(qconf, pn,
+ pkts_burst + j, k);
+ } else {
+ for (m = j; m != j + k; m++)
+ rte_pktmbuf_free(pkts_burst[m]);
+ }
+ }
+
+#endif /* APP_LOOKUP_METHOD */
+#else /* ENABLE_MULTI_BUFFER_OPTIMIZE == 0 */
+
/* Prefetch first packets */
for (j = 0; j < PREFETCH_OFFSET && j < nb_rx; j++) {
rte_prefetch0(rte_pktmbuf_mtod(
for (j = 0; j < (nb_rx - PREFETCH_OFFSET); j++) {
rte_prefetch0(rte_pktmbuf_mtod(pkts_burst[
j + PREFETCH_OFFSET], void *));
- l3fwd_simple_forward(pkts_burst[j], portid, qconf);
+ l3fwd_simple_forward(pkts_burst[j], portid,
+ qconf);
}
/* Forward remaining prefetched packets */
for (; j < nb_rx; j++) {
- l3fwd_simple_forward(pkts_burst[j], portid, qconf);
+ l3fwd_simple_forward(pkts_burst[j], portid,
+ qconf);
}
-#endif // End of #if((ENABLE_MULTI_BUFFER_OPTIMIZE == 1)&(APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH))
+#endif /* ENABLE_MULTI_BUFFER_OPTIMIZE */
+
}
}
}
if(size >= sizeof(s))
return -1;
- rte_snprintf(s, sizeof(s), "%.*s", size, p);
+ snprintf(s, sizeof(s), "%.*s", size, p);
if (rte_strsplit(s, sizeof(s), str_fld, _NUM_FLD, ',') != _NUM_FLD)
return -1;
for (i = 0; i < _NUM_FLD; i++){
printf("jumbo frame is enabled - disabling simple TX path\n");
port_conf.rxmode.jumbo_frame = 1;
- tx_conf.txq_flags = 0;
/* if no max-pkt-len set, use the default value ETHER_MAX_LEN */
if (0 == getopt_long(argc, argvopt, "", &lenopts, &option_index)) {
static void
print_ethaddr(const char *name, const struct ether_addr *eth_addr)
{
- printf ("%s%02X:%02X:%02X:%02X:%02X:%02X", name,
- eth_addr->addr_bytes[0],
- eth_addr->addr_bytes[1],
- eth_addr->addr_bytes[2],
- eth_addr->addr_bytes[3],
- eth_addr->addr_bytes[4],
- eth_addr->addr_bytes[5]);
+ char buf[ETHER_ADDR_FMT_SIZE];
+ ether_format_addr(buf, ETHER_ADDR_FMT_SIZE, eth_addr);
+ printf("%s%s", name, buf);
}
#if (APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH)
convert_ipv4_5tuple(&entry.key, &newkey);
ret = rte_hash_add_key (h,(void *) &newkey);
if (ret < 0) {
- rte_exit(EXIT_FAILURE, "Unable to add entry %u to the"
- "l3fwd hash.\n", i);
+ rte_exit(EXIT_FAILURE, "Unable to add entry %" PRIu32
+ " to the l3fwd hash.\n", i);
}
ipv4_l3fwd_out_if[ret] = entry.if_out;
}
- printf("Hash: Adding 0x%x keys\n", array_len);
+ printf("Hash: Adding 0x%" PRIx32 " keys\n", array_len);
}
#define BIT_16_TO_23 0x00ff0000
convert_ipv6_5tuple(&entry.key, &newkey);
ret = rte_hash_add_key (h, (void *) &newkey);
if (ret < 0) {
- rte_exit(EXIT_FAILURE, "Unable to add entry %u to the"
- "l3fwd hash.\n", i);
+ rte_exit(EXIT_FAILURE, "Unable to add entry %" PRIu32
+ " to the l3fwd hash.\n", i);
}
ipv6_l3fwd_out_if[ret] = entry.if_out;
}
- printf("Hash: Adding 0x%xkeys\n", array_len);
+ printf("Hash: Adding 0x%" PRIx32 "keys\n", array_len);
}
#define NUMBER_PORT_USED 4
char s[64];
/* create ipv4 hash */
- rte_snprintf(s, sizeof(s), "ipv4_l3fwd_hash_%d", socketid);
+ snprintf(s, sizeof(s), "ipv4_l3fwd_hash_%d", socketid);
ipv4_l3fwd_hash_params.name = s;
ipv4_l3fwd_hash_params.socket_id = socketid;
ipv4_l3fwd_lookup_struct[socketid] = rte_hash_create(&ipv4_l3fwd_hash_params);
"socket %d\n", socketid);
/* create ipv6 hash */
- rte_snprintf(s, sizeof(s), "ipv6_l3fwd_hash_%d", socketid);
+ snprintf(s, sizeof(s), "ipv6_l3fwd_hash_%d", socketid);
ipv6_l3fwd_hash_params.name = s;
ipv6_l3fwd_hash_params.socket_id = socketid;
ipv6_l3fwd_lookup_struct[socketid] = rte_hash_create(&ipv6_l3fwd_hash_params);
char s[64];
/* create the LPM table */
- rte_snprintf(s, sizeof(s), "IPV4_L3FWD_LPM_%d", socketid);
+ snprintf(s, sizeof(s), "IPV4_L3FWD_LPM_%d", socketid);
ipv4_l3fwd_lookup_struct[socketid] = rte_lpm_create(s, socketid,
IPV4_L3FWD_LPM_MAX_RULES, 0);
if (ipv4_l3fwd_lookup_struct[socketid] == NULL)
/* populate the LPM table */
for (i = 0; i < IPV4_L3FWD_NUM_ROUTES; i++) {
+
+ /* skip unused ports */
+ if ((1 << ipv4_l3fwd_route_array[i].if_out &
+ enabled_port_mask) == 0)
+ continue;
+
ret = rte_lpm_add(ipv4_l3fwd_lookup_struct[socketid],
ipv4_l3fwd_route_array[i].ip,
ipv4_l3fwd_route_array[i].depth,
}
/* create the LPM6 table */
- rte_snprintf(s, sizeof(s), "IPV6_L3FWD_LPM_%d", socketid);
+ snprintf(s, sizeof(s), "IPV6_L3FWD_LPM_%d", socketid);
config.max_rules = IPV6_L3FWD_LPM_MAX_RULES;
config.number_tbl8s = IPV6_L3FWD_LPM_NUMBER_TBL8S;
/* populate the LPM table */
for (i = 0; i < IPV6_L3FWD_NUM_ROUTES; i++) {
+
+ /* skip unused ports */
+ if ((1 << ipv6_l3fwd_route_array[i].if_out &
+ enabled_port_mask) == 0)
+ continue;
+
ret = rte_lpm6_add(ipv6_l3fwd_lookup_struct[socketid],
ipv6_l3fwd_route_array[i].ip,
ipv6_l3fwd_route_array[i].depth,
socketid, lcore_id, NB_SOCKETS);
}
if (pktmbuf_pool[socketid] == NULL) {
- rte_snprintf(s, sizeof(s), "mbuf_pool_%d", socketid);
+ snprintf(s, sizeof(s), "mbuf_pool_%d", socketid);
pktmbuf_pool[socketid] =
- rte_mempool_create(s, nb_mbuf, MBUF_SIZE, MEMPOOL_CACHE_SIZE,
- sizeof(struct rte_pktmbuf_pool_private),
- rte_pktmbuf_pool_init, NULL,
- rte_pktmbuf_init, NULL,
- socketid, 0);
+ rte_pktmbuf_pool_create(s, nb_mbuf,
+ MEMPOOL_CACHE_SIZE, 0, MBUF_DATA_SIZE,
+ socketid);
if (pktmbuf_pool[socketid] == NULL)
rte_exit(EXIT_FAILURE,
"Cannot init mbuf pool on socket %d\n", socketid);
}
int
-MAIN(int argc, char **argv)
+main(int argc, char **argv)
{
struct lcore_conf *qconf;
+ struct rte_eth_dev_info dev_info;
+ struct rte_eth_txconf *txconf;
int ret;
unsigned nb_ports;
uint16_t queueid;
if (ret < 0)
rte_exit(EXIT_FAILURE, "init_lcore_rx_queues failed\n");
-
- if (rte_eal_pci_probe() < 0)
- rte_exit(EXIT_FAILURE, "Cannot probe PCI\n");
-
nb_ports = rte_eth_dev_count();
if (nb_ports > RTE_MAX_ETHPORTS)
nb_ports = RTE_MAX_ETHPORTS;
print_ethaddr(" Address:", &ports_eth_addr[portid]);
printf(", ");
+ /*
+ * prepare dst and src MACs for each port.
+ */
+ *(uint64_t *)(val_eth + portid) =
+ ETHER_LOCAL_ADMIN_ADDR + ((uint64_t)portid << 40);
+ ether_addr_copy(&ports_eth_addr[portid],
+ (struct ether_addr *)(val_eth + portid) + 1);
+
/* init memory */
ret = init_mem(NB_MBUF);
if (ret < 0)
printf("txq=%u,%d,%d ", lcore_id, queueid, socketid);
fflush(stdout);
+
+ rte_eth_dev_info_get(portid, &dev_info);
+ txconf = &dev_info.default_txconf;
+ if (port_conf.rxmode.jumbo_frame)
+ txconf->txq_flags = 0;
ret = rte_eth_tx_queue_setup(portid, queueid, nb_txd,
- socketid, &tx_conf);
+ socketid, txconf);
if (ret < 0)
rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup: err=%d, "
"port=%d\n", ret, portid);
fflush(stdout);
ret = rte_eth_rx_queue_setup(portid, queueid, nb_rxd,
- socketid, &rx_conf, pktmbuf_pool[socketid]);
+ socketid,
+ NULL,
+ pktmbuf_pool[socketid]);
if (ret < 0)
rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup: err=%d,"
"port=%d\n", ret, portid);
git.droids-corp.org / dpdk.git / blobdiff