diff options
Diffstat (limited to 'drivers/net/tap')
-rw-r--r-- | drivers/net/tap/Makefile | 2 | ||||
-rw-r--r-- | drivers/net/tap/rte_eth_tap.c | 1009 | ||||
-rw-r--r-- | drivers/net/tap/rte_eth_tap.h | 16 | ||||
-rw-r--r-- | drivers/net/tap/tap_bpf.h | 2 | ||||
-rw-r--r-- | drivers/net/tap/tap_bpf_api.c | 2 | ||||
-rw-r--r-- | drivers/net/tap/tap_bpf_insns.h | 2 | ||||
-rw-r--r-- | drivers/net/tap/tap_bpf_program.c | 4 | ||||
-rw-r--r-- | drivers/net/tap/tap_flow.c | 144 | ||||
-rw-r--r-- | drivers/net/tap/tap_flow.h | 2 | ||||
-rw-r--r-- | drivers/net/tap/tap_intr.c | 4 | ||||
-rw-r--r-- | drivers/net/tap/tap_log.h | 10 | ||||
-rw-r--r-- | drivers/net/tap/tap_netlink.c | 20 | ||||
-rw-r--r-- | drivers/net/tap/tap_netlink.h | 2 | ||||
-rw-r--r-- | drivers/net/tap/tap_rss.h | 8 | ||||
-rw-r--r-- | drivers/net/tap/tap_tcmsgs.c | 11 | ||||
-rw-r--r-- | drivers/net/tap/tap_tcmsgs.h | 2 |
16 files changed, 870 insertions, 370 deletions
diff --git a/drivers/net/tap/Makefile b/drivers/net/tap/Makefile index ccc5c5fc..32433653 100644 --- a/drivers/net/tap/Makefile +++ b/drivers/net/tap/Makefile @@ -24,7 +24,7 @@ CFLAGS += -I. CFLAGS += $(WERROR_FLAGS) LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring LDLIBS += -lrte_ethdev -lrte_net -lrte_kvargs -lrte_hash -LDLIBS += -lrte_bus_vdev +LDLIBS += -lrte_bus_vdev -lrte_gso CFLAGS += -DTAP_MAX_QUEUES=$(TAP_MAX_QUEUES) diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c index f09db0ea..feb92b48 100644 --- a/drivers/net/tap/rte_eth_tap.c +++ b/drivers/net/tap/rte_eth_tap.c @@ -15,7 +15,9 @@ #include <rte_net.h> #include <rte_debug.h> #include <rte_ip.h> +#include <rte_string_fns.h> +#include <assert.h> #include <sys/types.h> #include <sys/stat.h> #include <sys/socket.h> @@ -34,6 +36,7 @@ #include <linux/if_ether.h> #include <fcntl.h> +#include <tap_rss.h> #include <rte_eth_tap.h> #include <tap_flow.h> #include <tap_netlink.h> @@ -42,13 +45,25 @@ /* Linux based path to the TUN device */ #define TUN_TAP_DEV_PATH "/dev/net/tun" #define DEFAULT_TAP_NAME "dtap" +#define DEFAULT_TUN_NAME "dtun" #define ETH_TAP_IFACE_ARG "iface" #define ETH_TAP_REMOTE_ARG "remote" #define ETH_TAP_MAC_ARG "mac" #define ETH_TAP_MAC_FIXED "fixed" +#define ETH_TAP_USR_MAC_FMT "xx:xx:xx:xx:xx:xx" +#define ETH_TAP_CMP_MAC_FMT "0123456789ABCDEFabcdef" +#define ETH_TAP_MAC_ARG_FMT ETH_TAP_MAC_FIXED "|" ETH_TAP_USR_MAC_FMT + +#define TAP_GSO_MBUFS_PER_CORE 128 +#define TAP_GSO_MBUF_SEG_SIZE 128 +#define TAP_GSO_MBUF_CACHE_SIZE 4 +#define TAP_GSO_MBUFS_NUM \ + (TAP_GSO_MBUFS_PER_CORE * TAP_GSO_MBUF_CACHE_SIZE) + static struct rte_vdev_driver pmd_tap_drv; +static struct rte_vdev_driver pmd_tun_drv; static const char *valid_arguments[] = { ETH_TAP_IFACE_ARG, @@ -57,7 +72,10 @@ static const char *valid_arguments[] = { NULL }; -static int tap_unit; +static unsigned int tap_unit; +static unsigned int tun_unit; + +static char tuntap_name[8]; static volatile uint32_t tap_trigger; /* Rx trigger */ @@ -65,7 +83,7 @@ static struct rte_eth_link pmd_link = { .link_speed = ETH_SPEED_NUM_10G, .link_duplex = ETH_LINK_FULL_DUPLEX, .link_status = ETH_LINK_DOWN, - .link_autoneg = ETH_LINK_AUTONEG + .link_autoneg = ETH_LINK_FIXED, }; static void @@ -84,13 +102,20 @@ enum ioctl_mode { static int tap_intr_handle_set(struct rte_eth_dev *dev, int set); -/* Tun/Tap allocation routine +/** + * Tun/Tap allocation routine + * + * @param[in] pmd + * Pointer to private structure. + * + * @param[in] is_keepalive + * Keepalive flag * - * name is the number of the interface to use, unless NULL to take the host - * supplied name. + * @return + * -1 on failure, fd on success */ static int -tun_alloc(struct pmd_internals *pmd) +tun_alloc(struct pmd_internals *pmd, int is_keepalive) { struct ifreq ifr; #ifdef IFF_MULTI_QUEUE @@ -104,51 +129,64 @@ tun_alloc(struct pmd_internals *pmd) * Do not set IFF_NO_PI as packet information header will be needed * to check if a received packet has been truncated. */ - ifr.ifr_flags = IFF_TAP; + ifr.ifr_flags = (pmd->type == ETH_TUNTAP_TYPE_TAP) ? + IFF_TAP : IFF_TUN | IFF_POINTOPOINT; snprintf(ifr.ifr_name, IFNAMSIZ, "%s", pmd->name); - RTE_LOG(DEBUG, PMD, "ifr_name '%s'\n", ifr.ifr_name); + TAP_LOG(DEBUG, "ifr_name '%s'", ifr.ifr_name); fd = open(TUN_TAP_DEV_PATH, O_RDWR); if (fd < 0) { - RTE_LOG(ERR, PMD, "Unable to create TAP interface\n"); + TAP_LOG(ERR, "Unable to create %s interface", tuntap_name); goto error; } #ifdef IFF_MULTI_QUEUE /* Grab the TUN features to verify we can work multi-queue */ if (ioctl(fd, TUNGETFEATURES, &features) < 0) { - RTE_LOG(ERR, PMD, "TAP unable to get TUN/TAP features\n"); + TAP_LOG(ERR, "%s unable to get TUN/TAP features", + tuntap_name); goto error; } - RTE_LOG(DEBUG, PMD, " TAP Features %08x\n", features); + TAP_LOG(DEBUG, "%s Features %08x", tuntap_name, features); if (features & IFF_MULTI_QUEUE) { - RTE_LOG(DEBUG, PMD, " Multi-queue support for %d queues\n", + TAP_LOG(DEBUG, " Multi-queue support for %d queues", RTE_PMD_TAP_MAX_QUEUES); ifr.ifr_flags |= IFF_MULTI_QUEUE; } else #endif { ifr.ifr_flags |= IFF_ONE_QUEUE; - RTE_LOG(DEBUG, PMD, " Single queue only support\n"); + TAP_LOG(DEBUG, " Single queue only support"); } /* Set the TUN/TAP configuration and set the name if needed */ if (ioctl(fd, TUNSETIFF, (void *)&ifr) < 0) { - RTE_LOG(WARNING, PMD, - "Unable to set TUNSETIFF for %s\n", - ifr.ifr_name); - perror("TUNSETIFF"); + TAP_LOG(WARNING, "Unable to set TUNSETIFF for %s: %s", + ifr.ifr_name, strerror(errno)); goto error; } + if (is_keepalive) { + /* + * Detach the TUN/TAP keep-alive queue + * to avoid traffic through it + */ + ifr.ifr_flags = IFF_DETACH_QUEUE; + if (ioctl(fd, TUNSETQUEUE, (void *)&ifr) < 0) { + TAP_LOG(WARNING, + "Unable to detach keep-alive queue for %s: %s", + ifr.ifr_name, strerror(errno)); + goto error; + } + } + /* Always set the file descriptor to non-blocking */ if (fcntl(fd, F_SETFL, O_NONBLOCK) < 0) { - RTE_LOG(WARNING, PMD, - "Unable to set %s to nonblocking\n", - ifr.ifr_name); - perror("F_SETFL, NONBLOCK"); + TAP_LOG(WARNING, + "Unable to set %s to nonblocking: %s", + ifr.ifr_name, strerror(errno)); goto error; } @@ -182,10 +220,11 @@ tun_alloc(struct pmd_internals *pmd) fcntl(fd, F_SETFL, flags | O_ASYNC); fcntl(fd, F_SETOWN, getpid()); } while (0); + if (errno) { /* Disable trigger globally in case of error */ tap_trigger = 0; - RTE_LOG(WARNING, PMD, "Rx trigger disabled: %s\n", + TAP_LOG(WARNING, "Rx trigger disabled: %s", strerror(errno)); } @@ -255,14 +294,9 @@ static uint64_t tap_rx_offload_get_port_capa(void) { /* - * In order to support legacy apps, - * report capabilities also as port capabilities. + * No specific port Rx offload capabilities. */ - return DEV_RX_OFFLOAD_SCATTER | - DEV_RX_OFFLOAD_IPV4_CKSUM | - DEV_RX_OFFLOAD_UDP_CKSUM | - DEV_RX_OFFLOAD_TCP_CKSUM | - DEV_RX_OFFLOAD_CRC_STRIP; + return 0; } static uint64_t @@ -275,21 +309,6 @@ tap_rx_offload_get_queue_capa(void) DEV_RX_OFFLOAD_CRC_STRIP; } -static bool -tap_rxq_are_offloads_valid(struct rte_eth_dev *dev, uint64_t offloads) -{ - uint64_t port_offloads = dev->data->dev_conf.rxmode.offloads; - uint64_t queue_supp_offloads = tap_rx_offload_get_queue_capa(); - uint64_t port_supp_offloads = tap_rx_offload_get_port_capa(); - - if ((offloads & (queue_supp_offloads | port_supp_offloads)) != - offloads) - return false; - if ((port_offloads ^ offloads) & port_supp_offloads) - return false; - return true; -} - /* Callback to handle the rx burst of packets to the correct interface and * file descriptor(s) in a multi-queue setup. */ @@ -389,13 +408,9 @@ static uint64_t tap_tx_offload_get_port_capa(void) { /* - * In order to support legacy apps, - * report capabilities also as port capabilities. + * No specific port Tx offload capabilities. */ - return DEV_TX_OFFLOAD_MULTI_SEGS | - DEV_TX_OFFLOAD_IPV4_CKSUM | - DEV_TX_OFFLOAD_UDP_CKSUM | - DEV_TX_OFFLOAD_TCP_CKSUM; + return 0; } static uint64_t @@ -404,28 +419,45 @@ tap_tx_offload_get_queue_capa(void) return DEV_TX_OFFLOAD_MULTI_SEGS | DEV_TX_OFFLOAD_IPV4_CKSUM | DEV_TX_OFFLOAD_UDP_CKSUM | - DEV_TX_OFFLOAD_TCP_CKSUM; + DEV_TX_OFFLOAD_TCP_CKSUM | + DEV_TX_OFFLOAD_TCP_TSO; } -static bool -tap_txq_are_offloads_valid(struct rte_eth_dev *dev, uint64_t offloads) +/* Finalize l4 checksum calculation */ +static void +tap_tx_l4_cksum(uint16_t *l4_cksum, uint16_t l4_phdr_cksum, + uint32_t l4_raw_cksum) { - uint64_t port_offloads = dev->data->dev_conf.txmode.offloads; - uint64_t queue_supp_offloads = tap_tx_offload_get_queue_capa(); - uint64_t port_supp_offloads = tap_tx_offload_get_port_capa(); - - if ((offloads & (queue_supp_offloads | port_supp_offloads)) != - offloads) - return false; - /* Verify we have no conflict with port offloads */ - if ((port_offloads ^ offloads) & port_supp_offloads) - return false; - return true; + if (l4_cksum) { + uint32_t cksum; + + cksum = __rte_raw_cksum_reduce(l4_raw_cksum); + cksum += l4_phdr_cksum; + + cksum = ((cksum & 0xffff0000) >> 16) + (cksum & 0xffff); + cksum = (~cksum) & 0xffff; + if (cksum == 0) + cksum = 0xffff; + *l4_cksum = cksum; + } } +/* Accumaulate L4 raw checksums */ static void -tap_tx_offload(char *packet, uint64_t ol_flags, unsigned int l2_len, - unsigned int l3_len) +tap_tx_l4_add_rcksum(char *l4_data, unsigned int l4_len, uint16_t *l4_cksum, + uint32_t *l4_raw_cksum) +{ + if (l4_cksum == NULL) + return; + + *l4_raw_cksum = __rte_raw_cksum(l4_data, l4_len, *l4_raw_cksum); +} + +/* L3 and L4 pseudo headers checksum offloads */ +static void +tap_tx_l3_cksum(char *packet, uint64_t ol_flags, unsigned int l2_len, + unsigned int l3_len, unsigned int l4_len, uint16_t **l4_cksum, + uint16_t *l4_phdr_cksum, uint32_t *l4_raw_cksum) { void *l3_hdr = packet + l2_len; @@ -438,38 +470,137 @@ tap_tx_offload(char *packet, uint64_t ol_flags, unsigned int l2_len, iph->hdr_checksum = (cksum == 0xffff) ? cksum : ~cksum; } if (ol_flags & PKT_TX_L4_MASK) { - uint16_t l4_len; - uint32_t cksum; - uint16_t *l4_cksum; void *l4_hdr; l4_hdr = packet + l2_len + l3_len; if ((ol_flags & PKT_TX_L4_MASK) == PKT_TX_UDP_CKSUM) - l4_cksum = &((struct udp_hdr *)l4_hdr)->dgram_cksum; + *l4_cksum = &((struct udp_hdr *)l4_hdr)->dgram_cksum; else if ((ol_flags & PKT_TX_L4_MASK) == PKT_TX_TCP_CKSUM) - l4_cksum = &((struct tcp_hdr *)l4_hdr)->cksum; + *l4_cksum = &((struct tcp_hdr *)l4_hdr)->cksum; else return; - *l4_cksum = 0; - if (ol_flags & PKT_TX_IPV4) { - struct ipv4_hdr *iph = l3_hdr; + **l4_cksum = 0; + if (ol_flags & PKT_TX_IPV4) + *l4_phdr_cksum = rte_ipv4_phdr_cksum(l3_hdr, 0); + else + *l4_phdr_cksum = rte_ipv6_phdr_cksum(l3_hdr, 0); + *l4_raw_cksum = __rte_raw_cksum(l4_hdr, l4_len, 0); + } +} - l4_len = rte_be_to_cpu_16(iph->total_length) - l3_len; - cksum = rte_ipv4_phdr_cksum(l3_hdr, 0); - } else { - struct ipv6_hdr *ip6h = l3_hdr; +static inline void +tap_write_mbufs(struct tx_queue *txq, uint16_t num_mbufs, + struct rte_mbuf **pmbufs, + uint16_t *num_packets, unsigned long *num_tx_bytes) +{ + int i; + uint16_t l234_hlen; - /* payload_len does not include ext headers */ - l4_len = rte_be_to_cpu_16(ip6h->payload_len) - - l3_len + sizeof(struct ipv6_hdr); - cksum = rte_ipv6_phdr_cksum(l3_hdr, 0); + for (i = 0; i < num_mbufs; i++) { + struct rte_mbuf *mbuf = pmbufs[i]; + struct iovec iovecs[mbuf->nb_segs + 2]; + struct tun_pi pi = { .flags = 0, .proto = 0x00 }; + struct rte_mbuf *seg = mbuf; + char m_copy[mbuf->data_len]; + int proto; + int n; + int j; + int k; /* current index in iovecs for copying segments */ + uint16_t seg_len; /* length of first segment */ + uint16_t nb_segs; + uint16_t *l4_cksum; /* l4 checksum (pseudo header + payload) */ + uint32_t l4_raw_cksum = 0; /* TCP/UDP payload raw checksum */ + uint16_t l4_phdr_cksum = 0; /* TCP/UDP pseudo header checksum */ + uint16_t is_cksum = 0; /* in case cksum should be offloaded */ + + l4_cksum = NULL; + if (txq->type == ETH_TUNTAP_TYPE_TUN) { + /* + * TUN and TAP are created with IFF_NO_PI disabled. + * For TUN PMD this mandatory as fields are used by + * Kernel tun.c to determine whether its IP or non IP + * packets. + * + * The logic fetches the first byte of data from mbuf + * then compares whether its v4 or v6. If first byte + * is 4 or 6, then protocol field is updated. + */ + char *buff_data = rte_pktmbuf_mtod(seg, void *); + proto = (*buff_data & 0xf0); + pi.proto = (proto == 0x40) ? + rte_cpu_to_be_16(ETHER_TYPE_IPv4) : + ((proto == 0x60) ? + rte_cpu_to_be_16(ETHER_TYPE_IPv6) : + 0x00); } - cksum += rte_raw_cksum(l4_hdr, l4_len); - cksum = ((cksum & 0xffff0000) >> 16) + (cksum & 0xffff); - cksum = (~cksum) & 0xffff; - if (cksum == 0) - cksum = 0xffff; - *l4_cksum = cksum; + + k = 0; + iovecs[k].iov_base = π + iovecs[k].iov_len = sizeof(pi); + k++; + + nb_segs = mbuf->nb_segs; + if (txq->csum && + ((mbuf->ol_flags & (PKT_TX_IP_CKSUM | PKT_TX_IPV4) || + (mbuf->ol_flags & PKT_TX_L4_MASK) == PKT_TX_UDP_CKSUM || + (mbuf->ol_flags & PKT_TX_L4_MASK) == PKT_TX_TCP_CKSUM))) { + is_cksum = 1; + + /* Support only packets with at least layer 4 + * header included in the first segment + */ + seg_len = rte_pktmbuf_data_len(mbuf); + l234_hlen = mbuf->l2_len + mbuf->l3_len + mbuf->l4_len; + if (seg_len < l234_hlen) + break; + + /* To change checksums, work on a * copy of l2, l3 + * headers + l4 pseudo header + */ + rte_memcpy(m_copy, rte_pktmbuf_mtod(mbuf, void *), + l234_hlen); + tap_tx_l3_cksum(m_copy, mbuf->ol_flags, + mbuf->l2_len, mbuf->l3_len, mbuf->l4_len, + &l4_cksum, &l4_phdr_cksum, + &l4_raw_cksum); + iovecs[k].iov_base = m_copy; + iovecs[k].iov_len = l234_hlen; + k++; + + /* Update next iovecs[] beyond l2, l3, l4 headers */ + if (seg_len > l234_hlen) { + iovecs[k].iov_len = seg_len - l234_hlen; + iovecs[k].iov_base = + rte_pktmbuf_mtod(seg, char *) + + l234_hlen; + tap_tx_l4_add_rcksum(iovecs[k].iov_base, + iovecs[k].iov_len, l4_cksum, + &l4_raw_cksum); + k++; + nb_segs++; + } + seg = seg->next; + } + + for (j = k; j <= nb_segs; j++) { + iovecs[j].iov_len = rte_pktmbuf_data_len(seg); + iovecs[j].iov_base = rte_pktmbuf_mtod(seg, void *); + if (is_cksum) + tap_tx_l4_add_rcksum(iovecs[j].iov_base, + iovecs[j].iov_len, l4_cksum, + &l4_raw_cksum); + seg = seg->next; + } + + if (is_cksum) + tap_tx_l4_cksum(l4_cksum, l4_phdr_cksum, l4_raw_cksum); + + /* copy the tx frame data */ + n = writev(txq->fd, iovecs, j); + if (n <= 0) + break; + (*num_packets)++; + (*num_tx_bytes) += rte_pktmbuf_pkt_len(mbuf); } } @@ -480,6 +611,7 @@ pmd_tx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) { struct tx_queue *txq = queue; uint16_t num_tx = 0; + uint16_t num_packets = 0; unsigned long num_tx_bytes = 0; uint32_t max_size; int i; @@ -487,53 +619,70 @@ pmd_tx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) if (unlikely(nb_pkts == 0)) return 0; + struct rte_mbuf *gso_mbufs[MAX_GSO_MBUFS]; max_size = *txq->mtu + (ETHER_HDR_LEN + ETHER_CRC_LEN + 4); for (i = 0; i < nb_pkts; i++) { - struct rte_mbuf *mbuf = bufs[num_tx]; - struct iovec iovecs[mbuf->nb_segs + 1]; - struct tun_pi pi = { .flags = 0 }; - struct rte_mbuf *seg = mbuf; - char m_copy[mbuf->data_len]; - int n; + struct rte_mbuf *mbuf_in = bufs[num_tx]; + struct rte_mbuf **mbuf; + uint16_t num_mbufs = 0; + uint16_t tso_segsz = 0; + int ret; + uint16_t hdrs_len; int j; + uint64_t tso; - /* stats.errs will be incremented */ - if (rte_pktmbuf_pkt_len(mbuf) > max_size) - break; + tso = mbuf_in->ol_flags & PKT_TX_TCP_SEG; + if (tso) { + struct rte_gso_ctx *gso_ctx = &txq->gso_ctx; - iovecs[0].iov_base = π - iovecs[0].iov_len = sizeof(pi); - for (j = 1; j <= mbuf->nb_segs; j++) { - iovecs[j].iov_len = rte_pktmbuf_data_len(seg); - iovecs[j].iov_base = - rte_pktmbuf_mtod(seg, void *); - seg = seg->next; - } - if (txq->csum && - ((mbuf->ol_flags & (PKT_TX_IP_CKSUM | PKT_TX_IPV4) || - (mbuf->ol_flags & PKT_TX_L4_MASK) == PKT_TX_UDP_CKSUM || - (mbuf->ol_flags & PKT_TX_L4_MASK) == PKT_TX_TCP_CKSUM))) { - /* Support only packets with all data in the same seg */ - if (mbuf->nb_segs > 1) + assert(gso_ctx != NULL); + + /* TCP segmentation implies TCP checksum offload */ + mbuf_in->ol_flags |= PKT_TX_TCP_CKSUM; + + /* gso size is calculated without ETHER_CRC_LEN */ + hdrs_len = mbuf_in->l2_len + mbuf_in->l3_len + + mbuf_in->l4_len; + tso_segsz = mbuf_in->tso_segsz + hdrs_len; + if (unlikely(tso_segsz == hdrs_len) || + tso_segsz > *txq->mtu) { + txq->stats.errs++; break; - /* To change checksums, work on a copy of data. */ - rte_memcpy(m_copy, rte_pktmbuf_mtod(mbuf, void *), - rte_pktmbuf_data_len(mbuf)); - tap_tx_offload(m_copy, mbuf->ol_flags, - mbuf->l2_len, mbuf->l3_len); - iovecs[1].iov_base = m_copy; + } + gso_ctx->gso_size = tso_segsz; + ret = rte_gso_segment(mbuf_in, /* packet to segment */ + gso_ctx, /* gso control block */ + (struct rte_mbuf **)&gso_mbufs, /* out mbufs */ + RTE_DIM(gso_mbufs)); /* max tso mbufs */ + + /* ret contains the number of new created mbufs */ + if (ret < 0) + break; + + mbuf = gso_mbufs; + num_mbufs = ret; + } else { + /* stats.errs will be incremented */ + if (rte_pktmbuf_pkt_len(mbuf_in) > max_size) + break; + + /* ret 0 indicates no new mbufs were created */ + ret = 0; + mbuf = &mbuf_in; + num_mbufs = 1; } - /* copy the tx frame data */ - n = writev(txq->fd, iovecs, mbuf->nb_segs + 1); - if (n <= 0) - break; + tap_write_mbufs(txq, num_mbufs, mbuf, + &num_packets, &num_tx_bytes); num_tx++; - num_tx_bytes += mbuf->pkt_len; - rte_pktmbuf_free(mbuf); + /* free original mbuf */ + rte_pktmbuf_free(mbuf_in); + /* free tso mbufs */ + for (j = 0; j < ret; j++) + rte_pktmbuf_free(mbuf[j]); } - txq->stats.opackets += num_tx; + txq->stats.opackets += num_packets; txq->stats.errs += nb_pkts - num_tx; txq->stats.obytes += num_tx_bytes; @@ -593,7 +742,9 @@ apply: case SIOCSIFMTU: break; default: - RTE_ASSERT(!"unsupported request type: must not happen"); + RTE_LOG(WARNING, PMD, "%s: ioctl() called with wrong arg\n", + pmd->name); + return -EINVAL; } if (ioctl(pmd->ioctl_sock, request, ifr) < 0) goto error; @@ -602,8 +753,8 @@ apply: return 0; error: - RTE_LOG(DEBUG, PMD, "%s: %s(%s) failed: %s(%d)\n", ifr->ifr_name, - __func__, tap_ioctl_req2str(request), strerror(errno), errno); + TAP_LOG(DEBUG, "%s(%s) failed: %s(%d)", ifr->ifr_name, + tap_ioctl_req2str(request), strerror(errno), errno); return -errno; } @@ -630,12 +781,22 @@ tap_link_set_up(struct rte_eth_dev *dev) static int tap_dev_start(struct rte_eth_dev *dev) { - int err; + int err, i; err = tap_intr_handle_set(dev, 1); if (err) return err; - return tap_link_set_up(dev); + + err = tap_link_set_up(dev); + if (err) + return err; + + for (i = 0; i < dev->data->nb_tx_queues; i++) + dev->data->tx_queue_state[i] = RTE_ETH_QUEUE_STATE_STARTED; + for (i = 0; i < dev->data->nb_rx_queues; i++) + dev->data->rx_queue_state[i] = RTE_ETH_QUEUE_STATE_STARTED; + + return err; } /* This function gets called when the current port gets stopped. @@ -643,6 +804,13 @@ tap_dev_start(struct rte_eth_dev *dev) static void tap_dev_stop(struct rte_eth_dev *dev) { + int i; + + for (i = 0; i < dev->data->nb_tx_queues; i++) + dev->data->tx_queue_state[i] = RTE_ETH_QUEUE_STATE_STOPPED; + for (i = 0; i < dev->data->nb_rx_queues; i++) + dev->data->rx_queue_state[i] = RTE_ETH_QUEUE_STATE_STOPPED; + tap_intr_handle_set(dev, 0); tap_link_set_down(dev); } @@ -650,39 +818,28 @@ tap_dev_stop(struct rte_eth_dev *dev) static int tap_dev_configure(struct rte_eth_dev *dev) { - uint64_t supp_tx_offloads = tap_tx_offload_get_port_capa(); - uint64_t tx_offloads = dev->data->dev_conf.txmode.offloads; - - if ((tx_offloads & supp_tx_offloads) != tx_offloads) { - rte_errno = ENOTSUP; - RTE_LOG(ERR, PMD, - "Some Tx offloads are not supported " - "requested 0x%" PRIx64 " supported 0x%" PRIx64 "\n", - tx_offloads, supp_tx_offloads); - return -rte_errno; - } if (dev->data->nb_rx_queues > RTE_PMD_TAP_MAX_QUEUES) { - RTE_LOG(ERR, PMD, - "%s: number of rx queues %d exceeds max num of queues %d\n", + TAP_LOG(ERR, + "%s: number of rx queues %d exceeds max num of queues %d", dev->device->name, dev->data->nb_rx_queues, RTE_PMD_TAP_MAX_QUEUES); return -1; } if (dev->data->nb_tx_queues > RTE_PMD_TAP_MAX_QUEUES) { - RTE_LOG(ERR, PMD, - "%s: number of tx queues %d exceeds max num of queues %d\n", + TAP_LOG(ERR, + "%s: number of tx queues %d exceeds max num of queues %d", dev->device->name, dev->data->nb_tx_queues, RTE_PMD_TAP_MAX_QUEUES); return -1; } - RTE_LOG(INFO, PMD, "%s: %p: TX configured queues number: %u\n", - dev->device->name, (void *)dev, dev->data->nb_tx_queues); + TAP_LOG(INFO, "%s: %p: TX configured queues number: %u", + dev->device->name, (void *)dev, dev->data->nb_tx_queues); - RTE_LOG(INFO, PMD, "%s: %p: RX configured queues number: %u\n", - dev->device->name, (void *)dev, dev->data->nb_rx_queues); + TAP_LOG(INFO, "%s: %p: RX configured queues number: %u", + dev->device->name, (void *)dev, dev->data->nb_rx_queues); return 0; } @@ -732,7 +889,6 @@ tap_dev_info(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) dev_info->max_rx_queues = RTE_PMD_TAP_MAX_QUEUES; dev_info->max_tx_queues = RTE_PMD_TAP_MAX_QUEUES; dev_info->min_rx_bufsize = 0; - dev_info->pci_dev = NULL; dev_info->speed_capa = tap_dev_speed_capa(); dev_info->rx_queue_offload_capa = tap_rx_offload_get_queue_capa(); dev_info->rx_offload_capa = tap_rx_offload_get_port_capa() | @@ -740,6 +896,12 @@ tap_dev_info(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info) dev_info->tx_queue_offload_capa = tap_tx_offload_get_queue_capa(); dev_info->tx_offload_capa = tap_tx_offload_get_port_capa() | dev_info->tx_queue_offload_capa; + dev_info->hash_key_size = TAP_RSS_HASH_KEY_SIZE; + /* + * limitation: TAP supports all of IP, UDP and TCP hash + * functions together and not in partial combinations + */ + dev_info->flow_type_rss_offloads = ~TAP_RSS_HF_MASK; } static int @@ -830,6 +992,15 @@ tap_dev_close(struct rte_eth_dev *dev) ioctl(internals->ioctl_sock, SIOCSIFFLAGS, &internals->remote_initial_flags); } + + if (internals->ka_fd != -1) { + close(internals->ka_fd); + internals->ka_fd = -1; + } + /* + * Since TUN device has no more opened file descriptors + * it will be removed from kernel + */ } static void @@ -929,48 +1100,103 @@ tap_allmulti_disable(struct rte_eth_dev *dev) tap_flow_implicit_destroy(pmd, TAP_REMOTE_ALLMULTI); } -static void +static int tap_mac_set(struct rte_eth_dev *dev, struct ether_addr *mac_addr) { struct pmd_internals *pmd = dev->data->dev_private; enum ioctl_mode mode = LOCAL_ONLY; struct ifreq ifr; + int ret; + + if (pmd->type == ETH_TUNTAP_TYPE_TUN) { + TAP_LOG(ERR, "%s: can't MAC address for TUN", + dev->device->name); + return -ENOTSUP; + } if (is_zero_ether_addr(mac_addr)) { - RTE_LOG(ERR, PMD, "%s: can't set an empty MAC address\n", + TAP_LOG(ERR, "%s: can't set an empty MAC address", dev->device->name); - return; + return -EINVAL; } /* Check the actual current MAC address on the tap netdevice */ - if (tap_ioctl(pmd, SIOCGIFHWADDR, &ifr, 0, LOCAL_ONLY) < 0) - return; + ret = tap_ioctl(pmd, SIOCGIFHWADDR, &ifr, 0, LOCAL_ONLY); + if (ret < 0) + return ret; if (is_same_ether_addr((struct ether_addr *)&ifr.ifr_hwaddr.sa_data, mac_addr)) - return; + return 0; /* Check the current MAC address on the remote */ - if (tap_ioctl(pmd, SIOCGIFHWADDR, &ifr, 0, REMOTE_ONLY) < 0) - return; + ret = tap_ioctl(pmd, SIOCGIFHWADDR, &ifr, 0, REMOTE_ONLY); + if (ret < 0) + return ret; if (!is_same_ether_addr((struct ether_addr *)&ifr.ifr_hwaddr.sa_data, mac_addr)) mode = LOCAL_AND_REMOTE; ifr.ifr_hwaddr.sa_family = AF_LOCAL; rte_memcpy(ifr.ifr_hwaddr.sa_data, mac_addr, ETHER_ADDR_LEN); - if (tap_ioctl(pmd, SIOCSIFHWADDR, &ifr, 1, mode) < 0) - return; + ret = tap_ioctl(pmd, SIOCSIFHWADDR, &ifr, 1, mode); + if (ret < 0) + return ret; rte_memcpy(&pmd->eth_addr, mac_addr, ETHER_ADDR_LEN); if (pmd->remote_if_index && !pmd->flow_isolate) { /* Replace MAC redirection rule after a MAC change */ - if (tap_flow_implicit_destroy(pmd, TAP_REMOTE_LOCAL_MAC) < 0) { - RTE_LOG(ERR, PMD, - "%s: Couldn't delete MAC redirection rule\n", + ret = tap_flow_implicit_destroy(pmd, TAP_REMOTE_LOCAL_MAC); + if (ret < 0) { + TAP_LOG(ERR, + "%s: Couldn't delete MAC redirection rule", dev->device->name); - return; + return ret; } - if (tap_flow_implicit_create(pmd, TAP_REMOTE_LOCAL_MAC) < 0) - RTE_LOG(ERR, PMD, - "%s: Couldn't add MAC redirection rule\n", + ret = tap_flow_implicit_create(pmd, TAP_REMOTE_LOCAL_MAC); + if (ret < 0) { + TAP_LOG(ERR, + "%s: Couldn't add MAC redirection rule", dev->device->name); + return ret; + } } + + return 0; +} + +static int +tap_gso_ctx_setup(struct rte_gso_ctx *gso_ctx, struct rte_eth_dev *dev) +{ + uint32_t gso_types; + char pool_name[64]; + + /* + * Create private mbuf pool with TAP_GSO_MBUF_SEG_SIZE bytes + * size per mbuf use this pool for both direct and indirect mbufs + */ + + struct rte_mempool *mp; /* Mempool for GSO packets */ + + /* initialize GSO context */ + gso_types = DEV_TX_OFFLOAD_TCP_TSO; + snprintf(pool_name, sizeof(pool_name), "mp_%s", dev->device->name); + mp = rte_mempool_lookup((const char *)pool_name); + if (!mp) { + mp = rte_pktmbuf_pool_create(pool_name, TAP_GSO_MBUFS_NUM, + TAP_GSO_MBUF_CACHE_SIZE, 0, + RTE_PKTMBUF_HEADROOM + TAP_GSO_MBUF_SEG_SIZE, + SOCKET_ID_ANY); + if (!mp) { + struct pmd_internals *pmd = dev->data->dev_private; + RTE_LOG(DEBUG, PMD, "%s: failed to create mbuf pool for device %s\n", + pmd->name, dev->device->name); + return -1; + } + } + + gso_ctx->direct_pool = mp; + gso_ctx->indirect_pool = mp; + gso_ctx->gso_types = gso_types; + gso_ctx->gso_size = 0; /* gso_size is set in tx_burst() per packet */ + gso_ctx->flag = 0; + + return 0; } static int @@ -979,52 +1205,62 @@ tap_setup_queue(struct rte_eth_dev *dev, uint16_t qid, int is_rx) { + int ret; int *fd; int *other_fd; const char *dir; struct pmd_internals *pmd = dev->data->dev_private; struct rx_queue *rx = &internals->rxq[qid]; struct tx_queue *tx = &internals->txq[qid]; + struct rte_gso_ctx *gso_ctx; if (is_rx) { fd = &rx->fd; other_fd = &tx->fd; dir = "rx"; + gso_ctx = NULL; } else { fd = &tx->fd; other_fd = &rx->fd; dir = "tx"; + gso_ctx = &tx->gso_ctx; } if (*fd != -1) { /* fd for this queue already exists */ - RTE_LOG(DEBUG, PMD, "%s: fd %d for %s queue qid %d exists\n", + TAP_LOG(DEBUG, "%s: fd %d for %s queue qid %d exists", pmd->name, *fd, dir, qid); + gso_ctx = NULL; } else if (*other_fd != -1) { /* Only other_fd exists. dup it */ *fd = dup(*other_fd); if (*fd < 0) { *fd = -1; - RTE_LOG(ERR, PMD, "%s: dup() failed.\n", - pmd->name); + TAP_LOG(ERR, "%s: dup() failed.", pmd->name); return -1; } - RTE_LOG(DEBUG, PMD, "%s: dup fd %d for %s queue qid %d (%d)\n", + TAP_LOG(DEBUG, "%s: dup fd %d for %s queue qid %d (%d)", pmd->name, *other_fd, dir, qid, *fd); } else { /* Both RX and TX fds do not exist (equal -1). Create fd */ - *fd = tun_alloc(pmd); + *fd = tun_alloc(pmd, 0); if (*fd < 0) { *fd = -1; /* restore original value */ - RTE_LOG(ERR, PMD, "%s: tun_alloc() failed.\n", - pmd->name); + TAP_LOG(ERR, "%s: tun_alloc() failed.", pmd->name); return -1; } - RTE_LOG(DEBUG, PMD, "%s: add %s queue for qid %d fd %d\n", + TAP_LOG(DEBUG, "%s: add %s queue for qid %d fd %d", pmd->name, dir, qid, *fd); } tx->mtu = &dev->data->mtu; rx->rxmode = &dev->data->dev_conf.rxmode; + if (gso_ctx) { + ret = tap_gso_ctx_setup(gso_ctx, dev); + if (ret) + return -1; + } + + tx->type = pmd->type; return *fd; } @@ -1049,25 +1285,12 @@ tap_rx_queue_setup(struct rte_eth_dev *dev, int i; if (rx_queue_id >= dev->data->nb_rx_queues || !mp) { - RTE_LOG(WARNING, PMD, - "nb_rx_queues %d too small or mempool NULL\n", + TAP_LOG(WARNING, + "nb_rx_queues %d too small or mempool NULL", dev->data->nb_rx_queues); return -1; } - /* Verify application offloads are valid for our port and queue. */ - if (!tap_rxq_are_offloads_valid(dev, rx_conf->offloads)) { - rte_errno = ENOTSUP; - RTE_LOG(ERR, PMD, - "%p: Rx queue offloads 0x%" PRIx64 - " don't match port offloads 0x%" PRIx64 - " or supported offloads 0x%" PRIx64 "\n", - (void *)dev, rx_conf->offloads, - dev->data->dev_conf.rxmode.offloads, - (tap_rx_offload_get_port_capa() | - tap_rx_offload_get_queue_capa())); - return -rte_errno; - } rxq->mp = mp; rxq->trigger_seen = 1; /* force initial burst */ rxq->in_port = dev->data->port_id; @@ -1075,8 +1298,8 @@ tap_rx_queue_setup(struct rte_eth_dev *dev, iovecs = rte_zmalloc_socket(dev->device->name, sizeof(*iovecs), 0, socket_id); if (!iovecs) { - RTE_LOG(WARNING, PMD, - "%s: Couldn't allocate %d RX descriptors\n", + TAP_LOG(WARNING, + "%s: Couldn't allocate %d RX descriptors", dev->device->name, nb_desc); return -ENOMEM; } @@ -1095,8 +1318,8 @@ tap_rx_queue_setup(struct rte_eth_dev *dev, for (i = 1; i <= nb_desc; i++) { *tmp = rte_pktmbuf_alloc(rxq->mp); if (!*tmp) { - RTE_LOG(WARNING, PMD, - "%s: couldn't allocate memory for queue %d\n", + TAP_LOG(WARNING, + "%s: couldn't allocate memory for queue %d", dev->device->name, rx_queue_id); ret = -ENOMEM; goto error; @@ -1108,7 +1331,7 @@ tap_rx_queue_setup(struct rte_eth_dev *dev, tmp = &(*tmp)->next; } - RTE_LOG(DEBUG, PMD, " RX TAP device name %s, qid %d on fd %d\n", + TAP_LOG(DEBUG, " RX TUNTAP device name %s, qid %d on fd %d", internals->name, rx_queue_id, internals->rxq[rx_queue_id].fd); return 0; @@ -1131,39 +1354,24 @@ tap_tx_queue_setup(struct rte_eth_dev *dev, struct pmd_internals *internals = dev->data->dev_private; struct tx_queue *txq; int ret; + uint64_t offloads; if (tx_queue_id >= dev->data->nb_tx_queues) return -1; dev->data->tx_queues[tx_queue_id] = &internals->txq[tx_queue_id]; txq = dev->data->tx_queues[tx_queue_id]; - /* - * Don't verify port offloads for application which - * use the old API. - */ - if (tx_conf != NULL && - !!(tx_conf->txq_flags & ETH_TXQ_FLAGS_IGNORE)) { - if (tap_txq_are_offloads_valid(dev, tx_conf->offloads)) { - txq->csum = !!(tx_conf->offloads & - (DEV_TX_OFFLOAD_IPV4_CKSUM | - DEV_TX_OFFLOAD_UDP_CKSUM | - DEV_TX_OFFLOAD_TCP_CKSUM)); - } else { - rte_errno = ENOTSUP; - RTE_LOG(ERR, PMD, - "%p: Tx queue offloads 0x%" PRIx64 - " don't match port offloads 0x%" PRIx64 - " or supported offloads 0x%" PRIx64, - (void *)dev, tx_conf->offloads, - dev->data->dev_conf.txmode.offloads, - tap_tx_offload_get_port_capa()); - return -rte_errno; - } - } + + offloads = tx_conf->offloads | dev->data->dev_conf.txmode.offloads; + txq->csum = !!(offloads & + (DEV_TX_OFFLOAD_IPV4_CKSUM | + DEV_TX_OFFLOAD_UDP_CKSUM | + DEV_TX_OFFLOAD_TCP_CKSUM)); + ret = tap_setup_queue(dev, internals, tx_queue_id, 0); if (ret == -1) return -1; - RTE_LOG(DEBUG, PMD, - " TX TAP device name %s, qid %d on fd %d csum %s\n", + TAP_LOG(DEBUG, + " TX TUNTAP device name %s, qid %d on fd %d csum %s", internals->name, tx_queue_id, internals->txq[tx_queue_id].fd, txq->csum ? "on" : "off"); @@ -1307,6 +1515,70 @@ tap_flow_ctrl_set(struct rte_eth_dev *dev __rte_unused, return 0; } +/** + * DPDK callback to update the RSS hash configuration. + * + * @param dev + * Pointer to Ethernet device structure. + * @param[in] rss_conf + * RSS configuration data. + * + * @return + * 0 on success, a negative errno value otherwise and rte_errno is set. + */ +static int +tap_rss_hash_update(struct rte_eth_dev *dev, + struct rte_eth_rss_conf *rss_conf) +{ + if (rss_conf->rss_hf & TAP_RSS_HF_MASK) { + rte_errno = EINVAL; + return -rte_errno; + } + if (rss_conf->rss_key && rss_conf->rss_key_len) { + /* + * Currently TAP RSS key is hard coded + * and cannot be updated + */ + TAP_LOG(ERR, + "port %u RSS key cannot be updated", + dev->data->port_id); + rte_errno = EINVAL; + return -rte_errno; + } + return 0; +} + +static int +tap_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id) +{ + dev->data->rx_queue_state[rx_queue_id] = RTE_ETH_QUEUE_STATE_STARTED; + + return 0; +} + +static int +tap_tx_queue_start(struct rte_eth_dev *dev, uint16_t tx_queue_id) +{ + dev->data->tx_queue_state[tx_queue_id] = RTE_ETH_QUEUE_STATE_STARTED; + + return 0; +} + +static int +tap_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rx_queue_id) +{ + dev->data->rx_queue_state[rx_queue_id] = RTE_ETH_QUEUE_STATE_STOPPED; + + return 0; +} + +static int +tap_tx_queue_stop(struct rte_eth_dev *dev, uint16_t tx_queue_id) +{ + dev->data->tx_queue_state[tx_queue_id] = RTE_ETH_QUEUE_STATE_STOPPED; + + return 0; +} static const struct eth_dev_ops ops = { .dev_start = tap_dev_start, .dev_stop = tap_dev_stop, @@ -1315,6 +1587,10 @@ static const struct eth_dev_ops ops = { .dev_infos_get = tap_dev_info, .rx_queue_setup = tap_rx_queue_setup, .tx_queue_setup = tap_tx_queue_setup, + .rx_queue_start = tap_rx_queue_start, + .tx_queue_start = tap_tx_queue_start, + .rx_queue_stop = tap_rx_queue_stop, + .tx_queue_stop = tap_tx_queue_stop, .rx_queue_release = tap_rx_queue_release, .tx_queue_release = tap_tx_queue_release, .flow_ctrl_get = tap_flow_ctrl_get, @@ -1332,12 +1608,14 @@ static const struct eth_dev_ops ops = { .stats_get = tap_stats_get, .stats_reset = tap_stats_reset, .dev_supported_ptypes_get = tap_dev_supported_ptypes_get, + .rss_hash_update = tap_rss_hash_update, .filter_ctrl = tap_dev_filter_ctrl, }; static int eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, - char *remote_iface, int fixed_mac_type) + char *remote_iface, struct ether_addr *mac_addr, + enum rte_tuntap_type type) { int numa_node = rte_socket_id(); struct rte_eth_dev *dev; @@ -1346,34 +1624,31 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, struct ifreq ifr; int i; - RTE_LOG(DEBUG, PMD, " TAP device on numa %u\n", rte_socket_id()); - - data = rte_zmalloc_socket(tap_name, sizeof(*data), 0, numa_node); - if (!data) { - RTE_LOG(ERR, PMD, "TAP Failed to allocate data\n"); - goto error_exit_nodev; - } + TAP_LOG(DEBUG, "%s device on numa %u", + tuntap_name, rte_socket_id()); dev = rte_eth_vdev_allocate(vdev, sizeof(*pmd)); if (!dev) { - RTE_LOG(ERR, PMD, "TAP Unable to allocate device struct\n"); + TAP_LOG(ERR, "%s Unable to allocate device struct", + tuntap_name); goto error_exit_nodev; } pmd = dev->data->dev_private; pmd->dev = dev; snprintf(pmd->name, sizeof(pmd->name), "%s", tap_name); + pmd->type = type; pmd->ioctl_sock = socket(AF_INET, SOCK_DGRAM, 0); if (pmd->ioctl_sock == -1) { - RTE_LOG(ERR, PMD, - "TAP Unable to get a socket for management: %s\n", - strerror(errno)); + TAP_LOG(ERR, + "%s Unable to get a socket for management: %s", + tuntap_name, strerror(errno)); goto error_exit; } /* Setup some default values */ - rte_memcpy(data, dev->data, sizeof(*data)); + data = dev->data; data->dev_private = pmd; data->dev_flags = RTE_ETH_DEV_INTR_LSC; data->numa_node = numa_node; @@ -1384,7 +1659,6 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, data->nb_rx_queues = 0; data->nb_tx_queues = 0; - dev->data = data; dev->dev_ops = &ops; dev->rx_pkt_burst = pmd_rx_burst; dev->tx_pkt_burst = pmd_tx_burst; @@ -1394,39 +1668,43 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, dev->intr_handle = &pmd->intr_handle; /* Presetup the fds to -1 as being not valid */ + pmd->ka_fd = -1; for (i = 0; i < RTE_PMD_TAP_MAX_QUEUES; i++) { pmd->rxq[i].fd = -1; pmd->txq[i].fd = -1; } - if (fixed_mac_type) { - /* fixed mac = 00:64:74:61:70:<iface_idx> */ - static int iface_idx; - char mac[ETHER_ADDR_LEN] = "\0dtap"; - - mac[ETHER_ADDR_LEN - 1] = iface_idx++; - rte_memcpy(&pmd->eth_addr, mac, ETHER_ADDR_LEN); - } else { - eth_random_addr((uint8_t *)&pmd->eth_addr); + if (pmd->type == ETH_TUNTAP_TYPE_TAP) { + if (is_zero_ether_addr(mac_addr)) + eth_random_addr((uint8_t *)&pmd->eth_addr); + else + rte_memcpy(&pmd->eth_addr, mac_addr, sizeof(*mac_addr)); } - /* Immediately create the netdevice (this will create the 1st queue). */ - /* rx queue */ - if (tap_setup_queue(dev, pmd, 0, 1) == -1) - goto error_exit; - /* tx queue */ - if (tap_setup_queue(dev, pmd, 0, 0) == -1) + /* + * Allocate a TUN device keep-alive file descriptor that will only be + * closed when the TUN device itself is closed or removed. + * This keep-alive file descriptor will guarantee that the TUN device + * exists even when all of its queues are closed + */ + pmd->ka_fd = tun_alloc(pmd, 1); + if (pmd->ka_fd == -1) { + TAP_LOG(ERR, "Unable to create %s interface", tuntap_name); goto error_exit; + } ifr.ifr_mtu = dev->data->mtu; if (tap_ioctl(pmd, SIOCSIFMTU, &ifr, 1, LOCAL_AND_REMOTE) < 0) goto error_exit; - memset(&ifr, 0, sizeof(struct ifreq)); - ifr.ifr_hwaddr.sa_family = AF_LOCAL; - rte_memcpy(ifr.ifr_hwaddr.sa_data, &pmd->eth_addr, ETHER_ADDR_LEN); - if (tap_ioctl(pmd, SIOCSIFHWADDR, &ifr, 0, LOCAL_ONLY) < 0) - goto error_exit; + if (pmd->type == ETH_TUNTAP_TYPE_TAP) { + memset(&ifr, 0, sizeof(struct ifreq)); + ifr.ifr_hwaddr.sa_family = AF_LOCAL; + rte_memcpy(ifr.ifr_hwaddr.sa_data, &pmd->eth_addr, + ETHER_ADDR_LEN); + if (tap_ioctl(pmd, SIOCSIFHWADDR, &ifr, 0, LOCAL_ONLY) < 0) + goto error_exit; + } /* * Set up everything related to rte_flow: @@ -1438,22 +1716,22 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, */ pmd->nlsk_fd = tap_nl_init(0); if (pmd->nlsk_fd == -1) { - RTE_LOG(WARNING, PMD, "%s: failed to create netlink socket.\n", + TAP_LOG(WARNING, "%s: failed to create netlink socket.", pmd->name); goto disable_rte_flow; } pmd->if_index = if_nametoindex(pmd->name); if (!pmd->if_index) { - RTE_LOG(ERR, PMD, "%s: failed to get if_index.\n", pmd->name); + TAP_LOG(ERR, "%s: failed to get if_index.", pmd->name); goto disable_rte_flow; } if (qdisc_create_multiq(pmd->nlsk_fd, pmd->if_index) < 0) { - RTE_LOG(ERR, PMD, "%s: failed to create multiq qdisc.\n", + TAP_LOG(ERR, "%s: failed to create multiq qdisc.", pmd->name); goto disable_rte_flow; } if (qdisc_create_ingress(pmd->nlsk_fd, pmd->if_index) < 0) { - RTE_LOG(ERR, PMD, "%s: failed to create ingress qdisc.\n", + TAP_LOG(ERR, "%s: failed to create ingress qdisc.", pmd->name); goto disable_rte_flow; } @@ -1462,7 +1740,7 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, if (strlen(remote_iface)) { pmd->remote_if_index = if_nametoindex(remote_iface); if (!pmd->remote_if_index) { - RTE_LOG(ERR, PMD, "%s: failed to get %s if_index.\n", + TAP_LOG(ERR, "%s: failed to get %s if_index.", pmd->name, remote_iface); goto error_remote; } @@ -1474,7 +1752,7 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, /* Replicate remote MAC address */ if (tap_ioctl(pmd, SIOCGIFHWADDR, &ifr, 0, REMOTE_ONLY) < 0) { - RTE_LOG(ERR, PMD, "%s: failed to get %s MAC address.\n", + TAP_LOG(ERR, "%s: failed to get %s MAC address.", pmd->name, pmd->remote_iface); goto error_remote; } @@ -1482,7 +1760,7 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, ETHER_ADDR_LEN); /* The desired MAC is already in ifreq after SIOCGIFHWADDR. */ if (tap_ioctl(pmd, SIOCSIFHWADDR, &ifr, 0, LOCAL_ONLY) < 0) { - RTE_LOG(ERR, PMD, "%s: failed to get %s MAC address.\n", + TAP_LOG(ERR, "%s: failed to get %s MAC address.", pmd->name, remote_iface); goto error_remote; } @@ -1495,7 +1773,7 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, qdisc_flush(pmd->nlsk_fd, pmd->remote_if_index); if (qdisc_create_ingress(pmd->nlsk_fd, pmd->remote_if_index) < 0) { - RTE_LOG(ERR, PMD, "%s: failed to create ingress qdisc.\n", + TAP_LOG(ERR, "%s: failed to create ingress qdisc.", pmd->remote_iface); goto error_remote; } @@ -1504,26 +1782,27 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, tap_flow_implicit_create(pmd, TAP_REMOTE_LOCAL_MAC) < 0 || tap_flow_implicit_create(pmd, TAP_REMOTE_BROADCAST) < 0 || tap_flow_implicit_create(pmd, TAP_REMOTE_BROADCASTV6) < 0) { - RTE_LOG(ERR, PMD, - "%s: failed to create implicit rules.\n", + TAP_LOG(ERR, + "%s: failed to create implicit rules.", pmd->name); goto error_remote; } } + rte_eth_dev_probing_finish(dev); return 0; disable_rte_flow: - RTE_LOG(ERR, PMD, " Disabling rte flow support: %s(%d)\n", + TAP_LOG(ERR, " Disabling rte flow support: %s(%d)", strerror(errno), errno); if (strlen(remote_iface)) { - RTE_LOG(ERR, PMD, "Remote feature requires flow support.\n"); + TAP_LOG(ERR, "Remote feature requires flow support."); goto error_exit; } return 0; error_remote: - RTE_LOG(ERR, PMD, " Can't set up remote feature: %s(%d)\n", + TAP_LOG(ERR, " Can't set up remote feature: %s(%d)", strerror(errno), errno); tap_flow_implicit_flush(pmd, NULL); @@ -1533,10 +1812,9 @@ error_exit: rte_eth_dev_release_port(dev); error_exit_nodev: - RTE_LOG(ERR, PMD, "TAP Unable to initialize %s\n", - rte_vdev_device_name(vdev)); + TAP_LOG(ERR, "%s Unable to initialize %s", + tuntap_name, rte_vdev_device_name(vdev)); - rte_free(data); return -EINVAL; } @@ -1548,7 +1826,7 @@ set_interface_name(const char *key __rte_unused, char *name = (char *)extra_args; if (value) - snprintf(name, RTE_ETH_NAME_MAX_LEN - 1, "%s", value); + strlcpy(name, value, RTE_ETH_NAME_MAX_LEN - 1); else snprintf(name, RTE_ETH_NAME_MAX_LEN - 1, "%s%d", DEFAULT_TAP_NAME, (tap_unit - 1)); @@ -1564,20 +1842,136 @@ set_remote_iface(const char *key __rte_unused, char *name = (char *)extra_args; if (value) - snprintf(name, RTE_ETH_NAME_MAX_LEN, "%s", value); + strlcpy(name, value, RTE_ETH_NAME_MAX_LEN); return 0; } +static int parse_user_mac(struct ether_addr *user_mac, + const char *value) +{ + unsigned int index = 0; + char mac_temp[strlen(ETH_TAP_USR_MAC_FMT) + 1], *mac_byte = NULL; + + if (user_mac == NULL || value == NULL) + return 0; + + strlcpy(mac_temp, value, sizeof(mac_temp)); + mac_byte = strtok(mac_temp, ":"); + + while ((mac_byte != NULL) && + (strlen(mac_byte) <= 2) && + (strlen(mac_byte) == strspn(mac_byte, + ETH_TAP_CMP_MAC_FMT))) { + user_mac->addr_bytes[index++] = strtoul(mac_byte, NULL, 16); + mac_byte = strtok(NULL, ":"); + } + + return index; +} + static int set_mac_type(const char *key __rte_unused, const char *value, void *extra_args) { - if (value && - !strncasecmp(ETH_TAP_MAC_FIXED, value, strlen(ETH_TAP_MAC_FIXED))) - *(int *)extra_args = 1; + struct ether_addr *user_mac = extra_args; + + if (!value) + return 0; + + if (!strncasecmp(ETH_TAP_MAC_FIXED, value, strlen(ETH_TAP_MAC_FIXED))) { + static int iface_idx; + + /* fixed mac = 00:64:74:61:70:<iface_idx> */ + memcpy((char *)user_mac->addr_bytes, "\0dtap", ETHER_ADDR_LEN); + user_mac->addr_bytes[ETHER_ADDR_LEN - 1] = iface_idx++ + '0'; + goto success; + } + + if (parse_user_mac(user_mac, value) != 6) + goto error; +success: + TAP_LOG(DEBUG, "TAP user MAC param (%s)", value); return 0; + +error: + TAP_LOG(ERR, "TAP user MAC (%s) is not in format (%s|%s)", + value, ETH_TAP_MAC_FIXED, ETH_TAP_USR_MAC_FMT); + return -1; +} + +/* + * Open a TUN interface device. TUN PMD + * 1) sets tap_type as false + * 2) intakes iface as argument. + * 3) as interface is virtual set speed to 10G + */ +static int +rte_pmd_tun_probe(struct rte_vdev_device *dev) +{ + const char *name, *params; + int ret; + struct rte_kvargs *kvlist = NULL; + char tun_name[RTE_ETH_NAME_MAX_LEN]; + char remote_iface[RTE_ETH_NAME_MAX_LEN]; + struct rte_eth_dev *eth_dev; + + strcpy(tuntap_name, "TUN"); + + name = rte_vdev_device_name(dev); + params = rte_vdev_device_args(dev); + memset(remote_iface, 0, RTE_ETH_NAME_MAX_LEN); + + if (rte_eal_process_type() == RTE_PROC_SECONDARY && + strlen(params) == 0) { + eth_dev = rte_eth_dev_attach_secondary(name); + if (!eth_dev) { + TAP_LOG(ERR, "Failed to probe %s", name); + return -1; + } + eth_dev->dev_ops = &ops; + eth_dev->device = &dev->device; + rte_eth_dev_probing_finish(eth_dev); + return 0; + } + + snprintf(tun_name, sizeof(tun_name), "%s%u", + DEFAULT_TUN_NAME, tun_unit++); + + if (params && (params[0] != '\0')) { + TAP_LOG(DEBUG, "parameters (%s)", params); + + kvlist = rte_kvargs_parse(params, valid_arguments); + if (kvlist) { + if (rte_kvargs_count(kvlist, ETH_TAP_IFACE_ARG) == 1) { + ret = rte_kvargs_process(kvlist, + ETH_TAP_IFACE_ARG, + &set_interface_name, + tun_name); + + if (ret == -1) + goto leave; + } + } + } + pmd_link.link_speed = ETH_SPEED_NUM_10G; + + TAP_LOG(NOTICE, "Initializing pmd_tun for %s as %s", + name, tun_name); + + ret = eth_dev_tap_create(dev, tun_name, remote_iface, 0, + ETH_TUNTAP_TYPE_TUN); + +leave: + if (ret == -1) { + TAP_LOG(ERR, "Failed to create pmd for %s as %s", + name, tun_name); + tun_unit--; /* Restore the unit number */ + } + rte_kvargs_free(kvlist); + + return ret; } /* Open a TAP interface device. @@ -1591,18 +1985,35 @@ rte_pmd_tap_probe(struct rte_vdev_device *dev) int speed; char tap_name[RTE_ETH_NAME_MAX_LEN]; char remote_iface[RTE_ETH_NAME_MAX_LEN]; - int fixed_mac_type = 0; + struct ether_addr user_mac = { .addr_bytes = {0} }; + struct rte_eth_dev *eth_dev; + + strcpy(tuntap_name, "TAP"); name = rte_vdev_device_name(dev); params = rte_vdev_device_args(dev); + if (rte_eal_process_type() == RTE_PROC_SECONDARY && + strlen(params) == 0) { + eth_dev = rte_eth_dev_attach_secondary(name); + if (!eth_dev) { + TAP_LOG(ERR, "Failed to probe %s", name); + return -1; + } + /* TODO: request info from primary to set up Rx and Tx */ + eth_dev->dev_ops = &ops; + eth_dev->device = &dev->device; + rte_eth_dev_probing_finish(eth_dev); + return 0; + } + speed = ETH_SPEED_NUM_10G; - snprintf(tap_name, sizeof(tap_name), "%s%d", + snprintf(tap_name, sizeof(tap_name), "%s%u", DEFAULT_TAP_NAME, tap_unit++); memset(remote_iface, 0, RTE_ETH_NAME_MAX_LEN); if (params && (params[0] != '\0')) { - RTE_LOG(DEBUG, PMD, "parameters (%s)\n", params); + TAP_LOG(DEBUG, "parameters (%s)", params); kvlist = rte_kvargs_parse(params, valid_arguments); if (kvlist) { @@ -1628,7 +2039,7 @@ rte_pmd_tap_probe(struct rte_vdev_device *dev) ret = rte_kvargs_process(kvlist, ETH_TAP_MAC_ARG, &set_mac_type, - &fixed_mac_type); + &user_mac); if (ret == -1) goto leave; } @@ -1636,14 +2047,15 @@ rte_pmd_tap_probe(struct rte_vdev_device *dev) } pmd_link.link_speed = speed; - RTE_LOG(NOTICE, PMD, "Initializing pmd_tap for %s as %s\n", + TAP_LOG(NOTICE, "Initializing pmd_tap for %s as %s", name, tap_name); - ret = eth_dev_tap_create(dev, tap_name, remote_iface, fixed_mac_type); + ret = eth_dev_tap_create(dev, tap_name, remote_iface, &user_mac, + ETH_TUNTAP_TYPE_TAP); leave: if (ret == -1) { - RTE_LOG(ERR, PMD, "Failed to create pmd for %s as %s\n", + TAP_LOG(ERR, "Failed to create pmd for %s as %s", name, tap_name); tap_unit--; /* Restore the unit number */ } @@ -1652,7 +2064,7 @@ leave: return ret; } -/* detach a TAP device. +/* detach a TUNTAP device. */ static int rte_pmd_tap_remove(struct rte_vdev_device *dev) @@ -1661,15 +2073,17 @@ rte_pmd_tap_remove(struct rte_vdev_device *dev) struct pmd_internals *internals; int i; - RTE_LOG(DEBUG, PMD, "Closing TUN/TAP Ethernet device on numa %u\n", - rte_socket_id()); - /* find the ethdev entry */ eth_dev = rte_eth_dev_allocated(rte_vdev_device_name(dev)); if (!eth_dev) return 0; internals = eth_dev->data->dev_private; + + TAP_LOG(DEBUG, "Closing %s Ethernet device on numa %u", + (internals->type == ETH_TUNTAP_TYPE_TAP) ? "TAP" : "TUN", + rte_socket_id()); + if (internals->nlsk_fd) { tap_flow_flush(eth_dev, NULL); tap_flow_implicit_flush(internals, NULL); @@ -1688,20 +2102,39 @@ rte_pmd_tap_remove(struct rte_vdev_device *dev) close(internals->ioctl_sock); rte_free(eth_dev->data->dev_private); - rte_free(eth_dev->data); - rte_eth_dev_release_port(eth_dev); + if (internals->ka_fd != -1) { + close(internals->ka_fd); + internals->ka_fd = -1; + } return 0; } +static struct rte_vdev_driver pmd_tun_drv = { + .probe = rte_pmd_tun_probe, + .remove = rte_pmd_tap_remove, +}; + static struct rte_vdev_driver pmd_tap_drv = { .probe = rte_pmd_tap_probe, .remove = rte_pmd_tap_remove, }; + RTE_PMD_REGISTER_VDEV(net_tap, pmd_tap_drv); +RTE_PMD_REGISTER_VDEV(net_tun, pmd_tun_drv); RTE_PMD_REGISTER_ALIAS(net_tap, eth_tap); +RTE_PMD_REGISTER_PARAM_STRING(net_tun, + ETH_TAP_IFACE_ARG "=<string> "); RTE_PMD_REGISTER_PARAM_STRING(net_tap, ETH_TAP_IFACE_ARG "=<string> " - ETH_TAP_MAC_ARG "=" ETH_TAP_MAC_FIXED " " + ETH_TAP_MAC_ARG "=" ETH_TAP_MAC_ARG_FMT " " ETH_TAP_REMOTE_ARG "=<string>"); +int tap_logtype; + +RTE_INIT(tap_init_log) +{ + tap_logtype = rte_log_register("pmd.net.tap"); + if (tap_logtype >= 0) + rte_log_set_level(tap_logtype, RTE_LOG_NOTICE); +} diff --git a/drivers/net/tap/rte_eth_tap.h b/drivers/net/tap/rte_eth_tap.h index 53a506ad..44e2773f 100644 --- a/drivers/net/tap/rte_eth_tap.h +++ b/drivers/net/tap/rte_eth_tap.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #ifndef _RTE_ETH_TAP_H_ @@ -15,12 +15,22 @@ #include <rte_ethdev_driver.h> #include <rte_ether.h> +#include <rte_gso.h> +#include "tap_log.h" #ifdef IFF_MULTI_QUEUE #define RTE_PMD_TAP_MAX_QUEUES TAP_MAX_QUEUES #else #define RTE_PMD_TAP_MAX_QUEUES 1 #endif +#define MAX_GSO_MBUFS 64 + +enum rte_tuntap_type { + ETH_TUNTAP_TYPE_UNKNOWN, + ETH_TUNTAP_TYPE_TUN, + ETH_TUNTAP_TYPE_TAP, + ETH_TUNTAP_TYPE_MAX, +}; struct pkt_stats { uint64_t opackets; /* Number of output packets */ @@ -47,15 +57,18 @@ struct rx_queue { struct tx_queue { int fd; + int type; /* Type field - TUN|TAP */ uint16_t *mtu; /* Pointer to MTU from dev_data */ uint16_t csum:1; /* Enable checksum offloading */ struct pkt_stats stats; /* Stats for this TX queue */ + struct rte_gso_ctx gso_ctx; /* GSO context */ }; struct pmd_internals { struct rte_eth_dev *dev; /* Ethernet device. */ char remote_iface[RTE_ETH_NAME_MAX_LEN]; /* Remote netdevice name */ char name[RTE_ETH_NAME_MAX_LEN]; /* Internal Tap device name */ + int type; /* Type field - TUN|TAP */ struct ether_addr eth_addr; /* Mac address of the device port */ struct ifreq remote_initial_flags; /* Remote netdevice flags on init */ int remote_if_index; /* remote netdevice IF_INDEX */ @@ -76,6 +89,7 @@ struct pmd_internals { struct rx_queue rxq[RTE_PMD_TAP_MAX_QUEUES]; /* List of RX queues */ struct tx_queue txq[RTE_PMD_TAP_MAX_QUEUES]; /* List of TX queues */ struct rte_intr_handle intr_handle; /* LSC interrupt handle. */ + int ka_fd; /* keep-alive file descriptor */ }; /* tap_intr.c */ diff --git a/drivers/net/tap/tap_bpf.h b/drivers/net/tap/tap_bpf.h index 1a70ffe2..9192686a 100644 --- a/drivers/net/tap/tap_bpf.h +++ b/drivers/net/tap/tap_bpf.h @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 - * Copyright 2017 Mellanox Technologies, Ltd. + * Copyright 2017 Mellanox Technologies, Ltd */ #ifndef __TAP_BPF_H__ diff --git a/drivers/net/tap/tap_bpf_api.c b/drivers/net/tap/tap_bpf_api.c index 109a681e..98f6a760 100644 --- a/drivers/net/tap/tap_bpf_api.c +++ b/drivers/net/tap/tap_bpf_api.c @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: BSD-3-Clause - * Copyright 2017 Mellanox Technologies, Ltd. + * Copyright 2017 Mellanox Technologies, Ltd */ #include <errno.h> diff --git a/drivers/net/tap/tap_bpf_insns.h b/drivers/net/tap/tap_bpf_insns.h index 89873b6d..79e3e66b 100644 --- a/drivers/net/tap/tap_bpf_insns.h +++ b/drivers/net/tap/tap_bpf_insns.h @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: BSD-3-Clause - * Copyright 2017 Mellanox Technologies, Ltd. + * Copyright 2017 Mellanox Technologies, Ltd */ #include <tap_bpf.h> diff --git a/drivers/net/tap/tap_bpf_program.c b/drivers/net/tap/tap_bpf_program.c index 8abb3b76..1cb73822 100644 --- a/drivers/net/tap/tap_bpf_program.c +++ b/drivers/net/tap/tap_bpf_program.c @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 - * Copyright 2017 Mellanox Technologies, Ltd. + * Copyright 2017 Mellanox Technologies, Ltd */ #include <stdint.h> @@ -84,7 +84,7 @@ struct ipv6_l3_l4_tuple { __u16 sport; } __attribute__((packed)); -static const __u8 def_rss_key[] = { +static const __u8 def_rss_key[TAP_RSS_HASH_KEY_SIZE] = { 0xd1, 0x81, 0xc6, 0x2c, 0xf7, 0xf4, 0xdb, 0x5b, 0x19, 0x83, 0xa2, 0xfc, diff --git a/drivers/net/tap/tap_flow.c b/drivers/net/tap/tap_flow.c index 551b2d83..0e01af62 100644 --- a/drivers/net/tap/tap_flow.c +++ b/drivers/net/tap/tap_flow.c @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #include <errno.h> @@ -270,13 +270,13 @@ static const struct tap_flow_items tap_flow_items[] = { .items = ITEMS(RTE_FLOW_ITEM_TYPE_IPV4, RTE_FLOW_ITEM_TYPE_IPV6), .mask = &(const struct rte_flow_item_vlan){ - .tpid = -1, /* DEI matching is not supported */ #if RTE_BYTE_ORDER == RTE_LITTLE_ENDIAN .tci = 0xffef, #else .tci = 0xefff, #endif + .inner_type = -1, }, .mask_sz = sizeof(struct rte_flow_item_vlan), .default_mask = &rte_flow_item_vlan_mask, @@ -537,7 +537,7 @@ tap_flow_create_eth(const struct rte_flow_item *item, void *data) if (!flow) return 0; msg = &flow->msg; - if (!is_zero_ether_addr(&spec->dst)) { + if (!is_zero_ether_addr(&mask->dst)) { tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_ETH_DST, ETHER_ADDR_LEN, &spec->dst.addr_bytes); tap_nlattr_add(&msg->nh, @@ -578,13 +578,19 @@ tap_flow_create_vlan(const struct rte_flow_item *item, void *data) /* use default mask if none provided */ if (!mask) mask = tap_flow_items[RTE_FLOW_ITEM_TYPE_VLAN].default_mask; - /* TC does not support tpid masking. Only accept if exact match. */ - if (mask->tpid && mask->tpid != 0xffff) + /* Outer TPID cannot be matched. */ + if (info->eth_type) return -1; /* Double-tagging not supported. */ - if (spec && mask->tpid && spec->tpid != htons(ETH_P_8021Q)) + if (info->vlan) return -1; info->vlan = 1; + if (mask->inner_type) { + /* TC does not support partial eth_type masking */ + if (mask->inner_type != RTE_BE16(0xffff)) + return -1; + info->eth_type = spec->inner_type; + } if (!flow) return 0; msg = &flow->msg; @@ -645,13 +651,13 @@ tap_flow_create_ipv4(const struct rte_flow_item *item, void *data) info->eth_type = htons(ETH_P_IP); if (!spec) return 0; - if (spec->hdr.dst_addr) { + if (mask->hdr.dst_addr) { tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_DST, spec->hdr.dst_addr); tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_DST_MASK, mask->hdr.dst_addr); } - if (spec->hdr.src_addr) { + if (mask->hdr.src_addr) { tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_SRC, spec->hdr.src_addr); tap_nlattr_add32(&msg->nh, TCA_FLOWER_KEY_IPV4_SRC_MASK, @@ -701,13 +707,13 @@ tap_flow_create_ipv6(const struct rte_flow_item *item, void *data) info->eth_type = htons(ETH_P_IPV6); if (!spec) return 0; - if (memcmp(spec->hdr.dst_addr, empty_addr, 16)) { + if (memcmp(mask->hdr.dst_addr, empty_addr, 16)) { tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_DST, sizeof(spec->hdr.dst_addr), &spec->hdr.dst_addr); tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_DST_MASK, sizeof(mask->hdr.dst_addr), &mask->hdr.dst_addr); } - if (memcmp(spec->hdr.src_addr, empty_addr, 16)) { + if (memcmp(mask->hdr.src_addr, empty_addr, 16)) { tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_SRC, sizeof(spec->hdr.src_addr), &spec->hdr.src_addr); tap_nlattr_add(&msg->nh, TCA_FLOWER_KEY_IPV6_SRC_MASK, @@ -756,10 +762,10 @@ tap_flow_create_udp(const struct rte_flow_item *item, void *data) tap_nlattr_add8(&msg->nh, TCA_FLOWER_KEY_IP_PROTO, IPPROTO_UDP); if (!spec) return 0; - if (spec->hdr.dst_port & mask->hdr.dst_port) + if (mask->hdr.dst_port) tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_UDP_DST, spec->hdr.dst_port); - if (spec->hdr.src_port & mask->hdr.src_port) + if (mask->hdr.src_port) tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_UDP_SRC, spec->hdr.src_port); return 0; @@ -802,10 +808,10 @@ tap_flow_create_tcp(const struct rte_flow_item *item, void *data) tap_nlattr_add8(&msg->nh, TCA_FLOWER_KEY_IP_PROTO, IPPROTO_TCP); if (!spec) return 0; - if (spec->hdr.dst_port & mask->hdr.dst_port) + if (mask->hdr.dst_port) tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_TCP_DST, spec->hdr.dst_port); - if (spec->hdr.src_port & mask->hdr.src_port) + if (mask->hdr.src_port) tap_nlattr_add16(&msg->nh, TCA_FLOWER_KEY_TCP_SRC, spec->hdr.src_port); return 0; @@ -1033,6 +1039,12 @@ priv_flow_process(struct pmd_internals *pmd, }; int action = 0; /* Only one action authorized for now */ + if (attr->transfer) { + rte_flow_error_set( + error, ENOTSUP, RTE_FLOW_ERROR_TYPE_ATTR_TRANSFER, + NULL, "transfer is not supported"); + return -rte_errno; + } if (attr->group > MAX_GROUP) { rte_flow_error_set( error, EINVAL, RTE_FLOW_ERROR_TYPE_ATTR_GROUP, @@ -1140,6 +1152,7 @@ priv_flow_process(struct pmd_internals *pmd, else goto end; } +actions: for (; actions->type != RTE_FLOW_ACTION_TYPE_END; ++actions) { int err = 0; @@ -1214,7 +1227,7 @@ priv_flow_process(struct pmd_internals *pmd, if (err) goto exit_action_not_supported; } - if (flow && rss) + if (flow) err = rss_add_actions(flow, pmd, rss, error); } else { goto exit_action_not_supported; @@ -1222,6 +1235,16 @@ priv_flow_process(struct pmd_internals *pmd, if (err) goto exit_action_not_supported; } + /* When fate is unknown, drop traffic. */ + if (!action) { + static const struct rte_flow_action drop[] = { + { .type = RTE_FLOW_ACTION_TYPE_DROP, }, + { .type = RTE_FLOW_ACTION_TYPE_END, }, + }; + + actions = drop; + goto actions; + } end: if (flow) tap_nlattr_nested_finish(&flow->msg); /* nested TCA_OPTIONS */ @@ -1376,8 +1399,8 @@ tap_flow_create(struct rte_eth_dev *dev, } err = tap_nl_recv_ack(pmd->nlsk_fd); if (err < 0) { - RTE_LOG(ERR, PMD, - "Kernel refused TC filter rule creation (%d): %s\n", + TAP_LOG(ERR, + "Kernel refused TC filter rule creation (%d): %s", errno, strerror(errno)); rte_flow_error_set(error, EEXIST, RTE_FLOW_ERROR_TYPE_HANDLE, NULL, @@ -1421,8 +1444,8 @@ tap_flow_create(struct rte_eth_dev *dev, } err = tap_nl_recv_ack(pmd->nlsk_fd); if (err < 0) { - RTE_LOG(ERR, PMD, - "Kernel refused TC filter rule creation (%d): %s\n", + TAP_LOG(ERR, + "Kernel refused TC filter rule creation (%d): %s", errno, strerror(errno)); rte_flow_error_set( error, ENOMEM, RTE_FLOW_ERROR_TYPE_HANDLE, @@ -1476,8 +1499,8 @@ tap_flow_destroy_pmd(struct pmd_internals *pmd, if (ret < 0 && errno == ENOENT) ret = 0; if (ret < 0) { - RTE_LOG(ERR, PMD, - "Kernel refused TC filter rule deletion (%d): %s\n", + TAP_LOG(ERR, + "Kernel refused TC filter rule deletion (%d): %s", errno, strerror(errno)); rte_flow_error_set( error, ENOTSUP, RTE_FLOW_ERROR_TYPE_HANDLE, NULL, @@ -1500,8 +1523,8 @@ tap_flow_destroy_pmd(struct pmd_internals *pmd, if (ret < 0 && errno == ENOENT) ret = 0; if (ret < 0) { - RTE_LOG(ERR, PMD, - "Kernel refused TC filter rule deletion (%d): %s\n", + TAP_LOG(ERR, + "Kernel refused TC filter rule deletion (%d): %s", errno, strerror(errno)); rte_flow_error_set( error, ENOMEM, RTE_FLOW_ERROR_TYPE_HANDLE, @@ -1545,10 +1568,14 @@ tap_flow_isolate(struct rte_eth_dev *dev, { struct pmd_internals *pmd = dev->data->dev_private; + /* normalize 'set' variable to contain 0 or 1 values */ if (set) - pmd->flow_isolate = 1; - else - pmd->flow_isolate = 0; + set = 1; + /* if already in the right isolation mode - nothing to do */ + if ((set ^ pmd->flow_isolate) == 0) + return 0; + /* mark the isolation mode for tap_flow_implicit_create() */ + pmd->flow_isolate = set; /* * If netdevice is there, setup appropriate flow rules immediately. * Otherwise it will be set when bringing up the netdevice (tun_alloc). @@ -1556,20 +1583,20 @@ tap_flow_isolate(struct rte_eth_dev *dev, if (!pmd->rxq[0].fd) return 0; if (set) { - struct rte_flow *flow; + struct rte_flow *remote_flow; while (1) { - flow = LIST_FIRST(&pmd->implicit_flows); - if (!flow) + remote_flow = LIST_FIRST(&pmd->implicit_flows); + if (!remote_flow) break; /* * Remove all implicit rules on the remote. * Keep the local rule to redirect packets on TX. * Keep also the last implicit local rule: ISOLATE. */ - if (flow->msg.t.tcm_ifindex == pmd->if_index) + if (remote_flow->msg.t.tcm_ifindex == pmd->if_index) break; - if (tap_flow_destroy_pmd(pmd, flow, NULL) < 0) + if (tap_flow_destroy_pmd(pmd, remote_flow, NULL) < 0) goto error; } /* Switch the TC rule according to pmd->flow_isolate */ @@ -1665,7 +1692,7 @@ int tap_flow_implicit_create(struct pmd_internals *pmd, remote_flow = rte_malloc(__func__, sizeof(struct rte_flow), 0); if (!remote_flow) { - RTE_LOG(ERR, PMD, "Cannot allocate memory for rte_flow\n"); + TAP_LOG(ERR, "Cannot allocate memory for rte_flow"); goto fail; } msg = &remote_flow->msg; @@ -1706,21 +1733,21 @@ int tap_flow_implicit_create(struct pmd_internals *pmd, tap_flow_set_handle(remote_flow); if (priv_flow_process(pmd, attr, items, actions, NULL, remote_flow, implicit_rte_flows[idx].mirred)) { - RTE_LOG(ERR, PMD, "rte flow rule validation failed\n"); + TAP_LOG(ERR, "rte flow rule validation failed"); goto fail; } err = tap_nl_send(pmd->nlsk_fd, &msg->nh); if (err < 0) { - RTE_LOG(ERR, PMD, "Failure sending nl request\n"); + TAP_LOG(ERR, "Failure sending nl request"); goto fail; } err = tap_nl_recv_ack(pmd->nlsk_fd); if (err < 0) { - /* Silently ignore re-entering remote promiscuous rule */ - if (errno == EEXIST && idx == TAP_REMOTE_PROMISC) + /* Silently ignore re-entering existing rule */ + if (errno == EEXIST) goto success; - RTE_LOG(ERR, PMD, - "Kernel refused TC filter rule creation (%d): %s\n", + TAP_LOG(ERR, + "Kernel refused TC filter rule creation (%d): %s", errno, strerror(errno)); goto fail; } @@ -1836,8 +1863,8 @@ static int rss_enable(struct pmd_internals *pmd, sizeof(struct rss_key), MAX_RSS_KEYS); if (pmd->map_fd < 0) { - RTE_LOG(ERR, PMD, - "Failed to create BPF map (%d): %s\n", + TAP_LOG(ERR, + "Failed to create BPF map (%d): %s", errno, strerror(errno)); rte_flow_error_set( error, ENOTSUP, RTE_FLOW_ERROR_TYPE_HANDLE, NULL, @@ -1854,7 +1881,7 @@ static int rss_enable(struct pmd_internals *pmd, for (i = 0; i < pmd->dev->data->nb_rx_queues; i++) { pmd->bpf_fd[i] = tap_flow_bpf_cls_q(i); if (pmd->bpf_fd[i] < 0) { - RTE_LOG(ERR, PMD, + TAP_LOG(ERR, "Failed to load BPF section %s for queue %d", SEC_NAME_CLS_Q, i); rte_flow_error_set( @@ -1868,7 +1895,7 @@ static int rss_enable(struct pmd_internals *pmd, rss_flow = rte_malloc(__func__, sizeof(struct rte_flow), 0); if (!rss_flow) { - RTE_LOG(ERR, PMD, + TAP_LOG(ERR, "Cannot allocate memory for rte_flow"); return -1; } @@ -1911,8 +1938,8 @@ static int rss_enable(struct pmd_internals *pmd, return -1; err = tap_nl_recv_ack(pmd->nlsk_fd); if (err < 0) { - RTE_LOG(ERR, PMD, - "Kernel refused TC filter rule creation (%d): %s\n", + TAP_LOG(ERR, + "Kernel refused TC filter rule creation (%d): %s", errno, strerror(errno)); return err; } @@ -2039,11 +2066,21 @@ static int rss_add_actions(struct rte_flow *flow, struct pmd_internals *pmd, struct rte_flow_error *error) { /* 4096 is the maximum number of instructions for a BPF program */ - int i; + unsigned int i; int err; struct rss_key rss_entry = { .hash_fields = 0, .key_size = 0 }; + /* Check supported RSS features */ + if (rss->func != RTE_ETH_HASH_FUNCTION_DEFAULT) + return rte_flow_error_set + (error, ENOTSUP, RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "non-default RSS hash functions are not supported"); + if (rss->level) + return rte_flow_error_set + (error, ENOTSUP, RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "a nonzero RSS encapsulation level is not supported"); + /* Get a new map key for a new RSS rule */ err = bpf_rss_key(KEY_CMD_GET, &flow->key_idx); if (err < 0) { @@ -2055,8 +2092,8 @@ static int rss_add_actions(struct rte_flow *flow, struct pmd_internals *pmd, } /* Update RSS map entry with queues */ - rss_entry.nb_queues = rss->num; - for (i = 0; i < rss->num; i++) + rss_entry.nb_queues = rss->queue_num; + for (i = 0; i < rss->queue_num; i++) rss_entry.queues[i] = rss->queue[i]; rss_entry.hash_fields = (1 << HASH_FIELD_IPV4_L3_L4) | (1 << HASH_FIELD_IPV6_L3_L4); @@ -2066,8 +2103,8 @@ static int rss_add_actions(struct rte_flow *flow, struct pmd_internals *pmd, &flow->key_idx, &rss_entry); if (err) { - RTE_LOG(ERR, PMD, - "Failed to update BPF map entry #%u (%d): %s\n", + TAP_LOG(ERR, + "Failed to update BPF map entry #%u (%d): %s", flow->key_idx, errno, strerror(errno)); rte_flow_error_set( error, ENOTSUP, RTE_FLOW_ERROR_TYPE_HANDLE, NULL, @@ -2085,8 +2122,8 @@ static int rss_add_actions(struct rte_flow *flow, struct pmd_internals *pmd, flow->bpf_fd[SEC_L3_L4] = tap_flow_bpf_calc_l3_l4_hash(flow->key_idx, pmd->map_fd); if (flow->bpf_fd[SEC_L3_L4] < 0) { - RTE_LOG(ERR, PMD, - "Failed to load BPF section %s (%d): %s\n", + TAP_LOG(ERR, + "Failed to load BPF section %s (%d): %s", sec_name[SEC_L3_L4], errno, strerror(errno)); rte_flow_error_set( error, ENOTSUP, RTE_FLOW_ERROR_TYPE_HANDLE, NULL, @@ -2147,9 +2184,8 @@ tap_dev_filter_ctrl(struct rte_eth_dev *dev, *(const void **)arg = &tap_flow_ops; return 0; default: - RTE_LOG(ERR, PMD, "%p: filter type (%d) not supported\n", - (void *)dev, filter_type); + TAP_LOG(ERR, "%p: filter type (%d) not supported", + dev, filter_type); } return -EINVAL; } - diff --git a/drivers/net/tap/tap_flow.h b/drivers/net/tap/tap_flow.h index ac6a952d..ac60a9ae 100644 --- a/drivers/net/tap/tap_flow.h +++ b/drivers/net/tap/tap_flow.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #ifndef _TAP_FLOW_H_ diff --git a/drivers/net/tap/tap_intr.c b/drivers/net/tap/tap_intr.c index b0e19914..fc590181 100644 --- a/drivers/net/tap/tap_intr.c +++ b/drivers/net/tap/tap_intr.c @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: BSD-3-Clause - * Copyright 2018 Mellanox Technologies, Ltd. + * Copyright 2018 Mellanox Technologies, Ltd */ /** @@ -62,7 +62,7 @@ tap_rx_intr_vec_install(struct rte_eth_dev *dev) intr_handle->intr_vec = malloc(sizeof(intr_handle->intr_vec[rxqs_n])); if (intr_handle->intr_vec == NULL) { rte_errno = ENOMEM; - RTE_LOG(ERR, PMD, + TAP_LOG(ERR, "failed to allocate memory for interrupt vector," " Rx interrupts will not be supported"); return -rte_errno; diff --git a/drivers/net/tap/tap_log.h b/drivers/net/tap/tap_log.h new file mode 100644 index 00000000..fa06843a --- /dev/null +++ b/drivers/net/tap/tap_log.h @@ -0,0 +1,10 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright 2017 6WIND S.A. + * Copyright 2017 Mellanox Technologies, Ltd + */ + +extern int tap_logtype; + +#define TAP_LOG(level, fmt, args...) \ + rte_log(RTE_LOG_ ## level, tap_logtype, "%s(): " fmt "\n", \ + __func__, ## args) diff --git a/drivers/net/tap/tap_netlink.c b/drivers/net/tap/tap_netlink.c index 82c8dc0e..6cb51009 100644 --- a/drivers/net/tap/tap_netlink.c +++ b/drivers/net/tap/tap_netlink.c @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #include <errno.h> @@ -13,6 +13,7 @@ #include <rte_malloc.h> #include <tap_netlink.h> #include <rte_random.h> +#include "tap_log.h" /* Must be quite large to support dumping a huge list of QDISC or filters. */ #define BUF_SIZE (32 * 1024) /* Size of the buffer to receive kernel messages */ @@ -45,19 +46,19 @@ tap_nl_init(uint32_t nl_groups) fd = socket(AF_NETLINK, SOCK_RAW | SOCK_CLOEXEC, NETLINK_ROUTE); if (fd < 0) { - RTE_LOG(ERR, PMD, "Unable to create a netlink socket\n"); + TAP_LOG(ERR, "Unable to create a netlink socket"); return -1; } if (setsockopt(fd, SOL_SOCKET, SO_SNDBUF, &sndbuf_size, sizeof(int))) { - RTE_LOG(ERR, PMD, "Unable to set socket buffer send size\n"); + TAP_LOG(ERR, "Unable to set socket buffer send size"); return -1; } if (setsockopt(fd, SOL_SOCKET, SO_RCVBUF, &rcvbuf_size, sizeof(int))) { - RTE_LOG(ERR, PMD, "Unable to set socket buffer receive size\n"); + TAP_LOG(ERR, "Unable to set socket buffer receive size"); return -1; } if (bind(fd, (struct sockaddr *)&local, sizeof(local)) < 0) { - RTE_LOG(ERR, PMD, "Unable to bind to the netlink socket\n"); + TAP_LOG(ERR, "Unable to bind to the netlink socket"); return -1; } return fd; @@ -76,7 +77,7 @@ int tap_nl_final(int nlsk_fd) { if (close(nlsk_fd)) { - RTE_LOG(ERR, PMD, "Failed to close netlink socket: %s (%d)\n", + TAP_LOG(ERR, "Failed to close netlink socket: %s (%d)", strerror(errno), errno); return -1; } @@ -117,7 +118,7 @@ tap_nl_send(int nlsk_fd, struct nlmsghdr *nh) nh->nlmsg_seq = (uint32_t)rte_rand(); send_bytes = sendmsg(nlsk_fd, &msg, 0); if (send_bytes < 0) { - RTE_LOG(ERR, PMD, "Failed to send netlink message: %s (%d)\n", + TAP_LOG(ERR, "Failed to send netlink message: %s (%d)", strerror(errno), errno); return -1; } @@ -300,9 +301,8 @@ tap_nlattr_nested_start(struct nlmsg *msg, uint16_t type) tail = rte_zmalloc(NULL, sizeof(struct nested_tail), 0); if (!tail) { - RTE_LOG(ERR, PMD, - "Couldn't allocate memory for nested netlink" - " attribute\n"); + TAP_LOG(ERR, + "Couldn't allocate memory for nested netlink attribute"); return -1; } diff --git a/drivers/net/tap/tap_netlink.h b/drivers/net/tap/tap_netlink.h index fafef840..faa73ba1 100644 --- a/drivers/net/tap/tap_netlink.h +++ b/drivers/net/tap/tap_netlink.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #ifndef _TAP_NETLINK_H_ diff --git a/drivers/net/tap/tap_rss.h b/drivers/net/tap/tap_rss.h index 3bb0d140..17606b2d 100644 --- a/drivers/net/tap/tap_rss.h +++ b/drivers/net/tap/tap_rss.h @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: BSD-3-Clause - * Copyright 2017 Mellanox Technologies, Ltd. + * Copyright 2017 Mellanox Technologies, Ltd */ #ifndef _TAP_RSS_H_ @@ -9,6 +9,12 @@ #define TAP_MAX_QUEUES 16 #endif +/* Fixed RSS hash key size in bytes. */ +#define TAP_RSS_HASH_KEY_SIZE 40 + +/* Supported RSS */ +#define TAP_RSS_HF_MASK (~(ETH_RSS_IP | ETH_RSS_UDP | ETH_RSS_TCP)) + /* hashed fields for RSS */ enum hash_field { HASH_FIELD_IPV4_L3, /* IPv4 src/dst addr */ diff --git a/drivers/net/tap/tap_tcmsgs.c b/drivers/net/tap/tap_tcmsgs.c index 954f13eb..3c9d0366 100644 --- a/drivers/net/tap/tap_tcmsgs.c +++ b/drivers/net/tap/tap_tcmsgs.c @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #include <inttypes.h> @@ -10,6 +10,7 @@ #include <rte_log.h> #include <tap_tcmsgs.h> +#include "tap_log.h" struct qdisc { uint32_t handle; @@ -81,8 +82,8 @@ qdisc_del(int nlsk_fd, uint16_t ifindex, struct qdisc *qinfo) if (!nlsk_fd) { fd = tap_nl_init(0); if (fd < 0) { - RTE_LOG(ERR, PMD, - "Could not delete QDISC: null netlink socket\n"); + TAP_LOG(ERR, + "Could not delete QDISC: null netlink socket"); return -1; } } else { @@ -261,7 +262,7 @@ qdisc_create_multiq(int nlsk_fd, uint16_t ifindex) err = qdisc_add_multiq(nlsk_fd, ifindex); if (err < 0 && errno != -EEXIST) { - RTE_LOG(ERR, PMD, "Could not add multiq qdisc (%d): %s\n", + TAP_LOG(ERR, "Could not add multiq qdisc (%d): %s", errno, strerror(errno)); return -1; } @@ -287,7 +288,7 @@ qdisc_create_ingress(int nlsk_fd, uint16_t ifindex) err = qdisc_add_ingress(nlsk_fd, ifindex); if (err < 0 && errno != -EEXIST) { - RTE_LOG(ERR, PMD, "Could not add ingress qdisc (%d): %s\n", + TAP_LOG(ERR, "Could not add ingress qdisc (%d): %s", errno, strerror(errno)); return -1; } diff --git a/drivers/net/tap/tap_tcmsgs.h b/drivers/net/tap/tap_tcmsgs.h index f72f8c5c..8cedea84 100644 --- a/drivers/net/tap/tap_tcmsgs.h +++ b/drivers/net/tap/tap_tcmsgs.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: BSD-3-Clause * Copyright 2017 6WIND S.A. - * Copyright 2017 Mellanox. + * Copyright 2017 Mellanox Technologies, Ltd */ #ifndef _TAP_TCMSGS_H_ |