diff options
Diffstat (limited to 'drivers/net/mlx5/mlx5.c')
| -rw-r--r-- | drivers/net/mlx5/mlx5.c | 525 |
1 files changed, 340 insertions, 185 deletions
diff --git a/drivers/net/mlx5/mlx5.c b/drivers/net/mlx5/mlx5.c index 45e0e8db..10ce3359 100644 --- a/drivers/net/mlx5/mlx5.c +++ b/drivers/net/mlx5/mlx5.c @@ -39,6 +39,7 @@ #include <stdlib.h> #include <errno.h> #include <net/if.h> +#include <sys/mman.h> /* Verbs header. */ /* ISO C doesn't support unnamed structs/unions, disabling -pedantic. */ @@ -56,6 +57,7 @@ #include <rte_pci.h> #include <rte_bus_pci.h> #include <rte_common.h> +#include <rte_eal_memconfig.h> #include <rte_kvargs.h> #include "mlx5.h" @@ -117,6 +119,10 @@ struct mlx5_args { int tx_vec_en; int rx_vec_en; }; + +/** Driver-specific log messages type. */ +int mlx5_logtype; + /** * Retrieve integer value from environment variable. * @@ -148,7 +154,7 @@ mlx5_getenv_int(const char *name) * A pointer to the callback data. * * @return - * a pointer to the allocate space. + * Allocated buffer, NULL otherwise and rte_errno is set. */ static void * mlx5_alloc_verbs_buf(size_t size, void *data) @@ -156,11 +162,22 @@ mlx5_alloc_verbs_buf(size_t size, void *data) struct priv *priv = data; void *ret; size_t alignment = sysconf(_SC_PAGESIZE); + unsigned int socket = SOCKET_ID_ANY; + + if (priv->verbs_alloc_ctx.type == MLX5_VERBS_ALLOC_TYPE_TX_QUEUE) { + const struct mlx5_txq_ctrl *ctrl = priv->verbs_alloc_ctx.obj; + + socket = ctrl->socket; + } else if (priv->verbs_alloc_ctx.type == + MLX5_VERBS_ALLOC_TYPE_RX_QUEUE) { + const struct mlx5_rxq_ctrl *ctrl = priv->verbs_alloc_ctx.obj; + socket = ctrl->socket; + } assert(data != NULL); - ret = rte_malloc_socket(__func__, size, alignment, - priv->dev->device->numa_node); - DEBUG("Extern alloc size: %lu, align: %lu: %p", size, alignment, ret); + ret = rte_malloc_socket(__func__, size, alignment, socket); + if (!ret && size) + rte_errno = ENOMEM; return ret; } @@ -176,7 +193,6 @@ static void mlx5_free_verbs_buf(void *ptr, void *data __rte_unused) { assert(data != NULL); - DEBUG("Extern free request: %p", ptr); rte_free(ptr); } @@ -191,17 +207,16 @@ mlx5_free_verbs_buf(void *ptr, void *data __rte_unused) static void mlx5_dev_close(struct rte_eth_dev *dev) { - struct priv *priv = mlx5_get_priv(dev); + struct priv *priv = dev->data->dev_private; unsigned int i; int ret; - priv_lock(priv); - DEBUG("%p: closing device \"%s\"", - (void *)dev, - ((priv->ctx != NULL) ? priv->ctx->device->name : "")); + DRV_LOG(DEBUG, "port %u closing device \"%s\"", + dev->data->port_id, + ((priv->ctx != NULL) ? priv->ctx->device->name : "")); /* In case mlx5_dev_stop() has not been called. */ - priv_dev_interrupt_handler_uninstall(priv, dev); - priv_dev_traffic_disable(priv, dev); + mlx5_dev_interrupt_handler_uninstall(dev); + mlx5_traffic_disable(dev); /* Prevent crashes when queues are still in use. */ dev->rx_pkt_burst = removed_rx_burst; dev->tx_pkt_burst = removed_tx_burst; @@ -209,7 +224,7 @@ mlx5_dev_close(struct rte_eth_dev *dev) /* XXX race condition if mlx5_rx_burst() is still running. */ usleep(1000); for (i = 0; (i != priv->rxqs_n); ++i) - mlx5_priv_rxq_release(priv, i); + mlx5_rxq_release(dev, i); priv->rxqs_n = 0; priv->rxqs = NULL; } @@ -217,7 +232,7 @@ mlx5_dev_close(struct rte_eth_dev *dev) /* XXX race condition if mlx5_tx_burst() is still running. */ usleep(1000); for (i = 0; (i != priv->txqs_n); ++i) - mlx5_priv_txq_release(priv, i); + mlx5_txq_release(dev, i); priv->txqs_n = 0; priv->txqs = NULL; } @@ -231,32 +246,40 @@ mlx5_dev_close(struct rte_eth_dev *dev) rte_free(priv->rss_conf.rss_key); if (priv->reta_idx != NULL) rte_free(priv->reta_idx); - priv_socket_uninit(priv); - ret = mlx5_priv_hrxq_ibv_verify(priv); + if (priv->primary_socket) + mlx5_socket_uninit(dev); + ret = mlx5_hrxq_ibv_verify(dev); if (ret) - WARN("%p: some Hash Rx queue still remain", (void *)priv); - ret = mlx5_priv_ind_table_ibv_verify(priv); + DRV_LOG(WARNING, "port %u some hash Rx queue still remain", + dev->data->port_id); + ret = mlx5_ind_table_ibv_verify(dev); if (ret) - WARN("%p: some Indirection table still remain", (void *)priv); - ret = mlx5_priv_rxq_ibv_verify(priv); + DRV_LOG(WARNING, "port %u some indirection table still remain", + dev->data->port_id); + ret = mlx5_rxq_ibv_verify(dev); if (ret) - WARN("%p: some Verbs Rx queue still remain", (void *)priv); - ret = mlx5_priv_rxq_verify(priv); + DRV_LOG(WARNING, "port %u some Verbs Rx queue still remain", + dev->data->port_id); + ret = mlx5_rxq_verify(dev); if (ret) - WARN("%p: some Rx Queues still remain", (void *)priv); - ret = mlx5_priv_txq_ibv_verify(priv); + DRV_LOG(WARNING, "port %u some Rx queues still remain", + dev->data->port_id); + ret = mlx5_txq_ibv_verify(dev); if (ret) - WARN("%p: some Verbs Tx queue still remain", (void *)priv); - ret = mlx5_priv_txq_verify(priv); + DRV_LOG(WARNING, "port %u some Verbs Tx queue still remain", + dev->data->port_id); + ret = mlx5_txq_verify(dev); if (ret) - WARN("%p: some Tx Queues still remain", (void *)priv); - ret = priv_flow_verify(priv); + DRV_LOG(WARNING, "port %u some Tx queues still remain", + dev->data->port_id); + ret = mlx5_flow_verify(dev); if (ret) - WARN("%p: some flows still remain", (void *)priv); - ret = priv_mr_verify(priv); + DRV_LOG(WARNING, "port %u some flows still remain", + dev->data->port_id); + ret = mlx5_mr_verify(dev); if (ret) - WARN("%p: some Memory Region still remain", (void *)priv); - priv_unlock(priv); + DRV_LOG(WARNING, "port %u some memory region still remain", + dev->data->port_id); memset(priv, 0, sizeof(*priv)); } @@ -394,7 +417,7 @@ mlx5_dev_idx(struct rte_pci_addr *pci_addr) * User data. * * @return - * 0 on success, negative errno value on failure. + * 0 on success, a negative errno value otherwise and rte_errno is set. */ static int mlx5_args_check(const char *key, const char *val, void *opaque) @@ -405,8 +428,9 @@ mlx5_args_check(const char *key, const char *val, void *opaque) errno = 0; tmp = strtoul(val, NULL, 0); if (errno) { - WARN("%s: \"%s\" is not a valid integer", key, val); - return errno; + rte_errno = errno; + DRV_LOG(WARNING, "%s: \"%s\" is not a valid integer", key, val); + return -rte_errno; } if (strcmp(MLX5_RXQ_CQE_COMP_EN, key) == 0) { args->cqe_comp = !!tmp; @@ -427,8 +451,9 @@ mlx5_args_check(const char *key, const char *val, void *opaque) } else if (strcmp(MLX5_RX_VEC_EN, key) == 0) { args->rx_vec_en = !!tmp; } else { - WARN("%s: unknown parameter", key); - return -EINVAL; + DRV_LOG(WARNING, "%s: unknown parameter", key); + rte_errno = EINVAL; + return -rte_errno; } return 0; } @@ -442,7 +467,7 @@ mlx5_args_check(const char *key, const char *val, void *opaque) * Device arguments structure. * * @return - * 0 on success, errno value on failure. + * 0 on success, a negative errno value otherwise and rte_errno is set. */ static int mlx5_args(struct mlx5_args *args, struct rte_devargs *devargs) @@ -474,9 +499,10 @@ mlx5_args(struct mlx5_args *args, struct rte_devargs *devargs) if (rte_kvargs_count(kvlist, params[i])) { ret = rte_kvargs_process(kvlist, params[i], mlx5_args_check, args); - if (ret != 0) { + if (ret) { + rte_errno = EINVAL; rte_kvargs_free(kvlist); - return ret; + return -rte_errno; } } } @@ -486,6 +512,112 @@ mlx5_args(struct mlx5_args *args, struct rte_devargs *devargs) static struct rte_pci_driver mlx5_driver; +/* + * Reserved UAR address space for TXQ UAR(hw doorbell) mapping, process + * local resource used by both primary and secondary to avoid duplicate + * reservation. + * The space has to be available on both primary and secondary process, + * TXQ UAR maps to this area using fixed mmap w/o double check. + */ +static void *uar_base; + +/** + * Reserve UAR address space for primary process. + * + * @param[in] dev + * Pointer to Ethernet device. + * + * @return + * 0 on success, a negative errno value otherwise and rte_errno is set. + */ +static int +mlx5_uar_init_primary(struct rte_eth_dev *dev) +{ + struct priv *priv = dev->data->dev_private; + void *addr = (void *)0; + int i; + const struct rte_mem_config *mcfg; + + if (uar_base) { /* UAR address space mapped. */ + priv->uar_base = uar_base; + return 0; + } + /* find out lower bound of hugepage segments */ + mcfg = rte_eal_get_configuration()->mem_config; + for (i = 0; i < RTE_MAX_MEMSEG && mcfg->memseg[i].addr; i++) { + if (addr) + addr = RTE_MIN(addr, mcfg->memseg[i].addr); + else + addr = mcfg->memseg[i].addr; + } + /* keep distance to hugepages to minimize potential conflicts. */ + addr = RTE_PTR_SUB(addr, MLX5_UAR_OFFSET + MLX5_UAR_SIZE); + /* anonymous mmap, no real memory consumption. */ + addr = mmap(addr, MLX5_UAR_SIZE, + PROT_NONE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (addr == MAP_FAILED) { + DRV_LOG(ERR, + "port %u failed to reserve UAR address space, please" + " adjust MLX5_UAR_SIZE or try --base-virtaddr", + dev->data->port_id); + rte_errno = ENOMEM; + return -rte_errno; + } + /* Accept either same addr or a new addr returned from mmap if target + * range occupied. + */ + DRV_LOG(INFO, "port %u reserved UAR address space: %p", + dev->data->port_id, addr); + priv->uar_base = addr; /* for primary and secondary UAR re-mmap. */ + uar_base = addr; /* process local, don't reserve again. */ + return 0; +} + +/** + * Reserve UAR address space for secondary process, align with + * primary process. + * + * @param[in] dev + * Pointer to Ethernet device. + * + * @return + * 0 on success, a negative errno value otherwise and rte_errno is set. + */ +static int +mlx5_uar_init_secondary(struct rte_eth_dev *dev) +{ + struct priv *priv = dev->data->dev_private; + void *addr; + + assert(priv->uar_base); + if (uar_base) { /* already reserved. */ + assert(uar_base == priv->uar_base); + return 0; + } + /* anonymous mmap, no real memory consumption. */ + addr = mmap(priv->uar_base, MLX5_UAR_SIZE, + PROT_NONE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if (addr == MAP_FAILED) { + DRV_LOG(ERR, "port %u UAR mmap failed: %p size: %llu", + dev->data->port_id, priv->uar_base, MLX5_UAR_SIZE); + rte_errno = ENXIO; + return -rte_errno; + } + if (priv->uar_base != addr) { + DRV_LOG(ERR, + "port %u UAR address %p size %llu occupied, please" + " adjust MLX5_UAR_OFFSET or try EAL parameter" + " --base-virtaddr", + dev->data->port_id, priv->uar_base, MLX5_UAR_SIZE); + rte_errno = ENXIO; + return -rte_errno; + } + uar_base = addr; /* process local, don't reserve again */ + DRV_LOG(INFO, "port %u reserved UAR address space: %p", + dev->data->port_id, addr); + return 0; +} + /** * Assign parameters from args into priv, only non default * values are considered. @@ -530,17 +662,17 @@ mlx5_args_assign(struct priv *priv, struct mlx5_args *args) * PCI device information. * * @return - * 0 on success, negative errno value on failure. + * 0 on success, a negative errno value otherwise and rte_errno is set. */ static int -mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) +mlx5_pci_probe(struct rte_pci_driver *pci_drv __rte_unused, + struct rte_pci_device *pci_dev) { - struct ibv_device **list; + struct ibv_device **list = NULL; struct ibv_device *ibv_dev; int err = 0; struct ibv_context *attr_ctx = NULL; struct ibv_device_attr_ex device_attr; - unsigned int sriov; unsigned int mps; unsigned int cqe_comp; unsigned int tunnel_en = 0; @@ -551,24 +683,25 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) struct ibv_counter_set_description cs_desc; #endif - (void)pci_drv; assert(pci_drv == &mlx5_driver); /* Get mlx5_dev[] index. */ idx = mlx5_dev_idx(&pci_dev->addr); if (idx == -1) { - ERROR("this driver cannot support any more adapters"); - return -ENOMEM; + DRV_LOG(ERR, "this driver cannot support any more adapters"); + err = ENOMEM; + goto error; } - DEBUG("using driver device index %d", idx); - + DRV_LOG(DEBUG, "using driver device index %d", idx); /* Save PCI address. */ mlx5_dev[idx].pci_addr = pci_dev->addr; list = ibv_get_device_list(&i); if (list == NULL) { assert(errno); + err = errno; if (errno == ENOSYS) - ERROR("cannot list devices, is ib_uverbs loaded?"); - return -errno; + DRV_LOG(ERR, + "cannot list devices, is ib_uverbs loaded?"); + goto error; } assert(i >= 0); /* @@ -579,7 +712,7 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) struct rte_pci_addr pci_addr; --i; - DEBUG("checking device \"%s\"", list[i]->name); + DRV_LOG(DEBUG, "checking device \"%s\"", list[i]->name); if (mlx5_ibv_device_to_pci_addr(list[i], &pci_addr)) continue; if ((pci_dev->addr.domain != pci_addr.domain) || @@ -587,14 +720,6 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) (pci_dev->addr.devid != pci_addr.devid) || (pci_dev->addr.function != pci_addr.function)) continue; - sriov = ((pci_dev->id.device_id == - PCI_DEVICE_ID_MELLANOX_CONNECTX4VF) || - (pci_dev->id.device_id == - PCI_DEVICE_ID_MELLANOX_CONNECTX4LXVF) || - (pci_dev->id.device_id == - PCI_DEVICE_ID_MELLANOX_CONNECTX5VF) || - (pci_dev->id.device_id == - PCI_DEVICE_ID_MELLANOX_CONNECTX5EXVF)); switch (pci_dev->id.device_id) { case PCI_DEVICE_ID_MELLANOX_CONNECTX4: tunnel_en = 1; @@ -609,30 +734,29 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) default: break; } - INFO("PCI information matches, using device \"%s\"" - " (SR-IOV: %s)", - list[i]->name, - sriov ? "true" : "false"); + DRV_LOG(INFO, "PCI information matches, using device \"%s\"", + list[i]->name); attr_ctx = ibv_open_device(list[i]); - err = errno; + rte_errno = errno; + err = rte_errno; break; } if (attr_ctx == NULL) { - ibv_free_device_list(list); switch (err) { case 0: - ERROR("cannot access device, is mlx5_ib loaded?"); - return -ENODEV; + DRV_LOG(ERR, + "cannot access device, is mlx5_ib loaded?"); + err = ENODEV; + break; case EINVAL: - ERROR("cannot use device, are drivers up to date?"); - return -EINVAL; + DRV_LOG(ERR, + "cannot use device, are drivers up to date?"); + break; } - assert(err > 0); - return -err; + goto error; } ibv_dev = list[i]; - - DEBUG("device opened"); + DRV_LOG(DEBUG, "device opened"); /* * Multi-packet send is supported by ConnectX-4 Lx PF as well * as all ConnectX-5 devices. @@ -640,14 +764,14 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) mlx5dv_query_device(attr_ctx, &attrs_out); if (attrs_out.flags & MLX5DV_CONTEXT_FLAGS_MPW_ALLOWED) { if (attrs_out.flags & MLX5DV_CONTEXT_FLAGS_ENHANCED_MPW) { - DEBUG("Enhanced MPW is supported"); + DRV_LOG(DEBUG, "enhanced MPW is supported"); mps = MLX5_MPW_ENHANCED; } else { - DEBUG("MPW is supported"); + DRV_LOG(DEBUG, "MPW is supported"); mps = MLX5_MPW; } } else { - DEBUG("MPW isn't supported"); + DRV_LOG(DEBUG, "MPW isn't supported"); mps = MLX5_MPW_DISABLED; } if (RTE_CACHE_LINE_SIZE == 128 && @@ -655,10 +779,13 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) cqe_comp = 0; else cqe_comp = 1; - if (ibv_query_device_ex(attr_ctx, NULL, &device_attr)) + err = ibv_query_device_ex(attr_ctx, NULL, &device_attr); + if (err) { + DEBUG("ibv_query_device_ex() failed"); goto error; - INFO("%u port(s) detected", device_attr.orig_attr.phys_port_cnt); - + } + DRV_LOG(INFO, "%u port(s) detected", + device_attr.orig_attr.phys_port_cnt); for (i = 0; i < device_attr.orig_attr.phys_port_cnt; i++) { char name[RTE_ETH_NAME_MAX_LEN]; uint32_t port = i + 1; /* ports are indexed from one */ @@ -667,11 +794,9 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) struct ibv_port_attr port_attr; struct ibv_pd *pd = NULL; struct priv *priv = NULL; - struct rte_eth_dev *eth_dev; + struct rte_eth_dev *eth_dev = NULL; struct ibv_device_attr_ex device_attr_ex; struct ether_addr mac; - uint16_t num_vfs = 0; - struct ibv_device_attr_ex device_attr; struct mlx5_args args = { .cqe_comp = MLX5_ARG_UNSET, .txq_inline = MLX5_ARG_UNSET, @@ -687,84 +812,85 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) snprintf(name, sizeof(name), PCI_PRI_FMT, pci_dev->addr.domain, pci_dev->addr.bus, pci_dev->addr.devid, pci_dev->addr.function); - mlx5_dev[idx].ports |= test; - if (rte_eal_process_type() == RTE_PROC_SECONDARY) { eth_dev = rte_eth_dev_attach_secondary(name); if (eth_dev == NULL) { - ERROR("can not attach rte ethdev"); - err = ENOMEM; + DRV_LOG(ERR, "can not attach rte ethdev"); + rte_errno = ENOMEM; + err = rte_errno; goto error; } eth_dev->device = &pci_dev->device; eth_dev->dev_ops = &mlx5_dev_sec_ops; - priv = eth_dev->data->dev_private; + err = mlx5_uar_init_secondary(eth_dev); + if (err) { + err = rte_errno; + goto error; + } /* Receive command fd from primary process */ - err = priv_socket_connect(priv); + err = mlx5_socket_connect(eth_dev); if (err < 0) { - err = -err; + err = rte_errno; goto error; } /* Remap UAR for Tx queues. */ - err = priv_tx_uar_remap(priv, err); - if (err < 0) { - err = -err; + err = mlx5_tx_uar_remap(eth_dev, err); + if (err) { + err = rte_errno; goto error; } - priv_dev_select_rx_function(priv, eth_dev); - priv_dev_select_tx_function(priv, eth_dev); + /* + * Ethdev pointer is still required as input since + * the primary device is not accessible from the + * secondary process. + */ + eth_dev->rx_pkt_burst = + mlx5_select_rx_function(eth_dev); + eth_dev->tx_pkt_burst = + mlx5_select_tx_function(eth_dev); continue; } - - DEBUG("using port %u (%08" PRIx32 ")", port, test); - + DRV_LOG(DEBUG, "using port %u (%08" PRIx32 ")", port, test); ctx = ibv_open_device(ibv_dev); if (ctx == NULL) { err = ENODEV; goto port_error; } - - ibv_query_device_ex(ctx, NULL, &device_attr); /* Check port status. */ err = ibv_query_port(ctx, port, &port_attr); if (err) { - ERROR("port query failed: %s", strerror(err)); + DRV_LOG(ERR, "port query failed: %s", strerror(err)); goto port_error; } - if (port_attr.link_layer != IBV_LINK_LAYER_ETHERNET) { - ERROR("port %d is not configured in Ethernet mode", - port); + DRV_LOG(ERR, + "port %d is not configured in Ethernet mode", + port); err = EINVAL; goto port_error; } - if (port_attr.state != IBV_PORT_ACTIVE) - DEBUG("port %d is not active: \"%s\" (%d)", - port, ibv_port_state_str(port_attr.state), - port_attr.state); - + DRV_LOG(DEBUG, "port %d is not active: \"%s\" (%d)", + port, ibv_port_state_str(port_attr.state), + port_attr.state); /* Allocate protection domain. */ pd = ibv_alloc_pd(ctx); if (pd == NULL) { - ERROR("PD allocation failure"); + DRV_LOG(ERR, "PD allocation failure"); err = ENOMEM; goto port_error; } - mlx5_dev[idx].ports |= test; - /* from rte_ethdev.c */ priv = rte_zmalloc("ethdev private structure", sizeof(*priv), RTE_CACHE_LINE_SIZE); if (priv == NULL) { - ERROR("priv allocation failure"); + DRV_LOG(ERR, "priv allocation failure"); err = ENOMEM; goto port_error; } - priv->ctx = ctx; strncpy(priv->ibdev_path, priv->ctx->device->ibdev_path, sizeof(priv->ibdev_path)); @@ -780,35 +906,37 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) priv->rx_vec_en = 1; err = mlx5_args(&args, pci_dev->device.devargs); if (err) { - ERROR("failed to process device arguments: %s", - strerror(err)); + DRV_LOG(ERR, "failed to process device arguments: %s", + strerror(err)); + err = rte_errno; goto port_error; } mlx5_args_assign(priv, &args); - if (ibv_query_device_ex(ctx, NULL, &device_attr_ex)) { - ERROR("ibv_query_device_ex() failed"); + err = ibv_query_device_ex(ctx, NULL, &device_attr_ex); + if (err) { + DRV_LOG(ERR, "ibv_query_device_ex() failed"); goto port_error; } - priv->hw_csum = !!(device_attr_ex.device_cap_flags_ex & IBV_DEVICE_RAW_IP_CSUM); - DEBUG("checksum offloading is %ssupported", - (priv->hw_csum ? "" : "not ")); + DRV_LOG(DEBUG, "checksum offloading is %ssupported", + (priv->hw_csum ? "" : "not ")); #ifdef HAVE_IBV_DEVICE_VXLAN_SUPPORT priv->hw_csum_l2tun = !!(exp_device_attr.exp_device_cap_flags & IBV_DEVICE_VXLAN_SUPPORT); #endif - DEBUG("Rx L2 tunnel checksum offloads are %ssupported", - (priv->hw_csum_l2tun ? "" : "not ")); + DRV_LOG(DEBUG, "Rx L2 tunnel checksum offloads are %ssupported", + (priv->hw_csum_l2tun ? "" : "not ")); #ifdef HAVE_IBV_DEVICE_COUNTERS_SET_SUPPORT priv->counter_set_supported = !!(device_attr.max_counter_sets); ibv_describe_counter_set(ctx, 0, &cs_desc); - DEBUG("counter type = %d, num of cs = %ld, attributes = %d", - cs_desc.counter_type, cs_desc.num_of_cs, - cs_desc.attributes); + DRV_LOG(DEBUG, + "counter type = %d, num of cs = %ld, attributes = %d", + cs_desc.counter_type, cs_desc.num_of_cs, + cs_desc.attributes); #endif priv->ind_table_max_size = device_attr_ex.rss_caps.max_rwq_indirection_table_size; @@ -817,27 +945,24 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) if (priv->ind_table_max_size > (unsigned int)ETH_RSS_RETA_SIZE_512) priv->ind_table_max_size = ETH_RSS_RETA_SIZE_512; - DEBUG("maximum RX indirection table size is %u", - priv->ind_table_max_size); + DRV_LOG(DEBUG, "maximum Rx indirection table size is %u", + priv->ind_table_max_size); priv->hw_vlan_strip = !!(device_attr_ex.raw_packet_caps & IBV_RAW_PACKET_CAP_CVLAN_STRIPPING); - DEBUG("VLAN stripping is %ssupported", - (priv->hw_vlan_strip ? "" : "not ")); + DRV_LOG(DEBUG, "VLAN stripping is %ssupported", + (priv->hw_vlan_strip ? "" : "not ")); - priv->hw_fcs_strip = - !!(device_attr_ex.orig_attr.device_cap_flags & - IBV_WQ_FLAGS_SCATTER_FCS); - DEBUG("FCS stripping configuration is %ssupported", - (priv->hw_fcs_strip ? "" : "not ")); + priv->hw_fcs_strip = !!(device_attr_ex.raw_packet_caps & + IBV_RAW_PACKET_CAP_SCATTER_FCS); + DRV_LOG(DEBUG, "FCS stripping configuration is %ssupported", + (priv->hw_fcs_strip ? "" : "not ")); #ifdef HAVE_IBV_WQ_FLAG_RX_END_PADDING priv->hw_padding = !!device_attr_ex.rx_pad_end_addr_align; #endif - DEBUG("hardware RX end alignment padding is %ssupported", - (priv->hw_padding ? "" : "not ")); - - priv_get_num_vfs(priv, &num_vfs); - priv->sriov = (num_vfs || sriov); + DRV_LOG(DEBUG, + "hardware Rx end alignment padding is %ssupported", + (priv->hw_padding ? "" : "not ")); priv->tso = ((priv->tso) && (device_attr_ex.tso_caps.max_tso > 0) && (device_attr_ex.tso_caps.supported_qpts & @@ -846,18 +971,21 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) priv->max_tso_payload_sz = device_attr_ex.tso_caps.max_tso; if (priv->mps && !mps) { - ERROR("multi-packet send not supported on this device" - " (" MLX5_TXQ_MPW_EN ")"); + DRV_LOG(ERR, + "multi-packet send not supported on this device" + " (" MLX5_TXQ_MPW_EN ")"); err = ENOTSUP; goto port_error; } else if (priv->mps && priv->tso) { - WARN("multi-packet send not supported in conjunction " - "with TSO. MPS disabled"); + DRV_LOG(WARNING, + "multi-packet send not supported in conjunction" + " with TSO. MPS disabled"); priv->mps = 0; } - INFO("%sMPS is %s", - priv->mps == MLX5_MPW_ENHANCED ? "Enhanced " : "", - priv->mps != MLX5_MPW_DISABLED ? "enabled" : "disabled"); + DRV_LOG(INFO, "%s MPS is %s", + priv->mps == MLX5_MPW_ENHANCED ? "enhanced " : "", + priv->mps != MLX5_MPW_DISABLED ? "enabled" : + "disabled"); /* Set default values for Enhanced MPW, a.k.a MPWv2. */ if (priv->mps == MLX5_MPW_ENHANCED) { if (args.txqs_inline == MLX5_ARG_UNSET) @@ -870,59 +998,71 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) MLX5_WQE_SIZE; } if (priv->cqe_comp && !cqe_comp) { - WARN("Rx CQE compression isn't supported"); + DRV_LOG(WARNING, "Rx CQE compression isn't supported"); priv->cqe_comp = 0; } + eth_dev = rte_eth_dev_allocate(name); + if (eth_dev == NULL) { + DRV_LOG(ERR, "can not allocate rte ethdev"); + err = ENOMEM; + goto port_error; + } + eth_dev->data->dev_private = priv; + priv->dev_data = eth_dev->data; + eth_dev->data->mac_addrs = priv->mac; + eth_dev->device = &pci_dev->device; + rte_eth_copy_pci_info(eth_dev, pci_dev); + eth_dev->device->driver = &mlx5_driver.driver; + err = mlx5_uar_init_primary(eth_dev); + if (err) { + err = rte_errno; + goto port_error; + } /* Configure the first MAC address by default. */ - if (priv_get_mac(priv, &mac.addr_bytes)) { - ERROR("cannot get MAC address, is mlx5_en loaded?" - " (errno: %s)", strerror(errno)); + if (mlx5_get_mac(eth_dev, &mac.addr_bytes)) { + DRV_LOG(ERR, + "port %u cannot get MAC address, is mlx5_en" + " loaded? (errno: %s)", + eth_dev->data->port_id, strerror(errno)); err = ENODEV; goto port_error; } - INFO("port %u MAC address is %02x:%02x:%02x:%02x:%02x:%02x", - priv->port, - mac.addr_bytes[0], mac.addr_bytes[1], - mac.addr_bytes[2], mac.addr_bytes[3], - mac.addr_bytes[4], mac.addr_bytes[5]); + DRV_LOG(INFO, + "port %u MAC address is %02x:%02x:%02x:%02x:%02x:%02x", + eth_dev->data->port_id, + mac.addr_bytes[0], mac.addr_bytes[1], + mac.addr_bytes[2], mac.addr_bytes[3], + mac.addr_bytes[4], mac.addr_bytes[5]); #ifndef NDEBUG { char ifname[IF_NAMESIZE]; - if (priv_get_ifname(priv, &ifname) == 0) - DEBUG("port %u ifname is \"%s\"", - priv->port, ifname); + if (mlx5_get_ifname(eth_dev, &ifname) == 0) + DRV_LOG(DEBUG, "port %u ifname is \"%s\"", + eth_dev->data->port_id, ifname); else - DEBUG("port %u ifname is unknown", priv->port); + DRV_LOG(DEBUG, "port %u ifname is unknown", + eth_dev->data->port_id); } #endif /* Get actual MTU if possible. */ - priv_get_mtu(priv, &priv->mtu); - DEBUG("port %u MTU is %u", priv->port, priv->mtu); - - eth_dev = rte_eth_dev_allocate(name); - if (eth_dev == NULL) { - ERROR("can not allocate rte ethdev"); - err = ENOMEM; + err = mlx5_get_mtu(eth_dev, &priv->mtu); + if (err) { + err = rte_errno; goto port_error; } - eth_dev->data->dev_private = priv; - eth_dev->data->mac_addrs = priv->mac; - eth_dev->device = &pci_dev->device; - rte_eth_copy_pci_info(eth_dev, pci_dev); - eth_dev->device->driver = &mlx5_driver.driver; + DRV_LOG(DEBUG, "port %u MTU is %u", eth_dev->data->port_id, + priv->mtu); /* * Initialize burst functions to prevent crashes before link-up. */ eth_dev->rx_pkt_burst = removed_rx_burst; eth_dev->tx_pkt_burst = removed_tx_burst; - priv->dev = eth_dev; eth_dev->dev_ops = &mlx5_dev_ops; /* Register MAC address. */ claim_zero(mlx5_mac_addr_add(eth_dev, &mac, 0, 0)); TAILQ_INIT(&priv->flows); TAILQ_INIT(&priv->ctrl_flows); - /* Hint libmlx5 to use PMD allocator for data plane resources */ struct mlx5dv_ctx_allocators alctr = { .alloc = &mlx5_alloc_verbs_buf, @@ -931,12 +1071,17 @@ mlx5_pci_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) }; mlx5dv_set_context_attr(ctx, MLX5DV_CTX_ATTR_BUF_ALLOCATORS, (void *)((uintptr_t)&alctr)); - /* Bring Ethernet device up. */ - DEBUG("forcing Ethernet interface up"); - priv_set_flags(priv, ~IFF_UP, IFF_UP); + DRV_LOG(DEBUG, "port %u forcing Ethernet interface up", + eth_dev->data->port_id); + mlx5_set_link_up(eth_dev); + /* + * Even though the interrupt handler is not installed yet, + * interrupts will still trigger on the asyn_fd from + * Verbs context returned by ibv_open_device(). + */ + mlx5_link_update(eth_dev, 0); continue; - port_error: if (priv) rte_free(priv); @@ -944,29 +1089,31 @@ port_error: claim_zero(ibv_dealloc_pd(pd)); if (ctx) claim_zero(ibv_close_device(ctx)); + if (eth_dev && rte_eal_process_type() == RTE_PROC_PRIMARY) + rte_eth_dev_release_port(eth_dev); break; } - /* * XXX if something went wrong in the loop above, there is a resource * leak (ctx, pd, priv, dpdk ethdev) but we can do nothing about it as * long as the dpdk does not provide a way to deallocate a ethdev and a * way to enumerate the registered ethdevs to free the previous ones. */ - /* no port found, complain */ if (!mlx5_dev[idx].ports) { - err = ENODEV; - goto error; + rte_errno = ENODEV; + err = rte_errno; } - error: if (attr_ctx) claim_zero(ibv_close_device(attr_ctx)); if (list) ibv_free_device_list(list); - assert(err >= 0); - return -err; + if (err) { + rte_errno = err; + return -rte_errno; + } + return 0; } static const struct rte_pci_id mlx5_pci_id_map[] = { @@ -1042,3 +1189,11 @@ rte_mlx5_pmd_init(void) RTE_PMD_EXPORT_NAME(net_mlx5, __COUNTER__); RTE_PMD_REGISTER_PCI_TABLE(net_mlx5, mlx5_pci_id_map); RTE_PMD_REGISTER_KMOD_DEP(net_mlx5, "* ib_uverbs & mlx5_core & mlx5_ib"); + +/** Initialize driver log type. */ +RTE_INIT(vdev_netvsc_init_log) +{ + mlx5_logtype = rte_log_register("pmd.net.mlx5"); + if (mlx5_logtype >= 0) + rte_log_set_level(mlx5_logtype, RTE_LOG_NOTICE); +} |