diff --git a/sys/dev/irdma/icrdma.c b/sys/dev/irdma/icrdma.c index 576a185b013f..a4f3904a820c 100644 --- a/sys/dev/irdma/icrdma.c +++ b/sys/dev/irdma/icrdma.c @@ -1,816 +1,816 @@ /*- * SPDX-License-Identifier: GPL-2.0 or Linux-OpenIB * - * Copyright (c) 2021 - 2023 Intel Corporation + * Copyright (c) 2021 - 2025 Intel Corporation * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU * General Public License (GPL) Version 2, available from the file * COPYING in the main directory of this source tree, or the * OpenFabrics.org BSD license below: * * 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. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #include #include #include #include #include #include #include #include #include #include "ice_rdma.h" #include "irdma_main.h" #include "icrdma_hw.h" #include "irdma_if.h" #include "irdma_di_if.h" /** * Driver version */ -char irdma_driver_version[] = "1.2.36-k"; +char irdma_driver_version[] = "1.2.37-k"; /** * irdma_init_tunable - prepare tunables * @rf: RDMA PCI function * @pf_id: id of the pf */ static void irdma_init_tunable(struct irdma_pci_f *rf, uint8_t pf_id) { struct sysctl_oid_list *irdma_oid_list; struct irdma_tunable_info *t_info = &rf->tun_info; char pf_name[16]; snprintf(pf_name, 15, "irdma%d", pf_id); sysctl_ctx_init(&t_info->irdma_sysctl_ctx); t_info->irdma_sysctl_tree = SYSCTL_ADD_NODE(&t_info->irdma_sysctl_ctx, SYSCTL_STATIC_CHILDREN(_dev), OID_AUTO, pf_name, CTLFLAG_RD, NULL, ""); irdma_oid_list = SYSCTL_CHILDREN(t_info->irdma_sysctl_tree); t_info->sws_sysctl_tree = SYSCTL_ADD_NODE(&t_info->irdma_sysctl_ctx, irdma_oid_list, OID_AUTO, "sw_stats", CTLFLAG_RD, NULL, ""); /* * debug mask setting */ SYSCTL_ADD_S32(&t_info->irdma_sysctl_ctx, irdma_oid_list, OID_AUTO, "debug", CTLFLAG_RWTUN, &rf->sc_dev.debug_mask, 0, "irdma debug"); /* * RoCEv2/iWARP setting RoCEv2 the default mode */ t_info->roce_ena = 1; SYSCTL_ADD_U8(&t_info->irdma_sysctl_ctx, irdma_oid_list, OID_AUTO, "roce_enable", CTLFLAG_RDTUN, &t_info->roce_ena, 0, "RoCEv2 mode enable"); rf->protocol_used = IRDMA_IWARP_PROTOCOL_ONLY; if (t_info->roce_ena == 1) rf->protocol_used = IRDMA_ROCE_PROTOCOL_ONLY; else if (t_info->roce_ena != 0) printf("%s:%d wrong roce_enable value (%d), using iWARP\n", __func__, __LINE__, t_info->roce_ena); printf("%s:%d protocol: %s, roce_enable value: %d\n", __func__, __LINE__, (rf->protocol_used == IRDMA_IWARP_PROTOCOL_ONLY) ? "iWARP" : "RoCEv2", t_info->roce_ena); snprintf(t_info->drv_ver, IRDMA_VER_LEN, "%s", irdma_driver_version); SYSCTL_ADD_STRING(&t_info->irdma_sysctl_ctx, irdma_oid_list, OID_AUTO, "drv_ver", CTLFLAG_RDTUN, t_info->drv_ver, IRDMA_VER_LEN, "driver version"); irdma_dcqcn_tunables_init(rf); irdma_sysctl_settings(rf); } /** * irdma_find_handler - obtain hdl object to identify pf * @p_dev: the peer interface structure */ static struct irdma_handler * irdma_find_handler(struct ice_rdma_peer *p_dev) { struct irdma_handler *hdl; unsigned long flags; spin_lock_irqsave(&irdma_handler_lock, flags); list_for_each_entry(hdl, &irdma_handlers, list) { if (!hdl->iwdev->rf->peer_info) continue; if (hdl->iwdev->rf->peer_info->dev == p_dev->dev) { spin_unlock_irqrestore(&irdma_handler_lock, flags); return hdl; } } spin_unlock_irqrestore(&irdma_handler_lock, flags); return NULL; } /** * peer_to_iwdev - return iwdev based on peer * @peer: the peer interface structure */ static struct irdma_device * peer_to_iwdev(struct ice_rdma_peer *peer) { struct irdma_handler *hdl; hdl = irdma_find_handler(peer); if (!hdl) { printf("%s:%d rdma handler not found\n", __func__, __LINE__); return NULL; } return hdl->iwdev; } /** * irdma_get_qos_info - save qos info from parameters to internal struct * @l2params: destination, qos, tc, mtu info structure * @qos_info: source, DCB settings structure */ static void irdma_get_qos_info(struct irdma_pci_f *rf, struct irdma_l2params *l2params, struct ice_qos_params *qos_info) { int i; char txt[7][128] = {"", "", "", "", "", "", ""}; u8 len; l2params->num_tc = qos_info->num_tc; l2params->num_apps = qos_info->num_apps; l2params->vsi_prio_type = qos_info->vsi_priority_type; l2params->vsi_rel_bw = qos_info->vsi_relative_bw; for (i = 0; i < l2params->num_tc; i++) { l2params->tc_info[i].egress_virt_up = qos_info->tc_info[i].egress_virt_up; l2params->tc_info[i].ingress_virt_up = qos_info->tc_info[i].ingress_virt_up; l2params->tc_info[i].prio_type = qos_info->tc_info[i].prio_type; l2params->tc_info[i].rel_bw = qos_info->tc_info[i].rel_bw; l2params->tc_info[i].tc_ctx = qos_info->tc_info[i].tc_ctx; } for (i = 0; i < IRDMA_MAX_USER_PRIORITY; i++) l2params->up2tc[i] = qos_info->up2tc[i]; if (qos_info->pfc_mode == IRDMA_QOS_MODE_DSCP) { l2params->dscp_mode = true; memcpy(l2params->dscp_map, qos_info->dscp_map, sizeof(l2params->dscp_map)); } if (!(rf->sc_dev.debug_mask & IRDMA_DEBUG_DCB)) return; for (i = 0; i < l2params->num_tc; i++) { len = strlen(txt[0]); snprintf(txt[0] + len, sizeof(txt[0]) - 5, " %d", l2params->tc_info[i].egress_virt_up); len = strlen(txt[1]); snprintf(txt[1] + len, sizeof(txt[1]) - 5, " %d", l2params->tc_info[i].ingress_virt_up); len = strlen(txt[2]); snprintf(txt[2] + len, sizeof(txt[2]) - 5, " %d", l2params->tc_info[i].prio_type); len = strlen(txt[3]); snprintf(txt[3] + len, sizeof(txt[3]) - 5, " %d", l2params->tc_info[i].rel_bw); len = strlen(txt[4]); snprintf(txt[4] + len, sizeof(txt[4]) - 5, " %lu", l2params->tc_info[i].tc_ctx); } len = strlen(txt[5]); for (i = 0; i < IRDMA_MAX_USER_PRIORITY; i++) len += snprintf(txt[5] + len, sizeof(txt[5]) - 5, " %d", l2params->up2tc[i]); len = strlen(txt[6]); for (i = 0; i < IRDMA_DSCP_NUM_VAL; i++) len += snprintf(txt[6] + len, sizeof(txt[6]) - 5, " %d", l2params->dscp_map[i]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "num_tc: %d\n", l2params->num_tc); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "num_apps: %d\n", l2params->num_apps); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "vsi_prio_type: %d\n", l2params->vsi_prio_type); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "vsi_rel_bw: %d\n", l2params->vsi_rel_bw); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "egress_virt_up: %s\n", txt[0]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "ingress_virt_up:%s\n", txt[1]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "prio_type: %s\n", txt[2]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "rel_bw: %s\n", txt[3]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "tc_ctx: %s\n", txt[4]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "up2tc: %s\n", txt[5]); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_DCB, "dscp_mode: %s\n", txt[6]); irdma_debug_buf(&rf->sc_dev, IRDMA_DEBUG_DCB, "l2params", l2params, sizeof(*l2params)); } /** * irdma_log_invalid_mtu - check mtu setting validity * @mtu: mtu value * @dev: hardware control device structure */ static void irdma_log_invalid_mtu(u16 mtu, struct irdma_sc_dev *dev) { if (mtu < IRDMA_MIN_MTU_IPV4) irdma_dev_warn(to_ibdev(dev), "MTU setting [%d] too low for RDMA traffic. Minimum MTU is 576 for IPv4\n", mtu); else if (mtu < IRDMA_MIN_MTU_IPV6) irdma_dev_warn(to_ibdev(dev), "MTU setting [%d] too low for RDMA traffic. Minimum MTU is 1280 for IPv6\n", mtu); } /** * irdma_get_event_name - convert type enum to string * @type: event type enum */ static const char * irdma_get_event_name(enum ice_rdma_event_type type) { switch (type) { case ICE_RDMA_EVENT_LINK_CHANGE: return "LINK CHANGE"; case ICE_RDMA_EVENT_MTU_CHANGE: return "MTU CHANGE"; case ICE_RDMA_EVENT_TC_CHANGE: return "TC CHANGE"; case ICE_RDMA_EVENT_API_CHANGE: return "API CHANGE"; case ICE_RDMA_EVENT_CRIT_ERR: return "CRITICAL ERROR"; case ICE_RDMA_EVENT_RESET: return "RESET"; case ICE_RDMA_EVENT_QSET_REGISTER: return "QSET REGISTER"; case ICE_RDMA_EVENT_VSI_FILTER_UPDATE: return "VSI FILTER UPDATE"; default: return "UNKNOWN"; } } /** * irdma_event_handler - handling events from lan driver * @peer: the peer interface structure * @event: event info structure */ static void irdma_event_handler(struct ice_rdma_peer *peer, struct ice_rdma_event *event) { struct irdma_device *iwdev; struct irdma_l2params l2params = {}; printf("%s:%d event_handler %s (%x) on pf %d (%d)\n", __func__, __LINE__, irdma_get_event_name(event->type), event->type, peer->pf_id, if_getdunit(peer->ifp)); iwdev = peer_to_iwdev(peer); if (!iwdev) { printf("%s:%d rdma device not found\n", __func__, __LINE__); return; } switch (event->type) { case ICE_RDMA_EVENT_LINK_CHANGE: printf("%s:%d PF: %x (%x), state: %d, speed: %lu\n", __func__, __LINE__, peer->pf_id, if_getdunit(peer->ifp), event->linkstate, event->baudrate); break; case ICE_RDMA_EVENT_MTU_CHANGE: if (iwdev->vsi.mtu != event->mtu) { l2params.mtu = event->mtu; l2params.mtu_changed = true; irdma_log_invalid_mtu(l2params.mtu, &iwdev->rf->sc_dev); irdma_change_l2params(&iwdev->vsi, &l2params); } break; case ICE_RDMA_EVENT_TC_CHANGE: /* * 1. check if it is pre or post 2. check if it is currently being done */ if (event->prep == iwdev->vsi.tc_change_pending) { printf("%s:%d can't process %s TC change if TC change is %spending\n", __func__, __LINE__, event->prep ? "pre" : "post", event->prep ? " " : "not "); goto done; } if (!atomic_inc_not_zero(&iwdev->rf->dev_ctx.event_rfcnt)) { printf("%s:%d (%d) EVENT_TC_CHANGE received, but not processed %d\n", __func__, __LINE__, if_getdunit(peer->ifp), atomic_read(&iwdev->rf->dev_ctx.event_rfcnt)); break; } if (event->prep) { iwdev->vsi.tc_change_pending = true; irdma_sc_suspend_resume_qps(&iwdev->vsi, IRDMA_OP_SUSPEND); wait_event_timeout(iwdev->suspend_wq, !atomic_read(&iwdev->vsi.qp_suspend_reqs), IRDMA_EVENT_TIMEOUT_MS * 10); irdma_ws_reset(&iwdev->vsi); printf("%s:%d TC change preparation done\n", __func__, __LINE__); } else { l2params.tc_changed = true; irdma_get_qos_info(iwdev->rf, &l2params, &event->port_qos); if (iwdev->rf->protocol_used != IRDMA_IWARP_PROTOCOL_ONLY) iwdev->dcb_vlan_mode = l2params.num_tc > 1 && !l2params.dscp_mode; irdma_check_fc_for_tc_update(&iwdev->vsi, &l2params); irdma_change_l2params(&iwdev->vsi, &l2params); printf("%s:%d TC change done\n", __func__, __LINE__); } atomic_dec(&iwdev->rf->dev_ctx.event_rfcnt); break; case ICE_RDMA_EVENT_CRIT_ERR: if (event->oicr_reg & IRDMAPFINT_OICR_PE_CRITERR_M) { u32 pe_criterr; #define IRDMA_Q1_RESOURCE_ERR 0x0001024d pe_criterr = readl(iwdev->rf->sc_dev.hw_regs[IRDMA_GLPE_CRITERR]); if (pe_criterr != IRDMA_Q1_RESOURCE_ERR) { irdma_pr_err("critical PE Error, GLPE_CRITERR=0x%08x\n", pe_criterr); iwdev->rf->reset = true; } else { irdma_dev_warn(to_ibdev(&iwdev->rf->sc_dev), "Q1 Resource Check\n"); } } if (event->oicr_reg & IRDMAPFINT_OICR_HMC_ERR_M) { irdma_pr_err("HMC Error\n"); iwdev->rf->reset = true; } if (iwdev->rf->reset) iwdev->rf->gen_ops.request_reset(iwdev->rf); break; case ICE_RDMA_EVENT_RESET: iwdev->rf->reset = true; break; default: printf("%s:%d event type unsupported: %d\n", __func__, __LINE__, event->type); } done: return; } /** * irdma_link_change - Callback for link state change * @peer: the peer interface structure * @linkstate: state of the link * @baudrate: speed of the link */ static void irdma_link_change(struct ice_rdma_peer *peer, int linkstate, uint64_t baudrate) { printf("%s:%d PF: %x (%x), state: %d, speed: %lu\n", __func__, __LINE__, peer->pf_id, if_getdunit(peer->ifp), linkstate, baudrate); } /** * irdma_finalize_task - Finish open or close phase in a separate thread * @context: instance holding peer and iwdev information * * Triggered from irdma_open or irdma_close to perform rt_init_hw or * rt_deinit_hw respectively. Does registration and unregistration of * the device. */ static void irdma_finalize_task(void *context, int pending) { struct irdma_task_arg *task_arg = (struct irdma_task_arg *)context; struct irdma_device *iwdev = task_arg->iwdev; struct irdma_pci_f *rf = iwdev->rf; struct ice_rdma_peer *peer = task_arg->peer; struct irdma_l2params l2params = {{{0}}}; struct ice_rdma_request req = {0}; int status = 0; if (iwdev->iw_status) { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_INIT, "Starting deferred closing %d (%d)\n", rf->peer_info->pf_id, if_getdunit(peer->ifp)); atomic_dec(&rf->dev_ctx.event_rfcnt); wait_event_timeout(iwdev->suspend_wq, !atomic_read(&rf->dev_ctx.event_rfcnt), IRDMA_MAX_TIMEOUT); if (atomic_read(&rf->dev_ctx.event_rfcnt) != 0) { printf("%s:%d (%d) waiting for event_rfcnt (%d) timeout, proceed with unload\n", __func__, __LINE__, if_getdunit(peer->ifp), atomic_read(&rf->dev_ctx.event_rfcnt)); } irdma_dereg_ipaddr_event_cb(rf); irdma_ib_unregister_device(iwdev); req.type = ICE_RDMA_EVENT_VSI_FILTER_UPDATE; req.enable_filter = false; IRDMA_DI_REQ_HANDLER(peer, &req); irdma_cleanup_dead_qps(&iwdev->vsi); irdma_rt_deinit_hw(iwdev); } else { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_INIT, "Starting deferred opening %d (%d)\n", rf->peer_info->pf_id, if_getdunit(peer->ifp)); irdma_get_qos_info(iwdev->rf, &l2params, &peer->initial_qos_info); if (iwdev->rf->protocol_used != IRDMA_IWARP_PROTOCOL_ONLY) iwdev->dcb_vlan_mode = l2params.num_tc > 1 && !l2params.dscp_mode; l2params.mtu = peer->mtu; status = irdma_rt_init_hw(iwdev, &l2params); if (status) { irdma_pr_err("RT init failed %d\n", status); ib_dealloc_device(&iwdev->ibdev); return; } status = irdma_ib_register_device(iwdev); if (status) { irdma_pr_err("Registration failed %d\n", status); irdma_rt_deinit_hw(iwdev); ib_dealloc_device(&iwdev->ibdev); } irdma_sw_stats_tunables_init(rf); req.type = ICE_RDMA_EVENT_VSI_FILTER_UPDATE; req.enable_filter = true; IRDMA_DI_REQ_HANDLER(peer, &req); irdma_reg_ipaddr_event_cb(rf); atomic_inc(&rf->dev_ctx.event_rfcnt); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_INIT, "Deferred opening finished %d (%d)\n", rf->peer_info->pf_id, if_getdunit(peer->ifp)); } } /** * irdma_alloc_pcidev - allocate memory for pcidev and populate data * @peer: the new peer interface structure * @rf: RDMA PCI function */ static int irdma_alloc_pcidev(struct ice_rdma_peer *peer, struct irdma_pci_f *rf) { rf->pcidev = kzalloc(sizeof(struct pci_dev), GFP_KERNEL); if (!rf->pcidev) { return -ENOMEM; } if (linux_pci_attach_device(rf->dev_ctx.dev, NULL, NULL, rf->pcidev)) return -ENOMEM; return 0; } /** * irdma_dealloc_pcidev - deallocate memory for pcidev * @rf: RDMA PCI function */ static void irdma_dealloc_pcidev(struct irdma_pci_f *rf) { linux_pci_detach_device(rf->pcidev); kfree(rf->pcidev); } /** * irdma_fill_device_info - assign initial values to rf variables * @iwdev: irdma device * @peer: the peer interface structure */ static void irdma_fill_device_info(struct irdma_device *iwdev, struct ice_rdma_peer *peer) { struct irdma_pci_f *rf = iwdev->rf; rf->peer_info = peer; rf->gen_ops.register_qset = irdma_register_qset; rf->gen_ops.unregister_qset = irdma_unregister_qset; rf->rdma_ver = IRDMA_GEN_2; rf->sc_dev.hw_attrs.uk_attrs.hw_rev = IRDMA_GEN_2; rf->rsrc_profile = IRDMA_HMC_PROFILE_DEFAULT; rf->rst_to = IRDMA_RST_TIMEOUT_HZ; rf->check_fc = irdma_check_fc_for_qp; rf->gen_ops.request_reset = irdma_request_reset; irdma_set_rf_user_cfg_params(rf); rf->default_vsi.vsi_idx = peer->pf_vsi_num; rf->dev_ctx.dev = peer->dev; rf->dev_ctx.mem_bus_space_tag = rman_get_bustag(peer->pci_mem); rf->dev_ctx.mem_bus_space_handle = rman_get_bushandle(peer->pci_mem); rf->dev_ctx.mem_bus_space_size = rman_get_size(peer->pci_mem); rf->hw.dev_context = &rf->dev_ctx; rf->hw.hw_addr = (u8 *)rman_get_virtual(peer->pci_mem); rf->msix_count = peer->msix.count; rf->msix_info.entry = peer->msix.base; rf->msix_info.vector = peer->msix.count; printf("%s:%d msix_info: %d %d %d\n", __func__, __LINE__, rf->msix_count, rf->msix_info.entry, rf->msix_info.vector); rf->iwdev = iwdev; iwdev->netdev = peer->ifp; iwdev->init_state = INITIAL_STATE; iwdev->vsi_num = peer->pf_vsi_num; iwdev->rcv_wnd = IRDMA_CM_DEFAULT_RCV_WND_SCALED; iwdev->rcv_wscale = IRDMA_CM_DEFAULT_RCV_WND_SCALE; iwdev->roce_cwnd = IRDMA_ROCE_CWND_DEFAULT; iwdev->roce_ackcreds = IRDMA_ROCE_ACKCREDS_DEFAULT; iwdev->roce_rtomin = 5; if (rf->protocol_used == IRDMA_ROCE_PROTOCOL_ONLY) { iwdev->roce_mode = true; } } /** * irdma_probe - Callback to probe a new RDMA peer device * @peer: the new peer interface structure * * Callback implementing the RDMA_PROBE function. Called by the ice driver to * notify the RDMA client driver that a new device has been created */ static int irdma_probe(struct ice_rdma_peer *peer) { struct irdma_device *iwdev; struct irdma_pci_f *rf; struct irdma_handler *hdl; int err = 0; irdma_pr_info("probe: irdma-%s peer=%p, peer->pf_id=%d, peer->ifp=%p, peer->ifp->if_dunit=%d, peer->pci_mem->r_bustag=%p\n", irdma_driver_version, peer, peer->pf_id, peer->ifp, if_getdunit(peer->ifp), (void *)(uintptr_t)peer->pci_mem->r_bustag); hdl = irdma_find_handler(peer); if (hdl) return -EBUSY; hdl = kzalloc(sizeof(*hdl), GFP_KERNEL); if (!hdl) return -ENOMEM; iwdev = (struct irdma_device *)ib_alloc_device(sizeof(*iwdev)); if (!iwdev) { kfree(hdl); return -ENOMEM; } iwdev->rf = kzalloc(sizeof(*rf), GFP_KERNEL); if (!iwdev->rf) { ib_dealloc_device(&iwdev->ibdev); kfree(hdl); return -ENOMEM; } hdl->iwdev = iwdev; iwdev->hdl = hdl; irdma_init_tunable(iwdev->rf, if_getdunit(peer->ifp)); irdma_fill_device_info(iwdev, peer); rf = iwdev->rf; if (irdma_alloc_pcidev(peer, rf)) goto err_pcidev; irdma_add_handler(hdl); if (irdma_ctrl_init_hw(rf)) { err = -EIO; goto err_ctrl_init; } rf->dev_ctx.task_arg.peer = peer; rf->dev_ctx.task_arg.iwdev = iwdev; rf->dev_ctx.task_arg.peer = peer; TASK_INIT(&hdl->deferred_task, 0, irdma_finalize_task, &rf->dev_ctx.task_arg); hdl->deferred_tq = taskqueue_create_fast("irdma_defer", M_NOWAIT, taskqueue_thread_enqueue, &hdl->deferred_tq); taskqueue_start_threads(&hdl->deferred_tq, 1, PI_NET, "irdma_defer_t"); taskqueue_enqueue(hdl->deferred_tq, &hdl->deferred_task); return 0; err_ctrl_init: irdma_del_handler(hdl); irdma_dealloc_pcidev(rf); err_pcidev: kfree(iwdev->rf); ib_dealloc_device(&iwdev->ibdev); kfree(hdl); return err; } /** * irdma_remove - Callback to remove an RDMA peer device * @peer: the new peer interface structure * * Callback implementing the RDMA_REMOVE function. Called by the ice driver to * notify the RDMA client driver that the device wille be delated */ static int irdma_remove(struct ice_rdma_peer *peer) { struct irdma_handler *hdl; struct irdma_device *iwdev; irdma_debug((struct irdma_sc_dev *)NULL, IRDMA_DEBUG_INIT, "removing %s irdma%d\n", __func__, if_getdunit(peer->ifp)); hdl = irdma_find_handler(peer); if (!hdl) return 0; iwdev = hdl->iwdev; if (iwdev->vsi.tc_change_pending) { iwdev->vsi.tc_change_pending = false; irdma_sc_suspend_resume_qps(&iwdev->vsi, IRDMA_OP_RESUME); } taskqueue_enqueue(hdl->deferred_tq, &hdl->deferred_task); taskqueue_drain(hdl->deferred_tq, &hdl->deferred_task); taskqueue_free(hdl->deferred_tq); hdl->iwdev->rf->dev_ctx.task_arg.iwdev = NULL; hdl->iwdev->rf->dev_ctx.task_arg.peer = NULL; sysctl_ctx_free(&iwdev->rf->tun_info.irdma_sysctl_ctx); hdl->iwdev->rf->tun_info.irdma_sysctl_tree = NULL; hdl->iwdev->rf->tun_info.sws_sysctl_tree = NULL; irdma_ctrl_deinit_hw(iwdev->rf); irdma_dealloc_pcidev(iwdev->rf); irdma_del_handler(iwdev->hdl); kfree(iwdev->hdl); kfree(iwdev->rf); ib_dealloc_device(&iwdev->ibdev); irdma_pr_info("IRDMA hardware deinitialization complete irdma%d\n", if_getdunit(peer->ifp)); return 0; } /** * irdma_open - Callback for operation open for RDMA device * @peer: the new peer interface structure * * Callback implementing the RDMA_OPEN function. Called by the ice driver to * notify the RDMA client driver that a new device has been initialized. */ static int irdma_open(struct ice_rdma_peer *peer) { struct irdma_device *iwdev; struct ice_rdma_event event = {0}; iwdev = peer_to_iwdev(peer); if (iwdev) { event.type = ICE_RDMA_EVENT_MTU_CHANGE; event.mtu = peer->mtu; irdma_event_handler(peer, &event); } else { irdma_probe(peer); } return 0; } /** * irdma_close - Callback to notify that a peer device is down * @peer: the RDMA peer device being stopped * * Callback implementing the RDMA_CLOSE function. Called by the ice driver to * notify the RDMA client driver that a peer device is being stopped. */ static int irdma_close(struct ice_rdma_peer *peer) { /* * This is called when ifconfig down or pf-reset is about to happen. */ struct irdma_device *iwdev; iwdev = peer_to_iwdev(peer); if (iwdev && iwdev->rf->reset) irdma_remove(peer); return 0; } /** * irdma_prep_for_unregister - ensure the driver is ready to unregister */ static void irdma_prep_for_unregister(void) { struct irdma_handler *hdl; unsigned long flags; bool hdl_valid; do { hdl_valid = false; spin_lock_irqsave(&irdma_handler_lock, flags); list_for_each_entry(hdl, &irdma_handlers, list) { if (!hdl->iwdev->rf->peer_info) continue; hdl_valid = true; break; } spin_unlock_irqrestore(&irdma_handler_lock, flags); if (!hdl || !hdl_valid) break; IRDMA_CLOSE(hdl->iwdev->rf->peer_info); IRDMA_REMOVE(hdl->iwdev->rf->peer_info); } while (1); } static kobj_method_t irdma_methods[] = { KOBJMETHOD(irdma_probe, irdma_probe), KOBJMETHOD(irdma_open, irdma_open), KOBJMETHOD(irdma_close, irdma_close), KOBJMETHOD(irdma_remove, irdma_remove), KOBJMETHOD(irdma_link_change, irdma_link_change), KOBJMETHOD(irdma_event_handler, irdma_event_handler), KOBJMETHOD_END }; /* declare irdma_class which extends the ice_rdma_di class */ DEFINE_CLASS_1(irdma, irdma_class, irdma_methods, sizeof(struct ice_rdma_peer), ice_rdma_di_class); static struct ice_rdma_info irdma_info = { .major_version = ICE_RDMA_MAJOR_VERSION, .minor_version = ICE_RDMA_MINOR_VERSION, .patch_version = ICE_RDMA_PATCH_VERSION, .rdma_class = &irdma_class, }; /** * irdma_module_event_handler - Module event handler callback * @mod: unused mod argument * @what: the module event to handle * @arg: unused module event argument * * Callback used by the FreeBSD module stack to notify the driver of module * events. Used to implement custom handling for certain module events such as * load and unload. */ static int irdma_module_event_handler(module_t __unused mod, int what, void __unused * arg) { switch (what) { case MOD_LOAD: printf("Loading irdma module\n"); return ice_rdma_register(&irdma_info); case MOD_UNLOAD: printf("Unloading irdma module\n"); irdma_prep_for_unregister(); ice_rdma_unregister(); return (0); default: return (EOPNOTSUPP); } return (0); } static moduledata_t irdma_moduledata = { "irdma", irdma_module_event_handler, NULL }; DECLARE_MODULE(irdma, irdma_moduledata, SI_SUB_LAST, SI_ORDER_ANY); MODULE_VERSION(irdma, 1); MODULE_DEPEND(irdma, ice, 1, 1, 1); MODULE_DEPEND(irdma, ibcore, 1, 1, 1); diff --git a/sys/dev/irdma/irdma_cm.c b/sys/dev/irdma/irdma_cm.c index d4d4f328fb43..f3ca761b32f6 100644 --- a/sys/dev/irdma/irdma_cm.c +++ b/sys/dev/irdma/irdma_cm.c @@ -1,4271 +1,4237 @@ /*- * SPDX-License-Identifier: GPL-2.0 or Linux-OpenIB * - * Copyright (c) 2015 - 2023 Intel Corporation + * Copyright (c) 2015 - 2025 Intel Corporation * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU * General Public License (GPL) Version 2, available from the file * COPYING in the main directory of this source tree, or the * OpenFabrics.org BSD license below: * * 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. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #include "irdma_main.h" static void irdma_cm_post_event(struct irdma_cm_event *event); static void irdma_disconnect_worker(struct work_struct *work); /** * irdma_free_sqbuf - put back puda buffer if refcount is 0 * @vsi: The VSI structure of the device * @bufp: puda buffer to free */ void irdma_free_sqbuf(struct irdma_sc_vsi *vsi, void *bufp) { struct irdma_puda_buf *buf = bufp; struct irdma_puda_rsrc *ilq = vsi->ilq; if (atomic_dec_and_test(&buf->refcount)) irdma_puda_ret_bufpool(ilq, buf); } /** * irdma_record_ird_ord - Record IRD/ORD passed in * @cm_node: connection's node * @conn_ird: connection IRD * @conn_ord: connection ORD */ static void irdma_record_ird_ord(struct irdma_cm_node *cm_node, u32 conn_ird, u32 conn_ord) { if (conn_ird > cm_node->dev->hw_attrs.max_hw_ird) conn_ird = cm_node->dev->hw_attrs.max_hw_ird; if (conn_ord > cm_node->dev->hw_attrs.max_hw_ord) conn_ord = cm_node->dev->hw_attrs.max_hw_ord; else if (!conn_ord && cm_node->send_rdma0_op == SEND_RDMA_READ_ZERO) conn_ord = 1; cm_node->ird_size = conn_ird; cm_node->ord_size = conn_ord; } /** * irdma_copy_ip_ntohl - copy IP address from network to host * @dst: IP address in host order * @src: IP address in network order (big endian) */ void irdma_copy_ip_ntohl(u32 *dst, __be32 *src) { *dst++ = ntohl(*src++); *dst++ = ntohl(*src++); *dst++ = ntohl(*src++); *dst = ntohl(*src); } /** * irdma_copy_ip_htonl - copy IP address from host to network order * @dst: IP address in network order (big endian) * @src: IP address in host order */ void irdma_copy_ip_htonl(__be32 *dst, u32 *src) { *dst++ = htonl(*src++); *dst++ = htonl(*src++); *dst++ = htonl(*src++); *dst = htonl(*src); } /** * irdma_get_addr_info * @cm_node: contains ip/tcp info * @cm_info: to get a copy of the cm_node ip/tcp info */ static void irdma_get_addr_info(struct irdma_cm_node *cm_node, struct irdma_cm_info *cm_info) { memset(cm_info, 0, sizeof(*cm_info)); cm_info->ipv4 = cm_node->ipv4; cm_info->vlan_id = cm_node->vlan_id; memcpy(cm_info->loc_addr, cm_node->loc_addr, sizeof(cm_info->loc_addr)); memcpy(cm_info->rem_addr, cm_node->rem_addr, sizeof(cm_info->rem_addr)); cm_info->loc_port = cm_node->loc_port; cm_info->rem_port = cm_node->rem_port; } /** * irdma_fill_sockaddr4 - fill in addr info for IPv4 connection * @cm_node: connection's node * @event: upper layer's cm event */ static inline void irdma_fill_sockaddr4(struct irdma_cm_node *cm_node, struct iw_cm_event *event) { struct sockaddr_in *laddr = (struct sockaddr_in *)&event->local_addr; struct sockaddr_in *raddr = (struct sockaddr_in *)&event->remote_addr; laddr->sin_family = AF_INET; raddr->sin_family = AF_INET; laddr->sin_port = htons(cm_node->loc_port); raddr->sin_port = htons(cm_node->rem_port); laddr->sin_addr.s_addr = htonl(cm_node->loc_addr[0]); raddr->sin_addr.s_addr = htonl(cm_node->rem_addr[0]); } /** * irdma_fill_sockaddr6 - fill in addr info for IPv6 connection * @cm_node: connection's node * @event: upper layer's cm event */ static inline void irdma_fill_sockaddr6(struct irdma_cm_node *cm_node, struct iw_cm_event *event) { struct sockaddr_in6 *laddr6 = (struct sockaddr_in6 *)&event->local_addr; struct sockaddr_in6 *raddr6 = (struct sockaddr_in6 *)&event->remote_addr; laddr6->sin6_family = AF_INET6; raddr6->sin6_family = AF_INET6; laddr6->sin6_port = htons(cm_node->loc_port); raddr6->sin6_port = htons(cm_node->rem_port); irdma_copy_ip_htonl(laddr6->sin6_addr.__u6_addr.__u6_addr32, cm_node->loc_addr); irdma_copy_ip_htonl(raddr6->sin6_addr.__u6_addr.__u6_addr32, cm_node->rem_addr); } /** * irdma_get_cmevent_info - for cm event upcall * @cm_node: connection's node * @cm_id: upper layers cm struct for the event * @event: upper layer's cm event */ static inline void irdma_get_cmevent_info(struct irdma_cm_node *cm_node, struct iw_cm_id *cm_id, struct iw_cm_event *event) { memcpy(&event->local_addr, &cm_id->m_local_addr, sizeof(event->local_addr)); memcpy(&event->remote_addr, &cm_id->m_remote_addr, sizeof(event->remote_addr)); if (cm_node) { event->private_data = cm_node->pdata_buf; event->private_data_len = (u8)cm_node->pdata.size; event->ird = cm_node->ird_size; event->ord = cm_node->ord_size; } } /** * irdma_send_cm_event - upcall cm's event handler * @cm_node: connection's node * @cm_id: upper layer's cm info struct * @type: Event type to indicate * @status: status for the event type */ static int irdma_send_cm_event(struct irdma_cm_node *cm_node, struct iw_cm_id *cm_id, enum iw_cm_event_type type, int status) { struct iw_cm_event event = {0}; event.event = type; event.status = status; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_node %p cm_id=%p state=%d accel=%d event_type=%d status=%d\n", cm_node, cm_id, cm_node->accelerated, cm_node->state, type, status); switch (type) { case IW_CM_EVENT_CONNECT_REQUEST: if (cm_node->ipv4) irdma_fill_sockaddr4(cm_node, &event); else irdma_fill_sockaddr6(cm_node, &event); event.provider_data = cm_node; event.private_data = cm_node->pdata_buf; event.private_data_len = (u8)cm_node->pdata.size; event.ird = cm_node->ird_size; break; case IW_CM_EVENT_CONNECT_REPLY: irdma_get_cmevent_info(cm_node, cm_id, &event); break; case IW_CM_EVENT_ESTABLISHED: event.ird = cm_node->ird_size; event.ord = cm_node->ord_size; break; case IW_CM_EVENT_DISCONNECT: case IW_CM_EVENT_CLOSE: /* Wait if we are in RTS but havent issued the iwcm event upcall */ if (!cm_node->accelerated) wait_for_completion(&cm_node->establish_comp); break; default: return -EINVAL; } return cm_id->event_handler(cm_id, &event); } /** * irdma_timer_list_prep - add connection nodes to a list to perform timer tasks * @cm_core: cm's core * @timer_list: a timer list to which cm_node will be selected */ static void irdma_timer_list_prep(struct irdma_cm_core *cm_core, struct list_head *timer_list) { struct irdma_cm_node *cm_node; int bkt; HASH_FOR_EACH_RCU(cm_core->cm_hash_tbl, bkt, cm_node, list) { if ((cm_node->close_entry || cm_node->send_entry) && atomic_inc_not_zero(&cm_node->refcnt)) list_add(&cm_node->timer_entry, timer_list); } } /** * irdma_create_event - create cm event * @cm_node: connection's node * @type: Event type to generate */ static struct irdma_cm_event * irdma_create_event(struct irdma_cm_node *cm_node, enum irdma_cm_event_type type) { struct irdma_cm_event *event; if (!cm_node->cm_id) return NULL; event = kzalloc(sizeof(*event), GFP_ATOMIC); if (!event) return NULL; event->type = type; event->cm_node = cm_node; memcpy(event->cm_info.rem_addr, cm_node->rem_addr, sizeof(event->cm_info.rem_addr)); memcpy(event->cm_info.loc_addr, cm_node->loc_addr, sizeof(event->cm_info.loc_addr)); event->cm_info.rem_port = cm_node->rem_port; event->cm_info.loc_port = cm_node->loc_port; event->cm_info.cm_id = cm_node->cm_id; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "node=%p event=%p type=%u dst=%x src=%x\n", cm_node, event, type, event->cm_info.loc_addr[0], event->cm_info.rem_addr[0]); irdma_cm_post_event(event); return event; } /** * irdma_free_retrans_entry - free send entry * @cm_node: connection's node */ static void irdma_free_retrans_entry(struct irdma_cm_node *cm_node) { struct irdma_device *iwdev = cm_node->iwdev; struct irdma_timer_entry *send_entry; send_entry = cm_node->send_entry; if (!send_entry) return; cm_node->send_entry = NULL; irdma_free_sqbuf(&iwdev->vsi, send_entry->sqbuf); kfree(send_entry); atomic_dec(&cm_node->refcnt); } /** * irdma_cleanup_retrans_entry - free send entry with lock * @cm_node: connection's node */ static void irdma_cleanup_retrans_entry(struct irdma_cm_node *cm_node) { unsigned long flags; spin_lock_irqsave(&cm_node->retrans_list_lock, flags); irdma_free_retrans_entry(cm_node); spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); } /** * irdma_form_ah_cm_frame - get a free packet and build frame with address handle * @cm_node: connection's node ionfo to use in frame * @options: pointer to options info * @hdr: pointer mpa header * @pdata: pointer to private data * @flags: indicates FIN or ACK */ static struct irdma_puda_buf * irdma_form_ah_cm_frame(struct irdma_cm_node *cm_node, struct irdma_kmem_info *options, struct irdma_kmem_info *hdr, struct irdma_mpa_priv_info *pdata, u8 flags) { struct irdma_puda_buf *sqbuf; struct irdma_sc_vsi *vsi = &cm_node->iwdev->vsi; u8 *buf; struct tcphdr *tcph; u16 pktsize; u32 opts_len = 0; u32 pd_len = 0; u32 hdr_len = 0; if (!cm_node->ah || !cm_node->ah->ah_info.ah_valid) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "AH invalid\n"); return NULL; } sqbuf = irdma_puda_get_bufpool(vsi->ilq); if (!sqbuf) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "SQ buf NULL\n"); return NULL; } sqbuf->ah_id = cm_node->ah->ah_info.ah_idx; buf = sqbuf->mem.va; if (options) opts_len = (u32)options->size; if (hdr) hdr_len = hdr->size; if (pdata) pd_len = pdata->size; pktsize = sizeof(*tcph) + opts_len + hdr_len + pd_len; memset(buf, 0, pktsize); sqbuf->totallen = pktsize; sqbuf->tcphlen = sizeof(*tcph) + opts_len; sqbuf->scratch = cm_node; tcph = (struct tcphdr *)buf; buf += sizeof(*tcph); tcph->th_sport = htons(cm_node->loc_port); tcph->th_dport = htons(cm_node->rem_port); tcph->th_seq = htonl(cm_node->tcp_cntxt.loc_seq_num); if (flags & SET_ACK) { cm_node->tcp_cntxt.loc_ack_num = cm_node->tcp_cntxt.rcv_nxt; tcph->th_ack = htonl(cm_node->tcp_cntxt.loc_ack_num); tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_ACK); } else { tcph->th_ack = 0; } if (flags & SET_SYN) { cm_node->tcp_cntxt.loc_seq_num++; tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_SYN); } else { cm_node->tcp_cntxt.loc_seq_num += hdr_len + pd_len; } if (flags & SET_FIN) { cm_node->tcp_cntxt.loc_seq_num++; tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_FIN); } if (flags & SET_RST) tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_RST); tcph->th_off = (u16)((sizeof(*tcph) + opts_len + 3) >> 2); sqbuf->tcphlen = tcph->th_off << 2; tcph->th_win = htons(cm_node->tcp_cntxt.rcv_wnd); tcph->th_urp = 0; if (opts_len) { memcpy(buf, options->addr, opts_len); buf += opts_len; } if (hdr_len) { memcpy(buf, hdr->addr, hdr_len); buf += hdr_len; } if (pdata && pdata->addr) memcpy(buf, pdata->addr, pdata->size); atomic_set(&sqbuf->refcount, 1); irdma_debug_buf(vsi->dev, IRDMA_DEBUG_ILQ, "TRANSMIT ILQ BUFFER", sqbuf->mem.va, sqbuf->totallen); return sqbuf; } /** * irdma_form_uda_cm_frame - get a free packet and build frame full tcpip packet * @cm_node: connection's node ionfo to use in frame * @options: pointer to options info * @hdr: pointer mpa header * @pdata: pointer to private data * @flags: indicates FIN or ACK */ static struct irdma_puda_buf * irdma_form_uda_cm_frame(struct irdma_cm_node *cm_node, struct irdma_kmem_info *options, struct irdma_kmem_info *hdr, struct irdma_mpa_priv_info *pdata, u8 flags) { struct irdma_puda_buf *sqbuf; struct irdma_sc_vsi *vsi = &cm_node->iwdev->vsi; u8 *buf; struct tcphdr *tcph; struct ip *iph; struct ip6_hdr *ip6h; struct ether_header *ethh; u16 pktsize; u16 eth_hlen = ETH_HLEN; u32 opts_len = 0; u32 pd_len = 0; u32 hdr_len = 0; u16 vtag; sqbuf = irdma_puda_get_bufpool(vsi->ilq); if (!sqbuf) return NULL; buf = sqbuf->mem.va; if (options) opts_len = (u32)options->size; if (hdr) hdr_len = hdr->size; if (pdata) pd_len = pdata->size; if (cm_node->vlan_id < VLAN_N_VID) eth_hlen += 4; if (cm_node->ipv4) pktsize = sizeof(*iph) + sizeof(*tcph); else pktsize = sizeof(*ip6h) + sizeof(*tcph); pktsize += opts_len + hdr_len + pd_len; memset(buf, 0, eth_hlen + pktsize); sqbuf->totallen = pktsize + eth_hlen; sqbuf->maclen = eth_hlen; sqbuf->tcphlen = sizeof(*tcph) + opts_len; sqbuf->scratch = cm_node; ethh = (struct ether_header *)buf; buf += eth_hlen; if (cm_node->do_lpb) sqbuf->do_lpb = true; if (cm_node->ipv4) { sqbuf->ipv4 = true; iph = (struct ip *)buf; buf += sizeof(*iph); tcph = (struct tcphdr *)buf; buf += sizeof(*tcph); ether_addr_copy(ethh->ether_dhost, cm_node->rem_mac); ether_addr_copy(ethh->ether_shost, cm_node->loc_mac); if (cm_node->vlan_id < VLAN_N_VID) { ((struct ether_vlan_header *)ethh)->evl_proto = htons(ETH_P_8021Q); vtag = (cm_node->user_pri << VLAN_PRIO_SHIFT) | cm_node->vlan_id; ((struct ether_vlan_header *)ethh)->evl_tag = htons(vtag); ((struct ether_vlan_header *)ethh)->evl_encap_proto = htons(ETH_P_IP); } else { ethh->ether_type = htons(ETH_P_IP); } iph->ip_v = IPVERSION; iph->ip_hl = 5; /* 5 * 4Byte words, IP headr len */ iph->ip_tos = cm_node->tos; iph->ip_len = htons(pktsize); iph->ip_id = htons(++cm_node->tcp_cntxt.loc_id); iph->ip_off = htons(0x4000); iph->ip_ttl = 0x40; iph->ip_p = IPPROTO_TCP; iph->ip_src.s_addr = htonl(cm_node->loc_addr[0]); iph->ip_dst.s_addr = htonl(cm_node->rem_addr[0]); } else { sqbuf->ipv4 = false; ip6h = (struct ip6_hdr *)buf; buf += sizeof(*ip6h); tcph = (struct tcphdr *)buf; buf += sizeof(*tcph); ether_addr_copy(ethh->ether_dhost, cm_node->rem_mac); ether_addr_copy(ethh->ether_shost, cm_node->loc_mac); if (cm_node->vlan_id < VLAN_N_VID) { ((struct ether_vlan_header *)ethh)->evl_proto = htons(ETH_P_8021Q); vtag = (cm_node->user_pri << VLAN_PRIO_SHIFT) | cm_node->vlan_id; ((struct ether_vlan_header *)ethh)->evl_tag = htons(vtag); ((struct ether_vlan_header *)ethh)->evl_encap_proto = htons(ETH_P_IPV6); } else { ethh->ether_type = htons(ETH_P_IPV6); } ip6h->ip6_vfc = 6 << 4; ip6h->ip6_vfc |= cm_node->tos >> 4; ip6h->ip6_flow = cm_node->tos << 20; ip6h->ip6_plen = htons(pktsize - sizeof(*ip6h)); ip6h->ip6_nxt = 6; ip6h->ip6_hops = 128; irdma_copy_ip_htonl(ip6h->ip6_src.__u6_addr.__u6_addr32, cm_node->loc_addr); irdma_copy_ip_htonl(ip6h->ip6_dst.__u6_addr.__u6_addr32, cm_node->rem_addr); } tcph->th_sport = htons(cm_node->loc_port); tcph->th_dport = htons(cm_node->rem_port); tcph->th_seq = htonl(cm_node->tcp_cntxt.loc_seq_num); if (flags & SET_ACK) { cm_node->tcp_cntxt.loc_ack_num = cm_node->tcp_cntxt.rcv_nxt; tcph->th_ack = htonl(cm_node->tcp_cntxt.loc_ack_num); tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_ACK); } else { tcph->th_ack = 0; } if (flags & SET_SYN) { cm_node->tcp_cntxt.loc_seq_num++; tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_SYN); } else { cm_node->tcp_cntxt.loc_seq_num += hdr_len + pd_len; } if (flags & SET_FIN) { cm_node->tcp_cntxt.loc_seq_num++; tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_FIN); } if (flags & SET_RST) tcp_set_flags(tcph, tcp_get_flags(tcph) | TH_RST); tcph->th_off = (u16)((sizeof(*tcph) + opts_len + 3) >> 2); sqbuf->tcphlen = tcph->th_off << 2; tcph->th_win = htons(cm_node->tcp_cntxt.rcv_wnd); tcph->th_urp = 0; if (opts_len) { memcpy(buf, options->addr, opts_len); buf += opts_len; } if (hdr_len) { memcpy(buf, hdr->addr, hdr_len); buf += hdr_len; } if (pdata && pdata->addr) memcpy(buf, pdata->addr, pdata->size); atomic_set(&sqbuf->refcount, 1); irdma_debug_buf(vsi->dev, IRDMA_DEBUG_ILQ, "TRANSMIT ILQ BUFFER", sqbuf->mem.va, sqbuf->totallen); return sqbuf; } /** * irdma_send_reset - Send RST packet * @cm_node: connection's node */ int irdma_send_reset(struct irdma_cm_node *cm_node) { struct irdma_puda_buf *sqbuf; int flags = SET_RST | SET_ACK; sqbuf = cm_node->cm_core->form_cm_frame(cm_node, NULL, NULL, NULL, flags); if (!sqbuf) return -ENOMEM; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "caller: %pS cm_node %p cm_id=%p accel=%d state=%d rem_port=0x%04x, loc_port=0x%04x rem_addr=%x loc_addr=%x\n", __builtin_return_address(0), cm_node, cm_node->cm_id, cm_node->accelerated, cm_node->state, cm_node->rem_port, cm_node->loc_port, cm_node->rem_addr[0], cm_node->loc_addr[0]); return irdma_schedule_cm_timer(cm_node, sqbuf, IRDMA_TIMER_TYPE_SEND, 0, 1); } /** * irdma_active_open_err - send event for active side cm error * @cm_node: connection's node * @reset: Flag to send reset or not */ static void irdma_active_open_err(struct irdma_cm_node *cm_node, bool reset) { irdma_cleanup_retrans_entry(cm_node); cm_node->cm_core->stats_connect_errs++; if (reset) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_node=%p state=%d\n", cm_node, cm_node->state); atomic_inc(&cm_node->refcnt); irdma_send_reset(cm_node); } cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_create_event(cm_node, IRDMA_CM_EVENT_ABORTED); } /** * irdma_passive_open_err - handle passive side cm error * @cm_node: connection's node * @reset: send reset or just free cm_node */ static void irdma_passive_open_err(struct irdma_cm_node *cm_node, bool reset) { irdma_cleanup_retrans_entry(cm_node); cm_node->cm_core->stats_passive_errs++; cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_node=%p state=%d\n", cm_node, cm_node->state); if (reset) irdma_send_reset(cm_node); else irdma_rem_ref_cm_node(cm_node); } /** * irdma_event_connect_error - to create connect error event * @event: cm information for connect event */ static void irdma_event_connect_error(struct irdma_cm_event *event) { struct irdma_qp *iwqp; struct iw_cm_id *cm_id; cm_id = event->cm_node->cm_id; if (!cm_id) return; iwqp = cm_id->provider_data; if (!iwqp || !iwqp->iwdev) return; iwqp->cm_id = NULL; cm_id->provider_data = NULL; irdma_send_cm_event(event->cm_node, cm_id, IW_CM_EVENT_CONNECT_REPLY, -ECONNRESET); irdma_rem_ref_cm_node(event->cm_node); } /** * irdma_process_options - process options from TCP header * @cm_node: connection's node * @optionsloc: point to start of options * @optionsize: size of all options * @syn_pkt: flag if syn packet */ static int irdma_process_options(struct irdma_cm_node *cm_node, u8 *optionsloc, u32 optionsize, u32 syn_pkt) { u32 tmp; u32 offset = 0; union all_known_options *all_options; char got_mss_option = 0; while (offset < optionsize) { all_options = (union all_known_options *)(optionsloc + offset); switch (all_options->base.optionnum) { case OPTION_NUM_EOL: offset = optionsize; break; case OPTION_NUM_NONE: offset += 1; continue; case OPTION_NUM_MSS: irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "MSS Length: %d Offset: %d Size: %d\n", all_options->mss.len, offset, optionsize); got_mss_option = 1; if (all_options->mss.len != 4) return -EINVAL; tmp = ntohs(all_options->mss.mss); if ((cm_node->ipv4 && (tmp + IRDMA_MTU_TO_MSS_IPV4) < IRDMA_MIN_MTU_IPV4) || (!cm_node->ipv4 && (tmp + IRDMA_MTU_TO_MSS_IPV6) < IRDMA_MIN_MTU_IPV6)) return -EINVAL; if (tmp < cm_node->tcp_cntxt.mss) cm_node->tcp_cntxt.mss = tmp; break; case OPTION_NUM_WINDOW_SCALE: cm_node->tcp_cntxt.snd_wscale = all_options->windowscale.shiftcount; break; default: irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Unsupported TCP Option: %x\n", all_options->base.optionnum); break; } offset += all_options->base.len; } if (!got_mss_option && syn_pkt) cm_node->tcp_cntxt.mss = IRDMA_CM_DEFAULT_MSS; return 0; } /** * irdma_handle_tcp_options - setup TCP context info after parsing TCP options * @cm_node: connection's node * @tcph: pointer tcp header * @optionsize: size of options rcvd * @passive: active or passive flag */ static int irdma_handle_tcp_options(struct irdma_cm_node *cm_node, struct tcphdr *tcph, int optionsize, int passive) { u8 *optionsloc = (u8 *)&tcph[1]; int ret; if (optionsize) { ret = irdma_process_options(cm_node, optionsloc, optionsize, (u32)tcp_get_flags(tcph) & TH_SYN); if (ret) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Node %p, Sending Reset\n", cm_node); if (passive) irdma_passive_open_err(cm_node, true); else irdma_active_open_err(cm_node, true); return ret; } } cm_node->tcp_cntxt.snd_wnd = ntohs(tcph->th_win) << cm_node->tcp_cntxt.snd_wscale; if (cm_node->tcp_cntxt.snd_wnd > cm_node->tcp_cntxt.max_snd_wnd) cm_node->tcp_cntxt.max_snd_wnd = cm_node->tcp_cntxt.snd_wnd; return 0; } /** * irdma_build_mpa_v1 - build a MPA V1 frame * @cm_node: connection's node * @start_addr: address where to build frame * @mpa_key: to do read0 or write0 */ static void irdma_build_mpa_v1(struct irdma_cm_node *cm_node, void *start_addr, u8 mpa_key) { struct ietf_mpa_v1 *mpa_frame = start_addr; switch (mpa_key) { case MPA_KEY_REQUEST: memcpy(mpa_frame->key, IEFT_MPA_KEY_REQ, IETF_MPA_KEY_SIZE); break; case MPA_KEY_REPLY: memcpy(mpa_frame->key, IEFT_MPA_KEY_REP, IETF_MPA_KEY_SIZE); break; default: break; } mpa_frame->flags = IETF_MPA_FLAGS_CRC; mpa_frame->rev = cm_node->mpa_frame_rev; mpa_frame->priv_data_len = htons(cm_node->pdata.size); } /** * irdma_build_mpa_v2 - build a MPA V2 frame * @cm_node: connection's node * @start_addr: buffer start address * @mpa_key: to do read0 or write0 */ static void irdma_build_mpa_v2(struct irdma_cm_node *cm_node, void *start_addr, u8 mpa_key) { struct ietf_mpa_v2 *mpa_frame = start_addr; struct ietf_rtr_msg *rtr_msg = &mpa_frame->rtr_msg; u16 ctrl_ird, ctrl_ord; /* initialize the upper 5 bytes of the frame */ irdma_build_mpa_v1(cm_node, start_addr, mpa_key); mpa_frame->flags |= IETF_MPA_V2_FLAG; if (cm_node->iwdev->iw_ooo) { mpa_frame->flags |= IETF_MPA_FLAGS_MARKERS; cm_node->rcv_mark_en = true; } mpa_frame->priv_data_len = cpu_to_be16(be16_to_cpu(mpa_frame->priv_data_len) + IETF_RTR_MSG_SIZE); /* initialize RTR msg */ if (cm_node->mpav2_ird_ord == IETF_NO_IRD_ORD) { ctrl_ird = IETF_NO_IRD_ORD; ctrl_ord = IETF_NO_IRD_ORD; } else { ctrl_ird = (cm_node->ird_size > IETF_NO_IRD_ORD) ? IETF_NO_IRD_ORD : cm_node->ird_size; ctrl_ord = (cm_node->ord_size > IETF_NO_IRD_ORD) ? IETF_NO_IRD_ORD : cm_node->ord_size; } ctrl_ird |= IETF_PEER_TO_PEER; switch (mpa_key) { case MPA_KEY_REQUEST: ctrl_ord |= IETF_RDMA0_WRITE; ctrl_ord |= IETF_RDMA0_READ; break; case MPA_KEY_REPLY: switch (cm_node->send_rdma0_op) { case SEND_RDMA_WRITE_ZERO: ctrl_ord |= IETF_RDMA0_WRITE; break; case SEND_RDMA_READ_ZERO: ctrl_ord |= IETF_RDMA0_READ; break; } break; default: break; } rtr_msg->ctrl_ird = htons(ctrl_ird); rtr_msg->ctrl_ord = htons(ctrl_ord); } /** * irdma_cm_build_mpa_frame - build mpa frame for mpa version 1 or version 2 * @cm_node: connection's node * @mpa: mpa: data buffer * @mpa_key: to do read0 or write0 */ static int irdma_cm_build_mpa_frame(struct irdma_cm_node *cm_node, struct irdma_kmem_info *mpa, u8 mpa_key) { int hdr_len = 0; switch (cm_node->mpa_frame_rev) { case IETF_MPA_V1: hdr_len = sizeof(struct ietf_mpa_v1); irdma_build_mpa_v1(cm_node, mpa->addr, mpa_key); break; case IETF_MPA_V2: hdr_len = sizeof(struct ietf_mpa_v2); irdma_build_mpa_v2(cm_node, mpa->addr, mpa_key); break; default: break; } return hdr_len; } /** * irdma_send_mpa_request - active node send mpa request to passive node * @cm_node: connection's node */ static int irdma_send_mpa_request(struct irdma_cm_node *cm_node) { struct irdma_puda_buf *sqbuf; cm_node->mpa_hdr.addr = &cm_node->mpa_v2_frame; cm_node->mpa_hdr.size = irdma_cm_build_mpa_frame(cm_node, &cm_node->mpa_hdr, MPA_KEY_REQUEST); if (!cm_node->mpa_hdr.size) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "mpa size = %d\n", cm_node->mpa_hdr.size); return -EINVAL; } sqbuf = cm_node->cm_core->form_cm_frame(cm_node, NULL, &cm_node->mpa_hdr, &cm_node->pdata, SET_ACK); if (!sqbuf) return -ENOMEM; return irdma_schedule_cm_timer(cm_node, sqbuf, IRDMA_TIMER_TYPE_SEND, 1, 0); } /** * irdma_send_mpa_reject - * @cm_node: connection's node * @pdata: reject data for connection * @plen: length of reject data */ static int irdma_send_mpa_reject(struct irdma_cm_node *cm_node, const void *pdata, u8 plen) { struct irdma_puda_buf *sqbuf; struct irdma_mpa_priv_info priv_info; cm_node->mpa_hdr.addr = &cm_node->mpa_v2_frame; cm_node->mpa_hdr.size = irdma_cm_build_mpa_frame(cm_node, &cm_node->mpa_hdr, MPA_KEY_REPLY); cm_node->mpa_v2_frame.flags |= IETF_MPA_FLAGS_REJECT; priv_info.addr = pdata; priv_info.size = plen; sqbuf = cm_node->cm_core->form_cm_frame(cm_node, NULL, &cm_node->mpa_hdr, &priv_info, SET_ACK | SET_FIN); if (!sqbuf) return -ENOMEM; cm_node->state = IRDMA_CM_STATE_FIN_WAIT1; return irdma_schedule_cm_timer(cm_node, sqbuf, IRDMA_TIMER_TYPE_SEND, 1, 0); } /** * irdma_negotiate_mpa_v2_ird_ord - negotiate MPAv2 IRD/ORD * @cm_node: connection's node * @buf: Data pointer */ static int irdma_negotiate_mpa_v2_ird_ord(struct irdma_cm_node *cm_node, u8 *buf) { struct ietf_mpa_v2 *mpa_v2_frame; struct ietf_rtr_msg *rtr_msg; u16 ird_size; u16 ord_size; u16 ctrl_ord; u16 ctrl_ird; mpa_v2_frame = (struct ietf_mpa_v2 *)buf; rtr_msg = &mpa_v2_frame->rtr_msg; /* parse rtr message */ ctrl_ord = ntohs(rtr_msg->ctrl_ord); ctrl_ird = ntohs(rtr_msg->ctrl_ird); ird_size = ctrl_ird & IETF_NO_IRD_ORD; ord_size = ctrl_ord & IETF_NO_IRD_ORD; if (!(ctrl_ird & IETF_PEER_TO_PEER)) return -EOPNOTSUPP; if (ird_size == IETF_NO_IRD_ORD || ord_size == IETF_NO_IRD_ORD) { cm_node->mpav2_ird_ord = IETF_NO_IRD_ORD; goto negotiate_done; } if (cm_node->state != IRDMA_CM_STATE_MPAREQ_SENT) { /* responder */ if (!ord_size && (ctrl_ord & IETF_RDMA0_READ)) cm_node->ird_size = 1; if (cm_node->ord_size > ird_size) cm_node->ord_size = ird_size; } else { /* initiator */ if (!ird_size && (ctrl_ord & IETF_RDMA0_READ)) /* Remote peer doesn't support RDMA0_READ */ return -EOPNOTSUPP; if (cm_node->ord_size > ird_size) cm_node->ord_size = ird_size; if (cm_node->ird_size < ord_size) /* no resources available */ return -EINVAL; } negotiate_done: if (ctrl_ord & IETF_RDMA0_READ) cm_node->send_rdma0_op = SEND_RDMA_READ_ZERO; else if (ctrl_ord & IETF_RDMA0_WRITE) cm_node->send_rdma0_op = SEND_RDMA_WRITE_ZERO; else /* Not supported RDMA0 operation */ return -EOPNOTSUPP; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "MPAV2 Negotiated ORD: %d, IRD: %d\n", cm_node->ord_size, cm_node->ird_size); return 0; } /** * irdma_parse_mpa - process an IETF MPA frame * @cm_node: connection's node * @buf: Data pointer * @type: to return accept or reject * @len: Len of mpa buffer */ static int irdma_parse_mpa(struct irdma_cm_node *cm_node, u8 *buf, u32 *type, u32 len) { struct ietf_mpa_v1 *mpa_frame; int mpa_hdr_len, priv_data_len, ret; *type = IRDMA_MPA_REQUEST_ACCEPT; if (len < sizeof(*mpa_frame)) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "ietf buffer small (%x)\n", len); return -EINVAL; } mpa_frame = (struct ietf_mpa_v1 *)buf; mpa_hdr_len = sizeof(*mpa_frame); priv_data_len = ntohs(mpa_frame->priv_data_len); if (priv_data_len > IETF_MAX_PRIV_DATA_LEN) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "private_data too big %d\n", priv_data_len); return -EOVERFLOW; } if (mpa_frame->rev != IETF_MPA_V1 && mpa_frame->rev != IETF_MPA_V2) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "unsupported mpa rev = %d\n", mpa_frame->rev); return -EINVAL; } if (mpa_frame->rev > cm_node->mpa_frame_rev) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "rev %d\n", mpa_frame->rev); return -EINVAL; } cm_node->mpa_frame_rev = mpa_frame->rev; if (cm_node->state != IRDMA_CM_STATE_MPAREQ_SENT) { if (memcmp(mpa_frame->key, IEFT_MPA_KEY_REQ, IETF_MPA_KEY_SIZE)) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Unexpected MPA Key received\n"); return -EINVAL; } } else { if (memcmp(mpa_frame->key, IEFT_MPA_KEY_REP, IETF_MPA_KEY_SIZE)) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Unexpected MPA Key received\n"); return -EINVAL; } } if (priv_data_len + mpa_hdr_len > len) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "ietf buffer len(%x + %x != %x)\n", priv_data_len, mpa_hdr_len, len); return -EOVERFLOW; } if (len > IRDMA_MAX_CM_BUF) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "ietf buffer large len = %d\n", len); return -EOVERFLOW; } switch (mpa_frame->rev) { case IETF_MPA_V2: mpa_hdr_len += IETF_RTR_MSG_SIZE; ret = irdma_negotiate_mpa_v2_ird_ord(cm_node, buf); if (ret) return ret; break; case IETF_MPA_V1: default: break; } memcpy(cm_node->pdata_buf, buf + mpa_hdr_len, priv_data_len); cm_node->pdata.size = priv_data_len; if (mpa_frame->flags & IETF_MPA_FLAGS_REJECT) *type = IRDMA_MPA_REQUEST_REJECT; if (mpa_frame->flags & IETF_MPA_FLAGS_MARKERS) cm_node->snd_mark_en = true; return 0; } /** * irdma_schedule_cm_timer * @cm_node: connection's node * @sqbuf: buffer to send * @type: if it is send or close * @send_retrans: if rexmits to be done * @close_when_complete: is cm_node to be removed * * note - cm_node needs to be protected before calling this. Encase in: * irdma_rem_ref_cm_node(cm_core, cm_node); * irdma_schedule_cm_timer(...) * atomic_inc(&cm_node->refcnt); */ int irdma_schedule_cm_timer(struct irdma_cm_node *cm_node, struct irdma_puda_buf *sqbuf, enum irdma_timer_type type, int send_retrans, int close_when_complete) { struct irdma_sc_vsi *vsi = &cm_node->iwdev->vsi; struct irdma_cm_core *cm_core = cm_node->cm_core; struct irdma_timer_entry *new_send; u32 was_timer_set; unsigned long flags; new_send = kzalloc(sizeof(*new_send), GFP_ATOMIC); if (!new_send) { if (type != IRDMA_TIMER_TYPE_CLOSE) irdma_free_sqbuf(vsi, sqbuf); return -ENOMEM; } new_send->retrycount = IRDMA_DEFAULT_RETRYS; new_send->retranscount = IRDMA_DEFAULT_RETRANS; new_send->sqbuf = sqbuf; new_send->timetosend = jiffies; new_send->type = type; new_send->send_retrans = send_retrans; new_send->close_when_complete = close_when_complete; if (type == IRDMA_TIMER_TYPE_CLOSE) { new_send->timetosend += (HZ / 10); if (cm_node->close_entry) { kfree(new_send); irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "already close entry\n"); return -EINVAL; } cm_node->close_entry = new_send; } else { /* type == IRDMA_TIMER_TYPE_SEND */ spin_lock_irqsave(&cm_node->retrans_list_lock, flags); cm_node->send_entry = new_send; atomic_inc(&cm_node->refcnt); spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); new_send->timetosend = jiffies + IRDMA_RETRY_TIMEOUT; atomic_inc(&sqbuf->refcount); irdma_puda_send_buf(vsi->ilq, sqbuf); if (!send_retrans) { irdma_cleanup_retrans_entry(cm_node); if (close_when_complete) irdma_rem_ref_cm_node(cm_node); return 0; } } spin_lock_irqsave(&cm_core->ht_lock, flags); was_timer_set = timer_pending(&cm_core->tcp_timer); if (!was_timer_set) { cm_core->tcp_timer.expires = new_send->timetosend; add_timer(&cm_core->tcp_timer); } spin_unlock_irqrestore(&cm_core->ht_lock, flags); return 0; } /** * irdma_retrans_expired - Could not rexmit the packet * @cm_node: connection's node */ static void irdma_retrans_expired(struct irdma_cm_node *cm_node) { enum irdma_cm_node_state state = cm_node->state; cm_node->state = IRDMA_CM_STATE_CLOSED; switch (state) { case IRDMA_CM_STATE_SYN_RCVD: case IRDMA_CM_STATE_CLOSING: irdma_rem_ref_cm_node(cm_node); break; case IRDMA_CM_STATE_FIN_WAIT1: case IRDMA_CM_STATE_LAST_ACK: irdma_send_reset(cm_node); break; default: atomic_inc(&cm_node->refcnt); irdma_send_reset(cm_node); irdma_create_event(cm_node, IRDMA_CM_EVENT_ABORTED); break; } } /** * irdma_handle_close_entry - for handling retry/timeouts * @cm_node: connection's node * @rem_node: flag for remove cm_node */ static void irdma_handle_close_entry(struct irdma_cm_node *cm_node, u32 rem_node) { struct irdma_timer_entry *close_entry = cm_node->close_entry; struct irdma_qp *iwqp; unsigned long flags; if (!close_entry) return; iwqp = (struct irdma_qp *)close_entry->sqbuf; if (iwqp) { spin_lock_irqsave(&iwqp->lock, flags); if (iwqp->cm_id) { iwqp->hw_tcp_state = IRDMA_TCP_STATE_CLOSED; iwqp->hw_iwarp_state = IRDMA_QP_STATE_ERROR; iwqp->last_aeq = IRDMA_AE_RESET_SENT; iwqp->ibqp_state = IB_QPS_ERR; spin_unlock_irqrestore(&iwqp->lock, flags); irdma_cm_disconn(iwqp); } else { spin_unlock_irqrestore(&iwqp->lock, flags); } } else if (rem_node) { /* TIME_WAIT state */ irdma_rem_ref_cm_node(cm_node); } kfree(close_entry); cm_node->close_entry = NULL; } /** * irdma_cm_timer_tick - system's timer expired callback * @t: Pointer to timer_list */ static void irdma_cm_timer_tick(struct timer_list *t) { unsigned long nexttimeout = jiffies + IRDMA_LONG_TIME; struct irdma_cm_node *cm_node; struct irdma_timer_entry *send_entry, *close_entry; struct list_head *list_core_temp; struct list_head *list_node; struct irdma_cm_core *cm_core = timer_container_of(cm_core, t, tcp_timer); struct irdma_sc_vsi *vsi; u32 settimer = 0; unsigned long timetosend; unsigned long flags; struct list_head timer_list; INIT_LIST_HEAD(&timer_list); rcu_read_lock(); irdma_timer_list_prep(cm_core, &timer_list); rcu_read_unlock(); list_for_each_safe(list_node, list_core_temp, &timer_list) { cm_node = container_of(list_node, struct irdma_cm_node, timer_entry); close_entry = cm_node->close_entry; if (close_entry) { if (time_after(close_entry->timetosend, jiffies)) { if (nexttimeout > close_entry->timetosend || !settimer) { nexttimeout = close_entry->timetosend; settimer = 1; } } else { irdma_handle_close_entry(cm_node, 1); } } spin_lock_irqsave(&cm_node->retrans_list_lock, flags); send_entry = cm_node->send_entry; if (!send_entry) goto done; if (time_after(send_entry->timetosend, jiffies)) { if (cm_node->state != IRDMA_CM_STATE_OFFLOADED) { if (nexttimeout > send_entry->timetosend || !settimer) { nexttimeout = send_entry->timetosend; settimer = 1; } } else { irdma_free_retrans_entry(cm_node); } goto done; } if (cm_node->state == IRDMA_CM_STATE_OFFLOADED || cm_node->state == IRDMA_CM_STATE_CLOSED) { irdma_free_retrans_entry(cm_node); goto done; } if (!send_entry->retranscount || !send_entry->retrycount) { irdma_free_retrans_entry(cm_node); spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); irdma_retrans_expired(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSED; spin_lock_irqsave(&cm_node->retrans_list_lock, flags); goto done; } spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); vsi = &cm_node->iwdev->vsi; if (!cm_node->ack_rcvd) { atomic_inc(&send_entry->sqbuf->refcount); irdma_puda_send_buf(vsi->ilq, send_entry->sqbuf); cm_node->cm_core->stats_pkt_retrans++; } spin_lock_irqsave(&cm_node->retrans_list_lock, flags); if (send_entry->send_retrans) { send_entry->retranscount--; timetosend = (IRDMA_RETRY_TIMEOUT << (IRDMA_DEFAULT_RETRANS - send_entry->retranscount)); send_entry->timetosend = jiffies + min(timetosend, IRDMA_MAX_TIMEOUT); if (nexttimeout > send_entry->timetosend || !settimer) { nexttimeout = send_entry->timetosend; settimer = 1; } } else { int close_when_complete; close_when_complete = send_entry->close_when_complete; irdma_free_retrans_entry(cm_node); if (close_when_complete) irdma_rem_ref_cm_node(cm_node); } done: spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); irdma_rem_ref_cm_node(cm_node); } if (settimer) { spin_lock_irqsave(&cm_core->ht_lock, flags); if (!timer_pending(&cm_core->tcp_timer)) { cm_core->tcp_timer.expires = nexttimeout; add_timer(&cm_core->tcp_timer); } spin_unlock_irqrestore(&cm_core->ht_lock, flags); } } /** * irdma_send_syn - send SYN packet * @cm_node: connection's node * @sendack: flag to set ACK bit or not */ int irdma_send_syn(struct irdma_cm_node *cm_node, u32 sendack) { struct irdma_puda_buf *sqbuf; int flags = SET_SYN; char optionsbuf[sizeof(struct option_mss) + sizeof(struct option_windowscale) + sizeof(struct option_base) + TCP_OPTIONS_PADDING]; struct irdma_kmem_info opts; int optionssize = 0; /* Sending MSS option */ union all_known_options *options; opts.addr = optionsbuf; if (!cm_node) return -EINVAL; options = (union all_known_options *)&optionsbuf[optionssize]; options->mss.optionnum = OPTION_NUM_MSS; options->mss.len = sizeof(struct option_mss); options->mss.mss = htons(cm_node->tcp_cntxt.mss); optionssize += sizeof(struct option_mss); options = (union all_known_options *)&optionsbuf[optionssize]; options->windowscale.optionnum = OPTION_NUM_WINDOW_SCALE; options->windowscale.len = sizeof(struct option_windowscale); options->windowscale.shiftcount = cm_node->tcp_cntxt.rcv_wscale; optionssize += sizeof(struct option_windowscale); options = (union all_known_options *)&optionsbuf[optionssize]; options->eol = OPTION_NUM_EOL; optionssize += 1; if (sendack) flags |= SET_ACK; opts.size = optionssize; sqbuf = cm_node->cm_core->form_cm_frame(cm_node, &opts, NULL, NULL, flags); if (!sqbuf) return -ENOMEM; return irdma_schedule_cm_timer(cm_node, sqbuf, IRDMA_TIMER_TYPE_SEND, 1, 0); } /** * irdma_send_ack - Send ACK packet * @cm_node: connection's node */ void irdma_send_ack(struct irdma_cm_node *cm_node) { struct irdma_puda_buf *sqbuf; struct irdma_sc_vsi *vsi = &cm_node->iwdev->vsi; sqbuf = cm_node->cm_core->form_cm_frame(cm_node, NULL, NULL, NULL, SET_ACK); if (sqbuf) irdma_puda_send_buf(vsi->ilq, sqbuf); } /** * irdma_send_fin - Send FIN pkt * @cm_node: connection's node */ static int irdma_send_fin(struct irdma_cm_node *cm_node) { struct irdma_puda_buf *sqbuf; sqbuf = cm_node->cm_core->form_cm_frame(cm_node, NULL, NULL, NULL, SET_ACK | SET_FIN); if (!sqbuf) return -ENOMEM; return irdma_schedule_cm_timer(cm_node, sqbuf, IRDMA_TIMER_TYPE_SEND, 1, 0); } /** * irdma_find_listener - find a cm node listening on this addr-port pair * @cm_core: cm's core * @dst_addr: listener ip addr * @ipv4: flag indicating IPv4 when true * @dst_port: listener tcp port num * @vlan_id: virtual LAN ID * @listener_state: state to match with listen node's */ static struct irdma_cm_listener * irdma_find_listener(struct irdma_cm_core *cm_core, u32 *dst_addr, bool ipv4, u16 dst_port, u16 vlan_id, enum irdma_cm_listener_state listener_state) { struct irdma_cm_listener *listen_node; static const u32 ip_zero[4] = {0, 0, 0, 0}; u32 listen_addr[4]; u16 listen_port; unsigned long flags; /* walk list and find cm_node associated with this session ID */ spin_lock_irqsave(&cm_core->listen_list_lock, flags); list_for_each_entry(listen_node, &cm_core->listen_list, list) { memcpy(listen_addr, listen_node->loc_addr, sizeof(listen_addr)); listen_port = listen_node->loc_port; if (listen_node->ipv4 != ipv4 || listen_port != dst_port || !(listener_state & listen_node->listener_state)) continue; /* compare node pair, return node handle if a match */ if (!memcmp(listen_addr, ip_zero, sizeof(listen_addr)) || (!memcmp(listen_addr, dst_addr, sizeof(listen_addr)) && vlan_id == listen_node->vlan_id)) { atomic_inc(&listen_node->refcnt); spin_unlock_irqrestore(&cm_core->listen_list_lock, flags); return listen_node; } } spin_unlock_irqrestore(&cm_core->listen_list_lock, flags); return NULL; } /** * irdma_del_multiple_qhash - Remove qhash and child listens * @iwdev: iWarp device * @cm_info: CM info for parent listen node * @cm_parent_listen_node: The parent listen node */ static int irdma_del_multiple_qhash(struct irdma_device *iwdev, struct irdma_cm_info *cm_info, struct irdma_cm_listener *cm_parent_listen_node) { struct irdma_cm_listener *child_listen_node; struct list_head *pos, *tpos; unsigned long flags; int ret = -EINVAL; spin_lock_irqsave(&iwdev->cm_core.listen_list_lock, flags); list_for_each_safe(pos, tpos, &cm_parent_listen_node->child_listen_list) { child_listen_node = list_entry(pos, struct irdma_cm_listener, child_listen_list); if (child_listen_node->ipv4) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "removing child listen for IP=%x, port=%d, vlan=%d\n", child_listen_node->loc_addr[0], child_listen_node->loc_port, child_listen_node->vlan_id); else irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "removing child listen for IP=%x:%x:%x:%x, port=%d, vlan=%d\n", IRDMA_PRINT_IP6(child_listen_node->loc_addr), child_listen_node->loc_port, child_listen_node->vlan_id); list_del(pos); memcpy(cm_info->loc_addr, child_listen_node->loc_addr, sizeof(cm_info->loc_addr)); cm_info->vlan_id = child_listen_node->vlan_id; if (child_listen_node->qhash_set) { ret = irdma_manage_qhash(iwdev, cm_info, IRDMA_QHASH_TYPE_TCP_SYN, IRDMA_QHASH_MANAGE_TYPE_DELETE, NULL, false); child_listen_node->qhash_set = false; } else { ret = 0; } irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Child listen node freed = %p\n", child_listen_node); kfree(child_listen_node); cm_parent_listen_node->cm_core->stats_listen_nodes_destroyed++; } spin_unlock_irqrestore(&iwdev->cm_core.listen_list_lock, flags); return ret; } static u8 irdma_iw_get_vlan_prio(u32 *loc_addr, u8 prio, bool ipv4) { return prio; } /** * irdma_get_vlan_mac_ipv6 - Get the vlan and mac for an IPv6 * address * @addr: local IPv6 address * @vlan_id: vlan id for the given IPv6 address * @mac: mac address for the given IPv6 address * * Returns the net_device of the IPv6 address and also sets the * vlan id and mac for that address. */ void irdma_get_vlan_mac_ipv6(struct iw_cm_id *cm_id, u32 *addr, u16 *vlan_id, u8 *mac) { if_t ip_dev = NULL; struct in6_addr laddr6; struct vnet *vnet = &init_net; struct ifaddr *ifa; u16 scope_id = 0; irdma_copy_ip_htonl(laddr6.__u6_addr.__u6_addr32, addr); if (vlan_id) *vlan_id = 0xFFFF; /* Match rdma_vlan_dev_vlan_id() */ if (mac) eth_zero_addr(mac); if (IN6_IS_SCOPE_LINKLOCAL(&laddr6) || IN6_IS_ADDR_MC_INTFACELOCAL(&laddr6)) scope_id = ntohs(laddr6.__u6_addr.__u6_addr16[1]); #ifdef VIMAGE vnet = irdma_cmid_to_vnet(cm_id); #endif ip_dev = ip6_ifp_find(vnet, laddr6, scope_id); if (ip_dev) { if (vlan_id) *vlan_id = rdma_vlan_dev_vlan_id(ip_dev); ifa = if_getifaddr(ip_dev); if (ifa && ifa->ifa_addr && mac) ether_addr_copy(mac, if_getlladdr(ip_dev)); } } /** * irdma_get_vlan_ipv4 - Returns the vlan_id for IPv4 address * @addr: local IPv4 address */ u16 irdma_get_vlan_ipv4(struct iw_cm_id *cm_id, u32 *addr) { u16 vlan_id = 0xFFFF; #ifdef INET if_t netdev; struct vnet *vnet = &init_net; #ifdef VIMAGE vnet = irdma_cmid_to_vnet(cm_id); #endif netdev = ip_ifp_find(vnet, htonl(addr[0])); if (netdev) { vlan_id = rdma_vlan_dev_vlan_id(netdev); dev_put(netdev); } #endif return vlan_id; } -static int -irdma_manage_qhash_wait(struct irdma_pci_f *rf, struct irdma_cm_info *cm_info) -{ - struct irdma_cqp_request *cqp_request = cm_info->cqp_request; - int cnt = rf->sc_dev.hw_attrs.max_cqp_compl_wait_time_ms * CQP_TIMEOUT_THRESHOLD; - u32 ret_val; - - if (!cqp_request) - return -ENOMEM; - do { - irdma_cqp_ce_handler(rf, &rf->ccq.sc_cq); - mdelay(1); - } while (!READ_ONCE(cqp_request->request_done) && --cnt); - - ret_val = cqp_request->compl_info.op_ret_val; - irdma_put_cqp_request(&rf->cqp, cqp_request); - if (cnt) { - if (!ret_val) - return 0; - return -EINVAL; - } - - return -ETIMEDOUT; -} - /** * irdma_add_mqh_ifa_cb - Adds multiple qhashes for IPv4/IPv6 * @arg: Calback argument structure from irdma_add_mqh * @ifa: Current address to compute against * @count: Current cumulative output of all callbacks in this iteration * * Adds a qhash and a child listen node for a single IPv4/IPv6 address * on the adapter and adds the associated qhash filter */ static u_int irdma_add_mqh_ifa_cb(void *arg, struct ifaddr *ifa, u_int count) { struct irdma_add_mqh_cbs *cbs = arg; struct irdma_cm_listener *child_listen_node; struct irdma_cm_info *cm_info = cbs->cm_info; struct irdma_device *iwdev = cbs->iwdev; struct irdma_cm_listener *cm_parent_listen_node = cbs->cm_listen_node; if_t ip_dev = ifa->ifa_ifp; unsigned long flags; int ret; if (count) return 0; child_listen_node = kzalloc(sizeof(*child_listen_node), GFP_ATOMIC); if (!child_listen_node) { irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "listener memory allocation\n"); return -ENOMEM; } memcpy(child_listen_node, cm_parent_listen_node, sizeof(*child_listen_node)); cm_info->vlan_id = rdma_vlan_dev_vlan_id(ip_dev); child_listen_node->vlan_id = cm_info->vlan_id; if (cm_info->ipv4) { irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Allocating child CM Listener forIP=%x, vlan_id=%d, MAC=%x:%x:%x:%x:%x:%x\n", ((struct sockaddr_in *)&ifa->ifa_addr)->sin_addr.s_addr, rdma_vlan_dev_vlan_id(ip_dev), if_getlladdr(ip_dev)[0], if_getlladdr(ip_dev)[1], if_getlladdr(ip_dev)[2], if_getlladdr(ip_dev)[3], if_getlladdr(ip_dev)[4], if_getlladdr(ip_dev)[5]); child_listen_node->loc_addr[0] = ntohl(((struct sockaddr_in *)ifa->ifa_addr)->sin_addr.s_addr); } else { irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "IP=%x:%x:%x:%x, vlan_id=%d, MAC=%x:%x:%x:%x:%x:%x\n", IRDMA_PRINT_IP6(&((struct sockaddr_in6 *)ifa->ifa_addr)->sin6_addr), rdma_vlan_dev_vlan_id(ip_dev), if_getlladdr(ip_dev)[0], if_getlladdr(ip_dev)[1], if_getlladdr(ip_dev)[2], if_getlladdr(ip_dev)[3], if_getlladdr(ip_dev)[4], if_getlladdr(ip_dev)[5]); irdma_copy_ip_ntohl(child_listen_node->loc_addr, ((struct sockaddr_in6 *)ifa->ifa_addr)->sin6_addr.__u6_addr.__u6_addr32); } memcpy(cm_info->loc_addr, child_listen_node->loc_addr, sizeof(cm_info->loc_addr)); if (!iwdev->vsi.dscp_mode) cm_info->user_pri = irdma_iw_get_vlan_prio(child_listen_node->loc_addr, cm_info->user_pri, cm_info->ipv4); - ret = irdma_manage_qhash(iwdev, cm_info, - IRDMA_QHASH_TYPE_TCP_SYN, - IRDMA_QHASH_MANAGE_TYPE_ADD, - NULL, false); - if (ret) { - kfree(child_listen_node); - return ret; - } - /* wait for qhash finish */ - ret = irdma_manage_qhash_wait(iwdev->rf, cm_info); + ret = irdma_add_qhash_wait_no_lock(iwdev, cm_info); if (ret) { kfree(child_listen_node); return ret; } child_listen_node->qhash_set = true; spin_lock_irqsave(&iwdev->cm_core.listen_list_lock, flags); list_add(&child_listen_node->child_listen_list, &cm_parent_listen_node->child_listen_list); spin_unlock_irqrestore(&iwdev->cm_core.listen_list_lock, flags); cm_parent_listen_node->cm_core->stats_listen_nodes_created++; return 0; } /** * irdma_add_mqh - Adds multiple qhashes * @iwdev: iWarp device * @cm_info: CM info for parent listen node * @cm_listen_node: The parent listen node */ static int irdma_add_mqh(struct irdma_device *iwdev, struct irdma_cm_info *cm_info, struct irdma_cm_listener *cm_listen_node) { struct epoch_tracker et; struct irdma_add_mqh_cbs cbs; struct if_iter iter; if_t ifp; int err = -ENOENT; cbs.iwdev = iwdev; cbs.cm_info = cm_info; cbs.cm_listen_node = cm_listen_node; VNET_ITERATOR_DECL(vnet_iter); VNET_LIST_RLOCK(); NET_EPOCH_ENTER(et); VNET_FOREACH(vnet_iter) { CURVNET_SET_QUIET(vnet_iter); for (ifp = if_iter_start(&iter); ifp != NULL; ifp = if_iter_next(&iter)) { if (!(if_getflags(ifp) & IFF_UP)) continue; if (((rdma_vlan_dev_vlan_id(ifp) >= VLAN_N_VID) || (rdma_vlan_dev_real_dev(ifp) != iwdev->netdev)) && ifp != iwdev->netdev) continue; if_addr_rlock(ifp); if (cm_info->ipv4) err = if_foreach_addr_type(ifp, AF_INET, irdma_add_mqh_ifa_cb, &cbs); else err = if_foreach_addr_type(ifp, AF_INET6, irdma_add_mqh_ifa_cb, &cbs); if_addr_runlock(ifp); } if_iter_finish(&iter); CURVNET_RESTORE(); } NET_EPOCH_EXIT(et); VNET_LIST_RUNLOCK(); return err; } /** * irdma_reset_list_prep - add connection nodes slated for reset to list * @cm_core: cm's core * @listener: pointer to listener node * @reset_list: a list to which cm_node will be selected */ static void irdma_reset_list_prep(struct irdma_cm_core *cm_core, struct irdma_cm_listener *listener, struct list_head *reset_list) { struct irdma_cm_node *cm_node; int bkt; HASH_FOR_EACH_RCU(cm_core->cm_hash_tbl, bkt, cm_node, list) { if (cm_node->listener == listener && !cm_node->accelerated && atomic_inc_not_zero(&cm_node->refcnt)) list_add(&cm_node->reset_entry, reset_list); } } /** * irdma_dec_refcnt_listen - delete listener and associated cm nodes * @cm_core: cm's core * @listener: pointer to listener node * @free_hanging_nodes: to free associated cm_nodes * @apbvt_del: flag to delete the apbvt */ static int irdma_dec_refcnt_listen(struct irdma_cm_core *cm_core, struct irdma_cm_listener *listener, int free_hanging_nodes, bool apbvt_del) { struct list_head *list_pos; struct list_head *list_temp; struct irdma_cm_node *cm_node; struct list_head reset_list; struct irdma_cm_info nfo; enum irdma_cm_node_state old_state; unsigned long flags; int err; /* free non-accelerated child nodes for this listener */ INIT_LIST_HEAD(&reset_list); if (free_hanging_nodes) { rcu_read_lock(); irdma_reset_list_prep(cm_core, listener, &reset_list); rcu_read_unlock(); } list_for_each_safe(list_pos, list_temp, &reset_list) { cm_node = container_of(list_pos, struct irdma_cm_node, reset_entry); if (cm_node->state >= IRDMA_CM_STATE_FIN_WAIT1) { irdma_rem_ref_cm_node(cm_node); continue; } irdma_cleanup_retrans_entry(cm_node); err = irdma_send_reset(cm_node); if (err) { cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "send reset failed\n"); } else { old_state = cm_node->state; cm_node->state = IRDMA_CM_STATE_LISTENER_DESTROYED; if (old_state != IRDMA_CM_STATE_MPAREQ_RCVD) irdma_rem_ref_cm_node(cm_node); } } if (atomic_dec_and_test(&listener->refcnt)) { spin_lock_irqsave(&cm_core->listen_list_lock, flags); list_del(&listener->list); spin_unlock_irqrestore(&cm_core->listen_list_lock, flags); if (apbvt_del) irdma_del_apbvt(listener->iwdev, listener->apbvt_entry); memcpy(nfo.loc_addr, listener->loc_addr, sizeof(nfo.loc_addr)); nfo.loc_port = listener->loc_port; nfo.ipv4 = listener->ipv4; nfo.vlan_id = listener->vlan_id; nfo.user_pri = listener->user_pri; nfo.qh_qpid = listener->iwdev->vsi.ilq->qp_id; if (!list_empty(&listener->child_listen_list)) { irdma_del_multiple_qhash(listener->iwdev, &nfo, listener); } else { if (listener->qhash_set) irdma_manage_qhash(listener->iwdev, &nfo, IRDMA_QHASH_TYPE_TCP_SYN, IRDMA_QHASH_MANAGE_TYPE_DELETE, NULL, false); } cm_core->stats_listen_destroyed++; cm_core->stats_listen_nodes_destroyed++; irdma_debug(&listener->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "loc_port=0x%04x loc_addr=%x cm_listen_node=%p cm_id=%p qhash_set=%d vlan_id=%d apbvt_del=%d\n", listener->loc_port, listener->loc_addr[0], listener, listener->cm_id, listener->qhash_set, listener->vlan_id, apbvt_del); kfree(listener); listener = NULL; return 0; } return -EINVAL; } /** * irdma_cm_del_listen - delete a listener * @cm_core: cm's core * @listener: passive connection's listener * @apbvt_del: flag to delete apbvt */ static int irdma_cm_del_listen(struct irdma_cm_core *cm_core, struct irdma_cm_listener *listener, bool apbvt_del) { listener->listener_state = IRDMA_CM_LISTENER_PASSIVE_STATE; listener->cm_id = NULL; return irdma_dec_refcnt_listen(cm_core, listener, 1, apbvt_del); } /** * irdma_find_node - find a cm node that matches the reference cm node * @cm_core: cm's core * @rem_port: remote tcp port num * @rem_addr: remote ip addr * @loc_port: local tcp port num * @loc_addr: local ip addr * @vlan_id: local VLAN ID */ struct irdma_cm_node * irdma_find_node(struct irdma_cm_core *cm_core, u16 rem_port, u32 *rem_addr, u16 loc_port, u32 *loc_addr, u16 vlan_id) { struct irdma_cm_node *cm_node; u32 key = (rem_port << 16) | loc_port; rcu_read_lock(); HASH_FOR_EACH_POSSIBLE_RCU(cm_core->cm_hash_tbl, cm_node, list, key) { if (cm_node->vlan_id == vlan_id && cm_node->loc_port == loc_port && cm_node->rem_port == rem_port && !memcmp(cm_node->loc_addr, loc_addr, sizeof(cm_node->loc_addr)) && !memcmp(cm_node->rem_addr, rem_addr, sizeof(cm_node->rem_addr))) { if (!atomic_inc_not_zero(&cm_node->refcnt)) goto exit; rcu_read_unlock(); return cm_node; } } exit: rcu_read_unlock(); /* no owner node */ return NULL; } /** * irdma_add_hte_node - add a cm node to the hash table * @cm_core: cm's core * @cm_node: connection's node */ static void irdma_add_hte_node(struct irdma_cm_core *cm_core, struct irdma_cm_node *cm_node) { unsigned long flags; u32 key = (cm_node->rem_port << 16) | cm_node->loc_port; spin_lock_irqsave(&cm_core->ht_lock, flags); HASH_ADD_RCU(cm_core->cm_hash_tbl, &cm_node->list, key); spin_unlock_irqrestore(&cm_core->ht_lock, flags); } /** * irdma_ipv4_is_lpb - check if loopback * @loc_addr: local addr to compare * @rem_addr: remote address */ bool irdma_ipv4_is_lpb(u32 loc_addr, u32 rem_addr) { return ipv4_is_loopback(htonl(rem_addr)) || (loc_addr == rem_addr); } /** * irdma_ipv6_is_lpb - check if loopback * @loc_addr: local addr to compare * @rem_addr: remote address */ bool irdma_ipv6_is_lpb(u32 *loc_addr, u32 *rem_addr) { struct in6_addr raddr6; irdma_copy_ip_htonl(raddr6.__u6_addr.__u6_addr32, rem_addr); return !memcmp(loc_addr, rem_addr, 16) || ipv6_addr_loopback(&raddr6); } /** * irdma_cm_create_ah - create a cm address handle * @cm_node: The connection manager node to create AH for * @wait: Provides option to wait for ah creation or not */ static int irdma_cm_create_ah(struct irdma_cm_node *cm_node, bool wait) { struct irdma_ah_info ah_info = {0}; struct irdma_device *iwdev = cm_node->iwdev; #ifdef VIMAGE struct vnet *vnet = irdma_cmid_to_vnet(cm_node->cm_id); #endif ether_addr_copy(ah_info.mac_addr, if_getlladdr(iwdev->netdev)); ah_info.hop_ttl = 0x40; ah_info.tc_tos = cm_node->tos; ah_info.vsi = &iwdev->vsi; if (cm_node->ipv4) { ah_info.ipv4_valid = true; ah_info.dest_ip_addr[0] = cm_node->rem_addr[0]; ah_info.src_ip_addr[0] = cm_node->loc_addr[0]; CURVNET_SET_QUIET(vnet); ah_info.do_lpbk = irdma_ipv4_is_lpb(ah_info.src_ip_addr[0], ah_info.dest_ip_addr[0]); CURVNET_RESTORE(); } else { memcpy(ah_info.dest_ip_addr, cm_node->rem_addr, sizeof(ah_info.dest_ip_addr)); memcpy(ah_info.src_ip_addr, cm_node->loc_addr, sizeof(ah_info.src_ip_addr)); ah_info.do_lpbk = irdma_ipv6_is_lpb(ah_info.src_ip_addr, ah_info.dest_ip_addr); } ah_info.vlan_tag = cm_node->vlan_id; if (cm_node->vlan_id < VLAN_N_VID) { ah_info.insert_vlan_tag = 1; ah_info.vlan_tag |= cm_node->user_pri << VLAN_PRIO_SHIFT; } ah_info.dst_arpindex = irdma_arp_table(iwdev->rf, ah_info.dest_ip_addr, NULL, IRDMA_ARP_RESOLVE); if (irdma_puda_create_ah(&iwdev->rf->sc_dev, &ah_info, wait, IRDMA_PUDA_RSRC_TYPE_ILQ, cm_node, &cm_node->ah)) return -ENOMEM; return 0; } /** * irdma_cm_free_ah - free a cm address handle * @cm_node: The connection manager node to create AH for */ static void irdma_cm_free_ah(struct irdma_cm_node *cm_node) { struct irdma_device *iwdev = cm_node->iwdev; irdma_puda_free_ah(&iwdev->rf->sc_dev, cm_node->ah); cm_node->ah = NULL; } /** * irdma_make_cm_node - create a new instance of a cm node * @cm_core: cm's core * @iwdev: iwarp device structure * @cm_info: quad info for connection * @listener: passive connection's listener */ static struct irdma_cm_node * irdma_make_cm_node(struct irdma_cm_core *cm_core, struct irdma_device *iwdev, struct irdma_cm_info *cm_info, struct irdma_cm_listener *listener) { struct irdma_cm_node *cm_node; int arpindex; if_t netdev = iwdev->netdev; /* create an hte and cm_node for this instance */ cm_node = kzalloc(sizeof(*cm_node), GFP_ATOMIC); if (!cm_node) return NULL; /* set our node specific transport info */ cm_node->ipv4 = cm_info->ipv4; cm_node->vlan_id = cm_info->vlan_id; if (cm_node->vlan_id >= VLAN_N_VID && iwdev->dcb_vlan_mode) cm_node->vlan_id = 0; cm_node->tos = cm_info->tos; cm_node->user_pri = cm_info->user_pri; if (listener) { if (listener->tos != cm_info->tos) irdma_dev_warn(&iwdev->ibdev, "application TOS[%d] and remote client TOS[%d] mismatch\n", listener->tos, cm_info->tos); if (iwdev->vsi.dscp_mode) { cm_node->user_pri = listener->user_pri; } else { cm_node->tos = max(listener->tos, cm_info->tos); cm_node->user_pri = rt_tos2priority(cm_node->tos); cm_node->user_pri = irdma_iw_get_vlan_prio(cm_info->loc_addr, cm_node->user_pri, cm_info->ipv4); } irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_DCB, "listener: TOS:[%d] UP:[%d]\n", cm_node->tos, cm_node->user_pri); } memcpy(cm_node->loc_addr, cm_info->loc_addr, sizeof(cm_node->loc_addr)); memcpy(cm_node->rem_addr, cm_info->rem_addr, sizeof(cm_node->rem_addr)); cm_node->loc_port = cm_info->loc_port; cm_node->rem_port = cm_info->rem_port; cm_node->mpa_frame_rev = IRDMA_CM_DEFAULT_MPA_VER; cm_node->send_rdma0_op = SEND_RDMA_READ_ZERO; cm_node->iwdev = iwdev; cm_node->dev = &iwdev->rf->sc_dev; cm_node->ird_size = cm_node->dev->hw_attrs.max_hw_ird; cm_node->ord_size = cm_node->dev->hw_attrs.max_hw_ord; cm_node->listener = listener; cm_node->cm_id = cm_info->cm_id; ether_addr_copy(cm_node->loc_mac, if_getlladdr(netdev)); spin_lock_init(&cm_node->retrans_list_lock); cm_node->ack_rcvd = false; init_completion(&cm_node->establish_comp); atomic_set(&cm_node->refcnt, 1); /* associate our parent CM core */ cm_node->cm_core = cm_core; cm_node->tcp_cntxt.loc_id = IRDMA_CM_DEFAULT_LOCAL_ID; cm_node->tcp_cntxt.rcv_wscale = iwdev->rcv_wscale; cm_node->tcp_cntxt.rcv_wnd = iwdev->rcv_wnd >> cm_node->tcp_cntxt.rcv_wscale; kc_set_loc_seq_num_mss(cm_node); arpindex = irdma_resolve_neigh_lpb_chk(iwdev, cm_node, cm_info); if (arpindex < 0) goto err; ether_addr_copy(cm_node->rem_mac, iwdev->rf->arp_table[arpindex].mac_addr); irdma_add_hte_node(cm_core, cm_node); cm_core->stats_nodes_created++; return cm_node; err: kfree(cm_node); return NULL; } static void irdma_destroy_connection(struct irdma_cm_node *cm_node) { struct irdma_cm_core *cm_core = cm_node->cm_core; struct irdma_qp *iwqp; struct irdma_cm_info nfo; /* if the node is destroyed before connection was accelerated */ if (!cm_node->accelerated && cm_node->accept_pend) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "node destroyed before established\n"); atomic_dec(&cm_node->listener->pend_accepts_cnt); } if (cm_node->close_entry) irdma_handle_close_entry(cm_node, 0); if (cm_node->listener) { irdma_dec_refcnt_listen(cm_core, cm_node->listener, 0, true); } else { if (cm_node->apbvt_set) { irdma_del_apbvt(cm_node->iwdev, cm_node->apbvt_entry); cm_node->apbvt_set = 0; } irdma_get_addr_info(cm_node, &nfo); if (cm_node->qhash_set) { nfo.qh_qpid = cm_node->iwdev->vsi.ilq->qp_id; irdma_manage_qhash(cm_node->iwdev, &nfo, IRDMA_QHASH_TYPE_TCP_ESTABLISHED, IRDMA_QHASH_MANAGE_TYPE_DELETE, NULL, false); cm_node->qhash_set = 0; } } iwqp = cm_node->iwqp; if (iwqp) { cm_node->cm_id->rem_ref(cm_node->cm_id); cm_node->cm_id = NULL; iwqp->cm_id = NULL; irdma_qp_rem_ref(&iwqp->ibqp); cm_node->iwqp = NULL; } else if (cm_node->qhash_set) { irdma_get_addr_info(cm_node, &nfo); nfo.qh_qpid = cm_node->iwdev->vsi.ilq->qp_id; irdma_manage_qhash(cm_node->iwdev, &nfo, IRDMA_QHASH_TYPE_TCP_ESTABLISHED, IRDMA_QHASH_MANAGE_TYPE_DELETE, NULL, false); cm_node->qhash_set = 0; } cm_core->cm_free_ah(cm_node); } /** * irdma_rem_ref_cm_node - destroy an instance of a cm node * @cm_node: connection's node */ void irdma_rem_ref_cm_node(struct irdma_cm_node *cm_node) { struct irdma_cm_core *cm_core = cm_node->cm_core; unsigned long flags; spin_lock_irqsave(&cm_core->ht_lock, flags); if (!atomic_dec_and_test(&cm_node->refcnt)) { spin_unlock_irqrestore(&cm_core->ht_lock, flags); return; } if (cm_node->iwqp) { cm_node->iwqp->cm_node = NULL; cm_node->iwqp->cm_id = NULL; } HASH_DEL_RCU(cm_core->cm_hash_tbl, &cm_node->list); cm_node->cm_core->stats_nodes_destroyed++; spin_unlock_irqrestore(&cm_core->ht_lock, flags); irdma_destroy_connection(cm_node); kfree_rcu(cm_node, rcu_head); } /** * irdma_handle_fin_pkt - FIN packet received * @cm_node: connection's node */ static void irdma_handle_fin_pkt(struct irdma_cm_node *cm_node) { switch (cm_node->state) { case IRDMA_CM_STATE_SYN_RCVD: case IRDMA_CM_STATE_SYN_SENT: case IRDMA_CM_STATE_ESTABLISHED: case IRDMA_CM_STATE_MPAREJ_RCVD: cm_node->tcp_cntxt.rcv_nxt++; irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_LAST_ACK; irdma_send_fin(cm_node); break; case IRDMA_CM_STATE_MPAREQ_SENT: irdma_create_event(cm_node, IRDMA_CM_EVENT_ABORTED); cm_node->tcp_cntxt.rcv_nxt++; irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSED; atomic_inc(&cm_node->refcnt); irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_FIN_WAIT1: cm_node->tcp_cntxt.rcv_nxt++; irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSING; irdma_send_ack(cm_node); /* * Wait for ACK as this is simultaneous close. After we receive ACK, do not send anything. Just rm the * node. */ break; case IRDMA_CM_STATE_FIN_WAIT2: cm_node->tcp_cntxt.rcv_nxt++; irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_TIME_WAIT; irdma_send_ack(cm_node); irdma_schedule_cm_timer(cm_node, NULL, IRDMA_TIMER_TYPE_CLOSE, 1, 0); break; case IRDMA_CM_STATE_TIME_WAIT: cm_node->tcp_cntxt.rcv_nxt++; irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_rem_ref_cm_node(cm_node); break; case IRDMA_CM_STATE_OFFLOADED: default: irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "bad state node state = %d\n", cm_node->state); break; } } /** * irdma_handle_rst_pkt - process received RST packet * @cm_node: connection's node * @rbuf: receive buffer */ static void irdma_handle_rst_pkt(struct irdma_cm_node *cm_node, struct irdma_puda_buf *rbuf) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "caller: %pS cm_node=%p state=%d rem_port=0x%04x loc_port=0x%04x rem_addr=%x loc_addr=%x\n", __builtin_return_address(0), cm_node, cm_node->state, cm_node->rem_port, cm_node->loc_port, cm_node->rem_addr[0], cm_node->loc_addr[0]); irdma_cleanup_retrans_entry(cm_node); switch (cm_node->state) { case IRDMA_CM_STATE_SYN_SENT: case IRDMA_CM_STATE_MPAREQ_SENT: switch (cm_node->mpa_frame_rev) { case IETF_MPA_V2: /* Drop down to MPA_V1 */ cm_node->mpa_frame_rev = IETF_MPA_V1; /* send a syn and goto syn sent state */ cm_node->state = IRDMA_CM_STATE_SYN_SENT; if (irdma_send_syn(cm_node, 0)) irdma_active_open_err(cm_node, false); break; case IETF_MPA_V1: default: irdma_active_open_err(cm_node, false); break; } break; case IRDMA_CM_STATE_MPAREQ_RCVD: atomic_inc(&cm_node->passive_state); break; case IRDMA_CM_STATE_ESTABLISHED: case IRDMA_CM_STATE_SYN_RCVD: case IRDMA_CM_STATE_LISTENING: irdma_passive_open_err(cm_node, false); break; case IRDMA_CM_STATE_OFFLOADED: irdma_active_open_err(cm_node, false); break; case IRDMA_CM_STATE_CLOSED: break; case IRDMA_CM_STATE_FIN_WAIT2: case IRDMA_CM_STATE_FIN_WAIT1: case IRDMA_CM_STATE_LAST_ACK: case IRDMA_CM_STATE_TIME_WAIT: cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_rem_ref_cm_node(cm_node); break; default: break; } } /** * irdma_handle_rcv_mpa - Process a recv'd mpa buffer * @cm_node: connection's node * @rbuf: receive buffer */ static void irdma_handle_rcv_mpa(struct irdma_cm_node *cm_node, struct irdma_puda_buf *rbuf) { int err; int datasize = rbuf->datalen; u8 *dataloc = rbuf->data; enum irdma_cm_event_type type = IRDMA_CM_EVENT_UNKNOWN; u32 res_type; err = irdma_parse_mpa(cm_node, dataloc, &res_type, datasize); if (err) { if (cm_node->state == IRDMA_CM_STATE_MPAREQ_SENT) irdma_active_open_err(cm_node, true); else irdma_passive_open_err(cm_node, true); return; } switch (cm_node->state) { case IRDMA_CM_STATE_ESTABLISHED: if (res_type == IRDMA_MPA_REQUEST_REJECT) irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "state for reject\n"); cm_node->state = IRDMA_CM_STATE_MPAREQ_RCVD; type = IRDMA_CM_EVENT_MPA_REQ; irdma_send_ack(cm_node); /* ACK received MPA request */ atomic_set(&cm_node->passive_state, IRDMA_PASSIVE_STATE_INDICATED); break; case IRDMA_CM_STATE_MPAREQ_SENT: irdma_cleanup_retrans_entry(cm_node); if (res_type == IRDMA_MPA_REQUEST_REJECT) { type = IRDMA_CM_EVENT_MPA_REJECT; cm_node->state = IRDMA_CM_STATE_MPAREJ_RCVD; } else { type = IRDMA_CM_EVENT_CONNECTED; cm_node->state = IRDMA_CM_STATE_OFFLOADED; } irdma_send_ack(cm_node); break; default: irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "wrong cm_node state=%d\n", cm_node->state); break; } irdma_create_event(cm_node, type); } /** * irdma_check_syn - Check for error on received syn ack * @cm_node: connection's node * @tcph: pointer tcp header */ static int irdma_check_syn(struct irdma_cm_node *cm_node, struct tcphdr *tcph) { if (ntohl(tcph->th_ack) != cm_node->tcp_cntxt.loc_seq_num) { irdma_active_open_err(cm_node, true); return 1; } return 0; } /** * irdma_check_seq - check seq numbers if OK * @cm_node: connection's node * @tcph: pointer tcp header */ static int irdma_check_seq(struct irdma_cm_node *cm_node, struct tcphdr *tcph) { u32 seq; u32 ack_seq; u32 loc_seq_num = cm_node->tcp_cntxt.loc_seq_num; u32 rcv_nxt = cm_node->tcp_cntxt.rcv_nxt; u32 rcv_wnd; int err = 0; seq = ntohl(tcph->th_seq); ack_seq = ntohl(tcph->th_ack); rcv_wnd = cm_node->tcp_cntxt.rcv_wnd; if (ack_seq != loc_seq_num || !between(seq, rcv_nxt, (rcv_nxt + rcv_wnd))) err = -1; if (err) irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "seq number err\n"); return err; } void irdma_add_conn_est_qh(struct irdma_cm_node *cm_node) { struct irdma_cm_info nfo; irdma_get_addr_info(cm_node, &nfo); nfo.qh_qpid = cm_node->iwdev->vsi.ilq->qp_id; irdma_manage_qhash(cm_node->iwdev, &nfo, IRDMA_QHASH_TYPE_TCP_ESTABLISHED, IRDMA_QHASH_MANAGE_TYPE_ADD, cm_node, false); cm_node->qhash_set = true; } /** * irdma_handle_syn_pkt - is for Passive node * @cm_node: connection's node * @rbuf: receive buffer */ static void irdma_handle_syn_pkt(struct irdma_cm_node *cm_node, struct irdma_puda_buf *rbuf) { struct tcphdr *tcph = (struct tcphdr *)rbuf->tcph; int err; u32 inc_sequence; int optionsize; optionsize = (tcph->th_off << 2) - sizeof(*tcph); inc_sequence = ntohl(tcph->th_seq); switch (cm_node->state) { case IRDMA_CM_STATE_SYN_SENT: case IRDMA_CM_STATE_MPAREQ_SENT: /* Rcvd syn on active open connection */ irdma_active_open_err(cm_node, 1); break; case IRDMA_CM_STATE_LISTENING: /* Passive OPEN */ if (atomic_read(&cm_node->listener->pend_accepts_cnt) > cm_node->listener->backlog) { cm_node->cm_core->stats_backlog_drops++; irdma_passive_open_err(cm_node, false); break; } err = irdma_handle_tcp_options(cm_node, tcph, optionsize, 1); if (err) { irdma_passive_open_err(cm_node, false); /* drop pkt */ break; } err = cm_node->cm_core->cm_create_ah(cm_node, false); if (err) { irdma_passive_open_err(cm_node, false); /* drop pkt */ break; } cm_node->tcp_cntxt.rcv_nxt = inc_sequence + 1; cm_node->accept_pend = 1; atomic_inc(&cm_node->listener->pend_accepts_cnt); cm_node->state = IRDMA_CM_STATE_SYN_RCVD; break; case IRDMA_CM_STATE_CLOSED: irdma_cleanup_retrans_entry(cm_node); atomic_inc(&cm_node->refcnt); irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_OFFLOADED: case IRDMA_CM_STATE_ESTABLISHED: case IRDMA_CM_STATE_FIN_WAIT1: case IRDMA_CM_STATE_FIN_WAIT2: case IRDMA_CM_STATE_MPAREQ_RCVD: case IRDMA_CM_STATE_LAST_ACK: case IRDMA_CM_STATE_CLOSING: case IRDMA_CM_STATE_UNKNOWN: default: break; } } /** * irdma_handle_synack_pkt - Process SYN+ACK packet (active side) * @cm_node: connection's node * @rbuf: receive buffer */ static void irdma_handle_synack_pkt(struct irdma_cm_node *cm_node, struct irdma_puda_buf *rbuf) { struct tcphdr *tcph = (struct tcphdr *)rbuf->tcph; int err; u32 inc_sequence; int optionsize; optionsize = (tcph->th_off << 2) - sizeof(*tcph); inc_sequence = ntohl(tcph->th_seq); switch (cm_node->state) { case IRDMA_CM_STATE_SYN_SENT: irdma_cleanup_retrans_entry(cm_node); /* active open */ if (irdma_check_syn(cm_node, tcph)) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "check syn fail\n"); return; } cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->th_ack); /* setup options */ err = irdma_handle_tcp_options(cm_node, tcph, optionsize, 0); if (err) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_node=%p tcp_options failed\n", cm_node); break; } irdma_cleanup_retrans_entry(cm_node); cm_node->tcp_cntxt.rcv_nxt = inc_sequence + 1; irdma_send_ack(cm_node); /* ACK for the syn_ack */ err = irdma_send_mpa_request(cm_node); if (err) { irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_node=%p irdma_send_mpa_request failed\n", cm_node); break; } cm_node->state = IRDMA_CM_STATE_MPAREQ_SENT; break; case IRDMA_CM_STATE_MPAREQ_RCVD: irdma_passive_open_err(cm_node, true); break; case IRDMA_CM_STATE_LISTENING: cm_node->tcp_cntxt.loc_seq_num = ntohl(tcph->th_ack); irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_CLOSED: cm_node->tcp_cntxt.loc_seq_num = ntohl(tcph->th_ack); irdma_cleanup_retrans_entry(cm_node); atomic_inc(&cm_node->refcnt); irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_ESTABLISHED: case IRDMA_CM_STATE_FIN_WAIT1: case IRDMA_CM_STATE_FIN_WAIT2: case IRDMA_CM_STATE_LAST_ACK: case IRDMA_CM_STATE_OFFLOADED: case IRDMA_CM_STATE_CLOSING: case IRDMA_CM_STATE_UNKNOWN: case IRDMA_CM_STATE_MPAREQ_SENT: default: break; } } /** * irdma_handle_ack_pkt - process packet with ACK * @cm_node: connection's node * @rbuf: receive buffer */ static int irdma_handle_ack_pkt(struct irdma_cm_node *cm_node, struct irdma_puda_buf *rbuf) { struct tcphdr *tcph = (struct tcphdr *)rbuf->tcph; u32 inc_sequence; int ret; int optionsize; u32 datasize = rbuf->datalen; optionsize = (tcph->th_off << 2) - sizeof(*tcph); if (irdma_check_seq(cm_node, tcph)) return -EINVAL; inc_sequence = ntohl(tcph->th_seq); switch (cm_node->state) { case IRDMA_CM_STATE_SYN_RCVD: irdma_cleanup_retrans_entry(cm_node); ret = irdma_handle_tcp_options(cm_node, tcph, optionsize, 1); if (ret) return ret; cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->th_ack); cm_node->state = IRDMA_CM_STATE_ESTABLISHED; if (datasize) { cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize; irdma_handle_rcv_mpa(cm_node, rbuf); } break; case IRDMA_CM_STATE_ESTABLISHED: irdma_cleanup_retrans_entry(cm_node); if (datasize) { cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize; irdma_handle_rcv_mpa(cm_node, rbuf); } break; case IRDMA_CM_STATE_MPAREQ_SENT: cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->th_ack); if (datasize) { cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize; cm_node->ack_rcvd = false; irdma_handle_rcv_mpa(cm_node, rbuf); } else { cm_node->ack_rcvd = true; } break; case IRDMA_CM_STATE_LISTENING: irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_CLOSED: irdma_cleanup_retrans_entry(cm_node); atomic_inc(&cm_node->refcnt); irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_LAST_ACK: case IRDMA_CM_STATE_CLOSING: irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_rem_ref_cm_node(cm_node); break; case IRDMA_CM_STATE_FIN_WAIT1: irdma_cleanup_retrans_entry(cm_node); cm_node->state = IRDMA_CM_STATE_FIN_WAIT2; break; case IRDMA_CM_STATE_SYN_SENT: case IRDMA_CM_STATE_FIN_WAIT2: case IRDMA_CM_STATE_OFFLOADED: case IRDMA_CM_STATE_MPAREQ_RCVD: case IRDMA_CM_STATE_UNKNOWN: default: irdma_cleanup_retrans_entry(cm_node); break; } return 0; } /** * irdma_process_pkt - process cm packet * @cm_node: connection's node * @rbuf: receive buffer */ static void irdma_process_pkt(struct irdma_cm_node *cm_node, struct irdma_puda_buf *rbuf) { enum irdma_tcpip_pkt_type pkt_type = IRDMA_PKT_TYPE_UNKNOWN; struct tcphdr *tcph = (struct tcphdr *)rbuf->tcph; u32 fin_set = 0; int err; if (tcp_get_flags(tcph) & TH_RST) { pkt_type = IRDMA_PKT_TYPE_RST; } else if (tcp_get_flags(tcph) & TH_SYN) { pkt_type = IRDMA_PKT_TYPE_SYN; if (tcp_get_flags(tcph) & TH_ACK) pkt_type = IRDMA_PKT_TYPE_SYNACK; } else if (tcp_get_flags(tcph) & TH_ACK) { pkt_type = IRDMA_PKT_TYPE_ACK; } if (tcp_get_flags(tcph) & TH_FIN) fin_set = 1; switch (pkt_type) { case IRDMA_PKT_TYPE_SYN: irdma_handle_syn_pkt(cm_node, rbuf); break; case IRDMA_PKT_TYPE_SYNACK: irdma_handle_synack_pkt(cm_node, rbuf); break; case IRDMA_PKT_TYPE_ACK: err = irdma_handle_ack_pkt(cm_node, rbuf); if (fin_set && !err) irdma_handle_fin_pkt(cm_node); break; case IRDMA_PKT_TYPE_RST: irdma_handle_rst_pkt(cm_node, rbuf); break; default: if (fin_set && (!irdma_check_seq(cm_node, (struct tcphdr *)rbuf->tcph))) irdma_handle_fin_pkt(cm_node); break; } } /** * irdma_make_listen_node - create a listen node with params * @cm_core: cm's core * @iwdev: iwarp device structure * @cm_info: quad info for connection */ static struct irdma_cm_listener * irdma_make_listen_node(struct irdma_cm_core *cm_core, struct irdma_device *iwdev, struct irdma_cm_info *cm_info) { struct irdma_cm_listener *listener; unsigned long flags; /* cannot have multiple matching listeners */ listener = irdma_find_listener(cm_core, cm_info->loc_addr, cm_info->ipv4, cm_info->loc_port, cm_info->vlan_id, IRDMA_CM_LISTENER_EITHER_STATE); if (listener && listener->listener_state == IRDMA_CM_LISTENER_ACTIVE_STATE) { atomic_dec(&listener->refcnt); return NULL; } if (!listener) { /* * create a CM listen node 1/2 node to compare incoming traffic to */ listener = kzalloc(sizeof(*listener), GFP_KERNEL); if (!listener) return NULL; cm_core->stats_listen_nodes_created++; memcpy(listener->loc_addr, cm_info->loc_addr, sizeof(listener->loc_addr)); listener->loc_port = cm_info->loc_port; INIT_LIST_HEAD(&listener->child_listen_list); atomic_set(&listener->refcnt, 1); } else { listener->reused_node = 1; } listener->cm_id = cm_info->cm_id; listener->ipv4 = cm_info->ipv4; listener->vlan_id = cm_info->vlan_id; atomic_set(&listener->pend_accepts_cnt, 0); listener->cm_core = cm_core; listener->iwdev = iwdev; listener->backlog = cm_info->backlog; listener->listener_state = IRDMA_CM_LISTENER_ACTIVE_STATE; if (!listener->reused_node) { spin_lock_irqsave(&cm_core->listen_list_lock, flags); list_add(&listener->list, &cm_core->listen_list); spin_unlock_irqrestore(&cm_core->listen_list_lock, flags); } return listener; } /** * irdma_create_cm_node - make a connection node with params * @cm_core: cm's core * @iwdev: iwarp device structure * @conn_param: connection parameters * @cm_info: quad info for connection * @caller_cm_node: pointer to cm_node structure to return */ static int irdma_create_cm_node(struct irdma_cm_core *cm_core, struct irdma_device *iwdev, struct iw_cm_conn_param *conn_param, struct irdma_cm_info *cm_info, struct irdma_cm_node **caller_cm_node) { struct irdma_cm_node *cm_node; u16 private_data_len = conn_param->private_data_len; const void *private_data = conn_param->private_data; /* create a CM connection node */ cm_node = irdma_make_cm_node(cm_core, iwdev, cm_info, NULL); if (!cm_node) return -ENOMEM; /* set our node side to client (active) side */ cm_node->tcp_cntxt.client = 1; cm_node->tcp_cntxt.rcv_wscale = IRDMA_CM_DEFAULT_RCV_WND_SCALE; irdma_record_ird_ord(cm_node, conn_param->ird, conn_param->ord); cm_node->pdata.size = private_data_len; cm_node->pdata.addr = cm_node->pdata_buf; memcpy(cm_node->pdata_buf, private_data, private_data_len); *caller_cm_node = cm_node; return 0; } /** * irdma_cm_reject - reject and teardown a connection * @cm_node: connection's node * @pdata: ptr to private data for reject * @plen: size of private data */ static int irdma_cm_reject(struct irdma_cm_node *cm_node, const void *pdata, u8 plen) { int ret; int passive_state; if (cm_node->tcp_cntxt.client) return 0; irdma_cleanup_retrans_entry(cm_node); passive_state = atomic_add_return(1, &cm_node->passive_state); if (passive_state == IRDMA_SEND_RESET_EVENT) { cm_node->state = IRDMA_CM_STATE_CLOSED; irdma_rem_ref_cm_node(cm_node); return 0; } if (cm_node->state == IRDMA_CM_STATE_LISTENER_DESTROYED) { irdma_rem_ref_cm_node(cm_node); return 0; } ret = irdma_send_mpa_reject(cm_node, pdata, plen); if (!ret) return 0; cm_node->state = IRDMA_CM_STATE_CLOSED; if (irdma_send_reset(cm_node)) irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "send reset failed\n"); return ret; } /** * irdma_cm_close - close of cm connection * @cm_node: connection's node */ static int irdma_cm_close(struct irdma_cm_node *cm_node) { switch (cm_node->state) { case IRDMA_CM_STATE_SYN_RCVD: case IRDMA_CM_STATE_SYN_SENT: case IRDMA_CM_STATE_ONE_SIDE_ESTABLISHED: case IRDMA_CM_STATE_ESTABLISHED: case IRDMA_CM_STATE_ACCEPTING: case IRDMA_CM_STATE_MPAREQ_SENT: case IRDMA_CM_STATE_MPAREQ_RCVD: irdma_cleanup_retrans_entry(cm_node); irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_CLOSE_WAIT: cm_node->state = IRDMA_CM_STATE_LAST_ACK; irdma_send_fin(cm_node); break; case IRDMA_CM_STATE_FIN_WAIT1: case IRDMA_CM_STATE_FIN_WAIT2: case IRDMA_CM_STATE_LAST_ACK: case IRDMA_CM_STATE_TIME_WAIT: case IRDMA_CM_STATE_CLOSING: return -EINVAL; case IRDMA_CM_STATE_LISTENING: irdma_cleanup_retrans_entry(cm_node); irdma_send_reset(cm_node); break; case IRDMA_CM_STATE_MPAREJ_RCVD: case IRDMA_CM_STATE_UNKNOWN: case IRDMA_CM_STATE_INITED: case IRDMA_CM_STATE_CLOSED: case IRDMA_CM_STATE_LISTENER_DESTROYED: irdma_rem_ref_cm_node(cm_node); break; case IRDMA_CM_STATE_OFFLOADED: if (cm_node->send_entry) irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "CM send_entry in OFFLOADED state\n"); irdma_rem_ref_cm_node(cm_node); break; } return 0; } /** * irdma_receive_ilq - recv an ETHERNET packet, and process it * through CM * @vsi: VSI structure of dev * @rbuf: receive buffer */ void irdma_receive_ilq(struct irdma_sc_vsi *vsi, struct irdma_puda_buf *rbuf) { struct irdma_cm_node *cm_node; struct irdma_cm_listener *listener; struct ip *iph; struct ip6_hdr *ip6h; struct tcphdr *tcph; struct irdma_cm_info cm_info = {0}; struct irdma_device *iwdev = vsi->back_vsi; struct irdma_cm_core *cm_core = &iwdev->cm_core; struct ether_vlan_header *ethh; u16 vtag; /* if vlan, then maclen = 18 else 14 */ iph = (struct ip *)rbuf->iph; irdma_debug_buf(vsi->dev, IRDMA_DEBUG_ILQ, "RECEIVE ILQ BUFFER", rbuf->mem.va, rbuf->totallen); if (iwdev->rf->sc_dev.hw_attrs.uk_attrs.hw_rev >= IRDMA_GEN_2) { if (rbuf->vlan_valid) { vtag = rbuf->vlan_id; cm_info.user_pri = (vtag & EVL_PRI_MASK) >> VLAN_PRIO_SHIFT; cm_info.vlan_id = vtag & EVL_VLID_MASK; } else { cm_info.vlan_id = 0xFFFF; } } else { ethh = rbuf->mem.va; if (ethh->evl_proto == htons(ETH_P_8021Q)) { vtag = ntohs(ethh->evl_tag); cm_info.user_pri = (vtag & EVL_PRI_MASK) >> VLAN_PRIO_SHIFT; cm_info.vlan_id = vtag & EVL_VLID_MASK; irdma_debug(&cm_core->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "vlan_id=%d\n", cm_info.vlan_id); } else { cm_info.vlan_id = 0xFFFF; } } tcph = (struct tcphdr *)rbuf->tcph; if (rbuf->ipv4) { cm_info.loc_addr[0] = ntohl(iph->ip_dst.s_addr); cm_info.rem_addr[0] = ntohl(iph->ip_src.s_addr); cm_info.ipv4 = true; cm_info.tos = iph->ip_tos; } else { ip6h = (struct ip6_hdr *)rbuf->iph; irdma_copy_ip_ntohl(cm_info.loc_addr, ip6h->ip6_dst.__u6_addr.__u6_addr32); irdma_copy_ip_ntohl(cm_info.rem_addr, ip6h->ip6_src.__u6_addr.__u6_addr32); cm_info.ipv4 = false; cm_info.tos = (ip6h->ip6_vfc << 4) | ip6h->ip6_flow; } cm_info.loc_port = ntohs(tcph->th_dport); cm_info.rem_port = ntohs(tcph->th_sport); cm_node = irdma_find_node(cm_core, cm_info.rem_port, cm_info.rem_addr, cm_info.loc_port, cm_info.loc_addr, cm_info.vlan_id); if (!cm_node) { /* * Only type of packet accepted are for the PASSIVE open (syn only) */ if (!(tcp_get_flags(tcph) & TH_SYN) || tcp_get_flags(tcph) & TH_ACK) return; listener = irdma_find_listener(cm_core, cm_info.loc_addr, cm_info.ipv4, cm_info.loc_port, cm_info.vlan_id, IRDMA_CM_LISTENER_ACTIVE_STATE); if (!listener) { cm_info.cm_id = NULL; irdma_debug(&cm_core->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "no listener found\n"); return; } cm_info.cm_id = listener->cm_id; cm_node = irdma_make_cm_node(cm_core, iwdev, &cm_info, listener); if (!cm_node) { irdma_debug(&cm_core->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "allocate node failed\n"); atomic_dec(&listener->refcnt); return; } if (!(tcp_get_flags(tcph) & (TH_RST | TH_FIN))) { cm_node->state = IRDMA_CM_STATE_LISTENING; } else { irdma_rem_ref_cm_node(cm_node); return; } atomic_inc(&cm_node->refcnt); } else if (cm_node->state == IRDMA_CM_STATE_OFFLOADED) { irdma_rem_ref_cm_node(cm_node); return; } irdma_process_pkt(cm_node, rbuf); irdma_rem_ref_cm_node(cm_node); } static int irdma_add_qh(struct irdma_cm_node *cm_node, bool active) { if (!active) irdma_add_conn_est_qh(cm_node); return 0; } static void irdma_cm_free_ah_nop(struct irdma_cm_node *cm_node) { } /** * irdma_setup_cm_core - setup top level instance of a cm core * @iwdev: iwarp device structure * @rdma_ver: HW version */ int irdma_setup_cm_core(struct irdma_device *iwdev, u8 rdma_ver) { struct irdma_cm_core *cm_core = &iwdev->cm_core; cm_core->iwdev = iwdev; cm_core->dev = &iwdev->rf->sc_dev; /* Handles CM event work items send to Iwarp core */ cm_core->event_wq = alloc_ordered_workqueue("iwarp-event-wq", 0); if (!cm_core->event_wq) return -ENOMEM; INIT_LIST_HEAD(&cm_core->listen_list); timer_setup(&cm_core->tcp_timer, irdma_cm_timer_tick, 0); spin_lock_init(&cm_core->ht_lock); spin_lock_init(&cm_core->listen_list_lock); spin_lock_init(&cm_core->apbvt_lock); switch (rdma_ver) { case IRDMA_GEN_1: cm_core->form_cm_frame = irdma_form_uda_cm_frame; cm_core->cm_create_ah = irdma_add_qh; cm_core->cm_free_ah = irdma_cm_free_ah_nop; break; case IRDMA_GEN_2: default: cm_core->form_cm_frame = irdma_form_ah_cm_frame; cm_core->cm_create_ah = irdma_cm_create_ah; cm_core->cm_free_ah = irdma_cm_free_ah; } return 0; } /** * irdma_cleanup_cm_core - deallocate a top level instance of a * cm core * @cm_core: cm's core */ void irdma_cleanup_cm_core(struct irdma_cm_core *cm_core) { if (!cm_core) return; del_timer_sync(&cm_core->tcp_timer); destroy_workqueue(cm_core->event_wq); cm_core->dev->ws_reset(&cm_core->iwdev->vsi); } /** * irdma_init_tcp_ctx - setup qp context * @cm_node: connection's node * @tcp_info: offload info for tcp * @iwqp: associate qp for the connection */ static void irdma_init_tcp_ctx(struct irdma_cm_node *cm_node, struct irdma_tcp_offload_info *tcp_info, struct irdma_qp *iwqp) { tcp_info->ipv4 = cm_node->ipv4; tcp_info->drop_ooo_seg = !iwqp->iwdev->iw_ooo; tcp_info->wscale = true; tcp_info->ignore_tcp_opt = true; tcp_info->ignore_tcp_uns_opt = true; tcp_info->no_nagle = false; tcp_info->ttl = IRDMA_DEFAULT_TTL; tcp_info->rtt_var = IRDMA_DEFAULT_RTT_VAR; tcp_info->ss_thresh = IRDMA_DEFAULT_SS_THRESH; tcp_info->rexmit_thresh = IRDMA_DEFAULT_REXMIT_THRESH; tcp_info->tcp_state = IRDMA_TCP_STATE_ESTABLISHED; tcp_info->snd_wscale = cm_node->tcp_cntxt.snd_wscale; tcp_info->rcv_wscale = cm_node->tcp_cntxt.rcv_wscale; tcp_info->snd_nxt = cm_node->tcp_cntxt.loc_seq_num; tcp_info->snd_wnd = cm_node->tcp_cntxt.snd_wnd; tcp_info->rcv_nxt = cm_node->tcp_cntxt.rcv_nxt; tcp_info->snd_max = cm_node->tcp_cntxt.loc_seq_num; tcp_info->snd_una = cm_node->tcp_cntxt.loc_seq_num; tcp_info->cwnd = 2 * cm_node->tcp_cntxt.mss; tcp_info->snd_wl1 = cm_node->tcp_cntxt.rcv_nxt; tcp_info->snd_wl2 = cm_node->tcp_cntxt.loc_seq_num; tcp_info->max_snd_window = cm_node->tcp_cntxt.max_snd_wnd; tcp_info->rcv_wnd = cm_node->tcp_cntxt.rcv_wnd << cm_node->tcp_cntxt.rcv_wscale; tcp_info->flow_label = 0; tcp_info->snd_mss = (u32)cm_node->tcp_cntxt.mss; tcp_info->tos = cm_node->tos; if (cm_node->vlan_id < VLAN_N_VID) { tcp_info->insert_vlan_tag = true; tcp_info->vlan_tag = cm_node->vlan_id; tcp_info->vlan_tag |= cm_node->user_pri << VLAN_PRIO_SHIFT; } tcp_info->src_port = cm_node->loc_port; tcp_info->dst_port = cm_node->rem_port; tcp_info->arp_idx = (u16)irdma_arp_table(iwqp->iwdev->rf, cm_node->rem_addr, NULL, IRDMA_ARP_RESOLVE); if (cm_node->ipv4) { tcp_info->dest_ip_addr[3] = cm_node->rem_addr[0]; tcp_info->local_ipaddr[3] = cm_node->loc_addr[0]; } else { memcpy(tcp_info->dest_ip_addr, cm_node->rem_addr, sizeof(tcp_info->dest_ip_addr)); memcpy(tcp_info->local_ipaddr, cm_node->loc_addr, sizeof(tcp_info->local_ipaddr)); } } /** * irdma_cm_init_tsa_conn - setup qp for RTS * @iwqp: associate qp for the connection * @cm_node: connection's node */ static void irdma_cm_init_tsa_conn(struct irdma_qp *iwqp, struct irdma_cm_node *cm_node) { struct irdma_iwarp_offload_info *iwarp_info; struct irdma_qp_host_ctx_info *ctx_info; iwarp_info = &iwqp->iwarp_info; ctx_info = &iwqp->ctx_info; ctx_info->tcp_info = &iwqp->tcp_info; ctx_info->send_cq_num = iwqp->iwscq->sc_cq.cq_uk.cq_id; ctx_info->rcv_cq_num = iwqp->iwrcq->sc_cq.cq_uk.cq_id; iwarp_info->ord_size = cm_node->ord_size; iwarp_info->ird_size = cm_node->ird_size; iwarp_info->rd_en = true; iwarp_info->rdmap_ver = 1; iwarp_info->ddp_ver = 1; iwarp_info->pd_id = iwqp->iwpd->sc_pd.pd_id; ctx_info->tcp_info_valid = true; ctx_info->iwarp_info_valid = true; ctx_info->user_pri = cm_node->user_pri; irdma_init_tcp_ctx(cm_node, &iwqp->tcp_info, iwqp); if (cm_node->snd_mark_en) { iwarp_info->snd_mark_en = true; iwarp_info->snd_mark_offset = (iwqp->tcp_info.snd_nxt & SNDMARKER_SEQNMASK) + cm_node->lsmm_size; } cm_node->state = IRDMA_CM_STATE_OFFLOADED; iwqp->tcp_info.tcp_state = IRDMA_TCP_STATE_ESTABLISHED; iwqp->tcp_info.src_mac_addr_idx = iwqp->iwdev->mac_ip_table_idx; if (cm_node->rcv_mark_en) { iwarp_info->rcv_mark_en = true; iwarp_info->align_hdrs = true; } irdma_sc_qp_setctx(&iwqp->sc_qp, iwqp->host_ctx.va, ctx_info); /* once tcp_info is set, no need to do it again */ ctx_info->tcp_info_valid = false; ctx_info->iwarp_info_valid = false; } /** * irdma_cm_disconn - when a connection is being closed * @iwqp: associated qp for the connection */ void irdma_cm_disconn(struct irdma_qp *iwqp) { struct irdma_device *iwdev = iwqp->iwdev; struct disconn_work *work; unsigned long flags; work = kzalloc(sizeof(*work), GFP_ATOMIC); if (!work) return; spin_lock_irqsave(&iwdev->rf->qptable_lock, flags); if (!iwdev->rf->qp_table[iwqp->ibqp.qp_num]) { spin_unlock_irqrestore(&iwdev->rf->qptable_lock, flags); irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "qp_id %d is already freed\n", iwqp->ibqp.qp_num); kfree(work); return; } irdma_qp_add_ref(&iwqp->ibqp); spin_unlock_irqrestore(&iwdev->rf->qptable_lock, flags); work->iwqp = iwqp; INIT_WORK(&work->work, irdma_disconnect_worker); queue_work(iwdev->cleanup_wq, &work->work); } /** * irdma_qp_disconnect - free qp and close cm * @iwqp: associate qp for the connection */ static void irdma_qp_disconnect(struct irdma_qp *iwqp) { struct irdma_device *iwdev = iwqp->iwdev; iwqp->active_conn = 0; /* close the CM node down if it is still active */ irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Call close API\n"); irdma_cm_close(iwqp->cm_node); } /** * irdma_cm_disconn_true - called by worker thread to disconnect qp * @iwqp: associate qp for the connection */ static void irdma_cm_disconn_true(struct irdma_qp *iwqp) { struct iw_cm_id *cm_id; struct irdma_device *iwdev; struct irdma_sc_qp *qp = &iwqp->sc_qp; u16 last_ae; u8 original_hw_tcp_state; u8 original_ibqp_state; int disconn_status = 0; int issue_disconn = 0; int issue_close = 0; int issue_flush = 0; unsigned long flags; int err; iwdev = iwqp->iwdev; spin_lock_irqsave(&iwqp->lock, flags); if (rdma_protocol_roce(&iwdev->ibdev, 1)) { struct ib_qp_attr attr; if (iwqp->flush_issued || iwqp->sc_qp.qp_uk.destroy_pending) { spin_unlock_irqrestore(&iwqp->lock, flags); return; } spin_unlock_irqrestore(&iwqp->lock, flags); attr.qp_state = IB_QPS_ERR; irdma_modify_qp_roce(&iwqp->ibqp, &attr, IB_QP_STATE, NULL); irdma_ib_qp_event(iwqp, qp->event_type); return; } cm_id = iwqp->cm_id; original_hw_tcp_state = iwqp->hw_tcp_state; original_ibqp_state = iwqp->ibqp_state; last_ae = iwqp->last_aeq; if (qp->term_flags) { issue_disconn = 1; issue_close = 1; iwqp->cm_id = NULL; irdma_terminate_del_timer(qp); if (!iwqp->flush_issued) { iwqp->flush_issued = 1; issue_flush = 1; } } else if ((original_hw_tcp_state == IRDMA_TCP_STATE_CLOSE_WAIT) || ((original_ibqp_state == IB_QPS_RTS) && (last_ae == IRDMA_AE_LLP_CONNECTION_RESET))) { issue_disconn = 1; if (last_ae == IRDMA_AE_LLP_CONNECTION_RESET) disconn_status = -ECONNRESET; } if (original_hw_tcp_state == IRDMA_TCP_STATE_CLOSED || original_hw_tcp_state == IRDMA_TCP_STATE_TIME_WAIT || last_ae == IRDMA_AE_RDMAP_ROE_BAD_LLP_CLOSE || last_ae == IRDMA_AE_BAD_CLOSE || last_ae == IRDMA_AE_LLP_CONNECTION_RESET || iwdev->rf->reset || !cm_id) { issue_close = 1; iwqp->cm_id = NULL; qp->term_flags = 0; if (!iwqp->flush_issued) { iwqp->flush_issued = 1; issue_flush = 1; } } spin_unlock_irqrestore(&iwqp->lock, flags); if (issue_flush && !iwqp->sc_qp.qp_uk.destroy_pending) { irdma_flush_wqes(iwqp, IRDMA_FLUSH_SQ | IRDMA_FLUSH_RQ | IRDMA_FLUSH_WAIT); if (qp->term_flags) irdma_ib_qp_event(iwqp, qp->event_type); } if (!cm_id || !cm_id->event_handler) return; spin_lock_irqsave(&iwdev->cm_core.ht_lock, flags); if (!iwqp->cm_node) { spin_unlock_irqrestore(&iwdev->cm_core.ht_lock, flags); return; } atomic_inc(&iwqp->cm_node->refcnt); spin_unlock_irqrestore(&iwdev->cm_core.ht_lock, flags); if (issue_disconn) { err = irdma_send_cm_event(iwqp->cm_node, cm_id, IW_CM_EVENT_DISCONNECT, disconn_status); if (err) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "disconnect event failed: - cm_id = %p\n", cm_id); } if (issue_close) { cm_id->provider_data = iwqp; err = irdma_send_cm_event(iwqp->cm_node, cm_id, IW_CM_EVENT_CLOSE, 0); if (err) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "close event failed: - cm_id = %p\n", cm_id); irdma_qp_disconnect(iwqp); } irdma_rem_ref_cm_node(iwqp->cm_node); } /** * irdma_disconnect_worker - worker for connection close * @work: points or disconn structure */ static void irdma_disconnect_worker(struct work_struct *work) { struct disconn_work *dwork = container_of(work, struct disconn_work, work); struct irdma_qp *iwqp = dwork->iwqp; kfree(dwork); irdma_cm_disconn_true(iwqp); irdma_qp_rem_ref(&iwqp->ibqp); } /** * irdma_free_lsmm_rsrc - free lsmm memory and deregister * @iwqp: associate qp for the connection */ void irdma_free_lsmm_rsrc(struct irdma_qp *iwqp) { struct irdma_device *iwdev; iwdev = iwqp->iwdev; if (iwqp->ietf_mem.va) { if (iwqp->lsmm_mr) iwdev->ibdev.dereg_mr(iwqp->lsmm_mr, NULL); irdma_free_dma_mem(iwdev->rf->sc_dev.hw, &iwqp->ietf_mem); iwqp->ietf_mem.va = NULL; } } /** * irdma_accept - registered call for connection to be accepted * @cm_id: cm information for passive connection * @conn_param: accpet parameters */ int irdma_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) { struct ib_qp *ibqp; struct irdma_qp *iwqp; struct irdma_device *iwdev; struct irdma_sc_dev *dev; struct irdma_cm_node *cm_node; struct ib_qp_attr attr = {0}; int passive_state; struct ib_mr *ibmr; struct irdma_pd *iwpd; u16 buf_len = 0; struct irdma_kmem_info accept; u64 tagged_offset; int wait_ret; int ret = 0; ibqp = irdma_get_qp(cm_id->device, conn_param->qpn); if (!ibqp) return -EINVAL; iwqp = to_iwqp(ibqp); iwdev = iwqp->iwdev; dev = &iwdev->rf->sc_dev; cm_node = cm_id->provider_data; if (((struct sockaddr_in *)&cm_id->local_addr)->sin_family == AF_INET) { cm_node->ipv4 = true; cm_node->vlan_id = irdma_get_vlan_ipv4(cm_id, cm_node->loc_addr); } else { cm_node->ipv4 = false; irdma_get_vlan_mac_ipv6(cm_id, cm_node->loc_addr, &cm_node->vlan_id, NULL); } irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Accept vlan_id=%d\n", cm_node->vlan_id); if (cm_node->state == IRDMA_CM_STATE_LISTENER_DESTROYED) { ret = -EINVAL; goto error; } passive_state = atomic_add_return(1, &cm_node->passive_state); if (passive_state == IRDMA_SEND_RESET_EVENT) { ret = -ECONNRESET; goto error; } buf_len = conn_param->private_data_len + IRDMA_MAX_IETF_SIZE; iwqp->ietf_mem.size = buf_len; iwqp->ietf_mem.va = irdma_allocate_dma_mem(dev->hw, &iwqp->ietf_mem, iwqp->ietf_mem.size, 1); if (!iwqp->ietf_mem.va) { ret = -ENOMEM; goto error; } cm_node->pdata.size = conn_param->private_data_len; accept.addr = iwqp->ietf_mem.va; accept.size = irdma_cm_build_mpa_frame(cm_node, &accept, MPA_KEY_REPLY); memcpy((u8 *)accept.addr + accept.size, conn_param->private_data, conn_param->private_data_len); if (cm_node->dev->ws_add(iwqp->sc_qp.vsi, cm_node->user_pri)) { ret = -ENOMEM; goto error; } iwqp->sc_qp.user_pri = cm_node->user_pri; irdma_qp_add_qos(&iwqp->sc_qp); if (cm_node->dev->hw_attrs.uk_attrs.hw_rev == IRDMA_GEN_2) iwdev->rf->check_fc(&iwdev->vsi, &iwqp->sc_qp); /* setup our first outgoing iWarp send WQE (the IETF frame response) */ iwpd = iwqp->iwpd; tagged_offset = (uintptr_t)iwqp->ietf_mem.va; ibmr = irdma_reg_phys_mr(&iwpd->ibpd, iwqp->ietf_mem.pa, buf_len, IB_ACCESS_LOCAL_WRITE, &tagged_offset); if (IS_ERR(ibmr)) { ret = -ENOMEM; goto error; } ibmr->pd = &iwpd->ibpd; ibmr->device = iwpd->ibpd.device; iwqp->lsmm_mr = ibmr; if (iwqp->page) iwqp->sc_qp.qp_uk.sq_base = kmap_local_page(iwqp->page); cm_node->lsmm_size = accept.size + conn_param->private_data_len; irdma_sc_send_lsmm(&iwqp->sc_qp, iwqp->ietf_mem.va, cm_node->lsmm_size, ibmr->lkey); if (iwqp->page) kunmap_local(iwqp->sc_qp.qp_uk.sq_base); iwqp->cm_id = cm_id; cm_node->cm_id = cm_id; cm_id->provider_data = iwqp; iwqp->active_conn = 0; iwqp->cm_node = cm_node; cm_node->iwqp = iwqp; irdma_cm_init_tsa_conn(iwqp, cm_node); irdma_qp_add_ref(&iwqp->ibqp); cm_id->add_ref(cm_id); attr.qp_state = IB_QPS_RTS; cm_node->qhash_set = false; cm_node->cm_core->cm_free_ah(cm_node); irdma_modify_qp(&iwqp->ibqp, &attr, IB_QP_STATE, NULL); if (dev->hw_attrs.uk_attrs.feature_flags & IRDMA_FEATURE_RTS_AE) { wait_ret = wait_event_interruptible_timeout(iwqp->waitq, iwqp->rts_ae_rcvd, IRDMA_MAX_TIMEOUT); if (!wait_ret) { irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Slow Connection: cm_node=%p, loc_port=%d, rem_port=%d, cm_id=%p\n", cm_node, cm_node->loc_port, cm_node->rem_port, cm_node->cm_id); ret = -ECONNRESET; goto error; } } irdma_send_cm_event(cm_node, cm_id, IW_CM_EVENT_ESTABLISHED, 0); cm_node->accelerated = true; complete(&cm_node->establish_comp); if (cm_node->accept_pend) { atomic_dec(&cm_node->listener->pend_accepts_cnt); cm_node->accept_pend = 0; } irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "rem_port=0x%04x, loc_port=0x%04x rem_addr=%x loc_addr=%x cm_node=%p cm_id=%p qp_id=%d\n\n", cm_node->rem_port, cm_node->loc_port, cm_node->rem_addr[0], cm_node->loc_addr[0], cm_node, cm_id, ibqp->qp_num); cm_node->cm_core->stats_accepts++; return 0; error: irdma_free_lsmm_rsrc(iwqp); irdma_rem_ref_cm_node(cm_node); return ret; } /** * irdma_reject - registered call for connection to be rejected * @cm_id: cm information for passive connection * @pdata: private data to be sent * @pdata_len: private data length */ int irdma_reject(struct iw_cm_id *cm_id, const void *pdata, u8 pdata_len) { struct irdma_device *iwdev; struct irdma_cm_node *cm_node; cm_node = cm_id->provider_data; cm_node->pdata.size = pdata_len; iwdev = to_iwdev(cm_id->device); if (!iwdev) return -EINVAL; cm_node->cm_core->stats_rejects++; if (pdata_len + sizeof(struct ietf_mpa_v2) > IRDMA_MAX_CM_BUF) return -EINVAL; return irdma_cm_reject(cm_node, pdata, pdata_len); } /** * irdma_connect - registered call for connection to be established * @cm_id: cm information for passive connection * @conn_param: Information about the connection */ int irdma_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) { struct ib_qp *ibqp; struct irdma_qp *iwqp; struct irdma_device *iwdev; struct irdma_cm_node *cm_node; struct irdma_cm_info cm_info; struct sockaddr_in *laddr; struct sockaddr_in *raddr; struct sockaddr_in6 *laddr6; struct sockaddr_in6 *raddr6; int ret = 0; ibqp = irdma_get_qp(cm_id->device, conn_param->qpn); if (!ibqp) return -EINVAL; iwqp = to_iwqp(ibqp); if (!iwqp) return -EINVAL; iwdev = iwqp->iwdev; if (!iwdev) return -EINVAL; laddr = (struct sockaddr_in *)&cm_id->m_local_addr; raddr = (struct sockaddr_in *)&cm_id->m_remote_addr; laddr6 = (struct sockaddr_in6 *)&cm_id->m_local_addr; raddr6 = (struct sockaddr_in6 *)&cm_id->m_remote_addr; if (!(laddr->sin_port) || !(raddr->sin_port)) return -EINVAL; iwqp->active_conn = 1; iwqp->cm_id = NULL; cm_id->provider_data = iwqp; /* set up the connection params for the node */ if (cm_id->remote_addr.ss_family == AF_INET) { if (iwdev->vsi.mtu < IRDMA_MIN_MTU_IPV4) return -EINVAL; cm_info.ipv4 = true; memset(cm_info.loc_addr, 0, sizeof(cm_info.loc_addr)); memset(cm_info.rem_addr, 0, sizeof(cm_info.rem_addr)); cm_info.loc_addr[0] = ntohl(laddr->sin_addr.s_addr); cm_info.rem_addr[0] = ntohl(raddr->sin_addr.s_addr); cm_info.loc_port = ntohs(laddr->sin_port); cm_info.rem_port = ntohs(raddr->sin_port); cm_info.vlan_id = irdma_get_vlan_ipv4(cm_id, cm_info.loc_addr); } else { if (iwdev->vsi.mtu < IRDMA_MIN_MTU_IPV6) return -EINVAL; cm_info.ipv4 = false; irdma_copy_ip_ntohl(cm_info.loc_addr, laddr6->sin6_addr.__u6_addr.__u6_addr32); irdma_copy_ip_ntohl(cm_info.rem_addr, raddr6->sin6_addr.__u6_addr.__u6_addr32); cm_info.loc_port = ntohs(laddr6->sin6_port); cm_info.rem_port = ntohs(raddr6->sin6_port); irdma_get_vlan_mac_ipv6(cm_id, cm_info.loc_addr, &cm_info.vlan_id, NULL); } cm_info.cm_id = cm_id; cm_info.qh_qpid = iwdev->vsi.ilq->qp_id; cm_info.tos = cm_id->tos; if (iwdev->vsi.dscp_mode) { cm_info.user_pri = iwqp->sc_qp.vsi->dscp_map[irdma_tos2dscp(cm_info.tos)]; } else { cm_info.user_pri = rt_tos2priority(cm_id->tos); cm_info.user_pri = irdma_iw_get_vlan_prio(cm_info.loc_addr, cm_info.user_pri, cm_info.ipv4); } if (iwqp->sc_qp.dev->ws_add(iwqp->sc_qp.vsi, cm_info.user_pri)) return -ENOMEM; iwqp->sc_qp.user_pri = cm_info.user_pri; irdma_qp_add_qos(&iwqp->sc_qp); if (iwdev->rf->sc_dev.hw_attrs.uk_attrs.hw_rev == IRDMA_GEN_2) iwdev->rf->check_fc(&iwdev->vsi, &iwqp->sc_qp); irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_DCB, "TOS:[%d] UP:[%d]\n", cm_id->tos, cm_info.user_pri); ret = irdma_create_cm_node(&iwdev->cm_core, iwdev, conn_param, &cm_info, &cm_node); if (ret) return ret; ret = cm_node->cm_core->cm_create_ah(cm_node, true); if (ret) goto err; if (irdma_manage_qhash(iwdev, &cm_info, IRDMA_QHASH_TYPE_TCP_ESTABLISHED, IRDMA_QHASH_MANAGE_TYPE_ADD, NULL, true)) { ret = -EINVAL; goto err; } cm_node->qhash_set = true; cm_node->apbvt_entry = irdma_add_apbvt(iwdev, cm_info.loc_port); if (!cm_node->apbvt_entry) { ret = -EINVAL; goto err; } cm_node->apbvt_set = true; iwqp->cm_node = cm_node; cm_node->iwqp = iwqp; iwqp->cm_id = cm_id; irdma_qp_add_ref(&iwqp->ibqp); cm_id->add_ref(cm_id); if (cm_node->state != IRDMA_CM_STATE_OFFLOADED) { cm_node->state = IRDMA_CM_STATE_SYN_SENT; ret = irdma_send_syn(cm_node, 0); if (ret) goto err; } irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "rem_port=0x%04x, loc_port=0x%04x rem_addr=%x loc_addr=%x cm_node=%p cm_id=%p qp_id = %d\n\n", cm_node->rem_port, cm_node->loc_port, cm_node->rem_addr[0], cm_node->loc_addr[0], cm_node, cm_id, ibqp->qp_num); return 0; err: if (cm_info.ipv4) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "connect() FAILED: dest addr=%x", cm_info.rem_addr[0]); else irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "connect() FAILED: dest addr=%x:%x:%x:%x", IRDMA_PRINT_IP6(cm_info.rem_addr)); irdma_rem_ref_cm_node(cm_node); iwdev->cm_core.stats_connect_errs++; return ret; } /** * irdma_create_listen - registered call creating listener * @cm_id: cm information for passive connection * @backlog: to max accept pending count */ int irdma_create_listen(struct iw_cm_id *cm_id, int backlog) { struct irdma_device *iwdev; struct irdma_cm_listener *cm_listen_node; struct irdma_cm_info cm_info = {0}; struct sockaddr_in *laddr; struct sockaddr_in6 *laddr6; bool wildcard = false; int err; iwdev = to_iwdev(cm_id->device); if (!iwdev) return -EINVAL; laddr = (struct sockaddr_in *)&cm_id->m_local_addr; laddr6 = (struct sockaddr_in6 *)&cm_id->m_local_addr; cm_info.qh_qpid = iwdev->vsi.ilq->qp_id; if (laddr->sin_family == AF_INET) { if (iwdev->vsi.mtu < IRDMA_MIN_MTU_IPV4) return -EINVAL; cm_info.ipv4 = true; cm_info.loc_addr[0] = ntohl(laddr->sin_addr.s_addr); cm_info.loc_port = ntohs(laddr->sin_port); if (laddr->sin_addr.s_addr != htonl(INADDR_ANY)) { cm_info.vlan_id = irdma_get_vlan_ipv4(cm_id, cm_info.loc_addr); } else { cm_info.vlan_id = 0xFFFF; wildcard = true; } } else { if (iwdev->vsi.mtu < IRDMA_MIN_MTU_IPV6) return -EINVAL; cm_info.ipv4 = false; irdma_copy_ip_ntohl(cm_info.loc_addr, laddr6->sin6_addr.__u6_addr.__u6_addr32); cm_info.loc_port = ntohs(laddr6->sin6_port); if (!IN6_IS_ADDR_UNSPECIFIED(&laddr6->sin6_addr)) { irdma_get_vlan_mac_ipv6(cm_id, cm_info.loc_addr, &cm_info.vlan_id, NULL); } else { cm_info.vlan_id = 0xFFFF; wildcard = true; } } if (cm_info.vlan_id >= VLAN_N_VID && iwdev->dcb_vlan_mode) cm_info.vlan_id = 0; cm_info.backlog = backlog; cm_info.cm_id = cm_id; cm_listen_node = irdma_make_listen_node(&iwdev->cm_core, iwdev, &cm_info); if (!cm_listen_node) { irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_listen_node == NULL\n"); return -ENOMEM; } cm_id->provider_data = cm_listen_node; cm_listen_node->tos = cm_id->tos; if (iwdev->vsi.dscp_mode) cm_listen_node->user_pri = iwdev->vsi.dscp_map[irdma_tos2dscp(cm_id->tos)]; else cm_listen_node->user_pri = rt_tos2priority(cm_id->tos); cm_info.user_pri = cm_listen_node->user_pri; if (!cm_listen_node->reused_node) { if (wildcard) { err = irdma_add_mqh(iwdev, &cm_info, cm_listen_node); if (err) goto error; } else { if (!iwdev->vsi.dscp_mode) cm_info.user_pri = cm_listen_node->user_pri = irdma_iw_get_vlan_prio(cm_info.loc_addr, cm_info.user_pri, cm_info.ipv4); err = irdma_manage_qhash(iwdev, &cm_info, IRDMA_QHASH_TYPE_TCP_SYN, IRDMA_QHASH_MANAGE_TYPE_ADD, NULL, true); if (err) goto error; cm_listen_node->qhash_set = true; } cm_listen_node->apbvt_entry = irdma_add_apbvt(iwdev, cm_info.loc_port); if (!cm_listen_node->apbvt_entry) goto error; } cm_id->add_ref(cm_id); cm_listen_node->cm_core->stats_listen_created++; irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "loc_port=0x%04x loc_addr=%x cm_listen_node=%p cm_id=%p qhash_set=%d vlan_id=%d\n", cm_listen_node->loc_port, cm_listen_node->loc_addr[0], cm_listen_node, cm_listen_node->cm_id, cm_listen_node->qhash_set, cm_listen_node->vlan_id); return 0; error: irdma_cm_del_listen(&iwdev->cm_core, cm_listen_node, false); return -EINVAL; } /** * irdma_destroy_listen - registered call to destroy listener * @cm_id: cm information for passive connection */ int irdma_destroy_listen(struct iw_cm_id *cm_id) { struct irdma_device *iwdev; iwdev = to_iwdev(cm_id->device); if (cm_id->provider_data) irdma_cm_del_listen(&iwdev->cm_core, cm_id->provider_data, true); else irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "cm_id->provider_data was NULL\n"); cm_id->rem_ref(cm_id); return 0; } /** * irdma_iw_teardown_list_prep - add conn nodes slated for tear * down to list * @cm_core: cm's core * @teardown_list: a list to which cm_node will be selected * @ipaddr: pointer to ip address * @nfo: pointer to cm_info structure instance * @disconnect_all: flag indicating disconnect all QPs */ static void irdma_iw_teardown_list_prep(struct irdma_cm_core *cm_core, struct list_head *teardown_list, u32 *ipaddr, struct irdma_cm_info *nfo, bool disconnect_all) { struct irdma_cm_node *cm_node; int bkt; HASH_FOR_EACH_RCU(cm_core->cm_hash_tbl, bkt, cm_node, list) { if ((disconnect_all || (nfo->vlan_id == cm_node->vlan_id && !memcmp(cm_node->loc_addr, ipaddr, nfo->ipv4 ? 4 : 16))) && atomic_inc_not_zero(&cm_node->refcnt)) list_add(&cm_node->teardown_entry, teardown_list); } } static inline bool irdma_ip_vlan_match(u32 *ip1, u16 vlan_id1, bool check_vlan, u32 *ip2, u16 vlan_id2, bool ipv4) { return (!check_vlan || vlan_id1 == vlan_id2) && !memcmp(ip1, ip2, ipv4 ? 4 : 16); } /** * irdma_roce_teardown_list_prep - add conn nodes slated for * tear down to list * @iwdev: RDMA device * @teardown_list: a list to which cm_node will be selected * @ipaddr: pointer to ip address * @nfo: pointer to cm_info structure instance * @disconnect_all: flag indicating disconnect all QPs */ static void irdma_roce_teardown_list_prep(struct irdma_device *iwdev, struct list_head *teardown_list, u32 *ipaddr, struct irdma_cm_info *nfo, bool disconnect_all) { struct irdma_sc_vsi *vsi = &iwdev->vsi; struct irdma_sc_qp *sc_qp; struct list_head *list_node; struct irdma_qp *qp; unsigned long flags; int i; for (i = 0; i < IRDMA_MAX_USER_PRIORITY; i++) { mutex_lock(&vsi->qos[i].qos_mutex); list_for_each(list_node, &vsi->qos[i].qplist) { u32 qp_ip[4]; sc_qp = container_of(list_node, struct irdma_sc_qp, list); if (sc_qp->qp_uk.qp_type != IRDMA_QP_TYPE_ROCE_RC) continue; qp = sc_qp->qp_uk.back_qp; if (!disconnect_all) { if (nfo->ipv4) qp_ip[0] = qp->udp_info.local_ipaddr[3]; else memcpy(qp_ip, &qp->udp_info.local_ipaddr[0], sizeof(qp_ip)); } if (disconnect_all || irdma_ip_vlan_match(qp_ip, qp->udp_info.vlan_tag & EVL_VLID_MASK, qp->udp_info.insert_vlan_tag, ipaddr, nfo->vlan_id, nfo->ipv4)) { spin_lock_irqsave(&iwdev->rf->qptable_lock, flags); if (iwdev->rf->qp_table[sc_qp->qp_uk.qp_id]) { irdma_qp_add_ref(&qp->ibqp); list_add(&qp->teardown_entry, teardown_list); } spin_unlock_irqrestore(&iwdev->rf->qptable_lock, flags); } } mutex_unlock(&vsi->qos[i].qos_mutex); } } /** * irdma_cm_event_connected - handle connected active node * @event: the info for cm_node of connection */ static void irdma_cm_event_connected(struct irdma_cm_event *event) { struct irdma_qp *iwqp; struct irdma_device *iwdev; struct irdma_cm_node *cm_node; struct irdma_sc_dev *dev; struct ib_qp_attr attr = {0}; struct iw_cm_id *cm_id; int status; bool read0; int wait_ret = 0; cm_node = event->cm_node; cm_id = cm_node->cm_id; iwqp = cm_id->provider_data; iwdev = iwqp->iwdev; dev = &iwdev->rf->sc_dev; if (iwqp->sc_qp.qp_uk.destroy_pending) { status = -ETIMEDOUT; goto error; } irdma_cm_init_tsa_conn(iwqp, cm_node); read0 = (cm_node->send_rdma0_op == SEND_RDMA_READ_ZERO); if (iwqp->page) iwqp->sc_qp.qp_uk.sq_base = kmap_local_page(iwqp->page); irdma_sc_send_rtt(&iwqp->sc_qp, read0); if (iwqp->page) kunmap_local(iwqp->sc_qp.qp_uk.sq_base); attr.qp_state = IB_QPS_RTS; cm_node->qhash_set = false; irdma_modify_qp(&iwqp->ibqp, &attr, IB_QP_STATE, NULL); if (dev->hw_attrs.uk_attrs.feature_flags & IRDMA_FEATURE_RTS_AE) { wait_ret = wait_event_interruptible_timeout(iwqp->waitq, iwqp->rts_ae_rcvd, IRDMA_MAX_TIMEOUT); if (!wait_ret) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "Slow Connection: cm_node=%p, loc_port=%d, rem_port=%d, cm_id=%p\n", cm_node, cm_node->loc_port, cm_node->rem_port, cm_node->cm_id); } irdma_send_cm_event(cm_node, cm_id, IW_CM_EVENT_CONNECT_REPLY, 0); cm_node->accelerated = true; complete(&cm_node->establish_comp); cm_node->cm_core->cm_free_ah(cm_node); return; error: iwqp->cm_id = NULL; cm_id->provider_data = NULL; irdma_send_cm_event(event->cm_node, cm_id, IW_CM_EVENT_CONNECT_REPLY, status); irdma_rem_ref_cm_node(event->cm_node); } /** * irdma_cm_event_reset - handle reset * @event: the info for cm_node of connection */ static void irdma_cm_event_reset(struct irdma_cm_event *event) { struct irdma_cm_node *cm_node = event->cm_node; struct iw_cm_id *cm_id = cm_node->cm_id; struct irdma_qp *iwqp; if (!cm_id) return; iwqp = cm_id->provider_data; if (!iwqp) return; irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "reset event %p - cm_id = %p\n", event->cm_node, cm_id); iwqp->cm_id = NULL; irdma_send_cm_event(cm_node, cm_node->cm_id, IW_CM_EVENT_DISCONNECT, -ECONNRESET); irdma_send_cm_event(cm_node, cm_node->cm_id, IW_CM_EVENT_CLOSE, 0); } /** * irdma_cm_event_handler - send event to cm upper layer * @work: pointer of cm event info. */ static void irdma_cm_event_handler(struct work_struct *work) { struct irdma_cm_event *event = container_of(work, struct irdma_cm_event, event_work); struct irdma_cm_node *cm_node; if (!event || !event->cm_node || !event->cm_node->cm_core) return; cm_node = event->cm_node; switch (event->type) { case IRDMA_CM_EVENT_MPA_REQ: irdma_send_cm_event(cm_node, cm_node->cm_id, IW_CM_EVENT_CONNECT_REQUEST, 0); break; case IRDMA_CM_EVENT_RESET: irdma_cm_event_reset(event); break; case IRDMA_CM_EVENT_CONNECTED: if (!event->cm_node->cm_id || event->cm_node->state != IRDMA_CM_STATE_OFFLOADED) break; irdma_cm_event_connected(event); break; case IRDMA_CM_EVENT_MPA_REJECT: if (!event->cm_node->cm_id || cm_node->state == IRDMA_CM_STATE_OFFLOADED) break; irdma_send_cm_event(cm_node, cm_node->cm_id, IW_CM_EVENT_CONNECT_REPLY, -ECONNREFUSED); break; case IRDMA_CM_EVENT_ABORTED: if (!event->cm_node->cm_id || event->cm_node->state == IRDMA_CM_STATE_OFFLOADED) break; irdma_event_connect_error(event); break; default: irdma_debug(&cm_node->iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "bad event type = %d\n", event->type); break; } irdma_rem_ref_cm_node(event->cm_node); kfree(event); } /** * irdma_cm_post_event - queue event request for worker thread * @event: cm node's info for up event call */ static void irdma_cm_post_event(struct irdma_cm_event *event) { atomic_inc(&event->cm_node->refcnt); INIT_WORK(&event->event_work, irdma_cm_event_handler); queue_work(event->cm_node->cm_core->event_wq, &event->event_work); } /** * irdma_cm_teardown_connections - teardown QPs * @iwdev: device pointer * @ipaddr: Pointer to IPv4 or IPv6 address * @nfo: Connection info * @disconnect_all: flag indicating disconnect all QPs * * teardown QPs where source or destination addr matches ip addr */ static void __unused irdma_cm_teardown_connections(struct irdma_device *iwdev, u32 *ipaddr, struct irdma_cm_info *nfo, bool disconnect_all) { struct irdma_cm_core *cm_core = &iwdev->cm_core; struct list_head *list_core_temp; struct list_head *list_node; struct irdma_cm_node *cm_node; struct list_head teardown_list; struct ib_qp_attr attr; struct irdma_qp *qp; INIT_LIST_HEAD(&teardown_list); rcu_read_lock(); irdma_iw_teardown_list_prep(cm_core, &teardown_list, ipaddr, nfo, disconnect_all); rcu_read_unlock(); attr.qp_state = IB_QPS_ERR; list_for_each_safe(list_node, list_core_temp, &teardown_list) { cm_node = container_of(list_node, struct irdma_cm_node, teardown_entry); irdma_modify_qp(&cm_node->iwqp->ibqp, &attr, IB_QP_STATE, NULL); if (iwdev->rf->reset) irdma_cm_disconn(cm_node->iwqp); irdma_rem_ref_cm_node(cm_node); } if (!rdma_protocol_roce(&iwdev->ibdev, 1)) return; INIT_LIST_HEAD(&teardown_list); irdma_roce_teardown_list_prep(iwdev, &teardown_list, ipaddr, nfo, disconnect_all); list_for_each_safe(list_node, list_core_temp, &teardown_list) { qp = container_of(list_node, struct irdma_qp, teardown_entry); irdma_modify_qp_roce(&qp->ibqp, &attr, IB_QP_STATE, NULL); irdma_ib_qp_event(qp, IRDMA_QP_EVENT_CATASTROPHIC); irdma_qp_rem_ref(&qp->ibqp); } } diff --git a/sys/dev/irdma/irdma_hw.c b/sys/dev/irdma/irdma_hw.c index 05004b1ccc83..64c05b8663e0 100644 --- a/sys/dev/irdma/irdma_hw.c +++ b/sys/dev/irdma/irdma_hw.c @@ -1,2825 +1,2909 @@ /*- * SPDX-License-Identifier: GPL-2.0 or Linux-OpenIB * - * Copyright (c) 2015 - 2023 Intel Corporation + * Copyright (c) 2015 - 2025 Intel Corporation * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU * General Public License (GPL) Version 2, available from the file * COPYING in the main directory of this source tree, or the * OpenFabrics.org BSD license below: * * 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. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #include "irdma_main.h" static struct irdma_rsrc_limits rsrc_limits_table[] = { [0] = { .qplimit = SZ_128, }, [1] = { .qplimit = SZ_1K, }, [2] = { .qplimit = SZ_2K, }, [3] = { .qplimit = SZ_4K, }, [4] = { .qplimit = SZ_16K, }, [5] = { .qplimit = SZ_64K, }, [6] = { .qplimit = SZ_128K, }, [7] = { .qplimit = SZ_256K, }, }; /* types of hmc objects */ static enum irdma_hmc_rsrc_type iw_hmc_obj_types[] = { IRDMA_HMC_IW_QP, IRDMA_HMC_IW_CQ, IRDMA_HMC_IW_HTE, IRDMA_HMC_IW_ARP, IRDMA_HMC_IW_APBVT_ENTRY, IRDMA_HMC_IW_MR, IRDMA_HMC_IW_XF, IRDMA_HMC_IW_XFFL, IRDMA_HMC_IW_Q1, IRDMA_HMC_IW_Q1FL, IRDMA_HMC_IW_PBLE, IRDMA_HMC_IW_TIMER, IRDMA_HMC_IW_FSIMC, IRDMA_HMC_IW_FSIAV, IRDMA_HMC_IW_RRF, IRDMA_HMC_IW_RRFFL, IRDMA_HMC_IW_HDR, IRDMA_HMC_IW_MD, IRDMA_HMC_IW_OOISC, IRDMA_HMC_IW_OOISCFFL, }; /** * irdma_iwarp_ce_handler - handle iwarp completions * @iwcq: iwarp cq receiving event */ static void irdma_iwarp_ce_handler(struct irdma_sc_cq *iwcq) { struct irdma_cq *cq = iwcq->back_cq; if (!cq->user_mode) atomic_set(&cq->armed, 0); if (cq->ibcq.comp_handler) cq->ibcq.comp_handler(&cq->ibcq, cq->ibcq.cq_context); } /** * irdma_puda_ce_handler - handle puda completion events * @rf: RDMA PCI function * @cq: puda completion q for event */ static void irdma_puda_ce_handler(struct irdma_pci_f *rf, struct irdma_sc_cq *cq) { struct irdma_sc_dev *dev = &rf->sc_dev; u32 compl_error; int status; do { status = irdma_puda_poll_cmpl(dev, cq, &compl_error); if (status == -ENOENT) break; if (status) { irdma_debug(dev, IRDMA_DEBUG_ERR, "puda status = %d\n", status); break; } if (compl_error) { irdma_debug(dev, IRDMA_DEBUG_ERR, "puda compl_err = 0x%x\n", compl_error); break; } } while (1); irdma_sc_ccq_arm(cq); } /** * irdma_process_ceq - handle ceq for completions * @rf: RDMA PCI function * @ceq: ceq having cq for completion */ static void irdma_process_ceq(struct irdma_pci_f *rf, struct irdma_ceq *ceq) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_sc_ceq *sc_ceq; struct irdma_sc_cq *cq; unsigned long flags; sc_ceq = &ceq->sc_ceq; do { spin_lock_irqsave(&ceq->ce_lock, flags); cq = irdma_sc_process_ceq(dev, sc_ceq); if (!cq) { spin_unlock_irqrestore(&ceq->ce_lock, flags); break; } if (cq->cq_type == IRDMA_CQ_TYPE_IWARP) irdma_iwarp_ce_handler(cq); spin_unlock_irqrestore(&ceq->ce_lock, flags); if (cq->cq_type == IRDMA_CQ_TYPE_CQP) queue_work(rf->cqp_cmpl_wq, &rf->cqp_cmpl_work); else if (cq->cq_type == IRDMA_CQ_TYPE_ILQ || cq->cq_type == IRDMA_CQ_TYPE_IEQ) irdma_puda_ce_handler(rf, cq); } while (1); } static void irdma_set_flush_fields(struct irdma_sc_qp *qp, struct irdma_aeqe_info *info) { struct qp_err_code qp_err; qp->sq_flush_code = info->sq; qp->rq_flush_code = info->rq; qp_err = irdma_ae_to_qp_err_code(info->ae_id); qp->flush_code = qp_err.flush_code; qp->event_type = qp_err.event_type; } /** * irdma_complete_cqp_request - perform post-completion cleanup * @cqp: device CQP * @cqp_request: CQP request * * Mark CQP request as done, wake up waiting thread or invoke * callback function and release/free CQP request. */ static void irdma_complete_cqp_request(struct irdma_cqp *cqp, struct irdma_cqp_request *cqp_request) { WRITE_ONCE(cqp_request->request_done, true); if (cqp_request->waiting) wake_up(&cqp_request->waitq); else if (cqp_request->callback_fcn) cqp_request->callback_fcn(cqp_request); irdma_put_cqp_request(cqp, cqp_request); } /** * irdma_process_aeq - handle aeq events * @rf: RDMA PCI function */ static void irdma_process_aeq(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_aeq *aeq = &rf->aeq; struct irdma_sc_aeq *sc_aeq = &aeq->sc_aeq; struct irdma_aeqe_info aeinfo; struct irdma_aeqe_info *info = &aeinfo; int ret; struct irdma_qp *iwqp = NULL; struct irdma_cq *iwcq = NULL; struct irdma_sc_qp *qp = NULL; struct irdma_device *iwdev = rf->iwdev; struct irdma_qp_host_ctx_info *ctx_info = NULL; unsigned long flags; u32 aeqcnt = 0; if (!sc_aeq->size) return; do { memset(info, 0, sizeof(*info)); ret = irdma_sc_get_next_aeqe(sc_aeq, info); if (ret) break; aeqcnt++; irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_AEQ, "ae_id = 0x%x (%s), is_qp = %d, qp_id = %d, tcp_state = %d, iwarp_state = %d, ae_src = %d\n", info->ae_id, irdma_get_ae_desc(info->ae_id), info->qp, info->qp_cq_id, info->tcp_state, info->iwarp_state, info->ae_src); if (info->qp) { spin_lock_irqsave(&rf->qptable_lock, flags); iwqp = rf->qp_table[info->qp_cq_id]; if (!iwqp) { spin_unlock_irqrestore(&rf->qptable_lock, flags); if (info->ae_id == IRDMA_AE_QP_SUSPEND_COMPLETE) { struct irdma_device *iwdev = rf->iwdev; if (!iwdev->vsi.tc_change_pending) continue; atomic_dec(&iwdev->vsi.qp_suspend_reqs); wake_up(&iwdev->suspend_wq); continue; } irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_AEQ, "qp_id %d is already freed\n", info->qp_cq_id); continue; } irdma_qp_add_ref(&iwqp->ibqp); spin_unlock_irqrestore(&rf->qptable_lock, flags); qp = &iwqp->sc_qp; spin_lock_irqsave(&iwqp->lock, flags); iwqp->hw_tcp_state = info->tcp_state; iwqp->hw_iwarp_state = info->iwarp_state; if (info->ae_id != IRDMA_AE_QP_SUSPEND_COMPLETE) iwqp->last_aeq = info->ae_id; spin_unlock_irqrestore(&iwqp->lock, flags); ctx_info = &iwqp->ctx_info; } else { if (info->ae_id != IRDMA_AE_CQ_OPERATION_ERROR) continue; } switch (info->ae_id) { struct irdma_cm_node *cm_node; case IRDMA_AE_LLP_CONNECTION_ESTABLISHED: cm_node = iwqp->cm_node; if (cm_node->accept_pend) { atomic_dec(&cm_node->listener->pend_accepts_cnt); cm_node->accept_pend = 0; } iwqp->rts_ae_rcvd = 1; wake_up_interruptible(&iwqp->waitq); break; case IRDMA_AE_LLP_FIN_RECEIVED: if (qp->term_flags) break; if (atomic_inc_return(&iwqp->close_timer_started) == 1) { iwqp->hw_tcp_state = IRDMA_TCP_STATE_CLOSE_WAIT; if (iwqp->ibqp_state == IB_QPS_RTS) { irdma_next_iw_state(iwqp, IRDMA_QP_STATE_CLOSING, 0, 0, 0); irdma_cm_disconn(iwqp); } irdma_schedule_cm_timer(iwqp->cm_node, (struct irdma_puda_buf *)iwqp, IRDMA_TIMER_TYPE_CLOSE, 1, 0); } break; case IRDMA_AE_LLP_CLOSE_COMPLETE: if (qp->term_flags) irdma_terminate_done(qp, 0); else irdma_cm_disconn(iwqp); break; case IRDMA_AE_BAD_CLOSE: case IRDMA_AE_RESET_SENT: irdma_next_iw_state(iwqp, IRDMA_QP_STATE_ERROR, 1, 0, 0); irdma_cm_disconn(iwqp); break; case IRDMA_AE_LLP_CONNECTION_RESET: if (atomic_read(&iwqp->close_timer_started)) break; irdma_cm_disconn(iwqp); break; case IRDMA_AE_QP_SUSPEND_COMPLETE: if (iwqp->iwdev->vsi.tc_change_pending) { if (!atomic_dec_return(&iwqp->sc_qp.vsi->qp_suspend_reqs)) wake_up(&iwqp->iwdev->suspend_wq); } if (iwqp->suspend_pending) { iwqp->suspend_pending = false; wake_up(&iwqp->iwdev->suspend_wq); } break; case IRDMA_AE_TERMINATE_SENT: irdma_terminate_send_fin(qp); break; case IRDMA_AE_LLP_TERMINATE_RECEIVED: irdma_terminate_received(qp, info); break; case IRDMA_AE_LCE_CQ_CATASTROPHIC: case IRDMA_AE_CQ_OPERATION_ERROR: irdma_dev_err(&iwdev->ibdev, "Processing CQ[0x%x] op error, AE 0x%04X\n", info->qp_cq_id, info->ae_id); spin_lock_irqsave(&rf->cqtable_lock, flags); iwcq = rf->cq_table[info->qp_cq_id]; if (!iwcq) { spin_unlock_irqrestore(&rf->cqtable_lock, flags); irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_AEQ, "cq_id %d is already freed\n", info->qp_cq_id); continue; } irdma_cq_add_ref(&iwcq->ibcq); spin_unlock_irqrestore(&rf->cqtable_lock, flags); if (iwcq->ibcq.event_handler) { struct ib_event ibevent; ibevent.device = iwcq->ibcq.device; ibevent.event = IB_EVENT_CQ_ERR; ibevent.element.cq = &iwcq->ibcq; iwcq->ibcq.event_handler(&ibevent, iwcq->ibcq.cq_context); } irdma_cq_rem_ref(&iwcq->ibcq); break; case IRDMA_AE_RESET_NOT_SENT: case IRDMA_AE_LLP_DOUBT_REACHABILITY: break; case IRDMA_AE_RESOURCE_EXHAUSTION: irdma_dev_err(&iwdev->ibdev, "Resource exhaustion reason: q1 = %d xmit or rreq = %d\n", info->ae_src == IRDMA_AE_SOURCE_RSRC_EXHT_Q1, info->ae_src == IRDMA_AE_SOURCE_RSRC_EXHT_XT_RR); break; case IRDMA_AE_PRIV_OPERATION_DENIED: case IRDMA_AE_RDMAP_ROE_BAD_LLP_CLOSE: case IRDMA_AE_STAG_ZERO_INVALID: case IRDMA_AE_IB_RREQ_AND_Q1_FULL: case IRDMA_AE_DDP_UBE_INVALID_DDP_VERSION: case IRDMA_AE_DDP_UBE_INVALID_MO: case IRDMA_AE_DDP_UBE_INVALID_QN: case IRDMA_AE_DDP_NO_L_BIT: case IRDMA_AE_RDMAP_ROE_INVALID_RDMAP_VERSION: case IRDMA_AE_RDMAP_ROE_UNEXPECTED_OPCODE: case IRDMA_AE_ROE_INVALID_RDMA_READ_REQUEST: case IRDMA_AE_ROE_INVALID_RDMA_WRITE_OR_READ_RESP: case IRDMA_AE_INVALID_ARP_ENTRY: case IRDMA_AE_INVALID_TCP_OPTION_RCVD: case IRDMA_AE_STALE_ARP_ENTRY: case IRDMA_AE_LLP_RECEIVED_MPA_CRC_ERROR: case IRDMA_AE_LLP_SEGMENT_TOO_SMALL: case IRDMA_AE_LLP_SYN_RECEIVED: case IRDMA_AE_LLP_TOO_MANY_RETRIES: case IRDMA_AE_LCE_QP_CATASTROPHIC: case IRDMA_AE_LCE_FUNCTION_CATASTROPHIC: case IRDMA_AE_LLP_TOO_MANY_RNRS: case IRDMA_AE_UDA_XMIT_DGRAM_TOO_LONG: default: irdma_dev_err(&iwdev->ibdev, "AEQ: abnormal ae_id = 0x%x (%s), is_qp = %d, qp_id = %d, ae_source = %d\n", info->ae_id, irdma_get_ae_desc(info->ae_id), info->qp, info->qp_cq_id, info->ae_src); if (rdma_protocol_roce(&iwqp->iwdev->ibdev, 1)) { ctx_info->roce_info->err_rq_idx_valid = info->err_rq_idx_valid; if (info->rq) { ctx_info->roce_info->err_rq_idx = info->wqe_idx; irdma_sc_qp_setctx_roce(&iwqp->sc_qp, iwqp->host_ctx.va, ctx_info); } irdma_set_flush_fields(qp, info); irdma_cm_disconn(iwqp); break; } ctx_info->iwarp_info->err_rq_idx_valid = info->err_rq_idx_valid; if (info->rq) { ctx_info->iwarp_info->err_rq_idx = info->wqe_idx; ctx_info->tcp_info_valid = false; ctx_info->iwarp_info_valid = true; irdma_sc_qp_setctx(&iwqp->sc_qp, iwqp->host_ctx.va, ctx_info); } if (iwqp->hw_iwarp_state != IRDMA_QP_STATE_RTS && iwqp->hw_iwarp_state != IRDMA_QP_STATE_TERMINATE) { irdma_next_iw_state(iwqp, IRDMA_QP_STATE_ERROR, 1, 0, 0); irdma_cm_disconn(iwqp); } else { irdma_terminate_connection(qp, info); } break; } if (info->qp) irdma_qp_rem_ref(&iwqp->ibqp); } while (1); if (aeqcnt) irdma_sc_repost_aeq_entries(dev, aeqcnt); } /** * irdma_ena_intr - set up device interrupts * @dev: hardware control device structure * @msix_id: id of the interrupt to be enabled */ static void irdma_ena_intr(struct irdma_sc_dev *dev, u32 msix_id) { dev->irq_ops->irdma_en_irq(dev, msix_id); } /** * irdma_dpc - tasklet for aeq and ceq 0 * @t: tasklet_struct ptr */ static void irdma_dpc(unsigned long t) { struct irdma_pci_f *rf = from_tasklet(rf, (struct tasklet_struct *)t, dpc_tasklet); if (rf->msix_shared) irdma_process_ceq(rf, rf->ceqlist); irdma_process_aeq(rf); irdma_ena_intr(&rf->sc_dev, rf->iw_msixtbl[0].idx); } /** * irdma_ceq_dpc - dpc handler for CEQ * @t: tasklet_struct ptr */ static void irdma_ceq_dpc(unsigned long t) { struct irdma_ceq *iwceq = from_tasklet(iwceq, (struct tasklet_struct *)t, dpc_tasklet); struct irdma_pci_f *rf = iwceq->rf; irdma_process_ceq(rf, iwceq); irdma_ena_intr(&rf->sc_dev, iwceq->msix_idx); } /** * irdma_save_msix_info - copy msix vector information to iwarp device * @rf: RDMA PCI function * * Allocate iwdev msix table and copy the msix info to the table * Return 0 if successful, otherwise return error */ static int irdma_save_msix_info(struct irdma_pci_f *rf) { struct irdma_qvlist_info *iw_qvlist; struct irdma_qv_info *iw_qvinfo; u16 ceq_idx; u32 i; u32 size; if (!rf->msix_count) { irdma_dev_err(to_ibdev(&rf->sc_dev), "No MSI-X vectors reserved for RDMA.\n"); return -EINVAL; } size = sizeof(struct irdma_msix_vector) * rf->msix_count; size += sizeof(*iw_qvlist); size += sizeof(*iw_qvinfo) * rf->msix_count - 1; rf->iw_msixtbl = kzalloc(size, GFP_KERNEL); if (!rf->iw_msixtbl) return -ENOMEM; rf->iw_qvlist = (struct irdma_qvlist_info *) (&rf->iw_msixtbl[rf->msix_count]); iw_qvlist = rf->iw_qvlist; iw_qvinfo = iw_qvlist->qv_info; iw_qvlist->num_vectors = rf->msix_count; if (rf->msix_count <= num_online_cpus()) rf->msix_shared = true; else if (rf->msix_count > num_online_cpus() + 1) rf->msix_count = num_online_cpus() + 1; for (i = 0, ceq_idx = 0; i < rf->msix_count; i++, iw_qvinfo++) { rf->iw_msixtbl[i].idx = rf->msix_info.entry + i; rf->iw_msixtbl[i].cpu_affinity = ceq_idx; if (!i) { iw_qvinfo->aeq_idx = 0; if (rf->msix_shared) iw_qvinfo->ceq_idx = ceq_idx++; else iw_qvinfo->ceq_idx = IRDMA_Q_INVALID_IDX; } else { iw_qvinfo->aeq_idx = IRDMA_Q_INVALID_IDX; iw_qvinfo->ceq_idx = ceq_idx++; } iw_qvinfo->itr_idx = IRDMA_IDX_NOITR; iw_qvinfo->v_idx = rf->iw_msixtbl[i].idx; } return 0; } /** * irdma_irq_handler - interrupt handler for aeq and ceq0 * @data: RDMA PCI function */ static void irdma_irq_handler(void *data) { struct irdma_pci_f *rf = data; tasklet_schedule(&rf->dpc_tasklet); } /** * irdma_ceq_handler - interrupt handler for ceq * @data: ceq pointer */ static void irdma_ceq_handler(void *data) { struct irdma_ceq *iwceq = data; tasklet_schedule(&iwceq->dpc_tasklet); } /** * irdma_free_irq - free device interrupts in FreeBSD manner * @rf: RDMA PCI function * @msix_vec: msix vector to disable irq * * The function is called when destroying irq. It tearsdown * the interrupt and release resources. */ static void irdma_free_irq(struct irdma_pci_f *rf, struct irdma_msix_vector *msix_vec) { if (msix_vec->tag) { bus_teardown_intr(rf->dev_ctx.dev, msix_vec->res, msix_vec->tag); msix_vec->tag = NULL; } if (msix_vec->res) { bus_release_resource(rf->dev_ctx.dev, SYS_RES_IRQ, msix_vec->idx + 1, msix_vec->res); msix_vec->res = NULL; } } /** * irdma_destroy_irq - destroy device interrupts * @rf: RDMA PCI function * @msix_vec: msix vector to disable irq * @dev_id: parameter to pass to free_irq (used during irq setup) * * The function is called when destroying aeq/ceq */ static void irdma_destroy_irq(struct irdma_pci_f *rf, struct irdma_msix_vector *msix_vec, void *dev_id) { struct irdma_sc_dev *dev = &rf->sc_dev; dev->irq_ops->irdma_dis_irq(dev, msix_vec->idx); irdma_free_irq(rf, msix_vec); if (rf == dev_id) { tasklet_kill(&rf->dpc_tasklet); } else { struct irdma_ceq *iwceq = (struct irdma_ceq *)dev_id; tasklet_kill(&iwceq->dpc_tasklet); } } /** * irdma_destroy_cqp - destroy control qp * @rf: RDMA PCI function * @free_hwcqp: 1 if hw cqp should be freed * * Issue destroy cqp request and * free the resources associated with the cqp */ static void irdma_destroy_cqp(struct irdma_pci_f *rf, bool free_hwcqp) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_cqp *cqp = &rf->cqp; int status = 0; if (rf->cqp_cmpl_wq) destroy_workqueue(rf->cqp_cmpl_wq); status = irdma_sc_cqp_destroy(dev->cqp, free_hwcqp); if (status) irdma_debug(dev, IRDMA_DEBUG_ERR, "Destroy CQP failed %d\n", status); irdma_cleanup_pending_cqp_op(rf); irdma_free_dma_mem(dev->hw, &cqp->sq); kfree(cqp->scratch_array); cqp->scratch_array = NULL; kfree(cqp->cqp_requests); cqp->cqp_requests = NULL; } static void irdma_destroy_virt_aeq(struct irdma_pci_f *rf) { struct irdma_aeq *aeq = &rf->aeq; u32 pg_cnt = DIV_ROUND_UP(aeq->mem.size, PAGE_SIZE); dma_addr_t *pg_arr = (dma_addr_t *) aeq->palloc.level1.addr; irdma_unmap_vm_page_list(&rf->hw, pg_arr, pg_cnt); irdma_free_pble(rf->pble_rsrc, &aeq->palloc); vfree(aeq->mem.va); } /** * irdma_destroy_aeq - destroy aeq * @rf: RDMA PCI function * * Issue a destroy aeq request and * free the resources associated with the aeq * The function is called during driver unload */ static void irdma_destroy_aeq(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_aeq *aeq = &rf->aeq; int status = -EBUSY; if (!rf->msix_shared) { rf->sc_dev.irq_ops->irdma_cfg_aeq(&rf->sc_dev, rf->iw_msixtbl->idx, false); irdma_destroy_irq(rf, rf->iw_msixtbl, rf); } if (rf->reset) goto exit; aeq->sc_aeq.size = 0; status = irdma_cqp_aeq_cmd(dev, &aeq->sc_aeq, IRDMA_OP_AEQ_DESTROY); if (status) irdma_debug(dev, IRDMA_DEBUG_ERR, "Destroy AEQ failed %d\n", status); exit: if (aeq->virtual_map) irdma_destroy_virt_aeq(rf); else irdma_free_dma_mem(dev->hw, &aeq->mem); } /** * irdma_destroy_ceq - destroy ceq * @rf: RDMA PCI function * @iwceq: ceq to be destroyed * * Issue a destroy ceq request and * free the resources associated with the ceq */ static void irdma_destroy_ceq(struct irdma_pci_f *rf, struct irdma_ceq *iwceq) { struct irdma_sc_dev *dev = &rf->sc_dev; int status; if (rf->reset) goto exit; status = irdma_sc_ceq_destroy(&iwceq->sc_ceq, 0, 1); if (status) { irdma_debug(dev, IRDMA_DEBUG_ERR, "CEQ destroy command failed %d\n", status); goto exit; } status = irdma_sc_cceq_destroy_done(&iwceq->sc_ceq); if (status) irdma_debug(dev, IRDMA_DEBUG_ERR, "CEQ destroy completion failed %d\n", status); exit: spin_lock_destroy(&iwceq->ce_lock); spin_lock_destroy(&iwceq->sc_ceq.req_cq_lock); kfree(iwceq->sc_ceq.reg_cq); irdma_free_dma_mem(dev->hw, &iwceq->mem); } /** * irdma_del_ceq_0 - destroy ceq 0 * @rf: RDMA PCI function * * Disable the ceq 0 interrupt and destroy the ceq 0 */ static void irdma_del_ceq_0(struct irdma_pci_f *rf) { struct irdma_ceq *iwceq = rf->ceqlist; struct irdma_msix_vector *msix_vec; if (rf->msix_shared) { msix_vec = &rf->iw_msixtbl[0]; rf->sc_dev.irq_ops->irdma_cfg_ceq(&rf->sc_dev, msix_vec->ceq_id, msix_vec->idx, false); irdma_destroy_irq(rf, msix_vec, rf); } else { msix_vec = &rf->iw_msixtbl[1]; irdma_destroy_irq(rf, msix_vec, iwceq); } irdma_destroy_ceq(rf, iwceq); rf->sc_dev.ceq_valid = false; rf->ceqs_count = 0; } /** * irdma_del_ceqs - destroy all ceq's except CEQ 0 * @rf: RDMA PCI function * * Go through all of the device ceq's, except 0, and for each * ceq disable the ceq interrupt and destroy the ceq */ static void irdma_del_ceqs(struct irdma_pci_f *rf) { struct irdma_ceq *iwceq = &rf->ceqlist[1]; struct irdma_msix_vector *msix_vec; u32 i = 0; if (rf->msix_shared) msix_vec = &rf->iw_msixtbl[1]; else msix_vec = &rf->iw_msixtbl[2]; for (i = 1; i < rf->ceqs_count; i++, msix_vec++, iwceq++) { rf->sc_dev.irq_ops->irdma_cfg_ceq(&rf->sc_dev, msix_vec->ceq_id, msix_vec->idx, false); irdma_destroy_irq(rf, msix_vec, iwceq); irdma_cqp_ceq_cmd(&rf->sc_dev, &iwceq->sc_ceq, IRDMA_OP_CEQ_DESTROY); spin_lock_destroy(&iwceq->ce_lock); spin_lock_destroy(&iwceq->sc_ceq.req_cq_lock); kfree(iwceq->sc_ceq.reg_cq); irdma_free_dma_mem(rf->sc_dev.hw, &iwceq->mem); } rf->ceqs_count = 1; } /** * irdma_destroy_ccq - destroy control cq * @rf: RDMA PCI function * * Issue destroy ccq request and * free the resources associated with the ccq */ static void irdma_destroy_ccq(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_ccq *ccq = &rf->ccq; int status = 0; if (!rf->reset) status = irdma_sc_ccq_destroy(dev->ccq, 0, true); if (status) irdma_debug(dev, IRDMA_DEBUG_ERR, "CCQ destroy failed %d\n", status); irdma_free_dma_mem(dev->hw, &ccq->mem_cq); } /** * irdma_close_hmc_objects_type - delete hmc objects of a given type * @dev: iwarp device * @obj_type: the hmc object type to be deleted * @hmc_info: host memory info struct * @privileged: permission to close HMC objects * @reset: true if called before reset */ static void irdma_close_hmc_objects_type(struct irdma_sc_dev *dev, enum irdma_hmc_rsrc_type obj_type, struct irdma_hmc_info *hmc_info, bool privileged, bool reset) { struct irdma_hmc_del_obj_info info = {0}; info.hmc_info = hmc_info; info.rsrc_type = obj_type; info.count = hmc_info->hmc_obj[obj_type].cnt; info.privileged = privileged; if (irdma_sc_del_hmc_obj(dev, &info, reset)) irdma_debug(dev, IRDMA_DEBUG_ERR, "del HMC obj of type %d failed\n", obj_type); } /** * irdma_del_hmc_objects - remove all device hmc objects * @dev: iwarp device * @hmc_info: hmc_info to free * @privileged: permission to delete HMC objects * @reset: true if called before reset * @vers: hardware version */ void irdma_del_hmc_objects(struct irdma_sc_dev *dev, struct irdma_hmc_info *hmc_info, bool privileged, bool reset, enum irdma_vers vers) { unsigned int i; for (i = 0; i < IW_HMC_OBJ_TYPE_NUM; i++) { if (dev->hmc_info->hmc_obj[iw_hmc_obj_types[i]].cnt) irdma_close_hmc_objects_type(dev, iw_hmc_obj_types[i], hmc_info, privileged, reset); if (vers == IRDMA_GEN_1 && i == IRDMA_HMC_IW_TIMER) break; } } /** * irdma_create_hmc_obj_type - create hmc object of a given type * @dev: hardware control device structure * @info: information for the hmc object to create */ static int irdma_create_hmc_obj_type(struct irdma_sc_dev *dev, struct irdma_hmc_create_obj_info *info) { return irdma_sc_create_hmc_obj(dev, info); } /** * irdma_create_hmc_objs - create all hmc objects for the device * @rf: RDMA PCI function * @privileged: permission to create HMC objects * @vers: HW version * * Create the device hmc objects and allocate hmc pages * Return 0 if successful, otherwise clean up and return error */ static int irdma_create_hmc_objs(struct irdma_pci_f *rf, bool privileged, enum irdma_vers vers) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_hmc_create_obj_info info = {0}; int i, status = 0; info.hmc_info = dev->hmc_info; info.privileged = privileged; info.entry_type = rf->sd_type; for (i = 0; i < IW_HMC_OBJ_TYPE_NUM; i++) { if (iw_hmc_obj_types[i] == IRDMA_HMC_IW_PBLE) continue; if (dev->hmc_info->hmc_obj[iw_hmc_obj_types[i]].cnt) { info.rsrc_type = iw_hmc_obj_types[i]; info.count = dev->hmc_info->hmc_obj[info.rsrc_type].cnt; info.add_sd_cnt = 0; status = irdma_create_hmc_obj_type(dev, &info); if (status) { irdma_debug(dev, IRDMA_DEBUG_ERR, "create obj type %d status = %d\n", iw_hmc_obj_types[i], status); break; } } if (vers == IRDMA_GEN_1 && i == IRDMA_HMC_IW_TIMER) break; } if (!status) return irdma_sc_static_hmc_pages_allocated(dev->cqp, 0, dev->hmc_fn_id, true, true); while (i) { i--; /* destroy the hmc objects of a given type */ if (dev->hmc_info->hmc_obj[iw_hmc_obj_types[i]].cnt) irdma_close_hmc_objects_type(dev, iw_hmc_obj_types[i], dev->hmc_info, privileged, false); } return status; } /** * irdma_obj_aligned_mem - get aligned memory from device allocated memory * @rf: RDMA PCI function * @memptr: points to the memory addresses * @size: size of memory needed * @mask: mask for the aligned memory * * Get aligned memory of the requested size and * update the memptr to point to the new aligned memory * Return 0 if successful, otherwise return no memory error */ static int irdma_obj_aligned_mem(struct irdma_pci_f *rf, struct irdma_dma_mem *memptr, u32 size, u32 mask) { unsigned long va, newva; unsigned long extra; va = (unsigned long)rf->obj_next.va; newva = va; if (mask) newva = ALIGN(va, (unsigned long)mask + 1ULL); extra = newva - va; memptr->va = (u8 *)va + extra; memptr->pa = rf->obj_next.pa + extra; memptr->size = size; if (((u8 *)memptr->va + size) > ((u8 *)rf->obj_mem.va + rf->obj_mem.size)) return -ENOMEM; rf->obj_next.va = (u8 *)memptr->va + size; rf->obj_next.pa = memptr->pa + size; return 0; } /** * irdma_create_cqp - create control qp * @rf: RDMA PCI function * * Return 0, if the cqp and all the resources associated with it * are successfully created, otherwise return error */ static int irdma_create_cqp(struct irdma_pci_f *rf) { u32 sqsize = IRDMA_CQP_SW_SQSIZE_2048; struct irdma_dma_mem mem; struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_cqp_init_info cqp_init_info = {0}; struct irdma_cqp *cqp = &rf->cqp; u16 maj_err, min_err; int i, status; cqp->cqp_requests = kcalloc(sqsize, sizeof(*cqp->cqp_requests), GFP_KERNEL); if (!cqp->cqp_requests) return -ENOMEM; cqp->scratch_array = kcalloc(sqsize, sizeof(*cqp->scratch_array), GFP_KERNEL); if (!cqp->scratch_array) { status = -ENOMEM; goto err_scratch; } dev->cqp = &cqp->sc_cqp; dev->cqp->dev = dev; cqp->sq.size = sizeof(struct irdma_cqp_sq_wqe) * sqsize; cqp->sq.va = irdma_allocate_dma_mem(dev->hw, &cqp->sq, cqp->sq.size, IRDMA_CQP_ALIGNMENT); if (!cqp->sq.va) { status = -ENOMEM; goto err_sq; } status = irdma_obj_aligned_mem(rf, &mem, sizeof(struct irdma_cqp_ctx), IRDMA_HOST_CTX_ALIGNMENT_M); if (status) goto err_ctx; dev->cqp->host_ctx_pa = mem.pa; dev->cqp->host_ctx = mem.va; /* populate the cqp init info */ cqp_init_info.dev = dev; cqp_init_info.sq_size = sqsize; cqp_init_info.sq = cqp->sq.va; cqp_init_info.sq_pa = cqp->sq.pa; cqp_init_info.host_ctx_pa = mem.pa; cqp_init_info.host_ctx = mem.va; cqp_init_info.hmc_profile = rf->rsrc_profile; cqp_init_info.scratch_array = cqp->scratch_array; cqp_init_info.protocol_used = rf->protocol_used; cqp_init_info.en_rem_endpoint_trk = rf->en_rem_endpoint_trk; memcpy(&cqp_init_info.dcqcn_params, &rf->dcqcn_params, sizeof(cqp_init_info.dcqcn_params)); switch (rf->rdma_ver) { case IRDMA_GEN_1: cqp_init_info.hw_maj_ver = IRDMA_CQPHC_HW_MAJVER_GEN_1; break; case IRDMA_GEN_2: cqp_init_info.hw_maj_ver = IRDMA_CQPHC_HW_MAJVER_GEN_2; break; } status = irdma_sc_cqp_init(dev->cqp, &cqp_init_info); if (status) { irdma_debug(dev, IRDMA_DEBUG_ERR, "cqp init status %d\n", status); goto err_ctx; } spin_lock_init(&cqp->req_lock); spin_lock_init(&cqp->compl_lock); status = irdma_sc_cqp_create(dev->cqp, &maj_err, &min_err); if (status) { irdma_debug(dev, IRDMA_DEBUG_ERR, "cqp create failed - status %d maj_err %d min_err %d\n", status, maj_err, min_err); goto err_ctx; } INIT_LIST_HEAD(&cqp->cqp_avail_reqs); INIT_LIST_HEAD(&cqp->cqp_pending_reqs); /* init the waitqueue of the cqp_requests and add them to the list */ for (i = 0; i < sqsize; i++) { init_waitqueue_head(&cqp->cqp_requests[i].waitq); list_add_tail(&cqp->cqp_requests[i].list, &cqp->cqp_avail_reqs); } init_waitqueue_head(&cqp->remove_wq); return 0; err_ctx: irdma_free_dma_mem(dev->hw, &cqp->sq); err_sq: kfree(cqp->scratch_array); cqp->scratch_array = NULL; err_scratch: kfree(cqp->cqp_requests); cqp->cqp_requests = NULL; return status; } /** * irdma_create_ccq - create control cq * @rf: RDMA PCI function * * Return 0, if the ccq and the resources associated with it * are successfully created, otherwise return error */ static int irdma_create_ccq(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_ccq_init_info info = {0}; struct irdma_ccq *ccq = &rf->ccq; int status; dev->ccq = &ccq->sc_cq; dev->ccq->dev = dev; info.dev = dev; ccq->shadow_area.size = sizeof(struct irdma_cq_shadow_area); ccq->mem_cq.size = sizeof(struct irdma_cqe) * IW_CCQ_SIZE; ccq->mem_cq.va = irdma_allocate_dma_mem(dev->hw, &ccq->mem_cq, ccq->mem_cq.size, IRDMA_CQ0_ALIGNMENT); if (!ccq->mem_cq.va) return -ENOMEM; status = irdma_obj_aligned_mem(rf, &ccq->shadow_area, ccq->shadow_area.size, IRDMA_SHADOWAREA_M); if (status) goto exit; ccq->sc_cq.back_cq = ccq; /* populate the ccq init info */ info.cq_base = ccq->mem_cq.va; info.cq_pa = ccq->mem_cq.pa; info.num_elem = IW_CCQ_SIZE; info.shadow_area = ccq->shadow_area.va; info.shadow_area_pa = ccq->shadow_area.pa; info.ceqe_mask = false; info.ceq_id_valid = true; info.shadow_read_threshold = 16; info.vsi = &rf->default_vsi; status = irdma_sc_ccq_init(dev->ccq, &info); if (!status) status = irdma_sc_ccq_create(dev->ccq, 0, true, true); exit: if (status) irdma_free_dma_mem(dev->hw, &ccq->mem_cq); return status; } /** * irdma_alloc_set_mac - set up a mac address table entry * @iwdev: irdma device * * Allocate a mac ip entry and add it to the hw table Return 0 * if successful, otherwise return error */ static int irdma_alloc_set_mac(struct irdma_device *iwdev) { int status; status = irdma_alloc_local_mac_entry(iwdev->rf, &iwdev->mac_ip_table_idx); if (!status) { status = irdma_add_local_mac_entry(iwdev->rf, (const u8 *)if_getlladdr(iwdev->netdev), (u8)iwdev->mac_ip_table_idx); if (status) irdma_del_local_mac_entry(iwdev->rf, (u8)iwdev->mac_ip_table_idx); } return status; } /** * irdma_irq_request - set up the msix interrupt vector * @rf: RDMA PCI function * @msix_vec: interrupt vector information * @handler: function pointer to associate with interrupt * @argument: argument passed to the handler * * Allocate interrupt resources and setup interrupt * Return 0 if successful, otherwise return error * Note that after this function bus_describe_intr shall * be called. */ static int irdma_irq_request(struct irdma_pci_f *rf, struct irdma_msix_vector *msix_vec, driver_intr_t handler, void *argument) { device_t dev = rf->dev_ctx.dev; int rid = msix_vec->idx + 1; int err, status; msix_vec->res = bus_alloc_resource_any(dev, SYS_RES_IRQ, &rid, RF_SHAREABLE | RF_ACTIVE); if (!msix_vec->res) { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_ERR, "Unable to allocate bus resource int[%d]\n", rid); return -EINVAL; } err = bus_setup_intr(dev, msix_vec->res, INTR_TYPE_NET | INTR_MPSAFE, NULL, handler, argument, &msix_vec->tag); if (err) { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_ERR, "Unable to register handler with %x status\n", err); status = -EINVAL; goto fail_intr; } return 0; fail_intr: bus_release_resource(dev, SYS_RES_IRQ, rid, msix_vec->res); msix_vec->res = NULL; return status; } /** * irdma_cfg_ceq_vector - set up the msix interrupt vector for * ceq * @rf: RDMA PCI function * @iwceq: ceq associated with the vector * @ceq_id: the id number of the iwceq * @msix_vec: interrupt vector information * * Allocate interrupt resources and enable irq handling * Return 0 if successful, otherwise return error */ static int irdma_cfg_ceq_vector(struct irdma_pci_f *rf, struct irdma_ceq *iwceq, u16 ceq_id, struct irdma_msix_vector *msix_vec) { int status; if (rf->msix_shared && !ceq_id) { snprintf(msix_vec->name, sizeof(msix_vec->name) - 1, "irdma-%s-AEQCEQ-0", dev_name(&rf->pcidev->dev)); tasklet_setup(&rf->dpc_tasklet, irdma_dpc); status = irdma_irq_request(rf, msix_vec, irdma_irq_handler, rf); if (status) return status; bus_describe_intr(rf->dev_ctx.dev, msix_vec->res, msix_vec->tag, "%s", msix_vec->name); } else { snprintf(msix_vec->name, sizeof(msix_vec->name) - 1, "irdma-%s-CEQ-%d", dev_name(&rf->pcidev->dev), ceq_id); tasklet_setup(&iwceq->dpc_tasklet, irdma_ceq_dpc); status = irdma_irq_request(rf, msix_vec, irdma_ceq_handler, iwceq); if (status) return status; bus_describe_intr(rf->dev_ctx.dev, msix_vec->res, msix_vec->tag, "%s", msix_vec->name); } msix_vec->ceq_id = ceq_id; rf->sc_dev.irq_ops->irdma_cfg_ceq(&rf->sc_dev, ceq_id, msix_vec->idx, true); return 0; } /** * irdma_cfg_aeq_vector - set up the msix vector for aeq * @rf: RDMA PCI function * * Allocate interrupt resources and enable irq handling * Return 0 if successful, otherwise return error */ static int irdma_cfg_aeq_vector(struct irdma_pci_f *rf) { struct irdma_msix_vector *msix_vec = rf->iw_msixtbl; int status = 0; if (!rf->msix_shared) { snprintf(msix_vec->name, sizeof(msix_vec->name) - 1, "irdma-%s-AEQ", dev_name(&rf->pcidev->dev)); tasklet_setup(&rf->dpc_tasklet, irdma_dpc); status = irdma_irq_request(rf, msix_vec, irdma_irq_handler, rf); if (status) return status; bus_describe_intr(rf->dev_ctx.dev, msix_vec->res, msix_vec->tag, "%s", msix_vec->name); } if (status) { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_ERR, "aeq irq config fail\n"); return status; } rf->sc_dev.irq_ops->irdma_cfg_aeq(&rf->sc_dev, msix_vec->idx, true); return 0; } /** * irdma_create_ceq - create completion event queue * @rf: RDMA PCI function * @iwceq: pointer to the ceq resources to be created * @ceq_id: the id number of the iwceq * @vsi: SC vsi struct * * Return 0, if the ceq and the resources associated with it * are successfully created, otherwise return error */ static int irdma_create_ceq(struct irdma_pci_f *rf, struct irdma_ceq *iwceq, u16 ceq_id, struct irdma_sc_vsi *vsi) { int status; struct irdma_ceq_init_info info = {0}; struct irdma_sc_dev *dev = &rf->sc_dev; u64 scratch; u32 ceq_size; info.ceq_id = ceq_id; iwceq->rf = rf; ceq_size = min(rf->sc_dev.hmc_info->hmc_obj[IRDMA_HMC_IW_CQ].cnt, dev->hw_attrs.max_hw_ceq_size); iwceq->mem.size = sizeof(struct irdma_ceqe) * ceq_size; iwceq->mem.va = irdma_allocate_dma_mem(dev->hw, &iwceq->mem, iwceq->mem.size, IRDMA_CEQ_ALIGNMENT); if (!iwceq->mem.va) return -ENOMEM; info.ceq_id = ceq_id; info.ceqe_base = iwceq->mem.va; info.ceqe_pa = iwceq->mem.pa; info.elem_cnt = ceq_size; info.reg_cq = kzalloc(sizeof(struct irdma_sc_cq *) * info.elem_cnt, GFP_KERNEL); iwceq->sc_ceq.ceq_id = ceq_id; info.dev = dev; info.vsi = vsi; scratch = (uintptr_t)&rf->cqp.sc_cqp; status = irdma_sc_ceq_init(&iwceq->sc_ceq, &info); if (!status) { if (dev->ceq_valid) status = irdma_cqp_ceq_cmd(&rf->sc_dev, &iwceq->sc_ceq, IRDMA_OP_CEQ_CREATE); else status = irdma_sc_cceq_create(&iwceq->sc_ceq, scratch); } if (status) { kfree(info.reg_cq); irdma_free_dma_mem(dev->hw, &iwceq->mem); } return status; } /** * irdma_setup_ceq_0 - create CEQ 0 and it's interrupt resource * @rf: RDMA PCI function * * Allocate a list for all device completion event queues * Create the ceq 0 and configure it's msix interrupt vector * Return 0, if successfully set up, otherwise return error */ static int irdma_setup_ceq_0(struct irdma_pci_f *rf) { struct irdma_ceq *iwceq; struct irdma_msix_vector *msix_vec; u32 i; int status = 0; u32 num_ceqs; num_ceqs = min(rf->msix_count, rf->sc_dev.hmc_fpm_misc.max_ceqs); rf->ceqlist = kcalloc(num_ceqs, sizeof(*rf->ceqlist), GFP_KERNEL); memset(rf->ceqlist, 0, num_ceqs * sizeof(*rf->ceqlist)); if (!rf->ceqlist) { status = -ENOMEM; goto exit; } iwceq = &rf->ceqlist[0]; status = irdma_create_ceq(rf, iwceq, 0, &rf->default_vsi); if (status) { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_ERR, "create ceq status = %d\n", status); goto exit; } spin_lock_init(&iwceq->ce_lock); i = rf->msix_shared ? 0 : 1; msix_vec = &rf->iw_msixtbl[i]; iwceq->irq = msix_vec->irq; iwceq->msix_idx = msix_vec->idx; status = irdma_cfg_ceq_vector(rf, iwceq, 0, msix_vec); if (status) { irdma_destroy_ceq(rf, iwceq); goto exit; } irdma_ena_intr(&rf->sc_dev, msix_vec->idx); rf->ceqs_count++; exit: if (status && !rf->ceqs_count) { kfree(rf->ceqlist); rf->ceqlist = NULL; return status; } rf->sc_dev.ceq_valid = true; return 0; } /** * irdma_setup_ceqs - manage the device ceq's and their interrupt resources * @rf: RDMA PCI function * @vsi: VSI structure for this CEQ * * Allocate a list for all device completion event queues * Create the ceq's and configure their msix interrupt vectors * Return 0, if ceqs are successfully set up, otherwise return error */ static int irdma_setup_ceqs(struct irdma_pci_f *rf, struct irdma_sc_vsi *vsi) { u32 i; u16 ceq_id; struct irdma_ceq *iwceq; struct irdma_msix_vector *msix_vec; int status; u32 num_ceqs; num_ceqs = min(rf->msix_count, rf->sc_dev.hmc_fpm_misc.max_ceqs); i = (rf->msix_shared) ? 1 : 2; for (ceq_id = 1; i < num_ceqs; i++, ceq_id++) { iwceq = &rf->ceqlist[ceq_id]; status = irdma_create_ceq(rf, iwceq, ceq_id, vsi); if (status) { irdma_debug(&rf->sc_dev, IRDMA_DEBUG_ERR, "create ceq status = %d\n", status); goto del_ceqs; } spin_lock_init(&iwceq->ce_lock); msix_vec = &rf->iw_msixtbl[i]; iwceq->irq = msix_vec->irq; iwceq->msix_idx = msix_vec->idx; status = irdma_cfg_ceq_vector(rf, iwceq, ceq_id, msix_vec); if (status) { irdma_destroy_ceq(rf, iwceq); goto del_ceqs; } irdma_ena_intr(&rf->sc_dev, msix_vec->idx); rf->ceqs_count++; } return 0; del_ceqs: irdma_del_ceqs(rf); return status; } static int irdma_create_virt_aeq(struct irdma_pci_f *rf, u32 size) { struct irdma_aeq *aeq = &rf->aeq; dma_addr_t *pg_arr; u32 pg_cnt; int status; if (rf->rdma_ver < IRDMA_GEN_2) return -EOPNOTSUPP; aeq->mem.size = sizeof(struct irdma_sc_aeqe) * size; aeq->mem.va = vzalloc(aeq->mem.size); if (!aeq->mem.va) return -ENOMEM; pg_cnt = DIV_ROUND_UP(aeq->mem.size, PAGE_SIZE); status = irdma_get_pble(rf->pble_rsrc, &aeq->palloc, pg_cnt, true); if (status) { vfree(aeq->mem.va); return status; } pg_arr = (dma_addr_t *) aeq->palloc.level1.addr; status = irdma_map_vm_page_list(&rf->hw, aeq->mem.va, pg_arr, pg_cnt); if (status) { irdma_free_pble(rf->pble_rsrc, &aeq->palloc); vfree(aeq->mem.va); return status; } return 0; } /** * irdma_create_aeq - create async event queue * @rf: RDMA PCI function * * Return 0, if the aeq and the resources associated with it * are successfully created, otherwise return error */ static int irdma_create_aeq(struct irdma_pci_f *rf) { struct irdma_aeq_init_info info = {0}; struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_aeq *aeq = &rf->aeq; struct irdma_hmc_info *hmc_info = rf->sc_dev.hmc_info; u32 aeq_size; u8 multiplier = (rf->protocol_used == IRDMA_IWARP_PROTOCOL_ONLY) ? 2 : 1; int status; aeq_size = multiplier * hmc_info->hmc_obj[IRDMA_HMC_IW_QP].cnt + hmc_info->hmc_obj[IRDMA_HMC_IW_CQ].cnt; aeq_size = min(aeq_size, dev->hw_attrs.max_hw_aeq_size); aeq->mem.size = sizeof(struct irdma_sc_aeqe) * aeq_size; aeq->mem.va = irdma_allocate_dma_mem(dev->hw, &aeq->mem, aeq->mem.size, IRDMA_AEQ_ALIGNMENT); if (aeq->mem.va) goto skip_virt_aeq; /* physically mapped aeq failed. setup virtual aeq */ status = irdma_create_virt_aeq(rf, aeq_size); if (status) return status; info.virtual_map = true; aeq->virtual_map = info.virtual_map; info.pbl_chunk_size = 1; info.first_pm_pbl_idx = aeq->palloc.level1.idx; skip_virt_aeq: info.aeqe_base = aeq->mem.va; info.aeq_elem_pa = aeq->mem.pa; info.elem_cnt = aeq_size; info.dev = dev; info.msix_idx = rf->iw_msixtbl->idx; status = irdma_sc_aeq_init(&aeq->sc_aeq, &info); if (status) goto err; status = irdma_cqp_aeq_cmd(dev, &aeq->sc_aeq, IRDMA_OP_AEQ_CREATE); if (status) goto err; return 0; err: if (aeq->virtual_map) irdma_destroy_virt_aeq(rf); else irdma_free_dma_mem(dev->hw, &aeq->mem); return status; } /** * irdma_setup_aeq - set up the device aeq * @rf: RDMA PCI function * * Create the aeq and configure its msix interrupt vector * Return 0 if successful, otherwise return error */ static int irdma_setup_aeq(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; int status; status = irdma_create_aeq(rf); if (status) return status; status = irdma_cfg_aeq_vector(rf); if (status) { irdma_destroy_aeq(rf); return status; } if (!rf->msix_shared) irdma_ena_intr(dev, rf->iw_msixtbl[0].idx); return 0; } /** * irdma_initialize_ilq - create iwarp local queue for cm * @iwdev: irdma device * * Return 0 if successful, otherwise return error */ static int irdma_initialize_ilq(struct irdma_device *iwdev) { struct irdma_puda_rsrc_info info = {0}; int status; info.type = IRDMA_PUDA_RSRC_TYPE_ILQ; info.cq_id = 1; info.qp_id = 1; info.count = 1; info.pd_id = 1; info.abi_ver = IRDMA_ABI_VER; info.sq_size = min(iwdev->rf->max_qp / 2, (u32)32768); info.rq_size = info.sq_size; info.buf_size = 1024; info.tx_buf_cnt = 2 * info.sq_size; info.receive = irdma_receive_ilq; info.xmit_complete = irdma_free_sqbuf; status = irdma_puda_create_rsrc(&iwdev->vsi, &info); if (status) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_ERR, "ilq create fail\n"); return status; } /** * irdma_initialize_ieq - create iwarp exception queue * @iwdev: irdma device * * Return 0 if successful, otherwise return error */ static int irdma_initialize_ieq(struct irdma_device *iwdev) { struct irdma_puda_rsrc_info info = {0}; int status; info.type = IRDMA_PUDA_RSRC_TYPE_IEQ; info.cq_id = 2; info.qp_id = iwdev->vsi.exception_lan_q; info.count = 1; info.pd_id = 2; info.abi_ver = IRDMA_ABI_VER; info.sq_size = min(iwdev->rf->max_qp / 2, (u32)32768); info.rq_size = info.sq_size; info.buf_size = iwdev->vsi.mtu + IRDMA_IPV4_PAD; info.tx_buf_cnt = 4096; status = irdma_puda_create_rsrc(&iwdev->vsi, &info); if (status) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_ERR, "ieq create fail\n"); return status; } /** * irdma_reinitialize_ieq - destroy and re-create ieq * @vsi: VSI structure */ void irdma_reinitialize_ieq(struct irdma_sc_vsi *vsi) { struct irdma_device *iwdev = vsi->back_vsi; struct irdma_pci_f *rf = iwdev->rf; irdma_puda_dele_rsrc(vsi, IRDMA_PUDA_RSRC_TYPE_IEQ, false); if (irdma_initialize_ieq(iwdev)) { iwdev->rf->reset = true; rf->gen_ops.request_reset(rf); } } /** * irdma_hmc_setup - create hmc objects for the device * @rf: RDMA PCI function * * Set up the device private memory space for the number and size of * the hmc objects and create the objects * Return 0 if successful, otherwise return error */ static int irdma_hmc_setup(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; int status; u32 qpcnt; qpcnt = rsrc_limits_table[rf->limits_sel].qplimit; rf->sd_type = IRDMA_SD_TYPE_DIRECT; status = irdma_cfg_fpm_val(dev, qpcnt); if (status) return status; status = irdma_create_hmc_objs(rf, true, rf->rdma_ver); return status; } /** * irdma_del_init_mem - deallocate memory resources * @rf: RDMA PCI function */ static void irdma_del_init_mem(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; kfree(dev->hmc_info->sd_table.sd_entry); dev->hmc_info->sd_table.sd_entry = NULL; vfree(rf->mem_rsrc); rf->mem_rsrc = NULL; irdma_free_dma_mem(&rf->hw, &rf->obj_mem); if (rf->rdma_ver != IRDMA_GEN_1) { kfree(rf->allocated_ws_nodes); rf->allocated_ws_nodes = NULL; } mutex_destroy(&dev->ws_mutex); kfree(rf->ceqlist); rf->ceqlist = NULL; kfree(rf->iw_msixtbl); rf->iw_msixtbl = NULL; kfree(rf->hmc_info_mem); rf->hmc_info_mem = NULL; } /** * irdma_initialize_dev - initialize device * @rf: RDMA PCI function * * Allocate memory for the hmc objects and initialize iwdev * Return 0 if successful, otherwise clean up the resources * and return error */ static int irdma_initialize_dev(struct irdma_pci_f *rf) { int status; struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_device_init_info info = {0}; struct irdma_dma_mem mem; u32 size; size = sizeof(struct irdma_hmc_pble_rsrc) + sizeof(struct irdma_hmc_info) + (sizeof(struct irdma_hmc_obj_info) * IRDMA_HMC_IW_MAX); rf->hmc_info_mem = kzalloc(size, GFP_KERNEL); if (!rf->hmc_info_mem) return -ENOMEM; rf->pble_rsrc = (struct irdma_hmc_pble_rsrc *)rf->hmc_info_mem; dev->hmc_info = &rf->hw.hmc; dev->hmc_info->hmc_obj = (struct irdma_hmc_obj_info *) (rf->pble_rsrc + 1); status = irdma_obj_aligned_mem(rf, &mem, IRDMA_QUERY_FPM_BUF_SIZE, IRDMA_FPM_QUERY_BUF_ALIGNMENT_M); if (status) goto error; info.fpm_query_buf_pa = mem.pa; info.fpm_query_buf = mem.va; status = irdma_obj_aligned_mem(rf, &mem, IRDMA_COMMIT_FPM_BUF_SIZE, IRDMA_FPM_COMMIT_BUF_ALIGNMENT_M); if (status) goto error; info.fpm_commit_buf_pa = mem.pa; info.fpm_commit_buf = mem.va; info.bar0 = rf->hw.hw_addr; info.hmc_fn_id = rf->peer_info->pf_id; /* * the debug_mask is already assigned at this point through sysctl and so the value shouldn't be overwritten */ info.debug_mask = rf->sc_dev.debug_mask; info.hw = &rf->hw; status = irdma_sc_dev_init(&rf->sc_dev, &info); if (status) goto error; return status; error: kfree(rf->hmc_info_mem); rf->hmc_info_mem = NULL; return status; } /** * irdma_rt_deinit_hw - clean up the irdma device resources * @iwdev: irdma device * * remove the mac ip entry and ipv4/ipv6 addresses, destroy the * device queues and free the pble and the hmc objects */ void irdma_rt_deinit_hw(struct irdma_device *iwdev) { struct irdma_sc_qp qp = {{0}}; irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_INIT, "state = %d\n", iwdev->init_state); switch (iwdev->init_state) { case IP_ADDR_REGISTERED: if (iwdev->rf->sc_dev.hw_attrs.uk_attrs.hw_rev == IRDMA_GEN_1) irdma_del_local_mac_entry(iwdev->rf, (u8)iwdev->mac_ip_table_idx); /* fallthrough */ case AEQ_CREATED: case PBLE_CHUNK_MEM: case CEQS_CREATED: case REM_ENDPOINT_TRK_CREATED: if (iwdev->rf->en_rem_endpoint_trk) { qp.dev = &iwdev->rf->sc_dev; qp.qp_uk.qp_id = IRDMA_REM_ENDPOINT_TRK_QPID; qp.qp_uk.qp_type = IRDMA_QP_TYPE_IWARP; irdma_cqp_qp_destroy_cmd(qp.dev, &qp); } /* fallthrough */ case IEQ_CREATED: if (!iwdev->roce_mode) irdma_puda_dele_rsrc(&iwdev->vsi, IRDMA_PUDA_RSRC_TYPE_IEQ, iwdev->rf->reset); /* fallthrough */ case ILQ_CREATED: if (!iwdev->roce_mode) irdma_puda_dele_rsrc(&iwdev->vsi, IRDMA_PUDA_RSRC_TYPE_ILQ, iwdev->rf->reset); break; default: irdma_dev_warn(&iwdev->ibdev, "bad init_state = %d\n", iwdev->init_state); break; } irdma_cleanup_cm_core(&iwdev->cm_core); if (iwdev->vsi.pestat) { irdma_vsi_stats_free(&iwdev->vsi); kfree(iwdev->vsi.pestat); } if (iwdev->cleanup_wq) destroy_workqueue(iwdev->cleanup_wq); } static int irdma_setup_init_state(struct irdma_pci_f *rf) { int status; status = irdma_save_msix_info(rf); if (status) return status; rf->obj_mem.size = 8192; rf->obj_mem.va = irdma_allocate_dma_mem(&rf->hw, &rf->obj_mem, rf->obj_mem.size, IRDMA_HW_PAGE_SIZE); if (!rf->obj_mem.va) { status = -ENOMEM; goto clean_msixtbl; } rf->obj_next = rf->obj_mem; status = irdma_initialize_dev(rf); if (status) goto clean_obj_mem; return 0; clean_obj_mem: irdma_free_dma_mem(&rf->hw, &rf->obj_mem); clean_msixtbl: kfree(rf->iw_msixtbl); rf->iw_msixtbl = NULL; return status; } /** * irdma_get_used_rsrc - determine resources used internally * @iwdev: irdma device * * Called at the end of open to get all internal allocations */ static void irdma_get_used_rsrc(struct irdma_device *iwdev) { iwdev->rf->used_pds = find_first_zero_bit(iwdev->rf->allocated_pds, iwdev->rf->max_pd); iwdev->rf->used_qps = find_first_zero_bit(iwdev->rf->allocated_qps, iwdev->rf->max_qp); iwdev->rf->used_cqs = find_first_zero_bit(iwdev->rf->allocated_cqs, iwdev->rf->max_cq); iwdev->rf->used_mrs = find_first_zero_bit(iwdev->rf->allocated_mrs, iwdev->rf->max_mr); } void irdma_ctrl_deinit_hw(struct irdma_pci_f *rf) { enum init_completion_state state = rf->init_state; rf->init_state = INVALID_STATE; if (rf->rsrc_created) { irdma_destroy_aeq(rf); irdma_destroy_pble_prm(rf->pble_rsrc); irdma_del_ceqs(rf); rf->rsrc_created = false; } switch (state) { case CEQ0_CREATED: irdma_del_ceq_0(rf); /* fallthrough */ case CCQ_CREATED: irdma_destroy_ccq(rf); /* fallthrough */ case HW_RSRC_INITIALIZED: case HMC_OBJS_CREATED: irdma_del_hmc_objects(&rf->sc_dev, rf->sc_dev.hmc_info, true, rf->reset, rf->rdma_ver); /* fallthrough */ case CQP_CREATED: irdma_destroy_cqp(rf, !rf->reset); /* fallthrough */ case INITIAL_STATE: irdma_del_init_mem(rf); break; case INVALID_STATE: default: irdma_dev_warn(&rf->iwdev->ibdev, "bad init_state = %d\n", rf->init_state); break; } } /** * irdma_rt_init_hw - Initializes runtime portion of HW * @iwdev: irdma device * @l2params: qos, tc, mtu info from netdev driver * * Create device queues ILQ, IEQ, CEQs and PBLEs. Setup irdma * device resource objects. */ int irdma_rt_init_hw(struct irdma_device *iwdev, struct irdma_l2params *l2params) { struct irdma_pci_f *rf = iwdev->rf; struct irdma_sc_dev *dev = &rf->sc_dev; struct irdma_sc_qp qp = {{0}}; struct irdma_vsi_init_info vsi_info = {0}; struct irdma_vsi_stats_info stats_info = {0}; int status; vsi_info.dev = dev; vsi_info.back_vsi = iwdev; vsi_info.params = l2params; vsi_info.pf_data_vsi_num = iwdev->vsi_num; vsi_info.register_qset = rf->gen_ops.register_qset; vsi_info.unregister_qset = rf->gen_ops.unregister_qset; vsi_info.exception_lan_q = 2; irdma_sc_vsi_init(&iwdev->vsi, &vsi_info); status = irdma_setup_cm_core(iwdev, rf->rdma_ver); if (status) return status; stats_info.pestat = kzalloc(sizeof(*stats_info.pestat), GFP_KERNEL); if (!stats_info.pestat) { irdma_cleanup_cm_core(&iwdev->cm_core); return -ENOMEM; } stats_info.fcn_id = dev->hmc_fn_id; status = irdma_vsi_stats_init(&iwdev->vsi, &stats_info); if (status) { irdma_cleanup_cm_core(&iwdev->cm_core); kfree(stats_info.pestat); return status; } do { if (!iwdev->roce_mode) { status = irdma_initialize_ilq(iwdev); if (status) break; iwdev->init_state = ILQ_CREATED; status = irdma_initialize_ieq(iwdev); if (status) break; iwdev->init_state = IEQ_CREATED; } if (iwdev->rf->en_rem_endpoint_trk) { qp.dev = dev; qp.qp_uk.qp_id = IRDMA_REM_ENDPOINT_TRK_QPID; qp.qp_uk.qp_type = IRDMA_QP_TYPE_IWARP; status = irdma_cqp_qp_create_cmd(dev, &qp); if (status) break; iwdev->init_state = REM_ENDPOINT_TRK_CREATED; } if (!rf->rsrc_created) { status = irdma_setup_ceqs(rf, &iwdev->vsi); if (status) break; iwdev->init_state = CEQS_CREATED; status = irdma_hmc_init_pble(&rf->sc_dev, rf->pble_rsrc); if (status) { irdma_del_ceqs(rf); break; } iwdev->init_state = PBLE_CHUNK_MEM; status = irdma_setup_aeq(rf); if (status) { irdma_destroy_pble_prm(rf->pble_rsrc); irdma_del_ceqs(rf); break; } iwdev->init_state = AEQ_CREATED; rf->rsrc_created = true; } if (iwdev->rf->sc_dev.hw_attrs.uk_attrs.hw_rev == IRDMA_GEN_1) irdma_alloc_set_mac(iwdev); irdma_add_ip(iwdev); iwdev->init_state = IP_ADDR_REGISTERED; /* * handles asynch cleanup tasks - disconnect CM , free qp, free cq bufs */ iwdev->cleanup_wq = alloc_workqueue("irdma-cleanup-wq", WQ_UNBOUND, WQ_UNBOUND_MAX_ACTIVE); if (!iwdev->cleanup_wq) return -ENOMEM; irdma_get_used_rsrc(iwdev); init_waitqueue_head(&iwdev->suspend_wq); return 0; } while (0); dev_err(&rf->pcidev->dev, "HW runtime init FAIL status = %d last cmpl = %d\n", status, iwdev->init_state); irdma_rt_deinit_hw(iwdev); return status; } /** * irdma_ctrl_init_hw - Initializes control portion of HW * @rf: RDMA PCI function * * Create admin queues, HMC obejcts and RF resource objects */ int irdma_ctrl_init_hw(struct irdma_pci_f *rf) { struct irdma_sc_dev *dev = &rf->sc_dev; int status; do { status = irdma_setup_init_state(rf); if (status) break; rf->init_state = INITIAL_STATE; status = irdma_create_cqp(rf); if (status) break; rf->init_state = CQP_CREATED; dev->feature_info[IRDMA_FEATURE_FW_INFO] = IRDMA_FW_VER_DEFAULT; if (rf->rdma_ver != IRDMA_GEN_1) { status = irdma_get_rdma_features(dev); if (status) break; } status = irdma_hmc_setup(rf); if (status) break; rf->init_state = HMC_OBJS_CREATED; status = irdma_initialize_hw_rsrc(rf); if (status) break; rf->init_state = HW_RSRC_INITIALIZED; status = irdma_create_ccq(rf); if (status) break; rf->init_state = CCQ_CREATED; status = irdma_setup_ceq_0(rf); if (status) break; rf->init_state = CEQ0_CREATED; /* Handles processing of CQP completions */ rf->cqp_cmpl_wq = alloc_ordered_workqueue("cqp_cmpl_wq", WQ_HIGHPRI | WQ_UNBOUND); if (!rf->cqp_cmpl_wq) { status = -ENOMEM; break; } INIT_WORK(&rf->cqp_cmpl_work, cqp_compl_worker); irdma_sc_ccq_arm(dev->ccq); return 0; } while (0); pr_err("IRDMA hardware initialization FAILED init_state=%d status=%d\n", rf->init_state, status); irdma_ctrl_deinit_hw(rf); return status; } /** * irdma_set_hw_rsrc - set hw memory resources. * @rf: RDMA PCI function */ static void irdma_set_hw_rsrc(struct irdma_pci_f *rf) { rf->allocated_qps = (void *)(rf->mem_rsrc + (sizeof(struct irdma_arp_entry) * rf->arp_table_size)); rf->allocated_cqs = &rf->allocated_qps[BITS_TO_LONGS(rf->max_qp)]; rf->allocated_mrs = &rf->allocated_cqs[BITS_TO_LONGS(rf->max_cq)]; rf->allocated_pds = &rf->allocated_mrs[BITS_TO_LONGS(rf->max_mr)]; rf->allocated_ahs = &rf->allocated_pds[BITS_TO_LONGS(rf->max_pd)]; rf->allocated_mcgs = &rf->allocated_ahs[BITS_TO_LONGS(rf->max_ah)]; rf->allocated_arps = &rf->allocated_mcgs[BITS_TO_LONGS(rf->max_mcg)]; rf->qp_table = (struct irdma_qp **) (&rf->allocated_arps[BITS_TO_LONGS(rf->arp_table_size)]); rf->cq_table = (struct irdma_cq **)(&rf->qp_table[rf->max_qp]); spin_lock_init(&rf->rsrc_lock); spin_lock_init(&rf->arp_lock); spin_lock_init(&rf->qptable_lock); spin_lock_init(&rf->cqtable_lock); spin_lock_init(&rf->qh_list_lock); } /** * irdma_calc_mem_rsrc_size - calculate memory resources size. * @rf: RDMA PCI function */ static u32 irdma_calc_mem_rsrc_size(struct irdma_pci_f *rf){ u32 rsrc_size; rsrc_size = sizeof(struct irdma_arp_entry) * rf->arp_table_size; rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->max_qp); rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->max_mr); rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->max_cq); rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->max_pd); rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->arp_table_size); rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->max_ah); rsrc_size += sizeof(unsigned long) * BITS_TO_LONGS(rf->max_mcg); rsrc_size += sizeof(struct irdma_qp **) * rf->max_qp; rsrc_size += sizeof(struct irdma_cq **) * rf->max_cq; return rsrc_size; } /** * irdma_initialize_hw_rsrc - initialize hw resource tracking array * @rf: RDMA PCI function */ u32 irdma_initialize_hw_rsrc(struct irdma_pci_f *rf) { u32 rsrc_size; u32 mrdrvbits; u32 ret; if (rf->rdma_ver != IRDMA_GEN_1) { rf->allocated_ws_nodes = kcalloc(BITS_TO_LONGS(IRDMA_MAX_WS_NODES), sizeof(unsigned long), GFP_KERNEL); if (!rf->allocated_ws_nodes) return -ENOMEM; set_bit(0, rf->allocated_ws_nodes); rf->max_ws_node_id = IRDMA_MAX_WS_NODES; } rf->max_cqe = rf->sc_dev.hw_attrs.uk_attrs.max_hw_cq_size; rf->max_qp = rf->sc_dev.hmc_info->hmc_obj[IRDMA_HMC_IW_QP].cnt; rf->max_mr = rf->sc_dev.hmc_info->hmc_obj[IRDMA_HMC_IW_MR].cnt; rf->max_cq = rf->sc_dev.hmc_info->hmc_obj[IRDMA_HMC_IW_CQ].cnt; rf->max_pd = rf->sc_dev.hw_attrs.max_hw_pds; rf->arp_table_size = rf->sc_dev.hmc_info->hmc_obj[IRDMA_HMC_IW_ARP].cnt; rf->max_ah = rf->sc_dev.hmc_info->hmc_obj[IRDMA_HMC_IW_FSIAV].cnt; rf->max_mcg = rf->max_qp; rsrc_size = irdma_calc_mem_rsrc_size(rf); rf->mem_rsrc = vzalloc(rsrc_size); if (!rf->mem_rsrc) { ret = -ENOMEM; goto mem_rsrc_vmalloc_fail; } rf->arp_table = (struct irdma_arp_entry *)rf->mem_rsrc; irdma_set_hw_rsrc(rf); set_bit(0, rf->allocated_mrs); set_bit(0, rf->allocated_qps); set_bit(0, rf->allocated_cqs); set_bit(0, rf->allocated_pds); set_bit(0, rf->allocated_arps); set_bit(0, rf->allocated_ahs); set_bit(0, rf->allocated_mcgs); set_bit(2, rf->allocated_qps); /* qp 2 IEQ */ set_bit(1, rf->allocated_qps); /* qp 1 ILQ */ set_bit(IRDMA_REM_ENDPOINT_TRK_QPID, rf->allocated_qps); /* qp 3 Remote Endpt trk */ set_bit(1, rf->allocated_cqs); set_bit(1, rf->allocated_pds); set_bit(2, rf->allocated_cqs); set_bit(2, rf->allocated_pds); INIT_LIST_HEAD(&rf->mc_qht_list.list); /* stag index mask has a minimum of 14 bits */ mrdrvbits = 24 - max(get_count_order(rf->max_mr), 14); rf->mr_stagmask = ~(((1 << mrdrvbits) - 1) << (32 - mrdrvbits)); return 0; mem_rsrc_vmalloc_fail: kfree(rf->allocated_ws_nodes); rf->allocated_ws_nodes = NULL; return ret; } /** * irdma_cqp_ce_handler - handle cqp completions * @rf: RDMA PCI function * @cq: cq for cqp completions */ void irdma_cqp_ce_handler(struct irdma_pci_f *rf, struct irdma_sc_cq *cq) { struct irdma_cqp_request *cqp_request; struct irdma_sc_dev *dev = &rf->sc_dev; u32 cqe_count = 0; struct irdma_ccq_cqe_info info; unsigned long flags; int ret; do { memset(&info, 0, sizeof(info)); spin_lock_irqsave(&rf->cqp.compl_lock, flags); ret = irdma_sc_ccq_get_cqe_info(cq, &info); spin_unlock_irqrestore(&rf->cqp.compl_lock, flags); if (ret) break; cqp_request = (struct irdma_cqp_request *) (uintptr_t)info.scratch; if (info.error && irdma_cqp_crit_err(dev, cqp_request->info.cqp_cmd, info.maj_err_code, info.min_err_code)) irdma_dev_err(&rf->iwdev->ibdev, "cqp opcode = 0x%x maj_err_code = 0x%x min_err_code = 0x%x\n", info.op_code, info.maj_err_code, info.min_err_code); if (cqp_request) { cqp_request->compl_info.maj_err_code = info.maj_err_code; cqp_request->compl_info.min_err_code = info.min_err_code; cqp_request->compl_info.op_ret_val = info.op_ret_val; cqp_request->compl_info.error = info.error; irdma_complete_cqp_request(&rf->cqp, cqp_request); } cqe_count++; } while (1); if (cqe_count) { irdma_process_bh(dev); irdma_sc_ccq_arm(dev->ccq); } } /** * cqp_compl_worker - Handle cqp completions * @work: Pointer to work structure */ void cqp_compl_worker(struct work_struct *work) { struct irdma_pci_f *rf = container_of(work, struct irdma_pci_f, cqp_cmpl_work); struct irdma_sc_cq *cq = &rf->ccq.sc_cq; irdma_cqp_ce_handler(rf, cq); } /** * irdma_lookup_apbvt_entry - lookup hash table for an existing apbvt entry corresponding to port * @cm_core: cm's core * @port: port to identify apbvt entry */ static struct irdma_apbvt_entry * irdma_lookup_apbvt_entry(struct irdma_cm_core *cm_core, u16 port) { struct irdma_apbvt_entry *entry; HASH_FOR_EACH_POSSIBLE(cm_core->apbvt_hash_tbl, entry, hlist, port) { if (entry->port == port) { entry->use_cnt++; return entry; } } return NULL; } /** * irdma_next_iw_state - modify qp state * @iwqp: iwarp qp to modify * @state: next state for qp * @del_hash: del hash * @term: term message * @termlen: length of term message */ void irdma_next_iw_state(struct irdma_qp *iwqp, u8 state, u8 del_hash, u8 term, u8 termlen) { struct irdma_modify_qp_info info = {0}; info.next_iwarp_state = state; info.remove_hash_idx = del_hash; info.cq_num_valid = true; info.arp_cache_idx_valid = true; info.dont_send_term = true; info.dont_send_fin = true; info.termlen = termlen; if (term & IRDMAQP_TERM_SEND_TERM_ONLY) info.dont_send_term = false; if (term & IRDMAQP_TERM_SEND_FIN_ONLY) info.dont_send_fin = false; if (iwqp->sc_qp.term_flags && state == IRDMA_QP_STATE_ERROR) info.reset_tcp_conn = true; iwqp->hw_iwarp_state = state; irdma_hw_modify_qp(iwqp->iwdev, iwqp, &info, 0); iwqp->iwarp_state = info.next_iwarp_state; } /** * irdma_del_local_mac_entry - remove a mac entry from the hw * table * @rf: RDMA PCI function * @idx: the index of the mac ip address to delete */ void irdma_del_local_mac_entry(struct irdma_pci_f *rf, u16 idx) { struct irdma_cqp *iwcqp = &rf->cqp; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; cqp_request = irdma_alloc_and_get_cqp_request(iwcqp, true); if (!cqp_request) return; cqp_info = &cqp_request->info; cqp_info->cqp_cmd = IRDMA_OP_DELETE_LOCAL_MAC_ENTRY; cqp_info->post_sq = 1; cqp_info->in.u.del_local_mac_entry.cqp = &iwcqp->sc_cqp; cqp_info->in.u.del_local_mac_entry.scratch = (uintptr_t)cqp_request; cqp_info->in.u.del_local_mac_entry.entry_idx = idx; cqp_info->in.u.del_local_mac_entry.ignore_ref_count = 0; irdma_handle_cqp_op(rf, cqp_request); irdma_put_cqp_request(iwcqp, cqp_request); } /** * irdma_add_local_mac_entry - add a mac ip address entry to the * hw table * @rf: RDMA PCI function * @mac_addr: pointer to mac address * @idx: the index of the mac ip address to add */ int irdma_add_local_mac_entry(struct irdma_pci_f *rf, const u8 *mac_addr, u16 idx) { struct irdma_local_mac_entry_info *info; struct irdma_cqp *iwcqp = &rf->cqp; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; int status; cqp_request = irdma_alloc_and_get_cqp_request(iwcqp, true); if (!cqp_request) return -ENOMEM; cqp_info = &cqp_request->info; cqp_info->post_sq = 1; info = &cqp_info->in.u.add_local_mac_entry.info; ether_addr_copy(info->mac_addr, mac_addr); info->entry_idx = idx; cqp_info->in.u.add_local_mac_entry.scratch = (uintptr_t)cqp_request; cqp_info->cqp_cmd = IRDMA_OP_ADD_LOCAL_MAC_ENTRY; cqp_info->in.u.add_local_mac_entry.cqp = &iwcqp->sc_cqp; cqp_info->in.u.add_local_mac_entry.scratch = (uintptr_t)cqp_request; status = irdma_handle_cqp_op(rf, cqp_request); irdma_put_cqp_request(iwcqp, cqp_request); return status; } /** * irdma_alloc_local_mac_entry - allocate a mac entry * @rf: RDMA PCI function * @mac_tbl_idx: the index of the new mac address * * Allocate a mac address entry and update the mac_tbl_idx * to hold the index of the newly created mac address * Return 0 if successful, otherwise return error */ int irdma_alloc_local_mac_entry(struct irdma_pci_f *rf, u16 *mac_tbl_idx) { struct irdma_cqp *iwcqp = &rf->cqp; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; int status = 0; cqp_request = irdma_alloc_and_get_cqp_request(iwcqp, true); if (!cqp_request) return -ENOMEM; cqp_info = &cqp_request->info; cqp_info->cqp_cmd = IRDMA_OP_ALLOC_LOCAL_MAC_ENTRY; cqp_info->post_sq = 1; cqp_info->in.u.alloc_local_mac_entry.cqp = &iwcqp->sc_cqp; cqp_info->in.u.alloc_local_mac_entry.scratch = (uintptr_t)cqp_request; status = irdma_handle_cqp_op(rf, cqp_request); if (!status) *mac_tbl_idx = (u16)cqp_request->compl_info.op_ret_val; irdma_put_cqp_request(iwcqp, cqp_request); return status; } /** * irdma_cqp_manage_apbvt_cmd - send cqp command manage apbvt * @iwdev: irdma device * @accel_local_port: port for apbvt * @add_port: add ordelete port */ static int irdma_cqp_manage_apbvt_cmd(struct irdma_device *iwdev, u16 accel_local_port, bool add_port) { struct irdma_apbvt_info *info; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; int status; cqp_request = irdma_alloc_and_get_cqp_request(&iwdev->rf->cqp, add_port); if (!cqp_request) return -ENOMEM; cqp_info = &cqp_request->info; info = &cqp_info->in.u.manage_apbvt_entry.info; memset(info, 0, sizeof(*info)); info->add = add_port; info->port = accel_local_port; cqp_info->cqp_cmd = IRDMA_OP_MANAGE_APBVT_ENTRY; cqp_info->post_sq = 1; cqp_info->in.u.manage_apbvt_entry.cqp = &iwdev->rf->cqp.sc_cqp; cqp_info->in.u.manage_apbvt_entry.scratch = (uintptr_t)cqp_request; irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_DEV, "%s: port=0x%04x\n", (!add_port) ? "DELETE" : "ADD", accel_local_port); status = irdma_handle_cqp_op(iwdev->rf, cqp_request); irdma_put_cqp_request(&iwdev->rf->cqp, cqp_request); return status; } /** * irdma_add_apbvt - add tcp port to HW apbvt table * @iwdev: irdma device * @port: port for apbvt */ struct irdma_apbvt_entry * irdma_add_apbvt(struct irdma_device *iwdev, u16 port) { struct irdma_cm_core *cm_core = &iwdev->cm_core; struct irdma_apbvt_entry *entry; unsigned long flags; spin_lock_irqsave(&cm_core->apbvt_lock, flags); entry = irdma_lookup_apbvt_entry(cm_core, port); if (entry) { spin_unlock_irqrestore(&cm_core->apbvt_lock, flags); return entry; } entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (!entry) { spin_unlock_irqrestore(&cm_core->apbvt_lock, flags); return NULL; } entry->port = port; entry->use_cnt = 1; HASH_ADD(cm_core->apbvt_hash_tbl, &entry->hlist, entry->port); spin_unlock_irqrestore(&cm_core->apbvt_lock, flags); if (irdma_cqp_manage_apbvt_cmd(iwdev, port, true)) { kfree(entry); return NULL; } return entry; } /** * irdma_del_apbvt - delete tcp port from HW apbvt table * @iwdev: irdma device * @entry: apbvt entry object */ void irdma_del_apbvt(struct irdma_device *iwdev, struct irdma_apbvt_entry *entry) { struct irdma_cm_core *cm_core = &iwdev->cm_core; unsigned long flags; spin_lock_irqsave(&cm_core->apbvt_lock, flags); if (--entry->use_cnt) { spin_unlock_irqrestore(&cm_core->apbvt_lock, flags); return; } HASH_DEL(cm_core->apbvt_hash_tbl, &entry->hlist); /* * apbvt_lock is held across CQP delete APBVT OP (non-waiting) to protect against race where add APBVT CQP can * race ahead of the delete APBVT for same port. */ irdma_cqp_manage_apbvt_cmd(iwdev, entry->port, false); kfree(entry); spin_unlock_irqrestore(&cm_core->apbvt_lock, flags); } /** * irdma_manage_arp_cache - manage hw arp cache * @rf: RDMA PCI function * @mac_addr: mac address ptr * @ip_addr: ip addr for arp cache * @action: add, delete or modify */ void irdma_manage_arp_cache(struct irdma_pci_f *rf, const unsigned char *mac_addr, u32 *ip_addr, u32 action) { struct irdma_add_arp_cache_entry_info *info; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; int arp_index; arp_index = irdma_arp_table(rf, ip_addr, mac_addr, action); if (arp_index == -1) return; cqp_request = irdma_alloc_and_get_cqp_request(&rf->cqp, false); if (!cqp_request) return; cqp_info = &cqp_request->info; if (action == IRDMA_ARP_ADD) { cqp_info->cqp_cmd = IRDMA_OP_ADD_ARP_CACHE_ENTRY; info = &cqp_info->in.u.add_arp_cache_entry.info; memset(info, 0, sizeof(*info)); info->arp_index = (u16)arp_index; info->permanent = true; ether_addr_copy(info->mac_addr, mac_addr); cqp_info->in.u.add_arp_cache_entry.scratch = (uintptr_t)cqp_request; cqp_info->in.u.add_arp_cache_entry.cqp = &rf->cqp.sc_cqp; } else { cqp_info->cqp_cmd = IRDMA_OP_DELETE_ARP_CACHE_ENTRY; cqp_info->in.u.del_arp_cache_entry.scratch = (uintptr_t)cqp_request; cqp_info->in.u.del_arp_cache_entry.cqp = &rf->cqp.sc_cqp; cqp_info->in.u.del_arp_cache_entry.arp_index = arp_index; } cqp_info->post_sq = 1; irdma_handle_cqp_op(rf, cqp_request); irdma_put_cqp_request(&rf->cqp, cqp_request); } /** * irdma_send_syn_cqp_callback - do syn/ack after qhash * @cqp_request: qhash cqp completion */ static void irdma_send_syn_cqp_callback(struct irdma_cqp_request *cqp_request) { struct irdma_cm_node *cm_node = cqp_request->param; irdma_send_syn(cm_node, 1); irdma_rem_ref_cm_node(cm_node); } /** - * irdma_manage_qhash - add or modify qhash + * irdma_qhash_info_prepare - fill info for qhash op * @iwdev: irdma device + * @cqp_info: cqp info * @cminfo: cm info for qhash * @etype: type (syn or quad) * @mtype: type of qhash - * @cmnode: cmnode associated with connection - * @wait: wait for completion */ -int -irdma_manage_qhash(struct irdma_device *iwdev, struct irdma_cm_info *cminfo, - enum irdma_quad_entry_type etype, - enum irdma_quad_hash_manage_type mtype, void *cmnode, - bool wait) +static void +irdma_qhash_info_prepare(struct irdma_device *iwdev, + struct cqp_cmds_info *cqp_info, + struct irdma_cm_info *cminfo, + enum irdma_quad_entry_type etype, + enum irdma_quad_hash_manage_type mtype) { struct irdma_qhash_table_info *info; - struct irdma_cqp *iwcqp = &iwdev->rf->cqp; - struct irdma_cqp_request *cqp_request; - struct cqp_cmds_info *cqp_info; - struct irdma_cm_node *cm_node = cmnode; - int status; - - cqp_request = irdma_alloc_and_get_cqp_request(iwcqp, wait); - if (!cqp_request) - return -ENOMEM; - cminfo->cqp_request = cqp_request; - if (!wait) - atomic_inc(&cqp_request->refcnt); - cqp_info = &cqp_request->info; info = &cqp_info->in.u.manage_qhash_table_entry.info; memset(info, 0, sizeof(*info)); info->vsi = &iwdev->vsi; info->manage = mtype; info->entry_type = etype; if (cminfo->vlan_id < VLAN_N_VID) { info->vlan_valid = true; info->vlan_id = cminfo->vlan_id; } else { info->vlan_valid = false; } info->ipv4_valid = cminfo->ipv4; info->user_pri = cminfo->user_pri; ether_addr_copy(info->mac_addr, if_getlladdr(iwdev->netdev)); info->qp_num = cminfo->qh_qpid; info->dest_port = cminfo->loc_port; info->dest_ip[0] = cminfo->loc_addr[0]; info->dest_ip[1] = cminfo->loc_addr[1]; info->dest_ip[2] = cminfo->loc_addr[2]; info->dest_ip[3] = cminfo->loc_addr[3]; if (etype == IRDMA_QHASH_TYPE_TCP_ESTABLISHED || etype == IRDMA_QHASH_TYPE_UDP_UNICAST || etype == IRDMA_QHASH_TYPE_UDP_MCAST || etype == IRDMA_QHASH_TYPE_ROCE_MCAST || etype == IRDMA_QHASH_TYPE_ROCEV2_HW) { info->src_port = cminfo->rem_port; info->src_ip[0] = cminfo->rem_addr[0]; info->src_ip[1] = cminfo->rem_addr[1]; info->src_ip[2] = cminfo->rem_addr[2]; info->src_ip[3] = cminfo->rem_addr[3]; } + cqp_info->cqp_cmd = IRDMA_OP_MANAGE_QHASH_TABLE_ENTRY; + cqp_info->post_sq = 1; +} + +/** + * irdma_add_qhash_wait_no_lock - add qhash, blocking w/o lock + * @iwdev: irdma device + * @cminfo: cm info for qhash + */ +int +irdma_add_qhash_wait_no_lock(struct irdma_device *iwdev, + struct irdma_cm_info *cminfo) +{ + struct irdma_qhash_table_info *info; + struct irdma_cqp *iwcqp = &iwdev->rf->cqp; + struct irdma_cqp_request *cqp_request; + struct cqp_cmds_info *cqp_info; + int cnt = iwdev->rf->sc_dev.hw_attrs.max_cqp_compl_wait_time_ms * CQP_TIMEOUT_THRESHOLD; + int status; + int ret_val; + + cqp_request = irdma_alloc_and_get_cqp_request(iwcqp, false); + if (!cqp_request) + return -ENOMEM; + + cqp_info = &cqp_request->info; + info = &cqp_info->in.u.manage_qhash_table_entry.info; + irdma_qhash_info_prepare(iwdev, cqp_info, cminfo, IRDMA_QHASH_TYPE_TCP_SYN, + IRDMA_QHASH_MANAGE_TYPE_ADD); + if (info->ipv4_valid) + irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, + "ADD caller: %pS loc_port=0x%04x rem_port=0x%04x loc_addr=%x rem_addr=%x mac=%x:%x:%x:%x:%x:%x, vlan_id=%d\n", + __builtin_return_address(0), info->src_port, + info->dest_port, info->src_ip[0], info->dest_ip[0], + info->mac_addr[0], info->mac_addr[1], + info->mac_addr[2], info->mac_addr[3], + info->mac_addr[4], info->mac_addr[5], + cminfo->vlan_id); + else + irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, + "ADD caller: %pS loc_port=0x%04x rem_port=0x%04x loc_addr=%x:%x:%x:%x rem_addr=%x:%x:%x:%x mac=%x:%x:%x:%x:%x:%x, vlan_id=%d\n", + __builtin_return_address(0), info->src_port, + info->dest_port, IRDMA_PRINT_IP6(info->src_ip), + IRDMA_PRINT_IP6(info->dest_ip), info->mac_addr[0], + info->mac_addr[1], info->mac_addr[2], + info->mac_addr[3], info->mac_addr[4], + info->mac_addr[5], cminfo->vlan_id); + + cqp_info->in.u.manage_qhash_table_entry.cqp = &iwdev->rf->cqp.sc_cqp; + cqp_info->in.u.manage_qhash_table_entry.scratch = (uintptr_t)cqp_request; + status = irdma_handle_cqp_op(iwdev->rf, cqp_request); + if (status) { + irdma_put_cqp_request(iwcqp, cqp_request); + irdma_dev_warn(&iwdev->ibdev, "manage_qhash cqp op failure %d\n", status); + return status; + } + + do { + irdma_cqp_ce_handler(iwdev->rf, &iwdev->rf->ccq.sc_cq); + mdelay(1); + } while (!READ_ONCE(cqp_request->request_done) && --cnt); + + ret_val = cqp_request->compl_info.op_ret_val; + status = (cnt) ? ret_val : -ETIMEDOUT; + + irdma_put_cqp_request(iwcqp, cqp_request); + + return status; +} + +/** + * irdma_manage_qhash - add or modify qhash + * @iwdev: irdma device + * @cminfo: cm info for qhash + * @etype: type (syn or quad) + * @mtype: type of qhash + * @cmnode: cmnode associated with connection + * @wait: wait for completion + */ +int +irdma_manage_qhash(struct irdma_device *iwdev, struct irdma_cm_info *cminfo, + enum irdma_quad_entry_type etype, + enum irdma_quad_hash_manage_type mtype, void *cmnode, + bool wait) +{ + struct irdma_qhash_table_info *info; + struct irdma_cqp *iwcqp = &iwdev->rf->cqp; + struct irdma_cqp_request *cqp_request; + struct cqp_cmds_info *cqp_info; + struct irdma_cm_node *cm_node = cmnode; + int status; + + cqp_request = irdma_alloc_and_get_cqp_request(iwcqp, wait); + if (!cqp_request) + return -ENOMEM; + + cqp_info = &cqp_request->info; + info = &cqp_info->in.u.manage_qhash_table_entry.info; + irdma_qhash_info_prepare(iwdev, cqp_info, cminfo, etype, mtype); if (cmnode) { cqp_request->callback_fcn = irdma_send_syn_cqp_callback; cqp_request->param = cmnode; if (!wait) atomic_inc(&cm_node->refcnt); } if (info->ipv4_valid) irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "%s caller: %pS loc_port=0x%04x rem_port=0x%04x loc_addr=%x rem_addr=%x mac=%x:%x:%x:%x:%x:%x, vlan_id=%d cm_node=%p\n", (!mtype) ? "DELETE" : "ADD", __builtin_return_address(0), info->src_port, info->dest_port, info->src_ip[0], info->dest_ip[0], info->mac_addr[0], info->mac_addr[1], info->mac_addr[2], info->mac_addr[3], info->mac_addr[4], info->mac_addr[5], cminfo->vlan_id, cmnode ? cmnode : NULL); else irdma_debug(&iwdev->rf->sc_dev, IRDMA_DEBUG_CM, "%s caller: %pS loc_port=0x%04x rem_port=0x%04x loc_addr=%x:%x:%x:%x rem_addr=%x:%x:%x:%x mac=%x:%x:%x:%x:%x:%x, vlan_id=%d cm_node=%p\n", (!mtype) ? "DELETE" : "ADD", __builtin_return_address(0), info->src_port, info->dest_port, IRDMA_PRINT_IP6(info->src_ip), IRDMA_PRINT_IP6(info->dest_ip), info->mac_addr[0], info->mac_addr[1], info->mac_addr[2], info->mac_addr[3], info->mac_addr[4], info->mac_addr[5], cminfo->vlan_id, cmnode ? cmnode : NULL); cqp_info->in.u.manage_qhash_table_entry.cqp = &iwdev->rf->cqp.sc_cqp; cqp_info->in.u.manage_qhash_table_entry.scratch = (uintptr_t)cqp_request; - cqp_info->cqp_cmd = IRDMA_OP_MANAGE_QHASH_TABLE_ENTRY; - cqp_info->post_sq = 1; status = irdma_handle_cqp_op(iwdev->rf, cqp_request); if (status && cm_node && !wait) irdma_rem_ref_cm_node(cm_node); irdma_put_cqp_request(iwcqp, cqp_request); return status; } /** * irdma_hw_flush_wqes - flush qp's wqe * @rf: RDMA PCI function * @qp: hardware control qp * @info: info for flush * @wait: flag wait for completion */ int irdma_hw_flush_wqes(struct irdma_pci_f *rf, struct irdma_sc_qp *qp, struct irdma_qp_flush_info *info, bool wait) { int status; struct irdma_qp_flush_info *hw_info; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; struct irdma_qp *iwqp = qp->qp_uk.back_qp; cqp_request = irdma_alloc_and_get_cqp_request(&rf->cqp, wait); if (!cqp_request) return -ENOMEM; cqp_info = &cqp_request->info; hw_info = &cqp_request->info.in.u.qp_flush_wqes.info; memcpy(hw_info, info, sizeof(*hw_info)); cqp_info->cqp_cmd = IRDMA_OP_QP_FLUSH_WQES; cqp_info->post_sq = 1; cqp_info->in.u.qp_flush_wqes.qp = qp; cqp_info->in.u.qp_flush_wqes.scratch = (uintptr_t)cqp_request; status = irdma_handle_cqp_op(rf, cqp_request); if (status) { qp->qp_uk.sq_flush_complete = true; qp->qp_uk.rq_flush_complete = true; irdma_put_cqp_request(&rf->cqp, cqp_request); return status; } if (!wait || cqp_request->compl_info.maj_err_code) goto put_cqp; if (info->rq) { if (cqp_request->compl_info.min_err_code == IRDMA_CQP_COMPL_SQ_WQE_FLUSHED || cqp_request->compl_info.min_err_code == 0) { /* RQ WQE flush was requested but did not happen */ qp->qp_uk.rq_flush_complete = true; } } if (info->sq) { if (cqp_request->compl_info.min_err_code == IRDMA_CQP_COMPL_RQ_WQE_FLUSHED || cqp_request->compl_info.min_err_code == 0) { /* SQ WQE flush was requested but did not happen */ qp->qp_uk.sq_flush_complete = true; } } irdma_debug(&rf->sc_dev, IRDMA_DEBUG_VERBS, "qp_id=%d qp_type=%d qpstate=%d ibqpstate=%d last_aeq=%d hw_iw_state=%d maj_err_code=%d min_err_code=%d\n", iwqp->ibqp.qp_num, rf->protocol_used, iwqp->iwarp_state, iwqp->ibqp_state, iwqp->last_aeq, iwqp->hw_iwarp_state, cqp_request->compl_info.maj_err_code, cqp_request->compl_info.min_err_code); put_cqp: irdma_put_cqp_request(&rf->cqp, cqp_request); return status; } /** * irdma_gen_ae - generate AE * @rf: RDMA PCI function * @qp: qp associated with AE * @info: info for ae * @wait: wait for completion */ void irdma_gen_ae(struct irdma_pci_f *rf, struct irdma_sc_qp *qp, struct irdma_gen_ae_info *info, bool wait) { struct irdma_gen_ae_info *ae_info; struct irdma_cqp_request *cqp_request; struct cqp_cmds_info *cqp_info; cqp_request = irdma_alloc_and_get_cqp_request(&rf->cqp, wait); if (!cqp_request) return; cqp_info = &cqp_request->info; ae_info = &cqp_request->info.in.u.gen_ae.info; memcpy(ae_info, info, sizeof(*ae_info)); cqp_info->cqp_cmd = IRDMA_OP_GEN_AE; cqp_info->post_sq = 1; cqp_info->in.u.gen_ae.qp = qp; cqp_info->in.u.gen_ae.scratch = (uintptr_t)cqp_request; irdma_handle_cqp_op(rf, cqp_request); irdma_put_cqp_request(&rf->cqp, cqp_request); } void irdma_flush_wqes(struct irdma_qp *iwqp, u32 flush_mask) { struct irdma_qp_flush_info info = {0}; struct irdma_pci_f *rf = iwqp->iwdev->rf; u8 flush_code = iwqp->sc_qp.flush_code; if (!(flush_mask & IRDMA_FLUSH_SQ) && !(flush_mask & IRDMA_FLUSH_RQ)) return; /* Set flush info fields */ info.sq = flush_mask & IRDMA_FLUSH_SQ; info.rq = flush_mask & IRDMA_FLUSH_RQ; /* Generate userflush errors in CQE */ info.sq_major_code = IRDMA_FLUSH_MAJOR_ERR; info.sq_minor_code = FLUSH_GENERAL_ERR; info.rq_major_code = IRDMA_FLUSH_MAJOR_ERR; info.rq_minor_code = FLUSH_GENERAL_ERR; info.userflushcode = true; if (flush_mask & IRDMA_REFLUSH) { if (info.sq) iwqp->sc_qp.flush_sq = false; if (info.rq) iwqp->sc_qp.flush_rq = false; } else { if (flush_code) { if (info.sq && iwqp->sc_qp.sq_flush_code) info.sq_minor_code = flush_code; if (info.rq && iwqp->sc_qp.rq_flush_code) info.rq_minor_code = flush_code; } if (irdma_upload_context && irdma_upload_qp_context(iwqp, 0, 1)) irdma_dev_warn(&iwqp->iwdev->ibdev, "failed to upload QP context\n"); if (!iwqp->user_mode) irdma_sched_qp_flush_work(iwqp); } /* Issue flush */ (void)irdma_hw_flush_wqes(rf, &iwqp->sc_qp, &info, flush_mask & IRDMA_FLUSH_WAIT); iwqp->flush_issued = true; } diff --git a/sys/dev/irdma/irdma_main.h b/sys/dev/irdma/irdma_main.h index 5b292ceafea8..9181f3b70463 100644 --- a/sys/dev/irdma/irdma_main.h +++ b/sys/dev/irdma/irdma_main.h @@ -1,610 +1,611 @@ /*- * SPDX-License-Identifier: GPL-2.0 or Linux-OpenIB * - * Copyright (c) 2015 - 2023 Intel Corporation + * Copyright (c) 2015 - 2025 Intel Corporation * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU * General Public License (GPL) Version 2, available from the file * COPYING in the main directory of this source tree, or the * OpenFabrics.org BSD license below: * * 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. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #ifndef IRDMA_MAIN_H #define IRDMA_MAIN_H #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "osdep.h" #include "irdma_defs.h" #include "irdma_hmc.h" #include "irdma_type.h" #include "irdma_ws.h" #include "irdma_protos.h" #include "irdma_pble.h" #include "irdma_cm.h" #include "fbsd_kcompat.h" #include "irdma-abi.h" #include "irdma_verbs.h" #include "irdma_user.h" #include "irdma_puda.h" extern struct list_head irdma_handlers; extern spinlock_t irdma_handler_lock; extern bool irdma_upload_context; #define IRDMA_FW_VER_DEFAULT 2 #define IRDMA_HW_VER 2 #define IRDMA_ARP_ADD 1 #define IRDMA_ARP_DELETE 2 #define IRDMA_ARP_RESOLVE 3 #define IRDMA_MACIP_ADD 1 #define IRDMA_MACIP_DELETE 2 #define IW_CCQ_SIZE (IRDMA_CQP_SW_SQSIZE_2048 + 1) #define IW_CEQ_SIZE 2048 #define IW_AEQ_SIZE 2048 #define RX_BUF_SIZE (1536 + 8) #define IW_REG0_SIZE (4 * 1024) #define IW_TX_TIMEOUT (6 * HZ) #define IW_FIRST_QPN 1 #define IW_SW_CONTEXT_ALIGN 1024 #define MAX_DPC_ITERATIONS 128 #define IRDMA_EVENT_TIMEOUT_MS 5000 #define IRDMA_VCHNL_EVENT_TIMEOUT_MS 10000 #define IRDMA_RST_TIMEOUT_HZ 4 #define IRDMA_NO_QSET 0xffff #define IW_CFG_FPM_QP_COUNT 32768 #define IRDMA_MAX_PAGES_PER_FMR 262144 #define IRDMA_MIN_PAGES_PER_FMR 1 #define IRDMA_CQP_COMPL_RQ_WQE_FLUSHED 2 #define IRDMA_CQP_COMPL_SQ_WQE_FLUSHED 3 #define IRDMA_Q_TYPE_PE_AEQ 0x80 #define IRDMA_Q_INVALID_IDX 0xffff #define IRDMA_REM_ENDPOINT_TRK_QPID 3 #define IRDMA_DRV_OPT_ENA_MPA_VER_0 0x00000001 #define IRDMA_DRV_OPT_DISABLE_MPA_CRC 0x00000002 #define IRDMA_DRV_OPT_DISABLE_FIRST_WRITE 0x00000004 #define IRDMA_DRV_OPT_DISABLE_INTF 0x00000008 #define IRDMA_DRV_OPT_ENA_MSI 0x00000010 #define IRDMA_DRV_OPT_DUAL_LOGICAL_PORT 0x00000020 #define IRDMA_DRV_OPT_NO_INLINE_DATA 0x00000080 #define IRDMA_DRV_OPT_DISABLE_INT_MOD 0x00000100 #define IRDMA_DRV_OPT_DISABLE_VIRT_WQ 0x00000200 #define IRDMA_DRV_OPT_ENA_PAU 0x00000400 #define IRDMA_DRV_OPT_MCAST_LOGPORT_MAP 0x00000800 #define IW_HMC_OBJ_TYPE_NUM ARRAY_SIZE(iw_hmc_obj_types) #define IRDMA_ROCE_CWND_DEFAULT 0x400 #define IRDMA_ROCE_ACKCREDS_DEFAULT 0x1E #define IRDMA_FLUSH_SQ BIT(0) #define IRDMA_FLUSH_RQ BIT(1) #define IRDMA_REFLUSH BIT(2) #define IRDMA_FLUSH_WAIT BIT(3) #define IRDMA_IRQ_NAME_STR_LEN 64 enum init_completion_state { INVALID_STATE = 0, INITIAL_STATE, CQP_CREATED, HMC_OBJS_CREATED, HW_RSRC_INITIALIZED, CCQ_CREATED, CEQ0_CREATED, /* Last state of probe */ ILQ_CREATED, IEQ_CREATED, REM_ENDPOINT_TRK_CREATED, CEQS_CREATED, PBLE_CHUNK_MEM, AEQ_CREATED, IP_ADDR_REGISTERED, /* Last state of open */ }; struct ae_desc { u16 id; const char *desc; }; struct irdma_rsrc_limits { u32 qplimit; u32 mrlimit; u32 cqlimit; }; struct irdma_cqp_err_info { u16 maj; u16 min; const char *desc; }; struct irdma_cqp_compl_info { u32 op_ret_val; u16 maj_err_code; u16 min_err_code; bool error; u8 op_code; }; struct irdma_cqp_request { struct cqp_cmds_info info; wait_queue_head_t waitq; struct list_head list; atomic_t refcnt; void (*callback_fcn)(struct irdma_cqp_request *cqp_request); void *param; struct irdma_cqp_compl_info compl_info; u8 request_done; /* READ/WRITE_ONCE macros operate on it */ bool waiting:1; bool dynamic:1; }; struct irdma_cqp { struct irdma_sc_cqp sc_cqp; spinlock_t req_lock; /* protect CQP request list */ spinlock_t compl_lock; /* protect CQP completion processing */ wait_queue_head_t waitq; wait_queue_head_t remove_wq; struct irdma_dma_mem sq; struct irdma_dma_mem host_ctx; u64 *scratch_array; struct irdma_cqp_request *cqp_requests; struct list_head cqp_avail_reqs; struct list_head cqp_pending_reqs; }; struct irdma_ccq { struct irdma_sc_cq sc_cq; struct irdma_dma_mem mem_cq; struct irdma_dma_mem shadow_area; }; struct irdma_ceq { struct irdma_sc_ceq sc_ceq; struct irdma_dma_mem mem; u32 irq; u32 msix_idx; struct irdma_pci_f *rf; struct tasklet_struct dpc_tasklet; spinlock_t ce_lock; /* sync cq destroy with cq completion event notification */ }; struct irdma_aeq { struct irdma_sc_aeq sc_aeq; struct irdma_dma_mem mem; struct irdma_pble_alloc palloc; bool virtual_map; }; struct irdma_arp_entry { u32 ip_addr[4]; u8 mac_addr[ETHER_ADDR_LEN]; }; struct irdma_msix_vector { u32 idx; u32 irq; u32 cpu_affinity; u16 ceq_id; char name[IRDMA_IRQ_NAME_STR_LEN]; struct resource *res; void *tag; }; struct irdma_mc_table_info { u32 mgn; u32 dest_ip[4]; bool lan_fwd:1; bool ipv4_valid:1; }; struct mc_table_list { struct list_head list; struct irdma_mc_table_info mc_info; struct irdma_mcast_grp_info mc_grp_ctx; }; struct irdma_qv_info { u32 v_idx; /* msix_vector */ u16 ceq_idx; u16 aeq_idx; u8 itr_idx; }; struct irdma_qvlist_info { u32 num_vectors; struct irdma_qv_info qv_info[1]; }; struct irdma_gen_ops { void (*request_reset)(struct irdma_pci_f *rf); int (*register_qset)(struct irdma_sc_vsi *vsi, struct irdma_ws_node *tc_node); void (*unregister_qset)(struct irdma_sc_vsi *vsi, struct irdma_ws_node *tc_node); }; struct irdma_pci_f { bool reset:1; bool rsrc_created:1; bool msix_shared:1; bool ftype:1; u8 rsrc_profile; u8 *hmc_info_mem; u8 *mem_rsrc; u8 rdma_ver; u8 rst_to; /* Not used in SRIOV VF mode */ u8 pf_id; enum irdma_protocol_used protocol_used; bool en_rem_endpoint_trk:1; bool dcqcn_ena:1; u32 sd_type; u32 msix_count; u32 max_mr; u32 max_qp; u32 max_cq; u32 max_ah; u32 next_ah; u32 max_mcg; u32 next_mcg; u32 max_pd; u32 next_qp; u32 next_cq; u32 next_pd; u32 max_mr_size; u32 max_cqe; u32 mr_stagmask; u32 used_pds; u32 used_cqs; u32 used_mrs; u32 used_qps; u32 arp_table_size; u32 next_arp_index; u32 ceqs_count; u32 next_ws_node_id; u32 max_ws_node_id; u32 limits_sel; unsigned long *allocated_ws_nodes; unsigned long *allocated_qps; unsigned long *allocated_cqs; unsigned long *allocated_mrs; unsigned long *allocated_pds; unsigned long *allocated_mcgs; unsigned long *allocated_ahs; unsigned long *allocated_arps; enum init_completion_state init_state; struct irdma_sc_dev sc_dev; struct irdma_dev_ctx dev_ctx; struct irdma_tunable_info tun_info; eventhandler_tag irdma_ifaddr_event; struct irdma_handler *hdl; struct pci_dev *pcidev; struct ice_rdma_peer *peer_info; struct irdma_hw hw; struct irdma_cqp cqp; struct irdma_ccq ccq; struct irdma_aeq aeq; struct irdma_ceq *ceqlist; struct irdma_hmc_pble_rsrc *pble_rsrc; struct irdma_arp_entry *arp_table; spinlock_t arp_lock; /*protect ARP table access*/ spinlock_t rsrc_lock; /* protect HW resource array access */ spinlock_t qptable_lock; /*protect QP table access*/ spinlock_t cqtable_lock; /*protect CQ table access*/ struct irdma_qp **qp_table; struct irdma_cq **cq_table; spinlock_t qh_list_lock; /* protect mc_qht_list */ struct mc_table_list mc_qht_list; struct irdma_msix_vector *iw_msixtbl; struct irdma_qvlist_info *iw_qvlist; struct tasklet_struct dpc_tasklet; struct msix_entry msix_info; struct irdma_dma_mem obj_mem; struct irdma_dma_mem obj_next; atomic_t vchnl_msgs; wait_queue_head_t vchnl_waitq; struct workqueue_struct *cqp_cmpl_wq; struct work_struct cqp_cmpl_work; struct irdma_sc_vsi default_vsi; void *back_fcn; struct irdma_gen_ops gen_ops; void (*check_fc)(struct irdma_sc_vsi *vsi, struct irdma_sc_qp *sc_qp); struct irdma_dcqcn_cc_params dcqcn_params; struct irdma_device *iwdev; }; struct irdma_device { struct ib_device ibdev; struct irdma_pci_f *rf; if_t netdev; struct notifier_block nb_netdevice_event; struct irdma_handler *hdl; struct workqueue_struct *cleanup_wq; struct irdma_sc_vsi vsi; struct irdma_cm_core cm_core; u32 roce_cwnd; u32 roce_ackcreds; u32 vendor_id; u32 vendor_part_id; u32 rcv_wnd; u16 mac_ip_table_idx; u16 vsi_num; u8 rcv_wscale; u8 iw_status; u8 roce_rtomin; u8 rd_fence_rate; bool override_rcv_wnd:1; bool override_cwnd:1; bool override_ackcreds:1; bool override_ooo:1; bool override_rd_fence_rate:1; bool override_rtomin:1; bool push_mode:1; bool roce_mode:1; bool roce_dcqcn_en:1; bool dcb_vlan_mode:1; bool iw_ooo:1; enum init_completion_state init_state; wait_queue_head_t suspend_wq; }; struct irdma_handler { struct list_head list; struct irdma_device *iwdev; struct task deferred_task; struct taskqueue *deferred_tq; bool shared_res_created; }; static inline struct irdma_device *to_iwdev(struct ib_device *ibdev) { return container_of(ibdev, struct irdma_device, ibdev); } static inline struct irdma_ucontext *to_ucontext(struct ib_ucontext *ibucontext) { return container_of(ibucontext, struct irdma_ucontext, ibucontext); } static inline struct irdma_user_mmap_entry * to_irdma_mmap_entry(struct rdma_user_mmap_entry *rdma_entry) { return container_of(rdma_entry, struct irdma_user_mmap_entry, rdma_entry); } static inline struct irdma_pd *to_iwpd(struct ib_pd *ibpd) { return container_of(ibpd, struct irdma_pd, ibpd); } static inline struct irdma_ah *to_iwah(struct ib_ah *ibah) { return container_of(ibah, struct irdma_ah, ibah); } static inline struct irdma_mr *to_iwmr(struct ib_mr *ibmr) { return container_of(ibmr, struct irdma_mr, ibmr); } static inline struct irdma_mr *to_iwmw(struct ib_mw *ibmw) { return container_of(ibmw, struct irdma_mr, ibmw); } static inline struct irdma_cq *to_iwcq(struct ib_cq *ibcq) { return container_of(ibcq, struct irdma_cq, ibcq); } static inline struct irdma_qp *to_iwqp(struct ib_qp *ibqp) { return container_of(ibqp, struct irdma_qp, ibqp); } static inline struct irdma_pci_f *dev_to_rf(struct irdma_sc_dev *dev) { return container_of(dev, struct irdma_pci_f, sc_dev); } /** * irdma_alloc_resource - allocate a resource * @iwdev: device pointer * @resource_array: resource bit array: * @max_resources: maximum resource number * @req_resources_num: Allocated resource number * @next: next free id **/ static inline int irdma_alloc_rsrc(struct irdma_pci_f *rf, unsigned long *rsrc_array, u32 max_rsrc, u32 *req_rsrc_num, u32 *next) { u32 rsrc_num; unsigned long flags; spin_lock_irqsave(&rf->rsrc_lock, flags); rsrc_num = find_next_zero_bit(rsrc_array, max_rsrc, *next); if (rsrc_num >= max_rsrc) { rsrc_num = find_first_zero_bit(rsrc_array, max_rsrc); if (rsrc_num >= max_rsrc) { spin_unlock_irqrestore(&rf->rsrc_lock, flags); irdma_debug(&rf->sc_dev, IRDMA_DEBUG_ERR, "resource [%d] allocation failed\n", rsrc_num); return -EOVERFLOW; } } __set_bit(rsrc_num, rsrc_array); *next = rsrc_num + 1; if (*next == max_rsrc) *next = 0; *req_rsrc_num = rsrc_num; spin_unlock_irqrestore(&rf->rsrc_lock, flags); return 0; } /** * irdma_free_resource - free a resource * @iwdev: device pointer * @resource_array: resource array for the resource_num * @resource_num: resource number to free **/ static inline void irdma_free_rsrc(struct irdma_pci_f *rf, unsigned long *rsrc_array, u32 rsrc_num) { unsigned long flags; spin_lock_irqsave(&rf->rsrc_lock, flags); __clear_bit(rsrc_num, rsrc_array); spin_unlock_irqrestore(&rf->rsrc_lock, flags); } int irdma_ctrl_init_hw(struct irdma_pci_f *rf); void irdma_ctrl_deinit_hw(struct irdma_pci_f *rf); int irdma_rt_init_hw(struct irdma_device *iwdev, struct irdma_l2params *l2params); void irdma_rt_deinit_hw(struct irdma_device *iwdev); void irdma_qp_add_ref(struct ib_qp *ibqp); void irdma_qp_rem_ref(struct ib_qp *ibqp); void irdma_free_lsmm_rsrc(struct irdma_qp *iwqp); struct ib_qp *irdma_get_qp(struct ib_device *ibdev, int qpn); void irdma_flush_wqes(struct irdma_qp *iwqp, u32 flush_mask); void irdma_manage_arp_cache(struct irdma_pci_f *rf, const unsigned char *mac_addr, u32 *ip_addr, u32 action); struct irdma_apbvt_entry *irdma_add_apbvt(struct irdma_device *iwdev, u16 port); void irdma_del_apbvt(struct irdma_device *iwdev, struct irdma_apbvt_entry *entry); struct irdma_cqp_request *irdma_alloc_and_get_cqp_request(struct irdma_cqp *cqp, bool wait); void irdma_free_cqp_request(struct irdma_cqp *cqp, struct irdma_cqp_request *cqp_request); void irdma_put_cqp_request(struct irdma_cqp *cqp, struct irdma_cqp_request *cqp_request); int irdma_alloc_local_mac_entry(struct irdma_pci_f *rf, u16 *mac_tbl_idx); int irdma_add_local_mac_entry(struct irdma_pci_f *rf, const u8 *mac_addr, u16 idx); void irdma_del_local_mac_entry(struct irdma_pci_f *rf, u16 idx); const char *irdma_get_ae_desc(u16 ae_id); u32 irdma_initialize_hw_rsrc(struct irdma_pci_f *rf); void irdma_port_ibevent(struct irdma_device *iwdev); void irdma_cm_disconn(struct irdma_qp *qp); bool irdma_cqp_crit_err(struct irdma_sc_dev *dev, u8 cqp_cmd, u16 maj_err_code, u16 min_err_code); int irdma_handle_cqp_op(struct irdma_pci_f *rf, struct irdma_cqp_request *cqp_request); int irdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, struct ib_udata *udata); int irdma_modify_qp_roce(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask, struct ib_udata *udata); void irdma_cq_add_ref(struct ib_cq *ibcq); void irdma_cq_rem_ref(struct ib_cq *ibcq); void irdma_cq_wq_destroy(struct irdma_pci_f *rf, struct irdma_sc_cq *cq); void irdma_cleanup_pending_cqp_op(struct irdma_pci_f *rf); int irdma_hw_modify_qp(struct irdma_device *iwdev, struct irdma_qp *iwqp, struct irdma_modify_qp_info *info, bool wait); int irdma_qp_suspend_resume(struct irdma_sc_qp *qp, bool suspend); int irdma_manage_qhash(struct irdma_device *iwdev, struct irdma_cm_info *cminfo, enum irdma_quad_entry_type etype, enum irdma_quad_hash_manage_type mtype, void *cmnode, bool wait); +int irdma_add_qhash_wait_no_lock(struct irdma_device *iwdev, struct irdma_cm_info *cminfo); void irdma_receive_ilq(struct irdma_sc_vsi *vsi, struct irdma_puda_buf *rbuf); void irdma_free_sqbuf(struct irdma_sc_vsi *vsi, void *bufp); void irdma_free_qp_rsrc(struct irdma_qp *iwqp); int irdma_setup_cm_core(struct irdma_device *iwdev, u8 ver); void irdma_cleanup_cm_core(struct irdma_cm_core *cm_core); void irdma_next_iw_state(struct irdma_qp *iwqp, u8 state, u8 del_hash, u8 term, u8 term_len); int irdma_send_syn(struct irdma_cm_node *cm_node, u32 sendack); int irdma_send_reset(struct irdma_cm_node *cm_node); struct irdma_cm_node *irdma_find_node(struct irdma_cm_core *cm_core, u16 rem_port, u32 *rem_addr, u16 loc_port, u32 *loc_addr, u16 vlan_id); int irdma_hw_flush_wqes(struct irdma_pci_f *rf, struct irdma_sc_qp *qp, struct irdma_qp_flush_info *info, bool wait); void irdma_gen_ae(struct irdma_pci_f *rf, struct irdma_sc_qp *qp, struct irdma_gen_ae_info *info, bool wait); void irdma_copy_ip_ntohl(u32 *dst, __be32 *src); void irdma_copy_ip_htonl(__be32 *dst, u32 *src); u16 irdma_get_vlan_ipv4(struct iw_cm_id *cm_id, u32 *addr); void irdma_get_vlan_mac_ipv6(struct iw_cm_id *cm_id, u32 *addr, u16 *vlan_id, u8 *mac); struct ib_mr *irdma_reg_phys_mr(struct ib_pd *ib_pd, u64 addr, u64 size, int acc, u64 *iova_start); int irdma_upload_qp_context(struct irdma_qp *iwqp, bool freeze, bool raw); void irdma_del_hmc_objects(struct irdma_sc_dev *dev, struct irdma_hmc_info *hmc_info, bool privileged, bool reset, enum irdma_vers vers); void irdma_cqp_ce_handler(struct irdma_pci_f *rf, struct irdma_sc_cq *cq); int irdma_ah_cqp_op(struct irdma_pci_f *rf, struct irdma_sc_ah *sc_ah, u8 cmd, bool wait, void (*callback_fcn)(struct irdma_cqp_request *cqp_request), void *cb_param); void irdma_udqp_qs_worker(struct work_struct *work); bool irdma_cq_empty(struct irdma_cq *iwcq); int irdma_netdevice_event(struct notifier_block *notifier, unsigned long event, void *ptr); void irdma_unregister_notifiers(struct irdma_device *iwdev); int irdma_register_notifiers(struct irdma_device *iwdev); void irdma_set_rf_user_cfg_params(struct irdma_pci_f *rf); void irdma_add_ip(struct irdma_device *iwdev); void irdma_add_handler(struct irdma_handler *hdl); void irdma_del_handler(struct irdma_handler *hdl); void cqp_compl_worker(struct work_struct *work); void irdma_cleanup_dead_qps(struct irdma_sc_vsi *vsi); #endif /* IRDMA_MAIN_H */