From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from EUR02-VE1-obe.outbound.protection.outlook.com (mail-eopbgr20074.outbound.protection.outlook.com [40.107.2.74]) by dpdk.org (Postfix) with ESMTP id 8D0E95B30 for ; Sat, 3 Nov 2018 07:18:45 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=Mellanox.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=fQKNr/vPz5Z2vYFWozU223Ni4xfOKekw7fwnaW2s4y4=; b=ml86jq/evFBT543eayB9Zsr98CQaAXB4UTDxcas7LgkgYlm+4exQqK6LiiKZB2otc83udxENXC0qLq0romPbRixYYSf20/h+jvoHV+FZOeez0EeYPDOym2H8pGRGq9CLd/H4dD30Ze/NVy1goqMOHkQFTM4fcrkWGT3I/2gpG0I= Received: from AM4PR05MB3265.eurprd05.prod.outlook.com (10.171.186.150) by AM4PR05MB3250.eurprd05.prod.outlook.com (10.171.186.147) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.1294.21; Sat, 3 Nov 2018 06:18:44 +0000 Received: from AM4PR05MB3265.eurprd05.prod.outlook.com ([fe80::544b:a68d:e6a5:ba6e]) by AM4PR05MB3265.eurprd05.prod.outlook.com ([fe80::544b:a68d:e6a5:ba6e%2]) with mapi id 15.20.1294.027; Sat, 3 Nov 2018 06:18:44 +0000 From: Slava Ovsiienko To: Shahaf Shuler CC: "dev@dpdk.org" , Yongseok Koh , Slava Ovsiienko Thread-Topic: [PATCH v5 11/13] net/mlx5: add E-switch VXLAN tunnel devices management Thread-Index: AQHUcz0SY18NZe2+MkWHDCRySWnz/Q== Date: Sat, 3 Nov 2018 06:18:44 +0000 Message-ID: <1541225876-8817-12-git-send-email-viacheslavo@mellanox.com> References: <1541181152-15788-2-git-send-email-viacheslavo@mellanox.com> <1541225876-8817-1-git-send-email-viacheslavo@mellanox.com> In-Reply-To: <1541225876-8817-1-git-send-email-viacheslavo@mellanox.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-clientproxiedby: CWLP123CA0088.GBRP123.PROD.OUTLOOK.COM (2603:10a6:401:5b::28) To AM4PR05MB3265.eurprd05.prod.outlook.com (2603:10a6:205:4::22) authentication-results: spf=none (sender IP is ) smtp.mailfrom=viacheslavo@mellanox.com; x-ms-exchange-messagesentrepresentingtype: 1 x-originating-ip: [37.142.13.130] x-ms-publictraffictype: Email x-microsoft-exchange-diagnostics: 1; AM4PR05MB3250; 6:zqRtKrdcRZ0bPjjWEUUjvU5z8jrGKd3V+8vdoQgP9/m1v4dLyii5MfLRUe3642UZDJD6X5NW03+k1mAtj10owAIVIXCiJkhNksb4BSxHD61+VbL3W7NnG7qgGgXFEth+2pmybnoRboNLWHzbUKcoSAJDnWJo8ESfVTY7Sf7JmM1g9RUrTbcMr+bMD7OYSLUXZjO5VeB7g5MutcDMYAmBeVTImStQDOOtLe9vDuBj+5bda63+qpAgsthvnSfsGdzlRmvFCSR0uPrXB3Yd6ZGKpc4TUqNDI6fRm1Nr/POJjhVUyGA5N7onNCYegIQ3WqQlGNY234reliqFYwKiRzNX7HtWO6ojxfsIbI8Oey6RQRLn+8pxeVBJFCLljwusxXkGjEA3vA2zOCBhwFCbhxFATWa1Cq9/EEizhHonPE+57/eolz5s3xSY8HxrggzAZXLtvvN7VdGllbreTks7S1U7Zg==; 5:ZyBDLOyEXDSViksrO88RMvr4nk9tAVzi8fhSsA0yRb+XWGrmaqF4hvQyWmYr1ouoVGrboMJ0eZIl7QUqIle0kRpYaWOHeBGAQM6HTtLIHVKUsstHO6Xmo9l1k3yIFD37QLFpk60/61HYKqRC7LQ/97N/UINtyPuMcq4DUtGYf6Y=; 7:mc2dwihDuR09XyOhIrZgcvUbnurpXuPmMxTMr2J6UB41kyFewaAxlry9VjsvkU8xfT9PPJjPRSO2tyj979Qfd9p4YwOQi9k2j8RKx72LB4ngiFjXcb49ftlCDrw+ApeCJlym5flEk72W+BQ47WaW/g== x-ms-office365-filtering-correlation-id: 7d28c0f0-fa0e-46be-93cb-08d6415434fc x-ms-office365-filtering-ht: Tenant x-microsoft-antispam: BCL:0; PCL:0; RULEID:(7020095)(4652040)(8989299)(5600074)(711020)(4618075)(4534185)(4627221)(201703031133081)(201702281549075)(8990200)(2017052603328)(7153060)(7193020); SRVR:AM4PR05MB3250; x-ms-traffictypediagnostic: AM4PR05MB3250: x-microsoft-antispam-prvs: x-exchange-antispam-report-test: UriScan:(211171220733660)(788757137089); x-ms-exchange-senderadcheck: 1 x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:(8211001083)(6040522)(2401047)(8121501046)(5005006)(3231382)(944501410)(52105095)(3002001)(10201501046)(93006095)(93001095)(6055026)(148016)(149066)(150057)(6041310)(201703131423095)(201702281528075)(20161123555045)(201703061421075)(201703061406153)(20161123562045)(20161123560045)(20161123564045)(20161123558120)(201708071742011)(7699051)(76991095); SRVR:AM4PR05MB3250; BCL:0; PCL:0; RULEID:; SRVR:AM4PR05MB3250; x-forefront-prvs: 08457955C4 x-forefront-antispam-report: SFV:NSPM; SFS:(10009020)(346002)(376002)(366004)(39860400002)(136003)(396003)(189003)(199004)(81156014)(4744004)(66066001)(107886003)(6436002)(71200400001)(71190400001)(6636002)(106356001)(68736007)(25786009)(316002)(52116002)(3846002)(6486002)(6116002)(4326008)(76176011)(97736004)(6862004)(486006)(476003)(478600001)(11346002)(305945005)(37006003)(5660300001)(7736002)(105586002)(2900100001)(53946003)(26005)(2906002)(99286004)(186003)(2616005)(53936002)(446003)(6512007)(6506007)(8676002)(256004)(8936002)(386003)(14454004)(14444005)(5024004)(36756003)(102836004)(81166006)(86362001)(54906003); DIR:OUT; SFP:1101; SCL:1; SRVR:AM4PR05MB3250; H:AM4PR05MB3265.eurprd05.prod.outlook.com; FPR:; SPF:None; LANG:en; PTR:InfoNoRecords; MX:1; A:1; received-spf: None (protection.outlook.com: mellanox.com does not designate permitted sender hosts) x-microsoft-antispam-message-info: zGLLh90z929IygacPLZ9llfdY2aZwrb3P06aWlov0Y4rcsIzo8Y6yoBK8ZRyojKhNXDkrllv//u9V72aqx24IMckf4YLpHvgYfpj+ye14GJnO9jPcQh4rL+BTcRi3PocLU9JJt74dejc9Rra5VO/ATgZP32ydj5xGwn2g00okX53coSUfb1bf68wBW7+NXNiwxBSUiqyMHibB33emVXEAjWB8a5JJMXqh0MtvH+9IKkd42U5YH/tLN24s6GVe7XrACwax+xSzbQKS3zl8JV/SUTESyipTm+aXHKowyrj067WTuiZvfINsCuBvPD5KNpkur8Aj6aj73j51rH08fCRCTb6ZG6RDgfGaBb+1y+fNfg= spamdiagnosticoutput: 1:99 spamdiagnosticmetadata: NSPM Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-OriginatorOrg: Mellanox.com X-MS-Exchange-CrossTenant-Network-Message-Id: 7d28c0f0-fa0e-46be-93cb-08d6415434fc X-MS-Exchange-CrossTenant-originalarrivaltime: 03 Nov 2018 06:18:44.3301 (UTC) X-MS-Exchange-CrossTenant-fromentityheader: Hosted X-MS-Exchange-CrossTenant-id: a652971c-7d2e-4d9b-a6a4-d149256f461b X-MS-Exchange-Transport-CrossTenantHeadersStamped: AM4PR05MB3250 Subject: [dpdk-dev] [PATCH v5 11/13] net/mlx5: add E-switch VXLAN tunnel devices management 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: , X-List-Received-Date: Sat, 03 Nov 2018 06:18:46 -0000 VXLAN interfaces are dynamically created for each local UDP port of outer networks and then used as targets for TC "flower" filters in order to perform encapsulation. These VXLAN interfaces are system-wide, the only one device with given UDP port can exist in the system (the attempt of creating another device with the same UDP local port returns EEXIST), so PMD should support the shared device instances database for PMD instances. These VXLAN implicitly created devices are called VTEPs (Virtual Tunnel End Points). Creation of the VTEP occurs at the moment of rule applying. The link is set up, root ingress qdisc is also initialized. Encapsulation VTEPs are created on per port basis, the single VTEP is attached to the outer interface and is shared for all encapsulation rules on this interface. The source UDP port is automatically selected in range 30000-60000. For decapsulaton one VTEP is created per every unique UDP local port to accept tunnel traffic. The name of created VTEP consists of prefix "vmlx_" and the number of UDP port in decimal digits without leading zeros (vmlx_4789). The VTEP can be preliminary created in the system before the launching application, it allows to share UDP ports between primary and secondary processes. Suggested-by: Adrien Mazarguil Signed-off-by: Viacheslav Ovsiienko Acked-by: Yongseok Koh --- drivers/net/mlx5/mlx5_flow_tcf.c | 461 +++++++++++++++++++++++++++++++++++= +++- 1 file changed, 455 insertions(+), 6 deletions(-) diff --git a/drivers/net/mlx5/mlx5_flow_tcf.c b/drivers/net/mlx5/mlx5_flow_= tcf.c index 37d9edf..252ddc0 100644 --- a/drivers/net/mlx5/mlx5_flow_tcf.c +++ b/drivers/net/mlx5/mlx5_flow_tcf.c @@ -3797,6 +3797,417 @@ struct pedit_parser { return -err; } =20 +#define MNL_BUF_EXTRA_SPACE 16 +#define MNL_REQUEST_SIZE_MIN 256 +#define MNL_REQUEST_SIZE_MAX 2048 +#define MNL_REQUEST_SIZE RTE_MIN(RTE_MAX(sysconf(_SC_PAGESIZE), \ + MNL_REQUEST_SIZE_MIN), MNL_REQUEST_SIZE_MAX) + +/* VTEP device list is shared between PMD port instances. */ +static LIST_HEAD(, tcf_vtep) vtep_list_vxlan =3D LIST_HEAD_INITIALIZER(); +static pthread_mutex_t vtep_list_mutex =3D PTHREAD_MUTEX_INITIALIZER; + +/** + * Deletes VTEP network device. + * + * @param[in] tcf + * Context object initialized by mlx5_flow_tcf_context_create(). + * @param[in] vtep + * Object represinting the network device to delete. Memory + * allocated for this object is freed by routine. + */ +static void +flow_tcf_vtep_delete(struct mlx5_flow_tcf_context *tcf, + struct tcf_vtep *vtep) +{ + struct nlmsghdr *nlh; + struct ifinfomsg *ifm; + alignas(struct nlmsghdr) + uint8_t buf[mnl_nlmsg_size(MNL_ALIGN(sizeof(*ifm))) + + MNL_BUF_EXTRA_SPACE]; + int ret; + + assert(!vtep->refcnt); + /* Delete only ifaces those we actually created. */ + if (vtep->created && vtep->ifindex) { + DRV_LOG(INFO, "VTEP delete (%d)", vtep->ifindex); + nlh =3D mnl_nlmsg_put_header(buf); + nlh->nlmsg_type =3D RTM_DELLINK; + nlh->nlmsg_flags =3D NLM_F_REQUEST; + ifm =3D mnl_nlmsg_put_extra_header(nlh, sizeof(*ifm)); + ifm->ifi_family =3D AF_UNSPEC; + ifm->ifi_index =3D vtep->ifindex; + assert(sizeof(buf) >=3D nlh->nlmsg_len); + ret =3D flow_tcf_nl_ack(tcf, nlh, 0, NULL, NULL); + if (ret) + DRV_LOG(WARNING, "netlink: error deleting vxlan" + " encap/decap ifindex %u", + ifm->ifi_index); + } + rte_free(vtep); +} + +/** + * Creates VTEP network device. + * + * @param[in] tcf + * Context object initialized by mlx5_flow_tcf_context_create(). + * @param[in] ifouter + * Outer interface to attach new-created VXLAN device + * If zero the VXLAN device will not be attached to any device. + * These VTEPs are used for decapsulation and can be precreated + * and shared between processes. + * @param[in] port + * UDP port of created VTEP device. + * @param[out] error + * Perform verbose error reporting if not NULL. + * + * @return + * Pointer to created device structure on success, + * NULL otherwise and rte_errno is set. + */ +#ifdef HAVE_IFLA_VXLAN_COLLECT_METADATA +static struct tcf_vtep* +flow_tcf_vtep_create(struct mlx5_flow_tcf_context *tcf, + unsigned int ifouter, + uint16_t port, struct rte_flow_error *error) +{ + struct tcf_vtep *vtep; + struct nlmsghdr *nlh; + struct ifinfomsg *ifm; + char name[sizeof(MLX5_VXLAN_DEVICE_PFX) + 24]; + alignas(struct nlmsghdr) + uint8_t buf[mnl_nlmsg_size(sizeof(*ifm)) + + SZ_NLATTR_DATA_OF(sizeof(name)) + + SZ_NLATTR_NEST * 2 + + SZ_NLATTR_STRZ_OF("vxlan") + + SZ_NLATTR_DATA_OF(sizeof(uint32_t)) + + SZ_NLATTR_DATA_OF(sizeof(uint16_t)) + + SZ_NLATTR_DATA_OF(sizeof(uint8_t)) * 3 + + MNL_BUF_EXTRA_SPACE]; + struct nlattr *na_info; + struct nlattr *na_vxlan; + rte_be16_t vxlan_port =3D rte_cpu_to_be_16(port); + int ret; + + vtep =3D rte_zmalloc(__func__, sizeof(*vtep), alignof(struct tcf_vtep)); + if (!vtep) { + rte_flow_error_set(error, ENOMEM, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "unable to allocate memory for VTEP"); + return NULL; + } + *vtep =3D (struct tcf_vtep){ + .port =3D port, + .local =3D LIST_HEAD_INITIALIZER(), + .neigh =3D LIST_HEAD_INITIALIZER(), + }; + memset(buf, 0, sizeof(buf)); + nlh =3D mnl_nlmsg_put_header(buf); + nlh->nlmsg_type =3D RTM_NEWLINK; + nlh->nlmsg_flags =3D NLM_F_REQUEST | NLM_F_CREATE | NLM_F_EXCL; + ifm =3D mnl_nlmsg_put_extra_header(nlh, sizeof(*ifm)); + ifm->ifi_family =3D AF_UNSPEC; + ifm->ifi_type =3D 0; + ifm->ifi_index =3D 0; + ifm->ifi_flags =3D IFF_UP; + ifm->ifi_change =3D 0xffffffff; + snprintf(name, sizeof(name), "%s%u", MLX5_VXLAN_DEVICE_PFX, port); + mnl_attr_put_strz(nlh, IFLA_IFNAME, name); + na_info =3D mnl_attr_nest_start(nlh, IFLA_LINKINFO); + assert(na_info); + mnl_attr_put_strz(nlh, IFLA_INFO_KIND, "vxlan"); + na_vxlan =3D mnl_attr_nest_start(nlh, IFLA_INFO_DATA); + if (ifouter) + mnl_attr_put_u32(nlh, IFLA_VXLAN_LINK, ifouter); + assert(na_vxlan); + mnl_attr_put_u8(nlh, IFLA_VXLAN_COLLECT_METADATA, 1); + mnl_attr_put_u8(nlh, IFLA_VXLAN_UDP_ZERO_CSUM6_RX, 1); + mnl_attr_put_u8(nlh, IFLA_VXLAN_LEARNING, 0); + mnl_attr_put_u16(nlh, IFLA_VXLAN_PORT, vxlan_port); + mnl_attr_nest_end(nlh, na_vxlan); + mnl_attr_nest_end(nlh, na_info); + assert(sizeof(buf) >=3D nlh->nlmsg_len); + ret =3D flow_tcf_nl_ack(tcf, nlh, 0, NULL, NULL); + if (ret) { + DRV_LOG(WARNING, + "netlink: VTEP %s create failure (%d)", + name, rte_errno); + if (rte_errno !=3D EEXIST || ifouter) + /* + * Some unhandled error occurred or device is + * for encapsulation and cannot be shared. + */ + goto error; + } else { + /* + * Mark device we actually created. + * We should explicitly delete + * when we do not need it anymore. + */ + vtep->created =3D 1; + } + /* Try to get ifindex of created of pre-existing device. */ + ret =3D if_nametoindex(name); + if (!ret) { + DRV_LOG(WARNING, + "VTEP %s failed to get index (%d)", name, errno); + rte_flow_error_set + (error, -errno, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "netlink: failed to retrieve VTEP ifindex"); + goto error; + } + vtep->ifindex =3D ret; + vtep->ifouter =3D ifouter; + memset(buf, 0, sizeof(buf)); + nlh =3D mnl_nlmsg_put_header(buf); + nlh->nlmsg_type =3D RTM_NEWLINK; + nlh->nlmsg_flags =3D NLM_F_REQUEST; + ifm =3D mnl_nlmsg_put_extra_header(nlh, sizeof(*ifm)); + ifm->ifi_family =3D AF_UNSPEC; + ifm->ifi_type =3D 0; + ifm->ifi_index =3D vtep->ifindex; + ifm->ifi_flags =3D IFF_UP; + ifm->ifi_change =3D IFF_UP; + ret =3D flow_tcf_nl_ack(tcf, nlh, 0, NULL, NULL); + if (ret) { + rte_flow_error_set(error, -errno, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "netlink: failed to set VTEP link up"); + DRV_LOG(WARNING, "netlink: VTEP %s set link up failure (%d)", + name, rte_errno); + goto clean; + } + ret =3D mlx5_flow_tcf_init(tcf, vtep->ifindex, error); + if (ret) { + DRV_LOG(WARNING, "VTEP %s init failure (%d)", name, rte_errno); + goto clean; + } + DRV_LOG(INFO, "VTEP create (%d, %d)", vtep->port, vtep->ifindex); + vtep->refcnt =3D 1; + return vtep; +clean: + flow_tcf_vtep_delete(tcf, vtep); + return NULL; +error: + rte_free(vtep); + return NULL; +} +#else +static struct tcf_vtep* +flow_tcf_vtep_create(struct mlx5_flow_tcf_context *tcf __rte_unused, + unsigned int ifouter __rte_unused, + uint16_t port __rte_unused, + struct rte_flow_error *error) +{ + rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "netlink: failed to create VTEP, " + "vxlan metadata are not supported by kernel"); + return NULL; +} +#endif /* HAVE_IFLA_VXLAN_COLLECT_METADATA */ + +/** + * Acquire target interface index for VXLAN tunneling decapsulation. + * In order to share the UDP port within the other interfaces the + * VXLAN device created as not attached to any interface (if created). + * + * @param[in] tcf + * Context object initialized by mlx5_flow_tcf_context_create(). + * @param[in] dev_flow + * Flow tcf object with tunnel structure pointer set. + * @param[out] error + * Perform verbose error reporting if not NULL. + * @return + * Interface descriptor pointer on success, + * NULL otherwise and rte_errno is set. + */ +static struct tcf_vtep* +flow_tcf_decap_vtep_acquire(struct mlx5_flow_tcf_context *tcf, + struct mlx5_flow *dev_flow, + struct rte_flow_error *error) +{ + struct tcf_vtep *vtep; + uint16_t port =3D dev_flow->tcf.vxlan_decap->udp_port; + + LIST_FOREACH(vtep, &vtep_list_vxlan, next) { + if (vtep->port =3D=3D port) + break; + } + if (vtep && vtep->ifouter) { + rte_flow_error_set(error, -errno, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "Failed to create decap VTEP with specified" + " UDP port, atatched device exists"); + return NULL; + } + if (vtep) { + /* Device exists, just increment the reference counter. */ + vtep->refcnt++; + assert(vtep->ifindex); + return vtep; + } + /* No decapsulation device exists, try to create the new one. */ + vtep =3D flow_tcf_vtep_create(tcf, 0, port, error); + if (vtep) + LIST_INSERT_HEAD(&vtep_list_vxlan, vtep, next); + return vtep; +} + +/** + * Aqcuire target interface index for VXLAN tunneling encapsulation. + * + * @param[in] tcf + * Context object initialized by mlx5_flow_tcf_context_create(). + * @param[in] ifouter + * Network interface index to attach VXLAN encap device to. + * @param[in] dev_flow + * Flow tcf object with tunnel structure pointer set. + * @param[out] error + * Perform verbose error reporting if not NULL. + * @return + * Interface descriptor pointer on success, + * NULL otherwise and rte_errno is set. + */ +static struct tcf_vtep* +flow_tcf_encap_vtep_acquire(struct mlx5_flow_tcf_context *tcf, + unsigned int ifouter, + struct mlx5_flow *dev_flow __rte_unused, + struct rte_flow_error *error) +{ + static uint16_t encap_port =3D MLX5_VXLAN_PORT_MIN - 1; + struct tcf_vtep *vtep; + + assert(ifouter); + /* Look whether the attached VTEP for encap is created. */ + LIST_FOREACH(vtep, &vtep_list_vxlan, next) { + if (vtep->ifouter =3D=3D ifouter) + break; + } + if (vtep) { + /* VTEP already exists, just increment the reference. */ + vtep->refcnt++; + } else { + uint16_t pcnt; + + /* Not found, we should create the new attached VTEP. */ + for (pcnt =3D 0; pcnt <=3D (MLX5_VXLAN_PORT_MAX + - MLX5_VXLAN_PORT_MIN); pcnt++) { + encap_port++; + /* Wraparound the UDP port index. */ + if (encap_port < MLX5_VXLAN_PORT_MIN || + encap_port > MLX5_VXLAN_PORT_MAX) + encap_port =3D MLX5_VXLAN_PORT_MIN; + /* Check whether UDP port is in already in use. */ + LIST_FOREACH(vtep, &vtep_list_vxlan, next) { + if (vtep->port =3D=3D encap_port) + break; + } + if (vtep) { + /* Port is in use, try the next one. */ + vtep =3D NULL; + continue; + } + vtep =3D flow_tcf_vtep_create(tcf, ifouter, + encap_port, error); + if (vtep) { + LIST_INSERT_HEAD(&vtep_list_vxlan, vtep, next); + break; + } + if (rte_errno !=3D EEXIST) + break; + } + if (!vtep) + return NULL; + } + assert(vtep->ifouter =3D=3D ifouter); + assert(vtep->ifindex); + return vtep; +} + +/** + * Acquires target interface index for tunneling of any type. + * Creates the new VTEP if needed. + * + * @param[in] tcf + * Context object initialized by mlx5_flow_tcf_context_create(). + * @param[in] ifouter + * Network interface index to attach VXLAN encap device to. + * @param[in] dev_flow + * Flow tcf object with tunnel structure pointer set. + * @param[out] error + * Perform verbose error reporting if not NULL. + * @return + * Interface descriptor pointer on success, + * NULL otherwise and rte_errno is set. + */ +static struct tcf_vtep* +flow_tcf_vtep_acquire(struct mlx5_flow_tcf_context *tcf, + unsigned int ifouter, + struct mlx5_flow *dev_flow, + struct rte_flow_error *error) +{ + struct tcf_vtep *vtep =3D NULL; + + assert(dev_flow->tcf.tunnel); + pthread_mutex_lock(&vtep_list_mutex); + switch (dev_flow->tcf.tunnel->type) { + case FLOW_TCF_TUNACT_VXLAN_ENCAP: + vtep =3D flow_tcf_encap_vtep_acquire(tcf, ifouter, + dev_flow, error); + break; + case FLOW_TCF_TUNACT_VXLAN_DECAP: + vtep =3D flow_tcf_decap_vtep_acquire(tcf, dev_flow, error); + break; + default: + rte_flow_error_set(error, ENOTSUP, + RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, + "unsupported tunnel type"); + break; + } + pthread_mutex_unlock(&vtep_list_mutex); + return vtep; +} + +/** + * Release tunneling interface by ifindex. Decrements reference + * counter and actually removes the device if counter is zero. + * + * @param[in] tcf + * Context object initialized by mlx5_flow_tcf_context_create(). + * @param[in] vtep + * VTEP device descriptor structure. + * @param[in] dev_flow + * Flow tcf object with tunnel structure pointer set. + */ +static void +flow_tcf_vtep_release(struct mlx5_flow_tcf_context *tcf, + struct tcf_vtep *vtep, + struct mlx5_flow *dev_flow) +{ + assert(dev_flow->tcf.tunnel); + pthread_mutex_lock(&vtep_list_mutex); + switch (dev_flow->tcf.tunnel->type) { + case FLOW_TCF_TUNACT_VXLAN_DECAP: + break; + case FLOW_TCF_TUNACT_VXLAN_ENCAP: + break; + default: + assert(false); + DRV_LOG(WARNING, "Unsupported tunnel type"); + break; + } + assert(vtep->refcnt); + if (--vtep->refcnt =3D=3D 0) { + LIST_REMOVE(vtep, next); + flow_tcf_vtep_delete(tcf, vtep); + } + pthread_mutex_unlock(&vtep_list_mutex); +} + + /** * Apply flow to E-Switch by sending Netlink message. * @@ -3822,11 +4233,35 @@ struct pedit_parser { dev_flow =3D LIST_FIRST(&flow->dev_flows); /* E-Switch flow can't be expanded. */ assert(!LIST_NEXT(dev_flow, next)); + if (dev_flow->tcf.applied) + return 0; nlh =3D dev_flow->tcf.nlh; nlh->nlmsg_type =3D RTM_NEWTFILTER; nlh->nlmsg_flags =3D NLM_F_REQUEST | NLM_F_CREATE | NLM_F_EXCL; - if (!flow_tcf_nl_ack(ctx, nlh, 0, NULL, NULL)) + if (dev_flow->tcf.tunnel) { + /* + * Replace the interface index, target for + * encapsulation, source for decapsulation. + */ + assert(!dev_flow->tcf.tunnel->vtep); + assert(dev_flow->tcf.tunnel->ifindex_ptr); + /* Acquire actual VTEP device when rule is being applied. */ + dev_flow->tcf.tunnel->vtep =3D + flow_tcf_vtep_acquire(ctx, + dev_flow->tcf.tunnel->ifindex_org, + dev_flow, error); + if (!dev_flow->tcf.tunnel->vtep) + return -rte_errno; + DRV_LOG(INFO, "Replace ifindex: %d->%d", + dev_flow->tcf.tunnel->vtep->ifindex, + dev_flow->tcf.tunnel->ifindex_org); + *dev_flow->tcf.tunnel->ifindex_ptr =3D + dev_flow->tcf.tunnel->vtep->ifindex; + } + if (!flow_tcf_nl_ack(ctx, nlh, 0, NULL, NULL)) { + dev_flow->tcf.applied =3D 1; return 0; + } return rte_flow_error_set(error, rte_errno, RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, "netlink: failed to create TC flow rule"); @@ -3855,10 +4290,20 @@ struct pedit_parser { return; /* E-Switch flow can't be expanded. */ assert(!LIST_NEXT(dev_flow, next)); - nlh =3D dev_flow->tcf.nlh; - nlh->nlmsg_type =3D RTM_DELTFILTER; - nlh->nlmsg_flags =3D NLM_F_REQUEST; - flow_tcf_nl_ack(ctx, nlh, 0, NULL, NULL); + if (dev_flow->tcf.applied) { + nlh =3D dev_flow->tcf.nlh; + nlh->nlmsg_type =3D RTM_DELTFILTER; + nlh->nlmsg_flags =3D NLM_F_REQUEST; + flow_tcf_nl_ack(ctx, nlh, 0, NULL, NULL); + if (dev_flow->tcf.tunnel) { + assert(dev_flow->tcf.tunnel->vtep); + flow_tcf_vtep_release(ctx, + dev_flow->tcf.tunnel->vtep, + dev_flow); + dev_flow->tcf.tunnel->vtep =3D NULL; + } + dev_flow->tcf.applied =3D 0; + } } =20 /** @@ -4385,7 +4830,9 @@ struct pedit_parser { struct nlmsghdr *nlh; struct tcmsg *tcm; alignas(struct nlmsghdr) - uint8_t buf[mnl_nlmsg_size(sizeof(*tcm) + 128)]; + uint8_t buf[mnl_nlmsg_size(sizeof(*tcm)) + + SZ_NLATTR_STRZ_OF("ingress") + + MNL_BUF_EXTRA_SPACE]; =20 /* Destroy existing ingress qdisc and everything attached to it. */ nlh =3D mnl_nlmsg_put_header(buf); @@ -4396,6 +4843,7 @@ struct pedit_parser { tcm->tcm_ifindex =3D ifindex; tcm->tcm_handle =3D TC_H_MAKE(TC_H_INGRESS, 0); tcm->tcm_parent =3D TC_H_INGRESS; + assert(sizeof(buf) >=3D nlh->nlmsg_len); /* Ignore errors when qdisc is already absent. */ if (flow_tcf_nl_ack(ctx, nlh, 0, NULL, NULL) && rte_errno !=3D EINVAL && rte_errno !=3D ENOENT) @@ -4413,6 +4861,7 @@ struct pedit_parser { tcm->tcm_handle =3D TC_H_MAKE(TC_H_INGRESS, 0); tcm->tcm_parent =3D TC_H_INGRESS; mnl_attr_put_strz_check(nlh, sizeof(buf), TCA_KIND, "ingress"); + assert(sizeof(buf) >=3D nlh->nlmsg_len); if (flow_tcf_nl_ack(ctx, nlh, 0, NULL, NULL)) return rte_flow_error_set(error, rte_errno, RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, --=20 1.8.3.1