drivers/net/ethernet/intel/ice/ice_main.c

// SPDX-License-Identifier: GPL-2.0
/* Copyright (c) 2018, Intel Corporation. */

/* Intel(R) Ethernet Connection E800 Series Linux Driver */

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <generated/utsrelease.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"
/* 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"

#define DRV_SUMMARY	"Intel(R) Ethernet Connection E800 Series Linux Driver"
static const char ice_driver_string[] = DRV_SUMMARY;
static const char ice_copyright[] = "Copyright (c) 2018, Intel Corporation.";

/* DDP Package file located in firmware search paths (e.g. /lib/firmware/) */
#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_LICENSE("GPL v2");
MODULE_FIRMWARE(ICE_DDP_PKG_FILE);

static int debug = -1;
module_param(debug, int, 0644);
#ifndef CONFIG_DYNAMIC_DEBUG
MODULE_PARM_DESC(debug, "netif level (0=none,...,16=all), hw debug_mask (0x8XXXXXXX)");
#else
MODULE_PARM_DESC(debug, "netif level (0=none,...,16=all)");
#endif /* !CONFIG_DYNAMIC_DEBUG */

static DEFINE_IDA(ice_aux_ida);

static struct workqueue_struct *ice_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)
{
	return dev && (dev->netdev_ops == &ice_netdev_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)
{
	u16 head, tail;

	head = ring->next_to_clean;
	tail = ring->next_to_use;

	if (head != tail)
		return (head < tail) ?
			tail - head : (tail + ring->count - head);
	return 0;
}

/**
 * ice_check_for_hang_subtask - check for and recover hung queues
 * @pf: pointer to PF struct
 */
static void ice_check_for_hang_subtask(struct ice_pf *pf)
{
	struct ice_vsi *vsi = NULL;
	struct ice_hw *hw;
	unsigned int i;
	int packets;
	u32 v;

	ice_for_each_vsi(pf, v)
		if (pf->vsi[v] && pf->vsi[v]->type == ICE_VSI_PF) {
			vsi = pf->vsi[v];
			break;
		}

	if (!vsi || test_bit(ICE_VSI_DOWN, vsi->state))
		return;

	if (!(vsi->netdev && netif_carrier_ok(vsi->netdev)))
		return;

	hw = &vsi->back->hw;

	for (i = 0; i < vsi->num_txq; i++) {
		struct ice_ring *tx_ring = vsi->tx_rings[i];

		if (tx_ring && tx_ring->desc) {
			/* If packet counter has not changed the queue is
			 * likely stalled, so force an interrupt for this
			 * queue.
			 *
			 * 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) {
				/* Trigger sw interrupt to revive the queue */
				ice_trigger_sw_intr(hw, tx_ring->q_vector);
				continue;
			}

			/* Memory barrier between read of packet count and call
			 * to ice_get_tx_pending()
			 */
			smp_rmb();
			tx_ring->tx_stats.prev_pkt =
			    ice_get_tx_pending(tx_ring) ? packets : -1;
		}
	}
}

/**
 * ice_init_mac_fltr - Set initial MAC filters
 * @pf: board private structure
 *
 * Set initial set of MAC filters for PF VSI; configure filters for permanent
 * address and broadcast address. If an error is encountered, netdevice will be
 * unregistered.
 */
static int ice_init_mac_fltr(struct ice_pf *pf)
{
	enum ice_status status;
	struct ice_vsi *vsi;
	u8 *perm_addr;

	vsi = ice_get_main_vsi(pf);
	if (!vsi)
		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;
}

/**
 * ice_add_mac_to_sync_list - creates list of MAC addresses to be synced
 * @netdev: the net device on which the sync is happening
 * @addr: MAC address to sync
 *
 * This is a callback function which is called by the in kernel device sync
 * functions (like __dev_uc_sync, __dev_mc_sync, etc). This function only
 * populates the tmp_sync_list, which is later used by ice_add_mac to add the
 * MAC filters from the hardware.
 */
static int ice_add_mac_to_sync_list(struct net_device *netdev, const u8 *addr)
{
	struct ice_netdev_priv *np = netdev_priv(netdev);
	struct ice_vsi *vsi = np->vsi;

	if (ice_fltr_add_mac_to_list(vsi, &vsi->tmp_sync_list, addr,
				     ICE_FWD_TO_VSI))
		return -EINVAL;

	return 0;
}

/**
 * ice_add_mac_to_unsync_list - creates list of MAC addresses to be unsynced
 * @netdev: the net device on which the unsync is happening
 * @addr: MAC address to unsync
 *
 * This is a callback function which is called by the in kernel device unsync
 * functions (like __dev_uc_unsync, __dev_mc_unsync, etc). This function only
 * populates the tmp_unsync_list, which is later used by ice_remove_mac to
 * delete the MAC filters from the hardware.
 */
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;

	return 0;
}

/**
 * ice_vsi_fltr_changed - check if filter state changed
 * @vsi: VSI to be checked
 *
 * returns true if filter state has changed, false otherwise.
 */
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);
}

/**
 * ice_cfg_promisc - Enable or disable 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)
{
	struct ice_hw *hw = &vsi->back->hw;
	enum ice_status status = 0;

	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);
	} 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);
	}

	if (status)
		return -EIO;

	return 0;
}

/**
 * ice_vsi_sync_fltr - Update the VSI filter list to the HW
 * @vsi: ptr to the VSI
 *
 * Push any outstanding VSI filter changes through the AdminQ.
 */
static int ice_vsi_sync_fltr(struct ice_vsi *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;

	if (!vsi->netdev)
		return -EINVAL;

	while (test_and_set_bit(ICE_CFG_BUSY, vsi->state))
		usleep_range(1000, 2000);

	changed_flags = vsi->current_netdev_flags ^ vsi->netdev->flags;
	vsi->current_netdev_flags = vsi->netdev->flags;

	INIT_LIST_HEAD(&vsi->tmp_sync_list);
	INIT_LIST_HEAD(&vsi->tmp_unsync_list);

	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);
		__dev_uc_sync(netdev, ice_add_mac_to_sync_list,
			      ice_add_mac_to_unsync_list);
		__dev_mc_sync(netdev, ice_add_mac_to_sync_list,
			      ice_add_mac_to_unsync_list);
		/* our temp lists are populated. release lock */
		netif_addr_unlock_bh(netdev);
	}

	/* Remove MAC addresses in the unsync list */
	status = ice_fltr_remove_mac_list(vsi, &vsi->tmp_unsync_list);
	ice_fltr_free_list(dev, &vsi->tmp_unsync_list);
	if (status) {
		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;
			goto out;
		}
	}

	/* Add MAC addresses in the sync list */
	status = 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) {
		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 &&
		    !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;
		}
	}
	/* 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);
			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);
			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;
			}
		}
	}

	if (((changed_flags & IFF_PROMISC) || promisc_forced_on) ||
	    test_bit(ICE_VSI_PROMISC_CHANGED, vsi->state)) {
		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 (err && err != -EEXIST) {
					netdev_err(netdev, "Error %d setting default VSI %i Rx rule\n",
						   err, vsi->vsi_num);
					vsi->current_netdev_flags &=
						~IFF_PROMISC;
					goto out_promisc;
				}
				ice_cfg_vlan_pruning(vsi, false, false);
			}
		} 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 (err) {
					netdev_err(netdev, "Error %d clearing default VSI %i Rx rule\n",
						   err, vsi->vsi_num);
					vsi->current_netdev_flags |=
						IFF_PROMISC;
					goto out_promisc;
				}
				if (vsi->num_vlan > 1)
					ice_cfg_vlan_pruning(vsi, true, false);
			}
		}
	}
	goto exit;

out_promisc:
	set_bit(ICE_VSI_PROMISC_CHANGED, vsi->state);
	goto exit;
out:
	/* if something went wrong then set the changed flag so we try again */
	set_bit(ICE_VSI_UMAC_FLTR_CHANGED, vsi->state);
	set_bit(ICE_VSI_MMAC_FLTR_CHANGED, vsi->state);
exit:
	clear_bit(ICE_CFG_BUSY, vsi->state);
	return err;
}

/**
 * ice_sync_fltr_subtask - Sync the VSI filter list with HW
 * @pf: board private structure
 */
static void ice_sync_fltr_subtask(struct ice_pf *pf)
{
	int v;

	if (!pf || !(test_bit(ICE_FLAG_FLTR_SYNC, pf->flags)))
		return;

	clear_bit(ICE_FLAG_FLTR_SYNC, pf->flags);

	ice_for_each_vsi(pf, v)
		if (pf->vsi[v] && ice_vsi_fltr_changed(pf->vsi[v]) &&
		    ice_vsi_sync_fltr(pf->vsi[v])) {
			/* come back and try again later */
			set_bit(ICE_FLAG_FLTR_SYNC, pf->flags);
			break;
		}
}

/**
 * ice_pf_dis_all_vsi - Pause all VSIs on a PF
 * @pf: the PF
 * @locked: is the rtnl_lock already held
 */
static void ice_pf_dis_all_vsi(struct ice_pf *pf, bool locked)
{
	int node;
	int v;

	ice_for_each_vsi(pf, v)
		if (pf->vsi[v])
			ice_dis_vsi(pf->vsi[v], locked);

	for (node = 0; node < ICE_MAX_PF_AGG_NODES; node++)
		pf->pf_agg_node[node].num_vsis = 0;

	for (node = 0; node < ICE_MAX_VF_AGG_NODES; node++)
		pf->vf_agg_node[node].num_vsis = 0;
}

/**
 * ice_prepare_for_reset - prep for the core to reset
 * @pf: board private structure
 *
 * Inform or close all dependent features in prep for reset.
 */
static void
ice_prepare_for_reset(struct ice_pf *pf)
{
	struct ice_hw *hw = &pf->hw;
	unsigned int i;

	/* already prepared for reset */
	if (test_bit(ICE_PREPARED_FOR_RESET, pf->state))
		return;

	ice_unplug_aux_dev(pf);

	/* Notify VFs of impending reset */
	if (ice_check_sq_alive(hw, &hw->mailboxq))
		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]);

	/* clear SW filtering DB */
	ice_clear_hw_tbls(hw);
	/* disable the VSIs and their queues that are not already DOWN */
	ice_pf_dis_all_vsi(pf, false);

	if (test_bit(ICE_FLAG_PTP_SUPPORTED, pf->flags))
		ice_ptp_release(pf);

	if (hw->port_info)
		ice_sched_clear_port(hw->port_info);

	ice_shutdown_all_ctrlq(hw);

	set_bit(ICE_PREPARED_FOR_RESET, pf->state);
}

/**
 * ice_do_reset - Initiate one of many types of resets
 * @pf: board private structure
 * @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)
{
	struct device *dev = ice_pf_to_dev(pf);
	struct ice_hw *hw = &pf->hw;

	dev_dbg(dev, "reset_type 0x%x requested\n", reset_type);

	ice_prepare_for_reset(pf);

	/* trigger the reset */
	if (ice_reset(hw, reset_type)) {
		dev_err(dev, "reset %d failed\n", reset_type);
		set_bit(ICE_RESET_FAILED, pf->state);
		clear_bit(ICE_RESET_OICR_RECV, pf->state);
		clear_bit(ICE_PREPARED_FOR_RESET, pf->state);
		clear_bit(ICE_PFR_REQ, pf->state);
		clear_bit(ICE_CORER_REQ, pf->state);
		clear_bit(ICE_GLOBR_REQ, pf->state);
		wake_up(&pf->reset_wait_queue);
		return;
	}

	/* PFR is a bit of a special case because it doesn't result in an OICR
	 * interrupt. So for PFR, rebuild after the reset and clear the reset-
	 * associated state bits.
	 */
	if (reset_type == ICE_RESET_PFR) {
		pf->pfr_count++;
		ice_rebuild(pf, 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_subtask - Set up for resetting the device and driver
 * @pf: board private structure
 */
static void ice_reset_subtask(struct ice_pf *pf)
{
	enum ice_reset_req reset_type = ICE_RESET_INVAL;

	/* When a CORER/GLOBR/EMPR is about to happen, the hardware triggers an
	 * OICR interrupt. The OICR handler (ice_misc_intr) determines what type
	 * of reset is pending and sets bits in pf->state indicating the reset
	 * type and ICE_RESET_OICR_RECV. So, if the latter bit is set
	 * prepare for pending reset if not already (for PF software-initiated
	 * global resets the software should already be prepared for it as
	 * indicated by ICE_PREPARED_FOR_RESET; for global resets initiated
	 * by firmware or software on other PFs, that bit is not set so prepare
	 * for the reset now), poll for reset done, rebuild and return.
	 */
	if (test_bit(ICE_RESET_OICR_RECV, pf->state)) {
		/* Perform the largest reset requested */
		if (test_and_clear_bit(ICE_CORER_RECV, pf->state))
			reset_type = ICE_RESET_CORER;
		if (test_and_clear_bit(ICE_GLOBR_RECV, pf->state))
			reset_type = ICE_RESET_GLOBR;
		if (test_and_clear_bit(ICE_EMPR_RECV, pf->state))
			reset_type = ICE_RESET_EMPR;
		/* return if no valid reset type requested */
		if (reset_type == ICE_RESET_INVAL)
			return;
		ice_prepare_for_reset(pf);

		/* make sure we are ready to rebuild */
		if (ice_check_reset(&pf->hw)) {
			set_bit(ICE_RESET_FAILED, pf->state);
		} else {
			/* done with reset. start rebuild */
			pf->hw.reset_ongoing = false;
			ice_rebuild(pf, reset_type);
			/* clear bit to resume normal operations, but
			 * ICE_NEEDS_RESTART bit is set in case rebuild failed
			 */
			clear_bit(ICE_RESET_OICR_RECV, pf->state);
			clear_bit(ICE_PREPARED_FOR_RESET, pf->state);
			clear_bit(ICE_PFR_REQ, pf->state);
			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);
		}

		return;
	}

	/* No pending resets to finish processing. Check for new resets */
	if (test_bit(ICE_PFR_REQ, pf->state))
		reset_type = ICE_RESET_PFR;
	if (test_bit(ICE_CORER_REQ, pf->state))
		reset_type = ICE_RESET_CORER;
	if (test_bit(ICE_GLOBR_REQ, pf->state))
		reset_type = ICE_RESET_GLOBR;
	/* If no valid reset type requested just return */
	if (reset_type == ICE_RESET_INVAL)
		return;

	/* reset if not already down or busy */
	if (!test_bit(ICE_DOWN, pf->state) &&
	    !test_bit(ICE_CFG_BUSY, pf->state)) {
		ice_do_reset(pf, reset_type);
	}
}

/**
 * ice_print_topo_conflict - print topology conflict message
 * @vsi: the VSI whose topology status is being checked
 */
static void ice_print_topo_conflict(struct ice_vsi *vsi)
{
	switch (vsi->port_info->phy.link_info.topo_media_conflict) {
	case ICE_AQ_LINK_TOPO_CONFLICT:
	case ICE_AQ_LINK_MEDIA_CONFLICT:
	case ICE_AQ_LINK_TOPO_UNREACH_PRT:
	case ICE_AQ_LINK_TOPO_UNDRUTIL_PRT:
	case ICE_AQ_LINK_TOPO_UNDRUTIL_MEDIA:
		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");
		break;
	default:
		break;
	}
}

/**
 * ice_print_link_msg - print link up or down message
 * @vsi: the VSI whose link status is being queried
 * @isup: boolean for if the link is now up or down
 */
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;

	if (!vsi)
		return;

	if (vsi->current_isup == isup)
		return;

	vsi->current_isup = isup;

	if (!isup) {
		netdev_info(vsi->netdev, "NIC Link is Down\n");
		return;
	}

	switch (vsi->port_info->phy.link_info.link_speed) {
	case ICE_AQ_LINK_SPEED_100GB:
		speed = "100 G";
		break;
	case ICE_AQ_LINK_SPEED_50GB:
		speed = "50 G";
		break;
	case ICE_AQ_LINK_SPEED_40GB:
		speed = "40 G";
		break;
	case ICE_AQ_LINK_SPEED_25GB:
		speed = "25 G";
		break;
	case ICE_AQ_LINK_SPEED_20GB:
		speed = "20 G";
		break;
	case ICE_AQ_LINK_SPEED_10GB:
		speed = "10 G";
		break;
	case ICE_AQ_LINK_SPEED_5GB:
		speed = "5 G";
		break;
	case ICE_AQ_LINK_SPEED_2500MB:
		speed = "2.5 G";
		break;
	case ICE_AQ_LINK_SPEED_1000MB:
		speed = "1 G";
		break;
	case ICE_AQ_LINK_SPEED_100MB:
		speed = "100 M";
		break;
	default:
		speed = "Unknown ";
		break;
	}

	switch (vsi->port_info->fc.current_mode) {
	case ICE_FC_FULL:
		fc = "Rx/Tx";
		break;
	case ICE_FC_TX_PAUSE:
		fc = "Tx";
		break;
	case ICE_FC_RX_PAUSE:
		fc = "Rx";
		break;
	case ICE_FC_NONE:
		fc = "None";
		break;
	default:
		fc = "Unknown";
		break;
	}

	/* Get FEC mode based on negotiated link info */
	switch (vsi->port_info->phy.link_info.fec_info) {
	case ICE_AQ_LINK_25G_RS_528_FEC_EN:
	case ICE_AQ_LINK_25G_RS_544_FEC_EN:
		fec = "RS-FEC";
		break;
	case ICE_AQ_LINK_25G_KR_FEC_EN:
		fec = "FC-FEC/BASE-R";
		break;
	default:
		fec = "NONE";
		break;
	}

	/* check if autoneg completed, might be false due to not supported */
	if (vsi->port_info->phy.link_info.an_info & ICE_AQ_AN_COMPLETED)
		an = "True";
	else
		an = "False";

	/* Get FEC mode requested based on PHY caps last SW configuration */
	caps = kzalloc(sizeof(*caps), GFP_KERNEL);
	if (!caps) {
		fec_req = "Unknown";
		an_advertised = "Unknown";
		goto done;
	}

	status = ice_aq_get_phy_caps(vsi->port_info, false,
				     ICE_AQC_REPORT_ACTIVE_CFG, caps, NULL);
	if (status)
		netdev_info(vsi->netdev, "Get phy capability failed.\n");

	an_advertised = ice_is_phy_caps_an_enabled(caps) ? "On" : "Off";

	if (caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_528_REQ ||
	    caps->link_fec_options & ICE_AQC_PHY_FEC_25G_RS_544_REQ)
		fec_req = "RS-FEC";
	else if (caps->link_fec_options & ICE_AQC_PHY_FEC_10G_KR_40G_KR4_REQ ||
		 caps->link_fec_options & ICE_AQC_PHY_FEC_25G_KR_REQ)
		fec_req = "FC-FEC/BASE-R";
	else
		fec_req = "NONE";

	kfree(caps);

done:
	netdev_info(vsi->netdev, "NIC Link is up %sbps Full Duplex, Requested FEC: %s, Negotiated FEC: %s, Autoneg Advertised: %s, Autoneg Negotiated: %s, Flow Control: %s\n",
		    speed, fec_req, fec, an_advertised, an, fc);
	ice_print_topo_conflict(vsi);
}

/**
 * ice_vsi_link_event - update the VSI's netdev
 * @vsi: the VSI on which the link event occurred
 * @link_up: whether or not the VSI needs to be set up or down
 */
static void ice_vsi_link_event(struct ice_vsi *vsi, bool link_up)
{
	if (!vsi)
		return;

	if (test_bit(ICE_VSI_DOWN, vsi->state) || !vsi->netdev)
		return;

	if (vsi->type == ICE_VSI_PF) {
		if (link_up == netif_carrier_ok(vsi->netdev))
			return;

		if (link_up) {
			netif_carrier_on(vsi->netdev);
			netif_tx_wake_all_queues(vsi->netdev);
		} else {
			netif_carrier_off(vsi->netdev);
			netif_tx_stop_all_queues(vsi->netdev);
		}
	}
}

/**
 * ice_set_dflt_mib - send a default config MIB to the FW
 * @pf: private PF struct
 *
 * This function sends a default configuration MIB to the FW.
 *
 * If this function errors out at any point, the driver is still able to
 * function.  The main impact is that LFC may not operate as expected.
 * Therefore an error state in this function should be treated with a DBG
 * message and continue on with driver rebuild/reenable.
 */
static void ice_set_dflt_mib(struct ice_pf *pf)
{
	struct device *dev = ice_pf_to_dev(pf);
	u8 mib_type, *buf, *lldpmib = NULL;
	u16 len, typelen, offset = 0;
	struct ice_lldp_org_tlv *tlv;
	struct ice_hw *hw = &pf->hw;
	u32 ouisubtype;

	mib_type = SET_LOCAL_MIB_TYPE_LOCAL_MIB;
	lldpmib = kzalloc(ICE_LLDPDU_SIZE, GFP_KERNEL);
	if (!lldpmib) {
		dev_dbg(dev, "%s Failed to allocate MIB memory\n",
			__func__);
		return;
	}

	/* Add ETS CFG TLV */
	tlv = (struct ice_lldp_org_tlv *)lldpmib;
	typelen = ((ICE_TLV_TYPE_ORG << ICE_LLDP_TLV_TYPE_S) |
		   ICE_IEEE_ETS_TLV_LEN);
	tlv->typelen = htons(typelen);
	ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) |
		      ICE_IEEE_SUBTYPE_ETS_CFG);
	tlv->ouisubtype = htonl(ouisubtype);

	buf = tlv->tlvinfo;
	buf[0] = 0;

	/* ETS CFG all UPs map to TC 0. Next 4 (1 - 4) Octets = 0.
	 * Octets 5 - 12 are BW values, set octet 5 to 100% BW.
	 * Octets 13 - 20 are TSA values - leave as zeros
	 */
	buf[5] = 0x64;
	len = (typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S;
	offset += len + 2;
	tlv = (struct ice_lldp_org_tlv *)
		((char *)tlv + sizeof(tlv->typelen) + len);

	/* Add ETS REC TLV */
	buf = tlv->tlvinfo;
	tlv->typelen = htons(typelen);

	ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) |
		      ICE_IEEE_SUBTYPE_ETS_REC);
	tlv->ouisubtype = htonl(ouisubtype);

	/* First octet of buf is reserved
	 * Octets 1 - 4 map UP to TC - all UPs map to zero
	 * Octets 5 - 12 are BW values - set TC 0 to 100%.
	 * Octets 13 - 20 are TSA value - leave as zeros
	 */
	buf[5] = 0x64;
	offset += len + 2;
	tlv = (struct ice_lldp_org_tlv *)
		((char *)tlv + sizeof(tlv->typelen) + len);

	/* Add PFC CFG TLV */
	typelen = ((ICE_TLV_TYPE_ORG << ICE_LLDP_TLV_TYPE_S) |
		   ICE_IEEE_PFC_TLV_LEN);
	tlv->typelen = htons(typelen);

	ouisubtype = ((ICE_IEEE_8021QAZ_OUI << ICE_LLDP_TLV_OUI_S) |
		      ICE_IEEE_SUBTYPE_PFC_CFG);
	tlv->ouisubtype = htonl(ouisubtype);

	/* Octet 1 left as all zeros - PFC disabled */
	buf[0] = 0x08;
	len = (typelen & ICE_LLDP_TLV_LEN_M) >> ICE_LLDP_TLV_LEN_S;
	offset += len + 2;

	if (ice_aq_set_lldp_mib(hw, mib_type, (void *)lldpmib, offset, NULL))
		dev_dbg(dev, "%s Failed to set default LLDP MIB\n", __func__);

	kfree(lldpmib);
}

/**
 * ice_check_module_power
 * @pf: pointer to PF struct
 * @link_cfg_err: bitmap from the link info structure
 *
 * check module power level returned by a previous call to aq_get_link_info
 * and print error messages if module power level is not supported
 */
static void ice_check_module_power(struct ice_pf *pf, u8 link_cfg_err)
{
	/* if module power level is supported, clear the flag */
	if (!(link_cfg_err & (ICE_AQ_LINK_INVAL_MAX_POWER_LIMIT |
			      ICE_AQ_LINK_MODULE_POWER_UNSUPPORTED))) {
		clear_bit(ICE_FLAG_MOD_POWER_UNSUPPORTED, pf->flags);
		return;
	}

	/* if ICE_FLAG_MOD_POWER_UNSUPPORTED was previously set and the
	 * above block didn't clear this bit, there's nothing to do
	 */
	if (test_bit(ICE_FLAG_MOD_POWER_UNSUPPORTED, pf->flags))
		return;

	if (link_cfg_err & ICE_AQ_LINK_INVAL_MAX_POWER_LIMIT) {
		dev_err(ice_pf_to_dev(pf), "The installed module is incompatible with the device's NVM image. Cannot start link\n");
		set_bit(ICE_FLAG_MOD_POWER_UNSUPPORTED, pf->flags);
	} else if (link_cfg_err & ICE_AQ_LINK_MODULE_POWER_UNSUPPORTED) {
		dev_err(ice_pf_to_dev(pf), "The module's power requirements exceed the device's power supply. Cannot start link\n");
		set_bit(ICE_FLAG_MOD_POWER_UNSUPPORTED, pf->flags);
	}
}

/**
 * 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
 * @link_up: true if the physical link is up and false if it is down
 * @link_speed: current link speed received from the link event
 *
 * Returns 0 on success and negative on failure
 */
static int
ice_link_event(struct ice_pf *pf, struct ice_port_info *pi, bool link_up,
	       u16 link_speed)
{
	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;

	phy_info = &pi->phy;
	phy_info->link_info_old = phy_info->link_info;

	old_link = !!(phy_info->link_info_old.link_info & ICE_AQ_LINK_UP);
	old_link_speed = phy_info->link_info_old.link_speed;

	/* update the link info structures and re-enable link events,
	 * don't bail on failure due to other book keeping needed
	 */
	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));

	ice_check_module_power(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.
	 */
	if (phy_info->link_info.link_info & ICE_AQ_LINK_UP)
		link_up = true;

	vsi = ice_get_main_vsi(pf);
	if (!vsi || !vsi->port_info)
		return -EINVAL;

	/* turn off PHY if media was removed */
	if (!test_bit(ICE_FLAG_NO_MEDIA, pf->flags) &&
	    !(pi->phy.link_info.link_info & ICE_AQ_MEDIA_AVAILABLE)) {
		set_bit(ICE_FLAG_NO_MEDIA, pf->flags);
		ice_set_link(vsi, false);
	}

	/* if the old link up/down and speed is the same as the new */
	if (link_up == old_link && link_speed == old_link_speed)
		return 0;

	if (ice_is_dcb_active(pf)) {
		if (test_bit(ICE_FLAG_DCB_ENA, pf->flags))
			ice_dcb_rebuild(pf);
	} else {
		if (link_up)
			ice_set_dflt_mib(pf);
	}
	ice_vsi_link_event(vsi, link_up);
	ice_print_link_msg(vsi, link_up);

	ice_vc_notify_link_state(pf);

	return 0;
}

/**
 * ice_watchdog_subtask - periodic tasks not using event driven scheduling
 * @pf: board private structure
 */
static void ice_watchdog_subtask(struct ice_pf *pf)
{
	int i;

	/* if interface is down do nothing */
	if (test_bit(ICE_DOWN, pf->state) ||
	    test_bit(ICE_CFG_BUSY, pf->state))
		return;

	/* make sure we don't do these things too often */
	if (time_before(jiffies,
			pf->serv_tmr_prev + pf->serv_tmr_period))
		return;

	pf->serv_tmr_prev = jiffies;

	/* Update the stats for active netdevs so the network stack
	 * can look at updated numbers whenever it cares to
	 */
	ice_update_pf_stats(pf);
	ice_for_each_vsi(pf, i)
		if (pf->vsi[i] && pf->vsi[i]->netdev)
			ice_update_vsi_stats(pf->vsi[i]);
}

/**
 * ice_init_link_events - enable/initialize link events
 * @pi: pointer to the port_info instance
 *
 * Returns -EIO on failure, 0 on success
 */
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));

	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",
			pi->lport);
		return -EIO;
	}

	if (ice_aq_get_link_info(pi, true, NULL, NULL)) {
		dev_dbg(ice_hw_to_dev(pi->hw), "Failed to enable link events for port %d\n",
			pi->lport);
		return -EIO;
	}

	return 0;
}

/**
 * ice_handle_link_event - handle link event via ARQ
 * @pf: PF that the link event is associated with
 * @event: event structure containing link status info
 */
static int
ice_handle_link_event(struct ice_pf *pf, struct ice_rq_event_info *event)
{
	struct ice_aqc_get_link_status_data *link_data;
	struct ice_port_info *port_info;
	int status;

	link_data = (struct ice_aqc_get_link_status_data *)event->msg_buf;
	port_info = pf->hw.port_info;
	if (!port_info)
		return -EINVAL;

	status = ice_link_event(pf, port_info,
				!!(link_data->link_info & ICE_AQ_LINK_UP),
				le16_to_cpu(link_data->link_speed));
	if (status)
		dev_dbg(ice_pf_to_dev(pf), "Could not process link event, error %d\n",
			status);

	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;

	u16 opcode;
	struct ice_rq_event_info *event;
	enum ice_aq_task_state state;
};

/**
 * 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
 * @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)
{
	struct device *dev = ice_pf_to_dev(pf);
	struct ice_aq_task *task;
	unsigned long start;
	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,
					       timeout);
	switch (task->state) {
	case ICE_AQ_TASK_WAITING:
		err = ret < 0 ? ret : -ETIMEDOUT;
		break;
	case ICE_AQ_TASK_CANCELED:
		err = ret < 0 ? ret : -ECANCELED;
		break;
	case ICE_AQ_TASK_COMPLETE:
		err = ret < 0 ? ret : 0;
		break;
	default:
		WARN(1, "Unexpected AdminQ wait task state %u", task->state);
		err = -EINVAL;
		break;
	}

	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);

	spin_lock_bh(&pf->aq_wait_lock);
	hlist_del(&task->entry);
	spin_unlock_bh(&pf->aq_wait_lock);
	kfree(task);

	return err;
}

/**
 * ice_aq_check_events - Check if any thread is waiting for an AdminQ event
 * @pf: pointer to the PF private structure
 * @opcode: the opcode of the event
 * @event: the event to check
 *
 * Loops over the current list of pending threads waiting for an AdminQ event.
 * For each matching task, copy the contents of the event into the task
 * structure and wake up the thread.
 *
 * If multiple threads wait for the same opcode, they will all be woken up.
 *
 * Note that event->msg_buf will only be duplicated if the event has a buffer
 * with enough space already allocated. Otherwise, only the descriptor and
 * message length will be copied.
 *
 * Returns: true if an event was found, false otherwise
 */
static void ice_aq_check_events(struct ice_pf *pf, u16 opcode,
				struct ice_rq_event_info *event)
{
	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)
			continue;

		memcpy(&task->event->desc, &event->desc, sizeof(event->desc));
		task->event->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,
			       event->buf_len);
			task->event->buf_len = event->buf_len;
		}

		task->state = ICE_AQ_TASK_COMPLETE;
		found = true;
	}
	spin_unlock_bh(&pf->aq_wait_lock);

	if (found)
		wake_up(&pf->aq_wait_queue);
}

/**
 * ice_aq_cancel_waiting_tasks - Immediately cancel all waiting tasks
 * @pf: the PF private structure
 *
 * Set all waiting tasks to ICE_AQ_TASK_CANCELED, and wake up their threads.
 * This will then cause ice_aq_wait_for_event to exit with -ECANCELED.
 */
static void ice_aq_cancel_waiting_tasks(struct ice_pf *pf)
{
	struct ice_aq_task *task;

	spin_lock_bh(&pf->aq_wait_lock);
	hlist_for_each_entry(task, &pf->aq_wait_list, entry)
		task->state = ICE_AQ_TASK_CANCELED;
	spin_unlock_bh(&pf->aq_wait_lock);

	wake_up(&pf->aq_wait_queue);
}

/**
 * __ice_clean_ctrlq - helper function to clean controlq rings
 * @pf: ptr to struct ice_pf
 * @q_type: specific Control queue type
 */
static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type)
{
	struct device *dev = ice_pf_to_dev(pf);
	struct ice_rq_event_info event;
	struct ice_hw *hw = &pf->hw;
	struct ice_ctl_q_info *cq;
	u16 pending, i = 0;
	const char *qtype;
	u32 oldval, val;

	/* Do not clean control queue if/when PF reset fails */
	if (test_bit(ICE_RESET_FAILED, pf->state))
		return 0;

	switch (q_type) {
	case ICE_CTL_Q_ADMIN:
		cq = &hw->adminq;
		qtype = "Admin";
		break;
	case ICE_CTL_Q_SB:
		cq = &hw->sbq;
		qtype = "Sideband";
		break;
	case ICE_CTL_Q_MAILBOX:
		cq = &hw->mailboxq;
		qtype = "Mailbox";
		/* we are going to try to detect a malicious VF, so set the
		 * state to begin detection
		 */
		hw->mbx_snapshot.mbx_buf.state = ICE_MAL_VF_DETECT_STATE_NEW_SNAPSHOT;
		break;
	default:
		dev_warn(dev, "Unknown control queue type 0x%x\n", q_type);
		return 0;
	}

	/* check for error indications - PF_xx_AxQLEN register layout for
	 * FW/MBX/SB are identical so just use defines for PF_FW_AxQLEN.
	 */
	val = rd32(hw, cq->rq.len);
	if (val & (PF_FW_ARQLEN_ARQVFE_M | PF_FW_ARQLEN_ARQOVFL_M |
		   PF_FW_ARQLEN_ARQCRIT_M)) {
		oldval = val;
		if (val & PF_FW_ARQLEN_ARQVFE_M)
			dev_dbg(dev, "%s Receive Queue VF Error detected\n",
				qtype);
		if (val & PF_FW_ARQLEN_ARQOVFL_M) {
			dev_dbg(dev, "%s Receive Queue Overflow Error detected\n",
				qtype);
		}
		if (val & PF_FW_ARQLEN_ARQCRIT_M)
			dev_dbg(dev, "%s Receive Queue Critical Error detected\n",
				qtype);
		val &= ~(PF_FW_ARQLEN_ARQVFE_M | PF_FW_ARQLEN_ARQOVFL_M |
			 PF_FW_ARQLEN_ARQCRIT_M);
		if (oldval != val)
			wr32(hw, cq->rq.len, val);
	}

	val = rd32(hw, cq->sq.len);
	if (val & (PF_FW_ATQLEN_ATQVFE_M | PF_FW_ATQLEN_ATQOVFL_M |
		   PF_FW_ATQLEN_ATQCRIT_M)) {
		oldval = val;
		if (val & PF_FW_ATQLEN_ATQVFE_M)
			dev_dbg(dev, "%s Send Queue VF Error detected\n",
				qtype);
		if (val & PF_FW_ATQLEN_ATQOVFL_M) {
			dev_dbg(dev, "%s Send Queue Overflow Error detected\n",
				qtype);
		}
		if (val & PF_FW_ATQLEN_ATQCRIT_M)
			dev_dbg(dev, "%s Send Queue Critical Error detected\n",
				qtype);
		val &= ~(PF_FW_ATQLEN_ATQVFE_M | PF_FW_ATQLEN_ATQOVFL_M |
			 PF_FW_ATQLEN_ATQCRIT_M);
		if (oldval != val)
			wr32(hw, cq->sq.len, val);
	}

	event.buf_len = cq->rq_buf_size;
	event.msg_buf = kzalloc(event.buf_len, GFP_KERNEL);
	if (!event.msg_buf)
		return 0;

	do {
		enum ice_status ret;
		u16 opcode;

		ret = ice_clean_rq_elem(hw, cq, &event, &pending);
		if (ret == ICE_ERR_AQ_NO_WORK)
			break;
		if (ret) {
			dev_err(dev, "%s Receive Queue event error %s\n", qtype,
				ice_stat_str(ret));
			break;
		}

		opcode = le16_to_cpu(event.desc.opcode);

		/* Notify any thread that might be waiting for this event */
		ice_aq_check_events(pf, opcode, &event);

		switch (opcode) {
		case ice_aqc_opc_get_link_status:
			if (ice_handle_link_event(pf, &event))
				dev_err(dev, "Could not handle link event\n");
			break;
		case ice_aqc_opc_event_lan_overflow:
			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);
			break;
		case ice_aqc_opc_fw_logging:
			ice_output_fw_log(hw, &event.desc, event.msg_buf);
			break;
		case ice_aqc_opc_lldp_set_mib_change:
			ice_dcb_process_lldp_set_mib_change(pf, &event);
			break;
		default:
			dev_dbg(dev, "%s Receive Queue unknown event 0x%04x ignored\n",
				qtype, opcode);
			break;
		}
	} while (pending && (i++ < ICE_DFLT_IRQ_WORK));

	kfree(event.msg_buf);

	return pending && (i == ICE_DFLT_IRQ_WORK);
}

/**
 * ice_ctrlq_pending - check if there is a difference between ntc and ntu
 * @hw: pointer to hardware info
 * @cq: control queue information
 *
 * returns true if there are pending messages in a queue, false if there aren't
 */
static bool ice_ctrlq_pending(struct ice_hw *hw, struct ice_ctl_q_info *cq)
{
	u16 ntu;

	ntu = (u16)(rd32(hw, cq->rq.head) & cq->rq.head_mask);
	return cq->rq.next_to_clean != ntu;
}

/**
 * ice_clean_adminq_subtask - clean the AdminQ rings
 * @pf: board private structure
 */
static void ice_clean_adminq_subtask(struct ice_pf *pf)
{
	struct ice_hw *hw = &pf->hw;

	if (!test_bit(ICE_ADMINQ_EVENT_PENDING, pf->state))
		return;

	if (__ice_clean_ctrlq(pf, ICE_CTL_Q_ADMIN))
		return;

	clear_bit(ICE_ADMINQ_EVENT_PENDING, pf->state);

	/* There might be a situation where new messages arrive to a control
	 * queue between processing the last message and clearing the
	 * EVENT_PENDING bit. So before exiting, check queue head again (using
	 * ice_ctrlq_pending) and process new messages if any.
	 */
	if (ice_ctrlq_pending(hw, &hw->adminq))
		__ice_clean_ctrlq(pf, ICE_CTL_Q_ADMIN);

	ice_flush(hw);
}

/**
 * ice_clean_mailboxq_subtask - clean the MailboxQ rings
 * @pf: board private structure
 */
static void ice_clean_mailboxq_subtask(struct ice_pf *pf)
{
	struct ice_hw *hw = &pf->hw;

	if (!test_bit(ICE_MAILBOXQ_EVENT_PENDING, pf->state))
		return;

	if (__ice_clean_ctrlq(pf, ICE_CTL_Q_MAILBOX))
		return;

	clear_bit(ICE_MAILBOXQ_EVENT_PENDING, pf->state);

	if (ice_ctrlq_pending(hw, &hw->mailboxq))
		__ice_clean_ctrlq(pf, ICE_CTL_Q_MAILBOX);

	ice_flush(hw);
}

/**
 * ice_clean_sbq_subtask - clean the Sideband Queue rings
 * @pf: board private structure
 */
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)) {
		clear_bit(ICE_SIDEBANDQ_EVENT_PENDING, pf->state);
		return;
	}

	if (!test_bit(ICE_SIDEBANDQ_EVENT_PENDING, pf->state))
		return;

	if (__ice_clean_ctrlq(pf, ICE_CTL_Q_SB))
		return;

	clear_bit(ICE_SIDEBANDQ_EVENT_PENDING, pf->state);

	if (ice_ctrlq_pending(hw, &hw->sbq))
		__ice_clean_ctrlq(pf, ICE_CTL_Q_SB);

	ice_flush(hw);
}

/**
 * ice_service_task_schedule - schedule the service task to wake up
 * @pf: board private structure
 *
 * If not already scheduled, this puts the task into the work queue.
 */
void ice_service_task_schedule(struct ice_pf *pf)
{
	if (!test_bit(ICE_SERVICE_DIS, pf->state) &&
	    !test_and_set_bit(ICE_SERVICE_SCHED, pf->state) &&
	    !test_bit(ICE_NEEDS_RESTART, pf->state))
		queue_work(ice_wq, &pf->serv_task);
}

/**
 * ice_service_task_complete - finish up the service task
 * @pf: board private structure
 */
static void ice_service_task_complete(struct ice_pf *pf)
{
	WARN_ON(!test_bit(ICE_SERVICE_SCHED, pf->state));

	/* force memory (pf->state) to sync before next service task */
	smp_mb__before_atomic();
	clear_bit(ICE_SERVICE_SCHED, pf->state);
}

/**
 * ice_service_task_stop - stop service task and cancel works
 * @pf: board private structure
 *
 * Return 0 if the ICE_SERVICE_DIS bit was not already set,
 * 1 otherwise.
 */
static int ice_service_task_stop(struct ice_pf *pf)
{
	int ret;

	ret = test_and_set_bit(ICE_SERVICE_DIS, pf->state);

	if (pf->serv_tmr.function)
		del_timer_sync(&pf->serv_tmr);
	if (pf->serv_task.func)
		cancel_work_sync(&pf->serv_task);

	clear_bit(ICE_SERVICE_SCHED, pf->state);
	return ret;
}

/**
 * ice_service_task_restart - restart service task and schedule works
 * @pf: board private structure
 *
 * This function is needed for suspend and resume works (e.g WoL scenario)
 */
static void ice_service_task_restart(struct ice_pf *pf)
{
	clear_bit(ICE_SERVICE_DIS, pf->state);
	ice_service_task_schedule(pf);
}

/**
 * ice_service_timer - timer callback to schedule service task
 * @t: pointer to timer_list
 */
static void ice_service_timer(struct timer_list *t)
{
	struct ice_pf *pf = from_timer(pf, t, serv_tmr);

	mod_timer(&pf->serv_tmr, round_jiffies(pf->serv_tmr_period + jiffies));
	ice_service_task_schedule(pf);
}

/**
 * ice_handle_mdd_event - handle malicious driver detect event
 * @pf: pointer to the PF structure
 *
 * Called from service task. OICR interrupt handler indicates MDD event.
 * VF MDD logging is guarded by net_ratelimit. Additional PF and VF log
 * messages are wrapped by netif_msg_[rx|tx]_err. Since VF Rx MDD events
 * disable the queue, the PF can be configured to reset the VF using ethtool
 * private flag mdd-auto-reset-vf.
 */
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;
	u32 reg;

	if (!test_and_clear_bit(ICE_MDD_EVENT_PENDING, pf->state)) {
		/* Since the VF MDD event logging is rate limited, check if
		 * there are pending MDD events.
		 */
		ice_print_vfs_mdd_events(pf);
		return;
	}

	/* 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);

		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_PQM, 0xffffffff);
	}

	reg = rd32(hw, GL_MDET_TX_TCLAN);
	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);

		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);
	}

	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);

		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);
		wr32(hw, GL_MDET_RX, 0xffffffff);
	}

	/* check to see if this PF caused an MDD event */
	reg = rd32(hw, PF_MDET_TX_PQM);
	if (reg & PF_MDET_TX_PQM_VALID_M) {
		wr32(hw, PF_MDET_TX_PQM, 0xFFFF);
		if (netif_msg_tx_err(pf))
			dev_info(dev, "Malicious Driver Detection event TX_PQM detected on PF\n");
	}

	reg = rd32(hw, PF_MDET_TX_TCLAN);
	if (reg & PF_MDET_TX_TCLAN_VALID_M) {
		wr32(hw, PF_MDET_TX_TCLAN, 0xFFFF);
		if (netif_msg_tx_err(pf))
			dev_info(dev, "Malicious Driver Detection event TX_TCLAN detected on PF\n");
	}

	reg = rd32(hw, PF_MDET_RX);
	if (reg & PF_MDET_RX_VALID_M) {
		wr32(hw, PF_MDET_RX, 0xFFFF);
		if (netif_msg_rx_err(pf))
			dev_info(dev, "Malicious Driver Detection event RX detected on PF\n");
	}

	/* 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];

		reg = rd32(hw, VP_MDET_TX_PQM(i));
		if (reg & VP_MDET_TX_PQM_VALID_M) {
			wr32(hw, VP_MDET_TX_PQM(i), 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);
		}

		reg = rd32(hw, VP_MDET_TX_TCLAN(i));
		if (reg & VP_MDET_TX_TCLAN_VALID_M) {
			wr32(hw, VP_MDET_TX_TCLAN(i), 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);
		}

		reg = rd32(hw, VP_MDET_TX_TDPU(i));
		if (reg & VP_MDET_TX_TDPU_VALID_M) {
			wr32(hw, VP_MDET_TX_TDPU(i), 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);
		}

		reg = rd32(hw, VP_MDET_RX(i));
		if (reg & VP_MDET_RX_VALID_M) {
			wr32(hw, VP_MDET_RX(i), 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);

			/* 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);
			}
		}
	}

	ice_print_vfs_mdd_events(pf);
}

/**
 * ice_force_phys_link_state - Force the physical link state
 * @vsi: VSI to force the physical link state to up/down
 * @link_up: true/false indicates to set the physical link to up/down
 *
 * Force the physical link state by getting the current PHY capabilities from
 * hardware and setting the PHY config based on the determined capabilities. If
 * link changes a link event will be triggered because both the Enable Automatic
 * Link Update and LESM Enable bits are set when setting the PHY capabilities.
 *
 * Returns 0 on success, negative on failure
 */
static int ice_force_phys_link_state(struct ice_vsi *vsi, bool link_up)
{
	struct ice_aqc_get_phy_caps_data *pcaps;
	struct ice_aqc_set_phy_cfg_data *cfg;
	struct ice_port_info *pi;
	struct device *dev;
	int retcode;

	if (!vsi || !vsi->port_info || !vsi->back)
		return -EINVAL;
	if (vsi->type != ICE_VSI_PF)
		return 0;

	dev = ice_pf_to_dev(vsi->back);

	pi = vsi->port_info;

	pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL);
	if (!pcaps)
		return -ENOMEM;

	retcode = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_ACTIVE_CFG, pcaps,
				      NULL);
	if (retcode) {
		dev_err(dev, "Failed to get phy capabilities, VSI %d error %d\n",
			vsi->vsi_num, retcode);
		retcode = -EIO;
		goto out;
	}

	/* No change in link */
	if (link_up == !!(pcaps->caps & ICE_AQC_PHY_EN_LINK) &&
	    link_up == !!(pi->phy.link_info.link_info & ICE_AQ_LINK_UP))
		goto out;

	/* Use the current user PHY configuration. The current user PHY
	 * configuration is initialized during probe from PHY capabilities
	 * software mode, and updated on set PHY configuration.
	 */
	cfg = kmemdup(&pi->phy.curr_user_phy_cfg, sizeof(*cfg), GFP_KERNEL);
	if (!cfg) {
		retcode = -ENOMEM;
		goto out;
	}

	cfg->caps |= ICE_AQ_PHY_ENA_AUTO_LINK_UPDT;
	if (link_up)
		cfg->caps |= ICE_AQ_PHY_ENA_LINK;
	else
		cfg->caps &= ~ICE_AQ_PHY_ENA_LINK;

	retcode = ice_aq_set_phy_cfg(&vsi->back->hw, pi, cfg, NULL);
	if (retcode) {
		dev_err(dev, "Failed to set phy config, VSI %d error %d\n",
			vsi->vsi_num, retcode);
		retcode = -EIO;
	}

	kfree(cfg);
out:
	kfree(pcaps);
	return retcode;
}

/**
 * ice_init_nvm_phy_type - Initialize the NVM PHY type
 * @pi: port info structure
 *
 * Initialize nvm_phy_type_[low|high] for link lenient mode support
 */
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;

	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);

	if (status) {
		dev_err(ice_pf_to_dev(pf), "Get PHY capability failed.\n");
		err = -EIO;
		goto out;
	}

	pf->nvm_phy_type_hi = pcaps->phy_type_high;
	pf->nvm_phy_type_lo = pcaps->phy_type_low;

out:
	kfree(pcaps);
	return err;
}

/**
 * ice_init_link_dflt_override - Initialize link default override
 * @pi: port info structure
 *
 * Initialize link default override and PHY total port shutdown during probe
 */
static void ice_init_link_dflt_override(struct ice_port_info *pi)
{
	struct ice_link_default_override_tlv *ldo;
	struct ice_pf *pf = pi->hw->back;

	ldo = &pf->link_dflt_override;
	if (ice_get_link_default_override(ldo, pi))
		return;

	if (!(ldo->options & ICE_LINK_OVERRIDE_PORT_DIS))
		return;

	/* Enable Total Port Shutdown (override/replace link-down-on-close
	 * ethtool private flag) for ports with Port Disable bit set.
	 */
	set_bit(ICE_FLAG_TOTAL_PORT_SHUTDOWN_ENA, pf->flags);
	set_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, pf->flags);
}

/**
 * ice_init_phy_cfg_dflt_override - Initialize PHY cfg default override settings
 * @pi: port info structure
 *
 * If default override is enabled, initialize the user PHY cfg speed and FEC
 * settings using the default override mask from the NVM.
 *
 * The PHY should only be configured with the default override settings the
 * first time media is available. The ICE_LINK_DEFAULT_OVERRIDE_PENDING state
 * is used to indicate that the user PHY cfg default override is initialized
 * and the PHY has not been configured with the default override settings. The
 * state is set here, and cleared in ice_configure_phy the first time the PHY is
 * configured.
 *
 * This function should be called only if the FW doesn't support default
 * configuration mode, as reported by ice_fw_supports_report_dflt_cfg.
 */
static void ice_init_phy_cfg_dflt_override(struct ice_port_info *pi)
{
	struct ice_link_default_override_tlv *ldo;
	struct ice_aqc_set_phy_cfg_data *cfg;
	struct ice_phy_info *phy = &pi->phy;
	struct ice_pf *pf = pi->hw->back;

	ldo = &pf->link_dflt_override;

	/* If link default override is enabled, use to mask NVM PHY capabilities
	 * for speed and FEC default configuration.
	 */
	cfg = &phy->curr_user_phy_cfg;

	if (ldo->phy_type_low || ldo->phy_type_high) {
		cfg->phy_type_low = pf->nvm_phy_type_lo &
				    cpu_to_le64(ldo->phy_type_low);
		cfg->phy_type_high = pf->nvm_phy_type_hi &
				     cpu_to_le64(ldo->phy_type_high);
	}
	cfg->link_fec_opt = ldo->fec_options;
	phy->curr_user_fec_req = ICE_FEC_AUTO;

	set_bit(ICE_LINK_DEFAULT_OVERRIDE_PENDING, pf->state);
}

/**
 * ice_init_phy_user_cfg - Initialize the PHY user configuration
 * @pi: port info structure
 *
 * Initialize the current user PHY configuration, speed, FEC, and FC requested
 * mode to default. The PHY defaults are from get PHY capabilities topology
 * with media so call when media is first available. An error is returned if
 * called when media is not available. The PHY initialization completed state is
 * set here.
 *
 * These configurations are used when setting PHY
 * configuration. The user PHY configuration is updated on set PHY
 * configuration. Returns 0 on success, negative on failure
 */
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;

	if (!(phy->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE))
		return -EIO;

	pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL);
	if (!pcaps)
		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);
	else
		status = ice_aq_get_phy_caps(pi, false, ICE_AQC_REPORT_TOPO_CAP_MEDIA,
					     pcaps, NULL);
	if (status) {
		dev_err(ice_pf_to_dev(pf), "Get PHY capability failed.\n");
		err = -EIO;
		goto err_out;
	}

	ice_copy_phy_caps_to_cfg(pi, pcaps, &pi->phy.curr_user_phy_cfg);

	/* check if lenient mode is supported and enabled */
	if (ice_fw_supports_link_override(pi->hw) &&
	    !(pcaps->module_compliance_enforcement &
	      ICE_AQC_MOD_ENFORCE_STRICT_MODE)) {
		set_bit(ICE_FLAG_LINK_LENIENT_MODE_ENA, pf->flags);

		/* if the FW supports default PHY configuration mode, then the driver
		 * does not have to apply link override settings. If not,
		 * initialize user PHY configuration with link override values
		 */
		if (!ice_fw_supports_report_dflt_cfg(pi->hw) &&
		    (pf->link_dflt_override.options & ICE_LINK_OVERRIDE_EN)) {
			ice_init_phy_cfg_dflt_override(pi);
			goto out;
		}
	}

	/* if link default override is not enabled, set user flow control and
	 * FEC settings based on what get_phy_caps returned
	 */
	phy->curr_user_fec_req = ice_caps_to_fec_mode(pcaps->caps,
						      pcaps->link_fec_options);
	phy->curr_user_fc_req = ice_caps_to_fc_mode(pcaps->caps);

out:
	phy->curr_user_speed_req = ICE_AQ_LINK_SPEED_M;
	set_bit(ICE_PHY_INIT_COMPLETE, pf->state);
err_out:
	kfree(pcaps);
	return err;
}

/**
 * ice_configure_phy - configure PHY
 * @vsi: VSI of PHY
 *
 * Set the PHY configuration. If the current PHY configuration is the same as
 * the curr_user_phy_cfg, then do nothing to avoid link flap. Otherwise
 * configure the based get PHY capabilities for topology with media.
 */
static int ice_configure_phy(struct ice_vsi *vsi)
{
	struct device *dev = ice_pf_to_dev(vsi->back);
	struct ice_port_info *pi = vsi->port_info;
	struct ice_aqc_get_phy_caps_data *pcaps;
	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;

	/* Ensure we have media as we cannot configure a medialess port */
	if (!(phy->link_info.link_info & ICE_AQ_MEDIA_AVAILABLE))
		return -EPERM;

	ice_print_topo_conflict(vsi);

	if (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))
		return ice_force_phys_link_state(vsi, true);

	pcaps = kzalloc(sizeof(*pcaps), GFP_KERNEL);
	if (!pcaps)
		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;
		goto done;
	}

	/* If PHY enable link is configured and configuration has not changed,
	 * there's nothing to do
	 */
	if (pcaps->caps & ICE_AQC_PHY_EN_LINK &&
	    ice_phy_caps_equals_cfg(pcaps, &phy->curr_user_phy_cfg))
		goto done;

	/* 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);
	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;
		goto done;
	}

	cfg = kzalloc(sizeof(*cfg), GFP_KERNEL);
	if (!cfg) {
		err = -ENOMEM;
		goto done;
	}

	ice_copy_phy_caps_to_cfg(pi, pcaps, cfg);

	/* Speed - If default override pending, use curr_user_phy_cfg set in
	 * ice_init_phy_user_cfg_ldo.
	 */
	if (test_and_clear_bit(ICE_LINK_DEFAULT_OVERRIDE_PENDING,
			       vsi->back->state)) {
		cfg->phy_type_low = phy->curr_user_phy_cfg.phy_type_low;
		cfg->phy_type_high = phy->curr_user_phy_cfg.phy_type_high;
	} else {
		u64 phy_low = 0, phy_high = 0;

		ice_update_phy_type(&phy_low, &phy_high,
				    pi->phy.curr_user_speed_req);
		cfg->phy_type_low = pcaps->phy_type_low & cpu_to_le64(phy_low);
		cfg->phy_type_high = pcaps->phy_type_high &
				     cpu_to_le64(phy_high);
	}

	/* Can't provide what was requested; use PHY capabilities */
	if (!cfg->phy_type_low && !cfg->phy_type_high) {
		cfg->phy_type_low = pcaps->phy_type_low;
		cfg->phy_type_high = pcaps->phy_type_high;
	}

	/* FEC */
	ice_cfg_phy_fec(pi, cfg, phy->curr_user_fec_req);

	/* Can't provide what was requested; use PHY capabilities */
	if (cfg->link_fec_opt !=
	    (cfg->link_fec_opt & pcaps->link_fec_options)) {
		cfg->caps |= pcaps->caps & ICE_AQC_PHY_EN_AUTO_FEC;
		cfg->link_fec_opt = pcaps->link_fec_options;
	}

	/* Flow Control - always supported; no need to check against
	 * capabilities
	 */
	ice_cfg_phy_fc(pi, cfg, phy->curr_user_fc_req);

	/* 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;
	}

	kfree(cfg);
done:
	kfree(pcaps);
	return err;
}

/**
 * ice_check_media_subtask - Check for media
 * @pf: pointer to PF struct
 *
 * If media is available, then initialize PHY user configuration if it is not
 * been, and configure the PHY if the interface is up.
 */
static void ice_check_media_subtask(struct ice_pf *pf)
{
	struct ice_port_info *pi;
	struct ice_vsi *vsi;
	int err;

	/* No need to check for media if it's already present */
	if (!test_bit(ICE_FLAG_NO_MEDIA, pf->flags))
		return;

	vsi = ice_get_main_vsi(pf);
	if (!vsi)
		return;

	/* Refresh link info and check if media is present */
	pi = vsi->port_info;
	err = ice_update_link_info(pi);
	if (err)
		return;

	ice_check_module_power(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))
			ice_init_phy_user_cfg(pi);

		/* PHY settings are reset on media insertion, reconfigure
		 * PHY to preserve settings.
		 */
		if (test_bit(ICE_VSI_DOWN, vsi->state) &&
		    test_bit(ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, vsi->back->flags))
			return;

		err = ice_configure_phy(vsi);
		if (!err)
			clear_bit(ICE_FLAG_NO_MEDIA, pf->flags);

		/* A Link Status Event will be generated; the event handler
		 * will complete bringing the interface up
		 */
	}
}

/**
 * ice_service_task - manage and run subtasks
 * @work: pointer to work_struct contained by the PF struct
 */
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 */

	/* process reset requests first */
	ice_reset_subtask(pf);

	/* bail if a reset/recovery cycle is pending or rebuild failed */
	if (ice_is_reset_in_progress(pf->state) ||
	    test_bit(ICE_SUSPENDED, pf->state) ||
	    test_bit(ICE_NEEDS_RESTART, pf->state)) {
		ice_service_task_complete(pf);
		return;
	}

	ice_clean_adminq_subtask(pf);
	ice_check_media_subtask(pf);
	ice_check_for_hang_subtask(pf);
	ice_sync_fltr_subtask(pf);
	ice_handle_mdd_event(pf);
	ice_watchdog_subtask(pf);

	if (ice_is_safe_mode(pf)) {
		ice_service_task_complete(pf);
		return;
	}

	ice_process_vflr_event(pf);
	ice_clean_mailboxq_subtask(pf);
	ice_clean_sbq_subtask(pf);
	ice_sync_arfs_fltrs(pf);
	ice_flush_fdir_ctx(pf);

	/* Clear ICE_SERVICE_SCHED flag to allow scheduling next event */
	ice_service_task_complete(pf);

	/* If the tasks have taken longer than one service timer period
	 * or there is more work to be done, reset the service timer to
	 * schedule the service task now.
	 */
	if (time_after(jiffies, (start_time + pf->serv_tmr_period)) ||
	    test_bit(ICE_MDD_EVENT_PENDING, pf->state) ||
	    test_bit(ICE_VFLR_EVENT_PENDING, pf->state) ||
	    test_bit(ICE_MAILBOXQ_EVENT_PENDING, pf->state) ||
	    test_bit(ICE_FD_VF_FLUSH_CTX, pf->state) ||
	    test_bit(ICE_SIDEBANDQ_EVENT_PENDING, pf->state) ||
	    test_bit(ICE_ADMINQ_EVENT_PENDING, pf->state))
		mod_timer(&pf->serv_tmr, jiffies);
}

/**
 * ice_set_ctrlq_len - helper function to set controlq length
 * @hw: pointer to the HW instance
 */
static void ice_set_ctrlq_len(struct ice_hw *hw)
{
	hw->adminq.num_rq_entries = ICE_AQ_LEN;
	hw->adminq.num_sq_entries = ICE_AQ_LEN;
	hw->adminq.rq_buf_size = ICE_AQ_MAX_BUF_LEN;
	hw->adminq.sq_buf_size = ICE_AQ_MAX_BUF_LEN;
	hw->mailboxq.num_rq_entries = PF_MBX_ARQLEN_ARQLEN_M;
	hw->mailboxq.num_sq_entries = ICE_MBXSQ_LEN;
	hw->mailboxq.rq_buf_size = ICE_MBXQ_MAX_BUF_LEN;
	hw->mailboxq.sq_buf_size = ICE_MBXQ_MAX_BUF_LEN;
	hw->sbq.num_rq_entries = ICE_SBQ_LEN;
	hw->sbq.num_sq_entries = ICE_SBQ_LEN;
	hw->sbq.rq_buf_size = ICE_SBQ_MAX_BUF_LEN;
	hw->sbq.sq_buf_size = ICE_SBQ_MAX_BUF_LEN;
}

/**
 * ice_schedule_reset - schedule a reset
 * @pf: board private structure
 * @reset: reset being requested
 */
int ice_schedule_reset(struct ice_pf *pf, enum ice_reset_req reset)
{
	struct device *dev = ice_pf_to_dev(pf);

	/* bail out if earlier reset has failed */
	if (test_bit(ICE_RESET_FAILED, pf->state)) {
		dev_dbg(dev, "earlier reset has failed\n");
		return -EIO;
	}
	/* bail if reset/recovery already in progress */
	if (ice_is_reset_in_progress(pf->state)) {
		dev_dbg(dev, "Reset already in progress\n");
		return -EBUSY;
	}

	ice_unplug_aux_dev(pf);

	switch (reset) {
	case ICE_RESET_PFR:
		set_bit(ICE_PFR_REQ, pf->state);
		break;
	case ICE_RESET_CORER:
		set_bit(ICE_CORER_REQ, pf->state);
		break;
	case ICE_RESET_GLOBR:
		set_bit(ICE_GLOBR_REQ, pf->state);
		break;
	default:
		return -EINVAL;
	}

	ice_service_task_schedule(pf);
	return 0;
}

/**
 * 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
 */
static int ice_vsi_ena_irq(struct ice_vsi *vsi)
{
	struct ice_hw *hw = &vsi->back->hw;
	int i;

	ice_for_each_q_vector(vsi, i)
		ice_irq_dynamic_ena(hw, vsi, vsi->q_vectors[i]);

	ice_flush(hw);
	return 0;
}

/**
 * ice_vsi_req_irq_msix - get MSI-X vectors from the OS for the VSI
 * @vsi: the VSI being configured
 * @basename: name for the vector
 */
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;
	int vector, err;
	int irq_num;

	dev = ice_pf_to_dev(pf);
	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;

		if (q_vector->tx.ring && q_vector->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) {
			snprintf(q_vector->name, sizeof(q_vector->name) - 1,
				 "%s-%s-%d", basename, "rx", rx_int_idx++);
		} else if (q_vector->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)
			err = devm_request_irq(dev, irq_num, vsi->irq_handler,
					       IRQF_SHARED, q_vector->name,
					       q_vector);
		else
			err = devm_request_irq(dev, irq_num, vsi->irq_handler,
					       0, q_vector->name, q_vector);
		if (err) {
			netdev_err(vsi->netdev, "MSIX request_irq failed, error: %d\n",
				   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);
	}

	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);
		devm_free_irq(dev, irq_num, &vsi->q_vectors[vector]);
	}
	return err;
}

/**
 * ice_xdp_alloc_setup_rings - Allocate and setup Tx rings for XDP
 * @vsi: VSI to setup Tx rings used by XDP
 *
 * Return 0 on success and negative value on error
 */
static int ice_xdp_alloc_setup_rings(struct ice_vsi *vsi)
{
	struct device *dev = ice_pf_to_dev(vsi->back);
	int i;

	for (i = 0; i < vsi->num_xdp_txq; i++) {
		u16 xdp_q_idx = vsi->alloc_txq + i;
		struct ice_ring *xdp_ring;

		xdp_ring = kzalloc(sizeof(*xdp_ring), GFP_KERNEL);

		if (!xdp_ring)
			goto free_xdp_rings;

		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;
		xdp_ring->count = vsi->num_tx_desc;
		WRITE_ONCE(vsi->xdp_rings[i], xdp_ring);
		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);
	}

	return 0;

free_xdp_rings:
	for (; i >= 0; i--)
		if (vsi->xdp_rings[i] && vsi->xdp_rings[i]->desc)
			ice_free_tx_ring(vsi->xdp_rings[i]);
	return -ENOMEM;
}

/**
 * ice_vsi_assign_bpf_prog - set or clear bpf prog pointer on VSI
 * @vsi: VSI to set the bpf prog on
 * @prog: the bpf prog pointer
 */
static void ice_vsi_assign_bpf_prog(struct ice_vsi *vsi, struct bpf_prog *prog)
{
	struct bpf_prog *old_prog;
	int i;

	old_prog = xchg(&vsi->xdp_prog, 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);
}

/**
 * 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
 *
 * Return 0 on success and negative value on error
 */
int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog)
{
	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,
		.pf_map = pf->avail_txqs,
		.pf_map_size = pf->max_pf_txqs,
		.q_count = vsi->num_xdp_txq,
		.scatter_count = ICE_MAX_SCATTER_TXQS,
		.vsi_map = vsi->txq_map,
		.vsi_map_offset = vsi->alloc_txq,
		.mapping_mode = ICE_VSI_MAP_CONTIG
	};
	enum ice_status status;
	struct device *dev;
	int i, v_idx;

	dev = ice_pf_to_dev(pf);
	vsi->xdp_rings = devm_kcalloc(dev, vsi->num_xdp_txq,
				      sizeof(*vsi->xdp_rings), GFP_KERNEL);
	if (!vsi->xdp_rings)
		return -ENOMEM;

	vsi->xdp_mapping_mode = xdp_qs_cfg.mapping_mode;
	if (__ice_vsi_get_qs(&xdp_qs_cfg))
		goto err_map_xdp;

	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))
		return 0;

	/* tell the Tx scheduler that right now we have
	 * additional queues
	 */
	for (i = 0; i < vsi->tc_cfg.numtc; i++)
		max_txqs[i] = vsi->num_txq + vsi->num_xdp_txq;

	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;
	}
	ice_vsi_assign_bpf_prog(vsi, prog);

	return 0;
clear_xdp_rings:
	for (i = 0; i < vsi->num_xdp_txq; i++)
		if (vsi->xdp_rings[i]) {
			kfree_rcu(vsi->xdp_rings[i], rcu);
			vsi->xdp_rings[i] = NULL;
		}

err_map_xdp:
	mutex_lock(&pf->avail_q_mutex);
	for (i = 0; i < vsi->num_xdp_txq; 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);
	return -ENOMEM;
}

/**
 * ice_destroy_xdp_rings - undo the configuration made by ice_prepare_xdp_rings
 * @vsi: VSI to remove XDP rings
 *
 * Detach XDP rings from irq vectors, clean up the PF bitmap and free
 * resources
 */
int ice_destroy_xdp_rings(struct ice_vsi *vsi)
{
	u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 };
	struct ice_pf *pf = vsi->back;
	int i, v_idx;

	/* 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
	 */
	if (ice_is_reset_in_progress(pf->state) || !vsi->q_vectors[0])
		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;
	}

free_qmap:
	mutex_lock(&pf->avail_q_mutex);
	for (i = 0; i < vsi->num_xdp_txq; 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++)
		if (vsi->xdp_rings[i]) {
			if (vsi->xdp_rings[i]->desc)
				ice_free_tx_ring(vsi->xdp_rings[i]);
			kfree_rcu(vsi->xdp_rings[i], rcu);
			vsi->xdp_rings[i] = NULL;
		}

	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])
		return 0;

	ice_vsi_assign_bpf_prog(vsi, NULL);

	/* notify Tx scheduler that we destroyed XDP queues and bring
	 * back the old number of child nodes
	 */
	for (i = 0; i < vsi->tc_cfg.numtc; i++)
		max_txqs[i] = vsi->num_txq;

	/* change number of XDP Tx queues to 0 */
	vsi->num_xdp_txq = 0;

	return ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc,
			       max_txqs);
}

/**
 * ice_vsi_rx_napi_schedule - Schedule napi on RX queues from VSI
 * @vsi: VSI to schedule napi on
 */
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];

		if (rx_ring->xsk_pool)
			napi_schedule(&rx_ring->q_vector->napi);
	}
}

/**
 * ice_xdp_setup_prog - Add or remove XDP eBPF program
 * @vsi: VSI to setup XDP for
 * @prog: XDP program
 * @extack: netlink extended ack
 */
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);
	int ret = 0, xdp_ring_err = 0;

	if (frame_size > vsi->rx_buf_len) {
		NL_SET_ERR_MSG_MOD(extack, "MTU too large for loading XDP");
		return -EOPNOTSUPP;
	}

	/* need to stop netdev while setting up the program for Rx rings */
	if (if_running && !test_and_set_bit(ICE_VSI_DOWN, vsi->state)) {
		ret = ice_down(vsi);
		if (ret) {
			NL_SET_ERR_MSG_MOD(extack, "Preparing device for XDP attach failed");
			return ret;
		}
	}

	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");
	} else if (ice_is_xdp_ena_vsi(vsi) && !prog) {
		xdp_ring_err = ice_destroy_xdp_rings(vsi);
		if (xdp_ring_err)
			NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Tx resources failed");
	} else {
		ice_vsi_assign_bpf_prog(vsi, prog);
	}

	if (if_running)
		ret = ice_up(vsi);

	if (!ret && prog)
		ice_vsi_rx_napi_schedule(vsi);

	return (ret || xdp_ring_err) ? -ENOMEM : 0;
}

/**
 * ice_xdp_safe_mode - XDP handler for safe mode
 * @dev: netdevice
 * @xdp: XDP command
 */
static int ice_xdp_safe_mode(struct net_device __always_unused *dev,
			     struct netdev_bpf *xdp)
{
	NL_SET_ERR_MSG_MOD(xdp->extack,
			   "Please provide working DDP firmware package in order to use XDP\n"
			   "Refer to Documentation/networking/device_drivers/ethernet/intel/ice.rst");
	return -EOPNOTSUPP;
}

/**
 * ice_xdp - implements XDP handler
 * @dev: netdevice
 * @xdp: XDP command
 */
static 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;

	if (vsi->type != ICE_VSI_PF) {
		NL_SET_ERR_MSG_MOD(xdp->extack, "XDP can be loaded only on PF VSI");
		return -EINVAL;
	}

	switch (xdp->command) {
	case XDP_SETUP_PROG:
		return ice_xdp_setup_prog(vsi, xdp->prog, xdp->extack);
	case XDP_SETUP_XSK_POOL:
		return ice_xsk_pool_setup(vsi, xdp->xsk.pool,
					  xdp->xsk.queue_id);
	default:
		return -EINVAL;
	}
}

/**
 * ice_ena_misc_vector - enable the non-queue interrupts
 * @pf: board private structure
 */
static void ice_ena_misc_vector(struct ice_pf *pf)
{
	struct ice_hw *hw = &pf->hw;
	u32 val;

	/* Disable anti-spoof detection interrupt to prevent spurious event
	 * interrupts during a function reset. Anti-spoof functionally is
	 * still supported.
	 */
	val = rd32(hw, GL_MDCK_TX_TDPU);
	val |= GL_MDCK_TX_TDPU_RCU_ANTISPOOF_ITR_DIS_M;
	wr32(hw, GL_MDCK_TX_TDPU, val);

	/* clear things first */
	wr32(hw, PFINT_OICR_ENA, 0);	/* disable all */
	rd32(hw, PFINT_OICR);		/* read to clear */

	val = (PFINT_OICR_ECC_ERR_M |
	       PFINT_OICR_MAL_DETECT_M |
	       PFINT_OICR_GRST_M |
	       PFINT_OICR_PCI_EXCEPTION_M |
	       PFINT_OICR_VFLR_M |
	       PFINT_OICR_HMC_ERR_M |
	       PFINT_OICR_PE_PUSH_M |
	       PFINT_OICR_PE_CRITERR_M);

	wr32(hw, PFINT_OICR_ENA, val);

	/* SW_ITR_IDX = 0, but don't change INTENA */
	wr32(hw, GLINT_DYN_CTL(pf->oicr_idx),
	     GLINT_DYN_CTL_SW_ITR_INDX_M | GLINT_DYN_CTL_INTENA_MSK_M);
}

/**
 * ice_misc_intr - misc interrupt handler
 * @irq: interrupt number
 * @data: pointer to a q_vector
 */
static irqreturn_t ice_misc_intr(int __always_unused irq, void *data)
{
	struct ice_pf *pf = (struct ice_pf *)data;
	struct ice_hw *hw = &pf->hw;
	irqreturn_t ret = IRQ_NONE;
	struct device *dev;
	u32 oicr, ena_mask;

	dev = ice_pf_to_dev(pf);
	set_bit(ICE_ADMINQ_EVENT_PENDING, pf->state);
	set_bit(ICE_MAILBOXQ_EVENT_PENDING, pf->state);
	set_bit(ICE_SIDEBANDQ_EVENT_PENDING, pf->state);

	oicr = rd32(hw, PFINT_OICR);
	ena_mask = rd32(hw, PFINT_OICR_ENA);

	if (oicr & PFINT_OICR_SWINT_M) {
		ena_mask &= ~PFINT_OICR_SWINT_M;
		pf->sw_int_count++;
	}

	if (oicr & PFINT_OICR_MAL_DETECT_M) {
		ena_mask &= ~PFINT_OICR_MAL_DETECT_M;
		set_bit(ICE_MDD_EVENT_PENDING, pf->state);
	}
	if (oicr & PFINT_OICR_VFLR_M) {
		/* disable any further VFLR event notifications */
		if (test_bit(ICE_VF_RESETS_DISABLED, pf->state)) {
			u32 reg = rd32(hw, PFINT_OICR_ENA);

			reg &= ~PFINT_OICR_VFLR_M;
			wr32(hw, PFINT_OICR_ENA, reg);
		} else {
			ena_mask &= ~PFINT_OICR_VFLR_M;
			set_bit(ICE_VFLR_EVENT_PENDING, pf->state);
		}
	}

	if (oicr & PFINT_OICR_GRST_M) {
		u32 reset;

		/* 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;

		if (reset == ICE_RESET_CORER)
			pf->corer_count++;
		else if (reset == ICE_RESET_GLOBR)
			pf->globr_count++;
		else if (reset == ICE_RESET_EMPR)
			pf->empr_count++;
		else
			dev_dbg(dev, "Invalid reset type %d\n", reset);

		/* If a reset cycle isn't already in progress, we set a bit in
		 * pf->state so that the service task can start a reset/rebuild.
		 */
		if (!test_and_set_bit(ICE_RESET_OICR_RECV, pf->state)) {
			if (reset == ICE_RESET_CORER)
				set_bit(ICE_CORER_RECV, pf->state);
			else if (reset == ICE_RESET_GLOBR)
				set_bit(ICE_GLOBR_RECV, pf->state);
			else
				set_bit(ICE_EMPR_RECV, pf->state);

			/* There are couple of different bits at play here.
			 * hw->reset_ongoing indicates whether the hardware is
			 * in reset. This is set to true when a reset interrupt
			 * is received and set back to false after the driver
			 * has determined that the hardware is out of reset.
			 *
			 * ICE_RESET_OICR_RECV in pf->state indicates
			 * that a post reset rebuild is required before the
			 * driver is operational again. This is set above.
			 *
			 * As this is the start of the reset/rebuild cycle, set
			 * both to indicate that.
			 */
			hw->reset_ongoing = true;
		}
	}

	if (oicr & PFINT_OICR_TSYN_TX_M) {
		ena_mask &= ~PFINT_OICR_TSYN_TX_M;
		ice_ptp_process_ts(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);
	}

#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;

		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 */
	oicr &= ena_mask;
	if (oicr) {
		dev_dbg(dev, "unhandled interrupt oicr=0x%08x\n", oicr);
		/* If a critical error is pending there is no choice but to
		 * reset the device.
		 */
		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);

	return ret;
}

/**
 * ice_dis_ctrlq_interrupts - disable control queue interrupts
 * @hw: pointer to HW structure
 */
static void ice_dis_ctrlq_interrupts(struct ice_hw *hw)
{
	/* disable Admin queue Interrupt causes */
	wr32(hw, PFINT_FW_CTL,
	     rd32(hw, PFINT_FW_CTL) & ~PFINT_FW_CTL_CAUSE_ENA_M);

	/* disable Mailbox queue Interrupt causes */
	wr32(hw, PFINT_MBX_CTL,
	     rd32(hw, PFINT_MBX_CTL) & ~PFINT_MBX_CTL_CAUSE_ENA_M);

	wr32(hw, PFINT_SB_CTL,
	     rd32(hw, PFINT_SB_CTL) & ~PFINT_SB_CTL_CAUSE_ENA_M);

	/* disable Control queue Interrupt causes */
	wr32(hw, PFINT_OICR_CTL,
	     rd32(hw, PFINT_OICR_CTL) & ~PFINT_OICR_CTL_CAUSE_ENA_M);

	ice_flush(hw);
}

/**
 * ice_free_irq_msix_misc - Unroll misc vector setup
 * @pf: board private structure
 */
static void ice_free_irq_msix_misc(struct ice_pf *pf)
{
	struct ice_hw *hw = &pf->hw;

	ice_dis_ctrlq_interrupts(hw);

	/* disable OICR interrupt */
	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);
	}

	pf->num_avail_sw_msix += 1;
	ice_free_res(pf->irq_tracker, pf->oicr_idx, ICE_RES_MISC_VEC_ID);
}

/**
 * ice_ena_ctrlq_interrupts - enable control queue interrupts
 * @hw: pointer to HW structure
 * @reg_idx: HW vector index to associate the control queue interrupts with
 */
static void ice_ena_ctrlq_interrupts(struct ice_hw *hw, u16 reg_idx)
{
	u32 val;

	val = ((reg_idx & PFINT_OICR_CTL_MSIX_INDX_M) |
	       PFINT_OICR_CTL_CAUSE_ENA_M);
	wr32(hw, PFINT_OICR_CTL, val);

	/* enable Admin queue Interrupt causes */
	val = ((reg_idx & PFINT_FW_CTL_MSIX_INDX_M) |
	       PFINT_FW_CTL_CAUSE_ENA_M);
	wr32(hw, PFINT_FW_CTL, val);

	/* enable Mailbox queue Interrupt causes */
	val = ((reg_idx & PFINT_MBX_CTL_MSIX_INDX_M) |
	       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);

	ice_flush(hw);
}

/**
 * ice_req_irq_msix_misc - Setup the misc vector to handle non queue events
 * @pf: board private structure
 *
 * This sets up the handler for MSIX 0, which is used to manage the
 * non-queue interrupts, e.g. AdminQ and errors. This is not used
 * when in MSI or Legacy interrupt mode.
 */
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;

	if (!pf->int_name[0])
		snprintf(pf->int_name, sizeof(pf->int_name) - 1, "%s-%s:misc",
			 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.
	 */
	if (ice_is_reset_in_progress(pf->state))
		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;

	pf->num_avail_sw_msix -= 1;
	pf->oicr_idx = (u16)oicr_idx;

	err = devm_request_irq(dev, pf->msix_entries[pf->oicr_idx].vector,
			       ice_misc_intr, 0, pf->int_name, 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;
		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),
	     ITR_REG_ALIGN(ICE_ITR_8K) >> ICE_ITR_GRAN_S);

	ice_flush(hw);
	ice_irq_dynamic_ena(hw, NULL, NULL);

	return 0;
}

/**
 * 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
 */
static void ice_set_ops(struct net_device *netdev)
{
	struct ice_pf *pf = ice_netdev_to_pf(netdev);

	if (ice_is_safe_mode(pf)) {
		netdev->netdev_ops = &ice_netdev_safe_mode_ops;
		ice_set_ethtool_safe_mode_ops(netdev);
		return;
	}

	netdev->netdev_ops = &ice_netdev_ops;
	netdev->udp_tunnel_nic_info = &pf->hw.udp_tunnel_nic;
	ice_set_ethtool_ops(netdev);
}

/**
 * ice_set_netdev_features - set features for the given netdev
 * @netdev: netdev instance
 */
static void ice_set_netdev_features(struct net_device *netdev)
{
	struct ice_pf *pf = ice_netdev_to_pf(netdev);
	netdev_features_t csumo_features;
	netdev_features_t vlano_features;
	netdev_features_t dflt_features;
	netdev_features_t tso_features;

	if (ice_is_safe_mode(pf)) {
		/* safe mode */
		netdev->features = NETIF_F_SG | NETIF_F_HIGHDMA;
		netdev->hw_features = netdev->features;
		return;
	}

	dflt_features = NETIF_F_SG	|
			NETIF_F_HIGHDMA	|
			NETIF_F_NTUPLE	|
			NETIF_F_RXHASH;

	csumo_features = NETIF_F_RXCSUM	  |
			 NETIF_F_IP_CSUM  |
			 NETIF_F_SCTP_CRC |
			 NETIF_F_IPV6_CSUM;

	vlano_features = NETIF_F_HW_VLAN_CTAG_FILTER |
			 NETIF_F_HW_VLAN_CTAG_TX     |
			 NETIF_F_HW_VLAN_CTAG_RX;

	tso_features = NETIF_F_TSO			|
		       NETIF_F_TSO_ECN			|
		       NETIF_F_TSO6			|
		       NETIF_F_GSO_GRE			|
		       NETIF_F_GSO_UDP_TUNNEL		|
		       NETIF_F_GSO_GRE_CSUM		|
		       NETIF_F_GSO_UDP_TUNNEL_CSUM	|
		       NETIF_F_GSO_PARTIAL		|
		       NETIF_F_GSO_IPXIP4		|
		       NETIF_F_GSO_IPXIP6		|
		       NETIF_F_GSO_UDP_L4;

	netdev->gso_partial_features |= NETIF_F_GSO_UDP_TUNNEL_CSUM |
					NETIF_F_GSO_GRE_CSUM;
	/* set features that user can change */
	netdev->hw_features = dflt_features | csumo_features |
			      vlano_features | tso_features;

	/* add support for HW_CSUM on packets with MPLS header */
	netdev->mpls_features =  NETIF_F_HW_CSUM;

	/* enable features */
	netdev->features |= netdev->hw_features;
	/* 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;

	/* Setup netdev TC information */
	ice_vsi_cfg_netdev_tc(vsi, vsi->tc_cfg.ena_tc);

	/* setup watchdog timeout value to be 5 second */
	netdev->watchdog_timeo = 5 * HZ;

	netdev->min_mtu = ETH_MIN_MTU;
	netdev->max_mtu = ICE_MAX_MTU;

	return 0;
}

/**
 * ice_fill_rss_lut - Fill the RSS lookup table with default values
 * @lut: Lookup table
 * @rss_table_size: Lookup table size
 * @rss_size: Range of queue number for hashing
 */
void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size)
{
	u16 i;

	for (i = 0; i < rss_table_size; i++)
		lut[i] = i % rss_size;
}

/**
 * ice_pf_vsi_setup - Set up a PF VSI
 * @pf: board private structure
 * @pi: pointer to the port_info instance
 *
 * Returns pointer to the successfully allocated VSI software struct