// SPDX-License-Identifier: GPL-2.0
/* Copyright(c) 2013 - 2021 Intel Corporation. */

#include <linux/etherdevice.h>
#include <linux/of_net.h>
#include <linux/pci.h>
#include <linux/bpf.h>
#include <generated/utsrelease.h>
#include <linux/crash_dump.h>

/* Local includes */
#include "i40e.h"
#include "i40e_diag.h"
#include "i40e_xsk.h"
#include <net/udp_tunnel.h>
#include <net/xdp_sock_drv.h>
/* All i40e tracepoints are defined by the include below, which
 * must be included exactly once across the whole kernel with
 * CREATE_TRACE_POINTS defined
 */
#define CREATE_TRACE_POINTS
#include "i40e_trace.h"

const char i40e_driver_name[] = "i40e";
static const char i40e_driver_string[] =
			"Intel(R) Ethernet Connection XL710 Network Driver";

static const char i40e_copyright[] = "Copyright (c) 2013 - 2019 Intel Corporation.";

/* a bit of forward declarations */
static void i40e_vsi_reinit_locked(struct i40e_vsi *vsi);
static void i40e_handle_reset_warning(struct i40e_pf *pf, bool lock_acquired);
static int i40e_add_vsi(struct i40e_vsi *vsi);
static int i40e_add_veb(struct i40e_veb *veb, struct i40e_vsi *vsi);
static int i40e_setup_pf_switch(struct i40e_pf *pf, bool reinit, bool lock_acquired);
static int i40e_setup_misc_vector(struct i40e_pf *pf);
static void i40e_determine_queue_usage(struct i40e_pf *pf);
static int i40e_setup_pf_filter_control(struct i40e_pf *pf);
static void i40e_prep_for_reset(struct i40e_pf *pf);
static void i40e_reset_and_rebuild(struct i40e_pf *pf, bool reinit,
				   bool lock_acquired);
static int i40e_reset(struct i40e_pf *pf);
static void i40e_rebuild(struct i40e_pf *pf, bool reinit, bool lock_acquired);
static int i40e_setup_misc_vector_for_recovery_mode(struct i40e_pf *pf);
static int i40e_restore_interrupt_scheme(struct i40e_pf *pf);
static bool i40e_check_recovery_mode(struct i40e_pf *pf);
static int i40e_init_recovery_mode(struct i40e_pf *pf, struct i40e_hw *hw);
static void i40e_fdir_sb_setup(struct i40e_pf *pf);
static int i40e_veb_get_bw_info(struct i40e_veb *veb);
static int i40e_get_capabilities(struct i40e_pf *pf,
				 enum i40e_admin_queue_opc list_type);
static bool i40e_is_total_port_shutdown_enabled(struct i40e_pf *pf);

/* i40e_pci_tbl - PCI Device ID Table
 *
 * Last entry must be all 0s
 *
 * { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
 *   Class, Class Mask, private data (not used) }
 */
static const struct pci_device_id i40e_pci_tbl[] = {
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_XL710), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QEMU), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_B), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_C), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_A), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_B), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_C), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_BC), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T4), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_BC), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_SFP), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_B), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_X722), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_QSFP_X722), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_1G_BASE_T_X722), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_10G_BASE_T_X722), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_I_X722), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_X722_A), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_20G_KR2_A), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_X710_N3000), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_XXV710_N3000), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_25G_B), 0},
	{PCI_VDEVICE(INTEL, I40E_DEV_ID_25G_SFP28), 0},
	/* required last entry */
	{0, }
};
MODULE_DEVICE_TABLE(pci, i40e_pci_tbl);

#define I40E_MAX_VF_COUNT 128
static int debug = -1;
module_param(debug, uint, 0);
MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all), Debug mask (0x8XXXXXXX)");

MODULE_AUTHOR("Intel Corporation, <e1000-devel@lists.sourceforge.net>");
MODULE_DESCRIPTION("Intel(R) Ethernet Connection XL710 Network Driver");
MODULE_LICENSE("GPL v2");

static struct workqueue_struct *i40e_wq;

static void netdev_hw_addr_refcnt(struct i40e_mac_filter *f,
				  struct net_device *netdev, int delta)
{
	struct netdev_hw_addr *ha;

	if (!f || !netdev)
		return;

	netdev_for_each_mc_addr(ha, netdev) {
		if (ether_addr_equal(ha->addr, f->macaddr)) {
			ha->refcount += delta;
			if (ha->refcount <= 0)
				ha->refcount = 1;
			break;
		}
	}
}

/**
 * i40e_allocate_dma_mem_d - OS specific memory alloc for shared code
 * @hw:   pointer to the HW structure
 * @mem:  ptr to mem struct to fill out
 * @size: size of memory requested
 * @alignment: what to align the allocation to
 **/
int i40e_allocate_dma_mem_d(struct i40e_hw *hw, struct i40e_dma_mem *mem,
			    u64 size, u32 alignment)
{
	struct i40e_pf *pf = (struct i40e_pf *)hw->back;

	mem->size = ALIGN(size, alignment);
	mem->va = dma_alloc_coherent(&pf->pdev->dev, mem->size, &mem->pa,
				     GFP_KERNEL);
	if (!mem->va)
		return -ENOMEM;

	return 0;
}

/**
 * i40e_free_dma_mem_d - OS specific memory free for shared code
 * @hw:   pointer to the HW structure
 * @mem:  ptr to mem struct to free
 **/
int i40e_free_dma_mem_d(struct i40e_hw *hw, struct i40e_dma_mem *mem)
{
	struct i40e_pf *pf = (struct i40e_pf *)hw->back;

	dma_free_coherent(&pf->pdev->dev, mem->size, mem->va, mem->pa);
	mem->va = NULL;
	mem->pa = 0;
	mem->size = 0;

	return 0;
}

/**
 * i40e_allocate_virt_mem_d - OS specific memory alloc for shared code
 * @hw:   pointer to the HW structure
 * @mem:  ptr to mem struct to fill out
 * @size: size of memory requested
 **/
int i40e_allocate_virt_mem_d(struct i40e_hw *hw, struct i40e_virt_mem *mem,
			     u32 size)
{
	mem->size = size;
	mem->va = kzalloc(size, GFP_KERNEL);

	if (!mem->va)
		return -ENOMEM;

	return 0;
}

/**
 * i40e_free_virt_mem_d - OS specific memory free for shared code
 * @hw:   pointer to the HW structure
 * @mem:  ptr to mem struct to free
 **/
int i40e_free_virt_mem_d(struct i40e_hw *hw, struct i40e_virt_mem *mem)
{
	/* it's ok to kfree a NULL pointer */
	kfree(mem->va);
	mem->va = NULL;
	mem->size = 0;

	return 0;
}

/**
 * i40e_get_lump - find a lump of free generic resource
 * @pf: board private structure
 * @pile: the pile of resource to search
 * @needed: the number of items needed
 * @id: an owner id to stick on the items assigned
 *
 * Returns the base item index of the lump, or negative for error
 **/
static int i40e_get_lump(struct i40e_pf *pf, struct i40e_lump_tracking *pile,
			 u16 needed, u16 id)
{
	int ret = -ENOMEM;
	int i, j;

	if (!pile || needed == 0 || id >= I40E_PILE_VALID_BIT) {
		dev_info(&pf->pdev->dev,
			 "param err: pile=%s needed=%d id=0x%04x\n",
			 pile ? "<valid>" : "<null>", needed, id);
		return -EINVAL;
	}

	/* Allocate last queue in the pile for FDIR VSI queue
	 * so it doesn't fragment the qp_pile
	 */
	if (pile == pf->qp_pile && pf->vsi[id]->type == I40E_VSI_FDIR) {
		if (pile->list[pile->num_entries - 1] & I40E_PILE_VALID_BIT) {
			dev_err(&pf->pdev->dev,
				"Cannot allocate queue %d for I40E_VSI_FDIR\n",
				pile->num_entries - 1);
			return -ENOMEM;
		}
		pile->list[pile->num_entries - 1] = id | I40E_PILE_VALID_BIT;
		return pile->num_entries - 1;
	}

	i = 0;
	while (i < pile->num_entries) {
		/* skip already allocated entries */
		if (pile->list[i] & I40E_PILE_VALID_BIT) {
			i++;
			continue;
		}

		/* do we have enough in this lump? */
		for (j = 0; (j < needed) && ((i+j) < pile->num_entries); j++) {
			if (pile->list[i+j] & I40E_PILE_VALID_BIT)
				break;
		}

		if (j == needed) {
			/* there was enough, so assign it to the requestor */
			for (j = 0; j < needed; j++)
				pile->list[i+j] = id | I40E_PILE_VALID_BIT;
			ret = i;
			break;
		}

		/* not enough, so skip over it and continue looking */
		i += j;
	}

	return ret;
}

/**
 * i40e_put_lump - return a lump of generic resource
 * @pile: the pile of resource to search
 * @index: the base item index
 * @id: the owner id of the items assigned
 *
 * Returns the count of items in the lump
 **/
static int i40e_put_lump(struct i40e_lump_tracking *pile, u16 index, u16 id)
{
	int valid_id = (id | I40E_PILE_VALID_BIT);
	int count = 0;
	u16 i;

	if (!pile || index >= pile->num_entries)
		return -EINVAL;

	for (i = index;
	     i < pile->num_entries && pile->list[i] == valid_id;
	     i++) {
		pile->list[i] = 0;
		count++;
	}


	return count;
}

/**
 * i40e_find_vsi_from_id - searches for the vsi with the given id
 * @pf: the pf structure to search for the vsi
 * @id: id of the vsi it is searching for
 **/
struct i40e_vsi *i40e_find_vsi_from_id(struct i40e_pf *pf, u16 id)
{
	int i;

	for (i = 0; i < pf->num_alloc_vsi; i++)
		if (pf->vsi[i] && (pf->vsi[i]->id == id))
			return pf->vsi[i];

	return NULL;
}

/**
 * i40e_service_event_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 i40e_service_event_schedule(struct i40e_pf *pf)
{
	if ((!test_bit(__I40E_DOWN, pf->state) &&
	     !test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state)) ||
	      test_bit(__I40E_RECOVERY_MODE, pf->state))
		queue_work(i40e_wq, &pf->service_task);
}

/**
 * i40e_tx_timeout - Respond to a Tx Hang
 * @netdev: network interface device structure
 * @txqueue: queue number timing out
 *
 * If any port has noticed a Tx timeout, it is likely that the whole
 * device is munged, not just the one netdev port, so go for the full
 * reset.
 **/
static void i40e_tx_timeout(struct net_device *netdev, unsigned int txqueue)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;
	struct i40e_pf *pf = vsi->back;
	struct i40e_ring *tx_ring = NULL;
	unsigned int i;
	u32 head, val;

	pf->tx_timeout_count++;

	/* with txqueue index, find the tx_ring struct */
	for (i = 0; i < vsi->num_queue_pairs; i++) {
		if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) {
			if (txqueue ==
			    vsi->tx_rings[i]->queue_index) {
				tx_ring = vsi->tx_rings[i];
				break;
			}
		}
	}

	if (time_after(jiffies, (pf->tx_timeout_last_recovery + HZ*20)))
		pf->tx_timeout_recovery_level = 1;  /* reset after some time */
	else if (time_before(jiffies,
		      (pf->tx_timeout_last_recovery + netdev->watchdog_timeo)))
		return;   /* don't do any new action before the next timeout */

	/* don't kick off another recovery if one is already pending */
	if (test_and_set_bit(__I40E_TIMEOUT_RECOVERY_PENDING, pf->state))
		return;

	if (tx_ring) {
		head = i40e_get_head(tx_ring);
		/* Read interrupt register */
		if (pf->flags & I40E_FLAG_MSIX_ENABLED)
			val = rd32(&pf->hw,
			     I40E_PFINT_DYN_CTLN(tx_ring->q_vector->v_idx +
						tx_ring->vsi->base_vector - 1));
		else
			val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0);

		netdev_info(netdev, "tx_timeout: VSI_seid: %d, Q %d, NTC: 0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x, INT: 0x%x\n",
			    vsi->seid, txqueue, tx_ring->next_to_clean,
			    head, tx_ring->next_to_use,
			    readl(tx_ring->tail), val);
	}

	pf->tx_timeout_last_recovery = jiffies;
	netdev_info(netdev, "tx_timeout recovery level %d, txqueue %d\n",
		    pf->tx_timeout_recovery_level, txqueue);

	switch (pf->tx_timeout_recovery_level) {
	case 1:
		set_bit(__I40E_PF_RESET_REQUESTED, pf->state);
		break;
	case 2:
		set_bit(__I40E_CORE_RESET_REQUESTED, pf->state);
		break;
	case 3:
		set_bit(__I40E_GLOBAL_RESET_REQUESTED, pf->state);
		break;
	default:
		netdev_err(netdev, "tx_timeout recovery unsuccessful, device is in non-recoverable state.\n");
		set_bit(__I40E_DOWN_REQUESTED, pf->state);
		set_bit(__I40E_VSI_DOWN_REQUESTED, vsi->state);
		break;
	}

	i40e_service_event_schedule(pf);
	pf->tx_timeout_recovery_level++;
}

/**
 * i40e_get_vsi_stats_struct - Get System Network Statistics
 * @vsi: the VSI we care about
 *
 * Returns the address of the device statistics structure.
 * The statistics are actually updated from the service task.
 **/
struct rtnl_link_stats64 *i40e_get_vsi_stats_struct(struct i40e_vsi *vsi)
{
	return &vsi->net_stats;
}

/**
 * i40e_get_netdev_stats_struct_tx - populate stats from a Tx ring
 * @ring: Tx ring to get statistics from
 * @stats: statistics entry to be updated
 **/
static void i40e_get_netdev_stats_struct_tx(struct i40e_ring *ring,
					    struct rtnl_link_stats64 *stats)
{
	u64 bytes, packets;
	unsigned int start;

	do {
		start = u64_stats_fetch_begin(&ring->syncp);
		packets = ring->stats.packets;
		bytes   = ring->stats.bytes;
	} while (u64_stats_fetch_retry(&ring->syncp, start));

	stats->tx_packets += packets;
	stats->tx_bytes   += bytes;
}

/**
 * i40e_get_netdev_stats_struct - Get statistics for netdev interface
 * @netdev: network interface device structure
 * @stats: data structure to store statistics
 *
 * Returns the address of the device statistics structure.
 * The statistics are actually updated from the service task.
 **/
static void i40e_get_netdev_stats_struct(struct net_device *netdev,
				  struct rtnl_link_stats64 *stats)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;
	struct rtnl_link_stats64 *vsi_stats = i40e_get_vsi_stats_struct(vsi);
	struct i40e_ring *ring;
	int i;

	if (test_bit(__I40E_VSI_DOWN, vsi->state))
		return;

	if (!vsi->tx_rings)
		return;

	rcu_read_lock();
	for (i = 0; i < vsi->num_queue_pairs; i++) {
		u64 bytes, packets;
		unsigned int start;

		ring = READ_ONCE(vsi->tx_rings[i]);
		if (!ring)
			continue;
		i40e_get_netdev_stats_struct_tx(ring, stats);

		if (i40e_enabled_xdp_vsi(vsi)) {
			ring = READ_ONCE(vsi->xdp_rings[i]);
			if (!ring)
				continue;
			i40e_get_netdev_stats_struct_tx(ring, stats);
		}

		ring = READ_ONCE(vsi->rx_rings[i]);
		if (!ring)
			continue;
		do {
			start   = u64_stats_fetch_begin(&ring->syncp);
			packets = ring->stats.packets;
			bytes   = ring->stats.bytes;
		} while (u64_stats_fetch_retry(&ring->syncp, start));

		stats->rx_packets += packets;
		stats->rx_bytes   += bytes;

	}
	rcu_read_unlock();

	/* following stats updated by i40e_watchdog_subtask() */
	stats->multicast	= vsi_stats->multicast;
	stats->tx_errors	= vsi_stats->tx_errors;
	stats->tx_dropped	= vsi_stats->tx_dropped;
	stats->rx_errors	= vsi_stats->rx_errors;
	stats->rx_dropped	= vsi_stats->rx_dropped;
	stats->rx_crc_errors	= vsi_stats->rx_crc_errors;
	stats->rx_length_errors	= vsi_stats->rx_length_errors;
}

/**
 * i40e_vsi_reset_stats - Resets all stats of the given vsi
 * @vsi: the VSI to have its stats reset
 **/
void i40e_vsi_reset_stats(struct i40e_vsi *vsi)
{
	struct rtnl_link_stats64 *ns;
	int i;

	if (!vsi)
		return;

	ns = i40e_get_vsi_stats_struct(vsi);
	memset(ns, 0, sizeof(*ns));
	memset(&vsi->net_stats_offsets, 0, sizeof(vsi->net_stats_offsets));
	memset(&vsi->eth_stats, 0, sizeof(vsi->eth_stats));
	memset(&vsi->eth_stats_offsets, 0, sizeof(vsi->eth_stats_offsets));
	if (vsi->rx_rings && vsi->rx_rings[0]) {
		for (i = 0; i < vsi->num_queue_pairs; i++) {
			memset(&vsi->rx_rings[i]->stats, 0,
			       sizeof(vsi->rx_rings[i]->stats));
			memset(&vsi->rx_rings[i]->rx_stats, 0,
			       sizeof(vsi->rx_rings[i]->rx_stats));
			memset(&vsi->tx_rings[i]->stats, 0,
			       sizeof(vsi->tx_rings[i]->stats));
			memset(&vsi->tx_rings[i]->tx_stats, 0,
			       sizeof(vsi->tx_rings[i]->tx_stats));
		}
	}
	vsi->stat_offsets_loaded = false;
}

/**
 * i40e_pf_reset_stats - Reset all of the stats for the given PF
 * @pf: the PF to be reset
 **/
void i40e_pf_reset_stats(struct i40e_pf *pf)
{
	int i;

	memset(&pf->stats, 0, sizeof(pf->stats));
	memset(&pf->stats_offsets, 0, sizeof(pf->stats_offsets));
	pf->stat_offsets_loaded = false;

	for (i = 0; i < I40E_MAX_VEB; i++) {
		if (pf->veb[i]) {
			memset(&pf->veb[i]->stats, 0,
			       sizeof(pf->veb[i]->stats));
			memset(&pf->veb[i]->stats_offsets, 0,
			       sizeof(pf->veb[i]->stats_offsets));
			memset(&pf->veb[i]->tc_stats, 0,
			       sizeof(pf->veb[i]->tc_stats));
			memset(&pf->veb[i]->tc_stats_offsets, 0,
			       sizeof(pf->veb[i]->tc_stats_offsets));
			pf->veb[i]->stat_offsets_loaded = false;
		}
	}
	pf->hw_csum_rx_error = 0;
}

/**
 * i40e_compute_pci_to_hw_id - compute index form PCI function.
 * @vsi: ptr to the VSI to read from.
 * @hw: ptr to the hardware info.
 **/
static u32 i40e_compute_pci_to_hw_id(struct i40e_vsi *vsi, struct i40e_hw *hw)
{
	int pf_count = i40e_get_pf_count(hw);

	if (vsi->type == I40E_VSI_SRIOV)
		return (hw->port * BIT(7)) / pf_count + vsi->vf_id;

	return hw->port + BIT(7);
}

/**
 * i40e_stat_update64 - read and update a 64 bit stat from the chip.
 * @hw: ptr to the hardware info.
 * @hireg: the high 32 bit reg to read.
 * @loreg: the low 32 bit reg to read.
 * @offset_loaded: has the initial offset been loaded yet.
 * @offset: ptr to current offset value.
 * @stat: ptr to the stat.
 *
 * Since the device stats are not reset at PFReset, they will not
 * be zeroed when the driver starts.  We'll save the first values read
 * and use them as offsets to be subtracted from the raw values in order
 * to report stats that count from zero.
 **/
static void i40e_stat_update64(struct i40e_hw *hw, u32 hireg, u32 loreg,
			       bool offset_loaded, u64 *offset, u64 *stat)
{
	u64 new_data;

	new_data = rd64(hw, loreg);

	if (!offset_loaded || new_data < *offset)
		*offset = new_data;
	*stat = new_data - *offset;
}

/**
 * i40e_stat_update48 - read and update a 48 bit stat from the chip
 * @hw: ptr to the hardware info
 * @hireg: the high 32 bit reg to read
 * @loreg: the low 32 bit reg to read
 * @offset_loaded: has the initial offset been loaded yet
 * @offset: ptr to current offset value
 * @stat: ptr to the stat
 *
 * Since the device stats are not reset at PFReset, they likely will not
 * be zeroed when the driver starts.  We'll save the first values read
 * and use them as offsets to be subtracted from the raw values in order
 * to report stats that count from zero.  In the process, we also manage
 * the potential roll-over.
 **/
static void i40e_stat_update48(struct i40e_hw *hw, u32 hireg, u32 loreg,
			       bool offset_loaded, u64 *offset, u64 *stat)
{
	u64 new_data;

	if (hw->device_id == I40E_DEV_ID_QEMU) {
		new_data = rd32(hw, loreg);
		new_data |= ((u64)(rd32(hw, hireg) & 0xFFFF)) << 32;
	} else {
		new_data = rd64(hw, loreg);
	}
	if (!offset_loaded)
		*offset = new_data;
	if (likely(new_data >= *offset))
		*stat = new_data - *offset;
	else
		*stat = (new_data + BIT_ULL(48)) - *offset;
	*stat &= 0xFFFFFFFFFFFFULL;
}

/**
 * i40e_stat_update32 - read and update a 32 bit stat from the chip
 * @hw: ptr to the hardware info
 * @reg: the hw reg to read
 * @offset_loaded: has the initial offset been loaded yet
 * @offset: ptr to current offset value
 * @stat: ptr to the stat
 **/
static void i40e_stat_update32(struct i40e_hw *hw, u32 reg,
			       bool offset_loaded, u64 *offset, u64 *stat)
{
	u32 new_data;

	new_data = rd32(hw, reg);
	if (!offset_loaded)
		*offset = new_data;
	if (likely(new_data >= *offset))
		*stat = (u32)(new_data - *offset);
	else
		*stat = (u32)((new_data + BIT_ULL(32)) - *offset);
}

/**
 * i40e_stat_update_and_clear32 - read and clear hw reg, update a 32 bit stat
 * @hw: ptr to the hardware info
 * @reg: the hw reg to read and clear
 * @stat: ptr to the stat
 **/
static void i40e_stat_update_and_clear32(struct i40e_hw *hw, u32 reg, u64 *stat)
{
	u32 new_data = rd32(hw, reg);

	wr32(hw, reg, 1); /* must write a nonzero value to clear register */
	*stat += new_data;
}

/**
 * i40e_stats_update_rx_discards - update rx_discards.
 * @vsi: ptr to the VSI to be updated.
 * @hw: ptr to the hardware info.
 * @stat_idx: VSI's stat_counter_idx.
 * @offset_loaded: ptr to the VSI's stat_offsets_loaded.
 * @stat_offset: ptr to stat_offset to store first read of specific register.
 * @stat: ptr to VSI's stat to be updated.
 **/
static void
i40e_stats_update_rx_discards(struct i40e_vsi *vsi, struct i40e_hw *hw,
			      int stat_idx, bool offset_loaded,
			      struct i40e_eth_stats *stat_offset,
			      struct i40e_eth_stats *stat)
{
	u64 rx_rdpc, rx_rxerr;

	i40e_stat_update32(hw, I40E_GLV_RDPC(stat_idx), offset_loaded,
			   &stat_offset->rx_discards, &rx_rdpc);
	i40e_stat_update64(hw,
			   I40E_GL_RXERR1H(i40e_compute_pci_to_hw_id(vsi, hw)),
			   I40E_GL_RXERR1L(i40e_compute_pci_to_hw_id(vsi, hw)),
			   offset_loaded, &stat_offset->rx_discards_other,
			   &rx_rxerr);

	stat->rx_discards = rx_rdpc + rx_rxerr;
}

/**
 * i40e_update_eth_stats - Update VSI-specific ethernet statistics counters.
 * @vsi: the VSI to be updated
 **/
void i40e_update_eth_stats(struct i40e_vsi *vsi)
{
	int stat_idx = le16_to_cpu(vsi->info.stat_counter_idx);
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	struct i40e_eth_stats *oes;
	struct i40e_eth_stats *es;     /* device's eth stats */

	es = &vsi->eth_stats;
	oes = &vsi->eth_stats_offsets;

	/* Gather up the stats that the hw collects */
	i40e_stat_update32(hw, I40E_GLV_TEPC(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->tx_errors, &es->tx_errors);
	i40e_stat_update32(hw, I40E_GLV_RDPC(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->rx_discards, &es->rx_discards);
	i40e_stat_update32(hw, I40E_GLV_RUPP(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->rx_unknown_protocol, &es->rx_unknown_protocol);

	i40e_stat_update48(hw, I40E_GLV_GORCH(stat_idx),
			   I40E_GLV_GORCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->rx_bytes, &es->rx_bytes);
	i40e_stat_update48(hw, I40E_GLV_UPRCH(stat_idx),
			   I40E_GLV_UPRCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->rx_unicast, &es->rx_unicast);
	i40e_stat_update48(hw, I40E_GLV_MPRCH(stat_idx),
			   I40E_GLV_MPRCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->rx_multicast, &es->rx_multicast);
	i40e_stat_update48(hw, I40E_GLV_BPRCH(stat_idx),
			   I40E_GLV_BPRCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->rx_broadcast, &es->rx_broadcast);

	i40e_stat_update48(hw, I40E_GLV_GOTCH(stat_idx),
			   I40E_GLV_GOTCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->tx_bytes, &es->tx_bytes);
	i40e_stat_update48(hw, I40E_GLV_UPTCH(stat_idx),
			   I40E_GLV_UPTCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->tx_unicast, &es->tx_unicast);
	i40e_stat_update48(hw, I40E_GLV_MPTCH(stat_idx),
			   I40E_GLV_MPTCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->tx_multicast, &es->tx_multicast);
	i40e_stat_update48(hw, I40E_GLV_BPTCH(stat_idx),
			   I40E_GLV_BPTCL(stat_idx),
			   vsi->stat_offsets_loaded,
			   &oes->tx_broadcast, &es->tx_broadcast);

	i40e_stats_update_rx_discards(vsi, hw, stat_idx,
				      vsi->stat_offsets_loaded, oes, es);

	vsi->stat_offsets_loaded = true;
}

/**
 * i40e_update_veb_stats - Update Switch component statistics
 * @veb: the VEB being updated
 **/
void i40e_update_veb_stats(struct i40e_veb *veb)
{
	struct i40e_pf *pf = veb->pf;
	struct i40e_hw *hw = &pf->hw;
	struct i40e_eth_stats *oes;
	struct i40e_eth_stats *es;     /* device's eth stats */
	struct i40e_veb_tc_stats *veb_oes;
	struct i40e_veb_tc_stats *veb_es;
	int i, idx = 0;

	idx = veb->stats_idx;
	es = &veb->stats;
	oes = &veb->stats_offsets;
	veb_es = &veb->tc_stats;
	veb_oes = &veb->tc_stats_offsets;

	/* Gather up the stats that the hw collects */
	i40e_stat_update32(hw, I40E_GLSW_TDPC(idx),
			   veb->stat_offsets_loaded,
			   &oes->tx_discards, &es->tx_discards);
	if (hw->revision_id > 0)
		i40e_stat_update32(hw, I40E_GLSW_RUPP(idx),
				   veb->stat_offsets_loaded,
				   &oes->rx_unknown_protocol,
				   &es->rx_unknown_protocol);
	i40e_stat_update48(hw, I40E_GLSW_GORCH(idx), I40E_GLSW_GORCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->rx_bytes, &es->rx_bytes);
	i40e_stat_update48(hw, I40E_GLSW_UPRCH(idx), I40E_GLSW_UPRCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->rx_unicast, &es->rx_unicast);
	i40e_stat_update48(hw, I40E_GLSW_MPRCH(idx), I40E_GLSW_MPRCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->rx_multicast, &es->rx_multicast);
	i40e_stat_update48(hw, I40E_GLSW_BPRCH(idx), I40E_GLSW_BPRCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->rx_broadcast, &es->rx_broadcast);

	i40e_stat_update48(hw, I40E_GLSW_GOTCH(idx), I40E_GLSW_GOTCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->tx_bytes, &es->tx_bytes);
	i40e_stat_update48(hw, I40E_GLSW_UPTCH(idx), I40E_GLSW_UPTCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->tx_unicast, &es->tx_unicast);
	i40e_stat_update48(hw, I40E_GLSW_MPTCH(idx), I40E_GLSW_MPTCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->tx_multicast, &es->tx_multicast);
	i40e_stat_update48(hw, I40E_GLSW_BPTCH(idx), I40E_GLSW_BPTCL(idx),
			   veb->stat_offsets_loaded,
			   &oes->tx_broadcast, &es->tx_broadcast);
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		i40e_stat_update48(hw, I40E_GLVEBTC_RPCH(i, idx),
				   I40E_GLVEBTC_RPCL(i, idx),
				   veb->stat_offsets_loaded,
				   &veb_oes->tc_rx_packets[i],
				   &veb_es->tc_rx_packets[i]);
		i40e_stat_update48(hw, I40E_GLVEBTC_RBCH(i, idx),
				   I40E_GLVEBTC_RBCL(i, idx),
				   veb->stat_offsets_loaded,
				   &veb_oes->tc_rx_bytes[i],
				   &veb_es->tc_rx_bytes[i]);
		i40e_stat_update48(hw, I40E_GLVEBTC_TPCH(i, idx),
				   I40E_GLVEBTC_TPCL(i, idx),
				   veb->stat_offsets_loaded,
				   &veb_oes->tc_tx_packets[i],
				   &veb_es->tc_tx_packets[i]);
		i40e_stat_update48(hw, I40E_GLVEBTC_TBCH(i, idx),
				   I40E_GLVEBTC_TBCL(i, idx),
				   veb->stat_offsets_loaded,
				   &veb_oes->tc_tx_bytes[i],
				   &veb_es->tc_tx_bytes[i]);
	}
	veb->stat_offsets_loaded = true;
}

/**
 * i40e_update_vsi_stats - Update the vsi statistics counters.
 * @vsi: the VSI to be updated
 *
 * There are a few instances where we store the same stat in a
 * couple of different structs.  This is partly because we have
 * the netdev stats that need to be filled out, which is slightly
 * different from the "eth_stats" defined by the chip and used in
 * VF communications.  We sort it out here.
 **/
static void i40e_update_vsi_stats(struct i40e_vsi *vsi)
{
	u64 rx_page, rx_buf, rx_reuse, rx_alloc, rx_waive, rx_busy;
	struct i40e_pf *pf = vsi->back;
	struct rtnl_link_stats64 *ons;
	struct rtnl_link_stats64 *ns;   /* netdev stats */
	struct i40e_eth_stats *oes;
	struct i40e_eth_stats *es;     /* device's eth stats */
	u64 tx_restart, tx_busy;
	struct i40e_ring *p;
	u64 bytes, packets;
	unsigned int start;
	u64 tx_linearize;
	u64 tx_force_wb;
	u64 tx_stopped;
	u64 rx_p, rx_b;
	u64 tx_p, tx_b;
	u16 q;

	if (test_bit(__I40E_VSI_DOWN, vsi->state) ||
	    test_bit(__I40E_CONFIG_BUSY, pf->state))
		return;

	ns = i40e_get_vsi_stats_struct(vsi);
	ons = &vsi->net_stats_offsets;
	es = &vsi->eth_stats;
	oes = &vsi->eth_stats_offsets;

	/* Gather up the netdev and vsi stats that the driver collects
	 * on the fly during packet processing
	 */
	rx_b = rx_p = 0;
	tx_b = tx_p = 0;
	tx_restart = tx_busy = tx_linearize = tx_force_wb = 0;
	tx_stopped = 0;
	rx_page = 0;
	rx_buf = 0;
	rx_reuse = 0;
	rx_alloc = 0;
	rx_waive = 0;
	rx_busy = 0;
	rcu_read_lock();
	for (q = 0; q < vsi->num_queue_pairs; q++) {
		/* locate Tx ring */
		p = READ_ONCE(vsi->tx_rings[q]);
		if (!p)
			continue;

		do {
			start = u64_stats_fetch_begin(&p->syncp);
			packets = p->stats.packets;
			bytes = p->stats.bytes;
		} while (u64_stats_fetch_retry(&p->syncp, start));
		tx_b += bytes;
		tx_p += packets;
		tx_restart += p->tx_stats.restart_queue;
		tx_busy += p->tx_stats.tx_busy;
		tx_linearize += p->tx_stats.tx_linearize;
		tx_force_wb += p->tx_stats.tx_force_wb;
		tx_stopped += p->tx_stats.tx_stopped;

		/* locate Rx ring */
		p = READ_ONCE(vsi->rx_rings[q]);
		if (!p)
			continue;

		do {
			start = u64_stats_fetch_begin(&p->syncp);
			packets = p->stats.packets;
			bytes = p->stats.bytes;
		} while (u64_stats_fetch_retry(&p->syncp, start));
		rx_b += bytes;
		rx_p += packets;
		rx_buf += p->rx_stats.alloc_buff_failed;
		rx_page += p->rx_stats.alloc_page_failed;
		rx_reuse += p->rx_stats.page_reuse_count;
		rx_alloc += p->rx_stats.page_alloc_count;
		rx_waive += p->rx_stats.page_waive_count;
		rx_busy += p->rx_stats.page_busy_count;

		if (i40e_enabled_xdp_vsi(vsi)) {
			/* locate XDP ring */
			p = READ_ONCE(vsi->xdp_rings[q]);
			if (!p)
				continue;

			do {
				start = u64_stats_fetch_begin(&p->syncp);
				packets = p->stats.packets;
				bytes = p->stats.bytes;
			} while (u64_stats_fetch_retry(&p->syncp, start));
			tx_b += bytes;
			tx_p += packets;
			tx_restart += p->tx_stats.restart_queue;
			tx_busy += p->tx_stats.tx_busy;
			tx_linearize += p->tx_stats.tx_linearize;
			tx_force_wb += p->tx_stats.tx_force_wb;
		}
	}
	rcu_read_unlock();
	vsi->tx_restart = tx_restart;
	vsi->tx_busy = tx_busy;
	vsi->tx_linearize = tx_linearize;
	vsi->tx_force_wb = tx_force_wb;
	vsi->tx_stopped = tx_stopped;
	vsi->rx_page_failed = rx_page;
	vsi->rx_buf_failed = rx_buf;
	vsi->rx_page_reuse = rx_reuse;
	vsi->rx_page_alloc = rx_alloc;
	vsi->rx_page_waive = rx_waive;
	vsi->rx_page_busy = rx_busy;

	ns->rx_packets = rx_p;
	ns->rx_bytes = rx_b;
	ns->tx_packets = tx_p;
	ns->tx_bytes = tx_b;

	/* update netdev stats from eth stats */
	i40e_update_eth_stats(vsi);
	ons->tx_errors = oes->tx_errors;
	ns->tx_errors = es->tx_errors;
	ons->multicast = oes->rx_multicast;
	ns->multicast = es->rx_multicast;
	ons->rx_dropped = oes->rx_discards;
	ns->rx_dropped = es->rx_discards;
	ons->tx_dropped = oes->tx_discards;
	ns->tx_dropped = es->tx_discards;

	/* pull in a couple PF stats if this is the main vsi */
	if (vsi == pf->vsi[pf->lan_vsi]) {
		ns->rx_crc_errors = pf->stats.crc_errors;
		ns->rx_errors = pf->stats.crc_errors + pf->stats.illegal_bytes;
		ns->rx_length_errors = pf->stats.rx_length_errors;
	}
}

/**
 * i40e_update_pf_stats - Update the PF statistics counters.
 * @pf: the PF to be updated
 **/
static void i40e_update_pf_stats(struct i40e_pf *pf)
{
	struct i40e_hw_port_stats *osd = &pf->stats_offsets;
	struct i40e_hw_port_stats *nsd = &pf->stats;
	struct i40e_hw *hw = &pf->hw;
	u32 val;
	int i;

	i40e_stat_update48(hw, I40E_GLPRT_GORCH(hw->port),
			   I40E_GLPRT_GORCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.rx_bytes, &nsd->eth.rx_bytes);
	i40e_stat_update48(hw, I40E_GLPRT_GOTCH(hw->port),
			   I40E_GLPRT_GOTCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.tx_bytes, &nsd->eth.tx_bytes);
	i40e_stat_update32(hw, I40E_GLPRT_RDPC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.rx_discards,
			   &nsd->eth.rx_discards);
	i40e_stat_update48(hw, I40E_GLPRT_UPRCH(hw->port),
			   I40E_GLPRT_UPRCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.rx_unicast,
			   &nsd->eth.rx_unicast);
	i40e_stat_update48(hw, I40E_GLPRT_MPRCH(hw->port),
			   I40E_GLPRT_MPRCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.rx_multicast,
			   &nsd->eth.rx_multicast);
	i40e_stat_update48(hw, I40E_GLPRT_BPRCH(hw->port),
			   I40E_GLPRT_BPRCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.rx_broadcast,
			   &nsd->eth.rx_broadcast);
	i40e_stat_update48(hw, I40E_GLPRT_UPTCH(hw->port),
			   I40E_GLPRT_UPTCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.tx_unicast,
			   &nsd->eth.tx_unicast);
	i40e_stat_update48(hw, I40E_GLPRT_MPTCH(hw->port),
			   I40E_GLPRT_MPTCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.tx_multicast,
			   &nsd->eth.tx_multicast);
	i40e_stat_update48(hw, I40E_GLPRT_BPTCH(hw->port),
			   I40E_GLPRT_BPTCL(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->eth.tx_broadcast,
			   &nsd->eth.tx_broadcast);

	i40e_stat_update32(hw, I40E_GLPRT_TDOLD(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_dropped_link_down,
			   &nsd->tx_dropped_link_down);

	i40e_stat_update32(hw, I40E_GLPRT_CRCERRS(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->crc_errors, &nsd->crc_errors);

	i40e_stat_update32(hw, I40E_GLPRT_ILLERRC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->illegal_bytes, &nsd->illegal_bytes);

	i40e_stat_update32(hw, I40E_GLPRT_MLFC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->mac_local_faults,
			   &nsd->mac_local_faults);
	i40e_stat_update32(hw, I40E_GLPRT_MRFC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->mac_remote_faults,
			   &nsd->mac_remote_faults);

	i40e_stat_update32(hw, I40E_GLPRT_RLEC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_length_errors,
			   &nsd->rx_length_errors);

	i40e_stat_update32(hw, I40E_GLPRT_LXONRXC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->link_xon_rx, &nsd->link_xon_rx);
	i40e_stat_update32(hw, I40E_GLPRT_LXONTXC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->link_xon_tx, &nsd->link_xon_tx);
	i40e_stat_update32(hw, I40E_GLPRT_LXOFFRXC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->link_xoff_rx, &nsd->link_xoff_rx);
	i40e_stat_update32(hw, I40E_GLPRT_LXOFFTXC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->link_xoff_tx, &nsd->link_xoff_tx);

	for (i = 0; i < 8; i++) {
		i40e_stat_update32(hw, I40E_GLPRT_PXOFFRXC(hw->port, i),
				   pf->stat_offsets_loaded,
				   &osd->priority_xoff_rx[i],
				   &nsd->priority_xoff_rx[i]);
		i40e_stat_update32(hw, I40E_GLPRT_PXONRXC(hw->port, i),
				   pf->stat_offsets_loaded,
				   &osd->priority_xon_rx[i],
				   &nsd->priority_xon_rx[i]);
		i40e_stat_update32(hw, I40E_GLPRT_PXONTXC(hw->port, i),
				   pf->stat_offsets_loaded,
				   &osd->priority_xon_tx[i],
				   &nsd->priority_xon_tx[i]);
		i40e_stat_update32(hw, I40E_GLPRT_PXOFFTXC(hw->port, i),
				   pf->stat_offsets_loaded,
				   &osd->priority_xoff_tx[i],
				   &nsd->priority_xoff_tx[i]);
		i40e_stat_update32(hw,
				   I40E_GLPRT_RXON2OFFCNT(hw->port, i),
				   pf->stat_offsets_loaded,
				   &osd->priority_xon_2_xoff[i],
				   &nsd->priority_xon_2_xoff[i]);
	}

	i40e_stat_update48(hw, I40E_GLPRT_PRC64H(hw->port),
			   I40E_GLPRT_PRC64L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_64, &nsd->rx_size_64);
	i40e_stat_update48(hw, I40E_GLPRT_PRC127H(hw->port),
			   I40E_GLPRT_PRC127L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_127, &nsd->rx_size_127);
	i40e_stat_update48(hw, I40E_GLPRT_PRC255H(hw->port),
			   I40E_GLPRT_PRC255L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_255, &nsd->rx_size_255);
	i40e_stat_update48(hw, I40E_GLPRT_PRC511H(hw->port),
			   I40E_GLPRT_PRC511L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_511, &nsd->rx_size_511);
	i40e_stat_update48(hw, I40E_GLPRT_PRC1023H(hw->port),
			   I40E_GLPRT_PRC1023L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_1023, &nsd->rx_size_1023);
	i40e_stat_update48(hw, I40E_GLPRT_PRC1522H(hw->port),
			   I40E_GLPRT_PRC1522L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_1522, &nsd->rx_size_1522);
	i40e_stat_update48(hw, I40E_GLPRT_PRC9522H(hw->port),
			   I40E_GLPRT_PRC9522L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_size_big, &nsd->rx_size_big);

	i40e_stat_update48(hw, I40E_GLPRT_PTC64H(hw->port),
			   I40E_GLPRT_PTC64L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_64, &nsd->tx_size_64);
	i40e_stat_update48(hw, I40E_GLPRT_PTC127H(hw->port),
			   I40E_GLPRT_PTC127L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_127, &nsd->tx_size_127);
	i40e_stat_update48(hw, I40E_GLPRT_PTC255H(hw->port),
			   I40E_GLPRT_PTC255L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_255, &nsd->tx_size_255);
	i40e_stat_update48(hw, I40E_GLPRT_PTC511H(hw->port),
			   I40E_GLPRT_PTC511L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_511, &nsd->tx_size_511);
	i40e_stat_update48(hw, I40E_GLPRT_PTC1023H(hw->port),
			   I40E_GLPRT_PTC1023L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_1023, &nsd->tx_size_1023);
	i40e_stat_update48(hw, I40E_GLPRT_PTC1522H(hw->port),
			   I40E_GLPRT_PTC1522L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_1522, &nsd->tx_size_1522);
	i40e_stat_update48(hw, I40E_GLPRT_PTC9522H(hw->port),
			   I40E_GLPRT_PTC9522L(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->tx_size_big, &nsd->tx_size_big);

	i40e_stat_update32(hw, I40E_GLPRT_RUC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_undersize, &nsd->rx_undersize);
	i40e_stat_update32(hw, I40E_GLPRT_RFC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_fragments, &nsd->rx_fragments);
	i40e_stat_update32(hw, I40E_GLPRT_ROC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_oversize, &nsd->rx_oversize);
	i40e_stat_update32(hw, I40E_GLPRT_RJC(hw->port),
			   pf->stat_offsets_loaded,
			   &osd->rx_jabber, &nsd->rx_jabber);

	/* FDIR stats */
	i40e_stat_update_and_clear32(hw,
			I40E_GLQF_PCNT(I40E_FD_ATR_STAT_IDX(hw->pf_id)),
			&nsd->fd_atr_match);
	i40e_stat_update_and_clear32(hw,
			I40E_GLQF_PCNT(I40E_FD_SB_STAT_IDX(hw->pf_id)),
			&nsd->fd_sb_match);
	i40e_stat_update_and_clear32(hw,
			I40E_GLQF_PCNT(I40E_FD_ATR_TUNNEL_STAT_IDX(hw->pf_id)),
			&nsd->fd_atr_tunnel_match);

	val = rd32(hw, I40E_PRTPM_EEE_STAT);
	nsd->tx_lpi_status =
		       (val & I40E_PRTPM_EEE_STAT_TX_LPI_STATUS_MASK) >>
			I40E_PRTPM_EEE_STAT_TX_LPI_STATUS_SHIFT;
	nsd->rx_lpi_status =
		       (val & I40E_PRTPM_EEE_STAT_RX_LPI_STATUS_MASK) >>
			I40E_PRTPM_EEE_STAT_RX_LPI_STATUS_SHIFT;
	i40e_stat_update32(hw, I40E_PRTPM_TLPIC,
			   pf->stat_offsets_loaded,
			   &osd->tx_lpi_count, &nsd->tx_lpi_count);
	i40e_stat_update32(hw, I40E_PRTPM_RLPIC,
			   pf->stat_offsets_loaded,
			   &osd->rx_lpi_count, &nsd->rx_lpi_count);

	if (pf->flags & I40E_FLAG_FD_SB_ENABLED &&
	    !test_bit(__I40E_FD_SB_AUTO_DISABLED, pf->state))
		nsd->fd_sb_status = true;
	else
		nsd->fd_sb_status = false;

	if (pf->flags & I40E_FLAG_FD_ATR_ENABLED &&
	    !test_bit(__I40E_FD_ATR_AUTO_DISABLED, pf->state))
		nsd->fd_atr_status = true;
	else
		nsd->fd_atr_status = false;

	pf->stat_offsets_loaded = true;
}

/**
 * i40e_update_stats - Update the various statistics counters.
 * @vsi: the VSI to be updated
 *
 * Update the various stats for this VSI and its related entities.
 **/
void i40e_update_stats(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;

	if (vsi == pf->vsi[pf->lan_vsi])
		i40e_update_pf_stats(pf);

	i40e_update_vsi_stats(vsi);
}

/**
 * i40e_count_filters - counts VSI mac filters
 * @vsi: the VSI to be searched
 *
 * Returns count of mac filters
 **/
int i40e_count_filters(struct i40e_vsi *vsi)
{
	struct i40e_mac_filter *f;
	struct hlist_node *h;
	int bkt;
	int cnt = 0;

	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist)
		++cnt;

	return cnt;
}

/**
 * i40e_find_filter - Search VSI filter list for specific mac/vlan filter
 * @vsi: the VSI to be searched
 * @macaddr: the MAC address
 * @vlan: the vlan
 *
 * Returns ptr to the filter object or NULL
 **/
static struct i40e_mac_filter *i40e_find_filter(struct i40e_vsi *vsi,
						const u8 *macaddr, s16 vlan)
{
	struct i40e_mac_filter *f;
	u64 key;

	if (!vsi || !macaddr)
		return NULL;

	key = i40e_addr_to_hkey(macaddr);
	hash_for_each_possible(vsi->mac_filter_hash, f, hlist, key) {
		if ((ether_addr_equal(macaddr, f->macaddr)) &&
		    (vlan == f->vlan))
			return f;
	}
	return NULL;
}

/**
 * i40e_find_mac - Find a mac addr in the macvlan filters list
 * @vsi: the VSI to be searched
 * @macaddr: the MAC address we are searching for
 *
 * Returns the first filter with the provided MAC address or NULL if
 * MAC address was not found
 **/
struct i40e_mac_filter *i40e_find_mac(struct i40e_vsi *vsi, const u8 *macaddr)
{
	struct i40e_mac_filter *f;
	u64 key;

	if (!vsi || !macaddr)
		return NULL;

	key = i40e_addr_to_hkey(macaddr);
	hash_for_each_possible(vsi->mac_filter_hash, f, hlist, key) {
		if ((ether_addr_equal(macaddr, f->macaddr)))
			return f;
	}
	return NULL;
}

/**
 * i40e_is_vsi_in_vlan - Check if VSI is in vlan mode
 * @vsi: the VSI to be searched
 *
 * Returns true if VSI is in vlan mode or false otherwise
 **/
bool i40e_is_vsi_in_vlan(struct i40e_vsi *vsi)
{
	/* If we have a PVID, always operate in VLAN mode */
	if (vsi->info.pvid)
		return true;

	/* We need to operate in VLAN mode whenever we have any filters with
	 * a VLAN other than I40E_VLAN_ALL. We could check the table each
	 * time, incurring search cost repeatedly. However, we can notice two
	 * things:
	 *
	 * 1) the only place where we can gain a VLAN filter is in
	 *    i40e_add_filter.
	 *
	 * 2) the only place where filters are actually removed is in
	 *    i40e_sync_filters_subtask.
	 *
	 * Thus, we can simply use a boolean value, has_vlan_filters which we
	 * will set to true when we add a VLAN filter in i40e_add_filter. Then
	 * we have to perform the full search after deleting filters in
	 * i40e_sync_filters_subtask, but we already have to search
	 * filters here and can perform the check at the same time. This
	 * results in avoiding embedding a loop for VLAN mode inside another
	 * loop over all the filters, and should maintain correctness as noted
	 * above.
	 */
	return vsi->has_vlan_filter;
}

/**
 * i40e_correct_mac_vlan_filters - Correct non-VLAN filters if necessary
 * @vsi: the VSI to configure
 * @tmp_add_list: list of filters ready to be added
 * @tmp_del_list: list of filters ready to be deleted
 * @vlan_filters: the number of active VLAN filters
 *
 * Update VLAN=0 and VLAN=-1 (I40E_VLAN_ANY) filters properly so that they
 * behave as expected. If we have any active VLAN filters remaining or about
 * to be added then we need to update non-VLAN filters to be marked as VLAN=0
 * so that they only match against untagged traffic. If we no longer have any
 * active VLAN filters, we need to make all non-VLAN filters marked as VLAN=-1
 * so that they match against both tagged and untagged traffic. In this way,
 * we ensure that we correctly receive the desired traffic. This ensures that
 * when we have an active VLAN we will receive only untagged traffic and
 * traffic matching active VLANs. If we have no active VLANs then we will
 * operate in non-VLAN mode and receive all traffic, tagged or untagged.
 *
 * Finally, in a similar fashion, this function also corrects filters when
 * there is an active PVID assigned to this VSI.
 *
 * In case of memory allocation failure return -ENOMEM. Otherwise, return 0.
 *
 * This function is only expected to be called from within
 * i40e_sync_vsi_filters.
 *
 * NOTE: This function expects to be called while under the
 * mac_filter_hash_lock
 */
static int i40e_correct_mac_vlan_filters(struct i40e_vsi *vsi,
					 struct hlist_head *tmp_add_list,
					 struct hlist_head *tmp_del_list,
					 int vlan_filters)
{
	s16 pvid = le16_to_cpu(vsi->info.pvid);
	struct i40e_mac_filter *f, *add_head;
	struct i40e_new_mac_filter *new;
	struct hlist_node *h;
	int bkt, new_vlan;

	/* To determine if a particular filter needs to be replaced we
	 * have the three following conditions:
	 *
	 * a) if we have a PVID assigned, then all filters which are
	 *    not marked as VLAN=PVID must be replaced with filters that
	 *    are.
	 * b) otherwise, if we have any active VLANS, all filters
	 *    which are marked as VLAN=-1 must be replaced with
	 *    filters marked as VLAN=0
	 * c) finally, if we do not have any active VLANS, all filters
	 *    which are marked as VLAN=0 must be replaced with filters
	 *    marked as VLAN=-1
	 */

	/* Update the filters about to be added in place */
	hlist_for_each_entry(new, tmp_add_list, hlist) {
		if (pvid && new->f->vlan != pvid)
			new->f->vlan = pvid;
		else if (vlan_filters && new->f->vlan == I40E_VLAN_ANY)
			new->f->vlan = 0;
		else if (!vlan_filters && new->f->vlan == 0)
			new->f->vlan = I40E_VLAN_ANY;
	}

	/* Update the remaining active filters */
	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
		/* Combine the checks for whether a filter needs to be changed
		 * and then determine the new VLAN inside the if block, in
		 * order to avoid duplicating code for adding the new filter
		 * then deleting the old filter.
		 */
		if ((pvid && f->vlan != pvid) ||
		    (vlan_filters && f->vlan == I40E_VLAN_ANY) ||
		    (!vlan_filters && f->vlan == 0)) {
			/* Determine the new vlan we will be adding */
			if (pvid)
				new_vlan = pvid;
			else if (vlan_filters)
				new_vlan = 0;
			else
				new_vlan = I40E_VLAN_ANY;

			/* Create the new filter */
			add_head = i40e_add_filter(vsi, f->macaddr, new_vlan);
			if (!add_head)
				return -ENOMEM;

			/* Create a temporary i40e_new_mac_filter */
			new = kzalloc(sizeof(*new), GFP_ATOMIC);
			if (!new)
				return -ENOMEM;

			new->f = add_head;
			new->state = add_head->state;

			/* Add the new filter to the tmp list */
			hlist_add_head(&new->hlist, tmp_add_list);

			/* Put the original filter into the delete list */
			f->state = I40E_FILTER_REMOVE;
			hash_del(&f->hlist);
			hlist_add_head(&f->hlist, tmp_del_list);
		}
	}

	vsi->has_vlan_filter = !!vlan_filters;

	return 0;
}

/**
 * i40e_get_vf_new_vlan - Get new vlan id on a vf
 * @vsi: the vsi to configure
 * @new_mac: new mac filter to be added
 * @f: existing mac filter, replaced with new_mac->f if new_mac is not NULL
 * @vlan_filters: the number of active VLAN filters
 * @trusted: flag if the VF is trusted
 *
 * Get new VLAN id based on current VLAN filters, trust, PVID
 * and vf-vlan-prune-disable flag.
 *
 * Returns the value of the new vlan filter or
 * the old value if no new filter is needed.
 */
static s16 i40e_get_vf_new_vlan(struct i40e_vsi *vsi,
				struct i40e_new_mac_filter *new_mac,
				struct i40e_mac_filter *f,
				int vlan_filters,
				bool trusted)
{
	s16 pvid = le16_to_cpu(vsi->info.pvid);
	struct i40e_pf *pf = vsi->back;
	bool is_any;

	if (new_mac)
		f = new_mac->f;

	if (pvid && f->vlan != pvid)
		return pvid;

	is_any = (trusted ||
		  !(pf->flags & I40E_FLAG_VF_VLAN_PRUNING));

	if ((vlan_filters && f->vlan == I40E_VLAN_ANY) ||
	    (!is_any && !vlan_filters && f->vlan == I40E_VLAN_ANY) ||
	    (is_any && !vlan_filters && f->vlan == 0)) {
		if (is_any)
			return I40E_VLAN_ANY;
		else
			return 0;
	}

	return f->vlan;
}

/**
 * i40e_correct_vf_mac_vlan_filters - Correct non-VLAN VF filters if necessary
 * @vsi: the vsi to configure
 * @tmp_add_list: list of filters ready to be added
 * @tmp_del_list: list of filters ready to be deleted
 * @vlan_filters: the number of active VLAN filters
 * @trusted: flag if the VF is trusted
 *
 * Correct VF VLAN filters based on current VLAN filters, trust, PVID
 * and vf-vlan-prune-disable flag.
 *
 * In case of memory allocation failure return -ENOMEM. Otherwise, return 0.
 *
 * This function is only expected to be called from within
 * i40e_sync_vsi_filters.
 *
 * NOTE: This function expects to be called while under the
 * mac_filter_hash_lock
 */
static int i40e_correct_vf_mac_vlan_filters(struct i40e_vsi *vsi,
					    struct hlist_head *tmp_add_list,
					    struct hlist_head *tmp_del_list,
					    int vlan_filters,
					    bool trusted)
{
	struct i40e_mac_filter *f, *add_head;
	struct i40e_new_mac_filter *new_mac;
	struct hlist_node *h;
	int bkt, new_vlan;

	hlist_for_each_entry(new_mac, tmp_add_list, hlist) {
		new_mac->f->vlan = i40e_get_vf_new_vlan(vsi, new_mac, NULL,
							vlan_filters, trusted);
	}

	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
		new_vlan = i40e_get_vf_new_vlan(vsi, NULL, f, vlan_filters,
						trusted);
		if (new_vlan != f->vlan) {
			add_head = i40e_add_filter(vsi, f->macaddr, new_vlan);
			if (!add_head)
				return -ENOMEM;
			/* Create a temporary i40e_new_mac_filter */
			new_mac = kzalloc(sizeof(*new_mac), GFP_ATOMIC);
			if (!new_mac)
				return -ENOMEM;
			new_mac->f = add_head;
			new_mac->state = add_head->state;

			/* Add the new filter to the tmp list */
			hlist_add_head(&new_mac->hlist, tmp_add_list);

			/* Put the original filter into the delete list */
			f->state = I40E_FILTER_REMOVE;
			hash_del(&f->hlist);
			hlist_add_head(&f->hlist, tmp_del_list);
		}
	}

	vsi->has_vlan_filter = !!vlan_filters;
	return 0;
}

/**
 * i40e_rm_default_mac_filter - Remove the default MAC filter set by NVM
 * @vsi: the PF Main VSI - inappropriate for any other VSI
 * @macaddr: the MAC address
 *
 * Remove whatever filter the firmware set up so the driver can manage
 * its own filtering intelligently.
 **/
static void i40e_rm_default_mac_filter(struct i40e_vsi *vsi, u8 *macaddr)
{
	struct i40e_aqc_remove_macvlan_element_data element;
	struct i40e_pf *pf = vsi->back;

	/* Only appropriate for the PF main VSI */
	if (vsi->type != I40E_VSI_MAIN)
		return;

	memset(&element, 0, sizeof(element));
	ether_addr_copy(element.mac_addr, macaddr);
	element.vlan_tag = 0;
	/* Ignore error returns, some firmware does it this way... */
	element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH;
	i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL);

	memset(&element, 0, sizeof(element));
	ether_addr_copy(element.mac_addr, macaddr);
	element.vlan_tag = 0;
	/* ...and some firmware does it this way. */
	element.flags = I40E_AQC_MACVLAN_DEL_PERFECT_MATCH |
			I40E_AQC_MACVLAN_DEL_IGNORE_VLAN;
	i40e_aq_remove_macvlan(&pf->hw, vsi->seid, &element, 1, NULL);
}

/**
 * i40e_add_filter - Add a mac/vlan filter to the VSI
 * @vsi: the VSI to be searched
 * @macaddr: the MAC address
 * @vlan: the vlan
 *
 * Returns ptr to the filter object or NULL when no memory available.
 *
 * NOTE: This function is expected to be called with mac_filter_hash_lock
 * being held.
 **/
struct i40e_mac_filter *i40e_add_filter(struct i40e_vsi *vsi,
					const u8 *macaddr, s16 vlan)
{
	struct i40e_mac_filter *f;
	u64 key;

	if (!vsi || !macaddr)
		return NULL;

	f = i40e_find_filter(vsi, macaddr, vlan);
	if (!f) {
		f = kzalloc(sizeof(*f), GFP_ATOMIC);
		if (!f)
			return NULL;

		/* Update the boolean indicating if we need to function in
		 * VLAN mode.
		 */
		if (vlan >= 0)
			vsi->has_vlan_filter = true;

		ether_addr_copy(f->macaddr, macaddr);
		f->vlan = vlan;
		f->state = I40E_FILTER_NEW;
		INIT_HLIST_NODE(&f->hlist);

		key = i40e_addr_to_hkey(macaddr);
		hash_add(vsi->mac_filter_hash, &f->hlist, key);

		vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
		set_bit(__I40E_MACVLAN_SYNC_PENDING, vsi->back->state);
	}

	/* If we're asked to add a filter that has been marked for removal, it
	 * is safe to simply restore it to active state. __i40e_del_filter
	 * will have simply deleted any filters which were previously marked
	 * NEW or FAILED, so if it is currently marked REMOVE it must have
	 * previously been ACTIVE. Since we haven't yet run the sync filters
	 * task, just restore this filter to the ACTIVE state so that the
	 * sync task leaves it in place
	 */
	if (f->state == I40E_FILTER_REMOVE)
		f->state = I40E_FILTER_ACTIVE;

	return f;
}

/**
 * __i40e_del_filter - Remove a specific filter from the VSI
 * @vsi: VSI to remove from
 * @f: the filter to remove from the list
 *
 * This function should be called instead of i40e_del_filter only if you know
 * the exact filter you will remove already, such as via i40e_find_filter or
 * i40e_find_mac.
 *
 * NOTE: This function is expected to be called with mac_filter_hash_lock
 * being held.
 * ANOTHER NOTE: This function MUST be called from within the context of
 * the "safe" variants of any list iterators, e.g. list_for_each_entry_safe()
 * instead of list_for_each_entry().
 **/
void __i40e_del_filter(struct i40e_vsi *vsi, struct i40e_mac_filter *f)
{
	if (!f)
		return;

	/* If the filter was never added to firmware then we can just delete it
	 * directly and we don't want to set the status to remove or else an
	 * admin queue command will unnecessarily fire.
	 */
	if ((f->state == I40E_FILTER_FAILED) ||
	    (f->state == I40E_FILTER_NEW)) {
		hash_del(&f->hlist);
		kfree(f);
	} else {
		f->state = I40E_FILTER_REMOVE;
	}

	vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
	set_bit(__I40E_MACVLAN_SYNC_PENDING, vsi->back->state);
}

/**
 * i40e_del_filter - Remove a MAC/VLAN filter from the VSI
 * @vsi: the VSI to be searched
 * @macaddr: the MAC address
 * @vlan: the VLAN
 *
 * NOTE: This function is expected to be called with mac_filter_hash_lock
 * being held.
 * ANOTHER NOTE: This function MUST be called from within the context of
 * the "safe" variants of any list iterators, e.g. list_for_each_entry_safe()
 * instead of list_for_each_entry().
 **/
void i40e_del_filter(struct i40e_vsi *vsi, const u8 *macaddr, s16 vlan)
{
	struct i40e_mac_filter *f;

	if (!vsi || !macaddr)
		return;

	f = i40e_find_filter(vsi, macaddr, vlan);
	__i40e_del_filter(vsi, f);
}

/**
 * i40e_add_mac_filter - Add a MAC filter for all active VLANs
 * @vsi: the VSI to be searched
 * @macaddr: the mac address to be filtered
 *
 * If we're not in VLAN mode, just add the filter to I40E_VLAN_ANY. Otherwise,
 * go through all the macvlan filters and add a macvlan filter for each
 * unique vlan that already exists. If a PVID has been assigned, instead only
 * add the macaddr to that VLAN.
 *
 * Returns last filter added on success, else NULL
 **/
struct i40e_mac_filter *i40e_add_mac_filter(struct i40e_vsi *vsi,
					    const u8 *macaddr)
{
	struct i40e_mac_filter *f, *add = NULL;
	struct hlist_node *h;
	int bkt;

	if (vsi->info.pvid)
		return i40e_add_filter(vsi, macaddr,
				       le16_to_cpu(vsi->info.pvid));

	if (!i40e_is_vsi_in_vlan(vsi))
		return i40e_add_filter(vsi, macaddr, I40E_VLAN_ANY);

	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
		if (f->state == I40E_FILTER_REMOVE)
			continue;
		add = i40e_add_filter(vsi, macaddr, f->vlan);
		if (!add)
			return NULL;
	}

	return add;
}

/**
 * i40e_del_mac_filter - Remove a MAC filter from all VLANs
 * @vsi: the VSI to be searched
 * @macaddr: the mac address to be removed
 *
 * Removes a given MAC address from a VSI regardless of what VLAN it has been
 * associated with.
 *
 * Returns 0 for success, or error
 **/
int i40e_del_mac_filter(struct i40e_vsi *vsi, const u8 *macaddr)
{
	struct i40e_mac_filter *f;
	struct hlist_node *h;
	bool found = false;
	int bkt;

	lockdep_assert_held(&vsi->mac_filter_hash_lock);
	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
		if (ether_addr_equal(macaddr, f->macaddr)) {
			__i40e_del_filter(vsi, f);
			found = true;
		}
	}

	if (found)
		return 0;
	else
		return -ENOENT;
}

/**
 * i40e_set_mac - NDO callback to set mac address
 * @netdev: network interface device structure
 * @p: pointer to an address structure
 *
 * Returns 0 on success, negative on failure
 **/
static int i40e_set_mac(struct net_device *netdev, void *p)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	struct sockaddr *addr = p;

	if (!is_valid_ether_addr(addr->sa_data))
		return -EADDRNOTAVAIL;

	if (ether_addr_equal(netdev->dev_addr, addr->sa_data)) {
		netdev_info(netdev, "already using mac address %pM\n",
			    addr->sa_data);
		return 0;
	}

	if (test_bit(__I40E_DOWN, pf->state) ||
	    test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state))
		return -EADDRNOTAVAIL;

	if (ether_addr_equal(hw->mac.addr, addr->sa_data))
		netdev_info(netdev, "returning to hw mac address %pM\n",
			    hw->mac.addr);
	else
		netdev_info(netdev, "set new mac address %pM\n", addr->sa_data);

	/* Copy the address first, so that we avoid a possible race with
	 * .set_rx_mode().
	 * - Remove old address from MAC filter
	 * - Copy new address
	 * - Add new address to MAC filter
	 */
	spin_lock_bh(&vsi->mac_filter_hash_lock);
	i40e_del_mac_filter(vsi, netdev->dev_addr);
	eth_hw_addr_set(netdev, addr->sa_data);
	i40e_add_mac_filter(vsi, netdev->dev_addr);
	spin_unlock_bh(&vsi->mac_filter_hash_lock);

	if (vsi->type == I40E_VSI_MAIN) {
		int ret;

		ret = i40e_aq_mac_address_write(hw, I40E_AQC_WRITE_TYPE_LAA_WOL,
						addr->sa_data, NULL);
		if (ret)
			netdev_info(netdev, "Ignoring error from firmware on LAA update, status %pe, AQ ret %s\n",
				    ERR_PTR(ret),
				    i40e_aq_str(hw, hw->aq.asq_last_status));
	}

	/* schedule our worker thread which will take care of
	 * applying the new filter changes
	 */
	i40e_service_event_schedule(pf);
	return 0;
}

/**
 * i40e_config_rss_aq - Prepare for RSS using AQ commands
 * @vsi: vsi structure
 * @seed: RSS hash seed
 * @lut: pointer to lookup table of lut_size
 * @lut_size: size of the lookup table
 **/
static int i40e_config_rss_aq(struct i40e_vsi *vsi, const u8 *seed,
			      u8 *lut, u16 lut_size)
{
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	int ret = 0;

	if (seed) {
		struct i40e_aqc_get_set_rss_key_data *seed_dw =
			(struct i40e_aqc_get_set_rss_key_data *)seed;
		ret = i40e_aq_set_rss_key(hw, vsi->id, seed_dw);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "Cannot set RSS key, err %pe aq_err %s\n",
				 ERR_PTR(ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
			return ret;
		}
	}
	if (lut) {
		bool pf_lut = vsi->type == I40E_VSI_MAIN;

		ret = i40e_aq_set_rss_lut(hw, vsi->id, pf_lut, lut, lut_size);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "Cannot set RSS lut, err %pe aq_err %s\n",
				 ERR_PTR(ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
			return ret;
		}
	}
	return ret;
}

/**
 * i40e_vsi_config_rss - Prepare for VSI(VMDq) RSS if used
 * @vsi: VSI structure
 **/
static int i40e_vsi_config_rss(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	u8 seed[I40E_HKEY_ARRAY_SIZE];
	u8 *lut;
	int ret;

	if (!(pf->hw_features & I40E_HW_RSS_AQ_CAPABLE))
		return 0;
	if (!vsi->rss_size)
		vsi->rss_size = min_t(int, pf->alloc_rss_size,
				      vsi->num_queue_pairs);
	if (!vsi->rss_size)
		return -EINVAL;
	lut = kzalloc(vsi->rss_table_size, GFP_KERNEL);
	if (!lut)
		return -ENOMEM;

	/* Use the user configured hash keys and lookup table if there is one,
	 * otherwise use default
	 */
	if (vsi->rss_lut_user)
		memcpy(lut, vsi->rss_lut_user, vsi->rss_table_size);
	else
		i40e_fill_rss_lut(pf, lut, vsi->rss_table_size, vsi->rss_size);
	if (vsi->rss_hkey_user)
		memcpy(seed, vsi->rss_hkey_user, I40E_HKEY_ARRAY_SIZE);
	else
		netdev_rss_key_fill((void *)seed, I40E_HKEY_ARRAY_SIZE);
	ret = i40e_config_rss_aq(vsi, seed, lut, vsi->rss_table_size);
	kfree(lut);
	return ret;
}

/**
 * i40e_vsi_setup_queue_map_mqprio - Prepares mqprio based tc_config
 * @vsi: the VSI being configured,
 * @ctxt: VSI context structure
 * @enabled_tc: number of traffic classes to enable
 *
 * Prepares VSI tc_config to have queue configurations based on MQPRIO options.
 **/
static int i40e_vsi_setup_queue_map_mqprio(struct i40e_vsi *vsi,
					   struct i40e_vsi_context *ctxt,
					   u8 enabled_tc)
{
	u16 qcount = 0, max_qcount, qmap, sections = 0;
	int i, override_q, pow, num_qps, ret;
	u8 netdev_tc = 0, offset = 0;

	if (vsi->type != I40E_VSI_MAIN)
		return -EINVAL;
	sections = I40E_AQ_VSI_PROP_QUEUE_MAP_VALID;
	sections |= I40E_AQ_VSI_PROP_SCHED_VALID;
	vsi->tc_config.numtc = vsi->mqprio_qopt.qopt.num_tc;
	vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1;
	num_qps = vsi->mqprio_qopt.qopt.count[0];

	/* find the next higher power-of-2 of num queue pairs */
	pow = ilog2(num_qps);
	if (!is_power_of_2(num_qps))
		pow++;
	qmap = (offset << I40E_AQ_VSI_TC_QUE_OFFSET_SHIFT) |
		(pow << I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT);

	/* Setup queue offset/count for all TCs for given VSI */
	max_qcount = vsi->mqprio_qopt.qopt.count[0];
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		/* See if the given TC is enabled for the given VSI */
		if (vsi->tc_config.enabled_tc & BIT(i)) {
			offset = vsi->mqprio_qopt.qopt.offset[i];
			qcount = vsi->mqprio_qopt.qopt.count[i];
			if (qcount > max_qcount)
				max_qcount = qcount;
			vsi->tc_config.tc_info[i].qoffset = offset;
			vsi->tc_config.tc_info[i].qcount = qcount;
			vsi->tc_config.tc_info[i].netdev_tc = netdev_tc++;
		} else {
			/* TC is not enabled so set the offset to
			 * default queue and allocate one queue
			 * for the given TC.
			 */
			vsi->tc_config.tc_info[i].qoffset = 0;
			vsi->tc_config.tc_info[i].qcount = 1;
			vsi->tc_config.tc_info[i].netdev_tc = 0;
		}
	}

	/* Set actual Tx/Rx queue pairs */
	vsi->num_queue_pairs = offset + qcount;

	/* Setup queue TC[0].qmap for given VSI context */
	ctxt->info.tc_mapping[0] = cpu_to_le16(qmap);
	ctxt->info.mapping_flags |= cpu_to_le16(I40E_AQ_VSI_QUE_MAP_CONTIG);
	ctxt->info.queue_mapping[0] = cpu_to_le16(vsi->base_queue);
	ctxt->info.valid_sections |= cpu_to_le16(sections);

	/* Reconfigure RSS for main VSI with max queue count */
	vsi->rss_size = max_qcount;
	ret = i40e_vsi_config_rss(vsi);
	if (ret) {
		dev_info(&vsi->back->pdev->dev,
			 "Failed to reconfig rss for num_queues (%u)\n",
			 max_qcount);
		return ret;
	}
	vsi->reconfig_rss = true;
	dev_dbg(&vsi->back->pdev->dev,
		"Reconfigured rss with num_queues (%u)\n", max_qcount);

	/* Find queue count available for channel VSIs and starting offset
	 * for channel VSIs
	 */
	override_q = vsi->mqprio_qopt.qopt.count[0];
	if (override_q && override_q < vsi->num_queue_pairs) {
		vsi->cnt_q_avail = vsi->num_queue_pairs - override_q;
		vsi->next_base_queue = override_q;
	}
	return 0;
}

/**
 * i40e_vsi_setup_queue_map - Setup a VSI queue map based on enabled_tc
 * @vsi: the VSI being setup
 * @ctxt: VSI context structure
 * @enabled_tc: Enabled TCs bitmap
 * @is_add: True if called before Add VSI
 *
 * Setup VSI queue mapping for enabled traffic classes.
 **/
static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi,
				     struct i40e_vsi_context *ctxt,
				     u8 enabled_tc,
				     bool is_add)
{
	struct i40e_pf *pf = vsi->back;
	u16 num_tc_qps = 0;
	u16 sections = 0;
	u8 netdev_tc = 0;
	u16 numtc = 1;
	u16 qcount;
	u8 offset;
	u16 qmap;
	int i;

	sections = I40E_AQ_VSI_PROP_QUEUE_MAP_VALID;
	offset = 0;
	/* zero out queue mapping, it will get updated on the end of the function */
	memset(ctxt->info.queue_mapping, 0, sizeof(ctxt->info.queue_mapping));

	if (vsi->type == I40E_VSI_MAIN) {
		/* This code helps add more queue to the VSI if we have
		 * more cores than RSS can support, the higher cores will
		 * be served by ATR or other filters. Furthermore, the
		 * non-zero req_queue_pairs says that user requested a new
		 * queue count via ethtool's set_channels, so use this
		 * value for queues distribution across traffic classes
		 * We need at least one queue pair for the interface
		 * to be usable as we see in else statement.
		 */
		if (vsi->req_queue_pairs > 0)
			vsi->num_queue_pairs = vsi->req_queue_pairs;
		else if (pf->flags & I40E_FLAG_MSIX_ENABLED)
			vsi->num_queue_pairs = pf->num_lan_msix;
		else
			vsi->num_queue_pairs = 1;
	}

	/* Number of queues per enabled TC */
	if (vsi->type == I40E_VSI_MAIN ||
	    (vsi->type == I40E_VSI_SRIOV && vsi->num_queue_pairs != 0))
		num_tc_qps = vsi->num_queue_pairs;
	else
		num_tc_qps = vsi->alloc_queue_pairs;

	if (enabled_tc && (vsi->back->flags & I40E_FLAG_DCB_ENABLED)) {
		/* Find numtc from enabled TC bitmap */
		for (i = 0, numtc = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
			if (enabled_tc & BIT(i)) /* TC is enabled */
				numtc++;
		}
		if (!numtc) {
			dev_warn(&pf->pdev->dev, "DCB is enabled but no TC enabled, forcing TC0\n");
			numtc = 1;
		}
		num_tc_qps = num_tc_qps / numtc;
		num_tc_qps = min_t(int, num_tc_qps,
				   i40e_pf_get_max_q_per_tc(pf));
	}

	vsi->tc_config.numtc = numtc;
	vsi->tc_config.enabled_tc = enabled_tc ? enabled_tc : 1;

	/* Do not allow use more TC queue pairs than MSI-X vectors exist */
	if (pf->flags & I40E_FLAG_MSIX_ENABLED)
		num_tc_qps = min_t(int, num_tc_qps, pf->num_lan_msix);

	/* Setup queue offset/count for all TCs for given VSI */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		/* See if the given TC is enabled for the given VSI */
		if (vsi->tc_config.enabled_tc & BIT(i)) {
			/* TC is enabled */
			int pow, num_qps;

			switch (vsi->type) {
			case I40E_VSI_MAIN:
				if (!(pf->flags & (I40E_FLAG_FD_SB_ENABLED |
				    I40E_FLAG_FD_ATR_ENABLED)) ||
				    vsi->tc_config.enabled_tc != 1) {
					qcount = min_t(int, pf->alloc_rss_size,
						       num_tc_qps);
					break;
				}
				fallthrough;
			case I40E_VSI_FDIR:
			case I40E_VSI_SRIOV:
			case I40E_VSI_VMDQ2:
			default:
				qcount = num_tc_qps;
				WARN_ON(i != 0);
				break;
			}
			vsi->tc_config.tc_info[i].qoffset = offset;
			vsi->tc_config.tc_info[i].qcount = qcount;

			/* find the next higher power-of-2 of num queue pairs */
			num_qps = qcount;
			pow = 0;
			while (num_qps && (BIT_ULL(pow) < qcount)) {
				pow++;
				num_qps >>= 1;
			}

			vsi->tc_config.tc_info[i].netdev_tc = netdev_tc++;
			qmap =
			    (offset << I40E_AQ_VSI_TC_QUE_OFFSET_SHIFT) |
			    (pow << I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT);

			offset += qcount;
		} else {
			/* TC is not enabled so set the offset to
			 * default queue and allocate one queue
			 * for the given TC.
			 */
			vsi->tc_config.tc_info[i].qoffset = 0;
			vsi->tc_config.tc_info[i].qcount = 1;
			vsi->tc_config.tc_info[i].netdev_tc = 0;

			qmap = 0;
		}
		ctxt->info.tc_mapping[i] = cpu_to_le16(qmap);
	}
	/* Do not change previously set num_queue_pairs for PFs and VFs*/
	if ((vsi->type == I40E_VSI_MAIN && numtc != 1) ||
	    (vsi->type == I40E_VSI_SRIOV && vsi->num_queue_pairs == 0) ||
	    (vsi->type != I40E_VSI_MAIN && vsi->type != I40E_VSI_SRIOV))
		vsi->num_queue_pairs = offset;

	/* Scheduler section valid can only be set for ADD VSI */
	if (is_add) {
		sections |= I40E_AQ_VSI_PROP_SCHED_VALID;

		ctxt->info.up_enable_bits = enabled_tc;
	}
	if (vsi->type == I40E_VSI_SRIOV) {
		ctxt->info.mapping_flags |=
				     cpu_to_le16(I40E_AQ_VSI_QUE_MAP_NONCONTIG);
		for (i = 0; i < vsi->num_queue_pairs; i++)
			ctxt->info.queue_mapping[i] =
					       cpu_to_le16(vsi->base_queue + i);
	} else {
		ctxt->info.mapping_flags |=
					cpu_to_le16(I40E_AQ_VSI_QUE_MAP_CONTIG);
		ctxt->info.queue_mapping[0] = cpu_to_le16(vsi->base_queue);
	}
	ctxt->info.valid_sections |= cpu_to_le16(sections);
}

/**
 * i40e_addr_sync - Callback for dev_(mc|uc)_sync to add address
 * @netdev: the netdevice
 * @addr: address to add
 *
 * Called by __dev_(mc|uc)_sync when an address needs to be added. We call
 * __dev_(uc|mc)_sync from .set_rx_mode and guarantee to hold the hash lock.
 */
static int i40e_addr_sync(struct net_device *netdev, const u8 *addr)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;

	if (i40e_add_mac_filter(vsi, addr))
		return 0;
	else
		return -ENOMEM;
}

/**
 * i40e_addr_unsync - Callback for dev_(mc|uc)_sync to remove address
 * @netdev: the netdevice
 * @addr: address to add
 *
 * Called by __dev_(mc|uc)_sync when an address needs to be removed. We call
 * __dev_(uc|mc)_sync from .set_rx_mode and guarantee to hold the hash lock.
 */
static int i40e_addr_unsync(struct net_device *netdev, const u8 *addr)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_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/VLAN 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;

	i40e_del_mac_filter(vsi, addr);

	return 0;
}

/**
 * i40e_set_rx_mode - NDO callback to set the netdev filters
 * @netdev: network interface device structure
 **/
static void i40e_set_rx_mode(struct net_device *netdev)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;

	spin_lock_bh(&vsi->mac_filter_hash_lock);

	__dev_uc_sync(netdev, i40e_addr_sync, i40e_addr_unsync);
	__dev_mc_sync(netdev, i40e_addr_sync, i40e_addr_unsync);

	spin_unlock_bh(&vsi->mac_filter_hash_lock);

	/* check for other flag changes */
	if (vsi->current_netdev_flags != vsi->netdev->flags) {
		vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
		set_bit(__I40E_MACVLAN_SYNC_PENDING, vsi->back->state);
	}
}

/**
 * i40e_undo_del_filter_entries - Undo the changes made to MAC filter entries
 * @vsi: Pointer to VSI struct
 * @from: Pointer to list which contains MAC filter entries - changes to
 *        those entries needs to be undone.
 *
 * MAC filter entries from this list were slated for deletion.
 **/
static void i40e_undo_del_filter_entries(struct i40e_vsi *vsi,
					 struct hlist_head *from)
{
	struct i40e_mac_filter *f;
	struct hlist_node *h;

	hlist_for_each_entry_safe(f, h, from, hlist) {
		u64 key = i40e_addr_to_hkey(f->macaddr);

		/* Move the element back into MAC filter list*/
		hlist_del(&f->hlist);
		hash_add(vsi->mac_filter_hash, &f->hlist, key);
	}
}

/**
 * i40e_undo_add_filter_entries - Undo the changes made to MAC filter entries
 * @vsi: Pointer to vsi struct
 * @from: Pointer to list which contains MAC filter entries - changes to
 *        those entries needs to be undone.
 *
 * MAC filter entries from this list were slated for addition.
 **/
static void i40e_undo_add_filter_entries(struct i40e_vsi *vsi,
					 struct hlist_head *from)
{
	struct i40e_new_mac_filter *new;
	struct hlist_node *h;

	hlist_for_each_entry_safe(new, h, from, hlist) {
		/* We can simply free the wrapper structure */
		hlist_del(&new->hlist);
		netdev_hw_addr_refcnt(new->f, vsi->netdev, -1);
		kfree(new);
	}
}

/**
 * i40e_next_filter - Get the next non-broadcast filter from a list
 * @next: pointer to filter in list
 *
 * Returns the next non-broadcast filter in the list. Required so that we
 * ignore broadcast filters within the list, since these are not handled via
 * the normal firmware update path.
 */
static
struct i40e_new_mac_filter *i40e_next_filter(struct i40e_new_mac_filter *next)
{
	hlist_for_each_entry_continue(next, hlist) {
		if (!is_broadcast_ether_addr(next->f->macaddr))
			return next;
	}

	return NULL;
}

/**
 * i40e_update_filter_state - Update filter state based on return data
 * from firmware
 * @count: Number of filters added
 * @add_list: return data from fw
 * @add_head: pointer to first filter in current batch
 *
 * MAC filter entries from list were slated to be added to device. Returns
 * number of successful filters. Note that 0 does NOT mean success!
 **/
static int
i40e_update_filter_state(int count,
			 struct i40e_aqc_add_macvlan_element_data *add_list,
			 struct i40e_new_mac_filter *add_head)
{
	int retval = 0;
	int i;

	for (i = 0; i < count; i++) {
		/* Always check status of each filter. We don't need to check
		 * the firmware return status because we pre-set the filter
		 * status to I40E_AQC_MM_ERR_NO_RES when sending the filter
		 * request to the adminq. Thus, if it no longer matches then
		 * we know the filter is active.
		 */
		if (add_list[i].match_method == I40E_AQC_MM_ERR_NO_RES) {
			add_head->state = I40E_FILTER_FAILED;
		} else {
			add_head->state = I40E_FILTER_ACTIVE;
			retval++;
		}

		add_head = i40e_next_filter(add_head);
		if (!add_head)
			break;
	}

	return retval;
}

/**
 * i40e_aqc_del_filters - Request firmware to delete a set of filters
 * @vsi: ptr to the VSI
 * @vsi_name: name to display in messages
 * @list: the list of filters to send to firmware
 * @num_del: the number of filters to delete
 * @retval: Set to -EIO on failure to delete
 *
 * Send a request to firmware via AdminQ to delete a set of filters. Uses
 * *retval instead of a return value so that success does not force ret_val to
 * be set to 0. This ensures that a sequence of calls to this function
 * preserve the previous value of *retval on successful delete.
 */
static
void i40e_aqc_del_filters(struct i40e_vsi *vsi, const char *vsi_name,
			  struct i40e_aqc_remove_macvlan_element_data *list,
			  int num_del, int *retval)
{
	struct i40e_hw *hw = &vsi->back->hw;
	enum i40e_admin_queue_err aq_status;
	int aq_ret;

	aq_ret = i40e_aq_remove_macvlan_v2(hw, vsi->seid, list, num_del, NULL,
					   &aq_status);

	/* Explicitly ignore and do not report when firmware returns ENOENT */
	if (aq_ret && !(aq_status == I40E_AQ_RC_ENOENT)) {
		*retval = -EIO;
		dev_info(&vsi->back->pdev->dev,
			 "ignoring delete macvlan error on %s, err %pe, aq_err %s\n",
			 vsi_name, ERR_PTR(aq_ret),
			 i40e_aq_str(hw, aq_status));
	}
}

/**
 * i40e_aqc_add_filters - Request firmware to add a set of filters
 * @vsi: ptr to the VSI
 * @vsi_name: name to display in messages
 * @list: the list of filters to send to firmware
 * @add_head: Position in the add hlist
 * @num_add: the number of filters to add
 *
 * Send a request to firmware via AdminQ to add a chunk of filters. Will set
 * __I40E_VSI_OVERFLOW_PROMISC bit in vsi->state if the firmware has run out of
 * space for more filters.
 */
static
void i40e_aqc_add_filters(struct i40e_vsi *vsi, const char *vsi_name,
			  struct i40e_aqc_add_macvlan_element_data *list,
			  struct i40e_new_mac_filter *add_head,
			  int num_add)
{
	struct i40e_hw *hw = &vsi->back->hw;
	enum i40e_admin_queue_err aq_status;
	int fcnt;

	i40e_aq_add_macvlan_v2(hw, vsi->seid, list, num_add, NULL, &aq_status);
	fcnt = i40e_update_filter_state(num_add, list, add_head);

	if (fcnt != num_add) {
		if (vsi->type == I40E_VSI_MAIN) {
			set_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state);
			dev_warn(&vsi->back->pdev->dev,
				 "Error %s adding RX filters on %s, promiscuous mode forced on\n",
				 i40e_aq_str(hw, aq_status), vsi_name);
		} else if (vsi->type == I40E_VSI_SRIOV ||
			   vsi->type == I40E_VSI_VMDQ1 ||
			   vsi->type == I40E_VSI_VMDQ2) {
			dev_warn(&vsi->back->pdev->dev,
				 "Error %s adding RX filters on %s, please set promiscuous on manually for %s\n",
				 i40e_aq_str(hw, aq_status), vsi_name,
					     vsi_name);
		} else {
			dev_warn(&vsi->back->pdev->dev,
				 "Error %s adding RX filters on %s, incorrect VSI type: %i.\n",
				 i40e_aq_str(hw, aq_status), vsi_name,
					     vsi->type);
		}
	}
}

/**
 * i40e_aqc_broadcast_filter - Set promiscuous broadcast flags
 * @vsi: pointer to the VSI
 * @vsi_name: the VSI name
 * @f: filter data
 *
 * This function sets or clears the promiscuous broadcast flags for VLAN
 * filters in order to properly receive broadcast frames. Assumes that only
 * broadcast filters are passed.
 *
 * Returns status indicating success or failure;
 **/
static int
i40e_aqc_broadcast_filter(struct i40e_vsi *vsi, const char *vsi_name,
			  struct i40e_mac_filter *f)
{
	bool enable = f->state == I40E_FILTER_NEW;
	struct i40e_hw *hw = &vsi->back->hw;
	int aq_ret;

	if (f->vlan == I40E_VLAN_ANY) {
		aq_ret = i40e_aq_set_vsi_broadcast(hw,
						   vsi->seid,
						   enable,
						   NULL);
	} else {
		aq_ret = i40e_aq_set_vsi_bc_promisc_on_vlan(hw,
							    vsi->seid,
							    enable,
							    f->vlan,
							    NULL);
	}

	if (aq_ret) {
		set_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state);
		dev_warn(&vsi->back->pdev->dev,
			 "Error %s, forcing overflow promiscuous on %s\n",
			 i40e_aq_str(hw, hw->aq.asq_last_status),
			 vsi_name);
	}

	return aq_ret;
}

/**
 * i40e_set_promiscuous - set promiscuous mode
 * @pf: board private structure
 * @promisc: promisc on or off
 *
 * There are different ways of setting promiscuous mode on a PF depending on
 * what state/environment we're in.  This identifies and sets it appropriately.
 * Returns 0 on success.
 **/
static int i40e_set_promiscuous(struct i40e_pf *pf, bool promisc)
{
	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
	struct i40e_hw *hw = &pf->hw;
	int aq_ret;

	if (vsi->type == I40E_VSI_MAIN &&
	    pf->lan_veb != I40E_NO_VEB &&
	    !(pf->flags & I40E_FLAG_MFP_ENABLED)) {
		/* set defport ON for Main VSI instead of true promisc
		 * this way we will get all unicast/multicast and VLAN
		 * promisc behavior but will not get VF or VMDq traffic
		 * replicated on the Main VSI.
		 */
		if (promisc)
			aq_ret = i40e_aq_set_default_vsi(hw,
							 vsi->seid,
							 NULL);
		else
			aq_ret = i40e_aq_clear_default_vsi(hw,
							   vsi->seid,
							   NULL);
		if (aq_ret) {
			dev_info(&pf->pdev->dev,
				 "Set default VSI failed, err %pe, aq_err %s\n",
				 ERR_PTR(aq_ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
		}
	} else {
		aq_ret = i40e_aq_set_vsi_unicast_promiscuous(
						  hw,
						  vsi->seid,
						  promisc, NULL,
						  true);
		if (aq_ret) {
			dev_info(&pf->pdev->dev,
				 "set unicast promisc failed, err %pe, aq_err %s\n",
				 ERR_PTR(aq_ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
		}
		aq_ret = i40e_aq_set_vsi_multicast_promiscuous(
						  hw,
						  vsi->seid,
						  promisc, NULL);
		if (aq_ret) {
			dev_info(&pf->pdev->dev,
				 "set multicast promisc failed, err %pe, aq_err %s\n",
				 ERR_PTR(aq_ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
		}
	}

	if (!aq_ret)
		pf->cur_promisc = promisc;

	return aq_ret;
}

/**
 * i40e_sync_vsi_filters - Update the VSI filter list to the HW
 * @vsi: ptr to the VSI
 *
 * Push any outstanding VSI filter changes through the AdminQ.
 *
 * Returns 0 or error value
 **/
int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
{
	struct hlist_head tmp_add_list, tmp_del_list;
	struct i40e_mac_filter *f;
	struct i40e_new_mac_filter *new, *add_head = NULL;
	struct i40e_hw *hw = &vsi->back->hw;
	bool old_overflow, new_overflow;
	unsigned int failed_filters = 0;
	unsigned int vlan_filters = 0;
	char vsi_name[16] = "PF";
	int filter_list_len = 0;
	u32 changed_flags = 0;
	struct hlist_node *h;
	struct i40e_pf *pf;
	int num_add = 0;
	int num_del = 0;
	int aq_ret = 0;
	int retval = 0;
	u16 cmd_flags;
	int list_size;
	int bkt;

	/* empty array typed pointers, kcalloc later */
	struct i40e_aqc_add_macvlan_element_data *add_list;
	struct i40e_aqc_remove_macvlan_element_data *del_list;

	while (test_and_set_bit(__I40E_VSI_SYNCING_FILTERS, vsi->state))
		usleep_range(1000, 2000);
	pf = vsi->back;

	old_overflow = test_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state);

	if (vsi->netdev) {
		changed_flags = vsi->current_netdev_flags ^ vsi->netdev->flags;
		vsi->current_netdev_flags = vsi->netdev->flags;
	}

	INIT_HLIST_HEAD(&tmp_add_list);
	INIT_HLIST_HEAD(&tmp_del_list);

	if (vsi->type == I40E_VSI_SRIOV)
		snprintf(vsi_name, sizeof(vsi_name) - 1, "VF %d", vsi->vf_id);
	else if (vsi->type != I40E_VSI_MAIN)
		snprintf(vsi_name, sizeof(vsi_name) - 1, "vsi %d", vsi->seid);

	if (vsi->flags & I40E_VSI_FLAG_FILTER_CHANGED) {
		vsi->flags &= ~I40E_VSI_FLAG_FILTER_CHANGED;

		spin_lock_bh(&vsi->mac_filter_hash_lock);
		/* Create a list of filters to delete. */
		hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
			if (f->state == I40E_FILTER_REMOVE) {
				/* Move the element into temporary del_list */
				hash_del(&f->hlist);
				hlist_add_head(&f->hlist, &tmp_del_list);

				/* Avoid counting removed filters */
				continue;
			}
			if (f->state == I40E_FILTER_NEW) {
				/* Create a temporary i40e_new_mac_filter */
				new = kzalloc(sizeof(*new), GFP_ATOMIC);
				if (!new)
					goto err_no_memory_locked;

				/* Store pointer to the real filter */
				new->f = f;
				new->state = f->state;

				/* Add it to the hash list */
				hlist_add_head(&new->hlist, &tmp_add_list);
			}

			/* Count the number of active (current and new) VLAN
			 * filters we have now. Does not count filters which
			 * are marked for deletion.
			 */
			if (f->vlan > 0)
				vlan_filters++;
		}

		if (vsi->type != I40E_VSI_SRIOV)
			retval = i40e_correct_mac_vlan_filters
				(vsi, &tmp_add_list, &tmp_del_list,
				 vlan_filters);
		else
			retval = i40e_correct_vf_mac_vlan_filters
				(vsi, &tmp_add_list, &tmp_del_list,
				 vlan_filters, pf->vf[vsi->vf_id].trusted);

		hlist_for_each_entry(new, &tmp_add_list, hlist)
			netdev_hw_addr_refcnt(new->f, vsi->netdev, 1);

		if (retval)
			goto err_no_memory_locked;

		spin_unlock_bh(&vsi->mac_filter_hash_lock);
	}

	/* Now process 'del_list' outside the lock */
	if (!hlist_empty(&tmp_del_list)) {
		filter_list_len = hw->aq.asq_buf_size /
			    sizeof(struct i40e_aqc_remove_macvlan_element_data);
		list_size = filter_list_len *
			    sizeof(struct i40e_aqc_remove_macvlan_element_data);
		del_list = kzalloc(list_size, GFP_ATOMIC);
		if (!del_list)
			goto err_no_memory;

		hlist_for_each_entry_safe(f, h, &tmp_del_list, hlist) {
			cmd_flags = 0;

			/* handle broadcast filters by updating the broadcast
			 * promiscuous flag and release filter list.
			 */
			if (is_broadcast_ether_addr(f->macaddr)) {
				i40e_aqc_broadcast_filter(vsi, vsi_name, f);

				hlist_del(&f->hlist);
				kfree(f);
				continue;
			}

			/* add to delete list */
			ether_addr_copy(del_list[num_del].mac_addr, f->macaddr);
			if (f->vlan == I40E_VLAN_ANY) {
				del_list[num_del].vlan_tag = 0;
				cmd_flags |= I40E_AQC_MACVLAN_DEL_IGNORE_VLAN;
			} else {
				del_list[num_del].vlan_tag =
					cpu_to_le16((u16)(f->vlan));
			}

			cmd_flags |= I40E_AQC_MACVLAN_DEL_PERFECT_MATCH;
			del_list[num_del].flags = cmd_flags;
			num_del++;

			/* flush a full buffer */
			if (num_del == filter_list_len) {
				i40e_aqc_del_filters(vsi, vsi_name, del_list,
						     num_del, &retval);
				memset(del_list, 0, list_size);
				num_del = 0;
			}
			/* Release memory for MAC filter entries which were
			 * synced up with HW.
			 */
			hlist_del(&f->hlist);
			kfree(f);
		}

		if (num_del) {
			i40e_aqc_del_filters(vsi, vsi_name, del_list,
					     num_del, &retval);
		}

		kfree(del_list);
		del_list = NULL;
	}

	if (!hlist_empty(&tmp_add_list)) {
		/* Do all the adds now. */
		filter_list_len = hw->aq.asq_buf_size /
			       sizeof(struct i40e_aqc_add_macvlan_element_data);
		list_size = filter_list_len *
			       sizeof(struct i40e_aqc_add_macvlan_element_data);
		add_list = kzalloc(list_size, GFP_ATOMIC);
		if (!add_list)
			goto err_no_memory;

		num_add = 0;
		hlist_for_each_entry_safe(new, h, &tmp_add_list, hlist) {
			/* handle broadcast filters by updating the broadcast
			 * promiscuous flag instead of adding a MAC filter.
			 */
			if (is_broadcast_ether_addr(new->f->macaddr)) {
				if (i40e_aqc_broadcast_filter(vsi, vsi_name,
							      new->f))
					new->state = I40E_FILTER_FAILED;
				else
					new->state = I40E_FILTER_ACTIVE;
				continue;
			}

			/* add to add array */
			if (num_add == 0)
				add_head = new;
			cmd_flags = 0;
			ether_addr_copy(add_list[num_add].mac_addr,
					new->f->macaddr);
			if (new->f->vlan == I40E_VLAN_ANY) {
				add_list[num_add].vlan_tag = 0;
				cmd_flags |= I40E_AQC_MACVLAN_ADD_IGNORE_VLAN;
			} else {
				add_list[num_add].vlan_tag =
					cpu_to_le16((u16)(new->f->vlan));
			}
			add_list[num_add].queue_number = 0;
			/* set invalid match method for later detection */
			add_list[num_add].match_method = I40E_AQC_MM_ERR_NO_RES;
			cmd_flags |= I40E_AQC_MACVLAN_ADD_PERFECT_MATCH;
			add_list[num_add].flags = cpu_to_le16(cmd_flags);
			num_add++;

			/* flush a full buffer */
			if (num_add == filter_list_len) {
				i40e_aqc_add_filters(vsi, vsi_name, add_list,
						     add_head, num_add);
				memset(add_list, 0, list_size);
				num_add = 0;
			}
		}
		if (num_add) {
			i40e_aqc_add_filters(vsi, vsi_name, add_list, add_head,
					     num_add);
		}
		/* Now move all of the filters from the temp add list back to
		 * the VSI's list.
		 */
		spin_lock_bh(&vsi->mac_filter_hash_lock);
		hlist_for_each_entry_safe(new, h, &tmp_add_list, hlist) {
			/* Only update the state if we're still NEW */
			if (new->f->state == I40E_FILTER_NEW)
				new->f->state = new->state;
			hlist_del(&new->hlist);
			netdev_hw_addr_refcnt(new->f, vsi->netdev, -1);
			kfree(new);
		}
		spin_unlock_bh(&vsi->mac_filter_hash_lock);
		kfree(add_list);
		add_list = NULL;
	}

	/* Determine the number of active and failed filters. */
	spin_lock_bh(&vsi->mac_filter_hash_lock);
	vsi->active_filters = 0;
	hash_for_each(vsi->mac_filter_hash, bkt, f, hlist) {
		if (f->state == I40E_FILTER_ACTIVE)
			vsi->active_filters++;
		else if (f->state == I40E_FILTER_FAILED)
			failed_filters++;
	}
	spin_unlock_bh(&vsi->mac_filter_hash_lock);

	/* Check if we are able to exit overflow promiscuous mode. We can
	 * safely exit if we didn't just enter, we no longer have any failed
	 * filters, and we have reduced filters below the threshold value.
	 */
	if (old_overflow && !failed_filters &&
	    vsi->active_filters < vsi->promisc_threshold) {
		dev_info(&pf->pdev->dev,
			 "filter logjam cleared on %s, leaving overflow promiscuous mode\n",
			 vsi_name);
		clear_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state);
		vsi->promisc_threshold = 0;
	}

	/* if the VF is not trusted do not do promisc */
	if ((vsi->type == I40E_VSI_SRIOV) && !pf->vf[vsi->vf_id].trusted) {
		clear_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state);
		goto out;
	}

	new_overflow = test_bit(__I40E_VSI_OVERFLOW_PROMISC, vsi->state);

	/* If we are entering overflow promiscuous, we need to calculate a new
	 * threshold for when we are safe to exit
	 */
	if (!old_overflow && new_overflow)
		vsi->promisc_threshold = (vsi->active_filters * 3) / 4;

	/* check for changes in promiscuous modes */
	if (changed_flags & IFF_ALLMULTI) {
		bool cur_multipromisc;

		cur_multipromisc = !!(vsi->current_netdev_flags & IFF_ALLMULTI);
		aq_ret = i40e_aq_set_vsi_multicast_promiscuous(&vsi->back->hw,
							       vsi->seid,
							       cur_multipromisc,
							       NULL);
		if (aq_ret) {
			retval = i40e_aq_rc_to_posix(aq_ret,
						     hw->aq.asq_last_status);
			dev_info(&pf->pdev->dev,
				 "set multi promisc failed on %s, err %pe aq_err %s\n",
				 vsi_name,
				 ERR_PTR(aq_ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
		} else {
			dev_info(&pf->pdev->dev, "%s allmulti mode.\n",
				 cur_multipromisc ? "entering" : "leaving");
		}
	}

	if ((changed_flags & IFF_PROMISC) || old_overflow != new_overflow) {
		bool cur_promisc;

		cur_promisc = (!!(vsi->current_netdev_flags & IFF_PROMISC) ||
			       new_overflow);
		aq_ret = i40e_set_promiscuous(pf, cur_promisc);
		if (aq_ret) {
			retval = i40e_aq_rc_to_posix(aq_ret,
						     hw->aq.asq_last_status);
			dev_info(&pf->pdev->dev,
				 "Setting promiscuous %s failed on %s, err %pe aq_err %s\n",
				 cur_promisc ? "on" : "off",
				 vsi_name,
				 ERR_PTR(aq_ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
		}
	}
out:
	/* if something went wrong then set the changed flag so we try again */
	if (retval)
		vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;

	clear_bit(__I40E_VSI_SYNCING_FILTERS, vsi->state);
	return retval;

err_no_memory:
	/* Restore elements on the temporary add and delete lists */
	spin_lock_bh(&vsi->mac_filter_hash_lock);
err_no_memory_locked:
	i40e_undo_del_filter_entries(vsi, &tmp_del_list);
	i40e_undo_add_filter_entries(vsi, &tmp_add_list);
	spin_unlock_bh(&vsi->mac_filter_hash_lock);

	vsi->flags |= I40E_VSI_FLAG_FILTER_CHANGED;
	clear_bit(__I40E_VSI_SYNCING_FILTERS, vsi->state);
	return -ENOMEM;
}

/**
 * i40e_sync_filters_subtask - Sync the VSI filter list with HW
 * @pf: board private structure
 **/
static void i40e_sync_filters_subtask(struct i40e_pf *pf)
{
	int v;

	if (!pf)
		return;
	if (!test_and_clear_bit(__I40E_MACVLAN_SYNC_PENDING, pf->state))
		return;
	if (test_bit(__I40E_VF_DISABLE, pf->state)) {
		set_bit(__I40E_MACVLAN_SYNC_PENDING, pf->state);
		return;
	}

	for (v = 0; v < pf->num_alloc_vsi; v++) {
		if (pf->vsi[v] &&
		    (pf->vsi[v]->flags & I40E_VSI_FLAG_FILTER_CHANGED) &&
		    !test_bit(__I40E_VSI_RELEASING, pf->vsi[v]->state)) {
			int ret = i40e_sync_vsi_filters(pf->vsi[v]);

			if (ret) {
				/* come back and try again later */
				set_bit(__I40E_MACVLAN_SYNC_PENDING,
					pf->state);
				break;
			}
		}
	}
}

/**
 * i40e_max_xdp_frame_size - returns the maximum allowed frame size for XDP
 * @vsi: the vsi
 **/
static int i40e_max_xdp_frame_size(struct i40e_vsi *vsi)
{
	if (PAGE_SIZE >= 8192 || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
		return I40E_RXBUFFER_2048;
	else
		return I40E_RXBUFFER_3072;
}

/**
 * i40e_change_mtu - NDO callback to change the Maximum Transfer Unit
 * @netdev: network interface device structure
 * @new_mtu: new value for maximum frame size
 *
 * Returns 0 on success, negative on failure
 **/
static int i40e_change_mtu(struct net_device *netdev, int new_mtu)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;
	struct i40e_pf *pf = vsi->back;

	if (i40e_enabled_xdp_vsi(vsi)) {
		int frame_size = new_mtu + I40E_PACKET_HDR_PAD;

		if (frame_size > i40e_max_xdp_frame_size(vsi))
			return -EINVAL;
	}

	netdev_dbg(netdev, "changing MTU from %d to %d\n",
		   netdev->mtu, new_mtu);
	netdev->mtu = new_mtu;
	if (netif_running(netdev))
		i40e_vsi_reinit_locked(vsi);
	set_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state);
	set_bit(__I40E_CLIENT_L2_CHANGE, pf->state);
	return 0;
}

/**
 * i40e_ioctl - Access the hwtstamp interface
 * @netdev: network interface device structure
 * @ifr: interface request data
 * @cmd: ioctl command
 **/
int i40e_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_pf *pf = np->vsi->back;

	switch (cmd) {
	case SIOCGHWTSTAMP:
		return i40e_ptp_get_ts_config(pf, ifr);
	case SIOCSHWTSTAMP:
		return i40e_ptp_set_ts_config(pf, ifr);
	default:
		return -EOPNOTSUPP;
	}
}

/**
 * i40e_vlan_stripping_enable - Turn on vlan stripping for the VSI
 * @vsi: the vsi being adjusted
 **/
void i40e_vlan_stripping_enable(struct i40e_vsi *vsi)
{
	struct i40e_vsi_context ctxt;
	int ret;

	/* Don't modify stripping options if a port VLAN is active */
	if (vsi->info.pvid)
		return;

	if ((vsi->info.valid_sections &
	     cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID)) &&
	    ((vsi->info.port_vlan_flags & I40E_AQ_VSI_PVLAN_MODE_MASK) == 0))
		return;  /* already enabled */

	vsi->info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
	vsi->info.port_vlan_flags = I40E_AQ_VSI_PVLAN_MODE_ALL |
				    I40E_AQ_VSI_PVLAN_EMOD_STR_BOTH;

	ctxt.seid = vsi->seid;
	ctxt.info = vsi->info;
	ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
	if (ret) {
		dev_info(&vsi->back->pdev->dev,
			 "update vlan stripping failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&vsi->back->hw,
				     vsi->back->hw.aq.asq_last_status));
	}
}

/**
 * i40e_vlan_stripping_disable - Turn off vlan stripping for the VSI
 * @vsi: the vsi being adjusted
 **/
void i40e_vlan_stripping_disable(struct i40e_vsi *vsi)
{
	struct i40e_vsi_context ctxt;
	int ret;

	/* Don't modify stripping options if a port VLAN is active */
	if (vsi->info.pvid)
		return;

	if ((vsi->info.valid_sections &
	     cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID)) &&
	    ((vsi->info.port_vlan_flags & I40E_AQ_VSI_PVLAN_EMOD_MASK) ==
	     I40E_AQ_VSI_PVLAN_EMOD_MASK))
		return;  /* already disabled */

	vsi->info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
	vsi->info.port_vlan_flags = I40E_AQ_VSI_PVLAN_MODE_ALL |
				    I40E_AQ_VSI_PVLAN_EMOD_NOTHING;

	ctxt.seid = vsi->seid;
	ctxt.info = vsi->info;
	ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
	if (ret) {
		dev_info(&vsi->back->pdev->dev,
			 "update vlan stripping failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&vsi->back->hw,
				     vsi->back->hw.aq.asq_last_status));
	}
}

/**
 * i40e_add_vlan_all_mac - Add a MAC/VLAN filter for each existing MAC address
 * @vsi: the vsi being configured
 * @vid: vlan id to be added (0 = untagged only , -1 = any)
 *
 * This is a helper function for adding a new MAC/VLAN filter with the
 * specified VLAN for each existing MAC address already in the hash table.
 * This function does *not* perform any accounting to update filters based on
 * VLAN mode.
 *
 * NOTE: this function expects to be called while under the
 * mac_filter_hash_lock
 **/
int i40e_add_vlan_all_mac(struct i40e_vsi *vsi, s16 vid)
{
	struct i40e_mac_filter *f, *add_f;
	struct hlist_node *h;
	int bkt;

	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
		/* If we're asked to add a filter that has been marked for
		 * removal, it is safe to simply restore it to active state.
		 * __i40e_del_filter will have simply deleted any filters which
		 * were previously marked NEW or FAILED, so if it is currently
		 * marked REMOVE it must have previously been ACTIVE. Since we
		 * haven't yet run the sync filters task, just restore this
		 * filter to the ACTIVE state so that the sync task leaves it
		 * in place.
		 */
		if (f->state == I40E_FILTER_REMOVE && f->vlan == vid) {
			f->state = I40E_FILTER_ACTIVE;
			continue;
		} else if (f->state == I40E_FILTER_REMOVE) {
			continue;
		}
		add_f = i40e_add_filter(vsi, f->macaddr, vid);
		if (!add_f) {
			dev_info(&vsi->back->pdev->dev,
				 "Could not add vlan filter %d for %pM\n",
				 vid, f->macaddr);
			return -ENOMEM;
		}
	}

	return 0;
}

/**
 * i40e_vsi_add_vlan - Add VSI membership for given VLAN
 * @vsi: the VSI being configured
 * @vid: VLAN id to be added
 **/
int i40e_vsi_add_vlan(struct i40e_vsi *vsi, u16 vid)
{
	int err;

	if (vsi->info.pvid)
		return -EINVAL;

	/* The network stack will attempt to add VID=0, with the intention to
	 * receive priority tagged packets with a VLAN of 0. Our HW receives
	 * these packets by default when configured to receive untagged
	 * packets, so we don't need to add a filter for this case.
	 * Additionally, HW interprets adding a VID=0 filter as meaning to
	 * receive *only* tagged traffic and stops receiving untagged traffic.
	 * Thus, we do not want to actually add a filter for VID=0
	 */
	if (!vid)
		return 0;

	/* Locked once because all functions invoked below iterates list*/
	spin_lock_bh(&vsi->mac_filter_hash_lock);
	err = i40e_add_vlan_all_mac(vsi, vid);
	spin_unlock_bh(&vsi->mac_filter_hash_lock);
	if (err)
		return err;

	/* schedule our worker thread which will take care of
	 * applying the new filter changes
	 */
	i40e_service_event_schedule(vsi->back);
	return 0;
}

/**
 * i40e_rm_vlan_all_mac - Remove MAC/VLAN pair for all MAC with the given VLAN
 * @vsi: the vsi being configured
 * @vid: vlan id to be removed (0 = untagged only , -1 = any)
 *
 * This function should be used to remove all VLAN filters which match the
 * given VID. It does not schedule the service event and does not take the
 * mac_filter_hash_lock so it may be combined with other operations under
 * a single invocation of the mac_filter_hash_lock.
 *
 * NOTE: this function expects to be called while under the
 * mac_filter_hash_lock
 */
void i40e_rm_vlan_all_mac(struct i40e_vsi *vsi, s16 vid)
{
	struct i40e_mac_filter *f;
	struct hlist_node *h;
	int bkt;

	hash_for_each_safe(vsi->mac_filter_hash, bkt, h, f, hlist) {
		if (f->vlan == vid)
			__i40e_del_filter(vsi, f);
	}
}

/**
 * i40e_vsi_kill_vlan - Remove VSI membership for given VLAN
 * @vsi: the VSI being configured
 * @vid: VLAN id to be removed
 **/
void i40e_vsi_kill_vlan(struct i40e_vsi *vsi, u16 vid)
{
	if (!vid || vsi->info.pvid)
		return;

	spin_lock_bh(&vsi->mac_filter_hash_lock);
	i40e_rm_vlan_all_mac(vsi, vid);
	spin_unlock_bh(&vsi->mac_filter_hash_lock);

	/* schedule our worker thread which will take care of
	 * applying the new filter changes
	 */
	i40e_service_event_schedule(vsi->back);
}

/**
 * i40e_vlan_rx_add_vid - Add a vlan id filter to HW offload
 * @netdev: network interface to be adjusted
 * @proto: unused protocol value
 * @vid: vlan id to be added
 *
 * net_device_ops implementation for adding vlan ids
 **/
static int i40e_vlan_rx_add_vid(struct net_device *netdev,
				__always_unused __be16 proto, u16 vid)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;
	int ret = 0;

	if (vid >= VLAN_N_VID)
		return -EINVAL;

	ret = i40e_vsi_add_vlan(vsi, vid);
	if (!ret)
		set_bit(vid, vsi->active_vlans);

	return ret;
}

/**
 * i40e_vlan_rx_add_vid_up - Add a vlan id filter to HW offload in UP path
 * @netdev: network interface to be adjusted
 * @proto: unused protocol value
 * @vid: vlan id to be added
 **/
static void i40e_vlan_rx_add_vid_up(struct net_device *netdev,
				    __always_unused __be16 proto, u16 vid)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;

	if (vid >= VLAN_N_VID)
		return;
	set_bit(vid, vsi->active_vlans);
}

/**
 * i40e_vlan_rx_kill_vid - Remove a vlan id filter from HW offload
 * @netdev: network interface to be adjusted
 * @proto: unused protocol value
 * @vid: vlan id to be removed
 *
 * net_device_ops implementation for removing vlan ids
 **/
static int i40e_vlan_rx_kill_vid(struct net_device *netdev,
				 __always_unused __be16 proto, u16 vid)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;

	/* return code is ignored as there is nothing a user
	 * can do about failure to remove and a log message was
	 * already printed from the other function
	 */
	i40e_vsi_kill_vlan(vsi, vid);

	clear_bit(vid, vsi->active_vlans);

	return 0;
}

/**
 * i40e_restore_vlan - Reinstate vlans when vsi/netdev comes back up
 * @vsi: the vsi being brought back up
 **/
static void i40e_restore_vlan(struct i40e_vsi *vsi)
{
	u16 vid;

	if (!vsi->netdev)
		return;

	if (vsi->netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
		i40e_vlan_stripping_enable(vsi);
	else
		i40e_vlan_stripping_disable(vsi);

	for_each_set_bit(vid, vsi->active_vlans, VLAN_N_VID)
		i40e_vlan_rx_add_vid_up(vsi->netdev, htons(ETH_P_8021Q),
					vid);
}

/**
 * i40e_vsi_add_pvid - Add pvid for the VSI
 * @vsi: the vsi being adjusted
 * @vid: the vlan id to set as a PVID
 **/
int i40e_vsi_add_pvid(struct i40e_vsi *vsi, u16 vid)
{
	struct i40e_vsi_context ctxt;
	int ret;

	vsi->info.valid_sections = cpu_to_le16(I40E_AQ_VSI_PROP_VLAN_VALID);
	vsi->info.pvid = cpu_to_le16(vid);
	vsi->info.port_vlan_flags = I40E_AQ_VSI_PVLAN_MODE_TAGGED |
				    I40E_AQ_VSI_PVLAN_INSERT_PVID |
				    I40E_AQ_VSI_PVLAN_EMOD_STR;

	ctxt.seid = vsi->seid;
	ctxt.info = vsi->info;
	ret = i40e_aq_update_vsi_params(&vsi->back->hw, &ctxt, NULL);
	if (ret) {
		dev_info(&vsi->back->pdev->dev,
			 "add pvid failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&vsi->back->hw,
				     vsi->back->hw.aq.asq_last_status));
		return -ENOENT;
	}

	return 0;
}

/**
 * i40e_vsi_remove_pvid - Remove the pvid from the VSI
 * @vsi: the vsi being adjusted
 *
 * Just use the vlan_rx_register() service to put it back to normal
 **/
void i40e_vsi_remove_pvid(struct i40e_vsi *vsi)
{
	vsi->info.pvid = 0;

	i40e_vlan_stripping_disable(vsi);
}

/**
 * i40e_vsi_setup_tx_resources - Allocate VSI Tx queue resources
 * @vsi: ptr to the VSI
 *
 * If this function returns with an error, then it's possible one or
 * more of the rings is populated (while the rest are not).  It is the
 * callers duty to clean those orphaned rings.
 *
 * Return 0 on success, negative on failure
 **/
static int i40e_vsi_setup_tx_resources(struct i40e_vsi *vsi)
{
	int i, err = 0;

	for (i = 0; i < vsi->num_queue_pairs && !err; i++)
		err = i40e_setup_tx_descriptors(vsi->tx_rings[i]);

	if (!i40e_enabled_xdp_vsi(vsi))
		return err;

	for (i = 0; i < vsi->num_queue_pairs && !err; i++)
		err = i40e_setup_tx_descriptors(vsi->xdp_rings[i]);

	return err;
}

/**
 * i40e_vsi_free_tx_resources - Free Tx resources for VSI queues
 * @vsi: ptr to the VSI
 *
 * Free VSI's transmit software resources
 **/
static void i40e_vsi_free_tx_resources(struct i40e_vsi *vsi)
{
	int i;

	if (vsi->tx_rings) {
		for (i = 0; i < vsi->num_queue_pairs; i++)
			if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc)
				i40e_free_tx_resources(vsi->tx_rings[i]);
	}

	if (vsi->xdp_rings) {
		for (i = 0; i < vsi->num_queue_pairs; i++)
			if (vsi->xdp_rings[i] && vsi->xdp_rings[i]->desc)
				i40e_free_tx_resources(vsi->xdp_rings[i]);
	}
}

/**
 * i40e_vsi_setup_rx_resources - Allocate VSI queues Rx resources
 * @vsi: ptr to the VSI
 *
 * If this function returns with an error, then it's possible one or
 * more of the rings is populated (while the rest are not).  It is the
 * callers duty to clean those orphaned rings.
 *
 * Return 0 on success, negative on failure
 **/
static int i40e_vsi_setup_rx_resources(struct i40e_vsi *vsi)
{
	int i, err = 0;

	for (i = 0; i < vsi->num_queue_pairs && !err; i++)
		err = i40e_setup_rx_descriptors(vsi->rx_rings[i]);
	return err;
}

/**
 * i40e_vsi_free_rx_resources - Free Rx Resources for VSI queues
 * @vsi: ptr to the VSI
 *
 * Free all receive software resources
 **/
static void i40e_vsi_free_rx_resources(struct i40e_vsi *vsi)
{
	int i;

	if (!vsi->rx_rings)
		return;

	for (i = 0; i < vsi->num_queue_pairs; i++)
		if (vsi->rx_rings[i] && vsi->rx_rings[i]->desc)
			i40e_free_rx_resources(vsi->rx_rings[i]);
}

/**
 * i40e_config_xps_tx_ring - Configure XPS for a Tx ring
 * @ring: The Tx ring to configure
 *
 * This enables/disables XPS for a given Tx descriptor ring
 * based on the TCs enabled for the VSI that ring belongs to.
 **/
static void i40e_config_xps_tx_ring(struct i40e_ring *ring)
{
	int cpu;

	if (!ring->q_vector || !ring->netdev || ring->ch)
		return;

	/* We only initialize XPS once, so as not to overwrite user settings */
	if (test_and_set_bit(__I40E_TX_XPS_INIT_DONE, ring->state))
		return;

	cpu = cpumask_local_spread(ring->q_vector->v_idx, -1);
	netif_set_xps_queue(ring->netdev, get_cpu_mask(cpu),
			    ring->queue_index);
}

/**
 * i40e_xsk_pool - Retrieve the AF_XDP buffer pool if XDP and ZC is enabled
 * @ring: The Tx or Rx ring
 *
 * Returns the AF_XDP buffer pool or NULL.
 **/
static struct xsk_buff_pool *i40e_xsk_pool(struct i40e_ring *ring)
{
	bool xdp_on = i40e_enabled_xdp_vsi(ring->vsi);
	int qid = ring->queue_index;

	if (ring_is_xdp(ring))
		qid -= ring->vsi->alloc_queue_pairs;

	if (!xdp_on || !test_bit(qid, ring->vsi->af_xdp_zc_qps))
		return NULL;

	return xsk_get_pool_from_qid(ring->vsi->netdev, qid);
}

/**
 * i40e_configure_tx_ring - Configure a transmit ring context and rest
 * @ring: The Tx ring to configure
 *
 * Configure the Tx descriptor ring in the HMC context.
 **/
static int i40e_configure_tx_ring(struct i40e_ring *ring)
{
	struct i40e_vsi *vsi = ring->vsi;
	u16 pf_q = vsi->base_queue + ring->queue_index;
	struct i40e_hw *hw = &vsi->back->hw;
	struct i40e_hmc_obj_txq tx_ctx;
	u32 qtx_ctl = 0;
	int err = 0;

	if (ring_is_xdp(ring))
		ring->xsk_pool = i40e_xsk_pool(ring);

	/* some ATR related tx ring init */
	if (vsi->back->flags & I40E_FLAG_FD_ATR_ENABLED) {
		ring->atr_sample_rate = vsi->back->atr_sample_rate;
		ring->atr_count = 0;
	} else {
		ring->atr_sample_rate = 0;
	}

	/* configure XPS */
	i40e_config_xps_tx_ring(ring);

	/* clear the context structure first */
	memset(&tx_ctx, 0, sizeof(tx_ctx));

	tx_ctx.new_context = 1;
	tx_ctx.base = (ring->dma / 128);
	tx_ctx.qlen = ring->count;
	tx_ctx.fd_ena = !!(vsi->back->flags & (I40E_FLAG_FD_SB_ENABLED |
					       I40E_FLAG_FD_ATR_ENABLED));
	tx_ctx.timesync_ena = !!(vsi->back->flags & I40E_FLAG_PTP);
	/* FDIR VSI tx ring can still use RS bit and writebacks */
	if (vsi->type != I40E_VSI_FDIR)
		tx_ctx.head_wb_ena = 1;
	tx_ctx.head_wb_addr = ring->dma +
			      (ring->count * sizeof(struct i40e_tx_desc));

	/* As part of VSI creation/update, FW allocates certain
	 * Tx arbitration queue sets for each TC enabled for
	 * the VSI. The FW returns the handles to these queue
	 * sets as part of the response buffer to Add VSI,
	 * Update VSI, etc. AQ commands. It is expected that
	 * these queue set handles be associated with the Tx
	 * queues by the driver as part of the TX queue context
	 * initialization. This has to be done regardless of
	 * DCB as by default everything is mapped to TC0.
	 */

	if (ring->ch)
		tx_ctx.rdylist =
			le16_to_cpu(ring->ch->info.qs_handle[ring->dcb_tc]);

	else
		tx_ctx.rdylist = le16_to_cpu(vsi->info.qs_handle[ring->dcb_tc]);

	tx_ctx.rdylist_act = 0;

	/* clear the context in the HMC */
	err = i40e_clear_lan_tx_queue_context(hw, pf_q);
	if (err) {
		dev_info(&vsi->back->pdev->dev,
			 "Failed to clear LAN Tx queue context on Tx ring %d (pf_q %d), error: %d\n",
			 ring->queue_index, pf_q, err);
		return -ENOMEM;
	}

	/* set the context in the HMC */
	err = i40e_set_lan_tx_queue_context(hw, pf_q, &tx_ctx);
	if (err) {
		dev_info(&vsi->back->pdev->dev,
			 "Failed to set LAN Tx queue context on Tx ring %d (pf_q %d, error: %d\n",
			 ring->queue_index, pf_q, err);
		return -ENOMEM;
	}

	/* Now associate this queue with this PCI function */
	if (ring->ch) {
		if (ring->ch->type == I40E_VSI_VMDQ2)
			qtx_ctl = I40E_QTX_CTL_VM_QUEUE;
		else
			return -EINVAL;

		qtx_ctl |= (ring->ch->vsi_number <<
			    I40E_QTX_CTL_VFVM_INDX_SHIFT) &
			    I40E_QTX_CTL_VFVM_INDX_MASK;
	} else {
		if (vsi->type == I40E_VSI_VMDQ2) {
			qtx_ctl = I40E_QTX_CTL_VM_QUEUE;
			qtx_ctl |= ((vsi->id) << I40E_QTX_CTL_VFVM_INDX_SHIFT) &
				    I40E_QTX_CTL_VFVM_INDX_MASK;
		} else {
			qtx_ctl = I40E_QTX_CTL_PF_QUEUE;
		}
	}

	qtx_ctl |= ((hw->pf_id << I40E_QTX_CTL_PF_INDX_SHIFT) &
		    I40E_QTX_CTL_PF_INDX_MASK);
	wr32(hw, I40E_QTX_CTL(pf_q), qtx_ctl);
	i40e_flush(hw);

	/* cache tail off for easier writes later */
	ring->tail = hw->hw_addr + I40E_QTX_TAIL(pf_q);

	return 0;
}

/**
 * i40e_rx_offset - Return expected offset into page to access data
 * @rx_ring: Ring we are requesting offset of
 *
 * Returns the offset value for ring into the data buffer.
 */
static unsigned int i40e_rx_offset(struct i40e_ring *rx_ring)
{
	return ring_uses_build_skb(rx_ring) ? I40E_SKB_PAD : 0;
}

/**
 * i40e_configure_rx_ring - Configure a receive ring context
 * @ring: The Rx ring to configure
 *
 * Configure the Rx descriptor ring in the HMC context.
 **/
static int i40e_configure_rx_ring(struct i40e_ring *ring)
{
	struct i40e_vsi *vsi = ring->vsi;
	u32 chain_len = vsi->back->hw.func_caps.rx_buf_chain_len;
	u16 pf_q = vsi->base_queue + ring->queue_index;
	struct i40e_hw *hw = &vsi->back->hw;
	struct i40e_hmc_obj_rxq rx_ctx;
	int err = 0;
	bool ok;
	int ret;

	bitmap_zero(ring->state, __I40E_RING_STATE_NBITS);

	/* clear the context structure first */
	memset(&rx_ctx, 0, sizeof(rx_ctx));

	if (ring->vsi->type == I40E_VSI_MAIN)
		xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq);

	ring->xsk_pool = i40e_xsk_pool(ring);
	if (ring->xsk_pool) {
		ring->rx_buf_len =
		  xsk_pool_get_rx_frame_size(ring->xsk_pool);
		/* For AF_XDP ZC, we disallow packets to span on
		 * multiple buffers, thus letting us skip that
		 * handling in the fast-path.
		 */
		chain_len = 1;
		ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
						 MEM_TYPE_XSK_BUFF_POOL,
						 NULL);
		if (ret)
			return ret;
		dev_info(&vsi->back->pdev->dev,
			 "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n",
			 ring->queue_index);

	} else {
		ring->rx_buf_len = vsi->rx_buf_len;
		if (ring->vsi->type == I40E_VSI_MAIN) {
			ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq,
							 MEM_TYPE_PAGE_SHARED,
							 NULL);
			if (ret)
				return ret;
		}
	}

	rx_ctx.dbuff = DIV_ROUND_UP(ring->rx_buf_len,
				    BIT_ULL(I40E_RXQ_CTX_DBUFF_SHIFT));

	rx_ctx.base = (ring->dma / 128);
	rx_ctx.qlen = ring->count;

	/* use 16 byte descriptors */
	rx_ctx.dsize = 0;

	/* descriptor type is always zero
	 * rx_ctx.dtype = 0;
	 */
	rx_ctx.hsplit_0 = 0;

	rx_ctx.rxmax = min_t(u16, vsi->max_frame, chain_len * ring->rx_buf_len);
	if (hw->revision_id == 0)
		rx_ctx.lrxqthresh = 0;
	else
		rx_ctx.lrxqthresh = 1;
	rx_ctx.crcstrip = 1;
	rx_ctx.l2tsel = 1;
	/* this controls whether VLAN is stripped from inner headers */
	rx_ctx.showiv = 0;
	/* set the prefena field to 1 because the manual says to */
	rx_ctx.prefena = 1;

	/* clear the context in the HMC */
	err = i40e_clear_lan_rx_queue_context(hw, pf_q);
	if (err) {
		dev_info(&vsi->back->pdev->dev,
			 "Failed to clear LAN Rx queue context on Rx ring %d (pf_q %d), error: %d\n",
			 ring->queue_index, pf_q, err);
		return -ENOMEM;
	}

	/* set the context in the HMC */
	err = i40e_set_lan_rx_queue_context(hw, pf_q, &rx_ctx);
	if (err) {
		dev_info(&vsi->back->pdev->dev,
			 "Failed to set LAN Rx queue context on Rx ring %d (pf_q %d), error: %d\n",
			 ring->queue_index, pf_q, err);
		return -ENOMEM;
	}

	/* configure Rx buffer alignment */
	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
		clear_ring_build_skb_enabled(ring);
	else
		set_ring_build_skb_enabled(ring);

	ring->rx_offset = i40e_rx_offset(ring);

	/* cache tail for quicker writes, and clear the reg before use */
	ring->tail = hw->hw_addr + I40E_QRX_TAIL(pf_q);
	writel(0, ring->tail);

	if (ring->xsk_pool) {
		xsk_pool_set_rxq_info(ring->xsk_pool, &ring->xdp_rxq);
		ok = i40e_alloc_rx_buffers_zc(ring, I40E_DESC_UNUSED(ring));
	} else {
		ok = !i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring));
	}
	if (!ok) {
		/* Log this in case the user has forgotten to give the kernel
		 * any buffers, even later in the application.
		 */
		dev_info(&vsi->back->pdev->dev,
			 "Failed to allocate some buffers on %sRx ring %d (pf_q %d)\n",
			 ring->xsk_pool ? "AF_XDP ZC enabled " : "",
			 ring->queue_index, pf_q);
	}

	return 0;
}

/**
 * i40e_vsi_configure_tx - Configure the VSI for Tx
 * @vsi: VSI structure describing this set of rings and resources
 *
 * Configure the Tx VSI for operation.
 **/
static int i40e_vsi_configure_tx(struct i40e_vsi *vsi)
{
	int err = 0;
	u16 i;

	for (i = 0; (i < vsi->num_queue_pairs) && !err; i++)
		err = i40e_configure_tx_ring(vsi->tx_rings[i]);

	if (err || !i40e_enabled_xdp_vsi(vsi))
		return err;

	for (i = 0; (i < vsi->num_queue_pairs) && !err; i++)
		err = i40e_configure_tx_ring(vsi->xdp_rings[i]);

	return err;
}

/**
 * i40e_calculate_vsi_rx_buf_len - Calculates buffer length
 *
 * @vsi: VSI to calculate rx_buf_len from
 */
static u16 i40e_calculate_vsi_rx_buf_len(struct i40e_vsi *vsi)
{
	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
		return I40E_RXBUFFER_2048;

#if (PAGE_SIZE < 8192)
	if (!I40E_2K_TOO_SMALL_WITH_PADDING && vsi->netdev->mtu <= ETH_DATA_LEN)
		return I40E_RXBUFFER_1536 - NET_IP_ALIGN;
#endif

	return PAGE_SIZE < 8192 ? I40E_RXBUFFER_3072 : I40E_RXBUFFER_2048;
}

/**
 * i40e_vsi_configure_rx - Configure the VSI for Rx
 * @vsi: the VSI being configured
 *
 * Configure the Rx VSI for operation.
 **/
static int i40e_vsi_configure_rx(struct i40e_vsi *vsi)
{
	int err = 0;
	u16 i;

	vsi->max_frame = I40E_MAX_RXBUFFER;
	vsi->rx_buf_len = i40e_calculate_vsi_rx_buf_len(vsi);

#if (PAGE_SIZE < 8192)
	if (vsi->netdev && !I40E_2K_TOO_SMALL_WITH_PADDING &&
	    vsi->netdev->mtu <= ETH_DATA_LEN)
		vsi->max_frame = I40E_RXBUFFER_1536 - NET_IP_ALIGN;
#endif

	/* set up individual rings */
	for (i = 0; i < vsi->num_queue_pairs && !err; i++)
		err = i40e_configure_rx_ring(vsi->rx_rings[i]);

	return err;
}

/**
 * i40e_vsi_config_dcb_rings - Update rings to reflect DCB TC
 * @vsi: ptr to the VSI
 **/
static void i40e_vsi_config_dcb_rings(struct i40e_vsi *vsi)
{
	struct i40e_ring *tx_ring, *rx_ring;
	u16 qoffset, qcount;
	int i, n;

	if (!(vsi->back->flags & I40E_FLAG_DCB_ENABLED)) {
		/* Reset the TC information */
		for (i = 0; i < vsi->num_queue_pairs; i++) {
			rx_ring = vsi->rx_rings[i];
			tx_ring = vsi->tx_rings[i];
			rx_ring->dcb_tc = 0;
			tx_ring->dcb_tc = 0;
		}
		return;
	}

	for (n = 0; n < I40E_MAX_TRAFFIC_CLASS; n++) {
		if (!(vsi->tc_config.enabled_tc & BIT_ULL(n)))
			continue;

		qoffset = vsi->tc_config.tc_info[n].qoffset;
		qcount = vsi->tc_config.tc_info[n].qcount;
		for (i = qoffset; i < (qoffset + qcount); i++) {
			rx_ring = vsi->rx_rings[i];
			tx_ring = vsi->tx_rings[i];
			rx_ring->dcb_tc = n;
			tx_ring->dcb_tc = n;
		}
	}
}

/**
 * i40e_set_vsi_rx_mode - Call set_rx_mode on a VSI
 * @vsi: ptr to the VSI
 **/
static void i40e_set_vsi_rx_mode(struct i40e_vsi *vsi)
{
	if (vsi->netdev)
		i40e_set_rx_mode(vsi->netdev);
}

/**
 * i40e_reset_fdir_filter_cnt - Reset flow director filter counters
 * @pf: Pointer to the targeted PF
 *
 * Set all flow director counters to 0.
 */
static void i40e_reset_fdir_filter_cnt(struct i40e_pf *pf)
{
	pf->fd_tcp4_filter_cnt = 0;
	pf->fd_udp4_filter_cnt = 0;
	pf->fd_sctp4_filter_cnt = 0;
	pf->fd_ip4_filter_cnt = 0;
	pf->fd_tcp6_filter_cnt = 0;
	pf->fd_udp6_filter_cnt = 0;
	pf->fd_sctp6_filter_cnt = 0;
	pf->fd_ip6_filter_cnt = 0;
}

/**
 * i40e_fdir_filter_restore - Restore the Sideband Flow Director filters
 * @vsi: Pointer to the targeted VSI
 *
 * This function replays the hlist on the hw where all the SB Flow Director
 * filters were saved.
 **/
static void i40e_fdir_filter_restore(struct i40e_vsi *vsi)
{
	struct i40e_fdir_filter *filter;
	struct i40e_pf *pf = vsi->back;
	struct hlist_node *node;

	if (!(pf->flags & I40E_FLAG_FD_SB_ENABLED))
		return;

	/* Reset FDir counters as we're replaying all existing filters */
	i40e_reset_fdir_filter_cnt(pf);

	hlist_for_each_entry_safe(filter, node,
				  &pf->fdir_filter_list, fdir_node) {
		i40e_add_del_fdir(vsi, filter, true);
	}
}

/**
 * i40e_vsi_configure - Set up the VSI for action
 * @vsi: the VSI being configured
 **/
static int i40e_vsi_configure(struct i40e_vsi *vsi)
{
	int err;

	i40e_set_vsi_rx_mode(vsi);
	i40e_restore_vlan(vsi);
	i40e_vsi_config_dcb_rings(vsi);
	err = i40e_vsi_configure_tx(vsi);
	if (!err)
		err = i40e_vsi_configure_rx(vsi);

	return err;
}

/**
 * i40e_vsi_configure_msix - MSIX mode Interrupt Config in the HW
 * @vsi: the VSI being configured
 **/
static void i40e_vsi_configure_msix(struct i40e_vsi *vsi)
{
	bool has_xdp = i40e_enabled_xdp_vsi(vsi);
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	u16 vector;
	int i, q;
	u32 qp;

	/* The interrupt indexing is offset by 1 in the PFINT_ITRn
	 * and PFINT_LNKLSTn registers, e.g.:
	 *   PFINT_ITRn[0..n-1] gets msix-1..msix-n  (qpair interrupts)
	 */
	qp = vsi->base_queue;
	vector = vsi->base_vector;
	for (i = 0; i < vsi->num_q_vectors; i++, vector++) {
		struct i40e_q_vector *q_vector = vsi->q_vectors[i];

		q_vector->rx.next_update = jiffies + 1;
		q_vector->rx.target_itr =
			ITR_TO_REG(vsi->rx_rings[i]->itr_setting);
		wr32(hw, I40E_PFINT_ITRN(I40E_RX_ITR, vector - 1),
		     q_vector->rx.target_itr >> 1);
		q_vector->rx.current_itr = q_vector->rx.target_itr;

		q_vector->tx.next_update = jiffies + 1;
		q_vector->tx.target_itr =
			ITR_TO_REG(vsi->tx_rings[i]->itr_setting);
		wr32(hw, I40E_PFINT_ITRN(I40E_TX_ITR, vector - 1),
		     q_vector->tx.target_itr >> 1);
		q_vector->tx.current_itr = q_vector->tx.target_itr;

		wr32(hw, I40E_PFINT_RATEN(vector - 1),
		     i40e_intrl_usec_to_reg(vsi->int_rate_limit));

		/* begin of linked list for RX queue assigned to this vector */
		wr32(hw, I40E_PFINT_LNKLSTN(vector - 1), qp);
		for (q = 0; q < q_vector->num_ringpairs; q++) {
			u32 nextqp = has_xdp ? qp + vsi->alloc_queue_pairs : qp;
			u32 val;

			val = I40E_QINT_RQCTL_CAUSE_ENA_MASK |
			      (I40E_RX_ITR << I40E_QINT_RQCTL_ITR_INDX_SHIFT) |
			      (vector << I40E_QINT_RQCTL_MSIX_INDX_SHIFT) |
			      (nextqp << I40E_QINT_RQCTL_NEXTQ_INDX_SHIFT) |
			      (I40E_QUEUE_TYPE_TX <<
			       I40E_QINT_RQCTL_NEXTQ_TYPE_SHIFT);

			wr32(hw, I40E_QINT_RQCTL(qp), val);

			if (has_xdp) {
				/* TX queue with next queue set to TX */
				val = I40E_QINT_TQCTL_CAUSE_ENA_MASK |
				      (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) |
				      (vector << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) |
				      (qp << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT) |
				      (I40E_QUEUE_TYPE_TX <<
				       I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT);

				wr32(hw, I40E_QINT_TQCTL(nextqp), val);
			}
			/* TX queue with next RX or end of linked list */
			val = I40E_QINT_TQCTL_CAUSE_ENA_MASK |
			      (I40E_TX_ITR << I40E_QINT_TQCTL_ITR_INDX_SHIFT) |
			      (vector << I40E_QINT_TQCTL_MSIX_INDX_SHIFT) |
			      ((qp + 1) << I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT) |
			      (I40E_QUEUE_TYPE_RX <<
			       I40E_QINT_TQCTL_NEXTQ_TYPE_SHIFT);

			/* Terminate the linked list */
			if (q == (q_vector->num_ringpairs - 1))
				val |= (I40E_QUEUE_END_OF_LIST <<
					I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT);

			wr32(hw, I40E_QINT_TQCTL(qp), val);
			qp++;
		}
	}

	i40e_flush(hw);
}

/**
 * i40e_enable_misc_int_causes - enable the non-queue interrupts
 * @pf: pointer to private device data structure
 **/
static void i40e_enable_misc_int_causes(struct i40e_pf *pf)
{
	struct i40e_hw *hw = &pf->hw;
	u32 val;

	/* clear things first */
	wr32(hw, I40E_PFINT_ICR0_ENA, 0);  /* disable all */
	rd32(hw, I40E_PFINT_ICR0);         /* read to clear */

	val = I40E_PFINT_ICR0_ENA_ECC_ERR_MASK       |
	      I40E_PFINT_ICR0_ENA_MAL_DETECT_MASK    |
	      I40E_PFINT_ICR0_ENA_GRST_MASK          |
	      I40E_PFINT_ICR0_ENA_PCI_EXCEPTION_MASK |
	      I40E_PFINT_ICR0_ENA_GPIO_MASK          |
	      I40E_PFINT_ICR0_ENA_HMC_ERR_MASK       |
	      I40E_PFINT_ICR0_ENA_VFLR_MASK          |
	      I40E_PFINT_ICR0_ENA_ADMINQ_MASK;

	if (pf->flags & I40E_FLAG_IWARP_ENABLED)
		val |= I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK;

	if (pf->flags & I40E_FLAG_PTP)
		val |= I40E_PFINT_ICR0_ENA_TIMESYNC_MASK;

	wr32(hw, I40E_PFINT_ICR0_ENA, val);

	/* SW_ITR_IDX = 0, but don't change INTENA */
	wr32(hw, I40E_PFINT_DYN_CTL0, I40E_PFINT_DYN_CTL0_SW_ITR_INDX_MASK |
					I40E_PFINT_DYN_CTL0_INTENA_MSK_MASK);

	/* OTHER_ITR_IDX = 0 */
	wr32(hw, I40E_PFINT_STAT_CTL0, 0);
}

/**
 * i40e_configure_msi_and_legacy - Legacy mode interrupt config in the HW
 * @vsi: the VSI being configured
 **/
static void i40e_configure_msi_and_legacy(struct i40e_vsi *vsi)
{
	u32 nextqp = i40e_enabled_xdp_vsi(vsi) ? vsi->alloc_queue_pairs : 0;
	struct i40e_q_vector *q_vector = vsi->q_vectors[0];
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;

	/* set the ITR configuration */
	q_vector->rx.next_update = jiffies + 1;
	q_vector->rx.target_itr = ITR_TO_REG(vsi->rx_rings[0]->itr_setting);
	wr32(hw, I40E_PFINT_ITR0(I40E_RX_ITR), q_vector->rx.target_itr >> 1);
	q_vector->rx.current_itr = q_vector->rx.target_itr;
	q_vector->tx.next_update = jiffies + 1;
	q_vector->tx.target_itr = ITR_TO_REG(vsi->tx_rings[0]->itr_setting);
	wr32(hw, I40E_PFINT_ITR0(I40E_TX_ITR), q_vector->tx.target_itr >> 1);
	q_vector->tx.current_itr = q_vector->tx.target_itr;

	i40e_enable_misc_int_causes(pf);

	/* FIRSTQ_INDX = 0, FIRSTQ_TYPE = 0 (rx) */
	wr32(hw, I40E_PFINT_LNKLST0, 0);

	/* Associate the queue pair to the vector and enable the queue
	 * interrupt RX queue in linked list with next queue set to TX
	 */
	wr32(hw, I40E_QINT_RQCTL(0), I40E_QINT_RQCTL_VAL(nextqp, 0, TX));

	if (i40e_enabled_xdp_vsi(vsi)) {
		/* TX queue in linked list with next queue set to TX */
		wr32(hw, I40E_QINT_TQCTL(nextqp),
		     I40E_QINT_TQCTL_VAL(nextqp, 0, TX));
	}

	/* last TX queue so the next RX queue doesn't matter */
	wr32(hw, I40E_QINT_TQCTL(0),
	     I40E_QINT_TQCTL_VAL(I40E_QUEUE_END_OF_LIST, 0, RX));
	i40e_flush(hw);
}

/**
 * i40e_irq_dynamic_disable_icr0 - Disable default interrupt generation for icr0
 * @pf: board private structure
 **/
void i40e_irq_dynamic_disable_icr0(struct i40e_pf *pf)
{
	struct i40e_hw *hw = &pf->hw;

	wr32(hw, I40E_PFINT_DYN_CTL0,
	     I40E_ITR_NONE << I40E_PFINT_DYN_CTLN_ITR_INDX_SHIFT);
	i40e_flush(hw);
}

/**
 * i40e_irq_dynamic_enable_icr0 - Enable default interrupt generation for icr0
 * @pf: board private structure
 **/
void i40e_irq_dynamic_enable_icr0(struct i40e_pf *pf)
{
	struct i40e_hw *hw = &pf->hw;
	u32 val;

	val = I40E_PFINT_DYN_CTL0_INTENA_MASK   |
	      I40E_PFINT_DYN_CTL0_CLEARPBA_MASK |
	      (I40E_ITR_NONE << I40E_PFINT_DYN_CTL0_ITR_INDX_SHIFT);

	wr32(hw, I40E_PFINT_DYN_CTL0, val);
	i40e_flush(hw);
}

/**
 * i40e_msix_clean_rings - MSIX mode Interrupt Handler
 * @irq: interrupt number
 * @data: pointer to a q_vector
 **/
static irqreturn_t i40e_msix_clean_rings(int irq, void *data)
{
	struct i40e_q_vector *q_vector = data;

	if (!q_vector->tx.ring && !q_vector->rx.ring)
		return IRQ_HANDLED;

	napi_schedule_irqoff(&q_vector->napi);

	return IRQ_HANDLED;
}

/**
 * i40e_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 i40e_irq_affinity_notify(struct irq_affinity_notify *notify,
				     const cpumask_t *mask)
{
	struct i40e_q_vector *q_vector =
		container_of(notify, struct i40e_q_vector, affinity_notify);

	cpumask_copy(&q_vector->affinity_mask, mask);
}

/**
 * i40e_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 i40e_irq_affinity_release(struct kref *ref) {}

/**
 * i40e_vsi_request_irq_msix - Initialize MSI-X interrupts
 * @vsi: the VSI being configured
 * @basename: name for the vector
 *
 * Allocates MSI-X vectors and requests interrupts from the kernel.
 **/
static int i40e_vsi_request_irq_msix(struct i40e_vsi *vsi, char *basename)
{
	int q_vectors = vsi->num_q_vectors;
	struct i40e_pf *pf = vsi->back;
	int base = vsi->base_vector;
	int rx_int_idx = 0;
	int tx_int_idx = 0;
	int vector, err;
	int irq_num;
	int cpu;

	for (vector = 0; vector < q_vectors; vector++) {
		struct i40e_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;
		}
		err = request_irq(irq_num,
				  vsi->irq_handler,
				  0,
				  q_vector->name,
				  q_vector);
		if (err) {
			dev_info(&pf->pdev->dev,
				 "MSIX request_irq failed, error: %d\n", err);
			goto free_queue_irqs;
		}

		/* register for affinity change notifications */
		q_vector->irq_num = irq_num;
		q_vector->affinity_notify.notify = i40e_irq_affinity_notify;
		q_vector->affinity_notify.release = i40e_irq_affinity_release;
		irq_set_affinity_notifier(irq_num, &q_vector->affinity_notify);
		/* Spread affinity hints out across online CPUs.
		 *
		 * get_cpu_mask returns a static constant mask with
		 * a permanent lifetime so it's ok to pass to
		 * irq_update_affinity_hint without making a copy.
		 */
		cpu = cpumask_local_spread(q_vector->v_idx, -1);
		irq_update_affinity_hint(irq_num, get_cpu_mask(cpu));
	}

	vsi->irqs_ready = true;
	return 0;

free_queue_irqs:
	while (vector) {
		vector--;
		irq_num = pf->msix_entries[base + vector].vector;
		irq_set_affinity_notifier(irq_num, NULL);
		irq_update_affinity_hint(irq_num, NULL);
		free_irq(irq_num, &vsi->q_vectors[vector]);
	}
	return err;
}

/**
 * i40e_vsi_disable_irq - Mask off queue interrupt generation on the VSI
 * @vsi: the VSI being un-configured
 **/
static void i40e_vsi_disable_irq(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	int base = vsi->base_vector;
	int i;

	/* disable interrupt causation from each queue */
	for (i = 0; i < vsi->num_queue_pairs; i++) {
		u32 val;

		val = rd32(hw, I40E_QINT_TQCTL(vsi->tx_rings[i]->reg_idx));
		val &= ~I40E_QINT_TQCTL_CAUSE_ENA_MASK;
		wr32(hw, I40E_QINT_TQCTL(vsi->tx_rings[i]->reg_idx), val);

		val = rd32(hw, I40E_QINT_RQCTL(vsi->rx_rings[i]->reg_idx));
		val &= ~I40E_QINT_RQCTL_CAUSE_ENA_MASK;
		wr32(hw, I40E_QINT_RQCTL(vsi->rx_rings[i]->reg_idx), val);

		if (!i40e_enabled_xdp_vsi(vsi))
			continue;
		wr32(hw, I40E_QINT_TQCTL(vsi->xdp_rings[i]->reg_idx), 0);
	}

	/* disable each interrupt */
	if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
		for (i = vsi->base_vector;
		     i < (vsi->num_q_vectors + vsi->base_vector); i++)
			wr32(hw, I40E_PFINT_DYN_CTLN(i - 1), 0);

		i40e_flush(hw);
		for (i = 0; i < vsi->num_q_vectors; i++)
			synchronize_irq(pf->msix_entries[i + base].vector);
	} else {
		/* Legacy and MSI mode - this stops all interrupt handling */
		wr32(hw, I40E_PFINT_ICR0_ENA, 0);
		wr32(hw, I40E_PFINT_DYN_CTL0, 0);
		i40e_flush(hw);
		synchronize_irq(pf->pdev->irq);
	}
}

/**
 * i40e_vsi_enable_irq - Enable IRQ for the given VSI
 * @vsi: the VSI being configured
 **/
static int i40e_vsi_enable_irq(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	int i;

	if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
		for (i = 0; i < vsi->num_q_vectors; i++)
			i40e_irq_dynamic_enable(vsi, i);
	} else {
		i40e_irq_dynamic_enable_icr0(pf);
	}

	i40e_flush(&pf->hw);
	return 0;
}

/**
 * i40e_free_misc_vector - Free the vector that handles non-queue events
 * @pf: board private structure
 **/
static void i40e_free_misc_vector(struct i40e_pf *pf)
{
	/* Disable ICR 0 */
	wr32(&pf->hw, I40E_PFINT_ICR0_ENA, 0);
	i40e_flush(&pf->hw);

	if (pf->flags & I40E_FLAG_MSIX_ENABLED && pf->msix_entries) {
		free_irq(pf->msix_entries[0].vector, pf);
		clear_bit(__I40E_MISC_IRQ_REQUESTED, pf->state);
	}
}

/**
 * i40e_intr - MSI/Legacy and non-queue interrupt handler
 * @irq: interrupt number
 * @data: pointer to a q_vector
 *
 * This is the handler used for all MSI/Legacy interrupts, and deals
 * with both queue and non-queue interrupts.  This is also used in
 * MSIX mode to handle the non-queue interrupts.
 **/
static irqreturn_t i40e_intr(int irq, void *data)
{
	struct i40e_pf *pf = (struct i40e_pf *)data;
	struct i40e_hw *hw = &pf->hw;
	irqreturn_t ret = IRQ_NONE;
	u32 icr0, icr0_remaining;
	u32 val, ena_mask;

	icr0 = rd32(hw, I40E_PFINT_ICR0);
	ena_mask = rd32(hw, I40E_PFINT_ICR0_ENA);

	/* if sharing a legacy IRQ, we might get called w/o an intr pending */
	if ((icr0 & I40E_PFINT_ICR0_INTEVENT_MASK) == 0)
		goto enable_intr;

	/* if interrupt but no bits showing, must be SWINT */
	if (((icr0 & ~I40E_PFINT_ICR0_INTEVENT_MASK) == 0) ||
	    (icr0 & I40E_PFINT_ICR0_SWINT_MASK))
		pf->sw_int_count++;

	if ((pf->flags & I40E_FLAG_IWARP_ENABLED) &&
	    (icr0 & I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK)) {
		ena_mask &= ~I40E_PFINT_ICR0_ENA_PE_CRITERR_MASK;
		dev_dbg(&pf->pdev->dev, "cleared PE_CRITERR\n");
		set_bit(__I40E_CORE_RESET_REQUESTED, pf->state);
	}

	/* only q0 is used in MSI/Legacy mode, and none are used in MSIX */
	if (icr0 & I40E_PFINT_ICR0_QUEUE_0_MASK) {
		struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
		struct i40e_q_vector *q_vector = vsi->q_vectors[0];

		/* We do not have a way to disarm Queue causes while leaving
		 * interrupt enabled for all other causes, ideally
		 * interrupt should be disabled while we are in NAPI but
		 * this is not a performance path and napi_schedule()
		 * can deal with rescheduling.
		 */
		if (!test_bit(__I40E_DOWN, pf->state))
			napi_schedule_irqoff(&q_vector->napi);
	}

	if (icr0 & I40E_PFINT_ICR0_ADMINQ_MASK) {
		ena_mask &= ~I40E_PFINT_ICR0_ENA_ADMINQ_MASK;
		set_bit(__I40E_ADMINQ_EVENT_PENDING, pf->state);
		i40e_debug(&pf->hw, I40E_DEBUG_NVM, "AdminQ event\n");
	}

	if (icr0 & I40E_PFINT_ICR0_MAL_DETECT_MASK) {
		ena_mask &= ~I40E_PFINT_ICR0_ENA_MAL_DETECT_MASK;
		set_bit(__I40E_MDD_EVENT_PENDING, pf->state);
	}

	if (icr0 & I40E_PFINT_ICR0_VFLR_MASK) {
		/* disable any further VFLR event notifications */
		if (test_bit(__I40E_VF_RESETS_DISABLED, pf->state)) {
			u32 reg = rd32(hw, I40E_PFINT_ICR0_ENA);

			reg &= ~I40E_PFINT_ICR0_VFLR_MASK;
			wr32(hw, I40E_PFINT_ICR0_ENA, reg);
		} else {
			ena_mask &= ~I40E_PFINT_ICR0_ENA_VFLR_MASK;
			set_bit(__I40E_VFLR_EVENT_PENDING, pf->state);
		}
	}

	if (icr0 & I40E_PFINT_ICR0_GRST_MASK) {
		if (!test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state))
			set_bit(__I40E_RESET_INTR_RECEIVED, pf->state);
		ena_mask &= ~I40E_PFINT_ICR0_ENA_GRST_MASK;
		val = rd32(hw, I40E_GLGEN_RSTAT);
		val = (val & I40E_GLGEN_RSTAT_RESET_TYPE_MASK)
		       >> I40E_GLGEN_RSTAT_RESET_TYPE_SHIFT;
		if (val == I40E_RESET_CORER) {
			pf->corer_count++;
		} else if (val == I40E_RESET_GLOBR) {
			pf->globr_count++;
		} else if (val == I40E_RESET_EMPR) {
			pf->empr_count++;
			set_bit(__I40E_EMP_RESET_INTR_RECEIVED, pf->state);
		}
	}

	if (icr0 & I40E_PFINT_ICR0_HMC_ERR_MASK) {
		icr0 &= ~I40E_PFINT_ICR0_HMC_ERR_MASK;
		dev_info(&pf->pdev->dev, "HMC error interrupt\n");
		dev_info(&pf->pdev->dev, "HMC error info 0x%x, HMC error data 0x%x\n",
			 rd32(hw, I40E_PFHMC_ERRORINFO),
			 rd32(hw, I40E_PFHMC_ERRORDATA));
	}

	if (icr0 & I40E_PFINT_ICR0_TIMESYNC_MASK) {
		u32 prttsyn_stat = rd32(hw, I40E_PRTTSYN_STAT_0);

		if (prttsyn_stat & I40E_PRTTSYN_STAT_0_EVENT0_MASK)
			schedule_work(&pf->ptp_extts0_work);

		if (prttsyn_stat & I40E_PRTTSYN_STAT_0_TXTIME_MASK)
			i40e_ptp_tx_hwtstamp(pf);

		icr0 &= ~I40E_PFINT_ICR0_ENA_TIMESYNC_MASK;
	}

	/* If a critical error is pending we have no choice but to reset the
	 * device.
	 * Report and mask out any remaining unexpected interrupts.
	 */
	icr0_remaining = icr0 & ena_mask;
	if (icr0_remaining) {
		dev_info(&pf->pdev->dev, "unhandled interrupt icr0=0x%08x\n",
			 icr0_remaining);
		if ((icr0_remaining & I40E_PFINT_ICR0_PE_CRITERR_MASK) ||
		    (icr0_remaining & I40E_PFINT_ICR0_PCI_EXCEPTION_MASK) ||
		    (icr0_remaining & I40E_PFINT_ICR0_ECC_ERR_MASK)) {
			dev_info(&pf->pdev->dev, "device will be reset\n");
			set_bit(__I40E_PF_RESET_REQUESTED, pf->state);
			i40e_service_event_schedule(pf);
		}
		ena_mask &= ~icr0_remaining;
	}
	ret = IRQ_HANDLED;

enable_intr:
	/* re-enable interrupt causes */
	wr32(hw, I40E_PFINT_ICR0_ENA, ena_mask);
	if (!test_bit(__I40E_DOWN, pf->state) ||
	    test_bit(__I40E_RECOVERY_MODE, pf->state)) {
		i40e_service_event_schedule(pf);
		i40e_irq_dynamic_enable_icr0(pf);
	}

	return ret;
}

/**
 * i40e_clean_fdir_tx_irq - Reclaim resources after transmit completes
 * @tx_ring:  tx ring to clean
 * @budget:   how many cleans we're allowed
 *
 * Returns true if there's any budget left (e.g. the clean is finished)
 **/
static bool i40e_clean_fdir_tx_irq(struct i40e_ring *tx_ring, int budget)
{
	struct i40e_vsi *vsi = tx_ring->vsi;
	u16 i = tx_ring->next_to_clean;
	struct i40e_tx_buffer *tx_buf;
	struct i40e_tx_desc *tx_desc;

	tx_buf = &tx_ring->tx_bi[i];
	tx_desc = I40E_TX_DESC(tx_ring, i);
	i -= tx_ring->count;

	do {
		struct i40e_tx_desc *eop_desc = tx_buf->next_to_watch;

		/* if next_to_watch is not set then there is no work pending */
		if (!eop_desc)
			break;

		/* prevent any other reads prior to eop_desc */
		smp_rmb();

		/* if the descriptor isn't done, no work yet to do */
		if (!(eop_desc->cmd_type_offset_bsz &
		      cpu_to_le64(I40E_TX_DESC_DTYPE_DESC_DONE)))
			break;

		/* clear next_to_watch to prevent false hangs */
		tx_buf->next_to_watch = NULL;

		tx_desc->buffer_addr = 0;
		tx_desc->cmd_type_offset_bsz = 0;
		/* move past filter desc */
		tx_buf++;
		tx_desc++;
		i++;
		if (unlikely(!i)) {
			i -= tx_ring->count;
			tx_buf = tx_ring->tx_bi;
			tx_desc = I40E_TX_DESC(tx_ring, 0);
		}
		/* unmap skb header data */
		dma_unmap_single(tx_ring->dev,
				 dma_unmap_addr(tx_buf, dma),
				 dma_unmap_len(tx_buf, len),
				 DMA_TO_DEVICE);
		if (tx_buf->tx_flags & I40E_TX_FLAGS_FD_SB)
			kfree(tx_buf->raw_buf);

		tx_buf->raw_buf = NULL;
		tx_buf->tx_flags = 0;
		tx_buf->next_to_watch = NULL;
		dma_unmap_len_set(tx_buf, len, 0);
		tx_desc->buffer_addr = 0;
		tx_desc->cmd_type_offset_bsz = 0;

		/* move us past the eop_desc for start of next FD desc */
		tx_buf++;
		tx_desc++;
		i++;
		if (unlikely(!i)) {
			i -= tx_ring->count;
			tx_buf = tx_ring->tx_bi;
			tx_desc = I40E_TX_DESC(tx_ring, 0);
		}

		/* update budget accounting */
		budget--;
	} while (likely(budget));

	i += tx_ring->count;
	tx_ring->next_to_clean = i;

	if (vsi->back->flags & I40E_FLAG_MSIX_ENABLED)
		i40e_irq_dynamic_enable(vsi, tx_ring->q_vector->v_idx);

	return budget > 0;
}

/**
 * i40e_fdir_clean_ring - Interrupt Handler for FDIR SB ring
 * @irq: interrupt number
 * @data: pointer to a q_vector
 **/
static irqreturn_t i40e_fdir_clean_ring(int irq, void *data)
{
	struct i40e_q_vector *q_vector = data;
	struct i40e_vsi *vsi;

	if (!q_vector->tx.ring)
		return IRQ_HANDLED;

	vsi = q_vector->tx.ring->vsi;
	i40e_clean_fdir_tx_irq(q_vector->tx.ring, vsi->work_limit);

	return IRQ_HANDLED;
}

/**
 * i40e_map_vector_to_qp - Assigns the queue pair to the vector
 * @vsi: the VSI being configured
 * @v_idx: vector index
 * @qp_idx: queue pair index
 **/
static void i40e_map_vector_to_qp(struct i40e_vsi *vsi, int v_idx, int qp_idx)
{
	struct i40e_q_vector *q_vector = vsi->q_vectors[v_idx];
	struct i40e_ring *tx_ring = vsi->tx_rings[qp_idx];
	struct i40e_ring *rx_ring = vsi->rx_rings[qp_idx];

	tx_ring->q_vector = q_vector;
	tx_ring->next = q_vector->tx.ring;
	q_vector->tx.ring = tx_ring;
	q_vector->tx.count++;

	/* Place XDP Tx ring in the same q_vector ring list as regular Tx */
	if (i40e_enabled_xdp_vsi(vsi)) {
		struct i40e_ring *xdp_ring = vsi->xdp_rings[qp_idx];

		xdp_ring->q_vector = q_vector;
		xdp_ring->next = q_vector->tx.ring;
		q_vector->tx.ring = xdp_ring;
		q_vector->tx.count++;
	}

	rx_ring->q_vector = q_vector;
	rx_ring->next = q_vector->rx.ring;
	q_vector->rx.ring = rx_ring;
	q_vector->rx.count++;
}

/**
 * i40e_vsi_map_rings_to_vectors - Maps descriptor rings to vectors
 * @vsi: the VSI being configured
 *
 * This function maps descriptor rings to the queue-specific vectors
 * we were allotted through the MSI-X enabling code.  Ideally, we'd have
 * one vector per queue pair, but on a constrained vector budget, we
 * group the queue pairs as "efficiently" as possible.
 **/
static void i40e_vsi_map_rings_to_vectors(struct i40e_vsi *vsi)
{
	int qp_remaining = vsi->num_queue_pairs;
	int q_vectors = vsi->num_q_vectors;
	int num_ringpairs;
	int v_start = 0;
	int qp_idx = 0;

	/* If we don't have enough vectors for a 1-to-1 mapping, we'll have to
	 * group them so there are multiple queues per vector.
	 * It is also important to go through all the vectors available to be
	 * sure that if we don't use all the vectors, that the remaining vectors
	 * are cleared. This is especially important when decreasing the
	 * number of queues in use.
	 */
	for (; v_start < q_vectors; v_start++) {
		struct i40e_q_vector *q_vector = vsi->q_vectors[v_start];

		num_ringpairs = DIV_ROUND_UP(qp_remaining, q_vectors - v_start);

		q_vector->num_ringpairs = num_ringpairs;
		q_vector->reg_idx = q_vector->v_idx + vsi->base_vector - 1;

		q_vector->rx.count = 0;
		q_vector->tx.count = 0;
		q_vector->rx.ring = NULL;
		q_vector->tx.ring = NULL;

		while (num_ringpairs--) {
			i40e_map_vector_to_qp(vsi, v_start, qp_idx);
			qp_idx++;
			qp_remaining--;
		}
	}
}

/**
 * i40e_vsi_request_irq - Request IRQ from the OS
 * @vsi: the VSI being configured
 * @basename: name for the vector
 **/
static int i40e_vsi_request_irq(struct i40e_vsi *vsi, char *basename)
{
	struct i40e_pf *pf = vsi->back;
	int err;

	if (pf->flags & I40E_FLAG_MSIX_ENABLED)
		err = i40e_vsi_request_irq_msix(vsi, basename);
	else if (pf->flags & I40E_FLAG_MSI_ENABLED)
		err = request_irq(pf->pdev->irq, i40e_intr, 0,
				  pf->int_name, pf);
	else
		err = request_irq(pf->pdev->irq, i40e_intr, IRQF_SHARED,
				  pf->int_name, pf);

	if (err)
		dev_info(&pf->pdev->dev, "request_irq failed, Error %d\n", err);

	return err;
}

#ifdef CONFIG_NET_POLL_CONTROLLER
/**
 * i40e_netpoll - A Polling 'interrupt' handler
 * @netdev: network interface device structure
 *
 * This is used by netconsole to send skbs without having to re-enable
 * interrupts.  It's not called while the normal interrupt routine is executing.
 **/
static void i40e_netpoll(struct net_device *netdev)
{
	struct i40e_netdev_priv *np = netdev_priv(netdev);
	struct i40e_vsi *vsi = np->vsi;
	struct i40e_pf *pf = vsi->back;
	int i;

	/* if interface is down do nothing */
	if (test_bit(__I40E_VSI_DOWN, vsi->state))
		return;

	if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
		for (i = 0; i < vsi->num_q_vectors; i++)
			i40e_msix_clean_rings(0, vsi->q_vectors[i]);
	} else {
		i40e_intr(pf->pdev->irq, netdev);
	}
}
#endif

#define I40E_QTX_ENA_WAIT_COUNT 50

/**
 * i40e_pf_txq_wait - Wait for a PF's Tx queue to be enabled or disabled
 * @pf: the PF being configured
 * @pf_q: the PF queue
 * @enable: enable or disable state of the queue
 *
 * This routine will wait for the given Tx queue of the PF to reach the
 * enabled or disabled state.
 * Returns -ETIMEDOUT in case of failing to reach the requested state after
 * multiple retries; else will return 0 in case of success.
 **/
static int i40e_pf_txq_wait(struct i40e_pf *pf, int pf_q, bool enable)
{
	int i;
	u32 tx_reg;

	for (i = 0; i < I40E_QUEUE_WAIT_RETRY_LIMIT; i++) {
		tx_reg = rd32(&pf->hw, I40E_QTX_ENA(pf_q));
		if (enable == !!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
			break;

		usleep_range(10, 20);
	}
	if (i >= I40E_QUEUE_WAIT_RETRY_LIMIT)
		return -ETIMEDOUT;

	return 0;
}

/**
 * i40e_control_tx_q - Start or stop a particular Tx queue
 * @pf: the PF structure
 * @pf_q: the PF queue to configure
 * @enable: start or stop the queue
 *
 * This function enables or disables a single queue. Note that any delay
 * required after the operation is expected to be handled by the caller of
 * this function.
 **/
static void i40e_control_tx_q(struct i40e_pf *pf, int pf_q, bool enable)
{
	struct i40e_hw *hw = &pf->hw;
	u32 tx_reg;
	int i;

	/* warn the TX unit of coming changes */
	i40e_pre_tx_queue_cfg(&pf->hw, pf_q, enable);
	if (!enable)
		usleep_range(10, 20);

	for (i = 0; i < I40E_QTX_ENA_WAIT_COUNT; i++) {
		tx_reg = rd32(hw, I40E_QTX_ENA(pf_q));
		if (((tx_reg >> I40E_QTX_ENA_QENA_REQ_SHIFT) & 1) ==
		    ((tx_reg >> I40E_QTX_ENA_QENA_STAT_SHIFT) & 1))
			break;
		usleep_range(1000, 2000);
	}

	/* Skip if the queue is already in the requested state */
	if (enable == !!(tx_reg & I40E_QTX_ENA_QENA_STAT_MASK))
		return;

	/* turn on/off the queue */
	if (enable) {
		wr32(hw, I40E_QTX_HEAD(pf_q), 0);
		tx_reg |= I40E_QTX_ENA_QENA_REQ_MASK;
	} else {
		tx_reg &= ~I40E_QTX_ENA_QENA_REQ_MASK;
	}

	wr32(hw, I40E_QTX_ENA(pf_q), tx_reg);
}

/**
 * i40e_control_wait_tx_q - Start/stop Tx queue and wait for completion
 * @seid: VSI SEID
 * @pf: the PF structure
 * @pf_q: the PF queue to configure
 * @is_xdp: true if the queue is used for XDP
 * @enable: start or stop the queue
 **/
int i40e_control_wait_tx_q(int seid, struct i40e_pf *pf, int pf_q,
			   bool is_xdp, bool enable)
{
	int ret;

	i40e_control_tx_q(pf, pf_q, enable);

	/* wait for the change to finish */
	ret = i40e_pf_txq_wait(pf, pf_q, enable);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "VSI seid %d %sTx ring %d %sable timeout\n",
			 seid, (is_xdp ? "XDP " : ""), pf_q,
			 (enable ? "en" : "dis"));
	}

	return ret;
}

/**
 * i40e_vsi_enable_tx - Start a VSI's rings
 * @vsi: the VSI being configured
 **/
static int i40e_vsi_enable_tx(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	int i, pf_q, ret = 0;

	pf_q = vsi->base_queue;
	for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
		ret = i40e_control_wait_tx_q(vsi->seid, pf,
					     pf_q,
					     false /*is xdp*/, true);
		if (ret)
			break;

		if (!i40e_enabled_xdp_vsi(vsi))
			continue;

		ret = i40e_control_wait_tx_q(vsi->seid, pf,
					     pf_q + vsi->alloc_queue_pairs,
					     true /*is xdp*/, true);
		if (ret)
			break;
	}
	return ret;
}

/**
 * i40e_pf_rxq_wait - Wait for a PF's Rx queue to be enabled or disabled
 * @pf: the PF being configured
 * @pf_q: the PF queue
 * @enable: enable or disable state of the queue
 *
 * This routine will wait for the given Rx queue of the PF to reach the
 * enabled or disabled state.
 * Returns -ETIMEDOUT in case of failing to reach the requested state after
 * multiple retries; else will return 0 in case of success.
 **/
static int i40e_pf_rxq_wait(struct i40e_pf *pf, int pf_q, bool enable)
{
	int i;
	u32 rx_reg;

	for (i = 0; i < I40E_QUEUE_WAIT_RETRY_LIMIT; i++) {
		rx_reg = rd32(&pf->hw, I40E_QRX_ENA(pf_q));
		if (enable == !!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
			break;

		usleep_range(10, 20);
	}
	if (i >= I40E_QUEUE_WAIT_RETRY_LIMIT)
		return -ETIMEDOUT;

	return 0;
}

/**
 * i40e_control_rx_q - Start or stop a particular Rx queue
 * @pf: the PF structure
 * @pf_q: the PF queue to configure
 * @enable: start or stop the queue
 *
 * This function enables or disables a single queue. Note that
 * any delay required after the operation is expected to be
 * handled by the caller of this function.
 **/
static void i40e_control_rx_q(struct i40e_pf *pf, int pf_q, bool enable)
{
	struct i40e_hw *hw = &pf->hw;
	u32 rx_reg;
	int i;

	for (i = 0; i < I40E_QTX_ENA_WAIT_COUNT; i++) {
		rx_reg = rd32(hw, I40E_QRX_ENA(pf_q));
		if (((rx_reg >> I40E_QRX_ENA_QENA_REQ_SHIFT) & 1) ==
		    ((rx_reg >> I40E_QRX_ENA_QENA_STAT_SHIFT) & 1))
			break;
		usleep_range(1000, 2000);
	}

	/* Skip if the queue is already in the requested state */
	if (enable == !!(rx_reg & I40E_QRX_ENA_QENA_STAT_MASK))
		return;

	/* turn on/off the queue */
	if (enable)
		rx_reg |= I40E_QRX_ENA_QENA_REQ_MASK;
	else
		rx_reg &= ~I40E_QRX_ENA_QENA_REQ_MASK;

	wr32(hw, I40E_QRX_ENA(pf_q), rx_reg);
}

/**
 * i40e_control_wait_rx_q
 * @pf: the PF structure
 * @pf_q: queue being configured
 * @enable: start or stop the rings
 *
 * This function enables or disables a single queue along with waiting
 * for the change to finish. The caller of this function should handle
 * the delays needed in the case of disabling queues.
 **/
int i40e_control_wait_rx_q(struct i40e_pf *pf, int pf_q, bool enable)
{
	int ret = 0;

	i40e_control_rx_q(pf, pf_q, enable);

	/* wait for the change to finish */
	ret = i40e_pf_rxq_wait(pf, pf_q, enable);
	if (ret)
		return ret;

	return ret;
}

/**
 * i40e_vsi_enable_rx - Start a VSI's rings
 * @vsi: the VSI being configured
 **/
static int i40e_vsi_enable_rx(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	int i, pf_q, ret = 0;

	pf_q = vsi->base_queue;
	for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
		ret = i40e_control_wait_rx_q(pf, pf_q, true);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "VSI seid %d Rx ring %d enable timeout\n",
				 vsi->seid, pf_q);
			break;
		}
	}

	return ret;
}

/**
 * i40e_vsi_start_rings - Start a VSI's rings
 * @vsi: the VSI being configured
 **/
int i40e_vsi_start_rings(struct i40e_vsi *vsi)
{
	int ret = 0;

	/* do rx first for enable and last for disable */
	ret = i40e_vsi_enable_rx(vsi);
	if (ret)
		return ret;
	ret = i40e_vsi_enable_tx(vsi);

	return ret;
}

#define I40E_DISABLE_TX_GAP_MSEC	50

/**
 * i40e_vsi_stop_rings - Stop a VSI's rings
 * @vsi: the VSI being configured
 **/
void i40e_vsi_stop_rings(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	int pf_q, err, q_end;

	/* When port TX is suspended, don't wait */
	if (test_bit(__I40E_PORT_SUSPENDED, vsi->back->state))
		return i40e_vsi_stop_rings_no_wait(vsi);

	q_end = vsi->base_queue + vsi->num_queue_pairs;
	for (pf_q = vsi->base_queue; pf_q < q_end; pf_q++)
		i40e_pre_tx_queue_cfg(&pf->hw, (u32)pf_q, false);

	for (pf_q = vsi->base_queue; pf_q < q_end; pf_q++) {
		err = i40e_control_wait_rx_q(pf, pf_q, false);
		if (err)
			dev_info(&pf->pdev->dev,
				 "VSI seid %d Rx ring %d disable timeout\n",
				 vsi->seid, pf_q);
	}

	msleep(I40E_DISABLE_TX_GAP_MSEC);
	pf_q = vsi->base_queue;
	for (pf_q = vsi->base_queue; pf_q < q_end; pf_q++)
		wr32(&pf->hw, I40E_QTX_ENA(pf_q), 0);

	i40e_vsi_wait_queues_disabled(vsi);
}

/**
 * i40e_vsi_stop_rings_no_wait - Stop a VSI's rings and do not delay
 * @vsi: the VSI being shutdown
 *
 * This function stops all the rings for a VSI but does not delay to verify
 * that rings have been disabled. It is expected that the caller is shutting
 * down multiple VSIs at once and will delay together for all the VSIs after
 * initiating the shutdown. This is particularly useful for shutting down lots
 * of VFs together. Otherwise, a large delay can be incurred while configuring
 * each VSI in serial.
 **/
void i40e_vsi_stop_rings_no_wait(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	int i, pf_q;

	pf_q = vsi->base_queue;
	for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
		i40e_control_tx_q(pf, pf_q, false);
		i40e_control_rx_q(pf, pf_q, false);
	}
}

/**
 * i40e_vsi_free_irq - Free the irq association with the OS
 * @vsi: the VSI being configured
 **/
static void i40e_vsi_free_irq(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	int base = vsi->base_vector;
	u32 val, qp;
	int i;

	if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
		if (!vsi->q_vectors)
			return;

		if (!vsi->irqs_ready)
			return;

		vsi->irqs_ready = false;
		for (i = 0; i < vsi->num_q_vectors; i++) {
			int irq_num;
			u16 vector;

			vector = i + base;
			irq_num = pf->msix_entries[vector].vector;

			/* free only the irqs that were actually requested */
			if (!vsi->q_vectors[i] ||
			    !vsi->q_vectors[i]->num_ringpairs)
				continue;

			/* clear the affinity notifier in the IRQ descriptor */
			irq_set_affinity_notifier(irq_num, NULL);
			/* remove our suggested affinity mask for this IRQ */
			irq_update_affinity_hint(irq_num, NULL);
			free_irq(irq_num, vsi->q_vectors[i]);

			/* Tear down the interrupt queue link list
			 *
			 * We know that they come in pairs and always
			 * the Rx first, then the Tx.  To clear the
			 * link list, stick the EOL value into the
			 * next_q field of the registers.
			 */
			val = rd32(hw, I40E_PFINT_LNKLSTN(vector - 1));
			qp = (val & I40E_PFINT_LNKLSTN_FIRSTQ_INDX_MASK)
				>> I40E_PFINT_LNKLSTN_FIRSTQ_INDX_SHIFT;
			val |= I40E_QUEUE_END_OF_LIST
				<< I40E_PFINT_LNKLSTN_FIRSTQ_INDX_SHIFT;
			wr32(hw, I40E_PFINT_LNKLSTN(vector - 1), val);

			while (qp != I40E_QUEUE_END_OF_LIST) {
				u32 next;

				val = rd32(hw, I40E_QINT_RQCTL(qp));

				val &= ~(I40E_QINT_RQCTL_MSIX_INDX_MASK  |
					 I40E_QINT_RQCTL_MSIX0_INDX_MASK |
					 I40E_QINT_RQCTL_CAUSE_ENA_MASK  |
					 I40E_QINT_RQCTL_INTEVENT_MASK);

				val |= (I40E_QINT_RQCTL_ITR_INDX_MASK |
					 I40E_QINT_RQCTL_NEXTQ_INDX_MASK);

				wr32(hw, I40E_QINT_RQCTL(qp), val);

				val = rd32(hw, I40E_QINT_TQCTL(qp));

				next = (val & I40E_QINT_TQCTL_NEXTQ_INDX_MASK)
					>> I40E_QINT_TQCTL_NEXTQ_INDX_SHIFT;

				val &= ~(I40E_QINT_TQCTL_MSIX_INDX_MASK  |
					 I40E_QINT_TQCTL_MSIX0_INDX_MASK |
					 I40E_QINT_TQCTL_CAUSE_ENA_MASK  |
					 I40E_QINT_TQCTL_INTEVENT_MASK);

				val |= (I40E_QINT_TQCTL_ITR_INDX_MASK |
					 I40E_QINT_TQCTL_NEXTQ_INDX_MASK);

				wr32(hw, I40E_QINT_TQCTL(qp), val);
				qp = next;
			}
		}
	} else {
		free_irq(pf->pdev->irq, pf);

		val = rd32(hw, I40E_PFINT_LNKLST0);
		qp = (val & I40E_PFINT_LNKLSTN_FIRSTQ_INDX_MASK)
			>> I40E_PFINT_LNKLSTN_FIRSTQ_INDX_SHIFT;
		val |= I40E_QUEUE_END_OF_LIST
			<< I40E_PFINT_LNKLST0_FIRSTQ_INDX_SHIFT;
		wr32(hw, I40E_PFINT_LNKLST0, val);

		val = rd32(hw, I40E_QINT_RQCTL(qp));
		val &= ~(I40E_QINT_RQCTL_MSIX_INDX_MASK  |
			 I40E_QINT_RQCTL_MSIX0_INDX_MASK |
			 I40E_QINT_RQCTL_CAUSE_ENA_MASK  |
			 I40E_QINT_RQCTL_INTEVENT_MASK);

		val |= (I40E_QINT_RQCTL_ITR_INDX_MASK |
			I40E_QINT_RQCTL_NEXTQ_INDX_MASK);

		wr32(hw, I40E_QINT_RQCTL(qp), val);

		val = rd32(hw, I40E_QINT_TQCTL(qp));

		val &= ~(I40E_QINT_TQCTL_MSIX_INDX_MASK  |
			 I40E_QINT_TQCTL_MSIX0_INDX_MASK |
			 I40E_QINT_TQCTL_CAUSE_ENA_MASK  |
			 I40E_QINT_TQCTL_INTEVENT_MASK);

		val |= (I40E_QINT_TQCTL_ITR_INDX_MASK |
			I40E_QINT_TQCTL_NEXTQ_INDX_MASK);

		wr32(hw, I40E_QINT_TQCTL(qp), val);
	}
}

/**
 * i40e_free_q_vector - Free memory allocated for specific interrupt vector
 * @vsi: the VSI being configured
 * @v_idx: Index of vector to be freed
 *
 * This function frees the memory allocated to the q_vector.  In addition if
 * NAPI is enabled it will delete any references to the NAPI struct prior
 * to freeing the q_vector.
 **/
static void i40e_free_q_vector(struct i40e_vsi *vsi, int v_idx)
{
	struct i40e_q_vector *q_vector = vsi->q_vectors[v_idx];
	struct i40e_ring *ring;

	if (!q_vector)
		return;

	/* disassociate q_vector from rings */
	i40e_for_each_ring(ring, q_vector->tx)
		ring->q_vector = NULL;

	i40e_for_each_ring(ring, q_vector->rx)
		ring->q_vector = NULL;

	/* only VSI w/ an associated netdev is set up w/ NAPI */
	if (vsi->netdev)
		netif_napi_del(&q_vector->napi);

	vsi->q_vectors[v_idx] = NULL;

	kfree_rcu(q_vector, rcu);
}

/**
 * i40e_vsi_free_q_vectors - Free memory allocated for interrupt vectors
 * @vsi: the VSI being un-configured
 *
 * This frees the memory allocated to the q_vectors and
 * deletes references to the NAPI struct.
 **/
static void i40e_vsi_free_q_vectors(struct i40e_vsi *vsi)
{
	int v_idx;

	for (v_idx = 0; v_idx < vsi->num_q_vectors; v_idx++)
		i40e_free_q_vector(vsi, v_idx);
}

/**
 * i40e_reset_interrupt_capability - Disable interrupt setup in OS
 * @pf: board private structure
 **/
static void i40e_reset_interrupt_capability(struct i40e_pf *pf)
{
	/* If we're in Legacy mode, the interrupt was cleaned in vsi_close */
	if (pf->flags & I40E_FLAG_MSIX_ENABLED) {
		pci_disable_msix(pf->pdev);
		kfree(pf->msix_entries);
		pf->msix_entries = NULL;
		kfree(pf->irq_pile);
		pf->irq_pile = NULL;
	} else if (pf->flags & I40E_FLAG_MSI_ENABLED) {
		pci_disable_msi(pf->pdev);
	}
	pf->flags &= ~(I40E_FLAG_MSIX_ENABLED | I40E_FLAG_MSI_ENABLED);
}

/**
 * i40e_clear_interrupt_scheme - Clear the current interrupt scheme settings
 * @pf: board private structure
 *
 * We go through and clear interrupt specific resources and reset the structure
 * to pre-load conditions
 **/
static void i40e_clear_interrupt_scheme(struct i40e_pf *pf)
{
	int i;

	if (test_bit(__I40E_MISC_IRQ_REQUESTED, pf->state))
		i40e_free_misc_vector(pf);

	i40e_put_lump(pf->irq_pile, pf->iwarp_base_vector,
		      I40E_IWARP_IRQ_PILE_ID);

	i40e_put_lump(pf->irq_pile, 0, I40E_PILE_VALID_BIT-1);
	for (i = 0; i < pf->num_alloc_vsi; i++)
		if (pf->vsi[i])
			i40e_vsi_free_q_vectors(pf->vsi[i]);
	i40e_reset_interrupt_capability(pf);
}

/**
 * i40e_napi_enable_all - Enable NAPI for all q_vectors in the VSI
 * @vsi: the VSI being configured
 **/
static void i40e_napi_enable_all(struct i40e_vsi *vsi)
{
	int q_idx;

	if (!vsi->netdev)
		return;

	for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) {
		struct i40e_q_vector *q_vector = vsi->q_vectors[q_idx];

		if (q_vector->rx.ring || q_vector->tx.ring)
			napi_enable(&q_vector->napi);
	}
}

/**
 * i40e_napi_disable_all - Disable NAPI for all q_vectors in the VSI
 * @vsi: the VSI being configured
 **/
static void i40e_napi_disable_all(struct i40e_vsi *vsi)
{
	int q_idx;

	if (!vsi->netdev)
		return;

	for (q_idx = 0; q_idx < vsi->num_q_vectors; q_idx++) {
		struct i40e_q_vector *q_vector = vsi->q_vectors[q_idx];

		if (q_vector->rx.ring || q_vector->tx.ring)
			napi_disable(&q_vector->napi);
	}
}

/**
 * i40e_vsi_close - Shut down a VSI
 * @vsi: the vsi to be quelled
 **/
static void i40e_vsi_close(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	if (!test_and_set_bit(__I40E_VSI_DOWN, vsi->state))
		i40e_down(vsi);
	i40e_vsi_free_irq(vsi);
	i40e_vsi_free_tx_resources(vsi);
	i40e_vsi_free_rx_resources(vsi);
	vsi->current_netdev_flags = 0;
	set_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state);
	if (test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state))
		set_bit(__I40E_CLIENT_RESET, pf->state);
}

/**
 * i40e_quiesce_vsi - Pause a given VSI
 * @vsi: the VSI being paused
 **/
static void i40e_quiesce_vsi(struct i40e_vsi *vsi)
{
	if (test_bit(__I40E_VSI_DOWN, vsi->state))
		return;

	set_bit(__I40E_VSI_NEEDS_RESTART, vsi->state);
	if (vsi->netdev && netif_running(vsi->netdev))
		vsi->netdev->netdev_ops->ndo_stop(vsi->netdev);
	else
		i40e_vsi_close(vsi);
}

/**
 * i40e_unquiesce_vsi - Resume a given VSI
 * @vsi: the VSI being resumed
 **/
static void i40e_unquiesce_vsi(struct i40e_vsi *vsi)
{
	if (!test_and_clear_bit(__I40E_VSI_NEEDS_RESTART, vsi->state))
		return;

	if (vsi->netdev && netif_running(vsi->netdev))
		vsi->netdev->netdev_ops->ndo_open(vsi->netdev);
	else
		i40e_vsi_open(vsi);   /* this clears the DOWN bit */
}

/**
 * i40e_pf_quiesce_all_vsi - Pause all VSIs on a PF
 * @pf: the PF
 **/
static void i40e_pf_quiesce_all_vsi(struct i40e_pf *pf)
{
	int v;

	for (v = 0; v < pf->num_alloc_vsi; v++) {
		if (pf->vsi[v])
			i40e_quiesce_vsi(pf->vsi[v]);
	}
}

/**
 * i40e_pf_unquiesce_all_vsi - Resume all VSIs on a PF
 * @pf: the PF
 **/
static void i40e_pf_unquiesce_all_vsi(struct i40e_pf *pf)
{
	int v;

	for (v = 0; v < pf->num_alloc_vsi; v++) {
		if (pf->vsi[v])
			i40e_unquiesce_vsi(pf->vsi[v]);
	}
}

/**
 * i40e_vsi_wait_queues_disabled - Wait for VSI's queues to be disabled
 * @vsi: the VSI being configured
 *
 * Wait until all queues on a given VSI have been disabled.
 **/
int i40e_vsi_wait_queues_disabled(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;
	int i, pf_q, ret;

	pf_q = vsi->base_queue;
	for (i = 0; i < vsi->num_queue_pairs; i++, pf_q++) {
		/* Check and wait for the Tx queue */
		ret = i40e_pf_txq_wait(pf, pf_q, false);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "VSI seid %d Tx ring %d disable timeout\n",
				 vsi->seid, pf_q);
			return ret;
		}

		if (!i40e_enabled_xdp_vsi(vsi))
			goto wait_rx;

		/* Check and wait for the XDP Tx queue */
		ret = i40e_pf_txq_wait(pf, pf_q + vsi->alloc_queue_pairs,
				       false);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "VSI seid %d XDP Tx ring %d disable timeout\n",
				 vsi->seid, pf_q);
			return ret;
		}
wait_rx:
		/* Check and wait for the Rx queue */
		ret = i40e_pf_rxq_wait(pf, pf_q, false);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "VSI seid %d Rx ring %d disable timeout\n",
				 vsi->seid, pf_q);
			return ret;
		}
	}

	return 0;
}

#ifdef CONFIG_I40E_DCB
/**
 * i40e_pf_wait_queues_disabled - Wait for all queues of PF VSIs to be disabled
 * @pf: the PF
 *
 * This function waits for the queues to be in disabled state for all the
 * VSIs that are managed by this PF.
 **/
static int i40e_pf_wait_queues_disabled(struct i40e_pf *pf)
{
	int v, ret = 0;

	for (v = 0; v < pf->hw.func_caps.num_vsis; v++) {
		if (pf->vsi[v]) {
			ret = i40e_vsi_wait_queues_disabled(pf->vsi[v]);
			if (ret)
				break;
		}
	}

	return ret;
}

#endif

/**
 * i40e_get_iscsi_tc_map - Return TC map for iSCSI APP
 * @pf: pointer to PF
 *
 * Get TC map for ISCSI PF type that will include iSCSI TC
 * and LAN TC.
 **/
static u8 i40e_get_iscsi_tc_map(struct i40e_pf *pf)
{
	struct i40e_dcb_app_priority_table app;
	struct i40e_hw *hw = &pf->hw;
	u8 enabled_tc = 1; /* TC0 is always enabled */
	u8 tc, i;
	/* Get the iSCSI APP TLV */
	struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config;

	for (i = 0; i < dcbcfg->numapps; i++) {
		app = dcbcfg->app[i];
		if (app.selector == I40E_APP_SEL_TCPIP &&
		    app.protocolid == I40E_APP_PROTOID_ISCSI) {
			tc = dcbcfg->etscfg.prioritytable[app.priority];
			enabled_tc |= BIT(tc);
			break;
		}
	}

	return enabled_tc;
}

/**
 * i40e_dcb_get_num_tc -  Get the number of TCs from DCBx config
 * @dcbcfg: the corresponding DCBx configuration structure
 *
 * Return the number of TCs from given DCBx configuration
 **/
static u8 i40e_dcb_get_num_tc(struct i40e_dcbx_config *dcbcfg)
{
	int i, tc_unused = 0;
	u8 num_tc = 0;
	u8 ret = 0;

	/* Scan the ETS Config Priority Table to find
	 * traffic class enabled for a given priority
	 * and create a bitmask of enabled TCs
	 */
	for (i = 0; i < I40E_MAX_USER_PRIORITY; i++)
		num_tc |= BIT(dcbcfg->etscfg.prioritytable[i]);

	/* Now scan the bitmask to check for
	 * contiguous TCs starting with TC0
	 */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		if (num_tc & BIT(i)) {
			if (!tc_unused) {
				ret++;
			} else {
				pr_err("Non-contiguous TC - Disabling DCB\n");
				return 1;
			}
		} else {
			tc_unused = 1;
		}
	}

	/* There is always at least TC0 */
	if (!ret)
		ret = 1;

	return ret;
}

/**
 * i40e_dcb_get_enabled_tc - Get enabled traffic classes
 * @dcbcfg: the corresponding DCBx configuration structure
 *
 * Query the current DCB configuration and return the number of
 * traffic classes enabled from the given DCBX config
 **/
static u8 i40e_dcb_get_enabled_tc(struct i40e_dcbx_config *dcbcfg)
{
	u8 num_tc = i40e_dcb_get_num_tc(dcbcfg);
	u8 enabled_tc = 1;
	u8 i;

	for (i = 0; i < num_tc; i++)
		enabled_tc |= BIT(i);

	return enabled_tc;
}

/**
 * i40e_mqprio_get_enabled_tc - Get enabled traffic classes
 * @pf: PF being queried
 *
 * Query the current MQPRIO configuration and return the number of
 * traffic classes enabled.
 **/
static u8 i40e_mqprio_get_enabled_tc(struct i40e_pf *pf)
{
	struct i40e_vsi *vsi = pf->vsi[pf->lan_vsi];
	u8 num_tc = vsi->mqprio_qopt.qopt.num_tc;
	u8 enabled_tc = 1, i;

	for (i = 1; i < num_tc; i++)
		enabled_tc |= BIT(i);
	return enabled_tc;
}

/**
 * i40e_pf_get_num_tc - Get enabled traffic classes for PF
 * @pf: PF being queried
 *
 * Return number of traffic classes enabled for the given PF
 **/
static u8 i40e_pf_get_num_tc(struct i40e_pf *pf)
{
	struct i40e_hw *hw = &pf->hw;
	u8 i, enabled_tc = 1;
	u8 num_tc = 0;
	struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config;

	if (i40e_is_tc_mqprio_enabled(pf))
		return pf->vsi[pf->lan_vsi]->mqprio_qopt.qopt.num_tc;

	/* If neither MQPRIO nor DCB is enabled, then always use single TC */
	if (!(pf->flags & I40E_FLAG_DCB_ENABLED))
		return 1;

	/* SFP mode will be enabled for all TCs on port */
	if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
		return i40e_dcb_get_num_tc(dcbcfg);

	/* MFP mode return count of enabled TCs for this PF */
	if (pf->hw.func_caps.iscsi)
		enabled_tc =  i40e_get_iscsi_tc_map(pf);
	else
		return 1; /* Only TC0 */

	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		if (enabled_tc & BIT(i))
			num_tc++;
	}
	return num_tc;
}

/**
 * i40e_pf_get_tc_map - Get bitmap for enabled traffic classes
 * @pf: PF being queried
 *
 * Return a bitmap for enabled traffic classes for this PF.
 **/
static u8 i40e_pf_get_tc_map(struct i40e_pf *pf)
{
	if (i40e_is_tc_mqprio_enabled(pf))
		return i40e_mqprio_get_enabled_tc(pf);

	/* If neither MQPRIO nor DCB is enabled for this PF then just return
	 * default TC
	 */
	if (!(pf->flags & I40E_FLAG_DCB_ENABLED))
		return I40E_DEFAULT_TRAFFIC_CLASS;

	/* SFP mode we want PF to be enabled for all TCs */
	if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
		return i40e_dcb_get_enabled_tc(&pf->hw.local_dcbx_config);

	/* MFP enabled and iSCSI PF type */
	if (pf->hw.func_caps.iscsi)
		return i40e_get_iscsi_tc_map(pf);
	else
		return I40E_DEFAULT_TRAFFIC_CLASS;
}

/**
 * i40e_vsi_get_bw_info - Query VSI BW Information
 * @vsi: the VSI being queried
 *
 * Returns 0 on success, negative value on failure
 **/
static int i40e_vsi_get_bw_info(struct i40e_vsi *vsi)
{
	struct i40e_aqc_query_vsi_ets_sla_config_resp bw_ets_config = {0};
	struct i40e_aqc_query_vsi_bw_config_resp bw_config = {0};
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	u32 tc_bw_max;
	int ret;
	int i;

	/* Get the VSI level BW configuration */
	ret = i40e_aq_query_vsi_bw_config(hw, vsi->seid, &bw_config, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "couldn't get PF vsi bw config, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		return -EINVAL;
	}

	/* Get the VSI level BW configuration per TC */
	ret = i40e_aq_query_vsi_ets_sla_config(hw, vsi->seid, &bw_ets_config,
					       NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "couldn't get PF vsi ets bw config, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		return -EINVAL;
	}

	if (bw_config.tc_valid_bits != bw_ets_config.tc_valid_bits) {
		dev_info(&pf->pdev->dev,
			 "Enabled TCs mismatch from querying VSI BW info 0x%08x 0x%08x\n",
			 bw_config.tc_valid_bits,
			 bw_ets_config.tc_valid_bits);
		/* Still continuing */
	}

	vsi->bw_limit = le16_to_cpu(bw_config.port_bw_limit);
	vsi->bw_max_quanta = bw_config.max_bw;
	tc_bw_max = le16_to_cpu(bw_ets_config.tc_bw_max[0]) |
		    (le16_to_cpu(bw_ets_config.tc_bw_max[1]) << 16);
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		vsi->bw_ets_share_credits[i] = bw_ets_config.share_credits[i];
		vsi->bw_ets_limit_credits[i] =
					le16_to_cpu(bw_ets_config.credits[i]);
		/* 3 bits out of 4 for each TC */
		vsi->bw_ets_max_quanta[i] = (u8)((tc_bw_max >> (i*4)) & 0x7);
	}

	return 0;
}

/**
 * i40e_vsi_configure_bw_alloc - Configure VSI BW allocation per TC
 * @vsi: the VSI being configured
 * @enabled_tc: TC bitmap
 * @bw_share: BW shared credits per TC
 *
 * Returns 0 on success, negative value on failure
 **/
static int i40e_vsi_configure_bw_alloc(struct i40e_vsi *vsi, u8 enabled_tc,
				       u8 *bw_share)
{
	struct i40e_aqc_configure_vsi_tc_bw_data bw_data;
	struct i40e_pf *pf = vsi->back;
	int ret;
	int i;

	/* There is no need to reset BW when mqprio mode is on.  */
	if (i40e_is_tc_mqprio_enabled(pf))
		return 0;
	if (!vsi->mqprio_qopt.qopt.hw && !(pf->flags & I40E_FLAG_DCB_ENABLED)) {
		ret = i40e_set_bw_limit(vsi, vsi->seid, 0);
		if (ret)
			dev_info(&pf->pdev->dev,
				 "Failed to reset tx rate for vsi->seid %u\n",
				 vsi->seid);
		return ret;
	}
	memset(&bw_data, 0, sizeof(bw_data));
	bw_data.tc_valid_bits = enabled_tc;
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
		bw_data.tc_bw_credits[i] = bw_share[i];

	ret = i40e_aq_config_vsi_tc_bw(&pf->hw, vsi->seid, &bw_data, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "AQ command Config VSI BW allocation per TC failed = %d\n",
			 pf->hw.aq.asq_last_status);
		return -EINVAL;
	}

	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
		vsi->info.qs_handle[i] = bw_data.qs_handles[i];

	return 0;
}

/**
 * i40e_vsi_config_netdev_tc - Setup the netdev TC configuration
 * @vsi: the VSI being configured
 * @enabled_tc: TC map to be enabled
 *
 **/
static void i40e_vsi_config_netdev_tc(struct i40e_vsi *vsi, u8 enabled_tc)
{
	struct net_device *netdev = vsi->netdev;
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	u8 netdev_tc = 0;
	int i;
	struct i40e_dcbx_config *dcbcfg = &hw->local_dcbx_config;

	if (!netdev)
		return;

	if (!enabled_tc) {
		netdev_reset_tc(netdev);
		return;
	}

	/* Set up actual enabled TCs on the VSI */
	if (netdev_set_num_tc(netdev, vsi->tc_config.numtc))
		return;

	/* set per TC queues for the VSI */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		/* Only set TC queues for enabled tcs
		 *
		 * e.g. For a VSI that has TC0 and TC3 enabled the
		 * enabled_tc bitmap would be 0x00001001; the driver
		 * will set the numtc for netdev as 2 that will be
		 * referenced by the netdev layer as TC 0 and 1.
		 */
		if (vsi->tc_config.enabled_tc & BIT(i))
			netdev_set_tc_queue(netdev,
					vsi->tc_config.tc_info[i].netdev_tc,
					vsi->tc_config.tc_info[i].qcount,
					vsi->tc_config.tc_info[i].qoffset);
	}

	if (i40e_is_tc_mqprio_enabled(pf))
		return;

	/* Assign UP2TC map for the VSI */
	for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
		/* Get the actual TC# for the UP */
		u8 ets_tc = dcbcfg->etscfg.prioritytable[i];
		/* Get the mapped netdev TC# for the UP */
		netdev_tc =  vsi->tc_config.tc_info[ets_tc].netdev_tc;
		netdev_set_prio_tc_map(netdev, i, netdev_tc);
	}
}

/**
 * i40e_vsi_update_queue_map - Update our copy of VSi info with new queue map
 * @vsi: the VSI being configured
 * @ctxt: the ctxt buffer returned from AQ VSI update param command
 **/
static void i40e_vsi_update_queue_map(struct i40e_vsi *vsi,
				      struct i40e_vsi_context *ctxt)
{
	/* copy just the sections touched not the entire info
	 * since not all sections are valid as returned by
	 * update vsi params
	 */
	vsi->info.mapping_flags = ctxt->info.mapping_flags;
	memcpy(&vsi->info.queue_mapping,
	       &ctxt->info.queue_mapping, sizeof(vsi->info.queue_mapping));
	memcpy(&vsi->info.tc_mapping, ctxt->info.tc_mapping,
	       sizeof(vsi->info.tc_mapping));
}

/**
 * i40e_update_adq_vsi_queues - update queue mapping for ADq VSI
 * @vsi: the VSI being reconfigured
 * @vsi_offset: offset from main VF VSI
 */
int i40e_update_adq_vsi_queues(struct i40e_vsi *vsi, int vsi_offset)
{
	struct i40e_vsi_context ctxt = {};
	struct i40e_pf *pf;
	struct i40e_hw *hw;
	int ret;

	if (!vsi)
		return I40E_ERR_PARAM;
	pf = vsi->back;
	hw = &pf->hw;

	ctxt.seid = vsi->seid;
	ctxt.pf_num = hw->pf_id;
	ctxt.vf_num = vsi->vf_id + hw->func_caps.vf_base_id + vsi_offset;
	ctxt.uplink_seid = vsi->uplink_seid;
	ctxt.connection_type = I40E_AQ_VSI_CONN_TYPE_NORMAL;
	ctxt.flags = I40E_AQ_VSI_TYPE_VF;
	ctxt.info = vsi->info;

	i40e_vsi_setup_queue_map(vsi, &ctxt, vsi->tc_config.enabled_tc,
				 false);
	if (vsi->reconfig_rss) {
		vsi->rss_size = min_t(int, pf->alloc_rss_size,
				      vsi->num_queue_pairs);
		ret = i40e_vsi_config_rss(vsi);
		if (ret) {
			dev_info(&pf->pdev->dev, "Failed to reconfig rss for num_queues\n");
			return ret;
		}
		vsi->reconfig_rss = false;
	}

	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev, "Update vsi config failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(hw, hw->aq.asq_last_status));
		return ret;
	}
	/* update the local VSI info with updated queue map */
	i40e_vsi_update_queue_map(vsi, &ctxt);
	vsi->info.valid_sections = 0;

	return ret;
}

/**
 * i40e_vsi_config_tc - Configure VSI Tx Scheduler for given TC map
 * @vsi: VSI to be configured
 * @enabled_tc: TC bitmap
 *
 * This configures a particular VSI for TCs that are mapped to the
 * given TC bitmap. It uses default bandwidth share for TCs across
 * VSIs to configure TC for a particular VSI.
 *
 * NOTE:
 * It is expected that the VSI queues have been quisced before calling
 * this function.
 **/
static int i40e_vsi_config_tc(struct i40e_vsi *vsi, u8 enabled_tc)
{
	u8 bw_share[I40E_MAX_TRAFFIC_CLASS] = {0};
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	struct i40e_vsi_context ctxt;
	int ret = 0;
	int i;

	/* Check if enabled_tc is same as existing or new TCs */
	if (vsi->tc_config.enabled_tc == enabled_tc &&
	    vsi->mqprio_qopt.mode != TC_MQPRIO_MODE_CHANNEL)
		return ret;

	/* Enable ETS TCs with equal BW Share for now across all VSIs */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		if (enabled_tc & BIT(i))
			bw_share[i] = 1;
	}

	ret = i40e_vsi_configure_bw_alloc(vsi, enabled_tc, bw_share);
	if (ret) {
		struct i40e_aqc_query_vsi_bw_config_resp bw_config = {0};

		dev_info(&pf->pdev->dev,
			 "Failed configuring TC map %d for VSI %d\n",
			 enabled_tc, vsi->seid);
		ret = i40e_aq_query_vsi_bw_config(hw, vsi->seid,
						  &bw_config, NULL);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "Failed querying vsi bw info, err %pe aq_err %s\n",
				 ERR_PTR(ret),
				 i40e_aq_str(hw, hw->aq.asq_last_status));
			goto out;
		}
		if ((bw_config.tc_valid_bits & enabled_tc) != enabled_tc) {
			u8 valid_tc = bw_config.tc_valid_bits & enabled_tc;

			if (!valid_tc)
				valid_tc = bw_config.tc_valid_bits;
			/* Always enable TC0, no matter what */
			valid_tc |= 1;
			dev_info(&pf->pdev->dev,
				 "Requested tc 0x%x, but FW reports 0x%x as valid. Attempting to use 0x%x.\n",
				 enabled_tc, bw_config.tc_valid_bits, valid_tc);
			enabled_tc = valid_tc;
		}

		ret = i40e_vsi_configure_bw_alloc(vsi, enabled_tc, bw_share);
		if (ret) {
			dev_err(&pf->pdev->dev,
				"Unable to  configure TC map %d for VSI %d\n",
				enabled_tc, vsi->seid);
			goto out;
		}
	}

	/* Update Queue Pairs Mapping for currently enabled UPs */
	ctxt.seid = vsi->seid;
	ctxt.pf_num = vsi->back->hw.pf_id;
	ctxt.vf_num = 0;
	ctxt.uplink_seid = vsi->uplink_seid;
	ctxt.info = vsi->info;
	if (i40e_is_tc_mqprio_enabled(pf)) {
		ret = i40e_vsi_setup_queue_map_mqprio(vsi, &ctxt, enabled_tc);
		if (ret)
			goto out;
	} else {
		i40e_vsi_setup_queue_map(vsi, &ctxt, enabled_tc, false);
	}

	/* On destroying the qdisc, reset vsi->rss_size, as number of enabled
	 * queues changed.
	 */
	if (!vsi->mqprio_qopt.qopt.hw && vsi->reconfig_rss) {
		vsi->rss_size = min_t(int, vsi->back->alloc_rss_size,
				      vsi->num_queue_pairs);
		ret = i40e_vsi_config_rss(vsi);
		if (ret) {
			dev_info(&vsi->back->pdev->dev,
				 "Failed to reconfig rss for num_queues\n");
			return ret;
		}
		vsi->reconfig_rss = false;
	}
	if (vsi->back->flags & I40E_FLAG_IWARP_ENABLED) {
		ctxt.info.valid_sections |=
				cpu_to_le16(I40E_AQ_VSI_PROP_QUEUE_OPT_VALID);
		ctxt.info.queueing_opt_flags |= I40E_AQ_VSI_QUE_OPT_TCP_ENA;
	}

	/* Update the VSI after updating the VSI queue-mapping
	 * information
	 */
	ret = i40e_aq_update_vsi_params(hw, &ctxt, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Update vsi tc config failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(hw, hw->aq.asq_last_status));
		goto out;
	}
	/* update the local VSI info with updated queue map */
	i40e_vsi_update_queue_map(vsi, &ctxt);
	vsi->info.valid_sections = 0;

	/* Update current VSI BW information */
	ret = i40e_vsi_get_bw_info(vsi);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Failed updating vsi bw info, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(hw, hw->aq.asq_last_status));
		goto out;
	}

	/* Update the netdev TC setup */
	i40e_vsi_config_netdev_tc(vsi, enabled_tc);
out:
	return ret;
}

/**
 * i40e_get_link_speed - Returns link speed for the interface
 * @vsi: VSI to be configured
 *
 **/
static int i40e_get_link_speed(struct i40e_vsi *vsi)
{
	struct i40e_pf *pf = vsi->back;

	switch (pf->hw.phy.link_info.link_speed) {
	case I40E_LINK_SPEED_40GB:
		return 40000;
	case I40E_LINK_SPEED_25GB:
		return 25000;
	case I40E_LINK_SPEED_20GB:
		return 20000;
	case I40E_LINK_SPEED_10GB:
		return 10000;
	case I40E_LINK_SPEED_1GB:
		return 1000;
	default:
		return -EINVAL;
	}
}

/**
 * i40e_bw_bytes_to_mbits - Convert max_tx_rate from bytes to mbits
 * @vsi: Pointer to vsi structure
 * @max_tx_rate: max TX rate in bytes to be converted into Mbits
 *
 * Helper function to convert units before send to set BW limit
 **/
static u64 i40e_bw_bytes_to_mbits(struct i40e_vsi *vsi, u64 max_tx_rate)
{
	if (max_tx_rate < I40E_BW_MBPS_DIVISOR) {
		dev_warn(&vsi->back->pdev->dev,
			 "Setting max tx rate to minimum usable value of 50Mbps.\n");
		max_tx_rate = I40E_BW_CREDIT_DIVISOR;
	} else {
		do_div(max_tx_rate, I40E_BW_MBPS_DIVISOR);
	}

	return max_tx_rate;
}

/**
 * i40e_set_bw_limit - setup BW limit for Tx traffic based on max_tx_rate
 * @vsi: VSI to be configured
 * @seid: seid of the channel/VSI
 * @max_tx_rate: max TX rate to be configured as BW limit
 *
 * Helper function to set BW limit for a given VSI
 **/
int i40e_set_bw_limit(struct i40e_vsi *vsi, u16 seid, u64 max_tx_rate)
{
	struct i40e_pf *pf = vsi->back;
	u64 credits = 0;
	int speed = 0;
	int ret = 0;

	speed = i40e_get_link_speed(vsi);
	if (max_tx_rate > speed) {
		dev_err(&pf->pdev->dev,
			"Invalid max tx rate %llu specified for VSI seid %d.",
			max_tx_rate, seid);
		return -EINVAL;
	}
	if (max_tx_rate && max_tx_rate < I40E_BW_CREDIT_DIVISOR) {
		dev_warn(&pf->pdev->dev,
			 "Setting max tx rate to minimum usable value of 50Mbps.\n");
		max_tx_rate = I40E_BW_CREDIT_DIVISOR;
	}

	/* Tx rate credits are in values of 50Mbps, 0 is disabled */
	credits = max_tx_rate;
	do_div(credits, I40E_BW_CREDIT_DIVISOR);
	ret = i40e_aq_config_vsi_bw_limit(&pf->hw, seid, credits,
					  I40E_MAX_BW_INACTIVE_ACCUM, NULL);
	if (ret)
		dev_err(&pf->pdev->dev,
			"Failed set tx rate (%llu Mbps) for vsi->seid %u, err %pe aq_err %s\n",
			max_tx_rate, seid, ERR_PTR(ret),
			i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
	return ret;
}

/**
 * i40e_remove_queue_channels - Remove queue channels for the TCs
 * @vsi: VSI to be configured
 *
 * Remove queue channels for the TCs
 **/
static void i40e_remove_queue_channels(struct i40e_vsi *vsi)
{
	enum i40e_admin_queue_err last_aq_status;
	struct i40e_cloud_filter *cfilter;
	struct i40e_channel *ch, *ch_tmp;
	struct i40e_pf *pf = vsi->back;
	struct hlist_node *node;
	int ret, i;

	/* Reset rss size that was stored when reconfiguring rss for
	 * channel VSIs with non-power-of-2 queue count.
	 */
	vsi->current_rss_size = 0;

	/* perform cleanup for channels if they exist */
	if (list_empty(&vsi->ch_list))
		return;

	list_for_each_entry_safe(ch, ch_tmp, &vsi->ch_list, list) {
		struct i40e_vsi *p_vsi;

		list_del(&ch->list);
		p_vsi = ch->parent_vsi;
		if (!p_vsi || !ch->initialized) {
			kfree(ch);
			continue;
		}
		/* Reset queue contexts */
		for (i = 0; i < ch->num_queue_pairs; i++) {
			struct i40e_ring *tx_ring, *rx_ring;
			u16 pf_q;

			pf_q = ch->base_queue + i;
			tx_ring = vsi->tx_rings[pf_q];
			tx_ring->ch = NULL;

			rx_ring = vsi->rx_rings[pf_q];
			rx_ring->ch = NULL;
		}

		/* Reset BW configured for this VSI via mqprio */
		ret = i40e_set_bw_limit(vsi, ch->seid, 0);
		if (ret)
			dev_info(&vsi->back->pdev->dev,
				 "Failed to reset tx rate for ch->seid %u\n",
				 ch->seid);

		/* delete cloud filters associated with this channel */
		hlist_for_each_entry_safe(cfilter, node,
					  &pf->cloud_filter_list, cloud_node) {
			if (cfilter->seid != ch->seid)
				continue;

			hash_del(&cfilter->cloud_node);
			if (cfilter->dst_port)
				ret = i40e_add_del_cloud_filter_big_buf(vsi,
									cfilter,
									false);
			else
				ret = i40e_add_del_cloud_filter(vsi, cfilter,
								false);
			last_aq_status = pf->hw.aq.asq_last_status;
			if (ret)
				dev_info(&pf->pdev->dev,
					 "Failed to delete cloud filter, err %pe aq_err %s\n",
					 ERR_PTR(ret),
					 i40e_aq_str(&pf->hw, last_aq_status));
			kfree(cfilter);
		}

		/* delete VSI from FW */
		ret = i40e_aq_delete_element(&vsi->back->hw, ch->seid,
					     NULL);
		if (ret)
			dev_err(&vsi->back->pdev->dev,
				"unable to remove channel (%d) for parent VSI(%d)\n",
				ch->seid, p_vsi->seid);
		kfree(ch);
	}
	INIT_LIST_HEAD(&vsi->ch_list);
}

/**
 * i40e_get_max_queues_for_channel
 * @vsi: ptr to VSI to which channels are associated with
 *
 * Helper function which returns max value among the queue counts set on the
 * channels/TCs created.
 **/
static int i40e_get_max_queues_for_channel(struct i40e_vsi *vsi)
{
	struct i40e_channel *ch, *ch_tmp;
	int max = 0;

	list_for_each_entry_safe(ch, ch_tmp, &vsi->ch_list, list) {
		if (!ch->initialized)
			continue;
		if (ch->num_queue_pairs > max)
			max = ch->num_queue_pairs;
	}

	return max;
}

/**
 * i40e_validate_num_queues - validate num_queues w.r.t channel
 * @pf: ptr to PF device
 * @num_queues: number of queues
 * @vsi: the parent VSI
 * @reconfig_rss: indicates should the RSS be reconfigured or not
 *
 * This function validates number of queues in the context of new channel
 * which is being established and determines if RSS should be reconfigured
 * or not for parent VSI.
 **/
static int i40e_validate_num_queues(struct i40e_pf *pf, int num_queues,
				    struct i40e_vsi *vsi, bool *reconfig_rss)
{
	int max_ch_queues;

	if (!reconfig_rss)
		return -EINVAL;

	*reconfig_rss = false;
	if (vsi->current_rss_size) {
		if (num_queues > vsi->current_rss_size) {
			dev_dbg(&pf->pdev->dev,
				"Error: num_queues (%d) > vsi's current_size(%d)\n",
				num_queues, vsi->current_rss_size);
			return -EINVAL;
		} else if ((num_queues < vsi->current_rss_size) &&
			   (!is_power_of_2(num_queues))) {
			dev_dbg(&pf->pdev->dev,
				"Error: num_queues (%d) < vsi's current_size(%d), but not power of 2\n",
				num_queues, vsi->current_rss_size);
			return -EINVAL;
		}
	}

	if (!is_power_of_2(num_queues)) {
		/* Find the max num_queues configured for channel if channel
		 * exist.
		 * if channel exist, then enforce 'num_queues' to be more than
		 * max ever queues configured for channel.
		 */
		max_ch_queues = i40e_get_max_queues_for_channel(vsi);
		if (num_queues < max_ch_queues) {
			dev_dbg(&pf->pdev->dev,
				"Error: num_queues (%d) < max queues configured for channel(%d)\n",
				num_queues, max_ch_queues);
			return -EINVAL;
		}
		*reconfig_rss = true;
	}

	return 0;
}

/**
 * i40e_vsi_reconfig_rss - reconfig RSS based on specified rss_size
 * @vsi: the VSI being setup
 * @rss_size: size of RSS, accordingly LUT gets reprogrammed
 *
 * This function reconfigures RSS by reprogramming LUTs using 'rss_size'
 **/
static int i40e_vsi_reconfig_rss(struct i40e_vsi *vsi, u16 rss_size)
{
	struct i40e_pf *pf = vsi->back;
	u8 seed[I40E_HKEY_ARRAY_SIZE];
	struct i40e_hw *hw = &pf->hw;
	int local_rss_size;
	u8 *lut;
	int ret;

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

	if (rss_size > vsi->rss_size)
		return -EINVAL;

	local_rss_size = min_t(int, vsi->rss_size, rss_size);
	lut = kzalloc(vsi->rss_table_size, GFP_KERNEL);
	if (!lut)
		return -ENOMEM;

	/* Ignoring user configured lut if there is one */
	i40e_fill_rss_lut(pf, lut, vsi->rss_table_size, local_rss_size);

	/* Use user configured hash key if there is one, otherwise
	 * use default.
	 */
	if (vsi->rss_hkey_user)
		memcpy(seed, vsi->rss_hkey_user, I40E_HKEY_ARRAY_SIZE);
	else
		netdev_rss_key_fill((void *)seed, I40E_HKEY_ARRAY_SIZE);

	ret = i40e_config_rss(vsi, seed, lut, vsi->rss_table_size);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Cannot set RSS lut, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(hw, hw->aq.asq_last_status));
		kfree(lut);
		return ret;
	}
	kfree(lut);

	/* Do the update w.r.t. storing rss_size */
	if (!vsi->orig_rss_size)
		vsi->orig_rss_size = vsi->rss_size;
	vsi->current_rss_size = local_rss_size;

	return ret;
}

/**
 * i40e_channel_setup_queue_map - Setup a channel queue map
 * @pf: ptr to PF device
 * @ctxt: VSI context structure
 * @ch: ptr to channel structure
 *
 * Setup queue map for a specific channel
 **/
static void i40e_channel_setup_queue_map(struct i40e_pf *pf,
					 struct i40e_vsi_context *ctxt,
					 struct i40e_channel *ch)
{
	u16 qcount, qmap, sections = 0;
	u8 offset = 0;
	int pow;

	sections = I40E_AQ_VSI_PROP_QUEUE_MAP_VALID;
	sections |= I40E_AQ_VSI_PROP_SCHED_VALID;

	qcount = min_t(int, ch->num_queue_pairs, pf->num_lan_msix);
	ch->num_queue_pairs = qcount;

	/* find the next higher power-of-2 of num queue pairs */
	pow = ilog2(qcount);
	if (!is_power_of_2(qcount))
		pow++;

	qmap = (offset << I40E_AQ_VSI_TC_QUE_OFFSET_SHIFT) |
		(pow << I40E_AQ_VSI_TC_QUE_NUMBER_SHIFT);

	/* Setup queue TC[0].qmap for given VSI context */
	ctxt->info.tc_mapping[0] = cpu_to_le16(qmap);

	ctxt->info.up_enable_bits = 0x1; /* TC0 enabled */
	ctxt->info.mapping_flags |= cpu_to_le16(I40E_AQ_VSI_QUE_MAP_CONTIG);
	ctxt->info.queue_mapping[0] = cpu_to_le16(ch->base_queue);
	ctxt->info.valid_sections |= cpu_to_le16(sections);
}

/**
 * i40e_add_channel - add a channel by adding VSI
 * @pf: ptr to PF device
 * @uplink_seid: underlying HW switching element (VEB) ID
 * @ch: ptr to channel structure
 *
 * Add a channel (VSI) using add_vsi and queue_map
 **/
static int i40e_add_channel(struct i40e_pf *pf, u16 uplink_seid,
			    struct i40e_channel *ch)
{
	struct i40e_hw *hw = &pf->hw;
	struct i40e_vsi_context ctxt;
	u8 enabled_tc = 0x1; /* TC0 enabled */
	int ret;

	if (ch->type != I40E_VSI_VMDQ2) {
		dev_info(&pf->pdev->dev,
			 "add new vsi failed, ch->type %d\n", ch->type);
		return -EINVAL;
	}

	memset(&ctxt, 0, sizeof(ctxt));
	ctxt.pf_num = hw->pf_id;
	ctxt.vf_num = 0;
	ctxt.uplink_seid = uplink_seid;
	ctxt.connection_type = I40E_AQ_VSI_CONN_TYPE_NORMAL;
	if (ch->type == I40E_VSI_VMDQ2)
		ctxt.flags = I40E_AQ_VSI_TYPE_VMDQ2;

	if (pf->flags & I40E_FLAG_VEB_MODE_ENABLED) {
		ctxt.info.valid_sections |=
		     cpu_to_le16(I40E_AQ_VSI_PROP_SWITCH_VALID);
		ctxt.info.switch_id =
		   cpu_to_le16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
	}

	/* Set queue map for a given VSI context */
	i40e_channel_setup_queue_map(pf, &ctxt, ch);

	/* Now time to create VSI */
	ret = i40e_aq_add_vsi(hw, &ctxt, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "add new vsi failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw,
				     pf->hw.aq.asq_last_status));
		return -ENOENT;
	}

	/* Success, update channel, set enabled_tc only if the channel
	 * is not a macvlan
	 */
	ch->enabled_tc = !i40e_is_channel_macvlan(ch) && enabled_tc;
	ch->seid = ctxt.seid;
	ch->vsi_number = ctxt.vsi_number;
	ch->stat_counter_idx = le16_to_cpu(ctxt.info.stat_counter_idx);

	/* copy just the sections touched not the entire info
	 * since not all sections are valid as returned by
	 * update vsi params
	 */
	ch->info.mapping_flags = ctxt.info.mapping_flags;
	memcpy(&ch->info.queue_mapping,
	       &ctxt.info.queue_mapping, sizeof(ctxt.info.queue_mapping));
	memcpy(&ch->info.tc_mapping, ctxt.info.tc_mapping,
	       sizeof(ctxt.info.tc_mapping));

	return 0;
}

static int i40e_channel_config_bw(struct i40e_vsi *vsi, struct i40e_channel *ch,
				  u8 *bw_share)
{
	struct i40e_aqc_configure_vsi_tc_bw_data bw_data;
	int ret;
	int i;

	memset(&bw_data, 0, sizeof(bw_data));
	bw_data.tc_valid_bits = ch->enabled_tc;
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
		bw_data.tc_bw_credits[i] = bw_share[i];

	ret = i40e_aq_config_vsi_tc_bw(&vsi->back->hw, ch->seid,
				       &bw_data, NULL);
	if (ret) {
		dev_info(&vsi->back->pdev->dev,
			 "Config VSI BW allocation per TC failed, aq_err: %d for new_vsi->seid %u\n",
			 vsi->back->hw.aq.asq_last_status, ch->seid);
		return -EINVAL;
	}

	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++)
		ch->info.qs_handle[i] = bw_data.qs_handles[i];

	return 0;
}

/**
 * i40e_channel_config_tx_ring - config TX ring associated with new channel
 * @pf: ptr to PF device
 * @vsi: the VSI being setup
 * @ch: ptr to channel structure
 *
 * Configure TX rings associated with channel (VSI) since queues are being
 * from parent VSI.
 **/
static int i40e_channel_config_tx_ring(struct i40e_pf *pf,
				       struct i40e_vsi *vsi,
				       struct i40e_channel *ch)
{
	u8 bw_share[I40E_MAX_TRAFFIC_CLASS] = {0};
	int ret;
	int i;

	/* Enable ETS TCs with equal BW Share for now across all VSIs */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		if (ch->enabled_tc & BIT(i))
			bw_share[i] = 1;
	}

	/* configure BW for new VSI */
	ret = i40e_channel_config_bw(vsi, ch, bw_share);
	if (ret) {
		dev_info(&vsi->back->pdev->dev,
			 "Failed configuring TC map %d for channel (seid %u)\n",
			 ch->enabled_tc, ch->seid);
		return ret;
	}

	for (i = 0; i < ch->num_queue_pairs; i++) {
		struct i40e_ring *tx_ring, *rx_ring;
		u16 pf_q;

		pf_q = ch->base_queue + i;

		/* Get to TX ring ptr of main VSI, for re-setup TX queue
		 * context
		 */
		tx_ring = vsi->tx_rings[pf_q];
		tx_ring->ch = ch;

		/* Get the RX ring ptr */
		rx_ring = vsi->rx_rings[pf_q];
		rx_ring->ch = ch;
	}

	return 0;
}

/**
 * i40e_setup_hw_channel - setup new channel
 * @pf: ptr to PF device
 * @vsi: the VSI being setup
 * @ch: ptr to channel structure
 * @uplink_seid: underlying HW switching element (VEB) 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 inline int i40e_setup_hw_channel(struct i40e_pf *pf,
					struct i40e_vsi *vsi,
					struct i40e_channel *ch,
					u16 uplink_seid, u8 type)
{
	int ret;

	ch->initialized = false;
	ch->base_queue = vsi->next_base_queue;
	ch->type = type;

	/* Proceed with creation of channel (VMDq2) VSI */
	ret = i40e_add_channel(pf, uplink_seid, ch);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "failed to add_channel using uplink_seid %u\n",
			 uplink_seid);
		return ret;
	}

	/* Mark the successful creation of channel */
	ch->initialized = true;

	/* Reconfigure TX queues using QTX_CTL register */
	ret = i40e_channel_config_tx_ring(pf, vsi, ch);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "failed to configure TX rings for channel %u\n",
			 ch->seid);
		return ret;
	}

	/* update 'next_base_queue' */
	vsi->next_base_queue = vsi->next_base_queue + ch->num_queue_pairs;
	dev_dbg(&pf->pdev->dev,
		"Added channel: vsi_seid %u, vsi_number %u, stat_counter_idx %u, num_queue_pairs %u, pf->next_base_queue %d\n",
		ch->seid, ch->vsi_number, ch->stat_counter_idx,
		ch->num_queue_pairs,
		vsi->next_base_queue);
	return ret;
}

/**
 * i40e_setup_channel - setup new channel using uplink element
 * @pf: ptr to PF device
 * @vsi: pointer to the VSI to set up the channel within
 * @ch: ptr to channel structure
 *
 * Setup new channel (VSI) based on specified type (VMDq2/VF)
 * and uplink switching element (uplink_seid)
 **/
static bool i40e_setup_channel(struct i40e_pf *pf, struct i40e_vsi *vsi,
			       struct i40e_channel *ch)
{
	u8 vsi_type;
	u16 seid;
	int ret;

	if (vsi->type == I40E_VSI_MAIN) {
		vsi_type = I40E_VSI_VMDQ2;
	} else {
		dev_err(&pf->pdev->dev, "unsupported parent vsi type(%d)\n",
			vsi->type);
		return false;
	}

	/* underlying switching element */
	seid = pf->vsi[pf->lan_vsi]->uplink_seid;

	/* create channel (VSI), configure TX rings */
	ret = i40e_setup_hw_channel(pf, vsi, ch, seid, vsi_type);
	if (ret) {
		dev_err(&pf->pdev->dev, "failed to setup hw_channel\n");
		return false;
	}

	return ch->initialized ? true : false;
}

/**
 * i40e_validate_and_set_switch_mode - sets up switch mode correctly
 * @vsi: ptr to VSI which has PF backing
 *
 * Sets up switch mode correctly if it needs to be changed and perform
 * what are allowed modes.
 **/
static int i40e_validate_and_set_switch_mode(struct i40e_vsi *vsi)
{
	u8 mode;
	struct i40e_pf *pf = vsi->back;
	struct i40e_hw *hw = &pf->hw;
	int ret;

	ret = i40e_get_capabilities(pf, i40e_aqc_opc_list_dev_capabilities);
	if (ret)
		return -EINVAL;

	if (hw->dev_caps.switch_mode) {
		/* if switch mode is set, support mode2 (non-tunneled for
		 * cloud filter) for now
		 */
		u32 switch_mode = hw->dev_caps.switch_mode &
				  I40E_SWITCH_MODE_MASK;
		if (switch_mode >= I40E_CLOUD_FILTER_MODE1) {
			if (switch_mode == I40E_CLOUD_FILTER_MODE2)
				return 0;
			dev_err(&pf->pdev->dev,
				"Invalid switch_mode (%d), only non-tunneled mode for cloud filter is supported\n",
				hw->dev_caps.switch_mode);
			return -EINVAL;
		}
	}

	/* Set Bit 7 to be valid */
	mode = I40E_AQ_SET_SWITCH_BIT7_VALID;

	/* Set L4type for TCP support */
	mode |= I40E_AQ_SET_SWITCH_L4_TYPE_TCP;

	/* Set cloud filter mode */
	mode |= I40E_AQ_SET_SWITCH_MODE_NON_TUNNEL;

	/* Prep mode field for set_switch_config */
	ret = i40e_aq_set_switch_config(hw, pf->last_sw_conf_flags,
					pf->last_sw_conf_valid_flags,
					mode, NULL);
	if (ret && hw->aq.asq_last_status != I40E_AQ_RC_ESRCH)
		dev_err(&pf->pdev->dev,
			"couldn't set switch config bits, err %pe aq_err %s\n",
			ERR_PTR(ret),
			i40e_aq_str(hw,
				    hw->aq.asq_last_status));

	return ret;
}

/**
 * i40e_create_queue_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.
 **/
int i40e_create_queue_channel(struct i40e_vsi *vsi,
			      struct i40e_channel *ch)
{
	struct i40e_pf *pf = vsi->back;
	bool reconfig_rss;
	int err;

	if (!ch)
		return -EINVAL;

	if (!ch->num_queue_pairs) {
		dev_err(&pf->pdev->dev, "Invalid num_queues requested: %d\n",
			ch->num_queue_pairs);
		return -EINVAL;
	}

	/* validate user requested num_queues for channel */
	err = i40e_validate_num_queues(pf, ch->num_queue_pairs, vsi,
				       &reconfig_rss);
	if (err) {
		dev_info(&pf->pdev->dev, "Failed to validate num_queues (%d)\n",
			 ch->num_queue_pairs);
		return -EINVAL;
	}

	/* By default we are in VEPA mode, if this is the first VF/VMDq
	 * VSI to be added switch to VEB mode.
	 */

	if (!(pf->flags & I40E_FLAG_VEB_MODE_ENABLED)) {
		pf->flags |= I40E_FLAG_VEB_MODE_ENABLED;

		if (vsi->type == I40E_VSI_MAIN) {
			if (i40e_is_tc_mqprio_enabled(pf))
				i40e_do_reset(pf, I40E_PF_RESET_FLAG, true);
			else
				i40e_do_reset_safe(pf, I40E_PF_RESET_FLAG);
		}
		/* now onwards for main VSI, number of queues will be value
		 * of TC0's queue count
		 */
	}

	/* By this time, vsi->cnt_q_avail shall be set to non-zero and
	 * it should be more than num_queues
	 */
	if (!vsi->cnt_q_avail || vsi->cnt_q_avail < ch->num_queue_pairs) {
		dev_dbg(&pf->pdev->dev,
			"Error: cnt_q_avail (%u) less than num_queues %d\n",
			vsi->cnt_q_avail, ch->num_queue_pairs);
		return -EINVAL;
	}

	/* reconfig_rss only if vsi type is MAIN_VSI */
	if (reconfig_rss && (vsi->type == I40E_VSI_MAIN)) {
		err = i40e_vsi_reconfig_rss(vsi, ch->num_queue_pairs);
		if (err) {
			dev_info(&pf->pdev->dev,
				 "Error: unable to reconfig rss for num_queues (%u)\n",
				 ch->num_queue_pairs);
			return -EINVAL;
		}
	}

	if (!i40e_setup_channel(pf, vsi, ch)) {
		dev_info(&pf->pdev->dev, "Failed to setup channel\n");
		return -EINVAL;
	}

	dev_info(&pf->pdev->dev,
		 "Setup channel (id:%u) utilizing num_queues %d\n",
		 ch->seid, ch->num_queue_pairs);

	/* configure VSI for BW limit */
	if (ch->max_tx_rate) {
		u64 credits = ch->max_tx_rate;

		if (i40e_set_bw_limit(vsi, ch->seid, ch->max_tx_rate))
			return -EINVAL;

		do_div(credits, I40E_BW_CREDIT_DIVISOR);
		dev_dbg(&pf->pdev->dev,
			"Set tx rate of %llu Mbps (count of 50Mbps %llu) for vsi->seid %u\n",
			ch->max_tx_rate,
			credits,
			ch->seid);
	}

	/* in case of VF, this will be main SRIOV VSI */
	ch->parent_vsi = vsi;

	/* and update main_vsi's count for queue_available to use */
	vsi->cnt_q_avail -= ch->num_queue_pairs;

	return 0;
}

/**
 * i40e_configure_queue_channels - Add queue channel for the given TCs
 * @vsi: VSI to be configured
 *
 * Configures queue channel mapping to the given TCs
 **/
static int i40e_configure_queue_channels(struct i40e_vsi *vsi)
{
	struct i40e_channel *ch;
	u64 max_rate = 0;
	int ret = 0, i;

	/* Create app vsi with the TCs. Main VSI with TC0 is already set up */
	vsi->tc_seid_map[0] = vsi->seid;
	for (i = 1; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		if (vsi->tc_config.enabled_tc & BIT(i)) {
			ch = kzalloc(sizeof(*ch), GFP_KERNEL);
			if (!ch) {
				ret = -ENOMEM;
				goto err_free;
			}

			INIT_LIST_HEAD(&ch->list);
			ch->num_queue_pairs =
				vsi->tc_config.tc_info[i].qcount;
			ch->base_queue =
				vsi->tc_config.tc_info[i].qoffset;

			/* Bandwidth limit through tc interface is in bytes/s,
			 * change to Mbit/s
			 */
			max_rate = vsi->mqprio_qopt.max_rate[i];
			do_div(max_rate, I40E_BW_MBPS_DIVISOR);
			ch->max_tx_rate = max_rate;

			list_add_tail(&ch->list, &vsi->ch_list);

			ret = i40e_create_queue_channel(vsi, ch);
			if (ret) {
				dev_err(&vsi->back->pdev->dev,
					"Failed creating queue channel with TC%d: queues %d\n",
					i, ch->num_queue_pairs);
				goto err_free;
			}
			vsi->tc_seid_map[i] = ch->seid;
		}
	}

	/* reset to reconfigure TX queue contexts */
	i40e_do_reset(vsi->back, I40E_PF_RESET_FLAG, true);
	return ret;

err_free:
	i40e_remove_queue_channels(vsi);
	return ret;
}

/**
 * i40e_veb_config_tc - Configure TCs for given VEB
 * @veb: given VEB
 * @enabled_tc: TC bitmap
 *
 * Configures given TC bitmap for VEB (switching) element
 **/
int i40e_veb_config_tc(struct i40e_veb *veb, u8 enabled_tc)
{
	struct i40e_aqc_configure_switching_comp_bw_config_data bw_data = {0};
	struct i40e_pf *pf = veb->pf;
	int ret = 0;
	int i;

	/* No TCs or already enabled TCs just return */
	if (!enabled_tc || veb->enabled_tc == enabled_tc)
		return ret;

	bw_data.tc_valid_bits = enabled_tc;
	/* bw_data.absolute_credits is not set (relative) */

	/* Enable ETS TCs with equal BW Share for now */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		if (enabled_tc & BIT(i))
			bw_data.tc_bw_share_credits[i] = 1;
	}

	ret = i40e_aq_config_switch_comp_bw_config(&pf->hw, veb->seid,
						   &bw_data, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "VEB bw config failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		goto out;
	}

	/* Update the BW information */
	ret = i40e_veb_get_bw_info(veb);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Failed getting veb bw config, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
	}

out:
	return ret;
}

#ifdef CONFIG_I40E_DCB
/**
 * i40e_dcb_reconfigure - Reconfigure all VEBs and VSIs
 * @pf: PF struct
 *
 * Reconfigure VEB/VSIs on a given PF; it is assumed that
 * the caller would've quiesce all the VSIs before calling
 * this function
 **/
static void i40e_dcb_reconfigure(struct i40e_pf *pf)
{
	u8 tc_map = 0;
	int ret;
	u8 v;

	/* Enable the TCs available on PF to all VEBs */
	tc_map = i40e_pf_get_tc_map(pf);
	if (tc_map == I40E_DEFAULT_TRAFFIC_CLASS)
		return;

	for (v = 0; v < I40E_MAX_VEB; v++) {
		if (!pf->veb[v])
			continue;
		ret = i40e_veb_config_tc(pf->veb[v], tc_map);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "Failed configuring TC for VEB seid=%d\n",
				 pf->veb[v]->seid);
			/* Will try to configure as many components */
		}
	}

	/* Update each VSI */
	for (v = 0; v < pf->num_alloc_vsi; v++) {
		if (!pf->vsi[v])
			continue;

		/* - Enable all TCs for the LAN VSI
		 * - For all others keep them at TC0 for now
		 */
		if (v == pf->lan_vsi)
			tc_map = i40e_pf_get_tc_map(pf);
		else
			tc_map = I40E_DEFAULT_TRAFFIC_CLASS;

		ret = i40e_vsi_config_tc(pf->vsi[v], tc_map);
		if (ret) {
			dev_info(&pf->pdev->dev,
				 "Failed configuring TC for VSI seid=%d\n",
				 pf->vsi[v]->seid);
			/* Will try to configure as many components */
		} else {
			/* Re-configure VSI vectors based on updated TC map */
			i40e_vsi_map_rings_to_vectors(pf->vsi[v]);
			if (pf->vsi[v]->netdev)
				i40e_dcbnl_set_all(pf->vsi[v]);
		}
	}
}

/**
 * i40e_resume_port_tx - Resume port Tx
 * @pf: PF struct
 *
 * Resume a port's Tx and issue a PF reset in case of failure to
 * resume.
 **/
static int i40e_resume_port_tx(struct i40e_pf *pf)
{
	struct i40e_hw *hw = &pf->hw;
	int ret;

	ret = i40e_aq_resume_port_tx(hw, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Resume Port Tx failed, err %pe aq_err %s\n",
			  ERR_PTR(ret),
			  i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		/* Schedule PF reset to recover */
		set_bit(__I40E_PF_RESET_REQUESTED, pf->state);
		i40e_service_event_schedule(pf);
	}

	return ret;
}

/**
 * i40e_suspend_port_tx - Suspend port Tx
 * @pf: PF struct
 *
 * Suspend a port's Tx and issue a PF reset in case of failure.
 **/
static int i40e_suspend_port_tx(struct i40e_pf *pf)
{
	struct i40e_hw *hw = &pf->hw;
	int ret;

	ret = i40e_aq_suspend_port_tx(hw, pf->mac_seid, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Suspend Port Tx failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		/* Schedule PF reset to recover */
		set_bit(__I40E_PF_RESET_REQUESTED, pf->state);
		i40e_service_event_schedule(pf);
	}

	return ret;
}

/**
 * i40e_hw_set_dcb_config - Program new DCBX settings into HW
 * @pf: PF being configured
 * @new_cfg: New DCBX configuration
 *
 * Program DCB settings into HW and reconfigure VEB/VSIs on
 * given PF. Uses "Set LLDP MIB" AQC to program the hardware.
 **/
static int i40e_hw_set_dcb_config(struct i40e_pf *pf,
				  struct i40e_dcbx_config *new_cfg)
{
	struct i40e_dcbx_config *old_cfg = &pf->hw.local_dcbx_config;
	int ret;

	/* Check if need reconfiguration */
	if (!memcmp(&new_cfg, &old_cfg, sizeof(new_cfg))) {
		dev_dbg(&pf->pdev->dev, "No Change in DCB Config required.\n");
		return 0;
	}

	/* Config change disable all VSIs */
	i40e_pf_quiesce_all_vsi(pf);

	/* Copy the new config to the current config */
	*old_cfg = *new_cfg;
	old_cfg->etsrec = old_cfg->etscfg;
	ret = i40e_set_dcb_config(&pf->hw);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Set DCB Config failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		goto out;
	}

	/* Changes in configuration update VEB/VSI */
	i40e_dcb_reconfigure(pf);
out:
	/* In case of reset do not try to resume anything */
	if (!test_bit(__I40E_RESET_RECOVERY_PENDING, pf->state)) {
		/* Re-start the VSIs if disabled */
		ret = i40e_resume_port_tx(pf);
		/* In case of error no point in resuming VSIs */
		if (ret)
			goto err;
		i40e_pf_unquiesce_all_vsi(pf);
	}
err:
	return ret;
}

/**
 * i40e_hw_dcb_config - Program new DCBX settings into HW
 * @pf: PF being configured
 * @new_cfg: New DCBX configuration
 *
 * Program DCB settings into HW and reconfigure VEB/VSIs on
 * given PF
 **/
int i40e_hw_dcb_config(struct i40e_pf *pf, struct i40e_dcbx_config *new_cfg)
{
	struct i40e_aqc_configure_switching_comp_ets_data ets_data;
	u8 prio_type[I40E_MAX_TRAFFIC_CLASS] = {0};
	u32 mfs_tc[I40E_MAX_TRAFFIC_CLASS];
	struct i40e_dcbx_config *old_cfg;
	u8 mode[I40E_MAX_TRAFFIC_CLASS];
	struct i40e_rx_pb_config pb_cfg;
	struct i40e_hw *hw = &pf->hw;
	u8 num_ports = hw->num_ports;
	bool need_reconfig;
	int ret = -EINVAL;
	u8 lltc_map = 0;
	u8 tc_map = 0;
	u8 new_numtc;
	u8 i;

	dev_dbg(&pf->pdev->dev, "Configuring DCB registers directly\n");
	/* Un-pack information to Program ETS HW via shared API
	 * numtc, tcmap
	 * LLTC map
	 * ETS/NON-ETS arbiter mode
	 * max exponent (credit refills)
	 * Total number of ports
	 * PFC priority bit-map
	 * Priority Table
	 * BW % per TC
	 * Arbiter mode between UPs sharing same TC
	 * TSA table (ETS or non-ETS)
	 * EEE enabled or not
	 * MFS TC table
	 */

	new_numtc = i40e_dcb_get_num_tc(new_cfg);

	memset(&ets_data, 0, sizeof(ets_data));
	for (i = 0; i < new_numtc; i++) {
		tc_map |= BIT(i);
		switch (new_cfg->etscfg.tsatable[i]) {
		case I40E_IEEE_TSA_ETS:
			prio_type[i] = I40E_DCB_PRIO_TYPE_ETS;
			ets_data.tc_bw_share_credits[i] =
					new_cfg->etscfg.tcbwtable[i];
			break;
		case I40E_IEEE_TSA_STRICT:
			prio_type[i] = I40E_DCB_PRIO_TYPE_STRICT;
			lltc_map |= BIT(i);
			ets_data.tc_bw_share_credits[i] =
					I40E_DCB_STRICT_PRIO_CREDITS;
			break;
		default:
			/* Invalid TSA type */
			need_reconfig = false;
			goto out;
		}
	}

	old_cfg = &hw->local_dcbx_config;
	/* Check if need reconfiguration */
	need_reconfig = i40e_dcb_need_reconfig(pf, old_cfg, new_cfg);

	/* If needed, enable/disable frame tagging, disable all VSIs
	 * and suspend port tx
	 */
	if (need_reconfig) {
		/* Enable DCB tagging only when more than one TC */
		if (new_numtc > 1)
			pf->flags |= I40E_FLAG_DCB_ENABLED;
		else
			pf->flags &= ~I40E_FLAG_DCB_ENABLED;

		set_bit(__I40E_PORT_SUSPENDED, pf->state);
		/* Reconfiguration needed quiesce all VSIs */
		i40e_pf_quiesce_all_vsi(pf);
		ret = i40e_suspend_port_tx(pf);
		if (ret)
			goto err;
	}

	/* Configure Port ETS Tx Scheduler */
	ets_data.tc_valid_bits = tc_map;
	ets_data.tc_strict_priority_flags = lltc_map;
	ret = i40e_aq_config_switch_comp_ets
		(hw, pf->mac_seid, &ets_data,
		 i40e_aqc_opc_modify_switching_comp_ets, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "Modify Port ETS failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		goto out;
	}

	/* Configure Rx ETS HW */
	memset(&mode, I40E_DCB_ARB_MODE_ROUND_ROBIN, sizeof(mode));
	i40e_dcb_hw_set_num_tc(hw, new_numtc);
	i40e_dcb_hw_rx_fifo_config(hw, I40E_DCB_ARB_MODE_ROUND_ROBIN,
				   I40E_DCB_ARB_MODE_STRICT_PRIORITY,
				   I40E_DCB_DEFAULT_MAX_EXPONENT,
				   lltc_map);
	i40e_dcb_hw_rx_cmd_monitor_config(hw, new_numtc, num_ports);
	i40e_dcb_hw_rx_ets_bw_config(hw, new_cfg->etscfg.tcbwtable, mode,
				     prio_type);
	i40e_dcb_hw_pfc_config(hw, new_cfg->pfc.pfcenable,
			       new_cfg->etscfg.prioritytable);
	i40e_dcb_hw_rx_up2tc_config(hw, new_cfg->etscfg.prioritytable);

	/* Configure Rx Packet Buffers in HW */
	for (i = 0; i < I40E_MAX_TRAFFIC_CLASS; i++) {
		mfs_tc[i] = pf->vsi[pf->lan_vsi]->netdev->mtu;
		mfs_tc[i] += I40E_PACKET_HDR_PAD;
	}

	i40e_dcb_hw_calculate_pool_sizes(hw, num_ports,
					 false, new_cfg->pfc.pfcenable,
					 mfs_tc, &pb_cfg);
	i40e_dcb_hw_rx_pb_config(hw, &pf->pb_cfg, &pb_cfg);

	/* Update the local Rx Packet buffer config */
	pf->pb_cfg = pb_cfg;

	/* Inform the FW about changes to DCB configuration */
	ret = i40e_aq_dcb_updated(&pf->hw, NULL);
	if (ret) {
		dev_info(&pf->pdev->dev,
			 "DCB Updated failed, err %pe aq_err %s\n",
			 ERR_PTR(ret),
			 i40e_aq_str(&pf->hw, pf->hw.aq.asq_last_status));
		goto out;
	}

	/* Update the port DCBx configuration */
	*old_cfg = *new_cfg;

	/* Changes in configuration update VEB/VSI */
	i40e_dcb_reconfigure(pf);
out:
	/* Re-start the VSIs if disabled */
	if (need_reconfig) {
		ret = i40e_resume_port_tx(pf);

		clear_bit(__I40E_PORT_SUSPENDED, pf->state);
		/* In case of error no point in resuming VSIs */
		if (ret)
			goto err;

		/* Wait for the PF's queues to be disabled */
		ret = i40e_pf_wait_queues_disabled(pf);
		if (ret) {
			/* Schedule PF reset to recover */
			set_bit(__I40E_PF_RESET_REQUESTED, pf->state);
			i40e_service_event_schedule(pf);
			goto err;
		} else {
			i40e_pf_unquiesce_all_vsi(pf);
			set_bit(__I40E_CLIENT_SERVICE_REQUESTED, pf->state);
			set_bit(__I40E_CLIENT_L2_CHANGE, pf->state);
		}
		/* registers are set, lets apply */
		if (pf->hw_features & I40E_HW_USE_SET_LLDP_MIB)
			ret = i40e_hw_set_dcb_config(pf, new_cfg);
	}

err:
	return ret;
}

/**
 * i40e_dcb_sw_default_config - Set default DCB configuration when DCB in SW
 * @pf: PF being queried
 *
 * Set default DCB configuration in case DCB is to be done in SW.
 **/
int i40e_dcb_sw_default_config(struct i40e_pf *pf)
{
	struct i40e_dcbx_config *dcb_cfg = &pf->hw.local_dcbx_config;
	struct i40e_aqc_configure_switching_comp_ets_data ets_data;
	struct i40e_hw *hw = &pf->hw;
	int err;

	if (pf->hw_features & I40E_HW_USE_SET_LLDP_MIB) {
		/* Update the local cached instance with TC0 ETS */
		memset(&pf->tmp_cfg, 0, sizeof(struct i40e_dcbx_config));
		pf->tmp_cfg.etscfg.willing = I40E_IEEE_DEFAULT_ETS_WILLING;
		pf->tmp_cfg.etscfg.maxtcs = 0;
		pf->tmp_cfg.etscfg.tcbwtable[0] = I40E_IEEE_DEFAULT_ETS_TCBW;
		pf->tmp_cfg.etscfg.tsatable[0] = I40E_IEEE_TSA_ETS;
		pf->tmp_cfg.pfc.willing = I40E_IEEE_DEFAULT_PFC_WILLING;
		pf->tmp_cfg.pfc.pfccap = I40E_MAX_TRAFFIC_CLASS;
		/* FW needs one App to configure HW */
		pf->tmp_cfg.numapps = I40E_IEEE_DEFAULT_NUM_APPS;
		pf->tmp_cfg.app[0].selector = I40E_APP_SEL_ETHTYPE;
		pf->tmp_cfg.app[0].priority = I40E_IEEE_DEFAULT_APP_PRIO;
		pf->