X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Ftap%2Frte_eth_tap.c;h=cca5852cc3ac90a9b50a1db16fb70c72bf2c9292;hb=9e5afc72c9094192400c219660a314f0778d9c50;hp=d926d4b06c4b50c32413768f52977539ddb0575c;hpb=f46900d03823aff1cb636a2f889caeee99c721b2;p=dpdk.git diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c index d926d4b06c..cca5852cc3 100644 --- a/drivers/net/tap/rte_eth_tap.c +++ b/drivers/net/tap/rte_eth_tap.c @@ -1,34 +1,5 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2016-2017 Intel Corporation. All rights reserved. - * All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * - * * Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * * Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in - * the documentation and/or other materials provided with the - * distribution. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2016-2017 Intel Corporation */ #include @@ -36,14 +7,15 @@ #include #include #include -#include +#include #include #include -#include +#include #include #include #include #include +#include #include #include @@ -53,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -60,7 +33,6 @@ #include #include #include -#include #include #include @@ -71,27 +43,32 @@ /* 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_SPEED_ARG "speed" #define ETH_TAP_REMOTE_ARG "remote" #define ETH_TAP_MAC_ARG "mac" #define ETH_TAP_MAC_FIXED "fixed" -#define FLOWER_KERNEL_VERSION KERNEL_VERSION(4, 2, 0) -#define FLOWER_VLAN_KERNEL_VERSION KERNEL_VERSION(4, 9, 0) +#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 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, - ETH_TAP_SPEED_ARG, ETH_TAP_REMOTE_ARG, ETH_TAP_MAC_ARG, NULL }; static int tap_unit; +static int tun_unit; + +static int tap_type; +static char tuntap_name[8]; static volatile uint32_t tap_trigger; /* Rx trigger */ @@ -99,7 +76,7 @@ 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_SPEED_AUTONEG + .link_autoneg = ETH_LINK_AUTONEG }; static void @@ -138,24 +115,26 @@ 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 = (tap_type) ? 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); fd = open(TUN_TAP_DEV_PATH, O_RDWR); if (fd < 0) { - RTE_LOG(ERR, PMD, "Unable to create TAP interface\n"); + RTE_LOG(ERR, PMD, "Unable to create %s interface\n", + 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"); + RTE_LOG(ERR, PMD, "%s unable to get TUN/TAP features\n", + tuntap_name); goto error; } - RTE_LOG(DEBUG, PMD, " TAP Features %08x\n", features); + RTE_LOG(DEBUG, PMD, "%s Features %08x\n", tuntap_name, features); if (features & IFF_MULTI_QUEUE) { RTE_LOG(DEBUG, PMD, " Multi-queue support for %d queues\n", @@ -285,6 +264,45 @@ tap_verify_csum(struct rte_mbuf *mbuf) } } +static uint64_t +tap_rx_offload_get_port_capa(void) +{ + /* + * In order to support legacy apps, + * report capabilities also as port 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; +} + +static uint64_t +tap_rx_offload_get_queue_capa(void) +{ + 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; +} + +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. */ @@ -309,8 +327,9 @@ pmd_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) int len; len = readv(rxq->fd, *rxq->iovecs, - 1 + (rxq->rxmode->enable_scatter ? - rxq->nb_rx_desc : 1)); + 1 + + (rxq->rxmode->offloads & DEV_RX_OFFLOAD_SCATTER ? + rxq->nb_rx_desc : 1)); if (len < (int)sizeof(struct tun_pi)) break; @@ -365,7 +384,7 @@ pmd_rx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) seg->next = NULL; mbuf->packet_type = rte_net_get_ptype(mbuf, NULL, RTE_PTYPE_ALL_MASK); - if (rxq->rxmode->hw_ip_checksum) + if (rxq->rxmode->offloads & DEV_RX_OFFLOAD_CHECKSUM) tap_verify_csum(mbuf); /* account for the receive frame */ @@ -379,6 +398,44 @@ end: return num_rx; } +static uint64_t +tap_tx_offload_get_port_capa(void) +{ + /* + * In order to support legacy apps, + * report capabilities also as port capabilities. + */ + return DEV_TX_OFFLOAD_MULTI_SEGS | + DEV_TX_OFFLOAD_IPV4_CKSUM | + DEV_TX_OFFLOAD_UDP_CKSUM | + DEV_TX_OFFLOAD_TCP_CKSUM; +} + +static uint64_t +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; +} + +static bool +tap_txq_are_offloads_valid(struct rte_eth_dev *dev, uint64_t offloads) +{ + 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; +} + static void tap_tx_offload(char *packet, uint64_t ol_flags, unsigned int l2_len, unsigned int l3_len) @@ -447,7 +504,7 @@ pmd_tx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) 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 tun_pi pi = { .flags = 0, .proto = 0x00 }; struct rte_mbuf *seg = mbuf; char m_copy[mbuf->data_len]; int n; @@ -457,6 +514,21 @@ pmd_tx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) if (rte_pktmbuf_pkt_len(mbuf) > max_size) break; + /* + * 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. + * compares whether its v4 or v6. If none matches default + * value 0x00 is taken for protocol field. + */ + char *buff_data = rte_pktmbuf_mtod(seg, void *); + j = (*buff_data & 0xf0); + if (j & (0x40 | 0x60)) + pi.proto = (j == 0x40) ? 0x0008 : 0xdd86; + iovecs[0].iov_base = π iovecs[0].iov_len = sizeof(pi); for (j = 1; j <= mbuf->nb_segs; j++) { @@ -465,9 +537,10 @@ pmd_tx_burst(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) rte_pktmbuf_mtod(seg, void *); seg = seg->next; } - if (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) { + 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) break; @@ -605,6 +678,17 @@ 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", @@ -676,18 +760,16 @@ 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_offload_capa = (DEV_RX_OFFLOAD_IPV4_CKSUM | - DEV_RX_OFFLOAD_UDP_CKSUM | - DEV_RX_OFFLOAD_TCP_CKSUM); - dev_info->tx_offload_capa = - (DEV_TX_OFFLOAD_IPV4_CKSUM | - DEV_TX_OFFLOAD_UDP_CKSUM | - DEV_TX_OFFLOAD_TCP_CKSUM); + dev_info->rx_queue_offload_capa = tap_rx_offload_get_queue_capa(); + dev_info->rx_offload_capa = tap_rx_offload_get_port_capa() | + dev_info->rx_queue_offload_capa; + 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; } -static void +static int tap_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *tap_stats) { unsigned int i, imax; @@ -728,6 +810,7 @@ tap_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *tap_stats) tap_stats->opackets = tx_total; tap_stats->oerrors = tx_err_total; tap_stats->obytes = tx_bytes_total; + return 0; } static void @@ -873,48 +956,58 @@ 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 (is_zero_ether_addr(mac_addr)) { RTE_LOG(ERR, PMD, "%s: can't set an empty MAC address\n", 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) { + ret = tap_flow_implicit_destroy(pmd, TAP_REMOTE_LOCAL_MAC); + if (ret < 0) { RTE_LOG(ERR, PMD, "%s: Couldn't delete MAC redirection rule\n", dev->device->name); - return; + return ret; } - if (tap_flow_implicit_create(pmd, TAP_REMOTE_LOCAL_MAC) < 0) + ret = tap_flow_implicit_create(pmd, TAP_REMOTE_LOCAL_MAC); + if (ret < 0) { RTE_LOG(ERR, PMD, "%s: Couldn't add MAC redirection rule\n", dev->device->name); + return ret; + } } + + return 0; } static int @@ -999,6 +1092,19 @@ tap_rx_queue_setup(struct rte_eth_dev *dev, 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; @@ -1039,7 +1145,7 @@ 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", + RTE_LOG(DEBUG, PMD, " RX TUNTAP device name %s, qid %d on fd %d\n", internals->name, rx_queue_id, internals->rxq[rx_queue_id].fd); return 0; @@ -1057,21 +1163,46 @@ tap_tx_queue_setup(struct rte_eth_dev *dev, uint16_t tx_queue_id, uint16_t nb_tx_desc __rte_unused, unsigned int socket_id __rte_unused, - const struct rte_eth_txconf *tx_conf __rte_unused) + const struct rte_eth_txconf *tx_conf) { struct pmd_internals *internals = dev->data->dev_private; + struct tx_queue *txq; int ret; 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; + } + } 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\n", - internals->name, tx_queue_id, internals->txq[tx_queue_id].fd); + RTE_LOG(DEBUG, PMD, + " TX TUNTAP device name %s, qid %d on fd %d csum %s\n", + internals->name, tx_queue_id, internals->txq[tx_queue_id].fd, + txq->csum ? "on" : "off"); return 0; } @@ -1122,34 +1253,49 @@ tap_dev_intr_handler(void *cb_arg) struct rte_eth_dev *dev = cb_arg; struct pmd_internals *pmd = dev->data->dev_private; - nl_recv(pmd->intr_handle.fd, tap_nl_msg_handler, dev); + tap_nl_recv(pmd->intr_handle.fd, tap_nl_msg_handler, dev); } static int -tap_intr_handle_set(struct rte_eth_dev *dev, int set) +tap_lsc_intr_handle_set(struct rte_eth_dev *dev, int set) { struct pmd_internals *pmd = dev->data->dev_private; /* In any case, disable interrupt if the conf is no longer there. */ if (!dev->data->dev_conf.intr_conf.lsc) { - if (pmd->intr_handle.fd != -1) - nl_final(pmd->intr_handle.fd); - rte_intr_callback_unregister( - &pmd->intr_handle, tap_dev_intr_handler, dev); + if (pmd->intr_handle.fd != -1) { + tap_nl_final(pmd->intr_handle.fd); + rte_intr_callback_unregister(&pmd->intr_handle, + tap_dev_intr_handler, dev); + } return 0; } if (set) { - pmd->intr_handle.fd = nl_init(RTMGRP_LINK); + pmd->intr_handle.fd = tap_nl_init(RTMGRP_LINK); if (unlikely(pmd->intr_handle.fd == -1)) return -EBADF; return rte_intr_callback_register( &pmd->intr_handle, tap_dev_intr_handler, dev); } - nl_final(pmd->intr_handle.fd); + tap_nl_final(pmd->intr_handle.fd); return rte_intr_callback_unregister(&pmd->intr_handle, tap_dev_intr_handler, dev); } +static int +tap_intr_handle_set(struct rte_eth_dev *dev, int set) +{ + int err; + + err = tap_lsc_intr_handle_set(dev, set); + if (err) + return err; + err = tap_rx_intr_vec_set(dev, set); + if (err && set) + tap_lsc_intr_handle_set(dev, 0); + return err; +} + static const uint32_t* tap_dev_supported_ptypes_get(struct rte_eth_dev *dev __rte_unused) { @@ -1228,7 +1374,7 @@ static const struct eth_dev_ops ops = { 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) { int numa_node = rte_socket_id(); struct rte_eth_dev *dev; @@ -1237,18 +1383,14 @@ 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; - } + RTE_LOG(DEBUG, PMD, "%s device on numa %u\n", + 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"); - goto error_exit; + RTE_LOG(ERR, PMD, "%s Unable to allocate device struct\n", + tuntap_name); + goto error_exit_nodev; } pmd = dev->data->dev_private; @@ -1258,15 +1400,15 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, 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)); + "%s Unable to get a socket for management: %s\n", + 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_DETACHABLE | RTE_ETH_DEV_INTR_LSC; + data->dev_flags = RTE_ETH_DEV_INTR_LSC; data->numa_node = numa_node; data->dev_link = pmd_link; @@ -1275,13 +1417,13 @@ 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; pmd->intr_handle.type = RTE_INTR_HANDLE_EXT; pmd->intr_handle.fd = -1; + dev->intr_handle = &pmd->intr_handle; /* Presetup the fds to -1 as being not valid */ for (i = 0; i < RTE_PMD_TAP_MAX_QUEUES; i++) { @@ -1289,15 +1431,11 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, pmd->txq[i].fd = -1; } - if (fixed_mac_type) { - /* fixed mac = 00:64:74:61:70: */ - 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 (tap_type) { + 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). */ @@ -1312,11 +1450,14 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, 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 (tap_type) { + 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: @@ -1326,7 +1467,7 @@ eth_dev_tap_create(struct rte_vdev_device *vdev, char *tap_name, * - rte_flow actual/implicit lists * - implicit rules */ - pmd->nlsk_fd = nl_init(0); + pmd->nlsk_fd = tap_nl_init(0); if (pmd->nlsk_fd == -1) { RTE_LOG(WARNING, PMD, "%s: failed to create netlink socket.\n", pmd->name); @@ -1418,10 +1559,14 @@ error_remote: tap_flow_implicit_flush(pmd, NULL); error_exit: - RTE_LOG(ERR, PMD, "TAP Unable to initialize %s\n", - rte_vdev_device_name(vdev)); + if (pmd->ioctl_sock > 0) + close(pmd->ioctl_sock); + rte_eth_dev_release_port(dev); + +error_exit_nodev: + RTE_LOG(ERR, PMD, "%s Unable to initialize %s\n", + tuntap_name, rte_vdev_device_name(vdev)); - rte_free(data); return -EINVAL; } @@ -1433,7 +1578,7 @@ 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)); @@ -1441,16 +1586,6 @@ set_interface_name(const char *key __rte_unused, return 0; } -static int -set_interface_speed(const char *key __rte_unused, - const char *value, - void *extra_args) -{ - *(int *)extra_args = (value) ? atoi(value) : ETH_SPEED_NUM_10G; - - return 0; -} - static int set_remote_iface(const char *key __rte_unused, const char *value, @@ -1459,20 +1594,119 @@ 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: */ + 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: + RTE_LOG(DEBUG, PMD, "TAP user MAC param (%s)\n", value); return 0; + +error: + RTE_LOG(ERR, PMD, "TAP user MAC (%s) is not in format (%s|%s)\n", + 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]; + + tap_type = 0; + 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 (params && (params[0] != '\0')) { + RTE_LOG(DEBUG, PMD, "parameters (%s)\n", 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; + + RTE_LOG(NOTICE, PMD, "Initializing pmd_tun for %s as %s\n", + name, tun_name); + + ret = eth_dev_tap_create(dev, tun_name, remote_iface, 0); + +leave: + if (ret == -1) { + RTE_LOG(ERR, PMD, "Failed to create pmd for %s as %s\n", + name, tun_name); + tun_unit--; /* Restore the unit number */ + } + rte_kvargs_free(kvlist); + + return ret; } /* Open a TAP interface device. @@ -1486,11 +1720,27 @@ 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; + + tap_type = 1; + 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) { + RTE_LOG(ERR, PMD, "Failed to probe %s\n", name); + return -1; + } + /* TODO: request info from primary to set up Rx and Tx */ + eth_dev->dev_ops = &ops; + return 0; + } + speed = ETH_SPEED_NUM_10G; snprintf(tap_name, sizeof(tap_name), "%s%d", DEFAULT_TAP_NAME, tap_unit++); @@ -1501,15 +1751,6 @@ rte_pmd_tap_probe(struct rte_vdev_device *dev) kvlist = rte_kvargs_parse(params, valid_arguments); if (kvlist) { - if (rte_kvargs_count(kvlist, ETH_TAP_SPEED_ARG) == 1) { - ret = rte_kvargs_process(kvlist, - ETH_TAP_SPEED_ARG, - &set_interface_speed, - &speed); - if (ret == -1) - goto leave; - } - if (rte_kvargs_count(kvlist, ETH_TAP_IFACE_ARG) == 1) { ret = rte_kvargs_process(kvlist, ETH_TAP_IFACE_ARG, @@ -1532,7 +1773,7 @@ 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; } @@ -1543,7 +1784,7 @@ rte_pmd_tap_probe(struct rte_vdev_device *dev) RTE_LOG(NOTICE, PMD, "Initializing pmd_tap for %s as %s\n", 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); leave: if (ret == -1) { @@ -1556,7 +1797,7 @@ leave: return ret; } -/* detach a TAP device. +/* detach a TUNTAP device. */ static int rte_pmd_tap_remove(struct rte_vdev_device *dev) @@ -1577,7 +1818,7 @@ rte_pmd_tap_remove(struct rte_vdev_device *dev) if (internals->nlsk_fd) { tap_flow_flush(eth_dev, NULL); tap_flow_implicit_flush(internals, NULL); - nl_final(internals->nlsk_fd); + tap_nl_final(internals->nlsk_fd); } for (i = 0; i < RTE_PMD_TAP_MAX_QUEUES; i++) { if (internals->rxq[i].fd != -1) { @@ -1592,21 +1833,27 @@ 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); 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 "= "); RTE_PMD_REGISTER_PARAM_STRING(net_tap, ETH_TAP_IFACE_ARG "= " - ETH_TAP_SPEED_ARG "= " - ETH_TAP_MAC_ARG "=" ETH_TAP_MAC_FIXED " " + ETH_TAP_MAC_ARG "=" ETH_TAP_MAC_ARG_FMT " " ETH_TAP_REMOTE_ARG "=");