From patchwork Thu Jun 18 19:06:02 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Matan Azrad X-Patchwork-Id: 71754 X-Patchwork-Delegate: maxime.coquelin@redhat.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id A39A0A04F1; Thu, 18 Jun 2020 21:06:12 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id B23D21BFF3; Thu, 18 Jun 2020 21:06:11 +0200 (CEST) Received: from mellanox.co.il (mail-il-dmz.mellanox.com [193.47.165.129]) by dpdk.org (Postfix) with ESMTP id 44F5F1BFEC for ; Thu, 18 Jun 2020 21:06:10 +0200 (CEST) Received: from Internal Mail-Server by MTLPINE1 (envelope-from matan@mellanox.com) with SMTP; 18 Jun 2020 22:06:09 +0300 Received: from pegasus25.mtr.labs.mlnx. (pegasus25.mtr.labs.mlnx [10.210.16.10]) by labmailer.mlnx (8.13.8/8.13.8) with ESMTP id 05IJ67gU003175; Thu, 18 Jun 2020 22:06:08 +0300 From: Matan Azrad To: Maxime Coquelin , Xiao Wang Cc: dev@dpdk.org Date: Thu, 18 Jun 2020 19:06:02 +0000 Message-Id: <1592507163-441875-2-git-send-email-matan@mellanox.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1592507163-441875-1-git-send-email-matan@mellanox.com> References: <1591113208-79169-2-git-send-email-matan@mellanox.com> <1592507163-441875-1-git-send-email-matan@mellanox.com> Subject: [dpdk-dev] [PATCH v2 1/2] common/mlx5: share kernel interface name getter X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Some configuration of the mlx5 port are done by the kernel net device associated to the IB device represents the PCI device. The DPDK mlx5 driver uses linux system calls, for example ioctl, in order to configure per port configurations requested by the DPDK user. One of the basic knowledges required to access the correct kernel net device is its name. Move function to get interface name from IB device path to the common library. Signed-off-by: Matan Azrad Reviewed-by: Maxime Coquelin --- drivers/common/mlx5/linux/mlx5_common_os.c | 89 ++++++++++++++++++++++++ drivers/common/mlx5/mlx5_common.h | 3 + drivers/common/mlx5/rte_common_mlx5_version.map | 2 + drivers/net/mlx5/linux/mlx5_ethdev_os.c | 91 +------------------------ drivers/net/mlx5/linux/mlx5_os.c | 4 +- drivers/net/mlx5/mlx5.h | 1 - 6 files changed, 99 insertions(+), 91 deletions(-) diff --git a/drivers/common/mlx5/linux/mlx5_common_os.c b/drivers/common/mlx5/linux/mlx5_common_os.c index 1b71347..aa67e5a 100644 --- a/drivers/common/mlx5/linux/mlx5_common_os.c +++ b/drivers/common/mlx5/linux/mlx5_common_os.c @@ -8,8 +8,11 @@ #ifdef RTE_IBVERBS_LINK_DLOPEN #include #endif +#include +#include #include +#include #include "mlx5_common.h" #include "mlx5_common_utils.h" @@ -137,6 +140,92 @@ return sysconf(_SC_PAGESIZE); } +/** + * Get kernel interface name from IB device path. + * + * @param[in] ibdev_path + * Pointer to IB device path. + * @param[out] ifname + * Interface name output buffer. + * + * @return + * 0 on success, a negative errno value otherwise and rte_errno is set. + */ +int +mlx5_get_ifname_sysfs(const char *ibdev_path, char *ifname) +{ + DIR *dir; + struct dirent *dent; + unsigned int dev_type = 0; + unsigned int dev_port_prev = ~0u; + char match[IF_NAMESIZE] = ""; + + MLX5_ASSERT(ibdev_path); + { + MKSTR(path, "%s/device/net", ibdev_path); + + dir = opendir(path); + if (dir == NULL) { + rte_errno = errno; + return -rte_errno; + } + } + while ((dent = readdir(dir)) != NULL) { + char *name = dent->d_name; + FILE *file; + unsigned int dev_port; + int r; + + if ((name[0] == '.') && + ((name[1] == '\0') || + ((name[1] == '.') && (name[2] == '\0')))) + continue; + + MKSTR(path, "%s/device/net/%s/%s", + ibdev_path, name, + (dev_type ? "dev_id" : "dev_port")); + + file = fopen(path, "rb"); + if (file == NULL) { + if (errno != ENOENT) + continue; + /* + * Switch to dev_id when dev_port does not exist as + * is the case with Linux kernel versions < 3.15. + */ +try_dev_id: + match[0] = '\0'; + if (dev_type) + break; + dev_type = 1; + dev_port_prev = ~0u; + rewinddir(dir); + continue; + } + r = fscanf(file, (dev_type ? "%x" : "%u"), &dev_port); + fclose(file); + if (r != 1) + continue; + /* + * Switch to dev_id when dev_port returns the same value for + * all ports. May happen when using a MOFED release older than + * 3.0 with a Linux kernel >= 3.15. + */ + if (dev_port == dev_port_prev) + goto try_dev_id; + dev_port_prev = dev_port; + if (dev_port == 0) + strlcpy(match, name, IF_NAMESIZE); + } + closedir(dir); + if (match[0] == '\0') { + rte_errno = ENOENT; + return -rte_errno; + } + strncpy(ifname, match, IF_NAMESIZE); + return 0; +} + #ifdef MLX5_GLUE /** diff --git a/drivers/common/mlx5/mlx5_common.h b/drivers/common/mlx5/mlx5_common.h index 77f10e6..82b4b13 100644 --- a/drivers/common/mlx5/mlx5_common.h +++ b/drivers/common/mlx5/mlx5_common.h @@ -198,6 +198,9 @@ enum mlx5_cqe_status { __rte_internal int mlx5_dev_to_pci_addr(const char *dev_path, struct rte_pci_addr *pci_addr); +__rte_internal +int mlx5_get_ifname_sysfs(const char *ibdev_path, char *ifname); + #define MLX5_CLASS_ARG_NAME "class" diff --git a/drivers/common/mlx5/rte_common_mlx5_version.map b/drivers/common/mlx5/rte_common_mlx5_version.map index 3f62400..514d7c5 100644 --- a/drivers/common/mlx5/rte_common_mlx5_version.map +++ b/drivers/common/mlx5/rte_common_mlx5_version.map @@ -36,6 +36,8 @@ INTERNAL { mlx5_devx_cmd_query_virtq; mlx5_devx_get_out_command_status; + mlx5_get_ifname_sysfs; + mlx5_mp_init_primary; mlx5_mp_uninit_primary; mlx5_mp_init_secondary; diff --git a/drivers/net/mlx5/linux/mlx5_ethdev_os.c b/drivers/net/mlx5/linux/mlx5_ethdev_os.c index ab47cb5..21105f6 100644 --- a/drivers/net/mlx5/linux/mlx5_ethdev_os.c +++ b/drivers/net/mlx5/linux/mlx5_ethdev_os.c @@ -128,96 +128,11 @@ struct ethtool_link_settings { #define ETHTOOL_LINK_MODE_200000baseCR4_Full_BIT 2 /* 66 - 64 */ #endif -/** - * Get master interface name from private structure. - * - * @param[in] dev - * Pointer to Ethernet device. - * @param[out] ifname - * Interface name output buffer. - * - * @return - * 0 on success, a negative errno value otherwise and rte_errno is set. - */ -int -mlx5_get_master_ifname(const char *ibdev_path, char (*ifname)[IF_NAMESIZE]) -{ - DIR *dir; - struct dirent *dent; - unsigned int dev_type = 0; - unsigned int dev_port_prev = ~0u; - char match[IF_NAMESIZE] = ""; - - MLX5_ASSERT(ibdev_path); - { - MKSTR(path, "%s/device/net", ibdev_path); - - dir = opendir(path); - if (dir == NULL) { - rte_errno = errno; - return -rte_errno; - } - } - while ((dent = readdir(dir)) != NULL) { - char *name = dent->d_name; - FILE *file; - unsigned int dev_port; - int r; - - if ((name[0] == '.') && - ((name[1] == '\0') || - ((name[1] == '.') && (name[2] == '\0')))) - continue; - - MKSTR(path, "%s/device/net/%s/%s", - ibdev_path, name, - (dev_type ? "dev_id" : "dev_port")); - - file = fopen(path, "rb"); - if (file == NULL) { - if (errno != ENOENT) - continue; - /* - * Switch to dev_id when dev_port does not exist as - * is the case with Linux kernel versions < 3.15. - */ -try_dev_id: - match[0] = '\0'; - if (dev_type) - break; - dev_type = 1; - dev_port_prev = ~0u; - rewinddir(dir); - continue; - } - r = fscanf(file, (dev_type ? "%x" : "%u"), &dev_port); - fclose(file); - if (r != 1) - continue; - /* - * Switch to dev_id when dev_port returns the same value for - * all ports. May happen when using a MOFED release older than - * 3.0 with a Linux kernel >= 3.15. - */ - if (dev_port == dev_port_prev) - goto try_dev_id; - dev_port_prev = dev_port; - if (dev_port == 0) - strlcpy(match, name, sizeof(match)); - } - closedir(dir); - if (match[0] == '\0') { - rte_errno = ENOENT; - return -rte_errno; - } - strncpy(*ifname, match, sizeof(*ifname)); - return 0; -} /** * Get interface name from private structure. * - * This is a port representor-aware version of mlx5_get_master_ifname(). + * This is a port representor-aware version of mlx5_get_ifname_sysfs(). * * @param[in] dev * Pointer to Ethernet device. @@ -238,8 +153,8 @@ struct ethtool_link_settings { ifindex = mlx5_ifindex(dev); if (!ifindex) { if (!priv->representor) - return mlx5_get_master_ifname(priv->sh->ibdev_path, - ifname); + return mlx5_get_ifname_sysfs(priv->sh->ibdev_path, + *ifname); rte_errno = ENXIO; return -rte_errno; } diff --git a/drivers/net/mlx5/linux/mlx5_os.c b/drivers/net/mlx5/linux/mlx5_os.c index 3792371..b330cd5 100644 --- a/drivers/net/mlx5/linux/mlx5_os.c +++ b/drivers/net/mlx5/linux/mlx5_os.c @@ -1637,8 +1637,8 @@ */ continue; } - ret = mlx5_get_master_ifname - (ibv_match[i]->ibdev_path, &ifname); + ret = mlx5_get_ifname_sysfs + (ibv_match[i]->ibdev_path, ifname); if (!ret) list[ns].ifindex = if_nametoindex(ifname); diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h index 5bd5acd..8ecb59c 100644 --- a/drivers/net/mlx5/mlx5.h +++ b/drivers/net/mlx5/mlx5.h @@ -731,7 +731,6 @@ int mlx5_hairpin_cap_get(struct rte_eth_dev *dev, /* mlx5_ethdev_os.c */ int mlx5_get_ifname(const struct rte_eth_dev *dev, char (*ifname)[IF_NAMESIZE]); -int mlx5_get_master_ifname(const char *ibdev_path, char (*ifname)[IF_NAMESIZE]); unsigned int mlx5_ifindex(const struct rte_eth_dev *dev); int mlx5_ifreq(const struct rte_eth_dev *dev, int req, struct ifreq *ifr); int mlx5_get_mtu(struct rte_eth_dev *dev, uint16_t *mtu); From patchwork Thu Jun 18 19:06:03 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Matan Azrad X-Patchwork-Id: 71756 X-Patchwork-Delegate: maxime.coquelin@redhat.com Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id E2746A04F1; Thu, 18 Jun 2020 21:06:25 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id EBEF81C027; Thu, 18 Jun 2020 21:06:20 +0200 (CEST) Received: from mellanox.co.il (mail-il-dmz.mellanox.com [193.47.165.129]) by dpdk.org (Postfix) with ESMTP id 13A1C1C01E for ; Thu, 18 Jun 2020 21:06:15 +0200 (CEST) Received: from Internal Mail-Server by MTLPINE1 (envelope-from matan@mellanox.com) with SMTP; 18 Jun 2020 22:06:12 +0300 Received: from pegasus25.mtr.labs.mlnx. (pegasus25.mtr.labs.mlnx [10.210.16.10]) by labmailer.mlnx (8.13.8/8.13.8) with ESMTP id 05IJ67gV003175; Thu, 18 Jun 2020 22:06:12 +0300 From: Matan Azrad To: Maxime Coquelin , Xiao Wang Cc: dev@dpdk.org Date: Thu, 18 Jun 2020 19:06:03 +0000 Message-Id: <1592507163-441875-3-git-send-email-matan@mellanox.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1592507163-441875-1-git-send-email-matan@mellanox.com> References: <1591113208-79169-2-git-send-email-matan@mellanox.com> <1592507163-441875-1-git-send-email-matan@mellanox.com> Subject: [dpdk-dev] [PATCH v2 2/2] vdpa/mlx5: support MTU feature X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" The guest virtio device may request MTU updating when the vhost backend device exposes a capability to support it. Exspose the MTU feature capability. At configuration time, check the requested MTU and update it in the HW device. Signed-off-by: Matan Azrad Reviewed-by: Maxime Coquelin --- doc/guides/rel_notes/release_20_08.rst | 1 + doc/guides/vdpadevs/features/mlx5.ini | 1 + drivers/vdpa/mlx5/mlx5_vdpa.c | 67 ++++++++++++++++++++++++++++++++-- 3 files changed, 65 insertions(+), 4 deletions(-) diff --git a/doc/guides/rel_notes/release_20_08.rst b/doc/guides/rel_notes/release_20_08.rst index e2c93b1..cc39984 100644 --- a/doc/guides/rel_notes/release_20_08.rst +++ b/doc/guides/rel_notes/release_20_08.rst @@ -79,6 +79,7 @@ New Features Updated Mellanox mlx5 vDPA driver with new features, including: * Added support for virtio queue statistics. + * Added support for MTU update. diff --git a/doc/guides/vdpadevs/features/mlx5.ini b/doc/guides/vdpadevs/features/mlx5.ini index 788d4e0..8bb9977 100644 --- a/doc/guides/vdpadevs/features/mlx5.ini +++ b/doc/guides/vdpadevs/features/mlx5.ini @@ -10,6 +10,7 @@ host tso4 = Y host tso6 = Y version 1 = Y log all = Y +mtu = Y any layout = Y guest announce = Y mq = Y diff --git a/drivers/vdpa/mlx5/mlx5_vdpa.c b/drivers/vdpa/mlx5/mlx5_vdpa.c index 94cac66..8b0b3b8 100644 --- a/drivers/vdpa/mlx5/mlx5_vdpa.c +++ b/drivers/vdpa/mlx5/mlx5_vdpa.c @@ -2,6 +2,11 @@ * Copyright 2019 Mellanox Technologies, Ltd */ #include +#include +#include +#include +#include +#include #include #include @@ -25,14 +30,19 @@ (1ULL << VIRTIO_NET_F_MQ) | \ (1ULL << VIRTIO_NET_F_GUEST_ANNOUNCE) | \ (1ULL << VIRTIO_F_ORDER_PLATFORM) | \ - (1ULL << VHOST_F_LOG_ALL)) + (1ULL << VHOST_F_LOG_ALL) | \ + (1ULL << VIRTIO_NET_F_MTU)) #define MLX5_VDPA_PROTOCOL_FEATURES \ ((1ULL << VHOST_USER_PROTOCOL_F_SLAVE_REQ) | \ (1ULL << VHOST_USER_PROTOCOL_F_SLAVE_SEND_FD) | \ (1ULL << VHOST_USER_PROTOCOL_F_HOST_NOTIFIER) | \ (1ULL << VHOST_USER_PROTOCOL_F_LOG_SHMFD) | \ - (1ULL << VHOST_USER_PROTOCOL_F_MQ)) + (1ULL << VHOST_USER_PROTOCOL_F_MQ) | \ + (1ULL << VHOST_USER_PROTOCOL_F_NET_MTU)) + +#define MLX5_VDPA_MAX_RETRIES 20 +#define MLX5_VDPA_USEC 1000 TAILQ_HEAD(mlx5_vdpa_privs, mlx5_vdpa_priv) priv_list = TAILQ_HEAD_INITIALIZER(priv_list); @@ -223,6 +233,55 @@ } static int +mlx5_vdpa_mtu_set(struct mlx5_vdpa_priv *priv) +{ + struct ifreq request; + uint16_t vhost_mtu = 0; + uint16_t kern_mtu = 0; + int ret = rte_vhost_get_mtu(priv->vid, &vhost_mtu); + int sock; + int retries = MLX5_VDPA_MAX_RETRIES; + + if (ret) { + DRV_LOG(DEBUG, "Cannot get vhost MTU - %d.", ret); + return ret; + } + if (!vhost_mtu) { + DRV_LOG(DEBUG, "Vhost MTU is 0."); + return ret; + } + ret = mlx5_get_ifname_sysfs(priv->ctx->device->ibdev_path, + request.ifr_name); + if (ret) { + DRV_LOG(DEBUG, "Cannot get kernel IF name - %d.", ret); + return ret; + } + sock = socket(PF_INET, SOCK_DGRAM, IPPROTO_IP); + if (sock == -1) { + DRV_LOG(DEBUG, "Cannot open IF socket."); + return sock; + } + while (retries--) { + ret = ioctl(sock, SIOCGIFMTU, &request); + if (ret == -1) + break; + kern_mtu = request.ifr_mtu; + DRV_LOG(DEBUG, "MTU: current %d requested %d.", (int)kern_mtu, + (int)vhost_mtu); + if (kern_mtu == vhost_mtu) + break; + request.ifr_mtu = vhost_mtu; + ret = ioctl(sock, SIOCSIFMTU, &request); + if (ret == -1) + break; + request.ifr_mtu = 0; + usleep(MLX5_VDPA_USEC); + } + close(sock); + return kern_mtu == vhost_mtu ? 0 : -1; +} + +static int mlx5_vdpa_dev_close(int vid) { int did = rte_vhost_get_vdpa_device_id(vid); @@ -265,6 +324,8 @@ return -1; } priv->vid = vid; + if (mlx5_vdpa_mtu_set(priv)) + DRV_LOG(WARNING, "MTU cannot be set on device %d.", did); if (mlx5_vdpa_pd_create(priv) || mlx5_vdpa_mem_register(priv) || mlx5_vdpa_direct_db_prepare(priv) || mlx5_vdpa_virtqs_prepare(priv) || mlx5_vdpa_steer_setup(priv) || @@ -513,8 +574,6 @@ return ret; } -#define MLX5_VDPA_MAX_RETRIES 20 -#define MLX5_VDPA_USEC 1000 static int mlx5_vdpa_roce_disable(struct rte_pci_addr *addr, struct ibv_device **ibv) {