diff options
Diffstat (limited to 'drivers/net/ethernet/intel/ice/ice_main.c')
| -rw-r--r-- | drivers/net/ethernet/intel/ice/ice_main.c | 6055 |
1 files changed, 4296 insertions, 1759 deletions
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index ef8d1815af56..4bb68e7a00f5 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -1,24 +1,32 @@ // SPDX-License-Identifier: GPL-2.0 -/* Copyright (c) 2018, Intel Corporation. */ +/* Copyright (c) 2018-2023, Intel Corporation. */ /* Intel(R) Ethernet Connection E800 Series Linux Driver */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <generated/utsrelease.h> +#include <linux/crash_dump.h> #include "ice.h" #include "ice_base.h" #include "ice_lib.h" #include "ice_fltr.h" #include "ice_dcb_lib.h" #include "ice_dcb_nl.h" -#include "ice_devlink.h" +#include "devlink/devlink.h" +#include "devlink/port.h" +#include "ice_sf_eth.h" +#include "ice_hwmon.h" /* Including ice_trace.h with CREATE_TRACE_POINTS defined will generate the * ice tracepoint functions. This must be done exactly once across the * ice driver. */ #define CREATE_TRACE_POINTS #include "ice_trace.h" +#include "ice_eswitch.h" +#include "ice_tc_lib.h" +#include "ice_vsi_vlan_ops.h" +#include <net/xdp_sock_drv.h> #define DRV_SUMMARY "Intel(R) Ethernet Connection E800 Series Linux Driver" static const char ice_driver_string[] = DRV_SUMMARY; @@ -28,8 +36,12 @@ static const char ice_copyright[] = "Copyright (c) 2018, Intel Corporation."; #define ICE_DDP_PKG_PATH "intel/ice/ddp/" #define ICE_DDP_PKG_FILE ICE_DDP_PKG_PATH "ice.pkg" -MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>"); MODULE_DESCRIPTION(DRV_SUMMARY); +MODULE_IMPORT_NS("LIBETH"); +MODULE_IMPORT_NS("LIBETH_XDP"); +MODULE_IMPORT_NS("LIBIE"); +MODULE_IMPORT_NS("LIBIE_ADMINQ"); +MODULE_IMPORT_NS("LIBIE_FWLOG"); MODULE_LICENSE("GPL v2"); MODULE_FIRMWARE(ICE_DDP_PKG_FILE); @@ -41,27 +53,53 @@ MODULE_PARM_DESC(debug, "netif level (0=none,...,16=all), hw debug_mask (0x8XXXX MODULE_PARM_DESC(debug, "netif level (0=none,...,16=all)"); #endif /* !CONFIG_DYNAMIC_DEBUG */ -static DEFINE_IDA(ice_aux_ida); +DEFINE_STATIC_KEY_FALSE(ice_xdp_locking_key); +EXPORT_SYMBOL(ice_xdp_locking_key); + +/** + * ice_hw_to_dev - Get device pointer from the hardware structure + * @hw: pointer to the device HW structure + * + * Used to access the device pointer from compilation units which can't easily + * include the definition of struct ice_pf without leading to circular header + * dependencies. + */ +struct device *ice_hw_to_dev(struct ice_hw *hw) +{ + struct ice_pf *pf = container_of(hw, struct ice_pf, hw); + + return &pf->pdev->dev; +} static struct workqueue_struct *ice_wq; +struct workqueue_struct *ice_lag_wq; static const struct net_device_ops ice_netdev_safe_mode_ops; static const struct net_device_ops ice_netdev_ops; -static int ice_vsi_open(struct ice_vsi *vsi); static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type); static void ice_vsi_release_all(struct ice_pf *pf); -bool netif_is_ice(struct net_device *dev) +static int ice_rebuild_channels(struct ice_pf *pf); +static void ice_remove_q_channels(struct ice_vsi *vsi, bool rem_adv_fltr); + +static int +ice_indr_setup_tc_cb(struct net_device *netdev, struct Qdisc *sch, + void *cb_priv, enum tc_setup_type type, void *type_data, + void *data, + void (*cleanup)(struct flow_block_cb *block_cb)); + +bool netif_is_ice(const struct net_device *dev) { - return dev && (dev->netdev_ops == &ice_netdev_ops); + return dev && (dev->netdev_ops == &ice_netdev_ops || + dev->netdev_ops == &ice_netdev_safe_mode_ops); } /** * ice_get_tx_pending - returns number of Tx descriptors not processed * @ring: the ring of descriptors */ -static u16 ice_get_tx_pending(struct ice_ring *ring) +static u16 ice_get_tx_pending(struct ice_tx_ring *ring) { u16 head, tail; @@ -100,10 +138,20 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) hw = &vsi->back->hw; - for (i = 0; i < vsi->num_txq; i++) { - struct ice_ring *tx_ring = vsi->tx_rings[i]; + ice_for_each_txq(vsi, i) { + struct ice_tx_ring *tx_ring = vsi->tx_rings[i]; + struct ice_ring_stats *ring_stats; - if (tx_ring && tx_ring->desc) { + if (!tx_ring) + continue; + if (ice_ring_ch_enabled(tx_ring)) + continue; + + ring_stats = tx_ring->ring_stats; + if (!ring_stats) + continue; + + if (tx_ring->desc) { /* If packet counter has not changed the queue is * likely stalled, so force an interrupt for this * queue. @@ -111,8 +159,8 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) * prev_pkt would be negative if there was no * pending work. */ - packets = tx_ring->stats.pkts & INT_MAX; - if (tx_ring->tx_stats.prev_pkt == packets) { + packets = ring_stats->stats.pkts & INT_MAX; + if (ring_stats->tx_stats.prev_pkt == packets) { /* Trigger sw interrupt to revive the queue */ ice_trigger_sw_intr(hw, tx_ring->q_vector); continue; @@ -122,7 +170,7 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) * to ice_get_tx_pending() */ smp_rmb(); - tx_ring->tx_stats.prev_pkt = + ring_stats->tx_stats.prev_pkt = ice_get_tx_pending(tx_ring) ? packets : -1; } } @@ -138,7 +186,6 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) */ static int ice_init_mac_fltr(struct ice_pf *pf) { - enum ice_status status; struct ice_vsi *vsi; u8 *perm_addr; @@ -147,11 +194,7 @@ static int ice_init_mac_fltr(struct ice_pf *pf) return -EINVAL; perm_addr = vsi->port_info->mac.perm_addr; - status = ice_fltr_add_mac_and_broadcast(vsi, perm_addr, ICE_FWD_TO_VSI); - if (status) - return -EIO; - - return 0; + return ice_fltr_add_mac_and_broadcast(vsi, perm_addr, ICE_FWD_TO_VSI); } /** @@ -191,6 +234,14 @@ static int ice_add_mac_to_unsync_list(struct net_device *netdev, const u8 *addr) struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; + /* Under some circumstances, we might receive a request to delete our + * own device address from our uc list. Because we store the device + * address in the VSI's MAC filter list, we need to ignore such + * requests and not delete our device address from this list. + */ + if (ether_addr_equal(addr, netdev->dev_addr)) + return 0; + if (ice_fltr_add_mac_to_list(vsi, &vsi->tmp_unsync_list, addr, ICE_FWD_TO_VSI)) return -EINVAL; @@ -207,44 +258,66 @@ static int ice_add_mac_to_unsync_list(struct net_device *netdev, const u8 *addr) static bool ice_vsi_fltr_changed(struct ice_vsi *vsi) { return test_bit(ICE_VSI_UMAC_FLTR_CHANGED, vsi->state) || - test_bit(ICE_VSI_MMAC_FLTR_CHANGED, vsi->state) || - test_bit(ICE_VSI_VLAN_FLTR_CHANGED, vsi->state); + test_bit(ICE_VSI_MMAC_FLTR_CHANGED, vsi->state); } /** - * ice_cfg_promisc - Enable or disable promiscuous mode for a given PF + * ice_set_promisc - Enable promiscuous mode for a given PF * @vsi: the VSI being configured * @promisc_m: mask of promiscuous config bits - * @set_promisc: enable or disable promisc flag request * */ -static int ice_cfg_promisc(struct ice_vsi *vsi, u8 promisc_m, bool set_promisc) +static int ice_set_promisc(struct ice_vsi *vsi, u8 promisc_m) { - struct ice_hw *hw = &vsi->back->hw; - enum ice_status status = 0; + int status; if (vsi->type != ICE_VSI_PF) return 0; - if (vsi->num_vlan > 1) { - status = ice_set_vlan_vsi_promisc(hw, vsi->idx, promisc_m, - set_promisc); + if (ice_vsi_has_non_zero_vlans(vsi)) { + promisc_m |= (ICE_PROMISC_VLAN_RX | ICE_PROMISC_VLAN_TX); + status = ice_fltr_set_vlan_vsi_promisc(&vsi->back->hw, vsi, + promisc_m); } else { - if (set_promisc) - status = ice_set_vsi_promisc(hw, vsi->idx, promisc_m, - 0); - else - status = ice_clear_vsi_promisc(hw, vsi->idx, promisc_m, - 0); + status = ice_fltr_set_vsi_promisc(&vsi->back->hw, vsi->idx, + promisc_m, 0); } + if (status && status != -EEXIST) + return status; - if (status) - return -EIO; - + netdev_dbg(vsi->netdev, "set promisc filter bits for VSI %i: 0x%x\n", + vsi->vsi_num, promisc_m); return 0; } /** + * ice_clear_promisc - Disable promiscuous mode for a given PF + * @vsi: the VSI being configured + * @promisc_m: mask of promiscuous config bits + * + */ +static int ice_clear_promisc(struct ice_vsi *vsi, u8 promisc_m) +{ + int status; + + if (vsi->type != ICE_VSI_PF) + return 0; + + if (ice_vsi_has_non_zero_vlans(vsi)) { + promisc_m |= (ICE_PROMISC_VLAN_RX | ICE_PROMISC_VLAN_TX); + status = ice_fltr_clear_vlan_vsi_promisc(&vsi->back->hw, vsi, + promisc_m); + } else { + status = ice_fltr_clear_vsi_promisc(&vsi->back->hw, vsi->idx, + promisc_m, 0); + } + + netdev_dbg(vsi->netdev, "clear promisc filter bits for VSI %i: 0x%x\n", + vsi->vsi_num, promisc_m); + return status; +} + +/** * ice_vsi_sync_fltr - Update the VSI filter list to the HW * @vsi: ptr to the VSI * @@ -252,15 +325,14 @@ static int ice_cfg_promisc(struct ice_vsi *vsi, u8 promisc_m, bool set_promisc) */ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) { + struct ice_vsi_vlan_ops *vlan_ops = ice_get_compat_vsi_vlan_ops(vsi); struct device *dev = ice_pf_to_dev(vsi->back); struct net_device *netdev = vsi->netdev; bool promisc_forced_on = false; struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; - enum ice_status status = 0; u32 changed_flags = 0; - u8 promisc_m; - int err = 0; + int err; if (!vsi->netdev) return -EINVAL; @@ -277,7 +349,6 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) if (ice_vsi_fltr_changed(vsi)) { clear_bit(ICE_VSI_UMAC_FLTR_CHANGED, vsi->state); clear_bit(ICE_VSI_MMAC_FLTR_CHANGED, vsi->state); - clear_bit(ICE_VSI_VLAN_FLTR_CHANGED, vsi->state); /* grab the netdev's addr_list_lock */ netif_addr_lock_bh(netdev); @@ -290,67 +361,51 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) } /* Remove MAC addresses in the unsync list */ - status = ice_fltr_remove_mac_list(vsi, &vsi->tmp_unsync_list); + err = ice_fltr_remove_mac_list(vsi, &vsi->tmp_unsync_list); ice_fltr_free_list(dev, &vsi->tmp_unsync_list); - if (status) { + if (err) { netdev_err(netdev, "Failed to delete MAC filters\n"); /* if we failed because of alloc failures, just bail */ - if (status == ICE_ERR_NO_MEMORY) { - err = -ENOMEM; + if (err == -ENOMEM) goto out; - } } /* Add MAC addresses in the sync list */ - status = ice_fltr_add_mac_list(vsi, &vsi->tmp_sync_list); + err = ice_fltr_add_mac_list(vsi, &vsi->tmp_sync_list); ice_fltr_free_list(dev, &vsi->tmp_sync_list); /* If filter is added successfully or already exists, do not go into * 'if' condition and report it as error. Instead continue processing * rest of the function. */ - if (status && status != ICE_ERR_ALREADY_EXISTS) { + if (err && err != -EEXIST) { netdev_err(netdev, "Failed to add MAC filters\n"); /* If there is no more space for new umac filters, VSI * should go into promiscuous mode. There should be some * space reserved for promiscuous filters. */ - if (hw->adminq.sq_last_status == ICE_AQ_RC_ENOSPC && + if (hw->adminq.sq_last_status == LIBIE_AQ_RC_ENOSPC && !test_and_set_bit(ICE_FLTR_OVERFLOW_PROMISC, vsi->state)) { promisc_forced_on = true; netdev_warn(netdev, "Reached MAC filter limit, forcing promisc mode on VSI %d\n", vsi->vsi_num); } else { - err = -EIO; goto out; } } + err = 0; /* check for changes in promiscuous modes */ if (changed_flags & IFF_ALLMULTI) { if (vsi->current_netdev_flags & IFF_ALLMULTI) { - if (vsi->num_vlan > 1) - promisc_m = ICE_MCAST_VLAN_PROMISC_BITS; - else - promisc_m = ICE_MCAST_PROMISC_BITS; - - err = ice_cfg_promisc(vsi, promisc_m, true); + err = ice_set_promisc(vsi, ICE_MCAST_PROMISC_BITS); if (err) { - netdev_err(netdev, "Error setting Multicast promiscuous mode on VSI %i\n", - vsi->vsi_num); vsi->current_netdev_flags &= ~IFF_ALLMULTI; goto out_promisc; } } else { /* !(vsi->current_netdev_flags & IFF_ALLMULTI) */ - if (vsi->num_vlan > 1) - promisc_m = ICE_MCAST_VLAN_PROMISC_BITS; - else - promisc_m = ICE_MCAST_PROMISC_BITS; - - err = ice_cfg_promisc(vsi, promisc_m, false); + err = ice_clear_promisc(vsi, ICE_MCAST_PROMISC_BITS); if (err) { - netdev_err(netdev, "Error clearing Multicast promiscuous mode on VSI %i\n", - vsi->vsi_num); vsi->current_netdev_flags |= IFF_ALLMULTI; goto out_promisc; } @@ -362,8 +417,8 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) clear_bit(ICE_VSI_PROMISC_CHANGED, vsi->state); if (vsi->current_netdev_flags & IFF_PROMISC) { /* Apply Rx filter rule to get traffic from wire */ - if (!ice_is_dflt_vsi_in_use(pf->first_sw)) { - err = ice_set_dflt_vsi(pf->first_sw, vsi); + if (!ice_is_dflt_vsi_in_use(vsi->port_info)) { + err = ice_set_dflt_vsi(vsi); if (err && err != -EEXIST) { netdev_err(netdev, "Error %d setting default VSI %i Rx rule\n", err, vsi->vsi_num); @@ -371,12 +426,23 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) ~IFF_PROMISC; goto out_promisc; } - ice_cfg_vlan_pruning(vsi, false, false); + err = 0; + vlan_ops->dis_rx_filtering(vsi); + + /* promiscuous mode implies allmulticast so + * that VSIs that are in promiscuous mode are + * subscribed to multicast packets coming to + * the port + */ + err = ice_set_promisc(vsi, + ICE_MCAST_PROMISC_BITS); + if (err) + goto out_promisc; } } else { /* Clear Rx filter to remove traffic from wire */ - if (ice_is_vsi_dflt_vsi(pf->first_sw, vsi)) { - err = ice_clear_dflt_vsi(pf->first_sw); + if (ice_is_vsi_dflt_vsi(vsi)) { + err = ice_clear_dflt_vsi(vsi); if (err) { netdev_err(netdev, "Error %d clearing default VSI %i Rx rule\n", err, vsi->vsi_num); @@ -384,8 +450,21 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) IFF_PROMISC; goto out_promisc; } - if (vsi->num_vlan > 1) - ice_cfg_vlan_pruning(vsi, true, false); + if (vsi->netdev->features & + NETIF_F_HW_VLAN_CTAG_FILTER) + vlan_ops->ena_rx_filtering(vsi); + } + + /* disable allmulti here, but only if allmulti is not + * still enabled for the netdev + */ + if (!(vsi->current_netdev_flags & IFF_ALLMULTI)) { + err = ice_clear_promisc(vsi, + ICE_MCAST_PROMISC_BITS); + if (err) { + netdev_err(netdev, "Error %d clearing multicast promiscuous on VSI %i\n", + err, vsi->vsi_num); + } } } } @@ -447,21 +526,28 @@ static void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked) } /** - * ice_prepare_for_reset - prep for the core to reset + * ice_prepare_for_reset - prep for reset * @pf: board private structure + * @reset_type: reset type requested * * Inform or close all dependent features in prep for reset. */ static void -ice_prepare_for_reset(struct ice_pf *pf) +ice_prepare_for_reset(struct ice_pf *pf, enum ice_reset_req reset_type) { struct ice_hw *hw = &pf->hw; - unsigned int i; + struct ice_vsi *vsi; + struct ice_vf *vf; + unsigned int bkt; + + dev_dbg(ice_pf_to_dev(pf), "reset_type=%d\n", reset_type); /* already prepared for reset */ if (test_bit(ICE_PREPARED_FOR_RESET, pf->state)) return; + synchronize_irq(pf->oicr_irq.virq); + ice_unplug_aux_dev(pf); /* Notify VFs of impending reset */ @@ -469,21 +555,68 @@ ice_prepare_for_reset(struct ice_pf *pf) ice_vc_notify_reset(pf); /* Disable VFs until reset is completed */ - ice_for_each_vf(pf, i) - ice_set_vf_state_qs_dis(&pf->vf[i]); + mutex_lock(&pf->vfs.table_lock); + ice_for_each_vf(pf, bkt, vf) + ice_set_vf_state_dis(vf); + mutex_unlock(&pf->vfs.table_lock); + + if (ice_is_eswitch_mode_switchdev(pf)) { + rtnl_lock(); + ice_eswitch_br_fdb_flush(pf->eswitch.br_offloads->bridge); + rtnl_unlock(); + } + + /* release ADQ specific HW and SW resources */ + vsi = ice_get_main_vsi(pf); + if (!vsi) + goto skip; + + /* to be on safe side, reset orig_rss_size so that normal flow + * of deciding rss_size can take precedence + */ + vsi->orig_rss_size = 0; + + if (test_bit(ICE_FLAG_TC_MQPRIO, pf->flags)) { + if (reset_type == ICE_RESET_PFR) { + vsi->old_ena_tc = vsi->all_enatc; + vsi->old_numtc = vsi->all_numtc; + } else { + ice_remove_q_channels(vsi, true); + + /* for other reset type, do not support channel rebuild + * hence reset needed info + */ + vsi->old_ena_tc = 0; + vsi->all_enatc = 0; + vsi->old_numtc = 0; + vsi->all_numtc = 0; + vsi->req_txq = 0; + vsi->req_rxq = 0; + clear_bit(ICE_FLAG_TC_MQPRIO, pf->flags); + memset(&vsi->mqprio_qopt, 0, sizeof(vsi->mqprio_qopt)); + } + } + + if (vsi->netdev) + netif_device_detach(vsi->netdev); +skip: /* clear SW filtering DB */ ice_clear_hw_tbls(hw); /* disable the VSIs and their queues that are not already DOWN */ + set_bit(ICE_VSI_REBUILD_PENDING, ice_get_main_vsi(pf)->state); ice_pf_dis_all_vsi(pf, false); if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) - ice_ptp_release(pf); + ice_ptp_prepare_for_reset(pf, reset_type); + + if (ice_is_feature_supported(pf, ICE_F_GNSS)) + ice_gnss_exit(pf); if (hw->port_info) ice_sched_clear_port(hw->port_info); - ice_shutdown_all_ctrlq(hw); + ice_shutdown_all_ctrlq(hw, false); set_bit(ICE_PREPARED_FOR_RESET, pf->state); } @@ -491,8 +624,7 @@ ice_prepare_for_reset(struct ice_pf *pf) /** * ice_do_reset - Initiate one of many types of resets * @pf: board private structure - * @reset_type: reset type requested - * before this function was called. + * @reset_type: reset type requested before this function was called. */ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type) { @@ -501,7 +633,12 @@ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type) dev_dbg(dev, "reset_type 0x%x requested\n", reset_type); - ice_prepare_for_reset(pf); + if (pf->lag && pf->lag->bonded && reset_type == ICE_RESET_PFR) { + dev_dbg(dev, "PFR on a bonded interface, promoting to CORER\n"); + reset_type = ICE_RESET_CORER; + } + + ice_prepare_for_reset(pf, reset_type); /* trigger the reset */ if (ice_reset(hw, reset_type)) { @@ -526,7 +663,7 @@ static void ice_do_reset(struct ice_pf *pf, enum ice_reset_req reset_type) clear_bit(ICE_PREPARED_FOR_RESET, pf->state); clear_bit(ICE_PFR_REQ, pf->state); wake_up(&pf->reset_wait_queue); - ice_reset_all_vfs(pf, true); + ice_reset_all_vfs(pf); } } @@ -559,7 +696,7 @@ static void ice_reset_subtask(struct ice_pf *pf) /* return if no valid reset type requested */ if (reset_type == ICE_RESET_INVAL) return; - ice_prepare_for_reset(pf); + ice_prepare_for_reset(pf, reset_type); /* make sure we are ready to rebuild */ if (ice_check_reset(&pf->hw)) { @@ -577,15 +714,20 @@ static void ice_reset_subtask(struct ice_pf *pf) clear_bit(ICE_CORER_REQ, pf->state); clear_bit(ICE_GLOBR_REQ, pf->state); wake_up(&pf->reset_wait_queue); - ice_reset_all_vfs(pf, true); + ice_reset_all_vfs(pf); } return; } /* No pending resets to finish processing. Check for new resets */ - if (test_bit(ICE_PFR_REQ, pf->state)) + if (test_bit(ICE_PFR_REQ, pf->state)) { reset_type = ICE_RESET_PFR; + if (pf->lag && pf->lag->bonded) { + dev_dbg(ice_pf_to_dev(pf), "PFR on a bonded interface, promoting to CORER\n"); + reset_type = ICE_RESET_CORER; + } + } if (test_bit(ICE_CORER_REQ, pf->state)) reset_type = ICE_RESET_CORER; if (test_bit(ICE_GLOBR_REQ, pf->state)) @@ -616,7 +758,10 @@ static void ice_print_topo_conflict(struct ice_vsi *vsi) netdev_info(vsi->netdev, "Potential misconfiguration of the Ethernet port detected. If it was not intended, please use the Intel (R) Ethernet Port Configuration Tool to address the issue.\n"); break; case ICE_AQ_LINK_TOPO_UNSUPP_MEDIA: - netdev_info(vsi->netdev, "Rx/Tx is disabled on this device because an unsupported module type was detected. Refer to the Intel(R) Ethernet Adapters and Devices User Guide for a list of supported modules.\n"); + if (test_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, vsi->back->flags)) + netdev_warn(vsi->netdev, "An unsupported module type was detected. Refer to the Intel(R) Ethernet Adapters and Devices User Guide for a list of supported modules\n"); + else + netdev_err(vsi->netdev, "Rx/Tx is disabled on this device because an unsupported module type was detected. Refer to the Intel(R) Ethernet Adapters and Devices User Guide for a list of supported modules.\n"); break; default: break; @@ -632,12 +777,12 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) { struct ice_aqc_get_phy_caps_data *caps; const char *an_advertised; - enum ice_status status; const char *fec_req; const char *speed; const char *fec; const char *fc; const char *an; + int status; if (!vsi) return; @@ -653,6 +798,9 @@ void ice_print_link_msg(struct ice_vsi *vsi, bool isup) } switch (vsi->port_info->phy.link_info.link_speed) { + case ICE_AQ_LINK_SPEED_200GB: + speed = "200 G"; + break; case ICE_AQ_LINK_SPEED_100GB: speed = "100 G"; break; @@ -830,7 +978,7 @@ static void ice_set_dflt_mib(struct ice_pf *pf) * Octets 13 - 20 are TSA values - leave as zeros */ buf[5] = 0x64; - len = (typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S; + len = FIELD_GET(ICE_LLDP_TLV_LEN_M, typelen); offset += len + 2; tlv = (struct ice_lldp_org_tlv *) ((char *)tlv + sizeof(tlv->typelen) + len); @@ -864,7 +1012,7 @@ static void ice_set_dflt_mib(struct ice_pf *pf) /* Octet 1 left as all zeros - PFC disabled */ buf[0] = 0x08; - len = (typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S; + len = FIELD_GET(ICE_LLDP_TLV_LEN_M, typelen); offset += len + 2; if (ice_aq_set_lldp_mib(hw, mib_type, (void *)lldpmib, offset, NULL)) @@ -874,6 +1022,29 @@ static void ice_set_dflt_mib(struct ice_pf *pf) } /** + * ice_check_phy_fw_load - check if PHY FW load failed + * @pf: pointer to PF struct + * @link_cfg_err: bitmap from the link info structure + * + * check if external PHY FW load failed and print an error message if it did + */ +static void ice_check_phy_fw_load(struct ice_pf *pf, u8 link_cfg_err) +{ + if (!(link_cfg_err & ICE_AQ_LINK_EXTERNAL_PHY_LOAD_FAILURE)) { + clear_bit(ICE_FLAG_PHY_FW_LOAD_FAILED, pf->flags); + return; + } + + if (test_bit(ICE_FLAG_PHY_FW_LOAD_FAILED, pf->flags)) + return; + + if (link_cfg_err & ICE_AQ_LINK_EXTERNAL_PHY_LOAD_FAILURE) { + dev_err(ice_pf_to_dev(pf), "Device failed to load the FW for the external PHY. Please download and install the latest NVM for your device and try again\n"); + set_bit(ICE_FLAG_PHY_FW_LOAD_FAILED, pf->flags); + } +} + +/** * ice_check_module_power * @pf: pointer to PF struct * @link_cfg_err: bitmap from the link info structure @@ -906,6 +1077,20 @@ static void ice_check_module_power(struct ice_pf *pf, u8 link_cfg_err) } /** + * ice_check_link_cfg_err - check if link configuration failed + * @pf: pointer to the PF struct + * @link_cfg_err: bitmap from the link info structure + * + * print if any link configuration failure happens due to the value in the + * link_cfg_err parameter in the link info structure + */ +static void ice_check_link_cfg_err(struct ice_pf *pf, u8 link_cfg_err) +{ + ice_check_module_power(pf, link_cfg_err); + ice_check_phy_fw_load(pf, link_cfg_err); +} + +/** * ice_link_event - process the link event * @pf: PF that the link event is associated with * @pi: port_info for the port that the link event is associated with @@ -920,10 +1105,10 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, { struct device *dev = ice_pf_to_dev(pf); struct ice_phy_info *phy_info; - enum ice_status status; struct ice_vsi *vsi; u16 old_link_speed; bool old_link; + int status; phy_info = &pi->phy; phy_info->link_info_old = phy_info->link_info; @@ -936,11 +1121,11 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, */ status = ice_update_link_info(pi); if (status) - dev_dbg(dev, "Failed to update link status on port %d, err %s aq_err %s\n", - pi->lport, ice_stat_str(status), - ice_aq_str(pi->hw->adminq.sq_last_status)); + dev_dbg(dev, "Failed to update link status on port %d, err %d aq_err %s\n", + pi->lport, status, + libie_aq_str(pi->hw->adminq.sq_last_status)); - ice_check_module_power(pf, pi->phy.link_info.link_cfg_err); + ice_check_link_cfg_err(pf, pi->phy.link_info.link_cfg_err); /* Check if the link state is up after updating link info, and treat * this event as an UP event since the link is actually UP now. @@ -963,6 +1148,11 @@ ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up, if (link_up == old_link && link_speed == old_link_speed) return 0; + if (!link_up && old_link) + pf->link_down_events++; + + ice_ptp_link_change(pf, link_up); + if (ice_is_dcb_active(pf)) { if (test_bit(ICE_FLAG_DCB_ENA, pf->flags)) ice_dcb_rebuild(pf); @@ -1018,7 +1208,8 @@ static int ice_init_link_events(struct ice_port_info *pi) u16 mask; mask = ~((u16)(ICE_AQ_LINK_EVENT_UPDOWN | ICE_AQ_LINK_EVENT_MEDIA_NA | - ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL)); + ICE_AQ_LINK_EVENT_MODULE_QUAL_FAIL | + ICE_AQ_LINK_EVENT_PHY_FW_LOAD_FAIL)); if (ice_aq_set_event_mask(pi->hw, pi->lport, mask, NULL)) { dev_dbg(ice_hw_to_dev(pi->hw), "Failed to set link event mask for port %d\n", @@ -1062,64 +1253,63 @@ ice_handle_link_event(struct ice_pf *pf, struct ice_rq_event_info *event) return status; } -enum ice_aq_task_state { - ICE_AQ_TASK_WAITING = 0, - ICE_AQ_TASK_COMPLETE, - ICE_AQ_TASK_CANCELED, -}; - -struct ice_aq_task { - struct hlist_node entry; +/** + * ice_aq_prep_for_event - Prepare to wait for an AdminQ event from firmware + * @pf: pointer to the PF private structure + * @task: intermediate helper storage and identifier for waiting + * @opcode: the opcode to wait for + * + * Prepares to wait for a specific AdminQ completion event on the ARQ for + * a given PF. Actual wait would be done by a call to ice_aq_wait_for_event(). + * + * Calls are separated to allow caller registering for event before sending + * the command, which mitigates a race between registering and FW responding. + * + * To obtain only the descriptor contents, pass an task->event with null + * msg_buf. If the complete data buffer is desired, allocate the + * task->event.msg_buf with enough space ahead of time. + */ +void ice_aq_prep_for_event(struct ice_pf *pf, struct ice_aq_task *task, + u16 opcode) +{ + INIT_HLIST_NODE(&task->entry); + task->opcode = opcode; + task->state = ICE_AQ_TASK_WAITING; - u16 opcode; - struct ice_rq_event_info *event; - enum ice_aq_task_state state; -}; + spin_lock_bh(&pf->aq_wait_lock); + hlist_add_head(&task->entry, &pf->aq_wait_list); + spin_unlock_bh(&pf->aq_wait_lock); +} /** * ice_aq_wait_for_event - Wait for an AdminQ event from firmware * @pf: pointer to the PF private structure - * @opcode: the opcode to wait for + * @task: ptr prepared by ice_aq_prep_for_event() * @timeout: how long to wait, in jiffies - * @event: storage for the event info * * Waits for a specific AdminQ completion event on the ARQ for a given PF. The * current thread will be put to sleep until the specified event occurs or * until the given timeout is reached. * - * To obtain only the descriptor contents, pass an event without an allocated - * msg_buf. If the complete data buffer is desired, allocate the - * event->msg_buf with enough space ahead of time. - * * Returns: zero on success, or a negative error code on failure. */ -int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, - struct ice_rq_event_info *event) +int ice_aq_wait_for_event(struct ice_pf *pf, struct ice_aq_task *task, + unsigned long timeout) { + enum ice_aq_task_state *state = &task->state; struct device *dev = ice_pf_to_dev(pf); - struct ice_aq_task *task; - unsigned long start; + unsigned long start = jiffies; long ret; int err; - task = kzalloc(sizeof(*task), GFP_KERNEL); - if (!task) - return -ENOMEM; - - INIT_HLIST_NODE(&task->entry); - task->opcode = opcode; - task->event = event; - task->state = ICE_AQ_TASK_WAITING; - - spin_lock_bh(&pf->aq_wait_lock); - hlist_add_head(&task->entry, &pf->aq_wait_list); - spin_unlock_bh(&pf->aq_wait_lock); - - start = jiffies; - - ret = wait_event_interruptible_timeout(pf->aq_wait_queue, task->state, + ret = wait_event_interruptible_timeout(pf->aq_wait_queue, + *state != ICE_AQ_TASK_WAITING, timeout); - switch (task->state) { + switch (*state) { + case ICE_AQ_TASK_NOT_PREPARED: + WARN(1, "call to %s without ice_aq_prep_for_event()", __func__); + err = -EINVAL; + break; case ICE_AQ_TASK_WAITING: err = ret < 0 ? ret : -ETIMEDOUT; break; @@ -1130,7 +1320,7 @@ int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, err = ret < 0 ? ret : 0; break; default: - WARN(1, "Unexpected AdminQ wait task state %u", task->state); + WARN(1, "Unexpected AdminQ wait task state %u", *state); err = -EINVAL; break; } @@ -1138,12 +1328,11 @@ int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, dev_dbg(dev, "Waited %u msecs (max %u msecs) for firmware response to op 0x%04x\n", jiffies_to_msecs(jiffies - start), jiffies_to_msecs(timeout), - opcode); + task->opcode); spin_lock_bh(&pf->aq_wait_lock); hlist_del(&task->entry); spin_unlock_bh(&pf->aq_wait_lock); - kfree(task); return err; } @@ -1169,23 +1358,26 @@ int ice_aq_wait_for_event(struct ice_pf *pf, u16 opcode, unsigned long timeout, static void ice_aq_check_events(struct ice_pf *pf, u16 opcode, struct ice_rq_event_info *event) { + struct ice_rq_event_info *task_ev; struct ice_aq_task *task; bool found = false; spin_lock_bh(&pf->aq_wait_lock); hlist_for_each_entry(task, &pf->aq_wait_list, entry) { - if (task->state || task->opcode != opcode) + if (task->state != ICE_AQ_TASK_WAITING) + continue; + if (task->opcode != opcode) continue; - memcpy(&task->event->desc, &event->desc, sizeof(event->desc)); - task->event->msg_len = event->msg_len; + task_ev = &task->event; + memcpy(&task_ev->desc, &event->desc, sizeof(event->desc)); + task_ev->msg_len = event->msg_len; /* Only copy the data buffer if a destination was set */ - if (task->event->msg_buf && - task->event->buf_len > event->buf_len) { - memcpy(task->event->msg_buf, event->msg_buf, + if (task_ev->msg_buf && task_ev->buf_len >= event->buf_len) { + memcpy(task_ev->msg_buf, event->msg_buf, event->buf_len); - task->event->buf_len = event->buf_len; + task_ev->buf_len = event->buf_len; } task->state = ICE_AQ_TASK_COMPLETE; @@ -1216,6 +1408,8 @@ static void ice_aq_cancel_waiting_tasks(struct ice_pf *pf) wake_up(&pf->aq_wait_queue); } +#define ICE_MBX_OVERFLOW_WATERMARK 64 + /** * __ice_clean_ctrlq - helper function to clean controlq rings * @pf: ptr to struct ice_pf @@ -1306,15 +1500,16 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type) return 0; do { - enum ice_status ret; + struct ice_mbx_data data = {}; u16 opcode; + int ret; ret = ice_clean_rq_elem(hw, cq, &event, &pending); - if (ret == ICE_ERR_AQ_NO_WORK) + if (ret == -EALREADY) break; if (ret) { - dev_err(dev, "%s Receive Queue event error %s\n", qtype, - ice_stat_str(ret)); + dev_err(dev, "%s Receive Queue event error %d\n", qtype, + ret); break; } @@ -1332,15 +1527,31 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type) ice_vf_lan_overflow_event(pf, &event); break; case ice_mbx_opc_send_msg_to_pf: - if (!ice_is_malicious_vf(pf, &event, i, pending)) - ice_vc_process_vf_msg(pf, &event); + if (ice_is_feature_supported(pf, ICE_F_MBX_LIMIT)) { + ice_vc_process_vf_msg(pf, &event, NULL); + ice_mbx_vf_dec_trig_e830(hw, &event); + } else { + u16 val = hw->mailboxq.num_rq_entries; + + data.max_num_msgs_mbx = val; + val = ICE_MBX_OVERFLOW_WATERMARK; + data.async_watermark_val = val; + data.num_msg_proc = i; + data.num_pending_arq = pending; + + ice_vc_process_vf_msg(pf, &event, &data); + } break; - case ice_aqc_opc_fw_logging: - ice_output_fw_log(hw, &event.desc, event.msg_buf); + case ice_aqc_opc_fw_logs_event: + libie_get_fwlog_data(&hw->fwlog, event.msg_buf, + le16_to_cpu(event.desc.datalen)); break; case ice_aqc_opc_lldp_set_mib_change: ice_dcb_process_lldp_set_mib_change(pf, &event); break; + case ice_aqc_opc_get_health_status: + ice_process_health_status_event(pf, &event); + break; default: dev_dbg(dev, "%s Receive Queue unknown event 0x%04x ignored\n", qtype, opcode); @@ -1425,8 +1636,10 @@ static void ice_clean_sbq_subtask(struct ice_pf *pf) { struct ice_hw *hw = &pf->hw; - /* Nothing to do here if sideband queue is not supported */ - if (!ice_is_sbq_supported(hw)) { + /* if mac_type is not generic, sideband is not supported + * and there's nothing to do here + */ + if (!ice_is_generic_mac(hw)) { clear_bit(ICE_SIDEBANDQ_EVENT_PENDING, pf->state); return; } @@ -1486,7 +1699,7 @@ static int ice_service_task_stop(struct ice_pf *pf) ret = test_and_set_bit(ICE_SERVICE_DIS, pf->state); if (pf->serv_tmr.function) - del_timer_sync(&pf->serv_tmr); + timer_delete_sync(&pf->serv_tmr); if (pf->serv_task.func) cancel_work_sync(&pf->serv_task); @@ -1512,13 +1725,46 @@ static void ice_service_task_restart(struct ice_pf *pf) */ static void ice_service_timer(struct timer_list *t) { - struct ice_pf *pf = from_timer(pf, t, serv_tmr); + struct ice_pf *pf = timer_container_of(pf, t, serv_tmr); mod_timer(&pf->serv_tmr, round_jiffies(pf->serv_tmr_period + jiffies)); ice_service_task_schedule(pf); } /** + * ice_mdd_maybe_reset_vf - reset VF after MDD event + * @pf: pointer to the PF structure + * @vf: pointer to the VF structure + * @reset_vf_tx: whether Tx MDD has occurred + * @reset_vf_rx: whether Rx MDD has occurred + * + * Since the queue can get stuck on VF MDD events, the PF can be configured to + * automatically reset the VF by enabling the private ethtool flag + * mdd-auto-reset-vf. + */ +static void ice_mdd_maybe_reset_vf(struct ice_pf *pf, struct ice_vf *vf, + bool reset_vf_tx, bool reset_vf_rx) +{ + struct device *dev = ice_pf_to_dev(pf); + + if (!test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags)) + return; + + /* VF MDD event counters will be cleared by reset, so print the event + * prior to reset. + */ + if (reset_vf_tx) + ice_print_vf_tx_mdd_event(vf); + + if (reset_vf_rx) + ice_print_vf_rx_mdd_event(vf); + + dev_info(dev, "PF-to-VF reset on PF %d VF %d due to MDD event\n", + pf->hw.pf_id, vf->vf_id); + ice_reset_vf(vf, ICE_VF_RESET_NOTIFY | ICE_VF_RESET_LOCK); +} + +/** * ice_handle_mdd_event - handle malicious driver detect event * @pf: pointer to the PF structure * @@ -1532,7 +1778,8 @@ static void ice_handle_mdd_event(struct ice_pf *pf) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - unsigned int i; + struct ice_vf *vf; + unsigned int bkt; u32 reg; if (!test_and_clear_bit(ICE_MDD_EVENT_PENDING, pf->state)) { @@ -1546,52 +1793,46 @@ static void ice_handle_mdd_event(struct ice_pf *pf) /* find what triggered an MDD event */ reg = rd32(hw, GL_MDET_TX_PQM); if (reg & GL_MDET_TX_PQM_VALID_M) { - u8 pf_num = (reg & GL_MDET_TX_PQM_PF_NUM_M) >> - GL_MDET_TX_PQM_PF_NUM_S; - u16 vf_num = (reg & GL_MDET_TX_PQM_VF_NUM_M) >> - GL_MDET_TX_PQM_VF_NUM_S; - u8 event = (reg & GL_MDET_TX_PQM_MAL_TYPE_M) >> - GL_MDET_TX_PQM_MAL_TYPE_S; - u16 queue = ((reg & GL_MDET_TX_PQM_QNUM_M) >> - GL_MDET_TX_PQM_QNUM_S); + u8 pf_num = FIELD_GET(GL_MDET_TX_PQM_PF_NUM_M, reg); + u16 vf_num = FIELD_GET(GL_MDET_TX_PQM_VF_NUM_M, reg); + u8 event = FIELD_GET(GL_MDET_TX_PQM_MAL_TYPE_M, reg); + u16 queue = FIELD_GET(GL_MDET_TX_PQM_QNUM_M, reg); if (netif_msg_tx_err(pf)) dev_info(dev, "Malicious Driver Detection event %d on TX queue %d PF# %d VF# %d\n", event, queue, pf_num, vf_num); + ice_report_mdd_event(pf, ICE_MDD_SRC_TX_PQM, pf_num, vf_num, + event, queue); wr32(hw, GL_MDET_TX_PQM, 0xffffffff); } - reg = rd32(hw, GL_MDET_TX_TCLAN); + reg = rd32(hw, GL_MDET_TX_TCLAN_BY_MAC(hw)); if (reg & GL_MDET_TX_TCLAN_VALID_M) { - u8 pf_num = (reg & GL_MDET_TX_TCLAN_PF_NUM_M) >> - GL_MDET_TX_TCLAN_PF_NUM_S; - u16 vf_num = (reg & GL_MDET_TX_TCLAN_VF_NUM_M) >> - GL_MDET_TX_TCLAN_VF_NUM_S; - u8 event = (reg & GL_MDET_TX_TCLAN_MAL_TYPE_M) >> - GL_MDET_TX_TCLAN_MAL_TYPE_S; - u16 queue = ((reg & GL_MDET_TX_TCLAN_QNUM_M) >> - GL_MDET_TX_TCLAN_QNUM_S); + u8 pf_num = FIELD_GET(GL_MDET_TX_TCLAN_PF_NUM_M, reg); + u16 vf_num = FIELD_GET(GL_MDET_TX_TCLAN_VF_NUM_M, reg); + u8 event = FIELD_GET(GL_MDET_TX_TCLAN_MAL_TYPE_M, reg); + u16 queue = FIELD_GET(GL_MDET_TX_TCLAN_QNUM_M, reg); if (netif_msg_tx_err(pf)) dev_info(dev, "Malicious Driver Detection event %d on TX queue %d PF# %d VF# %d\n", event, queue, pf_num, vf_num); - wr32(hw, GL_MDET_TX_TCLAN, 0xffffffff); + ice_report_mdd_event(pf, ICE_MDD_SRC_TX_TCLAN, pf_num, vf_num, + event, queue); + wr32(hw, GL_MDET_TX_TCLAN_BY_MAC(hw), U32_MAX); } reg = rd32(hw, GL_MDET_RX); if (reg & GL_MDET_RX_VALID_M) { - u8 pf_num = (reg & GL_MDET_RX_PF_NUM_M) >> - GL_MDET_RX_PF_NUM_S; - u16 vf_num = (reg & GL_MDET_RX_VF_NUM_M) >> - GL_MDET_RX_VF_NUM_S; - u8 event = (reg & GL_MDET_RX_MAL_TYPE_M) >> - GL_MDET_RX_MAL_TYPE_S; - u16 queue = ((reg & GL_MDET_RX_QNUM_M) >> - GL_MDET_RX_QNUM_S); + u8 pf_num = FIELD_GET(GL_MDET_RX_PF_NUM_M, reg); + u16 vf_num = FIELD_GET(GL_MDET_RX_VF_NUM_M, reg); + u8 event = FIELD_GET(GL_MDET_RX_MAL_TYPE_M, reg); + u16 queue = FIELD_GET(GL_MDET_RX_QNUM_M, reg); if (netif_msg_rx_err(pf)) dev_info(dev, "Malicious Driver Detection event %d on RX queue %d PF# %d VF# %d\n", event, queue, pf_num, vf_num); + ice_report_mdd_event(pf, ICE_MDD_SRC_RX, pf_num, vf_num, event, + queue); wr32(hw, GL_MDET_RX, 0xffffffff); } @@ -1603,9 +1844,9 @@ static void ice_handle_mdd_event(struct ice_pf *pf) dev_info(dev, "Malicious Driver Detection event TX_PQM detected on PF\n"); } - reg = rd32(hw, PF_MDET_TX_TCLAN); + reg = rd32(hw, PF_MDET_TX_TCLAN_BY_MAC(hw)); if (reg & PF_MDET_TX_TCLAN_VALID_M) { - wr32(hw, PF_MDET_TX_TCLAN, 0xFFFF); + wr32(hw, PF_MDET_TX_TCLAN_BY_MAC(hw), 0xffff); if (netif_msg_tx_err(pf)) dev_info(dev, "Malicious Driver Detection event TX_TCLAN detected on PF\n"); } @@ -1620,61 +1861,63 @@ static void ice_handle_mdd_event(struct ice_pf *pf) /* Check to see if one of the VFs caused an MDD event, and then * increment counters and set print pending */ - ice_for_each_vf(pf, i) { - struct ice_vf *vf = &pf->vf[i]; + mutex_lock(&pf->vfs.table_lock); + ice_for_each_vf(pf, bkt, vf) { + bool reset_vf_tx = false, reset_vf_rx = false; - reg = rd32(hw, VP_MDET_TX_PQM(i)); + reg = rd32(hw, VP_MDET_TX_PQM(vf->vf_id)); if (reg & VP_MDET_TX_PQM_VALID_M) { - wr32(hw, VP_MDET_TX_PQM(i), 0xFFFF); + wr32(hw, VP_MDET_TX_PQM(vf->vf_id), 0xFFFF); vf->mdd_tx_events.count++; set_bit(ICE_MDD_VF_PRINT_PENDING, pf->state); if (netif_msg_tx_err(pf)) dev_info(dev, "Malicious Driver Detection event TX_PQM detected on VF %d\n", - i); + vf->vf_id); + + reset_vf_tx = true; } - reg = rd32(hw, VP_MDET_TX_TCLAN(i)); + reg = rd32(hw, VP_MDET_TX_TCLAN(vf->vf_id)); if (reg & VP_MDET_TX_TCLAN_VALID_M) { - wr32(hw, VP_MDET_TX_TCLAN(i), 0xFFFF); + wr32(hw, VP_MDET_TX_TCLAN(vf->vf_id), 0xFFFF); vf->mdd_tx_events.count++; set_bit(ICE_MDD_VF_PRINT_PENDING, pf->state); if (netif_msg_tx_err(pf)) dev_info(dev, "Malicious Driver Detection event TX_TCLAN detected on VF %d\n", - i); + vf->vf_id); + + reset_vf_tx = true; } - reg = rd32(hw, VP_MDET_TX_TDPU(i)); + reg = rd32(hw, VP_MDET_TX_TDPU(vf->vf_id)); if (reg & VP_MDET_TX_TDPU_VALID_M) { - wr32(hw, VP_MDET_TX_TDPU(i), 0xFFFF); + wr32(hw, VP_MDET_TX_TDPU(vf->vf_id), 0xFFFF); vf->mdd_tx_events.count++; set_bit(ICE_MDD_VF_PRINT_PENDING, pf->state); if (netif_msg_tx_err(pf)) dev_info(dev, "Malicious Driver Detection event TX_TDPU detected on VF %d\n", - i); + vf->vf_id); + + reset_vf_tx = true; } - reg = rd32(hw, VP_MDET_RX(i)); + reg = rd32(hw, VP_MDET_RX(vf->vf_id)); if (reg & VP_MDET_RX_VALID_M) { - wr32(hw, VP_MDET_RX(i), 0xFFFF); + wr32(hw, VP_MDET_RX(vf->vf_id), 0xFFFF); vf->mdd_rx_events.count++; set_bit(ICE_MDD_VF_PRINT_PENDING, pf->state); if (netif_msg_rx_err(pf)) dev_info(dev, "Malicious Driver Detection event RX detected on VF %d\n", - i); + vf->vf_id); - /* Since the queue is disabled on VF Rx MDD events, the - * PF can be configured to reset the VF through ethtool - * private flag mdd-auto-reset-vf. - */ - if (test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags)) { - /* VF MDD event counters will be cleared by - * reset, so print the event prior to reset. - */ - ice_print_vf_rx_mdd_event(vf); - ice_reset_vf(&pf->vf[i], false); - } + reset_vf_rx = true; } + + if (reset_vf_tx || reset_vf_rx) + ice_mdd_maybe_reset_vf(pf, vf, reset_vf_tx, + reset_vf_rx); } + mutex_unlock(&pf->vfs.table_lock); ice_print_vfs_mdd_events(pf); } @@ -1765,19 +2008,17 @@ static int ice_init_nvm_phy_type(struct ice_port_info *pi) { struct ice_aqc_get_phy_caps_data *pcaps; struct ice_pf *pf = pi->hw->back; - enum ice_status status; - int err = 0; + int err; pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL); if (!pcaps) return -ENOMEM; - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_NO_MEDIA, pcaps, - NULL); + err = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_NO_MEDIA, + pcaps, NULL); - if (status) { + if (err) { dev_err(ice_pf_to_dev(pf), "Get PHY capability failed.\n"); - err = -EIO; goto out; } @@ -1876,8 +2117,7 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi) struct ice_aqc_get_phy_caps_data *pcaps; struct ice_phy_info *phy = &pi->phy; struct ice_pf *pf = pi->hw->back; - enum ice_status status; - int err = 0; + int err; if (!(phy->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE)) return -EIO; @@ -1887,14 +2127,13 @@ static int ice_init_phy_user_cfg(struct ice_port_info *pi) return -ENOMEM; if (ice_fw_supports_report_dflt_cfg(pi->hw)) - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG, - pcaps, NULL); + err = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG, + pcaps, NULL); else - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, - pcaps, NULL); - if (status) { + err = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, + pcaps, NULL); + if (err) { dev_err(ice_pf_to_dev(pf), "Get PHY capability failed.\n"); - err = -EIO; goto err_out; } @@ -1948,16 +2187,16 @@ static int ice_configure_phy(struct ice_vsi *vsi) struct ice_aqc_set_phy_cfg_data *cfg; struct ice_phy_info *phy = &pi->phy; struct ice_pf *pf = vsi->back; - enum ice_status status; - int err = 0; + int err; /* Ensure we have media as we cannot configure a medialess port */ if (!(phy->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE)) - return -EPERM; + return -ENOMEDIUM; ice_print_topo_conflict(vsi); - if (phy->link_info.topo_media_conflict == ICE_AQ_LINK_TOPO_UNSUPP_MEDIA) + if (!test_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, pf->flags) && + phy->link_info.topo_media_conflict == ICE_AQ_LINK_TOPO_UNSUPP_MEDIA) return -EPERM; if (test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, pf->flags)) @@ -1968,12 +2207,11 @@ static int ice_configure_phy(struct ice_vsi *vsi) return -ENOMEM; /* Get current PHY config */ - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps, - NULL); - if (status) { - dev_err(dev, "Failed to get PHY configuration, VSI %d error %s\n", - vsi->vsi_num, ice_stat_str(status)); - err = -EIO; + err = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps, + NULL); + if (err) { + dev_err(dev, "Failed to get PHY configuration, VSI %d error %d\n", + vsi->vsi_num, err); goto done; } @@ -1987,15 +2225,14 @@ static int ice_configure_phy(struct ice_vsi *vsi) /* Use PHY topology as baseline for configuration */ memset(pcaps, 0, sizeof(*pcaps)); if (ice_fw_supports_report_dflt_cfg(pi->hw)) - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG, - pcaps, NULL); + err = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_DFLT_CFG, + pcaps, NULL); else - status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, - pcaps, NULL); - if (status) { - dev_err(dev, "Failed to get PHY caps, VSI %d error %s\n", - vsi->vsi_num, ice_stat_str(status)); - err = -EIO; + err = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA, + pcaps, NULL); + if (err) { + dev_err(dev, "Failed to get PHY caps, VSI %d error %d\n", + vsi->vsi_num, err); goto done; } @@ -2048,12 +2285,10 @@ static int ice_configure_phy(struct ice_vsi *vsi) /* Enable link and link update */ cfg->caps |= ICE_AQ_PHY_ENA_AUTO_LINK_UPDT | ICE_AQ_PHY_ENA_LINK; - status = ice_aq_set_phy_cfg(&pf->hw, pi, cfg, NULL); - if (status) { - dev_err(dev, "Failed to set phy config, VSI %d error %s\n", - vsi->vsi_num, ice_stat_str(status)); - err = -EIO; - } + err = ice_aq_set_phy_cfg(&pf->hw, pi, cfg, NULL); + if (err) + dev_err(dev, "Failed to set phy config, VSI %d error %d\n", + vsi->vsi_num, err); kfree(cfg); done: @@ -2088,7 +2323,7 @@ static void ice_check_media_subtask(struct ice_pf *pf) if (err) return; - ice_check_module_power(pf, pi->phy.link_info.link_cfg_err); + ice_check_link_cfg_err(pf, pi->phy.link_info.link_cfg_err); if (pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) { if (!test_bit(ICE_PHY_INIT_COMPLETE, pf->state)) @@ -2111,6 +2346,18 @@ static void ice_check_media_subtask(struct ice_pf *pf) } } +static void ice_service_task_recovery_mode(struct work_struct *work) +{ + struct ice_pf *pf = container_of(work, struct ice_pf, serv_task); + + set_bit(ICE_ADMINQ_EVENT_PENDING, pf->state); + ice_clean_adminq_subtask(pf); + + ice_service_task_complete(pf); + + mod_timer(&pf->serv_tmr, jiffies + msecs_to_jiffies(100)); +} + /** * ice_service_task - manage and run subtasks * @work: pointer to work_struct contained by the PF struct @@ -2120,9 +2367,11 @@ static void ice_service_task(struct work_struct *work) struct ice_pf *pf = container_of(work, struct ice_pf, serv_task); unsigned long start_time = jiffies; - /* subtasks */ + if (pf->health_reporters.tx_hang_buf.tx_ring) { + ice_report_tx_hang(pf); + pf->health_reporters.tx_hang_buf.tx_ring = NULL; + } - /* process reset requests first */ ice_reset_subtask(pf); /* bail if a reset/recovery cycle is pending or rebuild failed */ @@ -2133,6 +2382,40 @@ static void ice_service_task(struct work_struct *work) return; } + if (test_and_clear_bit(ICE_AUX_ERR_PENDING, pf->state)) { + struct iidc_rdma_event *event; + + event = kzalloc(sizeof(*event), GFP_KERNEL); + if (event) { + set_bit(IIDC_RDMA_EVENT_CRIT_ERR, event->type); + /* report the entire OICR value to AUX driver */ + swap(event->reg, pf->oicr_err_reg); + ice_send_event_to_aux(pf, event); + kfree(event); + } + } + + /* unplug aux dev per request, if an unplug request came in + * while processing a plug request, this will handle it + */ + if (test_and_clear_bit(ICE_FLAG_UNPLUG_AUX_DEV, pf->flags)) + ice_unplug_aux_dev(pf); + + /* Plug aux device per request */ + if (test_and_clear_bit(ICE_FLAG_PLUG_AUX_DEV, pf->flags)) + ice_plug_aux_dev(pf); + + if (test_and_clear_bit(ICE_FLAG_MTU_CHANGED, pf->flags)) { + struct iidc_rdma_event *event; + + event = kzalloc(sizeof(*event), GFP_KERNEL); + if (event) { + set_bit(IIDC_RDMA_EVENT_AFTER_MTU_CHANGE, event->type); + ice_send_event_to_aux(pf, event); + kfree(event); + } + } + ice_clean_adminq_subtask(pf); ice_check_media_subtask(pf); ice_check_for_hang_subtask(pf); @@ -2208,8 +2491,6 @@ int ice_schedule_reset(struct ice_pf *pf, enum ice_reset_req reset) return -EBUSY; } - ice_unplug_aux_dev(pf); - switch (reset) { case ICE_RESET_PFR: set_bit(ICE_PFR_REQ, pf->state); @@ -2229,34 +2510,6 @@ int ice_schedule_reset(struct ice_pf *pf, enum ice_reset_req reset) } /** - * ice_irq_affinity_notify - Callback for affinity changes - * @notify: context as to what irq was changed - * @mask: the new affinity mask - * - * This is a callback function used by the irq_set_affinity_notifier function - * so that we may register to receive changes to the irq affinity masks. - */ -static void -ice_irq_affinity_notify(struct irq_affinity_notify *notify, - const cpumask_t *mask) -{ - struct ice_q_vector *q_vector = - container_of(notify, struct ice_q_vector, affinity_notify); - - cpumask_copy(&q_vector->affinity_mask, mask); -} - -/** - * ice_irq_affinity_release - Callback for affinity notifier release - * @ref: internal core kernel usage - * - * This is a callback function used by the irq_set_affinity_notifier function - * to inform the current notification subscriber that they will no longer - * receive notifications. - */ -static void ice_irq_affinity_release(struct kref __always_unused *ref) {} - -/** * ice_vsi_ena_irq - Enable IRQ for the given VSI * @vsi: the VSI being configured */ @@ -2281,7 +2534,6 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) { int q_vectors = vsi->num_q_vectors; struct ice_pf *pf = vsi->back; - int base = vsi->base_vector; struct device *dev; int rx_int_idx = 0; int tx_int_idx = 0; @@ -2292,23 +2544,23 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) for (vector = 0; vector < q_vectors; vector++) { struct ice_q_vector *q_vector = vsi->q_vectors[vector]; - irq_num = pf->msix_entries[base + vector].vector; + irq_num = q_vector->irq.virq; - if (q_vector->tx.ring && q_vector->rx.ring) { + if (q_vector->tx.tx_ring && q_vector->rx.rx_ring) { snprintf(q_vector->name, sizeof(q_vector->name) - 1, "%s-%s-%d", basename, "TxRx", rx_int_idx++); tx_int_idx++; - } else if (q_vector->rx.ring) { + } else if (q_vector->rx.rx_ring) { snprintf(q_vector->name, sizeof(q_vector->name) - 1, "%s-%s-%d", basename, "rx", rx_int_idx++); - } else if (q_vector->tx.ring) { + } else if (q_vector->tx.tx_ring) { snprintf(q_vector->name, sizeof(q_vector->name) - 1, "%s-%s-%d", basename, "tx", tx_int_idx++); } else { /* skip this unused q_vector */ continue; } - if (vsi->type == ICE_VSI_CTRL && vsi->vf_id != ICE_INVAL_VFID) + if (vsi->type == ICE_VSI_CTRL && vsi->vf) err = devm_request_irq(dev, irq_num, vsi->irq_handler, IRQF_SHARED, q_vector->name, q_vector); @@ -2320,31 +2572,21 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) err); goto free_q_irqs; } + } - /* register for affinity change notifications */ - if (!IS_ENABLED(CONFIG_RFS_ACCEL)) { - struct irq_affinity_notify *affinity_notify; - - affinity_notify = &q_vector->affinity_notify; - affinity_notify->notify = ice_irq_affinity_notify; - affinity_notify->release = ice_irq_affinity_release; - irq_set_affinity_notifier(irq_num, affinity_notify); - } - - /* assign the mask for this irq */ - irq_set_affinity_hint(irq_num, &q_vector->affinity_mask); + err = ice_set_cpu_rx_rmap(vsi); + if (err) { + netdev_err(vsi->netdev, "Failed to setup CPU RMAP on VSI %u: %pe\n", + vsi->vsi_num, ERR_PTR(err)); + goto free_q_irqs; } vsi->irqs_ready = true; return 0; free_q_irqs: - while (vector) { - vector--; - irq_num = pf->msix_entries[base + vector].vector; - if (!IS_ENABLED(CONFIG_RFS_ACCEL)) - irq_set_affinity_notifier(irq_num, NULL); - irq_set_affinity_hint(irq_num, NULL); + while (vector--) { + irq_num = vsi->q_vectors[vector]->irq.virq; devm_free_irq(dev, irq_num, &vsi->q_vectors[vector]); } return err; @@ -2359,20 +2601,27 @@ free_q_irqs: static int ice_xdp_alloc_setup_rings(struct ice_vsi *vsi) { struct device *dev = ice_pf_to_dev(vsi->back); - int i; + struct ice_tx_desc *tx_desc; + int i, j; - for (i = 0; i < vsi->num_xdp_txq; i++) { + ice_for_each_xdp_txq(vsi, i) { u16 xdp_q_idx = vsi->alloc_txq + i; - struct ice_ring *xdp_ring; + struct ice_ring_stats *ring_stats; + struct ice_tx_ring *xdp_ring; xdp_ring = kzalloc(sizeof(*xdp_ring), GFP_KERNEL); - if (!xdp_ring) goto free_xdp_rings; + ring_stats = kzalloc(sizeof(*ring_stats), GFP_KERNEL); + if (!ring_stats) { + ice_free_tx_ring(xdp_ring); + goto free_xdp_rings; + } + + xdp_ring->ring_stats = ring_stats; xdp_ring->q_index = xdp_q_idx; xdp_ring->reg_idx = vsi->txq_map[xdp_q_idx]; - xdp_ring->ring_active = false; xdp_ring->vsi = vsi; xdp_ring->netdev = NULL; xdp_ring->dev = dev; @@ -2381,15 +2630,23 @@ static int ice_xdp_alloc_setup_rings(struct ice_vsi *vsi) if (ice_setup_tx_ring(xdp_ring)) goto free_xdp_rings; ice_set_ring_xdp(xdp_ring); - xdp_ring->xsk_pool = ice_xsk_pool(xdp_ring); + spin_lock_init(&xdp_ring->tx_lock); + for (j = 0; j < xdp_ring->count; j++) { + tx_desc = ICE_TX_DESC(xdp_ring, j); + tx_desc->cmd_type_offset_bsz = 0; + } } return 0; free_xdp_rings: - for (; i >= 0; i--) - if (vsi->xdp_rings[i] && vsi->xdp_rings[i]->desc) + for (; i >= 0; i--) { + if (vsi->xdp_rings[i] && vsi->xdp_rings[i]->desc) { + kfree_rcu(vsi->xdp_rings[i]->ring_stats, rcu); + vsi->xdp_rings[i]->ring_stats = NULL; ice_free_tx_ring(vsi->xdp_rings[i]); + } + } return -ENOMEM; } @@ -2404,24 +2661,100 @@ static void ice_vsi_assign_bpf_prog(struct ice_vsi *vsi, struct bpf_prog *prog) int i; old_prog = xchg(&vsi->xdp_prog, prog); + ice_for_each_rxq(vsi, i) + WRITE_ONCE(vsi->rx_rings[i]->xdp_prog, vsi->xdp_prog); + if (old_prog) bpf_prog_put(old_prog); +} - ice_for_each_rxq(vsi, i) - WRITE_ONCE(vsi->rx_rings[i]->xdp_prog, vsi->xdp_prog); +static struct ice_tx_ring *ice_xdp_ring_from_qid(struct ice_vsi *vsi, int qid) +{ + struct ice_q_vector *q_vector; + struct ice_tx_ring *ring; + + if (static_key_enabled(&ice_xdp_locking_key)) + return vsi->xdp_rings[qid % vsi->num_xdp_txq]; + + q_vector = vsi->rx_rings[qid]->q_vector; + ice_for_each_tx_ring(ring, q_vector->tx) + if (ice_ring_is_xdp(ring)) + return ring; + + return NULL; +} + +/** + * ice_map_xdp_rings - Map XDP rings to interrupt vectors + * @vsi: the VSI with XDP rings being configured + * + * Map XDP rings to interrupt vectors and perform the configuration steps + * dependent on the mapping. + */ +void ice_map_xdp_rings(struct ice_vsi *vsi) +{ + int xdp_rings_rem = vsi->num_xdp_txq; + int v_idx, q_idx; + + /* follow the logic from ice_vsi_map_rings_to_vectors */ + ice_for_each_q_vector(vsi, v_idx) { + struct ice_q_vector *q_vector = vsi->q_vectors[v_idx]; + int xdp_rings_per_v, q_id, q_base; + + xdp_rings_per_v = DIV_ROUND_UP(xdp_rings_rem, + vsi->num_q_vectors - v_idx); + q_base = vsi->num_xdp_txq - xdp_rings_rem; + + for (q_id = q_base; q_id < (q_base + xdp_rings_per_v); q_id++) { + struct ice_tx_ring *xdp_ring = vsi->xdp_rings[q_id]; + + xdp_ring->q_vector = q_vector; + xdp_ring->next = q_vector->tx.tx_ring; + q_vector->tx.tx_ring = xdp_ring; + } + xdp_rings_rem -= xdp_rings_per_v; + } + + ice_for_each_rxq(vsi, q_idx) { + vsi->rx_rings[q_idx]->xdp_ring = ice_xdp_ring_from_qid(vsi, + q_idx); + ice_tx_xsk_pool(vsi, q_idx); + } +} + +/** + * ice_unmap_xdp_rings - Unmap XDP rings from interrupt vectors + * @vsi: the VSI with XDP rings being unmapped + */ +static void ice_unmap_xdp_rings(struct ice_vsi *vsi) +{ + int v_idx; + + ice_for_each_q_vector(vsi, v_idx) { + struct ice_q_vector *q_vector = vsi->q_vectors[v_idx]; + struct ice_tx_ring *ring; + + ice_for_each_tx_ring(ring, q_vector->tx) + if (!ring->tx_buf || !ice_ring_is_xdp(ring)) + break; + + /* restore the value of last node prior to XDP setup */ + q_vector->tx.tx_ring = ring; + } } /** * ice_prepare_xdp_rings - Allocate, configure and setup Tx rings for XDP * @vsi: VSI to bring up Tx rings used by XDP * @prog: bpf program that will be assigned to VSI + * @cfg_type: create from scratch or restore the existing configuration * * Return 0 on success and negative value on error */ -int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) +int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog, + enum ice_xdp_cfg cfg_type) { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; - int xdp_rings_rem = vsi->num_xdp_txq; struct ice_pf *pf = vsi->back; struct ice_qs_cfg xdp_qs_cfg = { .qs_mutex = &pf->avail_q_mutex, @@ -2433,9 +2766,8 @@ int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) .vsi_map_offset = vsi->alloc_txq, .mapping_mode = ICE_VSI_MAP_CONTIG }; - enum ice_status status; struct device *dev; - int i, v_idx; + int status, i; dev = ice_pf_to_dev(pf); vsi->xdp_rings = devm_kcalloc(dev, vsi->num_xdp_txq, @@ -2447,35 +2779,22 @@ int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) if (__ice_vsi_get_qs(&xdp_qs_cfg)) goto err_map_xdp; + if (static_key_enabled(&ice_xdp_locking_key)) + netdev_warn(vsi->netdev, + "Could not allocate one XDP Tx ring per CPU, XDP_TX/XDP_REDIRECT actions will be slower\n"); + if (ice_xdp_alloc_setup_rings(vsi)) goto clear_xdp_rings; - /* follow the logic from ice_vsi_map_rings_to_vectors */ - ice_for_each_q_vector(vsi, v_idx) { - struct ice_q_vector *q_vector = vsi->q_vectors[v_idx]; - int xdp_rings_per_v, q_id, q_base; - - xdp_rings_per_v = DIV_ROUND_UP(xdp_rings_rem, - vsi->num_q_vectors - v_idx); - q_base = vsi->num_xdp_txq - xdp_rings_rem; - - for (q_id = q_base; q_id < (q_base + xdp_rings_per_v); q_id++) { - struct ice_ring *xdp_ring = vsi->xdp_rings[q_id]; - - xdp_ring->q_vector = q_vector; - xdp_ring->next = q_vector->tx.ring; - q_vector->tx.ring = xdp_ring; - } - xdp_rings_rem -= xdp_rings_per_v; - } - /* omit the scheduler update if in reset path; XDP queues will be * taken into account at the end of ice_vsi_rebuild, where * ice_cfg_vsi_lan is being called */ - if (ice_is_reset_in_progress(pf->state)) + if (cfg_type == ICE_XDP_CFG_PART) return 0; + ice_map_xdp_rings(vsi); + /* tell the Tx scheduler that right now we have * additional queues */ @@ -2485,15 +2804,28 @@ int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); if (status) { - dev_err(dev, "Failed VSI LAN queue config for XDP, error: %s\n", - ice_stat_str(status)); - goto clear_xdp_rings; + dev_err(dev, "Failed VSI LAN queue config for XDP, error: %d\n", + status); + goto unmap_xdp_rings; } - ice_vsi_assign_bpf_prog(vsi, prog); + + /* assign the prog only when it's not already present on VSI; + * this flow is a subject of both ethtool -L and ndo_bpf flows; + * VSI rebuild that happens under ethtool -L can expose us to + * the bpf_prog refcount issues as we would be swapping same + * bpf_prog pointers from vsi->xdp_prog and calling bpf_prog_put + * on it as it would be treated as an 'old_prog'; for ndo_bpf + * this is not harmful as dev_xdp_install bumps the refcount + * before calling the op exposed by the driver; + */ + if (!ice_is_xdp_ena_vsi(vsi)) + ice_vsi_assign_bpf_prog(vsi, prog); return 0; +unmap_xdp_rings: + ice_unmap_xdp_rings(vsi); clear_xdp_rings: - for (i = 0; i < vsi->num_xdp_txq; i++) + ice_for_each_xdp_txq(vsi, i) if (vsi->xdp_rings[i]) { kfree_rcu(vsi->xdp_rings[i], rcu); vsi->xdp_rings[i] = NULL; @@ -2501,61 +2833,56 @@ clear_xdp_rings: err_map_xdp: mutex_lock(&pf->avail_q_mutex); - for (i = 0; i < vsi->num_xdp_txq; i++) { + ice_for_each_xdp_txq(vsi, i) { clear_bit(vsi->txq_map[i + vsi->alloc_txq], pf->avail_txqs); vsi->txq_map[i + vsi->alloc_txq] = ICE_INVAL_Q_INDEX; } mutex_unlock(&pf->avail_q_mutex); devm_kfree(dev, vsi->xdp_rings); + vsi->xdp_rings = NULL; + return -ENOMEM; } /** * ice_destroy_xdp_rings - undo the configuration made by ice_prepare_xdp_rings * @vsi: VSI to remove XDP rings + * @cfg_type: disable XDP permanently or allow it to be restored later * * Detach XDP rings from irq vectors, clean up the PF bitmap and free * resources */ -int ice_destroy_xdp_rings(struct ice_vsi *vsi) +int ice_destroy_xdp_rings(struct ice_vsi *vsi, enum ice_xdp_cfg cfg_type) { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; struct ice_pf *pf = vsi->back; - int i, v_idx; + int i; /* q_vectors are freed in reset path so there's no point in detaching - * rings; in case of rebuild being triggered not from reset bits - * in pf->state won't be set, so additionally check first q_vector - * against NULL + * rings */ - if (ice_is_reset_in_progress(pf->state) || !vsi->q_vectors[0]) + if (cfg_type == ICE_XDP_CFG_PART) goto free_qmap; - ice_for_each_q_vector(vsi, v_idx) { - struct ice_q_vector *q_vector = vsi->q_vectors[v_idx]; - struct ice_ring *ring; - - ice_for_each_ring(ring, q_vector->tx) - if (!ring->tx_buf || !ice_ring_is_xdp(ring)) - break; - - /* restore the value of last node prior to XDP setup */ - q_vector->tx.ring = ring; - } + ice_unmap_xdp_rings(vsi); free_qmap: mutex_lock(&pf->avail_q_mutex); - for (i = 0; i < vsi->num_xdp_txq; i++) { + ice_for_each_xdp_txq(vsi, i) { clear_bit(vsi->txq_map[i + vsi->alloc_txq], pf->avail_txqs); vsi->txq_map[i + vsi->alloc_txq] = ICE_INVAL_Q_INDEX; } mutex_unlock(&pf->avail_q_mutex); - for (i = 0; i < vsi->num_xdp_txq; i++) + ice_for_each_xdp_txq(vsi, i) if (vsi->xdp_rings[i]) { - if (vsi->xdp_rings[i]->desc) + if (vsi->xdp_rings[i]->desc) { + synchronize_rcu(); ice_free_tx_ring(vsi->xdp_rings[i]); + } + kfree_rcu(vsi->xdp_rings[i]->ring_stats, rcu); + vsi->xdp_rings[i]->ring_stats = NULL; kfree_rcu(vsi->xdp_rings[i], rcu); vsi->xdp_rings[i] = NULL; } @@ -2563,7 +2890,10 @@ free_qmap: devm_kfree(ice_pf_to_dev(pf), vsi->xdp_rings); vsi->xdp_rings = NULL; - if (ice_is_reset_in_progress(pf->state) || !vsi->q_vectors[0]) + if (static_key_enabled(&ice_xdp_locking_key)) + static_branch_dec(&ice_xdp_locking_key); + + if (cfg_type == ICE_XDP_CFG_PART) return 0; ice_vsi_assign_bpf_prog(vsi, NULL); @@ -2590,14 +2920,49 @@ static void ice_vsi_rx_napi_schedule(struct ice_vsi *vsi) int i; ice_for_each_rxq(vsi, i) { - struct ice_ring *rx_ring = vsi->rx_rings[i]; + struct ice_rx_ring *rx_ring = vsi->rx_rings[i]; - if (rx_ring->xsk_pool) + if (READ_ONCE(rx_ring->xsk_pool)) napi_schedule(&rx_ring->q_vector->napi); } } /** + * ice_vsi_determine_xdp_res - figure out how many Tx qs can XDP have + * @vsi: VSI to determine the count of XDP Tx qs + * + * returns 0 if Tx qs count is higher than at least half of CPU count, + * -ENOMEM otherwise + */ +int ice_vsi_determine_xdp_res(struct ice_vsi *vsi) +{ + u16 avail = ice_get_avail_txq_count(vsi->back); + u16 cpus = num_possible_cpus(); + + if (avail < cpus / 2) + return -ENOMEM; + + if (vsi->type == ICE_VSI_SF) + avail = vsi->alloc_txq; + + vsi->num_xdp_txq = min_t(u16, avail, cpus); + + if (vsi->num_xdp_txq < cpus) + static_branch_inc(&ice_xdp_locking_key); + + return 0; +} + +/** + * ice_max_xdp_frame_size - returns the maximum allowed frame size for XDP + * @vsi: Pointer to VSI structure + */ +static int ice_max_xdp_frame_size(struct ice_vsi *vsi) +{ + return ICE_RXBUF_3072; +} + +/** * ice_xdp_setup_prog - Add or remove XDP eBPF program * @vsi: VSI to setup XDP for * @prog: XDP program @@ -2607,17 +2972,30 @@ static int ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, struct netlink_ext_ack *extack) { - int frame_size = vsi->netdev->mtu + ICE_ETH_PKT_HDR_PAD; - bool if_running = netif_running(vsi->netdev); + unsigned int frame_size = vsi->netdev->mtu + ICE_ETH_PKT_HDR_PAD; int ret = 0, xdp_ring_err = 0; + bool if_running; - if (frame_size > vsi->rx_buf_len) { - NL_SET_ERR_MSG_MOD(extack, "MTU too large for loading XDP"); - return -EOPNOTSUPP; + if (prog && !prog->aux->xdp_has_frags) { + if (frame_size > ice_max_xdp_frame_size(vsi)) { + NL_SET_ERR_MSG_MOD(extack, + "MTU is too large for linear frames and XDP prog does not support frags"); + return -EOPNOTSUPP; + } + } + + /* hot swap progs and avoid toggling link */ + if (ice_is_xdp_ena_vsi(vsi) == !!prog || + test_bit(ICE_VSI_REBUILD_PENDING, vsi->state)) { + ice_vsi_assign_bpf_prog(vsi, prog); + return 0; } + if_running = netif_running(vsi->netdev) && + !test_and_set_bit(ICE_VSI_DOWN, vsi->state); + /* need to stop netdev while setting up the program for Rx rings */ - if (if_running && !test_and_set_bit(ICE_VSI_DOWN, vsi->state)) { + if (if_running) { ret = ice_down(vsi); if (ret) { NL_SET_ERR_MSG_MOD(extack, "Preparing device for XDP attach failed"); @@ -2626,18 +3004,27 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, } if (!ice_is_xdp_ena_vsi(vsi) && prog) { - vsi->num_xdp_txq = vsi->alloc_rxq; - xdp_ring_err = ice_prepare_xdp_rings(vsi, prog); - if (xdp_ring_err) - NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Tx resources failed"); + xdp_ring_err = ice_vsi_determine_xdp_res(vsi); + if (xdp_ring_err) { + NL_SET_ERR_MSG_MOD(extack, "Not enough Tx resources for XDP"); + goto resume_if; + } else { + xdp_ring_err = ice_prepare_xdp_rings(vsi, prog, + ICE_XDP_CFG_FULL); + if (xdp_ring_err) { + NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Tx resources failed"); + goto resume_if; + } + } + xdp_features_set_redirect_target(vsi->netdev, true); } else if (ice_is_xdp_ena_vsi(vsi) && !prog) { - xdp_ring_err = ice_destroy_xdp_rings(vsi); + xdp_features_clear_redirect_target(vsi->netdev); + xdp_ring_err = ice_destroy_xdp_rings(vsi, ICE_XDP_CFG_FULL); if (xdp_ring_err) NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Tx resources failed"); - } else { - ice_vsi_assign_bpf_prog(vsi, prog); } +resume_if: if (if_running) ret = ice_up(vsi); @@ -2666,25 +3053,32 @@ static int ice_xdp_safe_mode(struct net_device __always_unused *dev, * @dev: netdevice * @xdp: XDP command */ -static int ice_xdp(struct net_device *dev, struct netdev_bpf *xdp) +int ice_xdp(struct net_device *dev, struct netdev_bpf *xdp) { struct ice_netdev_priv *np = netdev_priv(dev); struct ice_vsi *vsi = np->vsi; + int ret; - if (vsi->type != ICE_VSI_PF) { - NL_SET_ERR_MSG_MOD(xdp->extack, "XDP can be loaded only on PF VSI"); + if (vsi->type != ICE_VSI_PF && vsi->type != ICE_VSI_SF) { + NL_SET_ERR_MSG_MOD(xdp->extack, "XDP can be loaded only on PF or SF VSI"); return -EINVAL; } + mutex_lock(&vsi->xdp_state_lock); + switch (xdp->command) { case XDP_SETUP_PROG: - return ice_xdp_setup_prog(vsi, xdp->prog, xdp->extack); + ret = ice_xdp_setup_prog(vsi, xdp->prog, xdp->extack); + break; case XDP_SETUP_XSK_POOL: - return ice_xsk_pool_setup(vsi, xdp->xsk.pool, - xdp->xsk.queue_id); + ret = ice_xsk_pool_setup(vsi, xdp->xsk.pool, xdp->xsk.queue_id); + break; default: - return -EINVAL; + ret = -EINVAL; } + + mutex_unlock(&vsi->xdp_state_lock); + return ret; } /** @@ -2694,6 +3088,7 @@ static int ice_xdp(struct net_device *dev, struct netdev_bpf *xdp) static void ice_ena_misc_vector(struct ice_pf *pf) { struct ice_hw *hw = &pf->hw; + u32 pf_intr_start_offset; u32 val; /* Disable anti-spoof detection interrupt to prevent spurious event @@ -2720,11 +3115,54 @@ static void ice_ena_misc_vector(struct ice_pf *pf) wr32(hw, PFINT_OICR_ENA, val); /* SW_ITR_IDX = 0, but don't change INTENA */ - wr32(hw, GLINT_DYN_CTL(pf->oicr_idx), + wr32(hw, GLINT_DYN_CTL(pf->oicr_irq.index), + GLINT_DYN_CTL_SW_ITR_INDX_M | GLINT_DYN_CTL_INTENA_MSK_M); + + if (!pf->hw.dev_caps.ts_dev_info.ts_ll_int_read) + return; + pf_intr_start_offset = rd32(hw, PFINT_ALLOC) & PFINT_ALLOC_FIRST; + wr32(hw, GLINT_DYN_CTL(pf->ll_ts_irq.index + pf_intr_start_offset), GLINT_DYN_CTL_SW_ITR_INDX_M | GLINT_DYN_CTL_INTENA_MSK_M); } /** + * ice_ll_ts_intr - ll_ts interrupt handler + * @irq: interrupt number + * @data: pointer to a q_vector + */ +static irqreturn_t ice_ll_ts_intr(int __always_unused irq, void *data) +{ + struct ice_pf *pf = data; + u32 pf_intr_start_offset; + struct ice_ptp_tx *tx; + unsigned long flags; + struct ice_hw *hw; + u32 val; + u8 idx; + + hw = &pf->hw; + tx = &pf->ptp.port.tx; + spin_lock_irqsave(&tx->lock, flags); + if (tx->init) { + ice_ptp_complete_tx_single_tstamp(tx); + + idx = find_next_bit_wrap(tx->in_use, tx->len, + tx->last_ll_ts_idx_read + 1); + if (idx != tx->len) + ice_ptp_req_tx_single_tstamp(tx, idx); + } + spin_unlock_irqrestore(&tx->lock, flags); + + val = GLINT_DYN_CTL_INTENA_M | GLINT_DYN_CTL_CLEARPBA_M | + (ICE_ITR_NONE << GLINT_DYN_CTL_ITR_INDX_S); + pf_intr_start_offset = rd32(hw, PFINT_ALLOC) & PFINT_ALLOC_FIRST; + wr32(hw, GLINT_DYN_CTL(pf->ll_ts_irq.index + pf_intr_start_offset), + val); + + return IRQ_HANDLED; +} + +/** * ice_misc_intr - misc interrupt handler * @irq: interrupt number * @data: pointer to a q_vector @@ -2732,8 +3170,8 @@ static void ice_ena_misc_vector(struct ice_pf *pf) static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) { struct ice_pf *pf = (struct ice_pf *)data; + irqreturn_t ret = IRQ_HANDLED; struct ice_hw *hw = &pf->hw; - irqreturn_t ret = IRQ_NONE; struct device *dev; u32 oicr, ena_mask; @@ -2772,8 +3210,8 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) /* we have a reset warning */ ena_mask &= ~PFINT_OICR_GRST_M; - reset = (rd32(hw, GLGEN_RSTAT) & GLGEN_RSTAT_RESET_TYPE_M) >> - GLGEN_RSTAT_RESET_TYPE_S; + reset = FIELD_GET(GLGEN_RSTAT_RESET_TYPE_M, + rd32(hw, GLGEN_RSTAT)); if (reset == ICE_RESET_CORER) pf->corer_count++; @@ -2814,34 +3252,32 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) if (oicr & PFINT_OICR_TSYN_TX_M) { ena_mask &= ~PFINT_OICR_TSYN_TX_M; - ice_ptp_process_ts(pf); + + ret = ice_ptp_ts_irq(pf); } if (oicr & PFINT_OICR_TSYN_EVNT_M) { u8 tmr_idx = hw->func_caps.ts_func_info.tmr_index_owned; u32 gltsyn_stat = rd32(hw, GLTSYN_STAT(tmr_idx)); - /* Save EVENTs from GTSYN register */ - pf->ptp.ext_ts_irq |= gltsyn_stat & (GLTSYN_STAT_EVENT0_M | - GLTSYN_STAT_EVENT1_M | - GLTSYN_STAT_EVENT2_M); ena_mask &= ~PFINT_OICR_TSYN_EVNT_M; - kthread_queue_work(pf->ptp.kworker, &pf->ptp.extts_work); + + if (ice_pf_src_tmr_owned(pf)) { + /* Save EVENTs from GLTSYN register */ + pf->ptp.ext_ts_irq |= gltsyn_stat & + (GLTSYN_STAT_EVENT0_M | + GLTSYN_STAT_EVENT1_M | + GLTSYN_STAT_EVENT2_M); + + ice_ptp_extts_event(pf); + } } #define ICE_AUX_CRIT_ERR (PFINT_OICR_PE_CRITERR_M | PFINT_OICR_HMC_ERR_M | PFINT_OICR_PE_PUSH_M) if (oicr & ICE_AUX_CRIT_ERR) { - struct iidc_event *event; - + pf->oicr_err_reg |= oicr; + set_bit(ICE_AUX_ERR_PENDING, pf->state); ena_mask &= ~ICE_AUX_CRIT_ERR; - event = kzalloc(sizeof(*event), GFP_KERNEL); - if (event) { - set_bit(IIDC_EVENT_CRIT_ERR, event->type); - /* report the entire OICR value to AUX driver */ - event->reg = oicr; - ice_send_event_to_aux(pf, event); - kfree(event); - } } /* Report any remaining unexpected interrupts */ @@ -2854,18 +3290,47 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) if (oicr & (PFINT_OICR_PCI_EXCEPTION_M | PFINT_OICR_ECC_ERR_M)) { set_bit(ICE_PFR_REQ, pf->state); - ice_service_task_schedule(pf); } } - ret = IRQ_HANDLED; - ice_service_task_schedule(pf); - ice_irq_dynamic_ena(hw, NULL, NULL); + if (ret == IRQ_HANDLED) + ice_irq_dynamic_ena(hw, NULL, NULL); return ret; } /** + * ice_misc_intr_thread_fn - misc interrupt thread function + * @irq: interrupt number + * @data: pointer to a q_vector + */ +static irqreturn_t ice_misc_intr_thread_fn(int __always_unused irq, void *data) +{ + struct ice_pf *pf = data; + struct ice_hw *hw; + + hw = &pf->hw; + + if (ice_is_reset_in_progress(pf->state)) + goto skip_irq; + + if (test_and_clear_bit(ICE_MISC_THREAD_TX_TSTAMP, pf->misc_thread)) { + /* Process outstanding Tx timestamps. If there is more work, + * re-arm the interrupt to trigger again. + */ + if (ice_ptp_process_ts(pf) == ICE_TX_TSTAMP_WORK_PENDING) { + wr32(hw, PFINT_OICR, PFINT_OICR_TSYN_TX_M); + ice_flush(hw); + } + } + +skip_irq: + ice_irq_dynamic_ena(hw, NULL, NULL); + + return IRQ_HANDLED; +} + +/** * ice_dis_ctrlq_interrupts - disable control queue interrupts * @hw: pointer to HW structure */ @@ -2890,11 +3355,26 @@ static void ice_dis_ctrlq_interrupts(struct ice_hw *hw) } /** + * ice_free_irq_msix_ll_ts- Unroll ll_ts vector setup + * @pf: board private structure + */ +static void ice_free_irq_msix_ll_ts(struct ice_pf *pf) +{ + int irq_num = pf->ll_ts_irq.virq; + + synchronize_irq(irq_num); + devm_free_irq(ice_pf_to_dev(pf), irq_num, pf); + + ice_free_irq(pf, pf->ll_ts_irq); +} + +/** * ice_free_irq_msix_misc - Unroll misc vector setup * @pf: board private structure */ static void ice_free_irq_msix_misc(struct ice_pf *pf) { + int misc_irq_num = pf->oicr_irq.virq; struct ice_hw *hw = &pf->hw; ice_dis_ctrlq_interrupts(hw); @@ -2903,14 +3383,12 @@ static void ice_free_irq_msix_misc(struct ice_pf *pf) wr32(hw, PFINT_OICR_ENA, 0); ice_flush(hw); - if (pf->msix_entries) { - synchronize_irq(pf->msix_entries[pf->oicr_idx].vector); - devm_free_irq(ice_pf_to_dev(pf), - pf->msix_entries[pf->oicr_idx].vector, pf); - } + synchronize_irq(misc_irq_num); + devm_free_irq(ice_pf_to_dev(pf), misc_irq_num, pf); - pf->num_avail_sw_msix += 1; - ice_free_res(pf->irq_tracker, pf->oicr_idx, ICE_RES_MISC_VEC_ID); + ice_free_irq(pf, pf->oicr_irq); + if (pf->hw.dev_caps.ts_dev_info.ts_ll_int_read) + ice_free_irq_msix_ll_ts(pf); } /** @@ -2936,10 +3414,12 @@ static void ice_ena_ctrlq_interrupts(struct ice_hw *hw, u16 reg_idx) PFINT_MBX_CTL_CAUSE_ENA_M); wr32(hw, PFINT_MBX_CTL, val); - /* This enables Sideband queue Interrupt causes */ - val = ((reg_idx & PFINT_SB_CTL_MSIX_INDX_M) | - PFINT_SB_CTL_CAUSE_ENA_M); - wr32(hw, PFINT_SB_CTL, val); + if (!hw->dev_caps.ts_dev_info.ts_ll_int_read) { + /* enable Sideband queue Interrupt causes */ + val = ((reg_idx & PFINT_SB_CTL_MSIX_INDX_M) | + PFINT_SB_CTL_CAUSE_ENA_M); + wr32(hw, PFINT_SB_CTL, val); + } ice_flush(hw); } @@ -2956,12 +3436,17 @@ static int ice_req_irq_msix_misc(struct ice_pf *pf) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - int oicr_idx, err = 0; + u32 pf_intr_start_offset; + struct msi_map irq; + int err = 0; if (!pf->int_name[0]) snprintf(pf->int_name, sizeof(pf->int_name) - 1, "%s-%s:misc", dev_driver_string(dev), dev_name(dev)); + if (!pf->int_name_ll_ts[0]) + snprintf(pf->int_name_ll_ts, sizeof(pf->int_name_ll_ts) - 1, + "%s-%s:ll_ts", dev_driver_string(dev), dev_name(dev)); /* Do not request IRQ but do enable OICR interrupt since settings are * lost during reset. Note that this function is called only during * rebuild path and not while reset is in progress. @@ -2970,28 +3455,50 @@ static int ice_req_irq_msix_misc(struct ice_pf *pf) goto skip_req_irq; /* reserve one vector in irq_tracker for misc interrupts */ - oicr_idx = ice_get_res(pf, pf->irq_tracker, 1, ICE_RES_MISC_VEC_ID); - if (oicr_idx < 0) - return oicr_idx; + irq = ice_alloc_irq(pf, false); + if (irq.index < 0) + return irq.index; + + pf->oicr_irq = irq; + err = devm_request_threaded_irq(dev, pf->oicr_irq.virq, ice_misc_intr, + ice_misc_intr_thread_fn, 0, + pf->int_name, pf); + if (err) { + dev_err(dev, "devm_request_threaded_irq for %s failed: %d\n", + pf->int_name, err); + ice_free_irq(pf, pf->oicr_irq); + return err; + } + + /* reserve one vector in irq_tracker for ll_ts interrupt */ + if (!pf->hw.dev_caps.ts_dev_info.ts_ll_int_read) + goto skip_req_irq; - pf->num_avail_sw_msix -= 1; - pf->oicr_idx = (u16)oicr_idx; + irq = ice_alloc_irq(pf, false); + if (irq.index < 0) + return irq.index; - err = devm_request_irq(dev, pf->msix_entries[pf->oicr_idx].vector, - ice_misc_intr, 0, pf->int_name, pf); + pf->ll_ts_irq = irq; + err = devm_request_irq(dev, pf->ll_ts_irq.virq, ice_ll_ts_intr, 0, + pf->int_name_ll_ts, pf); if (err) { dev_err(dev, "devm_request_irq for %s failed: %d\n", - pf->int_name, err); - ice_free_res(pf->irq_tracker, 1, ICE_RES_MISC_VEC_ID); - pf->num_avail_sw_msix += 1; + pf->int_name_ll_ts, err); + ice_free_irq(pf, pf->ll_ts_irq); return err; } skip_req_irq: ice_ena_misc_vector(pf); - ice_ena_ctrlq_interrupts(hw, pf->oicr_idx); - wr32(hw, GLINT_ITR(ICE_RX_ITR, pf->oicr_idx), + ice_ena_ctrlq_interrupts(hw, pf->oicr_irq.index); + /* This enables LL TS interrupt */ + pf_intr_start_offset = rd32(hw, PFINT_ALLOC) & PFINT_ALLOC_FIRST; + if (pf->hw.dev_caps.ts_dev_info.ts_ll_int_read) + wr32(hw, PFINT_SB_CTL, + ((pf->ll_ts_irq.index + pf_intr_start_offset) & + PFINT_SB_CTL_MSIX_INDX_M) | PFINT_SB_CTL_CAUSE_ENA_M); + wr32(hw, GLINT_ITR(ICE_RX_ITR, pf->oicr_irq.index), ITR_REG_ALIGN(ICE_ITR_8K) >> ICE_ITR_GRAN_S); ice_flush(hw); @@ -3001,31 +3508,12 @@ skip_req_irq: } /** - * ice_napi_add - register NAPI handler for the VSI - * @vsi: VSI for which NAPI handler is to be registered - * - * This function is only called in the driver's load path. Registering the NAPI - * handler is done in ice_vsi_alloc_q_vector() for all other cases (i.e. resume, - * reset/rebuild, etc.) - */ -static void ice_napi_add(struct ice_vsi *vsi) -{ - int v_idx; - - if (!vsi->netdev) - return; - - ice_for_each_q_vector(vsi, v_idx) - netif_napi_add(vsi->netdev, &vsi->q_vectors[v_idx]->napi, - ice_napi_poll, NAPI_POLL_WEIGHT); -} - -/** * ice_set_ops - set netdev and ethtools ops for the given netdev - * @netdev: netdev instance + * @vsi: the VSI associated with the new netdev */ -static void ice_set_ops(struct net_device *netdev) +static void ice_set_ops(struct ice_vsi *vsi) { + struct net_device *netdev = vsi->netdev; struct ice_pf *pf = ice_netdev_to_pf(netdev); if (ice_is_safe_mode(pf)) { @@ -3036,16 +3524,26 @@ static void ice_set_ops(struct net_device *netdev) netdev->netdev_ops = &ice_netdev_ops; netdev->udp_tunnel_nic_info = &pf->hw.udp_tunnel_nic; + netdev->xdp_metadata_ops = &ice_xdp_md_ops; ice_set_ethtool_ops(netdev); + + if (vsi->type != ICE_VSI_PF) + return; + + netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_XSK_ZEROCOPY | + NETDEV_XDP_ACT_RX_SG; + netdev->xdp_zc_max_segs = ICE_MAX_BUF_TXD; } /** * ice_set_netdev_features - set features for the given netdev * @netdev: netdev instance */ -static void ice_set_netdev_features(struct net_device *netdev) +void ice_set_netdev_features(struct net_device *netdev) { struct ice_pf *pf = ice_netdev_to_pf(netdev); + bool is_dvm_ena = ice_is_dvm_ena(&pf->hw); netdev_features_t csumo_features; netdev_features_t vlano_features; netdev_features_t dflt_features; @@ -3072,6 +3570,10 @@ static void ice_set_netdev_features(struct net_device *netdev) NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX; + /* Enable CTAG/STAG filtering by default in Double VLAN Mode (DVM) */ + if (is_dvm_ena) + vlano_features |= NETIF_F_HW_VLAN_STAG_FILTER; + tso_features = NETIF_F_TSO | NETIF_F_TSO_ECN | NETIF_F_TSO6 | @@ -3091,62 +3593,46 @@ static void ice_set_netdev_features(struct net_device *netdev) vlano_features | tso_features; /* add support for HW_CSUM on packets with MPLS header */ - netdev->mpls_features = NETIF_F_HW_CSUM; + netdev->mpls_features = NETIF_F_HW_CSUM | + NETIF_F_TSO | + NETIF_F_TSO6; /* enable features */ netdev->features |= netdev->hw_features; + + netdev->hw_features |= NETIF_F_HW_TC; + netdev->hw_features |= NETIF_F_LOOPBACK; + /* encap and VLAN devices inherit default, csumo and tso features */ netdev->hw_enc_features |= dflt_features | csumo_features | tso_features; netdev->vlan_features |= dflt_features | csumo_features | tso_features; -} -/** - * ice_cfg_netdev - Allocate, configure and register a netdev - * @vsi: the VSI associated with the new netdev - * - * Returns 0 on success, negative value on failure - */ -static int ice_cfg_netdev(struct ice_vsi *vsi) -{ - struct ice_netdev_priv *np; - struct net_device *netdev; - u8 mac_addr[ETH_ALEN]; - - netdev = alloc_etherdev_mqs(sizeof(*np), vsi->alloc_txq, - vsi->alloc_rxq); - if (!netdev) - return -ENOMEM; - - set_bit(ICE_VSI_NETDEV_ALLOCD, vsi->state); - vsi->netdev = netdev; - np = netdev_priv(netdev); - np->vsi = vsi; - - ice_set_netdev_features(netdev); - - ice_set_ops(netdev); - - if (vsi->type == ICE_VSI_PF) { - SET_NETDEV_DEV(netdev, ice_pf_to_dev(vsi->back)); - ether_addr_copy(mac_addr, vsi->port_info->mac.perm_addr); - ether_addr_copy(netdev->dev_addr, mac_addr); - ether_addr_copy(netdev->perm_addr, mac_addr); - } - - netdev->priv_flags |= IFF_UNICAST_FLT; + /* advertise support but don't enable by default since only one type of + * VLAN offload can be enabled at a time (i.e. CTAG or STAG). When one + * type turns on the other has to be turned off. This is enforced by the + * ice_fix_features() ndo callback. + */ + if (is_dvm_ena) + netdev->hw_features |= NETIF_F_HW_VLAN_STAG_RX | + NETIF_F_HW_VLAN_STAG_TX; - /* Setup netdev TC information */ - ice_vsi_cfg_netdev_tc(vsi, vsi->tc_cfg.ena_tc); + /* Leave CRC / FCS stripping enabled by default, but allow the value to + * be changed at runtime + */ + netdev->hw_features |= NETIF_F_RXFCS; - /* setup watchdog timeout value to be 5 second */ - netdev->watchdog_timeo = 5 * HZ; + /* Allow core to manage IRQs affinity */ + netif_set_affinity_auto(netdev); - netdev->min_mtu = ETH_MIN_MTU; - netdev->max_mtu = ICE_MAX_MTU; + /* Mutual exclusivity for TSO and GCS is enforced by the set features + * ndo callback. + */ + if (ice_is_feature_supported(pf, ICE_F_GCS)) + netdev->hw_features |= NETIF_F_HW_CSUM; - return 0; + netif_set_tso_max_size(netdev, ICE_MAX_TSO_SIZE); } /** @@ -3174,7 +3660,27 @@ void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size) static struct ice_vsi * ice_pf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) { - return ice_vsi_setup(pf, pi, ICE_VSI_PF, ICE_INVAL_VFID); + struct ice_vsi_cfg_params params = {}; + + params.type = ICE_VSI_PF; + params.port_info = pi; + params.flags = ICE_VSI_FLAG_INIT; + + return ice_vsi_setup(pf, ¶ms); +} + +static struct ice_vsi * +ice_chnl_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, + struct ice_channel *ch) +{ + struct ice_vsi_cfg_params params = {}; + + params.type = ICE_VSI_CHNL; + params.port_info = pi; + params.ch = ch; + params.flags = ICE_VSI_FLAG_INIT; + + return ice_vsi_setup(pf, ¶ms); } /** @@ -3188,7 +3694,13 @@ ice_pf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) static struct ice_vsi * ice_ctrl_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) { - return ice_vsi_setup(pf, pi, ICE_VSI_CTRL, ICE_INVAL_VFID); + struct ice_vsi_cfg_params params = {}; + + params.type = ICE_VSI_CTRL; + params.port_info = pi; + params.flags = ICE_VSI_FLAG_INIT; + + return ice_vsi_setup(pf, ¶ms); } /** @@ -3202,42 +3714,73 @@ ice_ctrl_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) struct ice_vsi * ice_lb_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) { - return ice_vsi_setup(pf, pi, ICE_VSI_LB, ICE_INVAL_VFID); + struct ice_vsi_cfg_params params = {}; + + params.type = ICE_VSI_LB; + params.port_info = pi; + params.flags = ICE_VSI_FLAG_INIT; + + return ice_vsi_setup(pf, ¶ms); } /** * ice_vlan_rx_add_vid - Add a VLAN ID filter to HW offload * @netdev: network interface to be adjusted - * @proto: unused protocol + * @proto: VLAN TPID * @vid: VLAN ID to be added * * net_device_ops implementation for adding VLAN IDs */ -static int -ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto, - u16 vid) +int ice_vlan_rx_add_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi_vlan_ops *vlan_ops; struct ice_vsi *vsi = np->vsi; + struct ice_vlan vlan; int ret; /* VLAN 0 is added by default during load/reset */ if (!vid) return 0; - /* Enable VLAN pruning when a VLAN other than 0 is added */ - if (!ice_vsi_is_vlan_pruning_ena(vsi)) { - ret = ice_cfg_vlan_pruning(vsi, true, false); + while (test_and_set_bit(ICE_CFG_BUSY, vsi->state)) + usleep_range(1000, 2000); + + /* Add multicast promisc rule for the VLAN ID to be added if + * all-multicast is currently enabled. + */ + if (vsi->current_netdev_flags & IFF_ALLMULTI) { + ret = ice_fltr_set_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_VLAN_PROMISC_BITS, + vid); if (ret) - return ret; + goto finish; } + vlan_ops = ice_get_compat_vsi_vlan_ops(vsi); + /* Add a switch rule for this VLAN ID so its corresponding VLAN tagged * packets aren't pruned by the device's internal switch on Rx */ - ret = ice_vsi_add_vlan(vsi, vid, ICE_FWD_TO_VSI); - if (!ret) - set_bit(ICE_VSI_VLAN_FLTR_CHANGED, vsi->state); + vlan = ICE_VLAN(be16_to_cpu(proto), vid, 0); + ret = vlan_ops->add_vlan(vsi, &vlan); + if (ret) + goto finish; + + /* If all-multicast is currently enabled and this VLAN ID is only one + * besides VLAN-0 we have to update look-up type of multicast promisc + * rule for VLAN-0 from ICE_SW_LKUP_PROMISC to ICE_SW_LKUP_PROMISC_VLAN. + */ + if ((vsi->current_netdev_flags & IFF_ALLMULTI) && + ice_vsi_num_non_zero_vlans(vsi) == 1) { + ice_fltr_clear_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_PROMISC_BITS, 0); + ice_fltr_set_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_VLAN_PROMISC_BITS, 0); + } + +finish: + clear_bit(ICE_CFG_BUSY, vsi->state); return ret; } @@ -3245,102 +3788,112 @@ ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto, /** * ice_vlan_rx_kill_vid - Remove a VLAN ID filter from HW offload * @netdev: network interface to be adjusted - * @proto: unused protocol + * @proto: VLAN TPID * @vid: VLAN ID to be removed * * net_device_ops implementation for removing VLAN IDs */ -static int -ice_vlan_rx_kill_vid(struct net_device *netdev, __always_unused __be16 proto, - u16 vid) +int ice_vlan_rx_kill_vid(struct net_device *netdev, __be16 proto, u16 vid) { struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi_vlan_ops *vlan_ops; struct ice_vsi *vsi = np->vsi; + struct ice_vlan vlan; int ret; /* don't allow removal of VLAN 0 */ if (!vid) return 0; - /* Make sure ice_vsi_kill_vlan is successful before updating VLAN + while (test_and_set_bit(ICE_CFG_BUSY, vsi->state)) + usleep_range(1000, 2000); + + ret = ice_clear_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_VLAN_PROMISC_BITS, vid); + if (ret) { + netdev_err(netdev, "Error clearing multicast promiscuous mode on VSI %i\n", + vsi->vsi_num); + vsi->current_netdev_flags |= IFF_ALLMULTI; + } + + vlan_ops = ice_get_compat_vsi_vlan_ops(vsi); + + /* Make sure VLAN delete is successful before updating VLAN * information */ - ret = ice_vsi_kill_vlan(vsi, vid); + vlan = ICE_VLAN(be16_to_cpu(proto), vid, 0); + ret = vlan_ops->del_vlan(vsi, &vlan); if (ret) - return ret; + goto finish; + + /* Remove multicast promisc rule for the removed VLAN ID if + * all-multicast is enabled. + */ + if (vsi->current_netdev_flags & IFF_ALLMULTI) + ice_fltr_clear_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_VLAN_PROMISC_BITS, vid); + + if (!ice_vsi_has_non_zero_vlans(vsi)) { + /* Update look-up type of multicast promisc rule for VLAN 0 + * from ICE_SW_LKUP_PROMISC_VLAN to ICE_SW_LKUP_PROMISC when + * all-multicast is enabled and VLAN 0 is the only VLAN rule. + */ + if (vsi->current_netdev_flags & IFF_ALLMULTI) { + ice_fltr_clear_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_VLAN_PROMISC_BITS, + 0); + ice_fltr_set_vsi_promisc(&vsi->back->hw, vsi->idx, + ICE_MCAST_PROMISC_BITS, 0); + } + } - /* Disable pruning when VLAN 0 is the only VLAN rule */ - if (vsi->num_vlan == 1 && ice_vsi_is_vlan_pruning_ena(vsi)) - ret = ice_cfg_vlan_pruning(vsi, false, false); +finish: + clear_bit(ICE_CFG_BUSY, vsi->state); - set_bit(ICE_VSI_VLAN_FLTR_CHANGED, vsi->state); return ret; } /** - * ice_setup_pf_sw - Setup the HW switch on startup or after reset - * @pf: board private structure - * - * Returns 0 on success, negative value on failure + * ice_rep_indr_tc_block_unbind + * @cb_priv: indirection block private data */ -static int ice_setup_pf_sw(struct ice_pf *pf) +static void ice_rep_indr_tc_block_unbind(void *cb_priv) { - struct ice_vsi *vsi; - int status = 0; + struct ice_indr_block_priv *indr_priv = cb_priv; - if (ice_is_reset_in_progress(pf->state)) - return -EBUSY; - - vsi = ice_pf_vsi_setup(pf, pf->hw.port_info); - if (!vsi) - return -ENOMEM; - - status = ice_cfg_netdev(vsi); - if (status) { - status = -ENODEV; - goto unroll_vsi_setup; - } - /* netdev has to be configured before setting frame size */ - ice_vsi_cfg_frame_size(vsi); - - /* Setup DCB netlink interface */ - ice_dcbnl_setup(vsi); + list_del(&indr_priv->list); + kfree(indr_priv); +} - /* registering the NAPI handler requires both the queues and - * netdev to be created, which are done in ice_pf_vsi_setup() - * and ice_cfg_netdev() respectively - */ - ice_napi_add(vsi); +/** + * ice_tc_indir_block_unregister - Unregister TC indirect block notifications + * @vsi: VSI struct which has the netdev + */ +static void ice_tc_indir_block_unregister(struct ice_vsi *vsi) +{ + struct ice_netdev_priv *np = netdev_priv(vsi->netdev); - status = ice_set_cpu_rx_rmap(vsi); - if (status) { - dev_err(ice_pf_to_dev(pf), "Failed to set CPU Rx map VSI %d error %d\n", - vsi->vsi_num, status); - status = -EINVAL; - goto unroll_napi_add; - } - status = ice_init_mac_fltr(pf); - if (status) - goto free_cpu_rx_map; + flow_indr_dev_unregister(ice_indr_setup_tc_cb, np, + ice_rep_indr_tc_block_unbind); +} - return status; +/** + * ice_tc_indir_block_register - Register TC indirect block notifications + * @vsi: VSI struct which has the netdev + * + * Returns 0 on success, negative value on failure + */ +static int ice_tc_indir_block_register(struct ice_vsi *vsi) +{ + struct ice_netdev_priv *np; -free_cpu_rx_map: - ice_free_cpu_rx_rmap(vsi); + if (!vsi || !vsi->netdev) + return -EINVAL; -unroll_napi_add: - if (vsi) { - ice_napi_del(vsi); - if (vsi->netdev) { - clear_bit(ICE_VSI_NETDEV_ALLOCD, vsi->state); - free_netdev(vsi->netdev); - vsi->netdev = NULL; - } - } + np = netdev_priv(vsi->netdev); -unroll_vsi_setup: - ice_vsi_release(vsi); - return status; + INIT_LIST_HEAD(&np->tc_indr_block_priv_list); + return flow_indr_dev_register(ice_indr_setup_tc_cb, np); } /** @@ -3387,12 +3940,16 @@ u16 ice_get_avail_rxq_count(struct ice_pf *pf) * ice_deinit_pf - Unrolls initialziations done by ice_init_pf * @pf: board private structure to initialize */ -static void ice_deinit_pf(struct ice_pf *pf) +void ice_deinit_pf(struct ice_pf *pf) { - ice_service_task_stop(pf); + /* note that we unroll also on ice_init_pf() failure here */ + + mutex_destroy(&pf->lag_mutex); + mutex_destroy(&pf->adev_mutex); mutex_destroy(&pf->sw_mutex); mutex_destroy(&pf->tc_mutex); mutex_destroy(&pf->avail_q_mutex); + mutex_destroy(&pf->vfs.table_lock); if (pf->avail_txqs) { bitmap_free(pf->avail_txqs); @@ -3404,8 +3961,19 @@ static void ice_deinit_pf(struct ice_pf *pf) pf->avail_rxqs = NULL; } + if (pf->txtime_txqs) { + bitmap_free(pf->txtime_txqs); + pf->txtime_txqs = NULL; + } + if (pf->ptp.clock) ptp_clock_unregister(pf->ptp.clock); + + if (!xa_empty(&pf->irq_tracker.entries)) + ice_free_irq_msix_misc(pf); + + xa_destroy(&pf->dyn_ports); + xa_destroy(&pf->sf_nums); } /** @@ -3417,19 +3985,16 @@ static void ice_set_pf_caps(struct ice_pf *pf) struct ice_hw_func_caps *func_caps = &pf->hw.func_caps; clear_bit(ICE_FLAG_RDMA_ENA, pf->flags); - clear_bit(ICE_FLAG_AUX_ENA, pf->flags); - if (func_caps->common_cap.rdma) { + if (func_caps->common_cap.rdma) set_bit(ICE_FLAG_RDMA_ENA, pf->flags); - set_bit(ICE_FLAG_AUX_ENA, pf->flags); - } clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); if (func_caps->common_cap.dcb) set_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); clear_bit(ICE_FLAG_SRIOV_CAPABLE, pf->flags); if (func_caps->common_cap.sr_iov_1_1) { set_bit(ICE_FLAG_SRIOV_CAPABLE, pf->flags); - pf->num_vfs_supported = min_t(int, func_caps->num_allocd_vfs, - ICE_MAX_VF_COUNT); + pf->vfs.num_supported = min_t(int, func_caps->num_allocd_vfs, + ICE_MAX_SRIOV_VFS); } clear_bit(ICE_FLAG_RSS_ENA, pf->flags); if (func_caps->common_cap.rss_table_size) @@ -3460,16 +4025,30 @@ static void ice_set_pf_caps(struct ice_pf *pf) pf->max_pf_rxqs = func_caps->common_cap.num_rxq; } +void ice_start_service_task(struct ice_pf *pf) +{ + timer_setup(&pf->serv_tmr, ice_service_timer, 0); + pf->serv_tmr_period = HZ; + INIT_WORK(&pf->serv_task, ice_service_task); + clear_bit(ICE_SERVICE_SCHED, pf->state); +} + /** * ice_init_pf - Initialize general software structures (struct ice_pf) * @pf: board private structure to initialize + * Return: 0 on success, negative errno otherwise. */ -static int ice_init_pf(struct ice_pf *pf) +int ice_init_pf(struct ice_pf *pf) { - ice_set_pf_caps(pf); + struct udp_tunnel_nic_info *udp_tunnel_nic = &pf->hw.udp_tunnel_nic; + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + int err = -ENOMEM; mutex_init(&pf->sw_mutex); mutex_init(&pf->tc_mutex); + mutex_init(&pf->adev_mutex); + mutex_init(&pf->lag_mutex); INIT_HLIST_HEAD(&pf->aq_wait_list); spin_lock_init(&pf->aq_wait_lock); @@ -3477,221 +4056,49 @@ static int ice_init_pf(struct ice_pf *pf) init_waitqueue_head(&pf->reset_wait_queue); - /* setup service timer and periodic service task */ - timer_setup(&pf->serv_tmr, ice_service_timer, 0); - pf->serv_tmr_period = HZ; - INIT_WORK(&pf->serv_task, ice_service_task); - clear_bit(ICE_SERVICE_SCHED, pf->state); - mutex_init(&pf->avail_q_mutex); - pf->avail_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); - if (!pf->avail_txqs) - return -ENOMEM; - - pf->avail_rxqs = bitmap_zalloc(pf->max_pf_rxqs, GFP_KERNEL); - if (!pf->avail_rxqs) { - devm_kfree(ice_pf_to_dev(pf), pf->avail_txqs); - pf->avail_txqs = NULL; - return -ENOMEM; - } - - return 0; -} - -/** - * ice_ena_msix_range - Request a range of MSIX vectors from the OS - * @pf: board private structure - * - * compute the number of MSIX vectors required (v_budget) and request from - * the OS. Return the number of vectors reserved or negative on failure - */ -static int ice_ena_msix_range(struct ice_pf *pf) -{ - int num_cpus, v_left, v_actual, v_other, v_budget = 0; - struct device *dev = ice_pf_to_dev(pf); - int needed, err, i; - - v_left = pf->hw.func_caps.common_cap.num_msix_vectors; - num_cpus = num_online_cpus(); - - /* reserve for LAN miscellaneous handler */ - needed = ICE_MIN_LAN_OICR_MSIX; - if (v_left < needed) - goto no_hw_vecs_left_err; - v_budget += needed; - v_left -= needed; - - /* reserve for flow director */ - if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) { - needed = ICE_FDIR_MSIX; - if (v_left < needed) - goto no_hw_vecs_left_err; - v_budget += needed; - v_left -= needed; - } - - /* total used for non-traffic vectors */ - v_other = v_budget; - - /* reserve vectors for LAN traffic */ - needed = num_cpus; - if (v_left < needed) - goto no_hw_vecs_left_err; - pf->num_lan_msix = needed; - v_budget += needed; - v_left -= needed; - - /* reserve vectors for RDMA auxiliary driver */ - if (test_bit(ICE_FLAG_RDMA_ENA, pf->flags)) { - needed = num_cpus + ICE_RDMA_NUM_AEQ_MSIX; - if (v_left < needed) - goto no_hw_vecs_left_err; - pf->num_rdma_msix = needed; - v_budget += needed; - v_left -= needed; - } - - pf->msix_entries = devm_kcalloc(dev, v_budget, - sizeof(*pf->msix_entries), GFP_KERNEL); - if (!pf->msix_entries) { - err = -ENOMEM; - goto exit_err; - } - - for (i = 0; i < v_budget; i++) - pf->msix_entries[i].entry = i; - - /* actually reserve the vectors */ - v_actual = pci_enable_msix_range(pf->pdev, pf->msix_entries, - ICE_MIN_MSIX, v_budget); - if (v_actual < 0) { - dev_err(dev, "unable to reserve MSI-X vectors\n"); - err = v_actual; - goto msix_err; - } - - if (v_actual < v_budget) { - dev_warn(dev, "not enough OS MSI-X vectors. requested = %d, obtained = %d\n", - v_budget, v_actual); - - if (v_actual < ICE_MIN_MSIX) { - /* error if we can't get minimum vectors */ - pci_disable_msix(pf->pdev); - err = -ERANGE; - goto msix_err; - } else { - int v_remain = v_actual - v_other; - int v_rdma = 0, v_min_rdma = 0; - - if (test_bit(ICE_FLAG_RDMA_ENA, pf->flags)) { - /* Need at least 1 interrupt in addition to - * AEQ MSIX - */ - v_rdma = ICE_RDMA_NUM_AEQ_MSIX + 1; - v_min_rdma = ICE_MIN_RDMA_MSIX; - } - - if (v_actual == ICE_MIN_MSIX || - v_remain < ICE_MIN_LAN_TXRX_MSIX + v_min_rdma) { - dev_warn(dev, "Not enough MSI-X vectors to support RDMA.\n"); - clear_bit(ICE_FLAG_RDMA_ENA, pf->flags); - - pf->num_rdma_msix = 0; - pf->num_lan_msix = ICE_MIN_LAN_TXRX_MSIX; - } else if ((v_remain < ICE_MIN_LAN_TXRX_MSIX + v_rdma) || - (v_remain - v_rdma < v_rdma)) { - /* Support minimum RDMA and give remaining - * vectors to LAN MSIX - */ - pf->num_rdma_msix = v_min_rdma; - pf->num_lan_msix = v_remain - v_min_rdma; - } else { - /* Split remaining MSIX with RDMA after - * accounting for AEQ MSIX - */ - pf->num_rdma_msix = (v_remain - ICE_RDMA_NUM_AEQ_MSIX) / 2 + - ICE_RDMA_NUM_AEQ_MSIX; - pf->num_lan_msix = v_remain - pf->num_rdma_msix; - } - - dev_notice(dev, "Enabled %d MSI-X vectors for LAN traffic.\n", - pf->num_lan_msix); - - if (test_bit(ICE_FLAG_RDMA_ENA, pf->flags)) - dev_notice(dev, "Enabled %d MSI-X vectors for RDMA.\n", - pf->num_rdma_msix); - } - } - - return v_actual; - -msix_err: - devm_kfree(dev, pf->msix_entries); - goto exit_err; - -no_hw_vecs_left_err: - dev_err(dev, "not enough device MSI-X vectors. requested = %d, available = %d\n", - needed, v_left); - err = -ERANGE; -exit_err: - pf->num_rdma_msix = 0; - pf->num_lan_msix = 0; - return err; -} -/** - * ice_dis_msix - Disable MSI-X interrupt setup in OS - * @pf: board private structure - */ -static void ice_dis_msix(struct ice_pf *pf) -{ - pci_disable_msix(pf->pdev); - devm_kfree(ice_pf_to_dev(pf), pf->msix_entries); - pf->msix_entries = NULL; -} - -/** - * ice_clear_interrupt_scheme - Undo things done by ice_init_interrupt_scheme - * @pf: board private structure - */ -static void ice_clear_interrupt_scheme(struct ice_pf *pf) -{ - ice_dis_msix(pf); - - if (pf->irq_tracker) { - devm_kfree(ice_pf_to_dev(pf), pf->irq_tracker); - pf->irq_tracker = NULL; - } -} - -/** - * ice_init_interrupt_scheme - Determine proper interrupt scheme - * @pf: board private structure to initialize - */ -static int ice_init_interrupt_scheme(struct ice_pf *pf) -{ - int vectors; + mutex_init(&pf->vfs.table_lock); + hash_init(pf->vfs.table); + if (ice_is_feature_supported(pf, ICE_F_MBX_LIMIT)) + wr32(&pf->hw, E830_MBX_PF_IN_FLIGHT_VF_MSGS_THRESH, + ICE_MBX_OVERFLOW_WATERMARK); + else + ice_mbx_init_snapshot(&pf->hw); - vectors = ice_ena_msix_range(pf); + xa_init(&pf->dyn_ports); + xa_init(&pf->sf_nums); - if (vectors < 0) - return vectors; + pf->avail_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); + pf->avail_rxqs = bitmap_zalloc(pf->max_pf_rxqs, GFP_KERNEL); + pf->txtime_txqs = bitmap_zalloc(pf->max_pf_txqs, GFP_KERNEL); + if (!pf->avail_txqs || !pf->avail_rxqs || !pf->txtime_txqs) + goto undo_init; + + udp_tunnel_nic->set_port = ice_udp_tunnel_set_port; + udp_tunnel_nic->unset_port = ice_udp_tunnel_unset_port; + udp_tunnel_nic->shared = &hw->udp_tunnel_shared; + udp_tunnel_nic->tables[0].n_entries = hw->tnl.valid_count[TNL_VXLAN]; + udp_tunnel_nic->tables[0].tunnel_types = UDP_TUNNEL_TYPE_VXLAN; + udp_tunnel_nic->tables[1].n_entries = hw->tnl.valid_count[TNL_GENEVE]; + udp_tunnel_nic->tables[1].tunnel_types = UDP_TUNNEL_TYPE_GENEVE; - /* set up vector assignment tracking */ - pf->irq_tracker = devm_kzalloc(ice_pf_to_dev(pf), - struct_size(pf->irq_tracker, list, vectors), - GFP_KERNEL); - if (!pf->irq_tracker) { - ice_dis_msix(pf); - return -ENOMEM; + /* In case of MSIX we are going to setup the misc vector right here + * to handle admin queue events etc. In case of legacy and MSI + * the misc functionality and queue processing is combined in + * the same vector and that gets setup at open. + */ + err = ice_req_irq_msix_misc(pf); + if (err) { + dev_err(dev, "setup of misc vector failed: %d\n", err); + goto undo_init; } - /* populate SW interrupts pool with number of OS granted IRQs. */ - pf->num_avail_sw_msix = (u16)vectors; - pf->irq_tracker->num_entries = (u16)vectors; - pf->irq_tracker->end = pf->irq_tracker->num_entries; - return 0; +undo_init: + /* deinit handles half-initialized pf just fine */ + ice_deinit_pf(pf); + return err; } /** @@ -3719,15 +4126,16 @@ bool ice_is_wol_supported(struct ice_hw *hw) * @vsi: VSI being changed * @new_rx: new number of Rx queues * @new_tx: new number of Tx queues + * @locked: is adev device_lock held * * Only change the number of queues if new_tx, or new_rx is non-0. * * Returns 0 on success. */ -int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx) +int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx, bool locked) { struct ice_pf *pf = vsi->back; - int err = 0, timeout = 50; + int i, err = 0, timeout = 50; if (!new_rx && !new_tx) return -EINVAL; @@ -3746,15 +4154,32 @@ int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx) /* set for the next time the netdev is started */ if (!netif_running(vsi->netdev)) { - ice_vsi_rebuild(vsi, false); + err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT); + if (err) + goto rebuild_err; dev_dbg(ice_pf_to_dev(pf), "Link is down, queue count change happens when link is brought up\n"); goto done; } ice_vsi_close(vsi); - ice_vsi_rebuild(vsi, false); - ice_pf_dcb_recfg(pf); + err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT); + if (err) + goto rebuild_err; + + ice_for_each_traffic_class(i) { + if (vsi->tc_cfg.ena_tc & BIT(i)) + netdev_set_tc_queue(vsi->netdev, + vsi->tc_cfg.tc_info[i].netdev_tc, + vsi->tc_cfg.tc_info[i].qcount_tx, + vsi->tc_cfg.tc_info[i].qoffset); + } + ice_pf_dcb_recfg(pf, locked); ice_vsi_open(vsi); + goto done; + +rebuild_err: + dev_err(ice_pf_to_dev(pf), "Error during VSI rebuild: %d. Unload and reload the driver.\n", + err); done: clear_bit(ICE_CFG_BUSY, pf->state); return err; @@ -3771,8 +4196,8 @@ static void ice_set_safe_mode_vlan_cfg(struct ice_pf *pf) { struct ice_vsi *vsi = ice_get_main_vsi(pf); struct ice_vsi_ctx *ctxt; - enum ice_status status; struct ice_hw *hw; + int status; if (!vsi) return; @@ -3797,18 +4222,17 @@ static void ice_set_safe_mode_vlan_cfg(struct ice_pf *pf) ctxt->info.sw_flags2 &= ~ICE_AQ_VSI_SW_FLAG_RX_VLAN_PRUNE_ENA; /* allow all VLANs on Tx and don't strip on Rx */ - ctxt->info.vlan_flags = ICE_AQ_VSI_VLAN_MODE_ALL | - ICE_AQ_VSI_VLAN_EMOD_NOTHING; + ctxt->info.inner_vlan_flags = ICE_AQ_VSI_INNER_VLAN_TX_MODE_ALL | + ICE_AQ_VSI_INNER_VLAN_EMODE_NOTHING; status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_err(ice_pf_to_dev(vsi->back), "Failed to update VSI for safe mode VLANs, err %s aq_err %s\n", - ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); + dev_err(ice_pf_to_dev(vsi->back), "Failed to update VSI for safe mode VLANs, err %d aq_err %s\n", + status, libie_aq_str(hw->adminq.sq_last_status)); } else { vsi->info.sec_flags = ctxt->info.sec_flags; vsi->info.sw_flags2 = ctxt->info.sw_flags2; - vsi->info.vlan_flags = ctxt->info.vlan_flags; + vsi->info.inner_vlan_flags = ctxt->info.inner_vlan_flags; } kfree(ctxt); @@ -3817,108 +4241,80 @@ static void ice_set_safe_mode_vlan_cfg(struct ice_pf *pf) /** * ice_log_pkg_init - log result of DDP package load * @hw: pointer to hardware info - * @status: status of package load + * @state: state of package load */ -static void -ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) +static void ice_log_pkg_init(struct ice_hw *hw, enum ice_ddp_state state) { - struct ice_pf *pf = (struct ice_pf *)hw->back; - struct device *dev = ice_pf_to_dev(pf); + struct ice_pf *pf = hw->back; + struct device *dev; - switch (*status) { - case ICE_SUCCESS: - /* The package download AdminQ command returned success because - * this download succeeded or ICE_ERR_AQ_NO_WORK since there is - * already a package loaded on the device. - */ - if (hw->pkg_ver.major == hw->active_pkg_ver.major && - hw->pkg_ver.minor == hw->active_pkg_ver.minor && - hw->pkg_ver.update == hw->active_pkg_ver.update && - hw->pkg_ver.draft == hw->active_pkg_ver.draft && - !memcmp(hw->pkg_name, hw->active_pkg_name, - sizeof(hw->pkg_name))) { - if (hw->pkg_dwnld_status == ICE_AQ_RC_EEXIST) - dev_info(dev, "DDP package already present on device: %s version %d.%d.%d.%d\n", - hw->active_pkg_name, - hw->active_pkg_ver.major, - hw->active_pkg_ver.minor, - hw->active_pkg_ver.update, - hw->active_pkg_ver.draft); - else - dev_info(dev, "The DDP package was successfully loaded: %s version %d.%d.%d.%d\n", - hw->active_pkg_name, - hw->active_pkg_ver.major, - hw->active_pkg_ver.minor, - hw->active_pkg_ver.update, - hw->active_pkg_ver.draft); - } else if (hw->active_pkg_ver.major != ICE_PKG_SUPP_VER_MAJ || - hw->active_pkg_ver.minor != ICE_PKG_SUPP_VER_MNR) { - dev_err(dev, "The device has a DDP package that is not supported by the driver. The device has package '%s' version %d.%d.x.x. The driver requires version %d.%d.x.x. Entering Safe Mode.\n", - hw->active_pkg_name, - hw->active_pkg_ver.major, - hw->active_pkg_ver.minor, - ICE_PKG_SUPP_VER_MAJ, ICE_PKG_SUPP_VER_MNR); - *status = ICE_ERR_NOT_SUPPORTED; - } else if (hw->active_pkg_ver.major == ICE_PKG_SUPP_VER_MAJ && - hw->active_pkg_ver.minor == ICE_PKG_SUPP_VER_MNR) { - dev_info(dev, "The driver could not load the DDP package file because a compatible DDP package is already present on the device. The device has package '%s' version %d.%d.%d.%d. The package file found by the driver: '%s' version %d.%d.%d.%d.\n", - hw->active_pkg_name, - hw->active_pkg_ver.major, - hw->active_pkg_ver.minor, - hw->active_pkg_ver.update, - hw->active_pkg_ver.draft, - hw->pkg_name, - hw->pkg_ver.major, - hw->pkg_ver.minor, - hw->pkg_ver.update, - hw->pkg_ver.draft); - } else { - dev_err(dev, "An unknown error occurred when loading the DDP package, please reboot the system. If the problem persists, update the NVM. Entering Safe Mode.\n"); - *status = ICE_ERR_NOT_SUPPORTED; - } + dev = ice_pf_to_dev(pf); + + switch (state) { + case ICE_DDP_PKG_SUCCESS: + dev_info(dev, "The DDP package was successfully loaded: %s version %d.%d.%d.%d\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + hw->active_pkg_ver.update, + hw->active_pkg_ver.draft); + break; + case ICE_DDP_PKG_SAME_VERSION_ALREADY_LOADED: + dev_info(dev, "DDP package already present on device: %s version %d.%d.%d.%d\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + hw->active_pkg_ver.update, + hw->active_pkg_ver.draft); + break; + case ICE_DDP_PKG_ALREADY_LOADED_NOT_SUPPORTED: + dev_err(dev, "The device has a DDP package that is not supported by the driver. The device has package '%s' version %d.%d.x.x. The driver requires version %d.%d.x.x. Entering Safe Mode.\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + ICE_PKG_SUPP_VER_MAJ, ICE_PKG_SUPP_VER_MNR); break; - case ICE_ERR_FW_DDP_MISMATCH: + case ICE_DDP_PKG_COMPATIBLE_ALREADY_LOADED: + dev_info(dev, "The driver could not load the DDP package file because a compatible DDP package is already present on the device. The device has package '%s' version %d.%d.%d.%d. The package file found by the driver: '%s' version %d.%d.%d.%d.\n", + hw->active_pkg_name, + hw->active_pkg_ver.major, + hw->active_pkg_ver.minor, + hw->active_pkg_ver.update, + hw->active_pkg_ver.draft, + hw->pkg_name, + hw->pkg_ver.major, + hw->pkg_ver.minor, + hw->pkg_ver.update, + hw->pkg_ver.draft); + break; + case ICE_DDP_PKG_FW_MISMATCH: dev_err(dev, "The firmware loaded on the device is not compatible with the DDP package. Please update the device's NVM. Entering safe mode.\n"); break; - case ICE_ERR_BUF_TOO_SHORT: - case ICE_ERR_CFG: + case ICE_DDP_PKG_INVALID_FILE: dev_err(dev, "The DDP package file is invalid. Entering Safe Mode.\n"); break; - case ICE_ERR_NOT_SUPPORTED: - /* Package File version not supported */ - if (hw->pkg_ver.major > ICE_PKG_SUPP_VER_MAJ || - (hw->pkg_ver.major == ICE_PKG_SUPP_VER_MAJ && - hw->pkg_ver.minor > ICE_PKG_SUPP_VER_MNR)) - dev_err(dev, "The DDP package file version is higher than the driver supports. Please use an updated driver. Entering Safe Mode.\n"); - else if (hw->pkg_ver.major < ICE_PKG_SUPP_VER_MAJ || - (hw->pkg_ver.major == ICE_PKG_SUPP_VER_MAJ && - hw->pkg_ver.minor < ICE_PKG_SUPP_VER_MNR)) - dev_err(dev, "The DDP package file version is lower than the driver supports. The driver requires version %d.%d.x.x. Please use an updated DDP Package file. Entering Safe Mode.\n", - ICE_PKG_SUPP_VER_MAJ, ICE_PKG_SUPP_VER_MNR); + case ICE_DDP_PKG_FILE_VERSION_TOO_HIGH: + dev_err(dev, "The DDP package file version is higher than the driver supports. Please use an updated driver. Entering Safe Mode.\n"); break; - case ICE_ERR_AQ_ERROR: - switch (hw->pkg_dwnld_status) { - case ICE_AQ_RC_ENOSEC: - case ICE_AQ_RC_EBADSIG: - dev_err(dev, "The DDP package could not be loaded because its signature is not valid. Please use a valid DDP Package. Entering Safe Mode.\n"); - return; - case ICE_AQ_RC_ESVN: - dev_err(dev, "The DDP Package could not be loaded because its security revision is too low. Please use an updated DDP Package. Entering Safe Mode.\n"); - return; - case ICE_AQ_RC_EBADMAN: - case ICE_AQ_RC_EBADBUF: - dev_err(dev, "An error occurred on the device while loading the DDP package. The device will be reset.\n"); - /* poll for reset to complete */ - if (ice_check_reset(hw)) - dev_err(dev, "Error resetting device. Please reload the driver\n"); - return; - default: - break; - } - fallthrough; + case ICE_DDP_PKG_FILE_VERSION_TOO_LOW: + dev_err(dev, "The DDP package file version is lower than the driver supports. The driver requires version %d.%d.x.x. Please use an updated DDP Package file. Entering Safe Mode.\n", + ICE_PKG_SUPP_VER_MAJ, ICE_PKG_SUPP_VER_MNR); + break; + case ICE_DDP_PKG_FILE_SIGNATURE_INVALID: + dev_err(dev, "The DDP package could not be loaded because its signature is not valid. Please use a valid DDP Package. Entering Safe Mode.\n"); + break; + case ICE_DDP_PKG_FILE_REVISION_TOO_LOW: + dev_err(dev, "The DDP Package could not be loaded because its security revision is too low. Please use an updated DDP Package. Entering Safe Mode.\n"); + break; + case ICE_DDP_PKG_LOAD_ERROR: + dev_err(dev, "An error occurred on the device while loading the DDP package. The device will be reset.\n"); + /* poll for reset to complete */ + if (ice_check_reset(hw)) + dev_err(dev, "Error resetting device. Please reload the driver\n"); + break; + case ICE_DDP_PKG_ERR: default: - dev_err(dev, "An unknown error (%d) occurred when loading the DDP package. Entering Safe Mode.\n", - *status); + dev_err(dev, "An unknown error occurred when loading the DDP package. Entering Safe Mode.\n"); break; } } @@ -3934,24 +4330,24 @@ ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) static void ice_load_pkg(const struct firmware *firmware, struct ice_pf *pf) { - enum ice_status status = ICE_ERR_PARAM; + enum ice_ddp_state state = ICE_DDP_PKG_ERR; struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; /* Load DDP Package */ if (firmware && !hw->pkg_copy) { - status = ice_copy_and_init_pkg(hw, firmware->data, - firmware->size); - ice_log_pkg_init(hw, &status); + state = ice_copy_and_init_pkg(hw, firmware->data, + firmware->size); + ice_log_pkg_init(hw, state); } else if (!firmware && hw->pkg_copy) { /* Reload package during rebuild after CORER/GLOBR reset */ - status = ice_init_pkg(hw, hw->pkg_copy, hw->pkg_size); - ice_log_pkg_init(hw, &status); + state = ice_init_pkg(hw, hw->pkg_copy, hw->pkg_size); + ice_log_pkg_init(hw, state); } else { dev_err(dev, "The DDP package file failed to load. Entering Safe Mode.\n"); } - if (status) { + if (!ice_is_init_pkg_successful(state)) { /* Safe Mode */ clear_bit(ICE_FLAG_ADV_FEATURES, pf->flags); return; @@ -3982,9 +4378,9 @@ static void ice_verify_cacheline_size(struct ice_pf *pf) * ice_send_version - update firmware with driver version * @pf: PF struct * - * Returns ICE_SUCCESS on success, else error code + * Returns 0 on success, else error code */ -static enum ice_status ice_send_version(struct ice_pf *pf) +static int ice_send_version(struct ice_pf *pf) { struct ice_driver_ver dv; @@ -4044,6 +4440,23 @@ err_vsi_open: return err; } +static void ice_deinit_fdir(struct ice_pf *pf) +{ + struct ice_vsi *vsi = ice_get_ctrl_vsi(pf); + + if (!vsi) + return; + + ice_vsi_manage_fdir(vsi, false); + ice_vsi_release(vsi); + if (pf->ctrl_vsi_idx != ICE_NO_VSI) { + pf->vsi[pf->ctrl_vsi_idx] = NULL; + pf->ctrl_vsi_idx = ICE_NO_VSI; + } + + mutex_destroy(&(&pf->hw)->fdir_fltr_lock); +} + /** * ice_get_opt_fw_name - return optional firmware file name or NULL * @pf: pointer to the PF instance @@ -4077,11 +4490,13 @@ static char *ice_get_opt_fw_name(struct ice_pf *pf) /** * ice_request_fw - Device initialization routine * @pf: pointer to the PF instance + * @firmware: double pointer to firmware struct + * + * Return: zero when successful, negative values otherwise. */ -static void ice_request_fw(struct ice_pf *pf) +static int ice_request_fw(struct ice_pf *pf, const struct firmware **firmware) { char *opt_fw_filename = ice_get_opt_fw_name(pf); - const struct firmware *firmware = NULL; struct device *dev = ice_pf_to_dev(pf); int err = 0; @@ -4090,29 +4505,126 @@ static void ice_request_fw(struct ice_pf *pf) * and warning messages for other errors. */ if (opt_fw_filename) { - err = firmware_request_nowarn(&firmware, opt_fw_filename, dev); - if (err) { - kfree(opt_fw_filename); - goto dflt_pkg_load; - } - - /* request for firmware was successful. Download to device */ - ice_load_pkg(firmware, pf); + err = firmware_request_nowarn(firmware, opt_fw_filename, dev); kfree(opt_fw_filename); - release_firmware(firmware); - return; + if (!err) + return err; + } + err = request_firmware(firmware, ICE_DDP_PKG_FILE, dev); + if (err) + dev_err(dev, "The DDP package file was not found or could not be read. Entering Safe Mode\n"); + + return err; +} + +/** + * ice_init_tx_topology - performs Tx topology initialization + * @hw: pointer to the hardware structure + * @firmware: pointer to firmware structure + * + * Return: zero when init was successful, negative values otherwise. + */ +static int +ice_init_tx_topology(struct ice_hw *hw, const struct firmware *firmware) +{ + u8 num_tx_sched_layers = hw->num_tx_sched_layers; + struct ice_pf *pf = hw->back; + struct device *dev; + int err; + + dev = ice_pf_to_dev(pf); + err = ice_cfg_tx_topo(hw, firmware->data, firmware->size); + if (!err) { + if (hw->num_tx_sched_layers > num_tx_sched_layers) + dev_info(dev, "Tx scheduling layers switching feature disabled\n"); + else + dev_info(dev, "Tx scheduling layers switching feature enabled\n"); + return 0; + } else if (err == -ENODEV) { + /* If we failed to re-initialize the device, we can no longer + * continue loading. + */ + dev_warn(dev, "Failed to initialize hardware after applying Tx scheduling configuration.\n"); + return err; + } else if (err == -EIO) { + dev_info(dev, "DDP package does not support Tx scheduling layers switching feature - please update to the latest DDP package and try again\n"); + return 0; + } else if (err == -EEXIST) { + return 0; } -dflt_pkg_load: - err = request_firmware(&firmware, ICE_DDP_PKG_FILE, dev); + /* Do not treat this as a fatal error. */ + dev_info(dev, "Failed to apply Tx scheduling configuration, err %pe\n", + ERR_PTR(err)); + return 0; +} + +/** + * ice_init_supported_rxdids - Initialize supported Rx descriptor IDs + * @hw: pointer to the hardware structure + * @pf: pointer to pf structure + * + * The pf->supported_rxdids bitmap is used to indicate to VFs which descriptor + * formats the PF hardware supports. The exact list of supported RXDIDs + * depends on the loaded DDP package. The IDs can be determined by reading the + * GLFLXP_RXDID_FLAGS register after the DDP package is loaded. + * + * Note that the legacy 32-byte RXDID 0 is always supported but is not listed + * in the DDP package. The 16-byte legacy descriptor is never supported by + * VFs. + */ +static void ice_init_supported_rxdids(struct ice_hw *hw, struct ice_pf *pf) +{ + pf->supported_rxdids = BIT(ICE_RXDID_LEGACY_1); + + for (int i = ICE_RXDID_FLEX_NIC; i < ICE_FLEX_DESC_RXDID_MAX_NUM; i++) { + u32 regval; + + regval = rd32(hw, GLFLXP_RXDID_FLAGS(i, 0)); + if ((regval >> GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_S) + & GLFLXP_RXDID_FLAGS_FLEXIFLAG_4N_M) + pf->supported_rxdids |= BIT(i); + } +} + +/** + * ice_init_ddp_config - DDP related configuration + * @hw: pointer to the hardware structure + * @pf: pointer to pf structure + * + * This function loads DDP file from the disk, then initializes Tx + * topology. At the end DDP package is loaded on the card. + * + * Return: zero when init was successful, negative values otherwise. + */ +static int ice_init_ddp_config(struct ice_hw *hw, struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + const struct firmware *firmware = NULL; + int err; + + err = ice_request_fw(pf, &firmware); if (err) { - dev_err(dev, "The DDP package file was not found or could not be read. Entering Safe Mode\n"); - return; + dev_err(dev, "Fail during requesting FW: %d\n", err); + return err; + } + + err = ice_init_tx_topology(hw, firmware); + if (err) { + dev_err(dev, "Fail during initialization of Tx topology: %d\n", + err); + release_firmware(firmware); + return err; } - /* request for firmware was successful. Download to device */ + /* Download firmware to device */ ice_load_pkg(firmware, pf); release_firmware(firmware); + + /* Initialize the supported Rx descriptor IDs after loading DDP */ + ice_init_supported_rxdids(hw, pf); + + return 0; } /** @@ -4143,134 +4655,98 @@ static void ice_print_wake_reason(struct ice_pf *pf) } /** - * ice_register_netdev - register netdev and devlink port - * @pf: pointer to the PF struct + * ice_register_netdev - register netdev + * @vsi: pointer to the VSI struct */ -static int ice_register_netdev(struct ice_pf *pf) +static int ice_register_netdev(struct ice_vsi *vsi) { - struct ice_vsi *vsi; - int err = 0; + int err; - vsi = ice_get_main_vsi(pf); if (!vsi || !vsi->netdev) return -EIO; err = register_netdev(vsi->netdev); if (err) - goto err_register_netdev; + return err; set_bit(ICE_VSI_NETDEV_REGISTERED, vsi->state); netif_carrier_off(vsi->netdev); netif_tx_stop_all_queues(vsi->netdev); - err = ice_devlink_create_port(vsi); - if (err) - goto err_devlink_create; - - devlink_port_type_eth_set(&vsi->devlink_port, vsi->netdev); return 0; -err_devlink_create: +} + +static void ice_unregister_netdev(struct ice_vsi *vsi) +{ + if (!vsi || !vsi->netdev) + return; + unregister_netdev(vsi->netdev); clear_bit(ICE_VSI_NETDEV_REGISTERED, vsi->state); -err_register_netdev: - free_netdev(vsi->netdev); - vsi->netdev = NULL; - clear_bit(ICE_VSI_NETDEV_ALLOCD, vsi->state); - return err; } /** - * ice_probe - Device initialization routine - * @pdev: PCI device information struct - * @ent: entry in ice_pci_tbl + * ice_cfg_netdev - Allocate, configure and register a netdev + * @vsi: the VSI associated with the new netdev * - * Returns 0 on success, negative on failure + * Returns 0 on success, negative value on failure */ -static int -ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) +static int ice_cfg_netdev(struct ice_vsi *vsi) { - struct device *dev = &pdev->dev; - struct ice_pf *pf; - struct ice_hw *hw; - int i, err; + struct ice_netdev_priv *np; + struct net_device *netdev; + u8 mac_addr[ETH_ALEN]; - /* this driver uses devres, see - * Documentation/driver-api/driver-model/devres.rst - */ - err = pcim_enable_device(pdev); - if (err) - return err; + netdev = alloc_etherdev_mqs(sizeof(*np), vsi->alloc_txq, + vsi->alloc_rxq); + if (!netdev) + return -ENOMEM; - err = pcim_iomap_regions(pdev, BIT(ICE_BAR0), dev_driver_string(dev)); - if (err) { - dev_err(dev, "BAR0 I/O map error %d\n", err); - return err; - } + set_bit(ICE_VSI_NETDEV_ALLOCD, vsi->state); + vsi->netdev = netdev; + np = netdev_priv(netdev); + np->vsi = vsi; - pf = ice_allocate_pf(dev); - if (!pf) - return -ENOMEM; + ice_set_netdev_features(netdev); + ice_set_ops(vsi); - /* set up for high or low DMA */ - err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)); - if (err) - err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32)); - if (err) { - dev_err(dev, "DMA configuration failed: 0x%x\n", err); - return err; + if (vsi->type == ICE_VSI_PF) { + SET_NETDEV_DEV(netdev, ice_pf_to_dev(vsi->back)); + ether_addr_copy(mac_addr, vsi->port_info->mac.perm_addr); + eth_hw_addr_set(netdev, mac_addr); } - pci_enable_pcie_error_reporting(pdev); - pci_set_master(pdev); - - pf->pdev = pdev; - pci_set_drvdata(pdev, pf); - set_bit(ICE_DOWN, pf->state); - /* Disable service task until DOWN bit is cleared */ - set_bit(ICE_SERVICE_DIS, pf->state); + netdev->priv_flags |= IFF_UNICAST_FLT; - hw = &pf->hw; - hw->hw_addr = pcim_iomap_table(pdev)[ICE_BAR0]; - pci_save_state(pdev); + /* Setup netdev TC information */ + ice_vsi_cfg_netdev_tc(vsi, vsi->tc_cfg.ena_tc); - hw->back = pf; - hw->vendor_id = pdev->vendor; - hw->device_id = pdev->device; - pci_read_config_byte(pdev, PCI_REVISION_ID, &hw->revision_id); - hw->subsystem_vendor_id = pdev->subsystem_vendor; - hw->subsystem_device_id = pdev->subsystem_device; - hw->bus.device = PCI_SLOT(pdev->devfn); - hw->bus.func = PCI_FUNC(pdev->devfn); - ice_set_ctrlq_len(hw); + netdev->max_mtu = ICE_MAX_MTU; - pf->msg_enable = netif_msg_init(debug, ICE_DFLT_NETIF_M); + return 0; +} - err = ice_devlink_register(pf); - if (err) { - dev_err(dev, "ice_devlink_register failed: %d\n", err); - goto err_exit_unroll; - } +static void ice_decfg_netdev(struct ice_vsi *vsi) +{ + clear_bit(ICE_VSI_NETDEV_ALLOCD, vsi->state); + free_netdev(vsi->netdev); + vsi->netdev = NULL; +} -#ifndef CONFIG_DYNAMIC_DEBUG - if (debug < -1) - hw->debug_mask = debug; -#endif +void ice_init_dev_hw(struct ice_pf *pf) +{ + struct ice_hw *hw = &pf->hw; + int err; - err = ice_init_hw(hw); - if (err) { - dev_err(dev, "ice_init_hw failed: %d\n", err); - err = -EIO; - goto err_exit_unroll; - } + ice_init_feature_support(pf); - ice_request_fw(pf); + err = ice_init_ddp_config(hw, pf); - /* if ice_request_fw fails, ICE_FLAG_ADV_FEATURES bit won't be + /* if ice_init_ddp_config fails, ICE_FLAG_ADV_FEATURES bit won't be * set in pf->state, which will cause ice_is_safe_mode to return * true */ - if (ice_is_safe_mode(pf)) { - dev_err(dev, "Package download failed. Advanced features disabled - Device now in Safe Mode\n"); + if (err || ice_is_safe_mode(pf)) { /* we already got function/device capabilities but these don't * reflect what the driver needs to do in safe mode. Instead of * adding conditional logic everywhere to ignore these @@ -4278,112 +4754,114 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) */ ice_set_safe_mode_caps(hw); } +} - err = ice_init_pf(pf); +int ice_init_dev(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + int err; + + ice_set_pf_caps(pf); + err = ice_init_interrupt_scheme(pf); if (err) { - dev_err(dev, "ice_init_pf failed: %d\n", err); - goto err_init_pf_unroll; + dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err); + return -EIO; } - ice_devlink_init_regions(pf); + ice_start_service_task(pf); - pf->hw.udp_tunnel_nic.set_port = ice_udp_tunnel_set_port; - pf->hw.udp_tunnel_nic.unset_port = ice_udp_tunnel_unset_port; - pf->hw.udp_tunnel_nic.flags = UDP_TUNNEL_NIC_INFO_MAY_SLEEP; - pf->hw.udp_tunnel_nic.shared = &pf->hw.udp_tunnel_shared; - i = 0; - if (pf->hw.tnl.valid_count[TNL_VXLAN]) { - pf->hw.udp_tunnel_nic.tables[i].n_entries = - pf->hw.tnl.valid_count[TNL_VXLAN]; - pf->hw.udp_tunnel_nic.tables[i].tunnel_types = - UDP_TUNNEL_TYPE_VXLAN; - i++; - } - if (pf->hw.tnl.valid_count[TNL_GENEVE]) { - pf->hw.udp_tunnel_nic.tables[i].n_entries = - pf->hw.tnl.valid_count[TNL_GENEVE]; - pf->hw.udp_tunnel_nic.tables[i].tunnel_types = - UDP_TUNNEL_TYPE_GENEVE; - i++; - } - - pf->num_alloc_vsi = hw->func_caps.guar_num_vsi; - if (!pf->num_alloc_vsi) { - err = -EIO; - goto err_init_pf_unroll; - } - if (pf->num_alloc_vsi > UDP_TUNNEL_NIC_MAX_SHARING_DEVICES) { - dev_warn(&pf->pdev->dev, - "limiting the VSI count due to UDP tunnel limitation %d > %d\n", - pf->num_alloc_vsi, UDP_TUNNEL_NIC_MAX_SHARING_DEVICES); - pf->num_alloc_vsi = UDP_TUNNEL_NIC_MAX_SHARING_DEVICES; - } + return 0; +} - pf->vsi = devm_kcalloc(dev, pf->num_alloc_vsi, sizeof(*pf->vsi), - GFP_KERNEL); - if (!pf->vsi) { - err = -ENOMEM; - goto err_init_pf_unroll; - } +void ice_deinit_dev(struct ice_pf *pf) +{ + ice_service_task_stop(pf); - err = ice_init_interrupt_scheme(pf); - if (err) { - dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err); - err = -EIO; - goto err_init_vsi_unroll; - } + /* Service task is already stopped, so call reset directly. */ + ice_reset(&pf->hw, ICE_RESET_PFR); + pci_wait_for_pending_transaction(pf->pdev); + ice_clear_interrupt_scheme(pf); +} - /* In case of MSIX we are going to setup the misc vector right here - * to handle admin queue events etc. In case of legacy and MSI - * the misc functionality and queue processing is combined in - * the same vector and that gets setup at open. - */ - err = ice_req_irq_msix_misc(pf); - if (err) { - dev_err(dev, "setup of misc vector failed: %d\n", err); - goto err_init_interrupt_unroll; - } +static void ice_init_features(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); - /* create switch struct for the switch element created by FW on boot */ - pf->first_sw = devm_kzalloc(dev, sizeof(*pf->first_sw), GFP_KERNEL); - if (!pf->first_sw) { - err = -ENOMEM; - goto err_msix_misc_unroll; - } + if (ice_is_safe_mode(pf)) + return; - if (hw->evb_veb) - pf->first_sw->bridge_mode = BRIDGE_MODE_VEB; - else - pf->first_sw->bridge_mode = BRIDGE_MODE_VEPA; + /* initialize DDP driven features */ + if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) + ice_ptp_init(pf); - pf->first_sw->pf = pf; + if (ice_is_feature_supported(pf, ICE_F_GNSS)) + ice_gnss_init(pf); - /* record the sw_id available for later use */ - pf->first_sw->sw_id = hw->port_info->sw_id; + if (ice_is_feature_supported(pf, ICE_F_CGU) || + ice_is_feature_supported(pf, ICE_F_PHY_RCLK)) + ice_dpll_init(pf); - err = ice_setup_pf_sw(pf); - if (err) { - dev_err(dev, "probe failed due to setup PF switch: %d\n", err); - goto err_alloc_sw_unroll; + /* Note: Flow director init failure is non-fatal to load */ + if (ice_init_fdir(pf)) + dev_err(dev, "could not initialize flow director\n"); + + /* Note: DCB init failure is non-fatal to load */ + if (ice_init_pf_dcb(pf, false)) { + clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); + clear_bit(ICE_FLAG_DCB_ENA, pf->flags); + } else { + ice_cfg_lldp_mib_change(&pf->hw, true); } - clear_bit(ICE_SERVICE_DIS, pf->state); + if (ice_init_lag(pf)) + dev_warn(dev, "Failed to init link aggregation support\n"); - /* tell the firmware we are up */ - err = ice_send_version(pf); - if (err) { - dev_err(dev, "probe failed sending driver version %s. error: %d\n", - UTS_RELEASE, err); - goto err_send_version_unroll; - } + ice_hwmon_init(pf); +} - /* since everything is good, start the service timer */ - mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period)); +static void ice_deinit_features(struct ice_pf *pf) +{ + if (ice_is_safe_mode(pf)) + return; + + ice_deinit_lag(pf); + if (test_bit(ICE_FLAG_DCB_CAPABLE, pf->flags)) + ice_cfg_lldp_mib_change(&pf->hw, false); + ice_deinit_fdir(pf); + if (ice_is_feature_supported(pf, ICE_F_GNSS)) + ice_gnss_exit(pf); + if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) + ice_ptp_release(pf); + if (test_bit(ICE_FLAG_DPLL, pf->flags)) + ice_dpll_deinit(pf); + if (pf->eswitch_mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) + xa_destroy(&pf->eswitch.reprs); +} + +static void ice_init_wakeup(struct ice_pf *pf) +{ + /* Save wakeup reason register for later use */ + pf->wakeup_reason = rd32(&pf->hw, PFPM_WUS); + + /* check for a power management event */ + ice_print_wake_reason(pf); + + /* clear wake status, all bits */ + wr32(&pf->hw, PFPM_WUS, U32_MAX); + + /* Disable WoL at init, wait for user to enable */ + device_set_wakeup_enable(ice_pf_to_dev(pf), false); +} + +static int ice_init_link(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + int err; err = ice_init_link_events(pf->hw.port_info); if (err) { dev_err(dev, "ice_init_link_events failed: %d\n", err); - goto err_send_version_unroll; + return err; } /* not a fatal error if this fails */ @@ -4398,7 +4876,8 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) ice_init_link_dflt_override(pf->hw.port_info); - ice_check_module_power(pf, pf->hw.port_info->phy.link_info.link_cfg_err); + ice_check_link_cfg_err(pf, + pf->hw.port_info->phy.link_info.link_cfg_err); /* if media available, initialize PHY settings */ if (pf->hw.port_info->phy.link_info.link_info & @@ -4418,98 +4897,454 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) set_bit(ICE_FLAG_NO_MEDIA, pf->flags); } - ice_verify_cacheline_size(pf); + return err; +} - /* Save wakeup reason register for later use */ - pf->wakeup_reason = rd32(hw, PFPM_WUS); +static int ice_init_pf_sw(struct ice_pf *pf) +{ + bool dvm = ice_is_dvm_ena(&pf->hw); + struct ice_vsi *vsi; + int err; - /* check for a power management event */ - ice_print_wake_reason(pf); + /* create switch struct for the switch element created by FW on boot */ + pf->first_sw = kzalloc(sizeof(*pf->first_sw), GFP_KERNEL); + if (!pf->first_sw) + return -ENOMEM; - /* clear wake status, all bits */ - wr32(hw, PFPM_WUS, U32_MAX); + if (pf->hw.evb_veb) + pf->first_sw->bridge_mode = BRIDGE_MODE_VEB; + else + pf->first_sw->bridge_mode = BRIDGE_MODE_VEPA; - /* Disable WoL at init, wait for user to enable */ - device_set_wakeup_enable(dev, false); + pf->first_sw->pf = pf; - if (ice_is_safe_mode(pf)) { - ice_set_safe_mode_vlan_cfg(pf); - goto probe_done; + /* record the sw_id available for later use */ + pf->first_sw->sw_id = pf->hw.port_info->sw_id; + + err = ice_aq_set_port_params(pf->hw.port_info, dvm, NULL); + if (err) + goto err_aq_set_port_params; + + vsi = ice_pf_vsi_setup(pf, pf->hw.port_info); + if (!vsi) { + err = -ENOMEM; + goto err_pf_vsi_setup; } - /* initialize DDP driven features */ - if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) - ice_ptp_init(pf); + return 0; - /* Note: Flow director init failure is non-fatal to load */ - if (ice_init_fdir(pf)) - dev_err(dev, "could not initialize flow director\n"); +err_pf_vsi_setup: +err_aq_set_port_params: + kfree(pf->first_sw); + return err; +} - /* Note: DCB init failure is non-fatal to load */ - if (ice_init_pf_dcb(pf, false)) { - clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); - clear_bit(ICE_FLAG_DCB_ENA, pf->flags); - } else { - ice_cfg_lldp_mib_change(&pf->hw, true); +static void ice_deinit_pf_sw(struct ice_pf *pf) +{ + struct ice_vsi *vsi = ice_get_main_vsi(pf); + + if (!vsi) + return; + + ice_vsi_release(vsi); + kfree(pf->first_sw); +} + +static int ice_alloc_vsis(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + + pf->num_alloc_vsi = pf->hw.func_caps.guar_num_vsi; + if (!pf->num_alloc_vsi) + return -EIO; + + if (pf->num_alloc_vsi > UDP_TUNNEL_NIC_MAX_SHARING_DEVICES) { + dev_warn(dev, + "limiting the VSI count due to UDP tunnel limitation %d > %d\n", + pf->num_alloc_vsi, UDP_TUNNEL_NIC_MAX_SHARING_DEVICES); + pf->num_alloc_vsi = UDP_TUNNEL_NIC_MAX_SHARING_DEVICES; } - if (ice_init_lag(pf)) - dev_warn(dev, "Failed to init link aggregation support\n"); + pf->vsi = devm_kcalloc(dev, pf->num_alloc_vsi, sizeof(*pf->vsi), + GFP_KERNEL); + if (!pf->vsi) + return -ENOMEM; + + pf->vsi_stats = devm_kcalloc(dev, pf->num_alloc_vsi, + sizeof(*pf->vsi_stats), GFP_KERNEL); + if (!pf->vsi_stats) { + devm_kfree(dev, pf->vsi); + return -ENOMEM; + } + + return 0; +} + +static void ice_dealloc_vsis(struct ice_pf *pf) +{ + devm_kfree(ice_pf_to_dev(pf), pf->vsi_stats); + pf->vsi_stats = NULL; + + pf->num_alloc_vsi = 0; + devm_kfree(ice_pf_to_dev(pf), pf->vsi); + pf->vsi = NULL; +} + +static int ice_init_devlink(struct ice_pf *pf) +{ + int err; + + err = ice_devlink_register_params(pf); + if (err) + return err; + + ice_devlink_init_regions(pf); + ice_devlink_register(pf); + ice_health_init(pf); - /* print PCI link speed and width */ - pcie_print_link_status(pf->pdev); + return 0; +} + +static void ice_deinit_devlink(struct ice_pf *pf) +{ + ice_health_deinit(pf); + ice_devlink_unregister(pf); + ice_devlink_destroy_regions(pf); + ice_devlink_unregister_params(pf); +} -probe_done: - err = ice_register_netdev(pf); +static int ice_init(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + int err; + + err = ice_init_pf(pf); + if (err) { + dev_err(dev, "ice_init_pf failed: %d\n", err); + return err; + } + + if (pf->hw.mac_type == ICE_MAC_E830) { + err = pci_enable_ptm(pf->pdev, NULL); + if (err) + dev_dbg(dev, "PCIe PTM not supported by PCIe bus/controller\n"); + } + + err = ice_alloc_vsis(pf); + if (err) + goto unroll_pf_init; + + err = ice_init_pf_sw(pf); + if (err) + goto err_init_pf_sw; + + ice_init_wakeup(pf); + + err = ice_init_link(pf); + if (err) + goto err_init_link; + + err = ice_send_version(pf); if (err) - goto err_netdev_reg; + goto err_init_link; + + ice_verify_cacheline_size(pf); + + if (ice_is_safe_mode(pf)) + ice_set_safe_mode_vlan_cfg(pf); + else + /* print PCI link speed and width */ + pcie_print_link_status(pf->pdev); /* ready to go, so clear down state bit */ clear_bit(ICE_DOWN, pf->state); - if (ice_is_aux_ena(pf)) { - pf->aux_idx = ida_alloc(&ice_aux_ida, GFP_KERNEL); - if (pf->aux_idx < 0) { - dev_err(dev, "Failed to allocate device ID for AUX driver\n"); - err = -ENOMEM; - goto err_netdev_reg; - } + clear_bit(ICE_SERVICE_DIS, pf->state); - err = ice_init_rdma(pf); - if (err) { - dev_err(dev, "Failed to initialize RDMA: %d\n", err); - err = -EIO; - goto err_init_aux_unroll; - } - } else { - dev_warn(dev, "RDMA is not supported on this device\n"); - } + /* since everything is good, start the service timer */ + mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period)); return 0; -err_init_aux_unroll: - pf->adev = NULL; - ida_free(&ice_aux_ida, pf->aux_idx); -err_netdev_reg: -err_send_version_unroll: - ice_vsi_release_all(pf); -err_alloc_sw_unroll: +err_init_link: + ice_deinit_pf_sw(pf); +err_init_pf_sw: + ice_dealloc_vsis(pf); +unroll_pf_init: + ice_deinit_pf(pf); + return err; +} + +static void ice_deinit(struct ice_pf *pf) +{ set_bit(ICE_SERVICE_DIS, pf->state); set_bit(ICE_DOWN, pf->state); - devm_kfree(dev, pf->first_sw); -err_msix_misc_unroll: - ice_free_irq_msix_misc(pf); -err_init_interrupt_unroll: - ice_clear_interrupt_scheme(pf); -err_init_vsi_unroll: - devm_kfree(dev, pf->vsi); -err_init_pf_unroll: + + ice_deinit_pf_sw(pf); + ice_dealloc_vsis(pf); ice_deinit_pf(pf); - ice_devlink_destroy_regions(pf); +} + +/** + * ice_load - load pf by init hw and starting VSI + * @pf: pointer to the pf instance + * + * This function has to be called under devl_lock. + */ +int ice_load(struct ice_pf *pf) +{ + struct ice_vsi *vsi; + int err; + + devl_assert_locked(priv_to_devlink(pf)); + + vsi = ice_get_main_vsi(pf); + + /* init channel list */ + INIT_LIST_HEAD(&vsi->ch_list); + + err = ice_cfg_netdev(vsi); + if (err) + return err; + + /* Setup DCB netlink interface */ + ice_dcbnl_setup(vsi); + + err = ice_init_mac_fltr(pf); + if (err) + goto err_init_mac_fltr; + + err = ice_devlink_create_pf_port(pf); + if (err) + goto err_devlink_create_pf_port; + + SET_NETDEV_DEVLINK_PORT(vsi->netdev, &pf->devlink_port); + + err = ice_register_netdev(vsi); + if (err) + goto err_register_netdev; + + err = ice_tc_indir_block_register(vsi); + if (err) + goto err_tc_indir_block_register; + + ice_napi_add(vsi); + + ice_init_features(pf); + + err = ice_init_rdma(pf); + if (err) + goto err_init_rdma; + + ice_service_task_restart(pf); + + clear_bit(ICE_DOWN, pf->state); + + return 0; + +err_init_rdma: + ice_deinit_features(pf); + ice_tc_indir_block_unregister(vsi); +err_tc_indir_block_register: + ice_unregister_netdev(vsi); +err_register_netdev: + ice_devlink_destroy_pf_port(pf); +err_devlink_create_pf_port: +err_init_mac_fltr: + ice_decfg_netdev(vsi); + return err; +} + +/** + * ice_unload - unload pf by stopping VSI and deinit hw + * @pf: pointer to the pf instance + * + * This function has to be called under devl_lock. + */ +void ice_unload(struct ice_pf *pf) +{ + struct ice_vsi *vsi = ice_get_main_vsi(pf); + + devl_assert_locked(priv_to_devlink(pf)); + + ice_deinit_rdma(pf); + ice_deinit_features(pf); + ice_tc_indir_block_unregister(vsi); + ice_unregister_netdev(vsi); + ice_devlink_destroy_pf_port(pf); + ice_decfg_netdev(vsi); +} + +static int ice_probe_recovery_mode(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + int err; + + dev_err(dev, "Firmware recovery mode detected. Limiting functionality. Refer to the Intel(R) Ethernet Adapters and Devices User Guide for details on firmware recovery mode\n"); + + INIT_HLIST_HEAD(&pf->aq_wait_list); + spin_lock_init(&pf->aq_wait_lock); + init_waitqueue_head(&pf->aq_wait_queue); + + timer_setup(&pf->serv_tmr, ice_service_timer, 0); + pf->serv_tmr_period = HZ; + INIT_WORK(&pf->serv_task, ice_service_task_recovery_mode); + clear_bit(ICE_SERVICE_SCHED, pf->state); + err = ice_create_all_ctrlq(&pf->hw); + if (err) + return err; + + scoped_guard(devl, priv_to_devlink(pf)) { + err = ice_init_devlink(pf); + if (err) + return err; + } + + ice_service_task_restart(pf); + + return 0; +} + +/** + * ice_probe - Device initialization routine + * @pdev: PCI device information struct + * @ent: entry in ice_pci_tbl + * + * Returns 0 on success, negative on failure + */ +static int +ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) +{ + struct device *dev = &pdev->dev; + bool need_dev_deinit = false; + struct ice_adapter *adapter; + struct ice_pf *pf; + struct ice_hw *hw; + int err; + + if (pdev->is_virtfn) { + dev_err(dev, "can't probe a virtual function\n"); + return -EINVAL; + } + + /* when under a kdump kernel initiate a reset before enabling the + * device in order to clear out any pending DMA transactions. These + * transactions can cause some systems to machine check when doing + * the pcim_enable_device() below. + */ + if (is_kdump_kernel()) { + pci_save_state(pdev); + pci_clear_master(pdev); + err = pcie_flr(pdev); + if (err) + return err; + pci_restore_state(pdev); + } + + /* this driver uses devres, see + * Documentation/driver-api/driver-model/devres.rst + */ + err = pcim_enable_device(pdev); + if (err) + return err; + + err = pcim_iomap_regions(pdev, BIT(ICE_BAR0), dev_driver_string(dev)); + if (err) { + dev_err(dev, "BAR0 I/O map error %d\n", err); + return err; + } + + pf = ice_allocate_pf(dev); + if (!pf) + return -ENOMEM; + + /* initialize Auxiliary index to invalid value */ + pf->aux_idx = -1; + + /* set up for high or low DMA */ + err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(64)); + if (err) { + dev_err(dev, "DMA configuration failed: 0x%x\n", err); + return err; + } + + pci_set_master(pdev); + pf->pdev = pdev; + pci_set_drvdata(pdev, pf); + set_bit(ICE_DOWN, pf->state); + /* Disable service task until DOWN bit is cleared */ + set_bit(ICE_SERVICE_DIS, pf->state); + + hw = &pf->hw; + hw->hw_addr = pcim_iomap_table(pdev)[ICE_BAR0]; + pci_save_state(pdev); + + hw->back = pf; + hw->port_info = NULL; + hw->vendor_id = pdev->vendor; + hw->device_id = pdev->device; + pci_read_config_byte(pdev, PCI_REVISION_ID, &hw->revision_id); + hw->subsystem_vendor_id = pdev->subsystem_vendor; + hw->subsystem_device_id = pdev->subsystem_device; + hw->bus.device = PCI_SLOT(pdev->devfn); + hw->bus.func = PCI_FUNC(pdev->devfn); + ice_set_ctrlq_len(hw); + + pf->msg_enable = netif_msg_init(debug, ICE_DFLT_NETIF_M); + +#ifndef CONFIG_DYNAMIC_DEBUG + if (debug < -1) + hw->debug_mask = debug; +#endif + + if (ice_is_recovery_mode(hw)) + return ice_probe_recovery_mode(pf); + + err = ice_init_hw(hw); + if (err) { + dev_err(dev, "ice_init_hw failed: %d\n", err); + return err; + } + + adapter = ice_adapter_get(pdev); + if (IS_ERR(adapter)) { + err = PTR_ERR(adapter); + goto unroll_hw_init; + } + pf->adapter = adapter; + + err = ice_init_dev(pf); + if (err) + goto unroll_adapter; + + err = ice_init(pf); + if (err) + goto unroll_dev_init; + + devl_lock(priv_to_devlink(pf)); + err = ice_load(pf); + if (err) + goto unroll_init; + + err = ice_init_devlink(pf); + if (err) + goto unroll_load; + devl_unlock(priv_to_devlink(pf)); + + return 0; + +unroll_load: + ice_unload(pf); +unroll_init: + devl_unlock(priv_to_devlink(pf)); + ice_deinit(pf); +unroll_dev_init: + need_dev_deinit = true; +unroll_adapter: + ice_adapter_put(pdev); +unroll_hw_init: ice_deinit_hw(hw); -err_exit_unroll: - ice_devlink_unregister(pf); - pci_disable_pcie_error_reporting(pdev); - pci_disable_device(pdev); + if (need_dev_deinit) + ice_deinit_dev(pf); return err; } @@ -4546,9 +5381,9 @@ static void ice_setup_mc_magic_wake(struct ice_pf *pf) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - enum ice_status status; u8 mac_addr[ETH_ALEN]; struct ice_vsi *vsi; + int status; u8 flags; if (!pf->wol_ena) @@ -4570,9 +5405,8 @@ static void ice_setup_mc_magic_wake(struct ice_pf *pf) status = ice_aq_manage_mac_write(hw, mac_addr, flags, NULL); if (status) - dev_err(dev, "Failed to enable Multicast Magic Packet wake, err %s aq_err %s\n", - ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); + dev_err(dev, "Failed to enable Multicast Magic Packet wake, err %d aq_err %s\n", + status, libie_aq_str(hw->adminq.sq_last_status)); } /** @@ -4584,56 +5418,49 @@ static void ice_remove(struct pci_dev *pdev) struct ice_pf *pf = pci_get_drvdata(pdev); int i; - if (!pf) - return; - for (i = 0; i < ICE_MAX_RESET_WAIT; i++) { if (!ice_is_reset_in_progress(pf->state)) break; msleep(100); } + if (ice_is_recovery_mode(&pf->hw)) { + ice_service_task_stop(pf); + scoped_guard(devl, priv_to_devlink(pf)) { + ice_deinit_devlink(pf); + } + return; + } + if (test_bit(ICE_FLAG_SRIOV_ENA, pf->flags)) { set_bit(ICE_VF_RESETS_DISABLED, pf->state); ice_free_vfs(pf); } - ice_service_task_stop(pf); - - ice_aq_cancel_waiting_tasks(pf); - ice_unplug_aux_dev(pf); - ida_free(&ice_aux_ida, pf->aux_idx); - set_bit(ICE_DOWN, pf->state); + ice_hwmon_exit(pf); - mutex_destroy(&(&pf->hw)->fdir_fltr_lock); - ice_deinit_lag(pf); - if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) - ice_ptp_release(pf); if (!ice_is_safe_mode(pf)) ice_remove_arfs(pf); - ice_setup_mc_magic_wake(pf); + + devl_lock(priv_to_devlink(pf)); + ice_dealloc_all_dynamic_ports(pf); + ice_deinit_devlink(pf); + + ice_unload(pf); + devl_unlock(priv_to_devlink(pf)); + + ice_deinit(pf); ice_vsi_release_all(pf); + + ice_setup_mc_magic_wake(pf); ice_set_wake(pf); - ice_free_irq_msix_misc(pf); - ice_for_each_vsi(pf, i) { - if (!pf->vsi[i]) - continue; - ice_vsi_free_q_vectors(pf->vsi[i]); - } - ice_deinit_pf(pf); - ice_devlink_destroy_regions(pf); + + ice_adapter_put(pdev); ice_deinit_hw(&pf->hw); - ice_devlink_unregister(pf); - /* Issue a PFR as part of the prescribed driver unload flow. Do not - * do it via ice_schedule_reset() since there is no need to rebuild - * and the service task is already stopped. - */ - ice_reset(&pf->hw, ICE_RESET_PFR); - pci_wait_for_pending_transaction(pdev); - ice_clear_interrupt_scheme(pf); - pci_disable_pcie_error_reporting(pdev); - pci_disable_device(pdev); + ice_deinit_dev(pf); + ice_aq_cancel_waiting_tasks(pf); + set_bit(ICE_DOWN, pf->state); } /** @@ -4652,7 +5479,6 @@ static void ice_shutdown(struct pci_dev *pdev) } } -#ifdef CONFIG_PM /** * ice_prepare_for_shutdown - prep for PCI shutdown * @pf: board private structure @@ -4677,7 +5503,7 @@ static void ice_prepare_for_shutdown(struct ice_pf *pf) if (pf->vsi[v]) pf->vsi[v]->vsi_num = 0; - ice_shutdown_all_ctrlq(hw); + ice_shutdown_all_ctrlq(hw, true); } /** @@ -4714,6 +5540,9 @@ static int ice_reinit_interrupt_scheme(struct ice_pf *pf) if (ret) goto err_reinit; ice_vsi_map_rings_to_vectors(pf->vsi[v]); + rtnl_lock(); + ice_vsi_set_napi_queues(pf->vsi[v]); + rtnl_unlock(); } ret = ice_req_irq_msix_misc(pf); @@ -4727,8 +5556,12 @@ static int ice_reinit_interrupt_scheme(struct ice_pf *pf) err_reinit: while (v--) - if (pf->vsi[v]) + if (pf->vsi[v]) { + rtnl_lock(); + ice_vsi_clear_napi_queues(pf->vsi[v]); + rtnl_unlock(); ice_vsi_free_q_vectors(pf->vsi[v]); + } return ret; } @@ -4740,7 +5573,7 @@ err_reinit: * Power Management callback to quiesce the device and prepare * for D3 transition. */ -static int __maybe_unused ice_suspend(struct device *dev) +static int ice_suspend(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); struct ice_pf *pf; @@ -4761,7 +5594,7 @@ static int __maybe_unused ice_suspend(struct device *dev) */ disabled = ice_service_task_stop(pf); - ice_unplug_aux_dev(pf); + ice_deinit_rdma(pf); /* Already suspended?, then there is nothing to do */ if (test_and_set_bit(ICE_SUSPENDED, pf->state)) { @@ -4793,9 +5626,11 @@ static int __maybe_unused ice_suspend(struct device *dev) ice_for_each_vsi(pf, v) { if (!pf->vsi[v]) continue; + rtnl_lock(); + ice_vsi_clear_napi_queues(pf->vsi[v]); + rtnl_unlock(); ice_vsi_free_q_vectors(pf->vsi[v]); } - ice_free_cpu_rx_rmap(ice_get_main_vsi(pf)); ice_clear_interrupt_scheme(pf); pci_save_state(pdev); @@ -4808,7 +5643,7 @@ static int __maybe_unused ice_suspend(struct device *dev) * ice_resume - PM callback for waking up from D3 * @dev: generic device information structure */ -static int __maybe_unused ice_resume(struct device *dev) +static int ice_resume(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); enum ice_reset_req reset_type; @@ -4818,7 +5653,6 @@ static int __maybe_unused ice_resume(struct device *dev) pci_set_power_state(pdev, PCI_D0); pci_restore_state(pdev); - pci_save_state(pdev); if (!pci_device_is_present(pdev)) return -ENODEV; @@ -4842,6 +5676,11 @@ static int __maybe_unused ice_resume(struct device *dev) if (ret) dev_err(dev, "Cannot restore interrupt scheme: %d\n", ret); + ret = ice_init_rdma(pf); + if (ret) + dev_err(dev, "Reinitialize RDMA during resume failed: %d\n", + ret); + clear_bit(ICE_DOWN, pf->state); /* Now perform PF reset and rebuild */ reset_type = ICE_RESET_PFR; @@ -4859,7 +5698,6 @@ static int __maybe_unused ice_resume(struct device *dev) return 0; } -#endif /* CONFIG_PM */ /** * ice_pci_err_detected - warning that PCI error has been detected @@ -4885,7 +5723,7 @@ ice_pci_err_detected(struct pci_dev *pdev, pci_channel_state_t err) if (!test_bit(ICE_PREPARED_FOR_RESET, pf->state)) { set_bit(ICE_PFR_REQ, pf->state); - ice_prepare_for_reset(pf); + ice_prepare_for_reset(pf, ICE_RESET_PFR); } } @@ -4914,7 +5752,6 @@ static pci_ers_result_t ice_pci_err_slot_reset(struct pci_dev *pdev) } else { pci_set_master(pdev); pci_restore_state(pdev); - pci_save_state(pdev); pci_wake_from_d3(pdev, false); /* Check for life */ @@ -4925,12 +5762,6 @@ static pci_ers_result_t ice_pci_err_slot_reset(struct pci_dev *pdev) result = PCI_ERS_RESULT_DISCONNECT; } - err = pci_aer_clear_nonfatal_status(pdev); - if (err) - dev_dbg(&pdev->dev, "pci_aer_clear_nonfatal_status() failed, error %d\n", - err); - /* non-fatal, continue */ - return result; } @@ -4957,7 +5788,7 @@ static void ice_pci_err_resume(struct pci_dev *pdev) return; } - ice_restore_all_vfs_msi_state(pdev); + ice_restore_all_vfs_msi_state(pf); ice_do_reset(pf, ICE_RESET_PFR); ice_service_task_restart(pf); @@ -4977,7 +5808,7 @@ static void ice_pci_err_reset_prepare(struct pci_dev *pdev) if (!test_bit(ICE_PREPARED_FOR_RESET, pf->state)) { set_bit(ICE_PFR_REQ, pf->state); - ice_prepare_for_reset(pf); + ice_prepare_for_reset(pf, ICE_RESET_PFR); } } } @@ -5000,35 +5831,61 @@ static void ice_pci_err_reset_done(struct pci_dev *pdev) * Class, Class Mask, private data (not used) } */ static const struct pci_device_id ice_pci_tbl[] = { - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_BACKPLANE), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_QSFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_SFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810_XXV_SFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_BACKPLANE), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_QSFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_SFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_10G_BASE_T), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_SGMII), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_BACKPLANE), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_QSFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_10G_BASE_T), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SGMII), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_BACKPLANE), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_10G_BASE_T), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SGMII), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_BACKPLANE), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_SFP), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_10G_BASE_T), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_1GBE), 0 }, - { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_QSFP), 0 }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_QSFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810C_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810_XXV_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810_XXV_QSFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E810_XXV_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_QSFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_10G_BASE_T) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823C_SGMII) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_QSFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_10G_BASE_T) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822C_SGMII) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_10G_BASE_T) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822L_SGMII) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_10G_BASE_T) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_1GBE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E823L_QSFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E822_SI_DFLT) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E825C_BACKPLANE), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E825C_QSFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E825C_SFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E825C_SGMII), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830CC_BACKPLANE) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830CC_QSFP56) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830CC_SFP) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830CC_SFP_DD) }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830C_BACKPLANE), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830_XXV_BACKPLANE), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830C_QSFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830_XXV_QSFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830C_SFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E830_XXV_SFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835CC_BACKPLANE), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835CC_QSFP56), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835CC_SFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835C_BACKPLANE), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835C_QSFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835C_SFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835_L_BACKPLANE), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835_L_QSFP), }, + { PCI_VDEVICE(INTEL, ICE_DEV_ID_E835_L_SFP), }, /* required last entry */ - { 0, } + {} }; MODULE_DEVICE_TABLE(pci, ice_pci_tbl); -static __maybe_unused SIMPLE_DEV_PM_OPS(ice_pm_ops, ice_suspend, ice_resume); +static DEFINE_SIMPLE_DEV_PM_OPS(ice_pm_ops, ice_suspend, ice_resume); static const struct pci_error_handlers ice_pci_err_handler = { .error_detected = ice_pci_err_detected, @@ -5043,11 +5900,11 @@ static struct pci_driver ice_driver = { .id_table = ice_pci_tbl, .probe = ice_probe, .remove = ice_remove, -#ifdef CONFIG_PM - .driver.pm = &ice_pm_ops, -#endif /* CONFIG_PM */ + .driver.pm = pm_sleep_ptr(&ice_pm_ops), .shutdown = ice_shutdown, .sriov_configure = ice_sriov_configure, + .sriov_get_vf_total_msix = ice_sriov_get_vf_total_msix, + .sriov_set_msix_vec_count = ice_sriov_set_msix_vec_count, .err_handler = &ice_pci_err_handler }; @@ -5059,23 +5916,48 @@ static struct pci_driver ice_driver = { */ static int __init ice_module_init(void) { - int status; + int status = -ENOMEM; pr_info("%s\n", ice_driver_string); pr_info("%s\n", ice_copyright); - ice_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, KBUILD_MODNAME); + ice_adv_lnk_speed_maps_init(); + + ice_wq = alloc_workqueue("%s", WQ_UNBOUND, 0, KBUILD_MODNAME); if (!ice_wq) { pr_err("Failed to create workqueue\n"); - return -ENOMEM; + return status; + } + + ice_lag_wq = alloc_ordered_workqueue("ice_lag_wq", 0); + if (!ice_lag_wq) { + pr_err("Failed to create LAG workqueue\n"); + goto err_dest_wq; } + ice_debugfs_init(); + status = pci_register_driver(&ice_driver); if (status) { pr_err("failed to register PCI driver, err %d\n", status); - destroy_workqueue(ice_wq); + goto err_dest_lag_wq; } + status = ice_sf_driver_register(); + if (status) { + pr_err("Failed to register SF driver, err %d\n", status); + goto err_sf_driver; + } + + return 0; + +err_sf_driver: + pci_unregister_driver(&ice_driver); +err_dest_lag_wq: + destroy_workqueue(ice_lag_wq); + ice_debugfs_exit(); +err_dest_wq: + destroy_workqueue(ice_wq); return status; } module_init(ice_module_init); @@ -5088,8 +5970,11 @@ module_init(ice_module_init); */ static void __exit ice_module_exit(void) { + ice_sf_driver_unregister(); pci_unregister_driver(&ice_driver); + ice_debugfs_exit(); destroy_workqueue(ice_wq); + destroy_workqueue(ice_lag_wq); pr_info("module unloaded\n"); } module_exit(ice_module_exit); @@ -5108,21 +5993,16 @@ static int ice_set_mac_address(struct net_device *netdev, void *pi) struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; struct sockaddr *addr = pi; - enum ice_status status; + u8 old_mac[ETH_ALEN]; u8 flags = 0; - int err = 0; u8 *mac; + int err; mac = (u8 *)addr->sa_data; if (!is_valid_ether_addr(mac)) return -EADDRNOTAVAIL; - if (ether_addr_equal(netdev->dev_addr, mac)) { - netdev_warn(netdev, "already using mac %pM\n", mac); - return 0; - } - if (test_bit(ICE_DOWN, pf->state) || ice_is_reset_in_progress(pf->state)) { netdev_err(netdev, "can't set mac %pM. device not ready\n", @@ -5130,48 +6010,60 @@ static int ice_set_mac_address(struct net_device *netdev, void *pi) return -EBUSY; } + if (ice_chnl_dmac_fltr_cnt(pf)) { + netdev_err(netdev, "can't set mac %pM. Device has tc-flower filters, delete all of them and try again\n", + mac); + return -EAGAIN; + } + + netif_addr_lock_bh(netdev); + ether_addr_copy(old_mac, netdev->dev_addr); + /* change the netdev's MAC address */ + eth_hw_addr_set(netdev, mac); + netif_addr_unlock_bh(netdev); + /* Clean up old MAC filter. Not an error if old filter doesn't exist */ - status = ice_fltr_remove_mac(vsi, netdev->dev_addr, ICE_FWD_TO_VSI); - if (status && status != ICE_ERR_DOES_NOT_EXIST) { + err = ice_fltr_remove_mac(vsi, old_mac, ICE_FWD_TO_VSI); + if (err && err != -ENOENT) { err = -EADDRNOTAVAIL; goto err_update_filters; } /* Add filter for new MAC. If filter exists, return success */ - status = ice_fltr_add_mac(vsi, mac, ICE_FWD_TO_VSI); - if (status == ICE_ERR_ALREADY_EXISTS) { + err = ice_fltr_add_mac(vsi, mac, ICE_FWD_TO_VSI); + if (err == -EEXIST) { /* Although this MAC filter is already present in hardware it's * possible in some cases (e.g. bonding) that dev_addr was * modified outside of the driver and needs to be restored back * to this value. */ - memcpy(netdev->dev_addr, mac, netdev->addr_len); netdev_dbg(netdev, "filter for MAC %pM already exists\n", mac); - return 0; - } - /* error if the new filter addition failed */ - if (status) + return 0; + } else if (err) { + /* error if the new filter addition failed */ err = -EADDRNOTAVAIL; + } err_update_filters: if (err) { netdev_err(netdev, "can't set MAC %pM. filter update failed\n", mac); + netif_addr_lock_bh(netdev); + eth_hw_addr_set(netdev, old_mac); + netif_addr_unlock_bh(netdev); return err; } - /* change the netdev's MAC address */ - memcpy(netdev->dev_addr, mac, netdev->addr_len); netdev_dbg(vsi->netdev, "updated MAC address to %pM\n", netdev->dev_addr); /* write new MAC address to the firmware */ flags = ICE_AQC_MAN_MAC_UPDATE_LAA_WOL; - status = ice_aq_manage_mac_write(hw, mac, flags, NULL); - if (status) { - netdev_err(netdev, "can't set MAC %pM. write to firmware failed error %s\n", - mac, ice_stat_str(status)); + err = ice_aq_manage_mac_write(hw, mac, flags, NULL); + if (err) { + netdev_err(netdev, "can't set MAC %pM. write to firmware failed error %d\n", + mac, err); } return 0; } @@ -5185,7 +6077,7 @@ static void ice_set_rx_mode(struct net_device *netdev) struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - if (!vsi) + if (!vsi || ice_is_switchdev_running(vsi->back)) return; /* Set the flags to synchronize filters @@ -5213,8 +6105,8 @@ ice_set_tx_maxrate(struct net_device *netdev, int queue_index, u32 maxrate) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - enum ice_status status; u16 q_handle; + int status; u8 tc; /* Validate maxrate requested is within permitted range */ @@ -5227,6 +6119,13 @@ ice_set_tx_maxrate(struct net_device *netdev, int queue_index, u32 maxrate) q_handle = vsi->tx_rings[queue_index]->q_handle; tc = ice_dcb_get_tc(vsi, queue_index); + vsi = ice_locate_vsi_using_queue(vsi, queue_index); + if (!vsi) { + netdev_err(netdev, "Invalid VSI for given queue %d\n", + queue_index); + return -EINVAL; + } + /* Set BW back to default, when user set maxrate to 0 */ if (!maxrate) status = ice_cfg_q_bw_dflt_lmt(vsi->port_info, vsi->idx, tc, @@ -5234,13 +6133,11 @@ ice_set_tx_maxrate(struct net_device *netdev, int queue_index, u32 maxrate) else status = ice_cfg_q_bw_lmt(vsi->port_info, vsi->idx, tc, q_handle, ICE_MAX_BW, maxrate * 1000); - if (status) { - netdev_err(netdev, "Unable to set Tx max rate, error %s\n", - ice_stat_str(status)); - return -EIO; - } + if (status) + netdev_err(netdev, "Unable to set Tx max rate, error %d\n", + status); - return 0; + return status; } /** @@ -5251,12 +6148,14 @@ ice_set_tx_maxrate(struct net_device *netdev, int queue_index, u32 maxrate) * @addr: the MAC address entry being added * @vid: VLAN ID * @flags: instructions from stack about fdb operation + * @notified: whether notification was emitted * @extack: netlink extended ack */ static int ice_fdb_add(struct ndmsg *ndm, struct nlattr __always_unused *tb[], struct net_device *dev, const unsigned char *addr, u16 vid, - u16 flags, struct netlink_ext_ack __always_unused *extack) + u16 flags, bool *notified, + struct netlink_ext_ack __always_unused *extack) { int err; @@ -5290,11 +6189,14 @@ ice_fdb_add(struct ndmsg *ndm, struct nlattr __always_unused *tb[], * @dev: the net device pointer * @addr: the MAC address entry being added * @vid: VLAN ID + * @notified: whether notification was emitted + * @extack: netlink extended ack */ static int ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[], struct net_device *dev, const unsigned char *addr, - __always_unused u16 vid) + __always_unused u16 vid, bool *notified, + struct netlink_ext_ack *extack) { int err; @@ -5313,6 +6215,273 @@ ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[], return err; } +#define NETIF_VLAN_OFFLOAD_FEATURES (NETIF_F_HW_VLAN_CTAG_RX | \ + NETIF_F_HW_VLAN_CTAG_TX | \ + NETIF_F_HW_VLAN_STAG_RX | \ + NETIF_F_HW_VLAN_STAG_TX) + +#define NETIF_VLAN_STRIPPING_FEATURES (NETIF_F_HW_VLAN_CTAG_RX | \ + NETIF_F_HW_VLAN_STAG_RX) + +#define NETIF_VLAN_FILTERING_FEATURES (NETIF_F_HW_VLAN_CTAG_FILTER | \ + NETIF_F_HW_VLAN_STAG_FILTER) + +/** + * ice_fix_features - fix the netdev features flags based on device limitations + * @netdev: ptr to the netdev that flags are being fixed on + * @features: features that need to be checked and possibly fixed + * + * Make sure any fixups are made to features in this callback. This enables the + * driver to not have to check unsupported configurations throughout the driver + * because that's the responsiblity of this callback. + * + * Single VLAN Mode (SVM) Supported Features: + * NETIF_F_HW_VLAN_CTAG_FILTER + * NETIF_F_HW_VLAN_CTAG_RX + * NETIF_F_HW_VLAN_CTAG_TX + * + * Double VLAN Mode (DVM) Supported Features: + * NETIF_F_HW_VLAN_CTAG_FILTER + * NETIF_F_HW_VLAN_CTAG_RX + * NETIF_F_HW_VLAN_CTAG_TX + * + * NETIF_F_HW_VLAN_STAG_FILTER + * NETIF_HW_VLAN_STAG_RX + * NETIF_HW_VLAN_STAG_TX + * + * Features that need fixing: + * Cannot simultaneously enable CTAG and STAG stripping and/or insertion. + * These are mutually exlusive as the VSI context cannot support multiple + * VLAN ethertypes simultaneously for stripping and/or insertion. If this + * is not done, then default to clearing the requested STAG offload + * settings. + * + * All supported filtering has to be enabled or disabled together. For + * example, in DVM, CTAG and STAG filtering have to be enabled and disabled + * together. If this is not done, then default to VLAN filtering disabled. + * These are mutually exclusive as there is currently no way to + * enable/disable VLAN filtering based on VLAN ethertype when using VLAN + * prune rules. + */ +static netdev_features_t +ice_fix_features(struct net_device *netdev, netdev_features_t features) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + netdev_features_t req_vlan_fltr, cur_vlan_fltr; + bool cur_ctag, cur_stag, req_ctag, req_stag; + + cur_vlan_fltr = netdev->features & NETIF_VLAN_FILTERING_FEATURES; + cur_ctag = cur_vlan_fltr & NETIF_F_HW_VLAN_CTAG_FILTER; + cur_stag = cur_vlan_fltr & NETIF_F_HW_VLAN_STAG_FILTER; + + req_vlan_fltr = features & NETIF_VLAN_FILTERING_FEATURES; + req_ctag = req_vlan_fltr & NETIF_F_HW_VLAN_CTAG_FILTER; + req_stag = req_vlan_fltr & NETIF_F_HW_VLAN_STAG_FILTER; + + if (req_vlan_fltr != cur_vlan_fltr) { + if (ice_is_dvm_ena(&np->vsi->back->hw)) { + if (req_ctag && req_stag) { + features |= NETIF_VLAN_FILTERING_FEATURES; + } else if (!req_ctag && !req_stag) { + features &= ~NETIF_VLAN_FILTERING_FEATURES; + } else if ((!cur_ctag && req_ctag && !cur_stag) || + (!cur_stag && req_stag && !cur_ctag)) { + features |= NETIF_VLAN_FILTERING_FEATURES; + netdev_warn(netdev, "802.1Q and 802.1ad VLAN filtering must be either both on or both off. VLAN filtering has been enabled for both types.\n"); + } else if ((cur_ctag && !req_ctag && cur_stag) || + (cur_stag && !req_stag && cur_ctag)) { + features &= ~NETIF_VLAN_FILTERING_FEATURES; + netdev_warn(netdev, "802.1Q and 802.1ad VLAN filtering must be either both on or both off. VLAN filtering has been disabled for both types.\n"); + } + } else { + if (req_vlan_fltr & NETIF_F_HW_VLAN_STAG_FILTER) + netdev_warn(netdev, "cannot support requested 802.1ad filtering setting in SVM mode\n"); + + if (req_vlan_fltr & NETIF_F_HW_VLAN_CTAG_FILTER) + features |= NETIF_F_HW_VLAN_CTAG_FILTER; + } + } + + if ((features & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX)) && + (features & (NETIF_F_HW_VLAN_STAG_RX | NETIF_F_HW_VLAN_STAG_TX))) { + netdev_warn(netdev, "cannot support CTAG and STAG VLAN stripping and/or insertion simultaneously since CTAG and STAG offloads are mutually exclusive, clearing STAG offload settings\n"); + features &= ~(NETIF_F_HW_VLAN_STAG_RX | + NETIF_F_HW_VLAN_STAG_TX); + } + + if (!(netdev->features & NETIF_F_RXFCS) && + (features & NETIF_F_RXFCS) && + (features & NETIF_VLAN_STRIPPING_FEATURES) && + !ice_vsi_has_non_zero_vlans(np->vsi)) { + netdev_warn(netdev, "Disabling VLAN stripping as FCS/CRC stripping is also disabled and there is no VLAN configured\n"); + features &= ~NETIF_VLAN_STRIPPING_FEATURES; + } + + return features; +} + +/** + * ice_set_rx_rings_vlan_proto - update rings with new stripped VLAN proto + * @vsi: PF's VSI + * @vlan_ethertype: VLAN ethertype (802.1Q or 802.1ad) in network byte order + * + * Store current stripped VLAN proto in ring packet context, + * so it can be accessed more efficiently by packet processing code. + */ +static void +ice_set_rx_rings_vlan_proto(struct ice_vsi *vsi, __be16 vlan_ethertype) +{ + u16 i; + + ice_for_each_alloc_rxq(vsi, i) + vsi->rx_rings[i]->pkt_ctx.vlan_proto = vlan_ethertype; +} + +/** + * ice_set_vlan_offload_features - set VLAN offload features for the PF VSI + * @vsi: PF's VSI + * @features: features used to determine VLAN offload settings + * + * First, determine the vlan_ethertype based on the VLAN offload bits in + * features. Then determine if stripping and insertion should be enabled or + * disabled. Finally enable or disable VLAN stripping and insertion. + */ +static int +ice_set_vlan_offload_features(struct ice_vsi *vsi, netdev_features_t features) +{ + bool enable_stripping = true, enable_insertion = true; + struct ice_vsi_vlan_ops *vlan_ops; + int strip_err = 0, insert_err = 0; + u16 vlan_ethertype = 0; + + vlan_ops = ice_get_compat_vsi_vlan_ops(vsi); + + if (features & (NETIF_F_HW_VLAN_STAG_RX | NETIF_F_HW_VLAN_STAG_TX)) + vlan_ethertype = ETH_P_8021AD; + else if (features & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX)) + vlan_ethertype = ETH_P_8021Q; + + if (!(features & (NETIF_F_HW_VLAN_STAG_RX | NETIF_F_HW_VLAN_CTAG_RX))) + enable_stripping = false; + if (!(features & (NETIF_F_HW_VLAN_STAG_TX | NETIF_F_HW_VLAN_CTAG_TX))) + enable_insertion = false; + + if (enable_stripping) + strip_err = vlan_ops->ena_stripping(vsi, vlan_ethertype); + else + strip_err = vlan_ops->dis_stripping(vsi); + + if (enable_insertion) + insert_err = vlan_ops->ena_insertion(vsi, vlan_ethertype); + else + insert_err = vlan_ops->dis_insertion(vsi); + + if (strip_err || insert_err) + return -EIO; + + ice_set_rx_rings_vlan_proto(vsi, enable_stripping ? + htons(vlan_ethertype) : 0); + + return 0; +} + +/** + * ice_set_vlan_filtering_features - set VLAN filtering features for the PF VSI + * @vsi: PF's VSI + * @features: features used to determine VLAN filtering settings + * + * Enable or disable Rx VLAN filtering based on the VLAN filtering bits in the + * features. + */ +static int +ice_set_vlan_filtering_features(struct ice_vsi *vsi, netdev_features_t features) +{ + struct ice_vsi_vlan_ops *vlan_ops = ice_get_compat_vsi_vlan_ops(vsi); + int err = 0; + + /* support Single VLAN Mode (SVM) and Double VLAN Mode (DVM) by checking + * if either bit is set. In switchdev mode Rx filtering should never be + * enabled. + */ + if ((features & + (NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_HW_VLAN_STAG_FILTER)) && + !ice_is_eswitch_mode_switchdev(vsi->back)) + err = vlan_ops->ena_rx_filtering(vsi); + else + err = vlan_ops->dis_rx_filtering(vsi); + + return err; +} + +/** + * ice_set_vlan_features - set VLAN settings based on suggested feature set + * @netdev: ptr to the netdev being adjusted + * @features: the feature set that the stack is suggesting + * + * Only update VLAN settings if the requested_vlan_features are different than + * the current_vlan_features. + */ +static int +ice_set_vlan_features(struct net_device *netdev, netdev_features_t features) +{ + netdev_features_t current_vlan_features, requested_vlan_features; + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + int err; + + current_vlan_features = netdev->features & NETIF_VLAN_OFFLOAD_FEATURES; + requested_vlan_features = features & NETIF_VLAN_OFFLOAD_FEATURES; + if (current_vlan_features ^ requested_vlan_features) { + if ((features & NETIF_F_RXFCS) && + (features & NETIF_VLAN_STRIPPING_FEATURES)) { + dev_err(ice_pf_to_dev(vsi->back), + "To enable VLAN stripping, you must first enable FCS/CRC stripping\n"); + return -EIO; + } + + err = ice_set_vlan_offload_features(vsi, features); + if (err) + return err; + } + + current_vlan_features = netdev->features & + NETIF_VLAN_FILTERING_FEATURES; + requested_vlan_features = features & NETIF_VLAN_FILTERING_FEATURES; + if (current_vlan_features ^ requested_vlan_features) { + err = ice_set_vlan_filtering_features(vsi, features); + if (err) + return err; + } + + return 0; +} + +/** + * ice_set_loopback - turn on/off loopback mode on underlying PF + * @vsi: ptr to VSI + * @ena: flag to indicate the on/off setting + */ +static int ice_set_loopback(struct ice_vsi *vsi, bool ena) +{ + bool if_running = netif_running(vsi->netdev); + int ret; + + if (if_running && !test_and_set_bit(ICE_VSI_DOWN, vsi->state)) { + ret = ice_down(vsi); + if (ret) { + netdev_err(vsi->netdev, "Preparing device to toggle loopback failed\n"); + return ret; + } + } + ret = ice_aq_set_mac_loopback(&vsi->back->hw, ena, NULL); + if (ret) + netdev_err(vsi->netdev, "Failed to toggle loopback state\n"); + if (if_running) + ret = ice_up(vsi); + + return ret; +} + /** * ice_set_features - set the netdev feature flags * @netdev: ptr to the netdev being adjusted @@ -5321,97 +6490,123 @@ ice_fdb_del(struct ndmsg *ndm, __always_unused struct nlattr *tb[], static int ice_set_features(struct net_device *netdev, netdev_features_t features) { + netdev_features_t changed = netdev->features ^ features; struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; int ret = 0; /* Don't set any netdev advanced features with device in Safe Mode */ - if (ice_is_safe_mode(vsi->back)) { - dev_err(ice_pf_to_dev(vsi->back), "Device is in Safe Mode - not enabling advanced netdev features\n"); + if (ice_is_safe_mode(pf)) { + dev_err(ice_pf_to_dev(pf), + "Device is in Safe Mode - not enabling advanced netdev features\n"); return ret; } /* Do not change setting during reset */ if (ice_is_reset_in_progress(pf->state)) { - dev_err(ice_pf_to_dev(vsi->back), "Device is resetting, changing advanced netdev features temporarily unavailable.\n"); + dev_err(ice_pf_to_dev(pf), + "Device is resetting, changing advanced netdev features temporarily unavailable.\n"); return -EBUSY; } /* Multiple features can be changed in one call so keep features in * separate if/else statements to guarantee each feature is checked */ - if (features & NETIF_F_RXHASH && !(netdev->features & NETIF_F_RXHASH)) - ice_vsi_manage_rss_lut(vsi, true); - else if (!(features & NETIF_F_RXHASH) && - netdev->features & NETIF_F_RXHASH) - ice_vsi_manage_rss_lut(vsi, false); - - if ((features & NETIF_F_HW_VLAN_CTAG_RX) && - !(netdev->features & NETIF_F_HW_VLAN_CTAG_RX)) - ret = ice_vsi_manage_vlan_stripping(vsi, true); - else if (!(features & NETIF_F_HW_VLAN_CTAG_RX) && - (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)) - ret = ice_vsi_manage_vlan_stripping(vsi, false); - - if ((features & NETIF_F_HW_VLAN_CTAG_TX) && - !(netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) - ret = ice_vsi_manage_vlan_insertion(vsi); - else if (!(features & NETIF_F_HW_VLAN_CTAG_TX) && - (netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) - ret = ice_vsi_manage_vlan_insertion(vsi); - - if ((features & NETIF_F_HW_VLAN_CTAG_FILTER) && - !(netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) - ret = ice_cfg_vlan_pruning(vsi, true, false); - else if (!(features & NETIF_F_HW_VLAN_CTAG_FILTER) && - (netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) - ret = ice_cfg_vlan_pruning(vsi, false, false); - - if ((features & NETIF_F_NTUPLE) && - !(netdev->features & NETIF_F_NTUPLE)) { - ice_vsi_manage_fdir(vsi, true); - ice_init_arfs(vsi); - } else if (!(features & NETIF_F_NTUPLE) && - (netdev->features & NETIF_F_NTUPLE)) { - ice_vsi_manage_fdir(vsi, false); - ice_clear_arfs(vsi); + if (changed & NETIF_F_RXHASH) + ice_vsi_manage_rss_lut(vsi, !!(features & NETIF_F_RXHASH)); + + ret = ice_set_vlan_features(netdev, features); + if (ret) + return ret; + + /* Turn on receive of FCS aka CRC, and after setting this + * flag the packet data will have the 4 byte CRC appended + */ + if (changed & NETIF_F_RXFCS) { + if ((features & NETIF_F_RXFCS) && + (features & NETIF_VLAN_STRIPPING_FEATURES)) { + dev_err(ice_pf_to_dev(vsi->back), + "To disable FCS/CRC stripping, you must first disable VLAN stripping\n"); + return -EIO; + } + + ice_vsi_cfg_crc_strip(vsi, !!(features & NETIF_F_RXFCS)); + ret = ice_down_up(vsi); + if (ret) + return ret; + } + + if (changed & NETIF_F_NTUPLE) { + bool ena = !!(features & NETIF_F_NTUPLE); + + ice_vsi_manage_fdir(vsi, ena); + ena ? ice_init_arfs(vsi) : ice_clear_arfs(vsi); + } + + /* don't turn off hw_tc_offload when ADQ is already enabled */ + if (!(features & NETIF_F_HW_TC) && ice_is_adq_active(pf)) { + dev_err(ice_pf_to_dev(pf), "ADQ is active, can't turn hw_tc_offload off\n"); + return -EACCES; + } + + if (changed & NETIF_F_HW_TC) { + bool ena = !!(features & NETIF_F_HW_TC); + + assign_bit(ICE_FLAG_CLS_FLOWER, pf->flags, ena); + } + + if (changed & NETIF_F_LOOPBACK) + ret = ice_set_loopback(vsi, !!(features & NETIF_F_LOOPBACK)); + + /* Due to E830 hardware limitations, TSO (NETIF_F_ALL_TSO) with GCS + * (NETIF_F_HW_CSUM) is not supported. + */ + if (ice_is_feature_supported(pf, ICE_F_GCS) && + ((features & NETIF_F_HW_CSUM) && (features & NETIF_F_ALL_TSO))) { + if (netdev->features & NETIF_F_HW_CSUM) + dev_err(ice_pf_to_dev(pf), "To enable TSO, you must first disable HW checksum.\n"); + else + dev_err(ice_pf_to_dev(pf), "To enable HW checksum, you must first disable TSO.\n"); + return -EIO; } return ret; } /** - * ice_vsi_vlan_setup - Setup VLAN offload properties on a VSI + * ice_vsi_vlan_setup - Setup VLAN offload properties on a PF VSI * @vsi: VSI to setup VLAN properties for */ static int ice_vsi_vlan_setup(struct ice_vsi *vsi) { - int ret = 0; + int err; - if (vsi->netdev->features & NETIF_F_HW_VLAN_CTAG_RX) - ret = ice_vsi_manage_vlan_stripping(vsi, true); - if (vsi->netdev->features & NETIF_F_HW_VLAN_CTAG_TX) - ret = ice_vsi_manage_vlan_insertion(vsi); + err = ice_set_vlan_offload_features(vsi, vsi->netdev->features); + if (err) + return err; - return ret; + err = ice_set_vlan_filtering_features(vsi, vsi->netdev->features); + if (err) + return err; + + return ice_vsi_add_vlan_zero(vsi); } /** - * ice_vsi_cfg - Setup the VSI + * ice_vsi_cfg_lan - Setup the VSI lan related config * @vsi: the VSI being configured * * Return 0 on success and negative value on error */ -int ice_vsi_cfg(struct ice_vsi *vsi) +int ice_vsi_cfg_lan(struct ice_vsi *vsi) { int err; - if (vsi->netdev) { + if (vsi->netdev && vsi->type == ICE_VSI_PF) { ice_set_rx_mode(vsi->netdev); err = ice_vsi_vlan_setup(vsi); - if (err) return err; } @@ -5427,77 +6622,59 @@ int ice_vsi_cfg(struct ice_vsi *vsi) } /* THEORY OF MODERATION: - * The below code creates custom DIM profiles for use by this driver, because - * the ice driver hardware works differently than the hardware that DIMLIB was + * The ice driver hardware works differently than the hardware that DIMLIB was * originally made for. ice hardware doesn't have packet count limits that * can trigger an interrupt, but it *does* have interrupt rate limit support, - * and this code adds that capability to be used by the driver when it's using - * DIMLIB. The DIMLIB code was always designed to be a suggestion to the driver - * for how to "respond" to traffic and interrupts, so this driver uses a - * slightly different set of moderation parameters to get best performance. + * which is hard-coded to a limit of 250,000 ints/second. + * If not using dynamic moderation, the INTRL value can be modified + * by ethtool rx-usecs-high. */ struct ice_dim { /* the throttle rate for interrupts, basically worst case delay before * an initial interrupt fires, value is stored in microseconds. */ u16 itr; - /* the rate limit for interrupts, which can cap a delay from a small - * ITR at a certain amount of interrupts per second. f.e. a 2us ITR - * could yield as much as 500,000 interrupts per second, but with a - * 10us rate limit, it limits to 100,000 interrupts per second. Value - * is stored in microseconds. - */ - u16 intrl; }; /* Make a different profile for Rx that doesn't allow quite so aggressive - * moderation at the high end (it maxes out at 128us or about 8k interrupts a - * second. The INTRL/rate parameters here are only useful to cap small ITR - * values, which is why for larger ITR's - like 128, which can only generate - * 8k interrupts per second, there is no point to rate limit and the values - * are set to zero. The rate limit values do affect latency, and so must - * be reasonably small so to not impact latency sensitive tests. + * moderation at the high end (it maxes out at 126us or about 8k interrupts a + * second. */ static const struct ice_dim rx_profile[] = { - {2, 10}, - {8, 16}, - {32, 0}, - {96, 0}, - {128, 0} + {2}, /* 500,000 ints/s, capped at 250K by INTRL */ + {8}, /* 125,000 ints/s */ + {16}, /* 62,500 ints/s */ + {62}, /* 16,129 ints/s */ + {126} /* 7,936 ints/s */ }; /* The transmit profile, which has the same sorts of values * as the previous struct */ static const struct ice_dim tx_profile[] = { - {2, 10}, - {8, 16}, - {64, 0}, - {128, 0}, - {256, 0} + {2}, /* 500,000 ints/s, capped at 250K by INTRL */ + {8}, /* 125,000 ints/s */ + {40}, /* 16,125 ints/s */ + {128}, /* 7,812 ints/s */ + {256} /* 3,906 ints/s */ }; static void ice_tx_dim_work(struct work_struct *work) { struct ice_ring_container *rc; - struct ice_q_vector *q_vector; struct dim *dim; - u16 itr, intrl; + u16 itr; dim = container_of(work, struct dim, work); - rc = container_of(dim, struct ice_ring_container, dim); - q_vector = container_of(rc, struct ice_q_vector, tx); + rc = dim->priv; - if (dim->profile_ix >= ARRAY_SIZE(tx_profile)) - dim->profile_ix = ARRAY_SIZE(tx_profile) - 1; + WARN_ON(dim->profile_ix >= ARRAY_SIZE(tx_profile)); /* look up the values in our local table */ itr = tx_profile[dim->profile_ix].itr; - intrl = tx_profile[dim->profile_ix].intrl; - ice_trace(tx_dim_work, q_vector, dim); + ice_trace(tx_dim_work, container_of(rc, struct ice_q_vector, tx), dim); ice_write_itr(rc, itr); - ice_write_intrl(q_vector, intrl); dim->state = DIM_START_MEASURE; } @@ -5505,28 +6682,65 @@ static void ice_tx_dim_work(struct work_struct *work) static void ice_rx_dim_work(struct work_struct *work) { struct ice_ring_container *rc; - struct ice_q_vector *q_vector; struct dim *dim; - u16 itr, intrl; + u16 itr; dim = container_of(work, struct dim, work); - rc = container_of(dim, struct ice_ring_container, dim); - q_vector = container_of(rc, struct ice_q_vector, rx); + rc = dim->priv; - if (dim->profile_ix >= ARRAY_SIZE(rx_profile)) - dim->profile_ix = ARRAY_SIZE(rx_profile) - 1; + WARN_ON(dim->profile_ix >= ARRAY_SIZE(rx_profile)); /* look up the values in our local table */ itr = rx_profile[dim->profile_ix].itr; - intrl = rx_profile[dim->profile_ix].intrl; - ice_trace(rx_dim_work, q_vector, dim); + ice_trace(rx_dim_work, container_of(rc, struct ice_q_vector, rx), dim); ice_write_itr(rc, itr); - ice_write_intrl(q_vector, intrl); dim->state = DIM_START_MEASURE; } +#define ICE_DIM_DEFAULT_PROFILE_IX 1 + +/** + * ice_init_moderation - set up interrupt moderation + * @q_vector: the vector containing rings to be configured + * + * Set up interrupt moderation registers, with the intent to do the right thing + * when called from reset or from probe, and whether or not dynamic moderation + * is enabled or not. Take special care to write all the registers in both + * dynamic moderation mode or not in order to make sure hardware is in a known + * state. + */ +static void ice_init_moderation(struct ice_q_vector *q_vector) +{ + struct ice_ring_container *rc; + bool tx_dynamic, rx_dynamic; + + rc = &q_vector->tx; + INIT_WORK(&rc->dim.work, ice_tx_dim_work); + rc->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; + rc->dim.profile_ix = ICE_DIM_DEFAULT_PROFILE_IX; + rc->dim.priv = rc; + tx_dynamic = ITR_IS_DYNAMIC(rc); + + /* set the initial TX ITR to match the above */ + ice_write_itr(rc, tx_dynamic ? + tx_profile[rc->dim.profile_ix].itr : rc->itr_setting); + + rc = &q_vector->rx; + INIT_WORK(&rc->dim.work, ice_rx_dim_work); + rc->dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; + rc->dim.profile_ix = ICE_DIM_DEFAULT_PROFILE_IX; + rc->dim.priv = rc; + rx_dynamic = ITR_IS_DYNAMIC(rc); + + /* set the initial RX ITR to match the above */ + ice_write_itr(rc, rx_dynamic ? rx_profile[rc->dim.profile_ix].itr : + rc->itr_setting); + + ice_set_q_vector_intrl(q_vector); +} + /** * ice_napi_enable_all - Enable NAPI for all q_vectors in the VSI * @vsi: the VSI being configured @@ -5541,13 +6755,9 @@ static void ice_napi_enable_all(struct ice_vsi *vsi) ice_for_each_q_vector(vsi, q_idx) { struct ice_q_vector *q_vector = vsi->q_vectors[q_idx]; - INIT_WORK(&q_vector->tx.dim.work, ice_tx_dim_work); - q_vector->tx.dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; + ice_init_moderation(q_vector); - INIT_WORK(&q_vector->rx.dim.work, ice_rx_dim_work); - q_vector->rx.dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE; - - if (q_vector->rx.ring || q_vector->tx.ring) + if (q_vector->rx.rx_ring || q_vector->tx.tx_ring) napi_enable(&q_vector->napi); } } @@ -5579,13 +6789,21 @@ static int ice_up_complete(struct ice_vsi *vsi) if (vsi->port_info && (vsi->port_info->phy.link_info.link_info & ICE_AQ_LINK_UP) && - vsi->netdev) { + ((vsi->netdev && (vsi->type == ICE_VSI_PF || + vsi->type == ICE_VSI_SF)))) { ice_print_link_msg(vsi, true); netif_tx_start_all_queues(vsi->netdev); netif_carrier_on(vsi->netdev); + ice_ptp_link_change(pf, true); } - ice_service_task_schedule(pf); + /* Perform an initial read of the statistics registers now to + * set the baseline so counters are ready when interface is up + */ + ice_update_eth_stats(vsi); + + if (vsi->type == ICE_VSI_PF) + ice_service_task_schedule(pf); return 0; } @@ -5598,7 +6816,7 @@ int ice_up(struct ice_vsi *vsi) { int err; - err = ice_vsi_cfg(vsi); + err = ice_vsi_cfg_lan(vsi); if (!err) err = ice_up_complete(vsi); @@ -5607,53 +6825,56 @@ int ice_up(struct ice_vsi *vsi) /** * ice_fetch_u64_stats_per_ring - get packets and bytes stats per ring - * @ring: Tx or Rx ring to read stats from + * @syncp: pointer to u64_stats_sync + * @stats: stats that pkts and bytes count will be taken from * @pkts: packets stats counter * @bytes: bytes stats counter * * This function fetches stats from the ring considering the atomic operations * that needs to be performed to read u64 values in 32 bit machine. */ -static void -ice_fetch_u64_stats_per_ring(struct ice_ring *ring, u64 *pkts, u64 *bytes) +void +ice_fetch_u64_stats_per_ring(struct u64_stats_sync *syncp, + struct ice_q_stats stats, u64 *pkts, u64 *bytes) { unsigned int start; - *pkts = 0; - *bytes = 0; - if (!ring) - return; do { - start = u64_stats_fetch_begin_irq(&ring->syncp); - *pkts = ring->stats.pkts; - *bytes = ring->stats.bytes; - } while (u64_stats_fetch_retry_irq(&ring->syncp, start)); + start = u64_stats_fetch_begin(syncp); + *pkts = stats.pkts; + *bytes = stats.bytes; + } while (u64_stats_fetch_retry(syncp, start)); } /** * ice_update_vsi_tx_ring_stats - Update VSI Tx ring stats counters * @vsi: the VSI to be updated + * @vsi_stats: the stats struct to be updated * @rings: rings to work on * @count: number of rings */ static void -ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, struct ice_ring **rings, - u16 count) +ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, + struct rtnl_link_stats64 *vsi_stats, + struct ice_tx_ring **rings, u16 count) { - struct rtnl_link_stats64 *vsi_stats = &vsi->net_stats; u16 i; for (i = 0; i < count; i++) { - struct ice_ring *ring; - u64 pkts, bytes; + struct ice_tx_ring *ring; + u64 pkts = 0, bytes = 0; ring = READ_ONCE(rings[i]); - ice_fetch_u64_stats_per_ring(ring, &pkts, &bytes); + if (!ring || !ring->ring_stats) + continue; + ice_fetch_u64_stats_per_ring(&ring->ring_stats->syncp, + ring->ring_stats->stats, &pkts, + &bytes); vsi_stats->tx_packets += pkts; vsi_stats->tx_bytes += bytes; - vsi->tx_restart += ring->tx_stats.restart_q; - vsi->tx_busy += ring->tx_stats.tx_busy; - vsi->tx_linearize += ring->tx_stats.tx_linearize; + vsi->tx_restart += ring->ring_stats->tx_stats.restart_q; + vsi->tx_busy += ring->ring_stats->tx_stats.tx_busy; + vsi->tx_linearize += ring->ring_stats->tx_stats.tx_linearize; } } @@ -5663,15 +6884,15 @@ ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, struct ice_ring **rings, */ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) { - struct rtnl_link_stats64 *vsi_stats = &vsi->net_stats; + struct rtnl_link_stats64 *net_stats, *stats_prev; + struct rtnl_link_stats64 *vsi_stats; + struct ice_pf *pf = vsi->back; u64 pkts, bytes; int i; - /* reset netdev stats */ - vsi_stats->tx_packets = 0; - vsi_stats->tx_bytes = 0; - vsi_stats->rx_packets = 0; - vsi_stats->rx_bytes = 0; + vsi_stats = kzalloc(sizeof(*vsi_stats), GFP_ATOMIC); + if (!vsi_stats) + return; /* reset non-netdev (extended) stats */ vsi->tx_restart = 0; @@ -5683,25 +6904,52 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) rcu_read_lock(); /* update Tx rings counters */ - ice_update_vsi_tx_ring_stats(vsi, vsi->tx_rings, vsi->num_txq); + ice_update_vsi_tx_ring_stats(vsi, vsi_stats, vsi->tx_rings, + vsi->num_txq); /* update Rx rings counters */ ice_for_each_rxq(vsi, i) { - struct ice_ring *ring = READ_ONCE(vsi->rx_rings[i]); + struct ice_rx_ring *ring = READ_ONCE(vsi->rx_rings[i]); + struct ice_ring_stats *ring_stats; - ice_fetch_u64_stats_per_ring(ring, &pkts, &bytes); + ring_stats = ring->ring_stats; + ice_fetch_u64_stats_per_ring(&ring_stats->syncp, + ring_stats->stats, &pkts, + &bytes); vsi_stats->rx_packets += pkts; vsi_stats->rx_bytes += bytes; - vsi->rx_buf_failed += ring->rx_stats.alloc_buf_failed; - vsi->rx_page_failed += ring->rx_stats.alloc_page_failed; + vsi->rx_buf_failed += ring_stats->rx_stats.alloc_buf_failed; + vsi->rx_page_failed += ring_stats->rx_stats.alloc_page_failed; } /* update XDP Tx rings counters */ if (ice_is_xdp_ena_vsi(vsi)) - ice_update_vsi_tx_ring_stats(vsi, vsi->xdp_rings, + ice_update_vsi_tx_ring_stats(vsi, vsi_stats, vsi->xdp_rings, vsi->num_xdp_txq); rcu_read_unlock(); + + net_stats = &vsi->net_stats; + stats_prev = &vsi->net_stats_prev; + + /* Update netdev counters, but keep in mind that values could start at + * random value after PF reset. And as we increase the reported stat by + * diff of Prev-Cur, we need to be sure that Prev is valid. If it's not, + * let's skip this round. + */ + if (likely(pf->stat_prev_loaded)) { + net_stats->tx_packets += vsi_stats->tx_packets - stats_prev->tx_packets; + net_stats->tx_bytes += vsi_stats->tx_bytes - stats_prev->tx_bytes; + net_stats->rx_packets += vsi_stats->rx_packets - stats_prev->rx_packets; + net_stats->rx_bytes += vsi_stats->rx_bytes - stats_prev->rx_bytes; + } + + stats_prev->tx_packets = vsi_stats->tx_packets; + stats_prev->tx_bytes = vsi_stats->tx_bytes; + stats_prev->rx_packets = vsi_stats->rx_packets; + stats_prev->rx_bytes = vsi_stats->rx_bytes; + + kfree(vsi_stats); } /** @@ -5734,13 +6982,11 @@ void ice_update_vsi_stats(struct ice_vsi *vsi) cur_ns->rx_crc_errors = pf->stats.crc_errors; cur_ns->rx_errors = pf->stats.crc_errors + pf->stats.illegal_bytes + - pf->stats.rx_len_errors + pf->stats.rx_undersize + pf->hw_csum_rx_error + pf->stats.rx_jabber + pf->stats.rx_fragments + pf->stats.rx_oversize; - cur_ns->rx_length_errors = pf->stats.rx_len_errors; /* record drops from the port level */ cur_ns->rx_missed_errors = pf->stats.eth.rx_discards; } @@ -5761,6 +7007,9 @@ void ice_update_pf_stats(struct ice_pf *pf) prev_ps = &pf->stats_prev; cur_ps = &pf->stats; + if (ice_is_reset_in_progress(pf->state)) + pf->stat_prev_loaded = false; + ice_stat_update40(hw, GLPRT_GORCL(port), pf->stat_prev_loaded, &prev_ps->eth.rx_bytes, &cur_ps->eth.rx_bytes); @@ -5902,7 +7151,6 @@ void ice_update_pf_stats(struct ice_pf *pf) * @netdev: network interface device structure * @stats: main device statistics structure */ -static void ice_get_stats64(struct net_device *netdev, struct rtnl_link_stats64 *stats) { struct ice_netdev_priv *np = netdev_priv(netdev); @@ -5953,7 +7201,7 @@ static void ice_napi_disable_all(struct ice_vsi *vsi) ice_for_each_q_vector(vsi, q_idx) { struct ice_q_vector *q_vector = vsi->q_vectors[q_idx]; - if (q_vector->rx.ring || q_vector->tx.ring) + if (q_vector->rx.rx_ring || q_vector->tx.tx_ring) napi_disable(&q_vector->napi); cancel_work_sync(&q_vector->tx.dim.work); @@ -5962,17 +7210,64 @@ static void ice_napi_disable_all(struct ice_vsi *vsi) } /** + * ice_vsi_dis_irq - Mask off queue interrupt generation on the VSI + * @vsi: the VSI being un-configured + */ +static void ice_vsi_dis_irq(struct ice_vsi *vsi) +{ + struct ice_pf *pf = vsi->back; + struct ice_hw *hw = &pf->hw; + u32 val; + int i; + + /* disable interrupt causation from each Rx queue; Tx queues are + * handled in ice_vsi_stop_tx_ring() + */ + if (vsi->rx_rings) { + ice_for_each_rxq(vsi, i) { + if (vsi->rx_rings[i]) { + u16 reg; + + reg = vsi->rx_rings[i]->reg_idx; + val = rd32(hw, QINT_RQCTL(reg)); + val &= ~QINT_RQCTL_CAUSE_ENA_M; + wr32(hw, QINT_RQCTL(reg), val); + } + } + } + + /* disable each interrupt */ + ice_for_each_q_vector(vsi, i) { + if (!vsi->q_vectors[i]) + continue; + wr32(hw, GLINT_DYN_CTL(vsi->q_vectors[i]->reg_idx), 0); + } + + ice_flush(hw); + + /* don't call synchronize_irq() for VF's from the host */ + if (vsi->type == ICE_VSI_VF) + return; + + ice_for_each_q_vector(vsi, i) + synchronize_irq(vsi->q_vectors[i]->irq.virq); +} + +/** * ice_down - Shutdown the connection * @vsi: The VSI being stopped + * + * Caller of this function is expected to set the vsi->state ICE_DOWN bit */ int ice_down(struct ice_vsi *vsi) { - int i, tx_err, rx_err, link_err = 0; + int i, tx_err, rx_err, vlan_err = 0; + + WARN_ON(!test_bit(ICE_VSI_DOWN, vsi->state)); - /* Caller of this function is expected to set the - * vsi->state ICE_DOWN bit - */ if (vsi->netdev) { + vlan_err = ice_vsi_del_vlan_zero(vsi); + ice_ptp_link_change(vsi->back, false); netif_carrier_off(vsi->netdev); netif_tx_disable(vsi->netdev); } @@ -5983,7 +7278,7 @@ int ice_down(struct ice_vsi *vsi) if (tx_err) netdev_err(vsi->netdev, "Failed stop Tx rings, VSI %d error %d\n", vsi->vsi_num, tx_err); - if (!tx_err && ice_is_xdp_ena_vsi(vsi)) { + if (!tx_err && vsi->xdp_rings) { tx_err = ice_vsi_stop_xdp_tx_rings(vsi); if (tx_err) netdev_err(vsi->netdev, "Failed stop XDP rings, VSI %d error %d\n", @@ -5997,20 +7292,17 @@ int ice_down(struct ice_vsi *vsi) ice_napi_disable_all(vsi); - if (test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, vsi->back->flags)) { - link_err = ice_force_phys_link_state(vsi, false); - if (link_err) - netdev_err(vsi->netdev, "Failed to set physical link down, VSI %d error %d\n", - vsi->vsi_num, link_err); - } - ice_for_each_txq(vsi, i) ice_clean_tx_ring(vsi->tx_rings[i]); + if (vsi->xdp_rings) + ice_for_each_xdp_txq(vsi, i) + ice_clean_tx_ring(vsi->xdp_rings[i]); + ice_for_each_rxq(vsi, i) ice_clean_rx_ring(vsi->rx_rings[i]); - if (tx_err || rx_err || link_err) { + if (tx_err || rx_err || vlan_err) { netdev_err(vsi->netdev, "Failed to close VSI 0x%04X on switch 0x%04X\n", vsi->vsi_num, vsi->vsw->sw_id); return -EIO; @@ -6020,6 +7312,31 @@ int ice_down(struct ice_vsi *vsi) } /** + * ice_down_up - shutdown the VSI connection and bring it up + * @vsi: the VSI to be reconnected + */ +int ice_down_up(struct ice_vsi *vsi) +{ + int ret; + + /* if DOWN already set, nothing to do */ + if (test_and_set_bit(ICE_VSI_DOWN, vsi->state)) + return 0; + + ret = ice_down(vsi); + if (ret) + return ret; + + ret = ice_up(vsi); + if (ret) { + netdev_err(vsi->netdev, "reallocating resources failed during netdev features change, may need to reload driver\n"); + return ret; + } + + return 0; +} + +/** * ice_vsi_setup_tx_rings - Allocate VSI Tx queue resources * @vsi: VSI having resources allocated * @@ -6036,12 +7353,13 @@ int ice_vsi_setup_tx_rings(struct ice_vsi *vsi) } ice_for_each_txq(vsi, i) { - struct ice_ring *ring = vsi->tx_rings[i]; + struct ice_tx_ring *ring = vsi->tx_rings[i]; if (!ring) return -EINVAL; - ring->netdev = vsi->netdev; + if (vsi->netdev) + ring->netdev = vsi->netdev; err = ice_setup_tx_ring(ring); if (err) break; @@ -6067,12 +7385,13 @@ int ice_vsi_setup_rx_rings(struct ice_vsi *vsi) } ice_for_each_rxq(vsi, i) { - struct ice_ring *ring = vsi->rx_rings[i]; + struct ice_rx_ring *ring = vsi->rx_rings[i]; if (!ring) return -EINVAL; - ring->netdev = vsi->netdev; + if (vsi->netdev) + ring->netdev = vsi->netdev; err = ice_setup_rx_ring(ring); if (err) break; @@ -6106,7 +7425,7 @@ int ice_vsi_open_ctrl(struct ice_vsi *vsi) if (err) goto err_setup_rx; - err = ice_vsi_cfg(vsi); + err = ice_vsi_cfg_lan(vsi); if (err) goto err_setup_rx; @@ -6145,7 +7464,7 @@ err_setup_tx: * * Returns 0 on success, negative value on error */ -static int ice_vsi_open(struct ice_vsi *vsi) +int ice_vsi_open(struct ice_vsi *vsi) { char int_name[ICE_INT_NAME_STR_LEN]; struct ice_pf *pf = vsi->back; @@ -6160,7 +7479,7 @@ static int ice_vsi_open(struct ice_vsi *vsi) if (err) goto err_setup_rx; - err = ice_vsi_cfg(vsi); + err = ice_vsi_cfg_lan(vsi); if (err) goto err_setup_rx; @@ -6170,14 +7489,21 @@ static int ice_vsi_open(struct ice_vsi *vsi) if (err) goto err_setup_rx; - /* Notify the stack of the actual queue counts. */ - err = netif_set_real_num_tx_queues(vsi->netdev, vsi->num_txq); - if (err) - goto err_set_qs; + if (bitmap_empty(pf->txtime_txqs, pf->max_pf_txqs)) + ice_vsi_cfg_netdev_tc(vsi, vsi->tc_cfg.ena_tc); - err = netif_set_real_num_rx_queues(vsi->netdev, vsi->num_rxq); - if (err) - goto err_set_qs; + if (vsi->type == ICE_VSI_PF || vsi->type == ICE_VSI_SF) { + /* Notify the stack of the actual queue counts. */ + err = netif_set_real_num_tx_queues(vsi->netdev, vsi->num_txq); + if (err) + goto err_set_qs; + + err = netif_set_real_num_rx_queues(vsi->netdev, vsi->num_rxq); + if (err) + goto err_set_qs; + + ice_vsi_set_napi_queues(vsi); + } err = ice_up_complete(vsi); if (err) @@ -6212,6 +7538,9 @@ static void ice_vsi_release_all(struct ice_pf *pf) if (!pf->vsi[i]) continue; + if (pf->vsi[i]->type == ICE_VSI_CHNL) + continue; + err = ice_vsi_release(pf->vsi[i]); if (err) dev_dbg(ice_pf_to_dev(pf), "Failed to release pf->vsi[%d], err %d, vsi_num = %d\n", @@ -6229,7 +7558,6 @@ static void ice_vsi_release_all(struct ice_pf *pf) static int ice_vsi_rebuild_by_type(struct ice_pf *pf, enum ice_vsi_type type) { struct device *dev = ice_pf_to_dev(pf); - enum ice_status status; int i, err; ice_for_each_vsi(pf, i) { @@ -6239,7 +7567,7 @@ static int ice_vsi_rebuild_by_type(struct ice_pf *pf, enum ice_vsi_type type) continue; /* rebuild the VSI */ - err = ice_vsi_rebuild(vsi, true); + err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_INIT); if (err) { dev_err(dev, "rebuild VSI failed, err %d, VSI index %d, type %s\n", err, vsi->idx, ice_vsi_type_str(type)); @@ -6247,12 +7575,11 @@ static int ice_vsi_rebuild_by_type(struct ice_pf *pf, enum ice_vsi_type type) } /* replay filters for the VSI */ - status = ice_replay_vsi(&pf->hw, vsi->idx); - if (status) { - dev_err(dev, "replay VSI failed, status %s, VSI index %d, type %s\n", - ice_stat_str(status), vsi->idx, - ice_vsi_type_str(type)); - return -EIO; + err = ice_replay_vsi(&pf->hw, vsi->idx); + if (err) { + dev_err(dev, "replay VSI failed, error %d, VSI index %d, type %s\n", + err, vsi->idx, ice_vsi_type_str(type)); + return err; } /* Re-map HW VSI number, using VSI handle that has been @@ -6313,9 +7640,10 @@ static void ice_update_pf_netdev_link(struct ice_pf *pf) */ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) { + struct ice_vsi *vsi = ice_get_main_vsi(pf); struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - enum ice_status ret; + bool dvm; int err; if (test_bit(ICE_DOWN, pf->state)) @@ -6323,10 +7651,20 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) dev_dbg(dev, "rebuilding PF after reset_type=%d\n", reset_type); - ret = ice_init_all_ctrlq(hw); - if (ret) { - dev_err(dev, "control queues init failed %s\n", - ice_stat_str(ret)); +#define ICE_EMP_RESET_SLEEP_MS 5000 + if (reset_type == ICE_RESET_EMPR) { + /* If an EMP reset has occurred, any previously pending flash + * update will have completed. We no longer know whether or + * not the NVM update EMP reset is restricted. + */ + pf->fw_emp_reset_disabled = false; + + msleep(ICE_EMP_RESET_SLEEP_MS); + } + + err = ice_init_all_ctrlq(hw); + if (err) { + dev_err(dev, "control queues init failed %d\n", err); goto err_init_ctrlq; } @@ -6340,39 +7678,38 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) ice_load_pkg(NULL, pf); } - ret = ice_clear_pf_cfg(hw); - if (ret) { - dev_err(dev, "clear PF configuration failed %s\n", - ice_stat_str(ret)); + err = ice_clear_pf_cfg(hw); + if (err) { + dev_err(dev, "clear PF configuration failed %d\n", err); goto err_init_ctrlq; } - if (pf->first_sw->dflt_vsi_ena) - dev_info(dev, "Clearing default VSI, re-enable after reset completes\n"); - /* clear the default VSI configuration if it exists */ - pf->first_sw->dflt_vsi = NULL; - pf->first_sw->dflt_vsi_ena = false; - ice_clear_pxe_mode(hw); - ret = ice_init_nvm(hw); - if (ret) { - dev_err(dev, "ice_init_nvm failed %s\n", ice_stat_str(ret)); + err = ice_init_nvm(hw); + if (err) { + dev_err(dev, "ice_init_nvm failed %d\n", err); goto err_init_ctrlq; } - ret = ice_get_caps(hw); - if (ret) { - dev_err(dev, "ice_get_caps failed %s\n", ice_stat_str(ret)); + err = ice_get_caps(hw); + if (err) { + dev_err(dev, "ice_get_caps failed %d\n", err); goto err_init_ctrlq; } - ret = ice_aq_set_mac_cfg(hw, ICE_AQ_SET_MAC_FRAME_SIZE_MAX, NULL); - if (ret) { - dev_err(dev, "set_mac_cfg failed %s\n", ice_stat_str(ret)); + err = ice_aq_set_mac_cfg(hw, ICE_AQ_SET_MAC_FRAME_SIZE_MAX, NULL); + if (err) { + dev_err(dev, "set_mac_cfg failed %d\n", err); goto err_init_ctrlq; } + dvm = ice_is_dvm_ena(hw); + + err = ice_aq_set_port_params(pf->hw.port_info, dvm, NULL); + if (err) + goto err_init_ctrlq; + err = ice_sched_init_port(hw->port_info); if (err) goto err_sched_init_port; @@ -6407,7 +7744,10 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) * fail. */ if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags)) - ice_ptp_init(pf); + ice_ptp_rebuild(pf, reset_type); + + if (ice_is_feature_supported(pf, ICE_F_GNSS)) + ice_gnss_init(pf); /* rebuild PF VSI */ err = ice_vsi_rebuild_by_type(pf, ICE_VSI_PF); @@ -6416,6 +7756,15 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) goto err_vsi_rebuild; } + if (reset_type == ICE_RESET_PFR) { + err = ice_rebuild_channels(pf); + if (err) { + dev_err(dev, "failed to rebuild and replay ADQ VSIs, err %d\n", + err); + goto err_vsi_rebuild; + } + } + /* If Flow Director is active */ if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) { err = ice_vsi_rebuild_by_type(pf, ICE_VSI_CTRL); @@ -6434,13 +7783,16 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) ice_rebuild_arfs(pf); } + if (vsi && vsi->netdev) + netif_device_attach(vsi->netdev); + ice_update_pf_netdev_link(pf); /* tell the firmware we are up */ - ret = ice_send_version(pf); - if (ret) { - dev_err(dev, "Rebuild failed due to error sending driver version: %s\n", - ice_stat_str(ret)); + err = ice_send_version(pf); + if (err) { + dev_err(dev, "Rebuild failed due to error sending driver version: %d\n", + err); goto err_vsi_rebuild; } @@ -6449,14 +7801,21 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) /* if we get here, reset flow is successful */ clear_bit(ICE_RESET_FAILED, pf->state); + ice_health_clear(pf); + ice_plug_aux_dev(pf); + if (ice_is_feature_supported(pf, ICE_F_SRIOV_LAG)) + ice_lag_rebuild(pf); + + /* Restore timestamp mode settings after VSI rebuild */ + ice_ptp_restore_timestamp_mode(pf); return; err_vsi_rebuild: err_sched_init_port: ice_sched_cleanup_all(hw); err_init_ctrlq: - ice_shutdown_all_ctrlq(hw); + ice_shutdown_all_ctrlq(hw, false); set_bit(ICE_RESET_FAILED, pf->state); clear_recovery: /* set this bit in PF state to control service task scheduling */ @@ -6465,30 +7824,18 @@ clear_recovery: } /** - * ice_max_xdp_frame_size - returns the maximum allowed frame size for XDP - * @vsi: Pointer to VSI structure - */ -static int ice_max_xdp_frame_size(struct ice_vsi *vsi) -{ - if (PAGE_SIZE >= 8192 || test_bit(ICE_FLAG_LEGACY_RX, vsi->back->flags)) - return ICE_RXBUF_2048 - XDP_PACKET_HEADROOM; - else - return ICE_RXBUF_3072; -} - -/** * ice_change_mtu - NDO callback to change the MTU * @netdev: network interface device structure * @new_mtu: new value for maximum frame size * * Returns 0 on success, negative on failure */ -static int ice_change_mtu(struct net_device *netdev, int new_mtu) +int ice_change_mtu(struct net_device *netdev, int new_mtu) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; - struct iidc_event *event; + struct bpf_prog *prog; u8 count = 0; int err = 0; @@ -6497,7 +7844,8 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) return 0; } - if (ice_is_xdp_ena_vsi(vsi)) { + prog = vsi->xdp_prog; + if (prog && !prog->aux->xdp_has_frags) { int frame_size = ice_max_xdp_frame_size(vsi); if (new_mtu + ICE_ETH_PKT_HDR_PAD > frame_size) { @@ -6523,176 +7871,18 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) return -EBUSY; } - event = kzalloc(sizeof(*event), GFP_KERNEL); - if (!event) - return -ENOMEM; - - set_bit(IIDC_EVENT_BEFORE_MTU_CHANGE, event->type); - ice_send_event_to_aux(pf, event); - clear_bit(IIDC_EVENT_BEFORE_MTU_CHANGE, event->type); - - netdev->mtu = (unsigned int)new_mtu; - - /* if VSI is up, bring it down and then back up */ - if (!test_and_set_bit(ICE_VSI_DOWN, vsi->state)) { - err = ice_down(vsi); - if (err) { - netdev_err(netdev, "change MTU if_down err %d\n", err); - goto event_after; - } - - err = ice_up(vsi); - if (err) { - netdev_err(netdev, "change MTU if_up err %d\n", err); - goto event_after; - } - } + WRITE_ONCE(netdev->mtu, (unsigned int)new_mtu); + err = ice_down_up(vsi); + if (err) + return err; netdev_dbg(netdev, "changed MTU to %d\n", new_mtu); -event_after: - set_bit(IIDC_EVENT_AFTER_MTU_CHANGE, event->type); - ice_send_event_to_aux(pf, event); - kfree(event); + set_bit(ICE_FLAG_MTU_CHANGED, pf->flags); return err; } /** - * ice_do_ioctl - Access the hwtstamp interface - * @netdev: network interface device structure - * @ifr: interface request data - * @cmd: ioctl command - */ -static int ice_do_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) -{ - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; - - switch (cmd) { - case SIOCGHWTSTAMP: - return ice_ptp_get_ts_config(pf, ifr); - case SIOCSHWTSTAMP: - return ice_ptp_set_ts_config(pf, ifr); - default: - return -EOPNOTSUPP; - } -} - -/** - * ice_aq_str - convert AQ err code to a string - * @aq_err: the AQ error code to convert - */ -const char *ice_aq_str(enum ice_aq_err aq_err) -{ - switch (aq_err) { - case ICE_AQ_RC_OK: - return "OK"; - case ICE_AQ_RC_EPERM: - return "ICE_AQ_RC_EPERM"; - case ICE_AQ_RC_ENOENT: - return "ICE_AQ_RC_ENOENT"; - case ICE_AQ_RC_ENOMEM: - return "ICE_AQ_RC_ENOMEM"; - case ICE_AQ_RC_EBUSY: - return "ICE_AQ_RC_EBUSY"; - case ICE_AQ_RC_EEXIST: - return "ICE_AQ_RC_EEXIST"; - case ICE_AQ_RC_EINVAL: - return "ICE_AQ_RC_EINVAL"; - case ICE_AQ_RC_ENOSPC: - return "ICE_AQ_RC_ENOSPC"; - case ICE_AQ_RC_ENOSYS: - return "ICE_AQ_RC_ENOSYS"; - case ICE_AQ_RC_EMODE: - return "ICE_AQ_RC_EMODE"; - case ICE_AQ_RC_ENOSEC: - return "ICE_AQ_RC_ENOSEC"; - case ICE_AQ_RC_EBADSIG: - return "ICE_AQ_RC_EBADSIG"; - case ICE_AQ_RC_ESVN: - return "ICE_AQ_RC_ESVN"; - case ICE_AQ_RC_EBADMAN: - return "ICE_AQ_RC_EBADMAN"; - case ICE_AQ_RC_EBADBUF: - return "ICE_AQ_RC_EBADBUF"; - } - - return "ICE_AQ_RC_UNKNOWN"; -} - -/** - * ice_stat_str - convert status err code to a string - * @stat_err: the status error code to convert - */ -const char *ice_stat_str(enum ice_status stat_err) -{ - switch (stat_err) { - case ICE_SUCCESS: - return "OK"; - case ICE_ERR_PARAM: - return "ICE_ERR_PARAM"; - case ICE_ERR_NOT_IMPL: - return "ICE_ERR_NOT_IMPL"; - case ICE_ERR_NOT_READY: - return "ICE_ERR_NOT_READY"; - case ICE_ERR_NOT_SUPPORTED: - return "ICE_ERR_NOT_SUPPORTED"; - case ICE_ERR_BAD_PTR: - return "ICE_ERR_BAD_PTR"; - case ICE_ERR_INVAL_SIZE: - return "ICE_ERR_INVAL_SIZE"; - case ICE_ERR_DEVICE_NOT_SUPPORTED: - return "ICE_ERR_DEVICE_NOT_SUPPORTED"; - case ICE_ERR_RESET_FAILED: - return "ICE_ERR_RESET_FAILED"; - case ICE_ERR_FW_API_VER: - return "ICE_ERR_FW_API_VER"; - case ICE_ERR_NO_MEMORY: - return "ICE_ERR_NO_MEMORY"; - case ICE_ERR_CFG: - return "ICE_ERR_CFG"; - case ICE_ERR_OUT_OF_RANGE: - return "ICE_ERR_OUT_OF_RANGE"; - case ICE_ERR_ALREADY_EXISTS: - return "ICE_ERR_ALREADY_EXISTS"; - case ICE_ERR_NVM: - return "ICE_ERR_NVM"; - case ICE_ERR_NVM_CHECKSUM: - return "ICE_ERR_NVM_CHECKSUM"; - case ICE_ERR_BUF_TOO_SHORT: - return "ICE_ERR_BUF_TOO_SHORT"; - case ICE_ERR_NVM_BLANK_MODE: - return "ICE_ERR_NVM_BLANK_MODE"; - case ICE_ERR_IN_USE: - return "ICE_ERR_IN_USE"; - case ICE_ERR_MAX_LIMIT: - return "ICE_ERR_MAX_LIMIT"; - case ICE_ERR_RESET_ONGOING: - return "ICE_ERR_RESET_ONGOING"; - case ICE_ERR_HW_TABLE: - return "ICE_ERR_HW_TABLE"; - case ICE_ERR_DOES_NOT_EXIST: - return "ICE_ERR_DOES_NOT_EXIST"; - case ICE_ERR_FW_DDP_MISMATCH: - return "ICE_ERR_FW_DDP_MISMATCH"; - case ICE_ERR_AQ_ERROR: - return "ICE_ERR_AQ_ERROR"; - case ICE_ERR_AQ_TIMEOUT: - return "ICE_ERR_AQ_TIMEOUT"; - case ICE_ERR_AQ_FULL: - return "ICE_ERR_AQ_FULL"; - case ICE_ERR_AQ_NO_WORK: - return "ICE_ERR_AQ_NO_WORK"; - case ICE_ERR_AQ_EMPTY: - return "ICE_ERR_AQ_EMPTY"; - case ICE_ERR_AQ_FW_CRITICAL: - return "ICE_ERR_AQ_FW_CRITICAL"; - } - - return "ICE_ERR_UNKNOWN"; -} - -/** * ice_set_rss_lut - Set RSS LUT * @vsi: Pointer to VSI structure * @lut: Lookup table @@ -6704,7 +7894,7 @@ int ice_set_rss_lut(struct ice_vsi *vsi, u8 *lut, u16 lut_size) { struct ice_aq_get_set_rss_lut_params params = {}; struct ice_hw *hw = &vsi->back->hw; - enum ice_status status; + int status; if (!lut) return -EINVAL; @@ -6715,14 +7905,11 @@ int ice_set_rss_lut(struct ice_vsi *vsi, u8 *lut, u16 lut_size) params.lut = lut; status = ice_aq_set_rss_lut(hw, ¶ms); - if (status) { - dev_err(ice_pf_to_dev(vsi->back), "Cannot set RSS lut, err %s aq_err %s\n", - ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); - return -EIO; - } + if (status) + dev_err(ice_pf_to_dev(vsi->back), "Cannot set RSS lut, err %d aq_err %s\n", + status, libie_aq_str(hw->adminq.sq_last_status)); - return 0; + return status; } /** @@ -6735,20 +7922,17 @@ int ice_set_rss_lut(struct ice_vsi *vsi, u8 *lut, u16 lut_size) int ice_set_rss_key(struct ice_vsi *vsi, u8 *seed) { struct ice_hw *hw = &vsi->back->hw; - enum ice_status status; + int status; if (!seed) return -EINVAL; status = ice_aq_set_rss_key(hw, vsi->idx, (struct ice_aqc_get_set_rss_keys *)seed); - if (status) { - dev_err(ice_pf_to_dev(vsi->back), "Cannot set RSS key, err %s aq_err %s\n", - ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); - return -EIO; - } + if (status) + dev_err(ice_pf_to_dev(vsi->back), "Cannot set RSS key, err %d aq_err %s\n", + status, libie_aq_str(hw->adminq.sq_last_status)); - return 0; + return status; } /** @@ -6763,7 +7947,7 @@ int ice_get_rss_lut(struct ice_vsi *vsi, u8 *lut, u16 lut_size) { struct ice_aq_get_set_rss_lut_params params = {}; struct ice_hw *hw = &vsi->back->hw; - enum ice_status status; + int status; if (!lut) return -EINVAL; @@ -6774,14 +7958,11 @@ int ice_get_rss_lut(struct ice_vsi *vsi, u8 *lut, u16 lut_size) params.lut = lut; status = ice_aq_get_rss_lut(hw, ¶ms); - if (status) { - dev_err(ice_pf_to_dev(vsi->back), "Cannot get RSS lut, err %s aq_err %s\n", - ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); - return -EIO; - } + if (status) + dev_err(ice_pf_to_dev(vsi->back), "Cannot get RSS lut, err %d aq_err %s\n", + status, libie_aq_str(hw->adminq.sq_last_status)); - return 0; + return status; } /** @@ -6794,20 +7975,70 @@ int ice_get_rss_lut(struct ice_vsi *vsi, u8 *lut, u16 lut_size) int ice_get_rss_key(struct ice_vsi *vsi, u8 *seed) { struct ice_hw *hw = &vsi->back->hw; - enum ice_status status; + int status; if (!seed) return -EINVAL; status = ice_aq_get_rss_key(hw, vsi->idx, (struct ice_aqc_get_set_rss_keys *)seed); - if (status) { - dev_err(ice_pf_to_dev(vsi->back), "Cannot get RSS key, err %s aq_err %s\n", - ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); - return -EIO; + if (status) + dev_err(ice_pf_to_dev(vsi->back), "Cannot get RSS key, err %d aq_err %s\n", + status, libie_aq_str(hw->adminq.sq_last_status)); + + return status; +} + +/** + * ice_set_rss_hfunc - Set RSS HASH function + * @vsi: Pointer to VSI structure + * @hfunc: hash function (ICE_AQ_VSI_Q_OPT_RSS_*) + * + * Returns 0 on success, negative on failure + */ +int ice_set_rss_hfunc(struct ice_vsi *vsi, u8 hfunc) +{ + struct ice_hw *hw = &vsi->back->hw; + struct ice_vsi_ctx *ctx; + bool symm; + int err; + + if (hfunc == vsi->rss_hfunc) + return 0; + + if (hfunc != ICE_AQ_VSI_Q_OPT_RSS_HASH_TPLZ && + hfunc != ICE_AQ_VSI_Q_OPT_RSS_HASH_SYM_TPLZ) + return -EOPNOTSUPP; + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); + if (!ctx) + return -ENOMEM; + + ctx->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_Q_OPT_VALID); + ctx->info.q_opt_rss = vsi->info.q_opt_rss; + ctx->info.q_opt_rss &= ~ICE_AQ_VSI_Q_OPT_RSS_HASH_M; + ctx->info.q_opt_rss |= + FIELD_PREP(ICE_AQ_VSI_Q_OPT_RSS_HASH_M, hfunc); + ctx->info.q_opt_tc = vsi->info.q_opt_tc; + ctx->info.q_opt_flags = vsi->info.q_opt_rss; + + err = ice_update_vsi(hw, vsi->idx, ctx, NULL); + if (err) { + dev_err(ice_pf_to_dev(vsi->back), "Failed to configure RSS hash for VSI %d, error %d\n", + vsi->vsi_num, err); + } else { + vsi->info.q_opt_rss = ctx->info.q_opt_rss; + vsi->rss_hfunc = hfunc; + netdev_info(vsi->netdev, "Hash function set to: %sToeplitz\n", + hfunc == ICE_AQ_VSI_Q_OPT_RSS_HASH_SYM_TPLZ ? + "Symmetric " : ""); } + kfree(ctx); + if (err) + return err; - return 0; + /* Fix the symmetry setting for all existing RSS configurations */ + symm = !!(hfunc == ICE_AQ_VSI_Q_OPT_RSS_HASH_SYM_TPLZ); + return ice_set_rss_cfg_symm(hw, vsi, symm); } /** @@ -6825,9 +8056,7 @@ static int ice_bridge_getlink(struct sk_buff *skb, u32 pid, u32 seq, struct net_device *dev, u32 filter_mask, int nlflags) { - struct ice_netdev_priv *np = netdev_priv(dev); - struct ice_vsi *vsi = np->vsi; - struct ice_pf *pf = vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(dev); u16 bmode; bmode = pf->first_sw->bridge_mode; @@ -6848,8 +8077,7 @@ static int ice_vsi_update_bridge_mode(struct ice_vsi *vsi, u16 bmode) struct ice_aqc_vsi_props *vsi_props; struct ice_hw *hw = &vsi->back->hw; struct ice_vsi_ctx *ctxt; - enum ice_status status; - int ret = 0; + int ret; vsi_props = &vsi->info; @@ -6867,12 +8095,10 @@ static int ice_vsi_update_bridge_mode(struct ice_vsi *vsi, u16 bmode) ctxt->info.sw_flags &= ~ICE_AQ_VSI_SW_FLAG_ALLOW_LB; ctxt->info.valid_sections = cpu_to_le16(ICE_AQ_VSI_PROP_SW_VALID); - status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); - if (status) { - dev_err(ice_pf_to_dev(vsi->back), "update VSI for bridge mode failed, bmode = %d err %s aq_err %s\n", - bmode, ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); - ret = -EIO; + ret = ice_update_vsi(hw, vsi->idx, ctxt, NULL); + if (ret) { + dev_err(ice_pf_to_dev(vsi->back), "update VSI for bridge mode failed, bmode = %d err %d aq_err %s\n", + bmode, ret, libie_aq_str(hw->adminq.sq_last_status)); goto out; } /* Update sw flags for book keeping */ @@ -6900,24 +8126,21 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, u16 __always_unused flags, struct netlink_ext_ack __always_unused *extack) { - struct ice_netdev_priv *np = netdev_priv(dev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(dev); struct nlattr *attr, *br_spec; struct ice_hw *hw = &pf->hw; - enum ice_status status; struct ice_sw *pf_sw; int rem, v, err = 0; pf_sw = pf->first_sw; /* find the attribute in the netlink message */ br_spec = nlmsg_find_attr(nlh, sizeof(struct ifinfomsg), IFLA_AF_SPEC); + if (!br_spec) + return -EINVAL; - nla_for_each_nested(attr, br_spec, rem) { - __u16 mode; + nla_for_each_nested_type(attr, IFLA_BRIDGE_MODE, br_spec, rem) { + __u16 mode = nla_get_u16(attr); - if (nla_type(attr) != IFLA_BRIDGE_MODE) - continue; - mode = nla_get_u16(attr); if (mode != BRIDGE_MODE_VEPA && mode != BRIDGE_MODE_VEB) return -EINVAL; /* Continue if bridge mode is not being flipped */ @@ -6938,14 +8161,14 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, /* Update the unicast switch filter rules for the corresponding * switch of the netdev */ - status = ice_update_sw_rule_bridge_mode(hw); - if (status) { - netdev_err(dev, "switch rule update failed, mode = %d err %s aq_err %s\n", - mode, ice_stat_str(status), - ice_aq_str(hw->adminq.sq_last_status)); + err = ice_update_sw_rule_bridge_mode(hw); + if (err) { + netdev_err(dev, "switch rule update failed, mode = %d err %d aq_err %s\n", + mode, err, + libie_aq_str(hw->adminq.sq_last_status)); /* revert hw->evb_veb */ hw->evb_veb = (pf_sw->bridge_mode == BRIDGE_MODE_VEB); - return -EIO; + return err; } pf_sw->bridge_mode = mode; @@ -6959,10 +8182,10 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, * @netdev: network interface device structure * @txqueue: Tx queue */ -static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) +void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) { struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_ring *tx_ring = NULL; + struct ice_tx_ring *tx_ring = NULL; struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; u32 i; @@ -6980,7 +8203,7 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) } /* now that we have an index, find the tx_ring struct */ - for (i = 0; i < vsi->num_txq; i++) + ice_for_each_txq(vsi, i) if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) if (txqueue == vsi->tx_rings[i]->q_index) { tx_ring = vsi->tx_rings[i]; @@ -6998,16 +8221,18 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) if (tx_ring) { struct ice_hw *hw = &pf->hw; - u32 head, val = 0; + u32 head, intr = 0; - head = (rd32(hw, QTX_COMM_HEAD(vsi->txq_map[txqueue])) & - QTX_COMM_HEAD_HEAD_M) >> QTX_COMM_HEAD_HEAD_S; + head = FIELD_GET(QTX_COMM_HEAD_HEAD_M, + rd32(hw, QTX_COMM_HEAD(vsi->txq_map[txqueue]))); /* Read interrupt register */ - val = rd32(hw, GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); + intr = rd32(hw, GLINT_DYN_CTL(tx_ring->q_vector->reg_idx)); netdev_info(netdev, "tx_timeout: VSI_num: %d, Q %u, NTC: 0x%x, HW_HEAD: 0x%x, NTU: 0x%x, INT: 0x%x\n", vsi->vsi_num, txqueue, tx_ring->next_to_clean, - head, tx_ring->next_to_use, val); + head, tx_ring->next_to_use, intr); + + ice_prep_tx_hang_report(pf, tx_ring, vsi->vsi_num, head, intr); } pf->tx_timeout_last_recovery = jiffies; @@ -7037,6 +8262,1291 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) } /** + * ice_setup_tc_cls_flower - flower classifier offloads + * @np: net device to configure + * @filter_dev: device on which filter is added + * @cls_flower: offload data + * @ingress: if the rule is added to an ingress block + * + * Return: 0 if the flower was successfully added or deleted, + * negative error code otherwise. + */ +static int +ice_setup_tc_cls_flower(struct ice_netdev_priv *np, + struct net_device *filter_dev, + struct flow_cls_offload *cls_flower, + bool ingress) +{ + struct ice_vsi *vsi = np->vsi; + + if (cls_flower->common.chain_index) + return -EOPNOTSUPP; + + switch (cls_flower->command) { + case FLOW_CLS_REPLACE: + return ice_add_cls_flower(filter_dev, vsi, cls_flower, ingress); + case FLOW_CLS_DESTROY: + return ice_del_cls_flower(vsi, cls_flower); + default: + return -EINVAL; + } +} + +/** + * ice_setup_tc_block_cb_ingress - callback handler for ingress TC block + * @type: TC SETUP type + * @type_data: TC flower offload data that contains user input + * @cb_priv: netdev private data + * + * Return: 0 if the setup was successful, negative error code otherwise. + */ +static int +ice_setup_tc_block_cb_ingress(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + struct ice_netdev_priv *np = cb_priv; + + switch (type) { + case TC_SETUP_CLSFLOWER: + return ice_setup_tc_cls_flower(np, np->vsi->netdev, + type_data, true); + default: + return -EOPNOTSUPP; + } +} + +/** + * ice_setup_tc_block_cb_egress - callback handler for egress TC block + * @type: TC SETUP type + * @type_data: TC flower offload data that contains user input + * @cb_priv: netdev private data + * + * Return: 0 if the setup was successful, negative error code otherwise. + */ +static int +ice_setup_tc_block_cb_egress(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + struct ice_netdev_priv *np = cb_priv; + + switch (type) { + case TC_SETUP_CLSFLOWER: + return ice_setup_tc_cls_flower(np, np->vsi->netdev, + type_data, false); + default: + return -EOPNOTSUPP; + } +} + +/** + * ice_validate_mqprio_qopt - Validate TCF input parameters + * @vsi: Pointer to VSI + * @mqprio_qopt: input parameters for mqprio queue configuration + * + * This function validates MQPRIO params, such as qcount (power of 2 wherever + * needed), and make sure user doesn't specify qcount and BW rate limit + * for TCs, which are more than "num_tc" + */ +static int +ice_validate_mqprio_qopt(struct ice_vsi *vsi, + struct tc_mqprio_qopt_offload *mqprio_qopt) +{ + int non_power_of_2_qcount = 0; + struct ice_pf *pf = vsi->back; + int max_rss_q_cnt = 0; + u64 sum_min_rate = 0; + struct device *dev; + int i, speed; + u8 num_tc; + + if (vsi->type != ICE_VSI_PF) + return -EINVAL; + + if (mqprio_qopt->qopt.offset[0] != 0 || + mqprio_qopt->qopt.num_tc < 1 || + mqprio_qopt->qopt.num_tc > ICE_CHNL_MAX_TC) + return -EINVAL; + + dev = ice_pf_to_dev(pf); + vsi->ch_rss_size = 0; + num_tc = mqprio_qopt->qopt.num_tc; + speed = ice_get_link_speed_kbps(vsi); + + for (i = 0; num_tc; i++) { + int qcount = mqprio_qopt->qopt.count[i]; + u64 max_rate, min_rate, rem; + + if (!qcount) + return -EINVAL; + + if (is_power_of_2(qcount)) { + if (non_power_of_2_qcount && + qcount > non_power_of_2_qcount) { + dev_err(dev, "qcount[%d] cannot be greater than non power of 2 qcount[%d]\n", + qcount, non_power_of_2_qcount); + return -EINVAL; + } + if (qcount > max_rss_q_cnt) + max_rss_q_cnt = qcount; + } else { + if (non_power_of_2_qcount && + qcount != non_power_of_2_qcount) { + dev_err(dev, "Only one non power of 2 qcount allowed[%d,%d]\n", + qcount, non_power_of_2_qcount); + return -EINVAL; + } + if (qcount < max_rss_q_cnt) { + dev_err(dev, "non power of 2 qcount[%d] cannot be less than other qcount[%d]\n", + qcount, max_rss_q_cnt); + return -EINVAL; + } + max_rss_q_cnt = qcount; + non_power_of_2_qcount = qcount; + } + + /* TC command takes input in K/N/Gbps or K/M/Gbit etc but + * converts the bandwidth rate limit into Bytes/s when + * passing it down to the driver. So convert input bandwidth + * from Bytes/s to Kbps + */ + max_rate = mqprio_qopt->max_rate[i]; + max_rate = div_u64(max_rate, ICE_BW_KBPS_DIVISOR); + + /* min_rate is minimum guaranteed rate and it can't be zero */ + min_rate = mqprio_qopt->min_rate[i]; + min_rate = div_u64(min_rate, ICE_BW_KBPS_DIVISOR); + sum_min_rate += min_rate; + + if (min_rate && min_rate < ICE_MIN_BW_LIMIT) { + dev_err(dev, "TC%d: min_rate(%llu Kbps) < %u Kbps\n", i, + min_rate, ICE_MIN_BW_LIMIT); + return -EINVAL; + } + + if (max_rate && max_rate > speed) { + dev_err(dev, "TC%d: max_rate(%llu Kbps) > link speed of %u Kbps\n", + i, max_rate, speed); + return -EINVAL; + } + + iter_div_u64_rem(min_rate, ICE_MIN_BW_LIMIT, &rem); + if (rem) { + dev_err(dev, "TC%d: Min Rate not multiple of %u Kbps", + i, ICE_MIN_BW_LIMIT); + return -EINVAL; + } + + iter_div_u64_rem(max_rate, ICE_MIN_BW_LIMIT, &rem); + if (rem) { + dev_err(dev, "TC%d: Max Rate not multiple of %u Kbps", + i, ICE_MIN_BW_LIMIT); + return -EINVAL; + } + + /* min_rate can't be more than max_rate, except when max_rate + * is zero (implies max_rate sought is max line rate). In such + * a case min_rate can be more than max. + */ + if (max_rate && min_rate > max_rate) { + dev_err(dev, "min_rate %llu Kbps can't be more than max_rate %llu Kbps\n", + min_rate, max_rate); + return -EINVAL; + } + + if (i >= mqprio_qopt->qopt.num_tc - 1) + break; + if (mqprio_qopt->qopt.offset[i + 1] != + (mqprio_qopt->qopt.offset[i] + qcount)) + return -EINVAL; + } + if (vsi->num_rxq < + (mqprio_qopt->qopt.offset[i] + mqprio_qopt->qopt.count[i])) + return -EINVAL; + if (vsi->num_txq < + (mqprio_qopt->qopt.offset[i] + mqprio_qopt->qopt.count[i])) + return -EINVAL; + + if (sum_min_rate && sum_min_rate > (u64)speed) { + dev_err(dev, "Invalid min Tx rate(%llu) Kbps > speed (%u) Kbps specified\n", + sum_min_rate, speed); + return -EINVAL; + } + + /* make sure vsi->ch_rss_size is set correctly based on TC's qcount */ + vsi->ch_rss_size = max_rss_q_cnt; + + return 0; +} + +/** + * ice_add_vsi_to_fdir - add a VSI to the flow director group for PF + * @pf: ptr to PF device + * @vsi: ptr to VSI + */ +static int ice_add_vsi_to_fdir(struct ice_pf *pf, struct ice_vsi *vsi) +{ + struct device *dev = ice_pf_to_dev(pf); + bool added = false; + struct ice_hw *hw; + int flow; + + if (!(vsi->num_gfltr || vsi->num_bfltr)) + return -EINVAL; + + hw = &pf->hw; + for (flow = 0; flow < ICE_FLTR_PTYPE_MAX; flow++) { + struct ice_fd_hw_prof *prof; + int tun, status; + u64 entry_h; + + if (!(hw->fdir_prof && hw->fdir_prof[flow] && + hw->fdir_prof[flow]->cnt)) + continue; + + for (tun = 0; tun < ICE_FD_HW_SEG_MAX; tun++) { + enum ice_flow_priority prio; + + /* add this VSI to FDir profile for this flow */ + prio = ICE_FLOW_PRIO_NORMAL; + prof = hw->fdir_prof[flow]; + status = ice_flow_add_entry(hw, ICE_BLK_FD, + prof->prof_id[tun], + prof->vsi_h[0], vsi->idx, + prio, prof->fdir_seg[tun], + &entry_h); + if (status) { + dev_err(dev, "channel VSI idx %d, not able to add to group %d\n", + vsi->idx, flow); + continue; + } + + prof->entry_h[prof->cnt][tun] = entry_h; + } + + /* store VSI for filter replay and delete */ + prof->vsi_h[prof->cnt] = vsi->idx; + prof->cnt++; + + added = true; + dev_dbg(dev, "VSI idx %d added to fdir group %d\n", vsi->idx, + flow); + } + + if (!added) + dev_dbg(dev, "VSI idx %d not added to fdir groups\n", vsi->idx); + + return 0; +} + +/** + * ice_add_channel - add a channel by adding VSI + * @pf: ptr to PF device + * @sw_id: underlying HW switching element ID + * @ch: ptr to channel structure + * + * Add a channel (VSI) using add_vsi and queue_map + */ +static int ice_add_channel(struct ice_pf *pf, u16 sw_id, struct ice_channel *ch) +{ + struct device *dev = ice_pf_to_dev(pf); + struct ice_vsi *vsi; + + if (ch->type != ICE_VSI_CHNL) { + dev_err(dev, "add new VSI failed, ch->type %d\n", ch->type); + return -EINVAL; + } + + vsi = ice_chnl_vsi_setup(pf, pf->hw.port_info, ch); + if (!vsi || vsi->type != ICE_VSI_CHNL) { + dev_err(dev, "create chnl VSI failure\n"); + return -EINVAL; + } + + ice_add_vsi_to_fdir(pf, vsi); + + ch->sw_id = sw_id; + ch->vsi_num = vsi->vsi_num; + ch->info.mapping_flags = vsi->info.mapping_flags; + ch->ch_vsi = vsi; + /* set the back pointer of channel for newly created VSI */ + vsi->ch = ch; + + memcpy(&ch->info.q_mapping, &vsi->info.q_mapping, + sizeof(vsi->info.q_mapping)); + memcpy(&ch->info.tc_mapping, vsi->info.tc_mapping, + sizeof(vsi->info.tc_mapping)); + + return 0; +} + +/** + * ice_chnl_cfg_res + * @vsi: the VSI being setup + * @ch: ptr to channel structure + * + * Configure channel specific resources such as rings, vector. + */ +static void ice_chnl_cfg_res(struct ice_vsi *vsi, struct ice_channel *ch) +{ + int i; + + for (i = 0; i < ch->num_txq; i++) { + struct ice_q_vector *tx_q_vector, *rx_q_vector; + struct ice_ring_container *rc; + struct ice_tx_ring *tx_ring; + struct ice_rx_ring *rx_ring; + + tx_ring = vsi->tx_rings[ch->base_q + i]; + rx_ring = vsi->rx_rings[ch->base_q + i]; + if (!tx_ring || !rx_ring) + continue; + + /* setup ring being channel enabled */ + tx_ring->ch = ch; + rx_ring->ch = ch; + + /* following code block sets up vector specific attributes */ + tx_q_vector = tx_ring->q_vector; + rx_q_vector = rx_ring->q_vector; + if (!tx_q_vector && !rx_q_vector) + continue; + + if (tx_q_vector) { + tx_q_vector->ch = ch; + /* setup Tx and Rx ITR setting if DIM is off */ + rc = &tx_q_vector->tx; + if (!ITR_IS_DYNAMIC(rc)) + ice_write_itr(rc, rc->itr_setting); + } + if (rx_q_vector) { + rx_q_vector->ch = ch; + /* setup Tx and Rx ITR setting if DIM is off */ + rc = &rx_q_vector->rx; + if (!ITR_IS_DYNAMIC(rc)) + ice_write_itr(rc, rc->itr_setting); + } + } + + /* it is safe to assume that, if channel has non-zero num_t[r]xq, then + * GLINT_ITR register would have written to perform in-context + * update, hence perform flush + */ + if (ch->num_txq || ch->num_rxq) + ice_flush(&vsi->back->hw); +} + +/** + * ice_cfg_chnl_all_res - configure channel resources + * @vsi: pte to main_vsi + * @ch: ptr to channel structure + * + * This function configures channel specific resources such as flow-director + * counter index, and other resources such as queues, vectors, ITR settings + */ +static void +ice_cfg_chnl_all_res(struct ice_vsi *vsi, struct ice_channel *ch) +{ + /* configure channel (aka ADQ) resources such as queues, vectors, + * ITR settings for channel specific vectors and anything else + */ + ice_chnl_cfg_res(vsi, ch); +} + +/** + * ice_setup_hw_channel - setup new channel + * @pf: ptr to PF device + * @vsi: the VSI being setup + * @ch: ptr to channel structure + * @sw_id: underlying HW switching element ID + * @type: type of channel to be created (VMDq2/VF) + * + * Setup new channel (VSI) based on specified type (VMDq2/VF) + * and configures Tx rings accordingly + */ +static int +ice_setup_hw_channel(struct ice_pf *pf, struct ice_vsi *vsi, + struct ice_channel *ch, u16 sw_id, u8 type) +{ + struct device *dev = ice_pf_to_dev(pf); + int ret; + + ch->base_q = vsi->next_base_q; + ch->type = type; + + ret = ice_add_channel(pf, sw_id, ch); + if (ret) { + dev_err(dev, "failed to add_channel using sw_id %u\n", sw_id); + return ret; + } + + /* configure/setup ADQ specific resources */ + ice_cfg_chnl_all_res(vsi, ch); + + /* make sure to update the next_base_q so that subsequent channel's + * (aka ADQ) VSI queue map is correct + */ + vsi->next_base_q = vsi->next_base_q + ch->num_rxq; + dev_dbg(dev, "added channel: vsi_num %u, num_rxq %u\n", ch->vsi_num, + ch->num_rxq); + + return 0; +} + +/** + * ice_setup_channel - setup new channel using uplink element + * @pf: ptr to PF device + * @vsi: the VSI being setup + * @ch: ptr to channel structure + * + * Setup new channel (VSI) based on specified type (VMDq2/VF) + * and uplink switching element + */ +static bool +ice_setup_channel(struct ice_pf *pf, struct ice_vsi *vsi, + struct ice_channel *ch) +{ + struct device *dev = ice_pf_to_dev(pf); + u16 sw_id; + int ret; + + if (vsi->type != ICE_VSI_PF) { + dev_err(dev, "unsupported parent VSI type(%d)\n", vsi->type); + return false; + } + + sw_id = pf->first_sw->sw_id; + + /* create channel (VSI) */ + ret = ice_setup_hw_channel(pf, vsi, ch, sw_id, ICE_VSI_CHNL); + if (ret) { + dev_err(dev, "failed to setup hw_channel\n"); + return false; + } + dev_dbg(dev, "successfully created channel()\n"); + + return ch->ch_vsi ? true : false; +} + +/** + * ice_set_bw_limit - setup BW limit for Tx traffic based on max_tx_rate + * @vsi: VSI to be configured + * @max_tx_rate: max Tx rate in Kbps to be configured as maximum BW limit + * @min_tx_rate: min Tx rate in Kbps to be configured as minimum BW limit + */ +static int +ice_set_bw_limit(struct ice_vsi *vsi, u64 max_tx_rate, u64 min_tx_rate) +{ + int err; + + err = ice_set_min_bw_limit(vsi, min_tx_rate); + if (err) + return err; + + return ice_set_max_bw_limit(vsi, max_tx_rate); +} + +/** + * ice_create_q_channel - function to create channel + * @vsi: VSI to be configured + * @ch: ptr to channel (it contains channel specific params) + * + * This function creates channel (VSI) using num_queues specified by user, + * reconfigs RSS if needed. + */ +static int ice_create_q_channel(struct ice_vsi *vsi, struct ice_channel *ch) +{ + struct ice_pf *pf = vsi->back; + struct device *dev; + + if (!ch) + return -EINVAL; + + dev = ice_pf_to_dev(pf); + if (!ch->num_txq || !ch->num_rxq) { + dev_err(dev, "Invalid num_queues requested: %d\n", ch->num_rxq); + return -EINVAL; + } + + if (!vsi->cnt_q_avail || vsi->cnt_q_avail < ch->num_txq) { + dev_err(dev, "cnt_q_avail (%u) less than num_queues %d\n", + vsi->cnt_q_avail, ch->num_txq); + return -EINVAL; + } + + if (!ice_setup_channel(pf, vsi, ch)) { + dev_info(dev, "Failed to setup channel\n"); + return -EINVAL; + } + /* configure BW rate limit */ + if (ch->ch_vsi && (ch->max_tx_rate || ch->min_tx_rate)) { + int ret; + + ret = ice_set_bw_limit(ch->ch_vsi, ch->max_tx_rate, + ch->min_tx_rate); + if (ret) + dev_err(dev, "failed to set Tx rate of %llu Kbps for VSI(%u)\n", + ch->max_tx_rate, ch->ch_vsi->vsi_num); + else + dev_dbg(dev, "set Tx rate of %llu Kbps for VSI(%u)\n", + ch->max_tx_rate, ch->ch_vsi->vsi_num); + } + + vsi->cnt_q_avail -= ch->num_txq; + + return 0; +} + +/** + * ice_rem_all_chnl_fltrs - removes all channel filters + * @pf: ptr to PF, TC-flower based filter are tracked at PF level + * + * Remove all advanced switch filters only if they are channel specific + * tc-flower based filter + */ +static void ice_rem_all_chnl_fltrs(struct ice_pf *pf) +{ + struct ice_tc_flower_fltr *fltr; + struct hlist_node *node; + + /* to remove all channel filters, iterate an ordered list of filters */ + hlist_for_each_entry_safe(fltr, node, + &pf->tc_flower_fltr_list, + tc_flower_node) { + struct ice_rule_query_data rule; + int status; + + /* for now process only channel specific filters */ + if (!ice_is_chnl_fltr(fltr)) + continue; + + rule.rid = fltr->rid; + rule.rule_id = fltr->rule_id; + rule.vsi_handle = fltr->dest_vsi_handle; + status = ice_rem_adv_rule_by_id(&pf->hw, &rule); + if (status) { + if (status == -ENOENT) + dev_dbg(ice_pf_to_dev(pf), "TC flower filter (rule_id %u) does not exist\n", + rule.rule_id); + else + dev_err(ice_pf_to_dev(pf), "failed to delete TC flower filter, status %d\n", + status); + } else if (fltr->dest_vsi) { + /* update advanced switch filter count */ + if (fltr->dest_vsi->type == ICE_VSI_CHNL) { + u32 flags = fltr->flags; + + fltr->dest_vsi->num_chnl_fltr--; + if (flags & (ICE_TC_FLWR_FIELD_DST_MAC | + ICE_TC_FLWR_FIELD_ENC_DST_MAC)) + pf->num_dmac_chnl_fltrs--; + } + } + + hlist_del(&fltr->tc_flower_node); + kfree(fltr); + } +} + +/** + * ice_remove_q_channels - Remove queue channels for the TCs + * @vsi: VSI to be configured + * @rem_fltr: delete advanced switch filter or not + * + * Remove queue channels for the TCs + */ +static void ice_remove_q_channels(struct ice_vsi *vsi, bool rem_fltr) +{ + struct ice_channel *ch, *ch_tmp; + struct ice_pf *pf = vsi->back; + int i; + + /* remove all tc-flower based filter if they are channel filters only */ + if (rem_fltr) + ice_rem_all_chnl_fltrs(pf); + + /* remove ntuple filters since queue configuration is being changed */ + if (vsi->netdev->features & NETIF_F_NTUPLE) { + struct ice_hw *hw = &pf->hw; + + mutex_lock(&hw->fdir_fltr_lock); + ice_fdir_del_all_fltrs(vsi); + mutex_unlock(&hw->fdir_fltr_lock); + } + + /* perform cleanup for channels if they exist */ + list_for_each_entry_safe(ch, ch_tmp, &vsi->ch_list, list) { + struct ice_vsi *ch_vsi; + + list_del(&ch->list); + ch_vsi = ch->ch_vsi; + if (!ch_vsi) { + kfree(ch); + continue; + } + + /* Reset queue contexts */ + for (i = 0; i < ch->num_rxq; i++) { + struct ice_tx_ring *tx_ring; + struct ice_rx_ring *rx_ring; + + tx_ring = vsi->tx_rings[ch->base_q + i]; + rx_ring = vsi->rx_rings[ch->base_q + i]; + if (tx_ring) { + tx_ring->ch = NULL; + if (tx_ring->q_vector) + tx_ring->q_vector->ch = NULL; + } + if (rx_ring) { + rx_ring->ch = NULL; + if (rx_ring->q_vector) + rx_ring->q_vector->ch = NULL; + } + } + + /* Release FD resources for the channel VSI */ + ice_fdir_rem_adq_chnl(&pf->hw, ch->ch_vsi->idx); + + /* clear the VSI from scheduler tree */ + ice_rm_vsi_lan_cfg(ch->ch_vsi->port_info, ch->ch_vsi->idx); + + /* Delete VSI from FW, PF and HW VSI arrays */ + ice_vsi_delete(ch->ch_vsi); + + /* free the channel */ + kfree(ch); + } + + /* clear the channel VSI map which is stored in main VSI */ + ice_for_each_chnl_tc(i) + vsi->tc_map_vsi[i] = NULL; + + /* reset main VSI's all TC information */ + vsi->all_enatc = 0; + vsi->all_numtc = 0; +} + +/** + * ice_rebuild_channels - rebuild channel + * @pf: ptr to PF + * + * Recreate channel VSIs and replay filters + */ +static int ice_rebuild_channels(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + struct ice_vsi *main_vsi; + bool rem_adv_fltr = true; + struct ice_channel *ch; + struct ice_vsi *vsi; + int tc_idx = 1; + int i, err; + + main_vsi = ice_get_main_vsi(pf); + if (!main_vsi) + return 0; + + if (!test_bit(ICE_FLAG_TC_MQPRIO, pf->flags) || + main_vsi->old_numtc == 1) + return 0; /* nothing to be done */ + + /* reconfigure main VSI based on old value of TC and cached values + * for MQPRIO opts + */ + err = ice_vsi_cfg_tc(main_vsi, main_vsi->old_ena_tc); + if (err) { + dev_err(dev, "failed configuring TC(ena_tc:0x%02x) for HW VSI=%u\n", + main_vsi->old_ena_tc, main_vsi->vsi_num); + return err; + } + + /* rebuild ADQ VSIs */ + ice_for_each_vsi(pf, i) { + enum ice_vsi_type type; + + vsi = pf->vsi[i]; + if (!vsi || vsi->type != ICE_VSI_CHNL) + continue; + + type = vsi->type; + + /* rebuild ADQ VSI */ + err = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_INIT); + if (err) { + dev_err(dev, "VSI (type:%s) at index %d rebuild failed, err %d\n", + ice_vsi_type_str(type), vsi->idx, err); + goto cleanup; + } + + /* Re-map HW VSI number, using VSI handle that has been + * previously validated in ice_replay_vsi() call above + */ + vsi->vsi_num = ice_get_hw_vsi_num(&pf->hw, vsi->idx); + + /* replay filters for the VSI */ + err = ice_replay_vsi(&pf->hw, vsi->idx); + if (err) { + dev_err(dev, "VSI (type:%s) replay failed, err %d, VSI index %d\n", + ice_vsi_type_str(type), err, vsi->idx); + rem_adv_fltr = false; + goto cleanup; + } + dev_info(dev, "VSI (type:%s) at index %d rebuilt successfully\n", + ice_vsi_type_str(type), vsi->idx); + + /* store ADQ VSI at correct TC index in main VSI's + * map of TC to VSI + */ + main_vsi->tc_map_vsi[tc_idx++] = vsi; + } + + /* ADQ VSI(s) has been rebuilt successfully, so setup + * channel for main VSI's Tx and Rx rings + */ + list_for_each_entry(ch, &main_vsi->ch_list, list) { + struct ice_vsi *ch_vsi; + + ch_vsi = ch->ch_vsi; + if (!ch_vsi) + continue; + + /* reconfig channel resources */ + ice_cfg_chnl_all_res(main_vsi, ch); + + /* replay BW rate limit if it is non-zero */ + if (!ch->max_tx_rate && !ch->min_tx_rate) + continue; + + err = ice_set_bw_limit(ch_vsi, ch->max_tx_rate, + ch->min_tx_rate); + if (err) + dev_err(dev, "failed (err:%d) to rebuild BW rate limit, max_tx_rate: %llu Kbps, min_tx_rate: %llu Kbps for VSI(%u)\n", + err, ch->max_tx_rate, ch->min_tx_rate, + ch_vsi->vsi_num); + else + dev_dbg(dev, "successfully rebuild BW rate limit, max_tx_rate: %llu Kbps, min_tx_rate: %llu Kbps for VSI(%u)\n", + ch->max_tx_rate, ch->min_tx_rate, + ch_vsi->vsi_num); + } + + /* reconfig RSS for main VSI */ + if (main_vsi->ch_rss_size) + ice_vsi_cfg_rss_lut_key(main_vsi); + + return 0; + +cleanup: + ice_remove_q_channels(main_vsi, rem_adv_fltr); + return err; +} + +/** + * ice_create_q_channels - Add queue channel for the given TCs + * @vsi: VSI to be configured + * + * Configures queue channel mapping to the given TCs + */ +static int ice_create_q_channels(struct ice_vsi *vsi) +{ + struct ice_pf *pf = vsi->back; + struct ice_channel *ch; + int ret = 0, i; + + ice_for_each_chnl_tc(i) { + if (!(vsi->all_enatc & BIT(i))) + continue; + + ch = kzalloc(sizeof(*ch), GFP_KERNEL); + if (!ch) { + ret = -ENOMEM; + goto err_free; + } + INIT_LIST_HEAD(&ch->list); + ch->num_rxq = vsi->mqprio_qopt.qopt.count[i]; + ch->num_txq = vsi->mqprio_qopt.qopt.count[i]; + ch->base_q = vsi->mqprio_qopt.qopt.offset[i]; + ch->max_tx_rate = vsi->mqprio_qopt.max_rate[i]; + ch->min_tx_rate = vsi->mqprio_qopt.min_rate[i]; + + /* convert to Kbits/s */ + if (ch->max_tx_rate) + ch->max_tx_rate = div_u64(ch->max_tx_rate, + ICE_BW_KBPS_DIVISOR); + if (ch->min_tx_rate) + ch->min_tx_rate = div_u64(ch->min_tx_rate, + ICE_BW_KBPS_DIVISOR); + + ret = ice_create_q_channel(vsi, ch); + if (ret) { + dev_err(ice_pf_to_dev(pf), + "failed creating channel TC:%d\n", i); + kfree(ch); + goto err_free; + } + list_add_tail(&ch->list, &vsi->ch_list); + vsi->tc_map_vsi[i] = ch->ch_vsi; + dev_dbg(ice_pf_to_dev(pf), + "successfully created channel: VSI %p\n", ch->ch_vsi); + } + return 0; + +err_free: + ice_remove_q_channels(vsi, false); + + return ret; +} + +/** + * ice_setup_tc_mqprio_qdisc - configure multiple traffic classes + * @netdev: net device to configure + * @type_data: TC offload data + */ +static int ice_setup_tc_mqprio_qdisc(struct net_device *netdev, void *type_data) +{ + struct tc_mqprio_qopt_offload *mqprio_qopt = type_data; + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + struct ice_pf *pf = vsi->back; + u16 mode, ena_tc_qdisc = 0; + int cur_txq, cur_rxq; + u8 hw = 0, num_tcf; + struct device *dev; + int ret, i; + + dev = ice_pf_to_dev(pf); + num_tcf = mqprio_qopt->qopt.num_tc; + hw = mqprio_qopt->qopt.hw; + mode = mqprio_qopt->mode; + if (!hw) { + clear_bit(ICE_FLAG_TC_MQPRIO, pf->flags); + vsi->ch_rss_size = 0; + memcpy(&vsi->mqprio_qopt, mqprio_qopt, sizeof(*mqprio_qopt)); + goto config_tcf; + } + + /* Generate queue region map for number of TCF requested */ + for (i = 0; i < num_tcf; i++) + ena_tc_qdisc |= BIT(i); + + switch (mode) { + case TC_MQPRIO_MODE_CHANNEL: + + if (pf->hw.port_info->is_custom_tx_enabled) { + dev_err(dev, "Custom Tx scheduler feature enabled, can't configure ADQ\n"); + return -EBUSY; + } + ice_tear_down_devlink_rate_tree(pf); + + ret = ice_validate_mqprio_qopt(vsi, mqprio_qopt); + if (ret) { + netdev_err(netdev, "failed to validate_mqprio_qopt(), ret %d\n", + ret); + return ret; + } + memcpy(&vsi->mqprio_qopt, mqprio_qopt, sizeof(*mqprio_qopt)); + set_bit(ICE_FLAG_TC_MQPRIO, pf->flags); + /* don't assume state of hw_tc_offload during driver load + * and set the flag for TC flower filter if hw_tc_offload + * already ON + */ + if (vsi->netdev->features & NETIF_F_HW_TC) + set_bit(ICE_FLAG_CLS_FLOWER, pf->flags); + break; + default: + return -EINVAL; + } + +config_tcf: + + /* Requesting same TCF configuration as already enabled */ + if (ena_tc_qdisc == vsi->tc_cfg.ena_tc && + mode != TC_MQPRIO_MODE_CHANNEL) + return 0; + + /* Pause VSI queues */ + ice_dis_vsi(vsi, true); + + if (!hw && !test_bit(ICE_FLAG_TC_MQPRIO, pf->flags)) + ice_remove_q_channels(vsi, true); + + if (!hw && !test_bit(ICE_FLAG_TC_MQPRIO, pf->flags)) { + vsi->req_txq = min_t(int, ice_get_avail_txq_count(pf), + num_online_cpus()); + vsi->req_rxq = min_t(int, ice_get_avail_rxq_count(pf), + num_online_cpus()); + } else { + /* logic to rebuild VSI, same like ethtool -L */ + u16 offset = 0, qcount_tx = 0, qcount_rx = 0; + + for (i = 0; i < num_tcf; i++) { + if (!(ena_tc_qdisc & BIT(i))) + continue; + + offset = vsi->mqprio_qopt.qopt.offset[i]; + qcount_rx = vsi->mqprio_qopt.qopt.count[i]; + qcount_tx = vsi->mqprio_qopt.qopt.count[i]; + } + vsi->req_txq = offset + qcount_tx; + vsi->req_rxq = offset + qcount_rx; + + /* store away original rss_size info, so that it gets reused + * form ice_vsi_rebuild during tc-qdisc delete stage - to + * determine, what should be the rss_sizefor main VSI + */ + vsi->orig_rss_size = vsi->rss_size; + } + + /* save current values of Tx and Rx queues before calling VSI rebuild + * for fallback option + */ + cur_txq = vsi->num_txq; + cur_rxq = vsi->num_rxq; + + /* proceed with rebuild main VSI using correct number of queues */ + ret = ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT); + if (ret) { + /* fallback to current number of queues */ + dev_info(dev, "Rebuild failed with new queues, try with current number of queues\n"); + vsi->req_txq = cur_txq; + vsi->req_rxq = cur_rxq; + clear_bit(ICE_RESET_FAILED, pf->state); + if (ice_vsi_rebuild(vsi, ICE_VSI_FLAG_NO_INIT)) { + dev_err(dev, "Rebuild of main VSI failed again\n"); + return ret; + } + } + + vsi->all_numtc = num_tcf; + vsi->all_enatc = ena_tc_qdisc; + ret = ice_vsi_cfg_tc(vsi, ena_tc_qdisc); + if (ret) { + netdev_err(netdev, "failed configuring TC for VSI id=%d\n", + vsi->vsi_num); + goto exit; + } + + if (test_bit(ICE_FLAG_TC_MQPRIO, pf->flags)) { + u64 max_tx_rate = vsi->mqprio_qopt.max_rate[0]; + u64 min_tx_rate = vsi->mqprio_qopt.min_rate[0]; + + /* set TC0 rate limit if specified */ + if (max_tx_rate || min_tx_rate) { + /* convert to Kbits/s */ + if (max_tx_rate) + max_tx_rate = div_u64(max_tx_rate, ICE_BW_KBPS_DIVISOR); + if (min_tx_rate) + min_tx_rate = div_u64(min_tx_rate, ICE_BW_KBPS_DIVISOR); + + ret = ice_set_bw_limit(vsi, max_tx_rate, min_tx_rate); + if (!ret) { + dev_dbg(dev, "set Tx rate max %llu min %llu for VSI(%u)\n", + max_tx_rate, min_tx_rate, vsi->vsi_num); + } else { + dev_err(dev, "failed to set Tx rate max %llu min %llu for VSI(%u)\n", + max_tx_rate, min_tx_rate, vsi->vsi_num); + goto exit; + } + } + ret = ice_create_q_channels(vsi); + if (ret) { + netdev_err(netdev, "failed configuring queue channels\n"); + goto exit; + } else { + netdev_dbg(netdev, "successfully configured channels\n"); + } + } + + if (vsi->ch_rss_size) + ice_vsi_cfg_rss_lut_key(vsi); + +exit: + /* if error, reset the all_numtc and all_enatc */ + if (ret) { + vsi->all_numtc = 0; + vsi->all_enatc = 0; + } + /* resume VSI */ + ice_ena_vsi(vsi, true); + + return ret; +} + +/** + * ice_cfg_txtime - configure Tx Time for the Tx ring + * @tx_ring: pointer to the Tx ring structure + * + * Return: 0 on success, negative value on failure. + */ +static int ice_cfg_txtime(struct ice_tx_ring *tx_ring) +{ + int err, timeout = 50; + struct ice_vsi *vsi; + struct device *dev; + struct ice_pf *pf; + u32 queue; + + if (!tx_ring) + return -EINVAL; + + vsi = tx_ring->vsi; + pf = vsi->back; + while (test_and_set_bit(ICE_CFG_BUSY, pf->state)) { + timeout--; + if (!timeout) + return -EBUSY; + usleep_range(1000, 2000); + } + + queue = tx_ring->q_index; + dev = ice_pf_to_dev(pf); + + /* Ignore return value, and always attempt to enable queue. */ + ice_qp_dis(vsi, queue); + + err = ice_qp_ena(vsi, queue); + if (err) + dev_err(dev, "Failed to enable Tx queue %d for TxTime configuration\n", + queue); + + clear_bit(ICE_CFG_BUSY, pf->state); + return err; +} + +/** + * ice_offload_txtime - set earliest TxTime first + * @netdev: network interface device structure + * @qopt_off: etf queue option offload from the skb to set + * + * Return: 0 on success, negative value on failure. + */ +static int ice_offload_txtime(struct net_device *netdev, + void *qopt_off) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_pf *pf = np->vsi->back; + struct tc_etf_qopt_offload *qopt; + struct ice_vsi *vsi = np->vsi; + struct ice_tx_ring *tx_ring; + int ret = 0; + + if (!ice_is_feature_supported(pf, ICE_F_TXTIME)) + return -EOPNOTSUPP; + + qopt = qopt_off; + if (!qopt_off || qopt->queue < 0 || qopt->queue >= vsi->num_txq) + return -EINVAL; + + if (qopt->enable) + set_bit(qopt->queue, pf->txtime_txqs); + else + clear_bit(qopt->queue, pf->txtime_txqs); + + if (netif_running(vsi->netdev)) { + tx_ring = vsi->tx_rings[qopt->queue]; + ret = ice_cfg_txtime(tx_ring); + if (ret) + goto err; + } + + netdev_info(netdev, "%s TxTime on queue: %i\n", + str_enable_disable(qopt->enable), qopt->queue); + return 0; + +err: + netdev_err(netdev, "Failed to %s TxTime on queue: %i\n", + str_enable_disable(qopt->enable), qopt->queue); + + if (qopt->enable) + clear_bit(qopt->queue, pf->txtime_txqs); + return ret; +} + +static LIST_HEAD(ice_block_cb_list); + +static int +ice_setup_tc(struct net_device *netdev, enum tc_setup_type type, + void *type_data) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + enum flow_block_binder_type binder_type; + struct iidc_rdma_core_dev_info *cdev; + struct ice_pf *pf = np->vsi->back; + flow_setup_cb_t *flower_handler; + bool locked = false; + int err; + + switch (type) { + case TC_SETUP_BLOCK: + binder_type = + ((struct flow_block_offload *)type_data)->binder_type; + + switch (binder_type) { + case FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS: + flower_handler = ice_setup_tc_block_cb_ingress; + break; + case FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS: + flower_handler = ice_setup_tc_block_cb_egress; + break; + default: + return -EOPNOTSUPP; + } + + return flow_block_cb_setup_simple(type_data, + &ice_block_cb_list, + flower_handler, + np, np, false); + case TC_SETUP_QDISC_MQPRIO: + if (ice_is_eswitch_mode_switchdev(pf)) { + netdev_err(netdev, "TC MQPRIO offload not supported, switchdev is enabled\n"); + return -EOPNOTSUPP; + } + + cdev = pf->cdev_info; + if (cdev && cdev->adev) { + mutex_lock(&pf->adev_mutex); + device_lock(&cdev->adev->dev); + locked = true; + if (cdev->adev->dev.driver) { + netdev_err(netdev, "Cannot change qdisc when RDMA is active\n"); + err = -EBUSY; + goto adev_unlock; + } + } + + /* setup traffic classifier for receive side */ + mutex_lock(&pf->tc_mutex); + err = ice_setup_tc_mqprio_qdisc(netdev, type_data); + mutex_unlock(&pf->tc_mutex); + +adev_unlock: + if (locked) { + device_unlock(&cdev->adev->dev); + mutex_unlock(&pf->adev_mutex); + } + return err; + case TC_SETUP_QDISC_ETF: + return ice_offload_txtime(netdev, type_data); + default: + return -EOPNOTSUPP; + } + return -EOPNOTSUPP; +} + +static struct ice_indr_block_priv * +ice_indr_block_priv_lookup(struct ice_netdev_priv *np, + struct net_device *netdev) +{ + struct ice_indr_block_priv *cb_priv; + + list_for_each_entry(cb_priv, &np->tc_indr_block_priv_list, list) { + if (!cb_priv->netdev) + return NULL; + if (cb_priv->netdev == netdev) + return cb_priv; + } + return NULL; +} + +static int +ice_indr_setup_block_cb(enum tc_setup_type type, void *type_data, + void *indr_priv) +{ + struct ice_indr_block_priv *priv = indr_priv; + struct ice_netdev_priv *np = priv->np; + + switch (type) { + case TC_SETUP_CLSFLOWER: + return ice_setup_tc_cls_flower(np, priv->netdev, + (struct flow_cls_offload *) + type_data, false); + default: + return -EOPNOTSUPP; + } +} + +static int +ice_indr_setup_tc_block(struct net_device *netdev, struct Qdisc *sch, + struct ice_netdev_priv *np, + struct flow_block_offload *f, void *data, + void (*cleanup)(struct flow_block_cb *block_cb)) +{ + struct ice_indr_block_priv *indr_priv; + struct flow_block_cb *block_cb; + + if (!ice_is_tunnel_supported(netdev) && + !(is_vlan_dev(netdev) && + vlan_dev_real_dev(netdev) == np->vsi->netdev)) + return -EOPNOTSUPP; + + if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) + return -EOPNOTSUPP; + + switch (f->command) { + case FLOW_BLOCK_BIND: + indr_priv = ice_indr_block_priv_lookup(np, netdev); + if (indr_priv) + return -EEXIST; + + indr_priv = kzalloc(sizeof(*indr_priv), GFP_KERNEL); + if (!indr_priv) + return -ENOMEM; + + indr_priv->netdev = netdev; + indr_priv->np = np; + list_add(&indr_priv->list, &np->tc_indr_block_priv_list); + + block_cb = + flow_indr_block_cb_alloc(ice_indr_setup_block_cb, + indr_priv, indr_priv, + ice_rep_indr_tc_block_unbind, + f, netdev, sch, data, np, + cleanup); + + if (IS_ERR(block_cb)) { + list_del(&indr_priv->list); + kfree(indr_priv); + return PTR_ERR(block_cb); + } + flow_block_cb_add(block_cb, f); + list_add_tail(&block_cb->driver_list, &ice_block_cb_list); + break; + case FLOW_BLOCK_UNBIND: + indr_priv = ice_indr_block_priv_lookup(np, netdev); + if (!indr_priv) + return -ENOENT; + + block_cb = flow_block_cb_lookup(f->block, + ice_indr_setup_block_cb, + indr_priv); + if (!block_cb) + return -ENOENT; + + flow_indr_block_cb_remove(block_cb, f); + + list_del(&block_cb->driver_list); + break; + default: + return -EOPNOTSUPP; + } + return 0; +} + +static int +ice_indr_setup_tc_cb(struct net_device *netdev, struct Qdisc *sch, + void *cb_priv, enum tc_setup_type type, void *type_data, + void *data, + void (*cleanup)(struct flow_block_cb *block_cb)) +{ + switch (type) { + case TC_SETUP_BLOCK: + return ice_indr_setup_tc_block(netdev, sch, cb_priv, type_data, + data, cleanup); + + default: + return -EOPNOTSUPP; + } +} + +/** * ice_open - Called when a network interface becomes active * @netdev: network interface device structure * @@ -7050,8 +9560,7 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) */ int ice_open(struct net_device *netdev) { - struct ice_netdev_priv *np = netdev_priv(netdev); - struct ice_pf *pf = np->vsi->back; + struct ice_pf *pf = ice_netdev_to_pf(netdev); if (ice_is_reset_in_progress(pf->state)) { netdev_err(netdev, "can't open net device while reset is in progress"); @@ -7076,7 +9585,6 @@ int ice_open_internal(struct net_device *netdev) struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; struct ice_port_info *pi; - enum ice_status status; int err; if (test_bit(ICE_NEEDS_RESTART, pf->state)) { @@ -7087,14 +9595,13 @@ int ice_open_internal(struct net_device *netdev) netif_carrier_off(netdev); pi = vsi->port_info; - status = ice_update_link_info(pi); - if (status) { - netdev_err(netdev, "Failed to get link info, error %s\n", - ice_stat_str(status)); - return -EIO; + err = ice_update_link_info(pi); + if (err) { + netdev_err(netdev, "Failed to get link info, error %d\n", err); + return err; } - ice_check_module_power(pf, pi->phy.link_info.link_cfg_err); + ice_check_link_cfg_err(pf, pi->phy.link_info.link_cfg_err); /* Set PHY if there is media, otherwise, turn off PHY */ if (pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE) { @@ -7151,6 +9658,22 @@ int ice_stop(struct net_device *netdev) return -EBUSY; } + if (test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, vsi->back->flags)) { + int link_err = ice_force_phys_link_state(vsi, false); + + if (link_err) { + if (link_err == -ENOMEDIUM) + netdev_info(vsi->netdev, "Skipping link reconfig - no media attached, VSI %d\n", + vsi->vsi_num); + else + netdev_err(vsi->netdev, "Failed to set physical link down, VSI %d error %d\n", + vsi->vsi_num, link_err); + + ice_vsi_close(vsi); + return -EIO; + } + } + ice_vsi_close(vsi); return 0; @@ -7167,6 +9690,7 @@ ice_features_check(struct sk_buff *skb, struct net_device __always_unused *netdev, netdev_features_t features) { + bool gso = skb_is_gso(skb); size_t len; /* No point in doing any of this if neither checksum nor GSO are @@ -7179,24 +9703,32 @@ ice_features_check(struct sk_buff *skb, /* We cannot support GSO if the MSS is going to be less than * 64 bytes. If it is then we need to drop support for GSO. */ - if (skb_is_gso(skb) && (skb_shinfo(skb)->gso_size < 64)) + if (gso && (skb_shinfo(skb)->gso_size < ICE_TXD_CTX_MIN_MSS)) features &= ~NETIF_F_GSO_MASK; - len = skb_network_header(skb) - skb->data; + len = skb_network_offset(skb); if (len > ICE_TXD_MACLEN_MAX || len & 0x1) goto out_rm_features; - len = skb_transport_header(skb) - skb_network_header(skb); + len = skb_network_header_len(skb); if (len > ICE_TXD_IPLEN_MAX || len & 0x1) goto out_rm_features; if (skb->encapsulation) { - len = skb_inner_network_header(skb) - skb_transport_header(skb); - if (len > ICE_TXD_L4LEN_MAX || len & 0x1) - goto out_rm_features; + /* this must work for VXLAN frames AND IPIP/SIT frames, and in + * the case of IPIP frames, the transport header pointer is + * after the inner header! So check to make sure that this + * is a GRE or UDP_TUNNEL frame before doing that math. + */ + if (gso && (skb_shinfo(skb)->gso_type & + (SKB_GSO_GRE | SKB_GSO_UDP_TUNNEL))) { + len = skb_inner_network_header(skb) - + skb_transport_header(skb); + if (len > ICE_TXD_L4LEN_MAX || len & 0x1) + goto out_rm_features; + } - len = skb_inner_transport_header(skb) - - skb_inner_network_header(skb); + len = skb_inner_network_header_len(skb); if (len > ICE_TXD_IPLEN_MAX || len & 0x1) goto out_rm_features; } @@ -7222,14 +9754,15 @@ static const struct net_device_ops ice_netdev_ops = { .ndo_open = ice_open, .ndo_stop = ice_stop, .ndo_start_xmit = ice_start_xmit, + .ndo_select_queue = ice_select_queue, .ndo_features_check = ice_features_check, + .ndo_fix_features = ice_fix_features, .ndo_set_rx_mode = ice_set_rx_mode, .ndo_set_mac_address = ice_set_mac_address, .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = ice_change_mtu, .ndo_get_stats64 = ice_get_stats64, .ndo_set_tx_maxrate = ice_set_tx_maxrate, - .ndo_do_ioctl = ice_do_ioctl, .ndo_set_vf_spoofchk = ice_set_vf_spoofchk, .ndo_set_vf_mac = ice_set_vf_mac, .ndo_get_vf_config = ice_get_vf_cfg, @@ -7237,8 +9770,10 @@ static const struct net_device_ops ice_netdev_ops = { .ndo_set_vf_vlan = ice_set_vf_port_vlan, .ndo_set_vf_link_state = ice_set_vf_link_state, .ndo_get_vf_stats = ice_get_vf_stats, + .ndo_set_vf_rate = ice_set_vf_bw, .ndo_vlan_rx_add_vid = ice_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = ice_vlan_rx_kill_vid, + .ndo_setup_tc = ice_setup_tc, .ndo_set_features = ice_set_features, .ndo_bridge_getlink = ice_bridge_getlink, .ndo_bridge_setlink = ice_bridge_setlink, @@ -7251,4 +9786,6 @@ static const struct net_device_ops ice_netdev_ops = { .ndo_bpf = ice_xdp, .ndo_xdp_xmit = ice_xdp_xmit, .ndo_xsk_wakeup = ice_xsk_wakeup, + .ndo_hwtstamp_get = ice_ptp_hwtstamp_get, + .ndo_hwtstamp_set = ice_ptp_hwtstamp_set, }; |