From 5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Tue, 21 Feb 2023 18:24:12 -0800 Subject: Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ... --- drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c | 1056 +++++++++++++++++++++++ 1 file changed, 1056 insertions(+) create mode 100644 drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c (limited to 'drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c') diff --git a/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c new file mode 100644 index 000000000..21c6cea8b --- /dev/null +++ b/drivers/infiniband/ulp/opa_vnic/opa_vnic_vema.c @@ -0,0 +1,1056 @@ +/* + * Copyright(c) 2017 Intel Corporation. + * Copyright(c) 2021 Cornelis Networks. + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * 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. + * + */ + +/* + * This file contains OPX Virtual Network Interface Controller (VNIC) + * Ethernet Management Agent (EMA) driver + */ + +#include +#include +#include +#include +#include +#include + +#include "opa_vnic_internal.h" + +char opa_vnic_driver_name[] = "opa_vnic"; + +/* + * The trap service level is kept in bits 3 to 7 in the trap_sl_rsvd + * field in the class port info MAD. + */ +#define GET_TRAP_SL_FROM_CLASS_PORT_INFO(x) (((x) >> 3) & 0x1f) + +/* Cap trap bursts to a reasonable limit good for normal cases */ +#define OPA_VNIC_TRAP_BURST_LIMIT 4 + +/* + * VNIC trap limit timeout. + * Inverse of cap2_mask response time out (1.0737 secs) = 0.9 + * secs approx IB spec 13.4.6.2.1 PortInfoSubnetTimeout and + * 13.4.9 Traps. + */ +#define OPA_VNIC_TRAP_TIMEOUT ((4096 * (1UL << 18)) / 1000) + +#define OPA_VNIC_UNSUP_ATTR \ + cpu_to_be16(IB_MGMT_MAD_STATUS_UNSUPPORTED_METHOD_ATTRIB) + +#define OPA_VNIC_INVAL_ATTR \ + cpu_to_be16(IB_MGMT_MAD_STATUS_INVALID_ATTRIB_VALUE) + +#define OPA_VNIC_CLASS_CAP_TRAP 0x1 + +/* Maximum number of VNIC ports supported */ +#define OPA_VNIC_MAX_NUM_VPORT 255 + +/** + * struct opa_vnic_vema_port -- VNIC VEMA port details + * @cport: pointer to port + * @mad_agent: pointer to mad agent for port + * @class_port_info: Class port info information. + * @tid: Transaction id + * @port_num: OPA port number + * @vports: vnic ports + * @event_handler: ib event handler + * @lock: adapter interface lock + */ +struct opa_vnic_vema_port { + struct opa_vnic_ctrl_port *cport; + struct ib_mad_agent *mad_agent; + struct opa_class_port_info class_port_info; + u64 tid; + u8 port_num; + struct xarray vports; + struct ib_event_handler event_handler; + + /* Lock to query/update network adapter */ + struct mutex lock; +}; + +static int opa_vnic_vema_add_one(struct ib_device *device); +static void opa_vnic_vema_rem_one(struct ib_device *device, + void *client_data); + +static struct ib_client opa_vnic_client = { + .name = opa_vnic_driver_name, + .add = opa_vnic_vema_add_one, + .remove = opa_vnic_vema_rem_one, +}; + +/** + * vema_get_vport_num -- Get the vnic from the mad + * @recvd_mad: Received mad + * + * Return: returns value of the vnic port number + */ +static inline u8 vema_get_vport_num(struct opa_vnic_vema_mad *recvd_mad) +{ + return be32_to_cpu(recvd_mad->mad_hdr.attr_mod) & 0xff; +} + +/** + * vema_get_vport_adapter -- Get vnic port adapter from recvd mad + * @recvd_mad: received mad + * @port: ptr to port struct on which MAD was recvd + * + * Return: vnic adapter + */ +static inline struct opa_vnic_adapter * +vema_get_vport_adapter(struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_port *port) +{ + u8 vport_num = vema_get_vport_num(recvd_mad); + + return xa_load(&port->vports, vport_num); +} + +/** + * vema_mac_tbl_req_ok -- Check if mac request has correct values + * @mac_tbl: mac table + * + * This function checks for the validity of the offset and number of + * entries required. + * + * Return: true if offset and num_entries are valid + */ +static inline bool vema_mac_tbl_req_ok(struct opa_veswport_mactable *mac_tbl) +{ + u16 offset, num_entries; + u16 req_entries = ((OPA_VNIC_EMA_DATA - sizeof(*mac_tbl)) / + sizeof(mac_tbl->tbl_entries[0])); + + offset = be16_to_cpu(mac_tbl->offset); + num_entries = be16_to_cpu(mac_tbl->num_entries); + + return ((num_entries <= req_entries) && + (offset + num_entries <= OPA_VNIC_MAC_TBL_MAX_ENTRIES)); +} + +/* + * Return the power on default values in the port info structure + * in big endian format as required by MAD. + */ +static inline void vema_get_pod_values(struct opa_veswport_info *port_info) +{ + memset(port_info, 0, sizeof(*port_info)); + port_info->vport.max_mac_tbl_ent = + cpu_to_be16(OPA_VNIC_MAC_TBL_MAX_ENTRIES); + port_info->vport.max_smac_ent = + cpu_to_be16(OPA_VNIC_MAX_SMAC_LIMIT); + port_info->vport.oper_state = OPA_VNIC_STATE_DROP_ALL; + port_info->vport.config_state = OPA_VNIC_STATE_DROP_ALL; + port_info->vesw.eth_mtu = cpu_to_be16(ETH_DATA_LEN); +} + +/** + * vema_add_vport -- Add a new vnic port + * @port: ptr to opa_vnic_vema_port struct + * @vport_num: vnic port number (to be added) + * + * Return a pointer to the vnic adapter structure + */ +static struct opa_vnic_adapter *vema_add_vport(struct opa_vnic_vema_port *port, + u8 vport_num) +{ + struct opa_vnic_ctrl_port *cport = port->cport; + struct opa_vnic_adapter *adapter; + + adapter = opa_vnic_add_netdev(cport->ibdev, port->port_num, vport_num); + if (!IS_ERR(adapter)) { + int rc; + + adapter->cport = cport; + rc = xa_insert(&port->vports, vport_num, adapter, GFP_KERNEL); + if (rc < 0) { + opa_vnic_rem_netdev(adapter); + adapter = ERR_PTR(rc); + } + } + + return adapter; +} + +/** + * vema_get_class_port_info -- Get class info for port + * @port: Port on whic MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function copies the latest class port info value set for the + * port and stores it for generating traps + */ +static void vema_get_class_port_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_class_port_info *port_info; + + port_info = (struct opa_class_port_info *)rsp_mad->data; + memcpy(port_info, &port->class_port_info, sizeof(*port_info)); + port_info->base_version = OPA_MGMT_BASE_VERSION; + port_info->class_version = OPA_EMA_CLASS_VERSION; + + /* + * Set capability mask bit indicating agent generates traps, + * and set the maximum number of VNIC ports supported. + */ + port_info->cap_mask = cpu_to_be16((OPA_VNIC_CLASS_CAP_TRAP | + (OPA_VNIC_MAX_NUM_VPORT << 8))); + + /* + * Since a get routine is always sent by the EM first we + * set the expected response time to + * 4.096 usec * 2^18 == 1.0737 sec here. + */ + port_info->cap_mask2_resp_time = cpu_to_be32(18); +} + +/** + * vema_set_class_port_info -- Get class info for port + * @port: Port on whic MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function updates the port class info for the specific vnic + * and sets up the response mad data + */ +static void vema_set_class_port_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + memcpy(&port->class_port_info, recvd_mad->data, + sizeof(port->class_port_info)); + + vema_get_class_port_info(port, recvd_mad, rsp_mad); +} + +/** + * vema_get_veswport_info -- Get veswport info + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + */ +static void vema_get_veswport_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_info *port_info = + (struct opa_veswport_info *)rsp_mad->data; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (adapter) { + memset(port_info, 0, sizeof(*port_info)); + opa_vnic_get_vesw_info(adapter, &port_info->vesw); + opa_vnic_get_per_veswport_info(adapter, + &port_info->vport); + } else { + vema_get_pod_values(port_info); + } +} + +/** + * vema_set_veswport_info -- Set veswport info + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function gets the port class infor for vnic + */ +static void vema_set_veswport_info(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_vnic_ctrl_port *cport = port->cport; + struct opa_veswport_info *port_info; + struct opa_vnic_adapter *adapter; + u8 vport_num; + + vport_num = vema_get_vport_num(recvd_mad); + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + adapter = vema_add_vport(port, vport_num); + if (IS_ERR(adapter)) { + c_err("failed to add vport %d: %ld\n", + vport_num, PTR_ERR(adapter)); + goto err_exit; + } + } + + port_info = (struct opa_veswport_info *)recvd_mad->data; + opa_vnic_set_vesw_info(adapter, &port_info->vesw); + opa_vnic_set_per_veswport_info(adapter, &port_info->vport); + + /* Process the new config settings */ + opa_vnic_process_vema_config(adapter); + + vema_get_veswport_info(port, recvd_mad, rsp_mad); + return; + +err_exit: + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; +} + +/** + * vema_get_mac_entries -- Get MAC entries in VNIC MAC table + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function gets the MAC entries that are programmed into + * the VNIC MAC forwarding table. It checks for the validity of + * the index into the MAC table and the number of entries that + * are to be retrieved. + */ +static void vema_get_mac_entries(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_mactable *mac_tbl_in, *mac_tbl_out; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + mac_tbl_in = (struct opa_veswport_mactable *)recvd_mad->data; + mac_tbl_out = (struct opa_veswport_mactable *)rsp_mad->data; + + if (vema_mac_tbl_req_ok(mac_tbl_in)) { + mac_tbl_out->offset = mac_tbl_in->offset; + mac_tbl_out->num_entries = mac_tbl_in->num_entries; + opa_vnic_query_mac_tbl(adapter, mac_tbl_out); + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + } +} + +/** + * vema_set_mac_entries -- Set MAC entries in VNIC MAC table + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function sets the MAC entries in the VNIC forwarding table + * It checks for the validity of the index and the number of forwarding + * table entries to be programmed. + */ +static void vema_set_mac_entries(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_mactable *mac_tbl; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + mac_tbl = (struct opa_veswport_mactable *)recvd_mad->data; + if (vema_mac_tbl_req_ok(mac_tbl)) { + if (opa_vnic_update_mac_tbl(adapter, mac_tbl)) + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + } + vema_get_mac_entries(port, recvd_mad, rsp_mad); +} + +/** + * vema_set_delete_vesw -- Reset VESW info to POD values + * @port: source port on which MAD was received + * @recvd_mad: pointer to the received mad + * @rsp_mad: pointer to respose mad + * + * This function clears all the fields of veswport info for the requested vesw + * and sets them back to the power-on default values. It does not delete the + * vesw. + */ +static void vema_set_delete_vesw(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_info *port_info = + (struct opa_veswport_info *)rsp_mad->data; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + vema_get_pod_values(port_info); + opa_vnic_set_vesw_info(adapter, &port_info->vesw); + opa_vnic_set_per_veswport_info(adapter, &port_info->vport); + + /* Process the new config settings */ + opa_vnic_process_vema_config(adapter); + + opa_vnic_release_mac_tbl(adapter); + + vema_get_veswport_info(port, recvd_mad, rsp_mad); +} + +/** + * vema_get_mac_list -- Get the unicast/multicast macs. + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + * @attr_id: Attribute ID indicating multicast or unicast mac list + */ +static void vema_get_mac_list(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad, + u16 attr_id) +{ + struct opa_veswport_iface_macs *macs_in, *macs_out; + int max_entries = (OPA_VNIC_EMA_DATA - sizeof(*macs_out)) / ETH_ALEN; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (!adapter) { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + return; + } + + macs_in = (struct opa_veswport_iface_macs *)recvd_mad->data; + macs_out = (struct opa_veswport_iface_macs *)rsp_mad->data; + + macs_out->start_idx = macs_in->start_idx; + if (macs_in->num_macs_in_msg) + macs_out->num_macs_in_msg = macs_in->num_macs_in_msg; + else + macs_out->num_macs_in_msg = cpu_to_be16(max_entries); + + if (attr_id == OPA_EM_ATTR_IFACE_MCAST_MACS) + opa_vnic_query_mcast_macs(adapter, macs_out); + else + opa_vnic_query_ucast_macs(adapter, macs_out); +} + +/** + * vema_get_summary_counters -- Gets summary counters. + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + */ +static void vema_get_summary_counters(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_summary_counters *cntrs; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (adapter) { + cntrs = (struct opa_veswport_summary_counters *)rsp_mad->data; + opa_vnic_get_summary_counters(adapter, cntrs); + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + } +} + +/** + * vema_get_error_counters -- Gets summary counters. + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + */ +static void vema_get_error_counters(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + struct opa_veswport_error_counters *cntrs; + struct opa_vnic_adapter *adapter; + + adapter = vema_get_vport_adapter(recvd_mad, port); + if (adapter) { + cntrs = (struct opa_veswport_error_counters *)rsp_mad->data; + opa_vnic_get_error_counters(adapter, cntrs); + } else { + rsp_mad->mad_hdr.status = OPA_VNIC_INVAL_ATTR; + } +} + +/** + * vema_get -- Process received get MAD + * @port: source port on which MAD was received + * @recvd_mad: Received mad + * @rsp_mad: Response mad to be built + */ +static void vema_get(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + u16 attr_id = be16_to_cpu(recvd_mad->mad_hdr.attr_id); + + switch (attr_id) { + case OPA_EM_ATTR_CLASS_PORT_INFO: + vema_get_class_port_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_INFO: + vema_get_veswport_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_MAC_ENTRIES: + vema_get_mac_entries(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_IFACE_UCAST_MACS: + case OPA_EM_ATTR_IFACE_MCAST_MACS: + vema_get_mac_list(port, recvd_mad, rsp_mad, attr_id); + break; + case OPA_EM_ATTR_VESWPORT_SUMMARY_COUNTERS: + vema_get_summary_counters(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_ERROR_COUNTERS: + vema_get_error_counters(port, recvd_mad, rsp_mad); + break; + default: + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + break; + } +} + +/** + * vema_set -- Process received set MAD + * @port: source port on which MAD was received + * @recvd_mad: Received mad contains fields to set vnic parameters + * @rsp_mad: Response mad to be built + */ +static void vema_set(struct opa_vnic_vema_port *port, + struct opa_vnic_vema_mad *recvd_mad, + struct opa_vnic_vema_mad *rsp_mad) +{ + u16 attr_id = be16_to_cpu(recvd_mad->mad_hdr.attr_id); + + switch (attr_id) { + case OPA_EM_ATTR_CLASS_PORT_INFO: + vema_set_class_port_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_INFO: + vema_set_veswport_info(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_VESWPORT_MAC_ENTRIES: + vema_set_mac_entries(port, recvd_mad, rsp_mad); + break; + case OPA_EM_ATTR_DELETE_VESW: + vema_set_delete_vesw(port, recvd_mad, rsp_mad); + break; + default: + rsp_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + break; + } +} + +/** + * vema_send -- Send handler for VEMA MAD agent + * @mad_agent: pointer to the mad agent + * @mad_wc: pointer to mad send work completion information + * + * Free all the data structures associated with the sent MAD + */ +static void vema_send(struct ib_mad_agent *mad_agent, + struct ib_mad_send_wc *mad_wc) +{ + rdma_destroy_ah(mad_wc->send_buf->ah, RDMA_DESTROY_AH_SLEEPABLE); + ib_free_send_mad(mad_wc->send_buf); +} + +/** + * vema_recv -- Recv handler for VEMA MAD agent + * @mad_agent: pointer to the mad agent + * @send_buf: Send buffer if found, else NULL + * @mad_wc: pointer to mad send work completion information + * + * Handle only set and get methods and respond to other methods + * as unsupported. Allocate response buffer and address handle + * for the response MAD. + */ +static void vema_recv(struct ib_mad_agent *mad_agent, + struct ib_mad_send_buf *send_buf, + struct ib_mad_recv_wc *mad_wc) +{ + struct opa_vnic_vema_port *port; + struct ib_ah *ah; + struct ib_mad_send_buf *rsp; + struct opa_vnic_vema_mad *vema_mad; + + if (!mad_wc || !mad_wc->recv_buf.mad) + return; + + port = mad_agent->context; + ah = ib_create_ah_from_wc(mad_agent->qp->pd, mad_wc->wc, + mad_wc->recv_buf.grh, mad_agent->port_num); + if (IS_ERR(ah)) + goto free_recv_mad; + + rsp = ib_create_send_mad(mad_agent, mad_wc->wc->src_qp, + mad_wc->wc->pkey_index, 0, + IB_MGMT_VENDOR_HDR, OPA_VNIC_EMA_DATA, + GFP_KERNEL, OPA_MGMT_BASE_VERSION); + if (IS_ERR(rsp)) + goto err_rsp; + + rsp->ah = ah; + vema_mad = rsp->mad; + memcpy(vema_mad, mad_wc->recv_buf.mad, IB_MGMT_VENDOR_HDR); + vema_mad->mad_hdr.method = IB_MGMT_METHOD_GET_RESP; + vema_mad->mad_hdr.status = 0; + + /* Lock ensures network adapter is not removed */ + mutex_lock(&port->lock); + + switch (mad_wc->recv_buf.mad->mad_hdr.method) { + case IB_MGMT_METHOD_GET: + vema_get(port, (struct opa_vnic_vema_mad *)mad_wc->recv_buf.mad, + vema_mad); + break; + case IB_MGMT_METHOD_SET: + vema_set(port, (struct opa_vnic_vema_mad *)mad_wc->recv_buf.mad, + vema_mad); + break; + default: + vema_mad->mad_hdr.status = OPA_VNIC_UNSUP_ATTR; + break; + } + mutex_unlock(&port->lock); + + if (!ib_post_send_mad(rsp, NULL)) { + /* + * with post send successful ah and send mad + * will be destroyed in send handler + */ + goto free_recv_mad; + } + + ib_free_send_mad(rsp); + +err_rsp: + rdma_destroy_ah(ah, RDMA_DESTROY_AH_SLEEPABLE); +free_recv_mad: + ib_free_recv_mad(mad_wc); +} + +/** + * vema_get_port -- Gets the opa_vnic_vema_port + * @cport: pointer to control dev + * @port_num: Port number + * + * This function loops through the ports and returns + * the opa_vnic_vema port structure that is associated + * with the OPA port number + * + * Return: ptr to requested opa_vnic_vema_port strucure + * if success, NULL if not + */ +static struct opa_vnic_vema_port * +vema_get_port(struct opa_vnic_ctrl_port *cport, u8 port_num) +{ + struct opa_vnic_vema_port *port = (void *)cport + sizeof(*cport); + + if (port_num > cport->num_ports) + return NULL; + + return port + (port_num - 1); +} + +/** + * opa_vnic_vema_send_trap -- This function sends a trap to the EM + * @adapter: pointer to vnic adapter + * @data: pointer to trap data filled by calling function + * @lid: issuers lid (encap_slid from vesw_port_info) + * + * This function is called from the VNIC driver to send a trap if there + * is somethng the EM should be notified about. These events currently + * are + * 1) UNICAST INTERFACE MACADDRESS changes + * 2) MULTICAST INTERFACE MACADDRESS changes + * 3) ETHERNET LINK STATUS changes + * While allocating the send mad the remote site qpn used is 1 + * as this is the well known QP. + * + */ +void opa_vnic_vema_send_trap(struct opa_vnic_adapter *adapter, + struct __opa_veswport_trap *data, u32 lid) +{ + struct opa_vnic_ctrl_port *cport = adapter->cport; + struct ib_mad_send_buf *send_buf; + struct opa_vnic_vema_port *port; + struct ib_device *ibp; + struct opa_vnic_vema_mad_trap *trap_mad; + struct opa_class_port_info *class; + struct rdma_ah_attr ah_attr; + struct ib_ah *ah; + struct opa_veswport_trap *trap; + u32 trap_lid; + u16 pkey_idx; + + if (!cport) + goto err_exit; + ibp = cport->ibdev; + port = vema_get_port(cport, data->opaportnum); + if (!port || !port->mad_agent) + goto err_exit; + + if (time_before(jiffies, adapter->trap_timeout)) { + if (adapter->trap_count == OPA_VNIC_TRAP_BURST_LIMIT) { + v_warn("Trap rate exceeded\n"); + goto err_exit; + } else { + adapter->trap_count++; + } + } else { + adapter->trap_count = 0; + } + + class = &port->class_port_info; + /* Set up address handle */ + memset(&ah_attr, 0, sizeof(ah_attr)); + ah_attr.type = rdma_ah_find_type(ibp, port->port_num); + rdma_ah_set_sl(&ah_attr, + GET_TRAP_SL_FROM_CLASS_PORT_INFO(class->trap_sl_rsvd)); + rdma_ah_set_port_num(&ah_attr, port->port_num); + trap_lid = be32_to_cpu(class->trap_lid); + /* + * check for trap lid validity, must not be zero + * The trap sink could change after we fashion the MAD but since traps + * are not guaranteed we won't use a lock as anyway the change will take + * place even with locking. + */ + if (!trap_lid) { + c_err("%s: Invalid dlid\n", __func__); + goto err_exit; + } + + rdma_ah_set_dlid(&ah_attr, trap_lid); + ah = rdma_create_ah(port->mad_agent->qp->pd, &ah_attr, 0); + if (IS_ERR(ah)) { + c_err("%s:Couldn't create new AH = %p\n", __func__, ah); + c_err("%s:dlid = %d, sl = %d, port = %d\n", __func__, + rdma_ah_get_dlid(&ah_attr), rdma_ah_get_sl(&ah_attr), + rdma_ah_get_port_num(&ah_attr)); + goto err_exit; + } + + if (ib_find_pkey(ibp, data->opaportnum, IB_DEFAULT_PKEY_FULL, + &pkey_idx) < 0) { + c_err("%s:full key not found, defaulting to partial\n", + __func__); + if (ib_find_pkey(ibp, data->opaportnum, IB_DEFAULT_PKEY_PARTIAL, + &pkey_idx) < 0) + pkey_idx = 1; + } + + send_buf = ib_create_send_mad(port->mad_agent, 1, pkey_idx, 0, + IB_MGMT_VENDOR_HDR, IB_MGMT_MAD_DATA, + GFP_ATOMIC, OPA_MGMT_BASE_VERSION); + if (IS_ERR(send_buf)) { + c_err("%s:Couldn't allocate send buf\n", __func__); + goto err_sndbuf; + } + + send_buf->ah = ah; + + /* Set up common MAD hdr */ + trap_mad = send_buf->mad; + trap_mad->mad_hdr.base_version = OPA_MGMT_BASE_VERSION; + trap_mad->mad_hdr.mgmt_class = OPA_MGMT_CLASS_INTEL_EMA; + trap_mad->mad_hdr.class_version = OPA_EMA_CLASS_VERSION; + trap_mad->mad_hdr.method = IB_MGMT_METHOD_TRAP; + port->tid++; + trap_mad->mad_hdr.tid = cpu_to_be64(port->tid); + trap_mad->mad_hdr.attr_id = IB_SMP_ATTR_NOTICE; + + /* Set up vendor OUI */ + trap_mad->oui[0] = INTEL_OUI_1; + trap_mad->oui[1] = INTEL_OUI_2; + trap_mad->oui[2] = INTEL_OUI_3; + + /* Setup notice attribute portion */ + trap_mad->notice.gen_type = OPA_INTEL_EMA_NOTICE_TYPE_INFO << 1; + trap_mad->notice.oui_1 = INTEL_OUI_1; + trap_mad->notice.oui_2 = INTEL_OUI_2; + trap_mad->notice.oui_3 = INTEL_OUI_3; + trap_mad->notice.issuer_lid = cpu_to_be32(lid); + + /* copy the actual trap data */ + trap = (struct opa_veswport_trap *)trap_mad->notice.raw_data; + trap->fabric_id = cpu_to_be16(data->fabric_id); + trap->veswid = cpu_to_be16(data->veswid); + trap->veswportnum = cpu_to_be32(data->veswportnum); + trap->opaportnum = cpu_to_be16(data->opaportnum); + trap->veswportindex = data->veswportindex; + trap->opcode = data->opcode; + + /* If successful send set up rate limit timeout else bail */ + if (ib_post_send_mad(send_buf, NULL)) { + ib_free_send_mad(send_buf); + } else { + if (adapter->trap_count) + return; + adapter->trap_timeout = jiffies + + usecs_to_jiffies(OPA_VNIC_TRAP_TIMEOUT); + return; + } + +err_sndbuf: + rdma_destroy_ah(ah, 0); +err_exit: + v_err("Aborting trap\n"); +} + +static void opa_vnic_event(struct ib_event_handler *handler, + struct ib_event *record) +{ + struct opa_vnic_vema_port *port = + container_of(handler, struct opa_vnic_vema_port, event_handler); + struct opa_vnic_ctrl_port *cport = port->cport; + struct opa_vnic_adapter *adapter; + unsigned long index; + + if (record->element.port_num != port->port_num) + return; + + c_dbg("OPA_VNIC received event %d on device %s port %d\n", + record->event, dev_name(&record->device->dev), + record->element.port_num); + + if (record->event != IB_EVENT_PORT_ERR && + record->event != IB_EVENT_PORT_ACTIVE) + return; + + xa_for_each(&port->vports, index, adapter) { + if (record->event == IB_EVENT_PORT_ACTIVE) + netif_carrier_on(adapter->netdev); + else + netif_carrier_off(adapter->netdev); + } +} + +/** + * vema_unregister -- Unregisters agent + * @cport: pointer to control port + * + * This deletes the registration by VEMA for MADs + */ +static void vema_unregister(struct opa_vnic_ctrl_port *cport) +{ + struct opa_vnic_adapter *adapter; + unsigned long index; + int i; + + for (i = 1; i <= cport->num_ports; i++) { + struct opa_vnic_vema_port *port = vema_get_port(cport, i); + + if (!port->mad_agent) + continue; + + /* Lock ensures no MAD is being processed */ + mutex_lock(&port->lock); + xa_for_each(&port->vports, index, adapter) + opa_vnic_rem_netdev(adapter); + mutex_unlock(&port->lock); + + ib_unregister_mad_agent(port->mad_agent); + port->mad_agent = NULL; + mutex_destroy(&port->lock); + xa_destroy(&port->vports); + ib_unregister_event_handler(&port->event_handler); + } +} + +/** + * vema_register -- Registers agent + * @cport: pointer to control port + * + * This function registers the handlers for the VEMA MADs + * + * Return: returns 0 on success. non zero otherwise + */ +static int vema_register(struct opa_vnic_ctrl_port *cport) +{ + struct ib_mad_reg_req reg_req = { + .mgmt_class = OPA_MGMT_CLASS_INTEL_EMA, + .mgmt_class_version = OPA_MGMT_BASE_VERSION, + .oui = { INTEL_OUI_1, INTEL_OUI_2, INTEL_OUI_3 } + }; + int i; + + set_bit(IB_MGMT_METHOD_GET, reg_req.method_mask); + set_bit(IB_MGMT_METHOD_SET, reg_req.method_mask); + + /* register ib event handler and mad agent for each port on dev */ + for (i = 1; i <= cport->num_ports; i++) { + struct opa_vnic_vema_port *port = vema_get_port(cport, i); + int ret; + + port->cport = cport; + port->port_num = i; + + INIT_IB_EVENT_HANDLER(&port->event_handler, + cport->ibdev, opa_vnic_event); + ib_register_event_handler(&port->event_handler); + + xa_init(&port->vports); + mutex_init(&port->lock); + port->mad_agent = ib_register_mad_agent(cport->ibdev, i, + IB_QPT_GSI, ®_req, + IB_MGMT_RMPP_VERSION, + vema_send, vema_recv, + port, 0); + if (IS_ERR(port->mad_agent)) { + ret = PTR_ERR(port->mad_agent); + port->mad_agent = NULL; + mutex_destroy(&port->lock); + vema_unregister(cport); + return ret; + } + } + + return 0; +} + +/** + * opa_vnic_ctrl_config_dev -- This function sends a trap to the EM + * by way of ib_modify_port to indicate support for ethernet on the + * fabric. + * @cport: pointer to control port + * @en: enable or disable ethernet on fabric support + */ +static void opa_vnic_ctrl_config_dev(struct opa_vnic_ctrl_port *cport, bool en) +{ + struct ib_port_modify pm = { 0 }; + int i; + + if (en) + pm.set_port_cap_mask = OPA_CAP_MASK3_IsEthOnFabricSupported; + else + pm.clr_port_cap_mask = OPA_CAP_MASK3_IsEthOnFabricSupported; + + for (i = 1; i <= cport->num_ports; i++) + ib_modify_port(cport->ibdev, i, IB_PORT_OPA_MASK_CHG, &pm); +} + +/** + * opa_vnic_vema_add_one -- Handle new ib device + * @device: ib device pointer + * + * Allocate the vnic control port and initialize it. + */ +static int opa_vnic_vema_add_one(struct ib_device *device) +{ + struct opa_vnic_ctrl_port *cport; + int rc, size = sizeof(*cport); + + if (!rdma_cap_opa_vnic(device)) + return -EOPNOTSUPP; + + size += device->phys_port_cnt * sizeof(struct opa_vnic_vema_port); + cport = kzalloc(size, GFP_KERNEL); + if (!cport) + return -ENOMEM; + + cport->num_ports = device->phys_port_cnt; + cport->ibdev = device; + + /* Initialize opa vnic management agent (vema) */ + rc = vema_register(cport); + if (!rc) + c_info("VNIC client initialized\n"); + + ib_set_client_data(device, &opa_vnic_client, cport); + opa_vnic_ctrl_config_dev(cport, true); + return 0; +} + +/** + * opa_vnic_vema_rem_one -- Handle ib device removal + * @device: ib device pointer + * @client_data: ib client data + * + * Uninitialize and free the vnic control port. + */ +static void opa_vnic_vema_rem_one(struct ib_device *device, + void *client_data) +{ + struct opa_vnic_ctrl_port *cport = client_data; + + c_info("removing VNIC client\n"); + opa_vnic_ctrl_config_dev(cport, false); + vema_unregister(cport); + kfree(cport); +} + +static int __init opa_vnic_init(void) +{ + int rc; + + rc = ib_register_client(&opa_vnic_client); + if (rc) + pr_err("VNIC driver register failed %d\n", rc); + + return rc; +} +module_init(opa_vnic_init); + +static void opa_vnic_deinit(void) +{ + ib_unregister_client(&opa_vnic_client); +} +module_exit(opa_vnic_deinit); + +MODULE_LICENSE("Dual BSD/GPL"); +MODULE_AUTHOR("Cornelis Networks"); +MODULE_DESCRIPTION("Cornelis OPX Virtual Network driver"); -- cgit v1.2.3