/* Broadcom NetXtreme-C/E network driver.
 *
 * Copyright (c) 2014-2015 Broadcom Corporation
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation.
 */

#include <linux/module.h>

#include <linux/stringify.h>
#include <linux/kernel.h>
#include <linux/timer.h>
#include <linux/errno.h>
#include <linux/ioport.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/interrupt.h>
#include <linux/pci.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/skbuff.h>
#include <linux/dma-mapping.h>
#include <linux/bitops.h>
#include <linux/io.h>
#include <linux/irq.h>
#include <linux/delay.h>
#include <asm/byteorder.h>
#include <asm/page.h>
#include <linux/time.h>
#include <linux/mii.h>
#include <linux/if.h>
#include <linux/if_vlan.h>
#include <net/ip.h>
#include <net/tcp.h>
#include <net/udp.h>
#include <net/checksum.h>
#include <net/ip6_checksum.h>
#if defined(CONFIG_VXLAN) || defined(CONFIG_VXLAN_MODULE)
#include <net/vxlan.h>
#endif
#ifdef CONFIG_NET_RX_BUSY_POLL
#include <net/busy_poll.h>
#endif
#include <linux/workqueue.h>
#include <linux/prefetch.h>
#include <linux/cache.h>
#include <linux/log2.h>
#include <linux/aer.h>
#include <linux/bitmap.h>
#include <linux/cpu_rmap.h>

#include "bnxt_hsi.h"
#include "bnxt.h"
#include "bnxt_sriov.h"
#include "bnxt_ethtool.h"

#define BNXT_TX_TIMEOUT         (5 * HZ)

static const char version[] =
        "Broadcom NetXtreme-C/E driver " DRV_MODULE_NAME " v" DRV_MODULE_VERSION "\n";

MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("Broadcom BCM573xx network driver");
MODULE_VERSION(DRV_MODULE_VERSION);

#define BNXT_RX_OFFSET (NET_SKB_PAD + NET_IP_ALIGN)
#define BNXT_RX_DMA_OFFSET NET_SKB_PAD
#define BNXT_RX_COPY_THRESH 256

#define BNXT_TX_PUSH_THRESH 164

enum board_idx {
        BCM57301,
        BCM57302,
        BCM57304,
        BCM57402,
        BCM57404,
        BCM57406,
        BCM57304_VF,
        BCM57404_VF,
};

/* indexed by enum above */
static const struct {
        char *name;
} board_info[] = {
        { "Broadcom BCM57301 NetXtreme-C Single-port 10Gb Ethernet" },
        { "Broadcom BCM57302 NetXtreme-C Dual-port 10Gb/25Gb Ethernet" },
        { "Broadcom BCM57304 NetXtreme-C Dual-port 10Gb/25Gb/40Gb/50Gb Ethernet" },
        { "Broadcom BCM57402 NetXtreme-E Dual-port 10Gb Ethernet" },
        { "Broadcom BCM57404 NetXtreme-E Dual-port 10Gb/25Gb Ethernet" },
        { "Broadcom BCM57406 NetXtreme-E Dual-port 10GBase-T Ethernet" },
        { "Broadcom BCM57304 NetXtreme-C Ethernet Virtual Function" },
        { "Broadcom BCM57404 NetXtreme-E Ethernet Virtual Function" },
};

static const struct pci_device_id bnxt_pci_tbl[] = {
        { PCI_VDEVICE(BROADCOM, 0x16c8), .driver_data = BCM57301 },
        { PCI_VDEVICE(BROADCOM, 0x16c9), .driver_data = BCM57302 },
        { PCI_VDEVICE(BROADCOM, 0x16ca), .driver_data = BCM57304 },
        { PCI_VDEVICE(BROADCOM, 0x16d0), .driver_data = BCM57402 },
        { PCI_VDEVICE(BROADCOM, 0x16d1), .driver_data = BCM57404 },
        { PCI_VDEVICE(BROADCOM, 0x16d2), .driver_data = BCM57406 },
#ifdef CONFIG_BNXT_SRIOV
        { PCI_VDEVICE(BROADCOM, 0x16cb), .driver_data = BCM57304_VF },
        { PCI_VDEVICE(BROADCOM, 0x16d3), .driver_data = BCM57404_VF },
#endif
        { 0 }
};

MODULE_DEVICE_TABLE(pci, bnxt_pci_tbl);

static const u16 bnxt_vf_req_snif[] = {
        HWRM_FUNC_CFG,
        HWRM_PORT_PHY_QCFG,
        HWRM_CFA_L2_FILTER_ALLOC,
};

static bool bnxt_vf_pciid(enum board_idx idx)
{
        return (idx == BCM57304_VF || idx == BCM57404_VF);
}

#define DB_CP_REARM_FLAGS       (DB_KEY_CP | DB_IDX_VALID)
#define DB_CP_FLAGS             (DB_KEY_CP | DB_IDX_VALID | DB_IRQ_DIS)
#define DB_CP_IRQ_DIS_FLAGS     (DB_KEY_CP | DB_IRQ_DIS)

#define BNXT_CP_DB_REARM(db, raw_cons)                                  \
                writel(DB_CP_REARM_FLAGS | RING_CMP(raw_cons), db)

#define BNXT_CP_DB(db, raw_cons)                                        \
                writel(DB_CP_FLAGS | RING_CMP(raw_cons), db)

#define BNXT_CP_DB_IRQ_DIS(db)                                          \
                writel(DB_CP_IRQ_DIS_FLAGS, db)

static inline u32 bnxt_tx_avail(struct bnxt *bp, struct bnxt_tx_ring_info *txr)
{
        /* Tell compiler to fetch tx indices from memory. */
        barrier();

        return bp->tx_ring_size -
                ((txr->tx_prod - txr->tx_cons) & bp->tx_ring_mask);
}

static const u16 bnxt_lhint_arr[] = {
        TX_BD_FLAGS_LHINT_512_AND_SMALLER,
        TX_BD_FLAGS_LHINT_512_TO_1023,
        TX_BD_FLAGS_LHINT_1024_TO_2047,
        TX_BD_FLAGS_LHINT_1024_TO_2047,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
        TX_BD_FLAGS_LHINT_2048_AND_LARGER,
};

static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
        struct bnxt *bp = netdev_priv(dev);
        struct tx_bd *txbd;
        struct tx_bd_ext *txbd1;
        struct netdev_queue *txq;
        int i;
        dma_addr_t mapping;
        unsigned int length, pad = 0;
        u32 len, free_size, vlan_tag_flags, cfa_action, flags;
        u16 prod, last_frag;
        struct pci_dev *pdev = bp->pdev;
        struct bnxt_tx_ring_info *txr;
        struct bnxt_sw_tx_bd *tx_buf;

        i = skb_get_queue_mapping(skb);
        if (unlikely(i >= bp->tx_nr_rings)) {
                dev_kfree_skb_any(skb);
                return NETDEV_TX_OK;
        }

        txr = &bp->tx_ring[i];
        txq = netdev_get_tx_queue(dev, i);
        prod = txr->tx_prod;

        free_size = bnxt_tx_avail(bp, txr);
        if (unlikely(free_size < skb_shinfo(skb)->nr_frags + 2)) {
                netif_tx_stop_queue(txq);
                return NETDEV_TX_BUSY;
        }

        length = skb->len;
        len = skb_headlen(skb);
        last_frag = skb_shinfo(skb)->nr_frags;

        txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];

        txbd->tx_bd_opaque = prod;

        tx_buf = &txr->tx_buf_ring[prod];
        tx_buf->skb = skb;
        tx_buf->nr_frags = last_frag;

        vlan_tag_flags = 0;
        cfa_action = 0;
        if (skb_vlan_tag_present(skb)) {
                vlan_tag_flags = TX_BD_CFA_META_KEY_VLAN |
                                 skb_vlan_tag_get(skb);
                /* Currently supports 8021Q, 8021AD vlan offloads
                 * QINQ1, QINQ2, QINQ3 vlan headers are deprecated
                 */
                if (skb->vlan_proto == htons(ETH_P_8021Q))
                        vlan_tag_flags |= 1 << TX_BD_CFA_META_TPID_SHIFT;
        }

        if (free_size == bp->tx_ring_size && length <= bp->tx_push_thresh) {
                struct tx_push_buffer *tx_push_buf = txr->tx_push;
                struct tx_push_bd *tx_push = &tx_push_buf->push_bd;
                struct tx_bd_ext *tx_push1 = &tx_push->txbd2;
                void *pdata = tx_push_buf->data;
                u64 *end;
                int j, push_len;

                /* Set COAL_NOW to be ready quickly for the next push */
                tx_push->tx_bd_len_flags_type =
                        cpu_to_le32((length << TX_BD_LEN_SHIFT) |
                                        TX_BD_TYPE_LONG_TX_BD |
                                        TX_BD_FLAGS_LHINT_512_AND_SMALLER |
                                        TX_BD_FLAGS_COAL_NOW |
                                        TX_BD_FLAGS_PACKET_END |
                                        (2 << TX_BD_FLAGS_BD_CNT_SHIFT));

                if (skb->ip_summed == CHECKSUM_PARTIAL)
                        tx_push1->tx_bd_hsize_lflags =
                                        cpu_to_le32(TX_BD_FLAGS_TCP_UDP_CHKSUM);
                else
                        tx_push1->tx_bd_hsize_lflags = 0;

                tx_push1->tx_bd_cfa_meta = cpu_to_le32(vlan_tag_flags);
                tx_push1->tx_bd_cfa_action = cpu_to_le32(cfa_action);

                end = pdata + length;
                end = PTR_ALIGN(end, 8) - 1;
                *end = 0;

                skb_copy_from_linear_data(skb, pdata, len);
                pdata += len;
                for (j = 0; j < last_frag; j++) {
                        skb_frag_t *frag = &skb_shinfo(skb)->frags[j];
                        void *fptr;

                        fptr = skb_frag_address_safe(frag);
                        if (!fptr)
                                goto normal_tx;

                        memcpy(pdata, fptr, skb_frag_size(frag));
                        pdata += skb_frag_size(frag);
                }

                txbd->tx_bd_len_flags_type = tx_push->tx_bd_len_flags_type;
                txbd->tx_bd_haddr = txr->data_mapping;
                prod = NEXT_TX(prod);
                txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
                memcpy(txbd, tx_push1, sizeof(*txbd));
                prod = NEXT_TX(prod);
                tx_push->doorbell =
                        cpu_to_le32(DB_KEY_TX_PUSH | DB_LONG_TX_PUSH | prod);
                txr->tx_prod = prod;

                netdev_tx_sent_queue(txq, skb->len);

                push_len = (length + sizeof(*tx_push) + 7) / 8;
                if (push_len > 16) {
                        __iowrite64_copy(txr->tx_doorbell, tx_push_buf, 16);
                        __iowrite64_copy(txr->tx_doorbell + 4, tx_push_buf + 1,
                                         push_len - 16);
                } else {
                        __iowrite64_copy(txr->tx_doorbell, tx_push_buf,
                                         push_len);
                }

                tx_buf->is_push = 1;
                goto tx_done;
        }

normal_tx:
        if (length < BNXT_MIN_PKT_SIZE) {
                pad = BNXT_MIN_PKT_SIZE - length;
                if (skb_pad(skb, pad)) {
                        /* SKB already freed. */
                        tx_buf->skb = NULL;
                        return NETDEV_TX_OK;
                }
                length = BNXT_MIN_PKT_SIZE;
        }

        mapping = dma_map_single(&pdev->dev, skb->data, len, DMA_TO_DEVICE);

        if (unlikely(dma_mapping_error(&pdev->dev, mapping))) {
                dev_kfree_skb_any(skb);
                tx_buf->skb = NULL;
                return NETDEV_TX_OK;
        }

        dma_unmap_addr_set(tx_buf, mapping, mapping);
        flags = (len << TX_BD_LEN_SHIFT) | TX_BD_TYPE_LONG_TX_BD |
                ((last_frag + 2) << TX_BD_FLAGS_BD_CNT_SHIFT);

        txbd->tx_bd_haddr = cpu_to_le64(mapping);

        prod = NEXT_TX(prod);
        txbd1 = (struct tx_bd_ext *)
                &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];

        txbd1->tx_bd_hsize_lflags = 0;
        if (skb_is_gso(skb)) {
                u32 hdr_len;

                if (skb->encapsulation)
                        hdr_len = skb_inner_network_offset(skb) +
                                skb_inner_network_header_len(skb) +
                                inner_tcp_hdrlen(skb);
                else
                        hdr_len = skb_transport_offset(skb) +
                                tcp_hdrlen(skb);

                txbd1->tx_bd_hsize_lflags = cpu_to_le32(TX_BD_FLAGS_LSO |
                                        TX_BD_FLAGS_T_IPID |
                                        (hdr_len << (TX_BD_HSIZE_SHIFT - 1)));
                length = skb_shinfo(skb)->gso_size;
                txbd1->tx_bd_mss = cpu_to_le32(length);
                length += hdr_len;
        } else if (skb->ip_summed == CHECKSUM_PARTIAL) {
                txbd1->tx_bd_hsize_lflags =
                        cpu_to_le32(TX_BD_FLAGS_TCP_UDP_CHKSUM);
                txbd1->tx_bd_mss = 0;
        }

        length >>= 9;
        flags |= bnxt_lhint_arr[length];
        txbd->tx_bd_len_flags_type = cpu_to_le32(flags);

        txbd1->tx_bd_cfa_meta = cpu_to_le32(vlan_tag_flags);
        txbd1->tx_bd_cfa_action = cpu_to_le32(cfa_action);
        for (i = 0; i < last_frag; i++) {
                skb_frag_t *frag = &skb_shinfo(skb)->frags[i];

                prod = NEXT_TX(prod);
                txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];

                len = skb_frag_size(frag);
                mapping = skb_frag_dma_map(&pdev->dev, frag, 0, len,
                                           DMA_TO_DEVICE);

                if (unlikely(dma_mapping_error(&pdev->dev, mapping)))
                        goto tx_dma_error;

                tx_buf = &txr->tx_buf_ring[prod];
                dma_unmap_addr_set(tx_buf, mapping, mapping);

                txbd->tx_bd_haddr = cpu_to_le64(mapping);

                flags = len << TX_BD_LEN_SHIFT;
                txbd->tx_bd_len_flags_type = cpu_to_le32(flags);
        }

        flags &= ~TX_BD_LEN;
        txbd->tx_bd_len_flags_type =
                cpu_to_le32(((len + pad) << TX_BD_LEN_SHIFT) | flags |
                            TX_BD_FLAGS_PACKET_END);

        netdev_tx_sent_queue(txq, skb->len);

        /* Sync BD data before updating doorbell */
        wmb();

        prod = NEXT_TX(prod);
        txr->tx_prod = prod;

        writel(DB_KEY_TX | prod, txr->tx_doorbell);
        writel(DB_KEY_TX | prod, txr->tx_doorbell);

tx_done:

        mmiowb();

        if (unlikely(bnxt_tx_avail(bp, txr) <= MAX_SKB_FRAGS + 1)) {
                netif_tx_stop_queue(txq);

                /* netif_tx_stop_queue() must be done before checking
                 * tx index in bnxt_tx_avail() below, because in
                 * bnxt_tx_int(), we update tx index before checking for
                 * netif_tx_queue_stopped().
                 */
                smp_mb();
                if (bnxt_tx_avail(bp, txr) > bp->tx_wake_thresh)
                        netif_tx_wake_queue(txq);
        }
        return NETDEV_TX_OK;

tx_dma_error:
        last_frag = i;

        /* start back at beginning and unmap skb */
        prod = txr->tx_prod;
        tx_buf = &txr->tx_buf_ring[prod];
        tx_buf->skb = NULL;
        dma_unmap_single(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
                         skb_headlen(skb), PCI_DMA_TODEVICE);
        prod = NEXT_TX(prod);

        /* unmap remaining mapped pages */
        for (i = 0; i < last_frag; i++) {
                prod = NEXT_TX(prod);
                tx_buf = &txr->tx_buf_ring[prod];
                dma_unmap_page(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
                               skb_frag_size(&skb_shinfo(skb)->frags[i]),
                               PCI_DMA_TODEVICE);
        }

        dev_kfree_skb_any(skb);
        return NETDEV_TX_OK;
}

static void bnxt_tx_int(struct bnxt *bp, struct bnxt_napi *bnapi, int nr_pkts)
{
        struct bnxt_tx_ring_info *txr = bnapi->tx_ring;
        int index = txr - &bp->tx_ring[0];
        struct netdev_queue *txq = netdev_get_tx_queue(bp->dev, index);
        u16 cons = txr->tx_cons;
        struct pci_dev *pdev = bp->pdev;
        int i;
        unsigned int tx_bytes = 0;

        for (i = 0; i < nr_pkts; i++) {
                struct bnxt_sw_tx_bd *tx_buf;
                struct sk_buff *skb;
                int j, last;

                tx_buf = &txr->tx_buf_ring[cons];
                cons = NEXT_TX(cons);
                skb = tx_buf->skb;
                tx_buf->skb = NULL;

                if (tx_buf->is_push) {
                        tx_buf->is_push = 0;
                        goto next_tx_int;
                }

                dma_unmap_single(&pdev->dev, dma_unmap_addr(tx_buf, mapping),
                                 skb_headlen(skb), PCI_DMA_TODEVICE);
                last = tx_buf->nr_frags;

                for (j = 0; j < last; j++) {
                        cons = NEXT_TX(cons);
                        tx_buf = &txr->tx_buf_ring[cons];
                        dma_unmap_page(
                                &pdev->dev,
                                dma_unmap_addr(tx_buf, mapping),
                                skb_frag_size(&skb_shinfo(skb)->frags[j]),
                                PCI_DMA_TODEVICE);
                }

next_tx_int:
                cons = NEXT_TX(cons);

                tx_bytes += skb->len;
                dev_kfree_skb_any(skb);
        }

        netdev_tx_completed_queue(txq, nr_pkts, tx_bytes);
        txr->tx_cons = cons;

        /* Need to make the tx_cons update visible to bnxt_start_xmit()
         * before checking for netif_tx_queue_stopped().  Without the
         * memory barrier, there is a small possibility that bnxt_start_xmit()
         * will miss it and cause the queue to be stopped forever.
         */
        smp_mb();

        if (unlikely(netif_tx_queue_stopped(txq)) &&
            (bnxt_tx_avail(bp, txr) > bp->tx_wake_thresh)) {
                __netif_tx_lock(txq, smp_processor_id());
                if (netif_tx_queue_stopped(txq) &&
                    bnxt_tx_avail(bp, txr) > bp->tx_wake_thresh &&
                    txr->dev_state != BNXT_DEV_STATE_CLOSING)
                        netif_tx_wake_queue(txq);
                __netif_tx_unlock(txq);
        }
}

static inline u8 *__bnxt_alloc_rx_data(struct bnxt *bp, dma_addr_t *mapping,
                                       gfp_t gfp)
{
        u8 *data;
        struct pci_dev *pdev = bp->pdev;

        data = kmalloc(bp->rx_buf_size, gfp);
        if (!data)
                return NULL;

        *mapping = dma_map_single(&pdev->dev, data + BNXT_RX_DMA_OFFSET,
                                  bp->rx_buf_use_size, PCI_DMA_FROMDEVICE);

        if (dma_mapping_error(&pdev->dev, *mapping)) {
                kfree(data);
                data = NULL;
        }
        return data;
}

static inline int bnxt_alloc_rx_data(struct bnxt *bp,
                                     struct bnxt_rx_ring_info *rxr,
                                     u16 prod, gfp_t gfp)
{
        struct rx_bd *rxbd = &rxr->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)];
        struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[prod];
        u8 *data;
        dma_addr_t mapping;

        data = __bnxt_alloc_rx_data(bp, &mapping, gfp);
        if (!data)
                return -ENOMEM;

        rx_buf->data = data;
        dma_unmap_addr_set(rx_buf, mapping, mapping);

        rxbd->rx_bd_haddr = cpu_to_le64(mapping);

        return 0;
}

static void bnxt_reuse_rx_data(struct bnxt_rx_ring_info *rxr, u16 cons,
                               u8 *data)
{
        u16 prod = rxr->rx_prod;
        struct bnxt_sw_rx_bd *cons_rx_buf, *prod_rx_buf;
        struct rx_bd *cons_bd, *prod_bd;

        prod_rx_buf = &rxr->rx_buf_ring[prod];
        cons_rx_buf = &rxr->rx_buf_ring[cons];

        prod_rx_buf->data = data;

        dma_unmap_addr_set(prod_rx_buf, mapping,
                           dma_unmap_addr(cons_rx_buf, mapping));

        prod_bd = &rxr->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)];
        cons_bd = &rxr->rx_desc_ring[RX_RING(cons)][RX_IDX(cons)];

        prod_bd->rx_bd_haddr = cons_bd->rx_bd_haddr;
}

static inline u16 bnxt_find_next_agg_idx(struct bnxt_rx_ring_info *rxr, u16 idx)
{
        u16 next, max = rxr->rx_agg_bmap_size;

        next = find_next_zero_bit(rxr->rx_agg_bmap, max, idx);
        if (next >= max)
                next = find_first_zero_bit(rxr->rx_agg_bmap, max);
        return next;
}

static inline int bnxt_alloc_rx_page(struct bnxt *bp,
                                     struct bnxt_rx_ring_info *rxr,
                                     u16 prod, gfp_t gfp)
{
        struct rx_bd *rxbd =
                &rxr->rx_agg_desc_ring[RX_RING(prod)][RX_IDX(prod)];
        struct bnxt_sw_rx_agg_bd *rx_agg_buf;
        struct pci_dev *pdev = bp->pdev;
        struct page *page;
        dma_addr_t mapping;
        u16 sw_prod = rxr->rx_sw_agg_prod;
        unsigned int offset = 0;

        if (PAGE_SIZE > BNXT_RX_PAGE_SIZE) {
                page = rxr->rx_page;
                if (!page) {
                        page = alloc_page(gfp);
                        if (!page)
                                return -ENOMEM;
                        rxr->rx_page = page;
                        rxr->rx_page_offset = 0;
                }
                offset = rxr->rx_page_offset;
                rxr->rx_page_offset += BNXT_RX_PAGE_SIZE;
                if (rxr->rx_page_offset == PAGE_SIZE)
                        rxr->rx_page = NULL;
                else
                        get_page(page);
        } else {
                page = alloc_page(gfp);
                if (!page)
                        return -ENOMEM;
        }

        mapping = dma_map_page(&pdev->dev, page, offset, BNXT_RX_PAGE_SIZE,
                               PCI_DMA_FROMDEVICE);
        if (dma_mapping_error(&pdev->dev, mapping)) {
                __free_page(page);
                return -EIO;
        }

        if (unlikely(test_bit(sw_prod, rxr->rx_agg_bmap)))
                sw_prod = bnxt_find_next_agg_idx(rxr, sw_prod);

        __set_bit(sw_prod, rxr->rx_agg_bmap);
        rx_agg_buf = &rxr->rx_agg_ring[sw_prod];
        rxr->rx_sw_agg_prod = NEXT_RX_AGG(sw_prod);

        rx_agg_buf->page = page;
        rx_agg_buf->offset = offset;
        rx_agg_buf->mapping = mapping;
        rxbd->rx_bd_haddr = cpu_to_le64(mapping);
        rxbd->rx_bd_opaque = sw_prod;
        return 0;
}

static void bnxt_reuse_rx_agg_bufs(struct bnxt_napi *bnapi, u16 cp_cons,
                                   u32 agg_bufs)
{
        struct bnxt *bp = bnapi->bp;
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
        u16 prod = rxr->rx_agg_prod;
        u16 sw_prod = rxr->rx_sw_agg_prod;
        u32 i;

        for (i = 0; i < agg_bufs; i++) {
                u16 cons;
                struct rx_agg_cmp *agg;
                struct bnxt_sw_rx_agg_bd *cons_rx_buf, *prod_rx_buf;
                struct rx_bd *prod_bd;
                struct page *page;

                agg = (struct rx_agg_cmp *)
                        &cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
                cons = agg->rx_agg_cmp_opaque;
                __clear_bit(cons, rxr->rx_agg_bmap);

                if (unlikely(test_bit(sw_prod, rxr->rx_agg_bmap)))
                        sw_prod = bnxt_find_next_agg_idx(rxr, sw_prod);

                __set_bit(sw_prod, rxr->rx_agg_bmap);
                prod_rx_buf = &rxr->rx_agg_ring[sw_prod];
                cons_rx_buf = &rxr->rx_agg_ring[cons];

                /* It is possible for sw_prod to be equal to cons, so
                 * set cons_rx_buf->page to NULL first.
                 */
                page = cons_rx_buf->page;
                cons_rx_buf->page = NULL;
                prod_rx_buf->page = page;
                prod_rx_buf->offset = cons_rx_buf->offset;

                prod_rx_buf->mapping = cons_rx_buf->mapping;

                prod_bd = &rxr->rx_agg_desc_ring[RX_RING(prod)][RX_IDX(prod)];

                prod_bd->rx_bd_haddr = cpu_to_le64(cons_rx_buf->mapping);
                prod_bd->rx_bd_opaque = sw_prod;

                prod = NEXT_RX_AGG(prod);
                sw_prod = NEXT_RX_AGG(sw_prod);
                cp_cons = NEXT_CMP(cp_cons);
        }
        rxr->rx_agg_prod = prod;
        rxr->rx_sw_agg_prod = sw_prod;
}

static struct sk_buff *bnxt_rx_skb(struct bnxt *bp,
                                   struct bnxt_rx_ring_info *rxr, u16 cons,
                                   u16 prod, u8 *data, dma_addr_t dma_addr,
                                   unsigned int len)
{
        int err;
        struct sk_buff *skb;

        err = bnxt_alloc_rx_data(bp, rxr, prod, GFP_ATOMIC);
        if (unlikely(err)) {
                bnxt_reuse_rx_data(rxr, cons, data);
                return NULL;
        }

        skb = build_skb(data, 0);
        dma_unmap_single(&bp->pdev->dev, dma_addr, bp->rx_buf_use_size,
                         PCI_DMA_FROMDEVICE);
        if (!skb) {
                kfree(data);
                return NULL;
        }

        skb_reserve(skb, BNXT_RX_OFFSET);
        skb_put(skb, len);
        return skb;
}

static struct sk_buff *bnxt_rx_pages(struct bnxt *bp, struct bnxt_napi *bnapi,
                                     struct sk_buff *skb, u16 cp_cons,
                                     u32 agg_bufs)
{
        struct pci_dev *pdev = bp->pdev;
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
        u16 prod = rxr->rx_agg_prod;
        u32 i;

        for (i = 0; i < agg_bufs; i++) {
                u16 cons, frag_len;
                struct rx_agg_cmp *agg;
                struct bnxt_sw_rx_agg_bd *cons_rx_buf;
                struct page *page;
                dma_addr_t mapping;

                agg = (struct rx_agg_cmp *)
                        &cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];
                cons = agg->rx_agg_cmp_opaque;
                frag_len = (le32_to_cpu(agg->rx_agg_cmp_len_flags_type) &
                            RX_AGG_CMP_LEN) >> RX_AGG_CMP_LEN_SHIFT;

                cons_rx_buf = &rxr->rx_agg_ring[cons];
                skb_fill_page_desc(skb, i, cons_rx_buf->page,
                                   cons_rx_buf->offset, frag_len);
                __clear_bit(cons, rxr->rx_agg_bmap);

                /* It is possible for bnxt_alloc_rx_page() to allocate
                 * a sw_prod index that equals the cons index, so we
                 * need to clear the cons entry now.
                 */
                mapping = dma_unmap_addr(cons_rx_buf, mapping);
                page = cons_rx_buf->page;
                cons_rx_buf->page = NULL;

                if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_ATOMIC) != 0) {
                        struct skb_shared_info *shinfo;
                        unsigned int nr_frags;

                        shinfo = skb_shinfo(skb);
                        nr_frags = --shinfo->nr_frags;
                        __skb_frag_set_page(&shinfo->frags[nr_frags], NULL);

                        dev_kfree_skb(skb);

                        cons_rx_buf->page = page;

                        /* Update prod since possibly some pages have been
                         * allocated already.
                         */
                        rxr->rx_agg_prod = prod;
                        bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs - i);
                        return NULL;
                }

                dma_unmap_page(&pdev->dev, mapping, BNXT_RX_PAGE_SIZE,
                               PCI_DMA_FROMDEVICE);

                skb->data_len += frag_len;
                skb->len += frag_len;
                skb->truesize += PAGE_SIZE;

                prod = NEXT_RX_AGG(prod);
                cp_cons = NEXT_CMP(cp_cons);
        }
        rxr->rx_agg_prod = prod;
        return skb;
}

static int bnxt_agg_bufs_valid(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
                               u8 agg_bufs, u32 *raw_cons)
{
        u16 last;
        struct rx_agg_cmp *agg;

        *raw_cons = ADV_RAW_CMP(*raw_cons, agg_bufs);
        last = RING_CMP(*raw_cons);
        agg = (struct rx_agg_cmp *)
                &cpr->cp_desc_ring[CP_RING(last)][CP_IDX(last)];
        return RX_AGG_CMP_VALID(agg, *raw_cons);
}

static inline struct sk_buff *bnxt_copy_skb(struct bnxt_napi *bnapi, u8 *data,
                                            unsigned int len,
                                            dma_addr_t mapping)
{
        struct bnxt *bp = bnapi->bp;
        struct pci_dev *pdev = bp->pdev;
        struct sk_buff *skb;

        skb = napi_alloc_skb(&bnapi->napi, len);
        if (!skb)
                return NULL;

        dma_sync_single_for_cpu(&pdev->dev, mapping,
                                bp->rx_copy_thresh, PCI_DMA_FROMDEVICE);

        memcpy(skb->data - BNXT_RX_OFFSET, data, len + BNXT_RX_OFFSET);

        dma_sync_single_for_device(&pdev->dev, mapping,
                                   bp->rx_copy_thresh,
                                   PCI_DMA_FROMDEVICE);

        skb_put(skb, len);
        return skb;
}

static int bnxt_discard_rx(struct bnxt *bp, struct bnxt_napi *bnapi,
                           u32 *raw_cons, void *cmp)
{
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        struct rx_cmp *rxcmp = cmp;
        u32 tmp_raw_cons = *raw_cons;
        u8 cmp_type, agg_bufs = 0;

        cmp_type = RX_CMP_TYPE(rxcmp);

        if (cmp_type == CMP_TYPE_RX_L2_CMP) {
                agg_bufs = (le32_to_cpu(rxcmp->rx_cmp_misc_v1) &
                            RX_CMP_AGG_BUFS) >>
                           RX_CMP_AGG_BUFS_SHIFT;
        } else if (cmp_type == CMP_TYPE_RX_L2_TPA_END_CMP) {
                struct rx_tpa_end_cmp *tpa_end = cmp;

                agg_bufs = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) &
                            RX_TPA_END_CMP_AGG_BUFS) >>
                           RX_TPA_END_CMP_AGG_BUFS_SHIFT;
        }

        if (agg_bufs) {
                if (!bnxt_agg_bufs_valid(bp, cpr, agg_bufs, &tmp_raw_cons))
                        return -EBUSY;
        }
        *raw_cons = tmp_raw_cons;
        return 0;
}

static void bnxt_sched_reset(struct bnxt *bp, struct bnxt_rx_ring_info *rxr)
{
        if (!rxr->bnapi->in_reset) {
                rxr->bnapi->in_reset = true;
                set_bit(BNXT_RESET_TASK_SP_EVENT, &bp->sp_event);
                schedule_work(&bp->sp_task);
        }
        rxr->rx_next_cons = 0xffff;
}

static void bnxt_tpa_start(struct bnxt *bp, struct bnxt_rx_ring_info *rxr,
                           struct rx_tpa_start_cmp *tpa_start,
                           struct rx_tpa_start_cmp_ext *tpa_start1)
{
        u8 agg_id = TPA_START_AGG_ID(tpa_start);
        u16 cons, prod;
        struct bnxt_tpa_info *tpa_info;
        struct bnxt_sw_rx_bd *cons_rx_buf, *prod_rx_buf;
        struct rx_bd *prod_bd;
        dma_addr_t mapping;

        cons = tpa_start->rx_tpa_start_cmp_opaque;
        prod = rxr->rx_prod;
        cons_rx_buf = &rxr->rx_buf_ring[cons];
        prod_rx_buf = &rxr->rx_buf_ring[prod];
        tpa_info = &rxr->rx_tpa[agg_id];

        if (unlikely(cons != rxr->rx_next_cons)) {
                bnxt_sched_reset(bp, rxr);
                return;
        }

        prod_rx_buf->data = tpa_info->data;

        mapping = tpa_info->mapping;
        dma_unmap_addr_set(prod_rx_buf, mapping, mapping);

        prod_bd = &rxr->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)];

        prod_bd->rx_bd_haddr = cpu_to_le64(mapping);

        tpa_info->data = cons_rx_buf->data;
        cons_rx_buf->data = NULL;
        tpa_info->mapping = dma_unmap_addr(cons_rx_buf, mapping);

        tpa_info->len =
                le32_to_cpu(tpa_start->rx_tpa_start_cmp_len_flags_type) >>
                                RX_TPA_START_CMP_LEN_SHIFT;
        if (likely(TPA_START_HASH_VALID(tpa_start))) {
                u32 hash_type = TPA_START_HASH_TYPE(tpa_start);

                tpa_info->hash_type = PKT_HASH_TYPE_L4;
                tpa_info->gso_type = SKB_GSO_TCPV4;
                /* RSS profiles 1 and 3 with extract code 0 for inner 4-tuple */
                if (hash_type == 3)
                        tpa_info->gso_type = SKB_GSO_TCPV6;
                tpa_info->rss_hash =
                        le32_to_cpu(tpa_start->rx_tpa_start_cmp_rss_hash);
        } else {
                tpa_info->hash_type = PKT_HASH_TYPE_NONE;
                tpa_info->gso_type = 0;
                if (netif_msg_rx_err(bp))
                        netdev_warn(bp->dev, "TPA packet without valid hash\n");
        }
        tpa_info->flags2 = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_flags2);
        tpa_info->metadata = le32_to_cpu(tpa_start1->rx_tpa_start_cmp_metadata);

        rxr->rx_prod = NEXT_RX(prod);
        cons = NEXT_RX(cons);
        rxr->rx_next_cons = NEXT_RX(cons);
        cons_rx_buf = &rxr->rx_buf_ring[cons];

        bnxt_reuse_rx_data(rxr, cons, cons_rx_buf->data);
        rxr->rx_prod = NEXT_RX(rxr->rx_prod);
        cons_rx_buf->data = NULL;
}

static void bnxt_abort_tpa(struct bnxt *bp, struct bnxt_napi *bnapi,
                           u16 cp_cons, u32 agg_bufs)
{
        if (agg_bufs)
                bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs);
}

#define BNXT_IPV4_HDR_SIZE      (sizeof(struct iphdr) + sizeof(struct tcphdr))
#define BNXT_IPV6_HDR_SIZE      (sizeof(struct ipv6hdr) + sizeof(struct tcphdr))

static inline struct sk_buff *bnxt_gro_skb(struct bnxt_tpa_info *tpa_info,
                                           struct rx_tpa_end_cmp *tpa_end,
                                           struct rx_tpa_end_cmp_ext *tpa_end1,
                                           struct sk_buff *skb)
{
#ifdef CONFIG_INET
        struct tcphdr *th;
        int payload_off, tcp_opt_len = 0;
        int len, nw_off;
        u16 segs;

        segs = TPA_END_TPA_SEGS(tpa_end);
        if (segs == 1)
                return skb;

        NAPI_GRO_CB(skb)->count = segs;
        skb_shinfo(skb)->gso_size =
                le32_to_cpu(tpa_end1->rx_tpa_end_cmp_seg_len);
        skb_shinfo(skb)->gso_type = tpa_info->gso_type;
        payload_off = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) &
                       RX_TPA_END_CMP_PAYLOAD_OFFSET) >>
                      RX_TPA_END_CMP_PAYLOAD_OFFSET_SHIFT;
        if (TPA_END_GRO_TS(tpa_end))
                tcp_opt_len = 12;

        if (tpa_info->gso_type == SKB_GSO_TCPV4) {
                struct iphdr *iph;

                nw_off = payload_off - BNXT_IPV4_HDR_SIZE - tcp_opt_len -
                         ETH_HLEN;
                skb_set_network_header(skb, nw_off);
                iph = ip_hdr(skb);
                skb_set_transport_header(skb, nw_off + sizeof(struct iphdr));
                len = skb->len - skb_transport_offset(skb);
                th = tcp_hdr(skb);
                th->check = ~tcp_v4_check(len, iph->saddr, iph->daddr, 0);
        } else if (tpa_info->gso_type == SKB_GSO_TCPV6) {
                struct ipv6hdr *iph;

                nw_off = payload_off - BNXT_IPV6_HDR_SIZE - tcp_opt_len -
                         ETH_HLEN;
                skb_set_network_header(skb, nw_off);
                iph = ipv6_hdr(skb);
                skb_set_transport_header(skb, nw_off + sizeof(struct ipv6hdr));
                len = skb->len - skb_transport_offset(skb);
                th = tcp_hdr(skb);
                th->check = ~tcp_v6_check(len, &iph->saddr, &iph->daddr, 0);
        } else {
                dev_kfree_skb_any(skb);
                return NULL;
        }
        tcp_gro_complete(skb);

        if (nw_off) { /* tunnel */
                struct udphdr *uh = NULL;

                if (skb->protocol == htons(ETH_P_IP)) {
                        struct iphdr *iph = (struct iphdr *)skb->data;

                        if (iph->protocol == IPPROTO_UDP)
                                uh = (struct udphdr *)(iph + 1);
                } else {
                        struct ipv6hdr *iph = (struct ipv6hdr *)skb->data;

                        if (iph->nexthdr == IPPROTO_UDP)
                                uh = (struct udphdr *)(iph + 1);
                }
                if (uh) {

                        if (uh->check)
                                skb_shinfo(skb)->gso_type |=
                                        SKB_GSO_UDP_TUNNEL_CSUM;
                        else
                                skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL;
                }
        }
#endif
        return skb;
}

static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
                                           struct bnxt_napi *bnapi,
                                           u32 *raw_cons,
                                           struct rx_tpa_end_cmp *tpa_end,
                                           struct rx_tpa_end_cmp_ext *tpa_end1,
                                           bool *agg_event)
{
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
        u8 agg_id = TPA_END_AGG_ID(tpa_end);
        u8 *data, agg_bufs;
        u16 cp_cons = RING_CMP(*raw_cons);
        unsigned int len;
        struct bnxt_tpa_info *tpa_info;
        dma_addr_t mapping;
        struct sk_buff *skb;

        if (unlikely(bnapi->in_reset)) {
                int rc = bnxt_discard_rx(bp, bnapi, raw_cons, tpa_end);

                if (rc < 0)
                        return ERR_PTR(-EBUSY);
                return NULL;
        }

        tpa_info = &rxr->rx_tpa[agg_id];
        data = tpa_info->data;
        prefetch(data);
        len = tpa_info->len;
        mapping = tpa_info->mapping;

        agg_bufs = (le32_to_cpu(tpa_end->rx_tpa_end_cmp_misc_v1) &
                    RX_TPA_END_CMP_AGG_BUFS) >> RX_TPA_END_CMP_AGG_BUFS_SHIFT;

        if (agg_bufs) {
                if (!bnxt_agg_bufs_valid(bp, cpr, agg_bufs, raw_cons))
                        return ERR_PTR(-EBUSY);

                *agg_event = true;
                cp_cons = NEXT_CMP(cp_cons);
        }

        if (unlikely(agg_bufs > MAX_SKB_FRAGS)) {
                bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
                netdev_warn(bp->dev, "TPA frags %d exceeded MAX_SKB_FRAGS %d\n",
                            agg_bufs, (int)MAX_SKB_FRAGS);
                return NULL;
        }

        if (len <= bp->rx_copy_thresh) {
                skb = bnxt_copy_skb(bnapi, data, len, mapping);
                if (!skb) {
                        bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
                        return NULL;
                }
        } else {
                u8 *new_data;
                dma_addr_t new_mapping;

                new_data = __bnxt_alloc_rx_data(bp, &new_mapping, GFP_ATOMIC);
                if (!new_data) {
                        bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
                        return NULL;
                }

                tpa_info->data = new_data;
                tpa_info->mapping = new_mapping;

                skb = build_skb(data, 0);
                dma_unmap_single(&bp->pdev->dev, mapping, bp->rx_buf_use_size,
                                 PCI_DMA_FROMDEVICE);

                if (!skb) {
                        kfree(data);
                        bnxt_abort_tpa(bp, bnapi, cp_cons, agg_bufs);
                        return NULL;
                }
                skb_reserve(skb, BNXT_RX_OFFSET);
                skb_put(skb, len);
        }

        if (agg_bufs) {
                skb = bnxt_rx_pages(bp, bnapi, skb, cp_cons, agg_bufs);
                if (!skb) {
                        /* Page reuse already handled by bnxt_rx_pages(). */
                        return NULL;
                }
        }
        skb->protocol = eth_type_trans(skb, bp->dev);

        if (tpa_info->hash_type != PKT_HASH_TYPE_NONE)
                skb_set_hash(skb, tpa_info->rss_hash, tpa_info->hash_type);

        if (tpa_info->flags2 & RX_CMP_FLAGS2_META_FORMAT_VLAN) {
                netdev_features_t features = skb->dev->features;
                u16 vlan_proto = tpa_info->metadata >>
                        RX_CMP_FLAGS2_METADATA_TPID_SFT;

                if (((features & NETIF_F_HW_VLAN_CTAG_RX) &&
                     vlan_proto == ETH_P_8021Q) ||
                    ((features & NETIF_F_HW_VLAN_STAG_RX) &&
                     vlan_proto == ETH_P_8021AD)) {
                        __vlan_hwaccel_put_tag(skb, htons(vlan_proto),
                                               tpa_info->metadata &
                                               RX_CMP_FLAGS2_METADATA_VID_MASK);
                }
        }

        skb_checksum_none_assert(skb);
        if (likely(tpa_info->flags2 & RX_TPA_START_CMP_FLAGS2_L4_CS_CALC)) {
                skb->ip_summed = CHECKSUM_UNNECESSARY;
                skb->csum_level =
                        (tpa_info->flags2 & RX_CMP_FLAGS2_T_L4_CS_CALC) >> 3;
        }

        if (TPA_END_GRO(tpa_end))
                skb = bnxt_gro_skb(tpa_info, tpa_end, tpa_end1, skb);

        return skb;
}

/* returns the following:
 * 1       - 1 packet successfully received
 * 0       - successful TPA_START, packet not completed yet
 * -EBUSY  - completion ring does not have all the agg buffers yet
 * -ENOMEM - packet aborted due to out of memory
 * -EIO    - packet aborted due to hw error indicated in BD
 */
static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_napi *bnapi, u32 *raw_cons,
                       bool *agg_event)
{
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;
        struct net_device *dev = bp->dev;
        struct rx_cmp *rxcmp;
        struct rx_cmp_ext *rxcmp1;
        u32 tmp_raw_cons = *raw_cons;
        u16 cons, prod, cp_cons = RING_CMP(tmp_raw_cons);
        struct bnxt_sw_rx_bd *rx_buf;
        unsigned int len;
        u8 *data, agg_bufs, cmp_type;
        dma_addr_t dma_addr;
        struct sk_buff *skb;
        int rc = 0;

        rxcmp = (struct rx_cmp *)
                        &cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];

        tmp_raw_cons = NEXT_RAW_CMP(tmp_raw_cons);
        cp_cons = RING_CMP(tmp_raw_cons);
        rxcmp1 = (struct rx_cmp_ext *)
                        &cpr->cp_desc_ring[CP_RING(cp_cons)][CP_IDX(cp_cons)];

        if (!RX_CMP_VALID(rxcmp1, tmp_raw_cons))
                return -EBUSY;

        cmp_type = RX_CMP_TYPE(rxcmp);

        prod = rxr->rx_prod;

        if (cmp_type == CMP_TYPE_RX_L2_TPA_START_CMP) {
                bnxt_tpa_start(bp, rxr, (struct rx_tpa_start_cmp *)rxcmp,
                               (struct rx_tpa_start_cmp_ext *)rxcmp1);

                goto next_rx_no_prod;

        } else if (cmp_type == CMP_TYPE_RX_L2_TPA_END_CMP) {
                skb = bnxt_tpa_end(bp, bnapi, &tmp_raw_cons,
                                   (struct rx_tpa_end_cmp *)rxcmp,
                                   (struct rx_tpa_end_cmp_ext *)rxcmp1,
                                   agg_event);

                if (unlikely(IS_ERR(skb)))
                        return -EBUSY;

                rc = -ENOMEM;
                if (likely(skb)) {
                        skb_record_rx_queue(skb, bnapi->index);
                        skb_mark_napi_id(skb, &bnapi->napi);
                        if (bnxt_busy_polling(bnapi))
                                netif_receive_skb(skb);
                        else
                                napi_gro_receive(&bnapi->napi, skb);
                        rc = 1;
                }
                goto next_rx_no_prod;
        }

        cons = rxcmp->rx_cmp_opaque;
        rx_buf = &rxr->rx_buf_ring[cons];
        data = rx_buf->data;
        if (unlikely(cons != rxr->rx_next_cons)) {
                int rc1 = bnxt_discard_rx(bp, bnapi, raw_cons, rxcmp);

                bnxt_sched_reset(bp, rxr);
                return rc1;
        }
        prefetch(data);

        agg_bufs = (le32_to_cpu(rxcmp->rx_cmp_misc_v1) & RX_CMP_AGG_BUFS) >>
                                RX_CMP_AGG_BUFS_SHIFT;

        if (agg_bufs) {
                if (!bnxt_agg_bufs_valid(bp, cpr, agg_bufs, &tmp_raw_cons))
                        return -EBUSY;

                cp_cons = NEXT_CMP(cp_cons);
                *agg_event = true;
        }

        rx_buf->data = NULL;
        if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L2_ERRORS) {
                bnxt_reuse_rx_data(rxr, cons, data);
                if (agg_bufs)
                        bnxt_reuse_rx_agg_bufs(bnapi, cp_cons, agg_bufs);

                rc = -EIO;
                goto next_rx;
        }

        len = le32_to_cpu(rxcmp->rx_cmp_len_flags_type) >> RX_CMP_LEN_SHIFT;
        dma_addr = dma_unmap_addr(rx_buf, mapping);

        if (len <= bp->rx_copy_thresh) {
                skb = bnxt_copy_skb(bnapi, data, len, dma_addr);
                bnxt_reuse_rx_data(rxr, cons, data);
                if (!skb) {
                        rc = -ENOMEM;
                        goto next_rx;
                }
        } else {
                skb = bnxt_rx_skb(bp, rxr, cons, prod, data, dma_addr, len);
                if (!skb) {
                        rc = -ENOMEM;
                        goto next_rx;
                }
        }

        if (agg_bufs) {
                skb = bnxt_rx_pages(bp, bnapi, skb, cp_cons, agg_bufs);
                if (!skb) {
                        rc = -ENOMEM;
                        goto next_rx;
                }
        }

        if (RX_CMP_HASH_VALID(rxcmp)) {
                u32 hash_type = RX_CMP_HASH_TYPE(rxcmp);
                enum pkt_hash_types type = PKT_HASH_TYPE_L4;

                /* RSS profiles 1 and 3 with extract code 0 for inner 4-tuple */
                if (hash_type != 1 && hash_type != 3)
                        type = PKT_HASH_TYPE_L3;
                skb_set_hash(skb, le32_to_cpu(rxcmp->rx_cmp_rss_hash), type);
        }

        skb->protocol = eth_type_trans(skb, dev);

        if (rxcmp1->rx_cmp_flags2 &
            cpu_to_le32(RX_CMP_FLAGS2_META_FORMAT_VLAN)) {
                netdev_features_t features = skb->dev->features;
                u32 meta_data = le32_to_cpu(rxcmp1->rx_cmp_meta_data);
                u16 vlan_proto = meta_data >> RX_CMP_FLAGS2_METADATA_TPID_SFT;

                if (((features & NETIF_F_HW_VLAN_CTAG_RX) &&
                     vlan_proto == ETH_P_8021Q) ||
                    ((features & NETIF_F_HW_VLAN_STAG_RX) &&
                     vlan_proto == ETH_P_8021AD))
                        __vlan_hwaccel_put_tag(skb, htons(vlan_proto),
                                               meta_data &
                                               RX_CMP_FLAGS2_METADATA_VID_MASK);
        }

        skb_checksum_none_assert(skb);
        if (RX_CMP_L4_CS_OK(rxcmp1)) {
                if (dev->features & NETIF_F_RXCSUM) {
                        skb->ip_summed = CHECKSUM_UNNECESSARY;
                        skb->csum_level = RX_CMP_ENCAP(rxcmp1);
                }
        } else {
                if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L4_CS_ERR_BITS) {
                        if (dev->features & NETIF_F_RXCSUM)
                                cpr->rx_l4_csum_errors++;
                }
        }

        skb_record_rx_queue(skb, bnapi->index);
        skb_mark_napi_id(skb, &bnapi->napi);
        if (bnxt_busy_polling(bnapi))
                netif_receive_skb(skb);
        else
                napi_gro_receive(&bnapi->napi, skb);
        rc = 1;

next_rx:
        rxr->rx_prod = NEXT_RX(prod);
        rxr->rx_next_cons = NEXT_RX(cons);

next_rx_no_prod:
        *raw_cons = tmp_raw_cons;

        return rc;
}

static int bnxt_async_event_process(struct bnxt *bp,
                                    struct hwrm_async_event_cmpl *cmpl)
{
        u16 event_id = le16_to_cpu(cmpl->event_id);

        /* TODO CHIMP_FW: Define event id's for link change, error etc */
        switch (event_id) {
        case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_LINK_STATUS_CHANGE:
                set_bit(BNXT_LINK_CHNG_SP_EVENT, &bp->sp_event);
                break;
        case HWRM_ASYNC_EVENT_CMPL_EVENT_ID_PF_DRVR_UNLOAD:
                set_bit(BNXT_HWRM_PF_UNLOAD_SP_EVENT, &bp->sp_event);
                break;
        default:
                netdev_err(bp->dev, "unhandled ASYNC event (id 0x%x)\n",
                           event_id);
                goto async_event_process_exit;
        }
        schedule_work(&bp->sp_task);
async_event_process_exit:
        return 0;
}

static int bnxt_hwrm_handler(struct bnxt *bp, struct tx_cmp *txcmp)
{
        u16 cmpl_type = TX_CMP_TYPE(txcmp), vf_id, seq_id;
        struct hwrm_cmpl *h_cmpl = (struct hwrm_cmpl *)txcmp;
        struct hwrm_fwd_req_cmpl *fwd_req_cmpl =
                                (struct hwrm_fwd_req_cmpl *)txcmp;

        switch (cmpl_type) {
        case CMPL_BASE_TYPE_HWRM_DONE:
                seq_id = le16_to_cpu(h_cmpl->sequence_id);
                if (seq_id == bp->hwrm_intr_seq_id)
                        bp->hwrm_intr_seq_id = HWRM_SEQ_ID_INVALID;
                else
                        netdev_err(bp->dev, "Invalid hwrm seq id %d\n", seq_id);
                break;

        case CMPL_BASE_TYPE_HWRM_FWD_REQ:
                vf_id = le16_to_cpu(fwd_req_cmpl->source_id);

                if ((vf_id < bp->pf.first_vf_id) ||
                    (vf_id >= bp->pf.first_vf_id + bp->pf.active_vfs)) {
                        netdev_err(bp->dev, "Msg contains invalid VF id %x\n",
                                   vf_id);
                        return -EINVAL;
                }

                set_bit(vf_id - bp->pf.first_vf_id, bp->pf.vf_event_bmap);
                set_bit(BNXT_HWRM_EXEC_FWD_REQ_SP_EVENT, &bp->sp_event);
                schedule_work(&bp->sp_task);
                break;

        case CMPL_BASE_TYPE_HWRM_ASYNC_EVENT:
                bnxt_async_event_process(bp,
                                         (struct hwrm_async_event_cmpl *)txcmp);

        default:
                break;
        }

        return 0;
}

static irqreturn_t bnxt_msix(int irq, void *dev_instance)
{
        struct bnxt_napi *bnapi = dev_instance;
        struct bnxt *bp = bnapi->bp;
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        u32 cons = RING_CMP(cpr->cp_raw_cons);

        prefetch(&cpr->cp_desc_ring[CP_RING(cons)][CP_IDX(cons)]);
        napi_schedule(&bnapi->napi);
        return IRQ_HANDLED;
}

static inline int bnxt_has_work(struct bnxt *bp, struct bnxt_cp_ring_info *cpr)
{
        u32 raw_cons = cpr->cp_raw_cons;
        u16 cons = RING_CMP(raw_cons);
        struct tx_cmp *txcmp;

        txcmp = &cpr->cp_desc_ring[CP_RING(cons)][CP_IDX(cons)];

        return TX_CMP_VALID(txcmp, raw_cons);
}

static irqreturn_t bnxt_inta(int irq, void *dev_instance)
{
        struct bnxt_napi *bnapi = dev_instance;
        struct bnxt *bp = bnapi->bp;
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        u32 cons = RING_CMP(cpr->cp_raw_cons);
        u32 int_status;

        prefetch(&cpr->cp_desc_ring[CP_RING(cons)][CP_IDX(cons)]);

        if (!bnxt_has_work(bp, cpr)) {
                int_status = readl(bp->bar0 + BNXT_CAG_REG_LEGACY_INT_STATUS);
                /* return if erroneous interrupt */
                if (!(int_status & (0x10000 << cpr->cp_ring_struct.fw_ring_id)))
                        return IRQ_NONE;
        }

        /* disable ring IRQ */
        BNXT_CP_DB_IRQ_DIS(cpr->cp_doorbell);

        /* Return here if interrupt is shared and is disabled. */
        if (unlikely(atomic_read(&bp->intr_sem) != 0))
                return IRQ_HANDLED;

        napi_schedule(&bnapi->napi);
        return IRQ_HANDLED;
}

static int bnxt_poll_work(struct bnxt *bp, struct bnxt_napi *bnapi, int budget)
{
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        u32 raw_cons = cpr->cp_raw_cons;
        u32 cons;
        int tx_pkts = 0;
        int rx_pkts = 0;
        bool rx_event = false;
        bool agg_event = false;
        struct tx_cmp *txcmp;

        while (1) {
                int rc;

                cons = RING_CMP(raw_cons);
                txcmp = &cpr->cp_desc_ring[CP_RING(cons)][CP_IDX(cons)];

                if (!TX_CMP_VALID(txcmp, raw_cons))
                        break;

                /* The valid test of the entry must be done first before
                 * reading any further.
                 */
                rmb();
                if (TX_CMP_TYPE(txcmp) == CMP_TYPE_TX_L2_CMP) {
                        tx_pkts++;
                        /* return full budget so NAPI will complete. */
                        if (unlikely(tx_pkts > bp->tx_wake_thresh))
                                rx_pkts = budget;
                } else if ((TX_CMP_TYPE(txcmp) & 0x30) == 0x10) {
                        rc = bnxt_rx_pkt(bp, bnapi, &raw_cons, &agg_event);
                        if (likely(rc >= 0))
                                rx_pkts += rc;
                        else if (rc == -EBUSY)  /* partial completion */
                                break;
                        rx_event = true;
                } else if (unlikely((TX_CMP_TYPE(txcmp) ==
                                     CMPL_BASE_TYPE_HWRM_DONE) ||
                                    (TX_CMP_TYPE(txcmp) ==
                                     CMPL_BASE_TYPE_HWRM_FWD_REQ) ||
                                    (TX_CMP_TYPE(txcmp) ==
                                     CMPL_BASE_TYPE_HWRM_ASYNC_EVENT))) {
                        bnxt_hwrm_handler(bp, txcmp);
                }
                raw_cons = NEXT_RAW_CMP(raw_cons);

                if (rx_pkts == budget)
                        break;
        }

        cpr->cp_raw_cons = raw_cons;
        /* ACK completion ring before freeing tx ring and producing new
         * buffers in rx/agg rings to prevent overflowing the completion
         * ring.
         */
        BNXT_CP_DB(cpr->cp_doorbell, cpr->cp_raw_cons);

        if (tx_pkts)
                bnxt_tx_int(bp, bnapi, tx_pkts);

        if (rx_event) {
                struct bnxt_rx_ring_info *rxr = bnapi->rx_ring;

                writel(DB_KEY_RX | rxr->rx_prod, rxr->rx_doorbell);
                writel(DB_KEY_RX | rxr->rx_prod, rxr->rx_doorbell);
                if (agg_event) {
                        writel(DB_KEY_RX | rxr->rx_agg_prod,
                               rxr->rx_agg_doorbell);
                        writel(DB_KEY_RX | rxr->rx_agg_prod,
                               rxr->rx_agg_doorbell);
                }
        }
        return rx_pkts;
}

static int bnxt_poll(struct napi_struct *napi, int budget)
{
        struct bnxt_napi *bnapi = container_of(napi, struct bnxt_napi, napi);
        struct bnxt *bp = bnapi->bp;
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        int work_done = 0;

        if (!bnxt_lock_napi(bnapi))
                return budget;

        while (1) {
                work_done += bnxt_poll_work(bp, bnapi, budget - work_done);

                if (work_done >= budget)
                        break;

                if (!bnxt_has_work(bp, cpr)) {
                        napi_complete(napi);
                        BNXT_CP_DB_REARM(cpr->cp_doorbell, cpr->cp_raw_cons);
                        break;
                }
        }
        mmiowb();
        bnxt_unlock_napi(bnapi);
        return work_done;
}

#ifdef CONFIG_NET_RX_BUSY_POLL
static int bnxt_busy_poll(struct napi_struct *napi)
{
        struct bnxt_napi *bnapi = container_of(napi, struct bnxt_napi, napi);
        struct bnxt *bp = bnapi->bp;
        struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
        int rx_work, budget = 4;

        if (atomic_read(&bp->intr_sem) != 0)
                return LL_FLUSH_FAILED;

        if (!bnxt_lock_poll(bnapi))
                return LL_FLUSH_BUSY;

        rx_work = bnxt_poll_work(bp, bnapi, budget);

        BNXT_CP_DB_REARM(cpr->cp_doorbell, cpr->cp_raw_cons);

        bnxt_unlock_poll(bnapi);
        return rx_work;
}
#endif

static void bnxt_free_tx_skbs(struct bnxt *bp)
{
        int i, max_idx;
        struct pci_dev *pdev = bp->pdev;

        if (!bp->tx_ring)
                return;

        max_idx = bp->tx_nr_pages * TX_DESC_CNT;
        for (i = 0; i < bp->tx_nr_rings; i++) {
                struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
                int j;

                for (j = 0; j < max_idx;) {
                        struct bnxt_sw_tx_bd *tx_buf = &txr->tx_buf_ring[j];
                        struct sk_buff *skb = tx_buf->skb;
                        int k, last;

                        if (!skb) {
                                j++;
                                continue;
                        }

                        tx_buf->skb = NULL;

                        if (tx_buf->is_push) {
                                dev_kfree_skb(skb);
                                j += 2;
                                continue;
                        }

                        dma_unmap_single(&pdev->dev,
                                         dma_unmap_addr(tx_buf, mapping),
                                         skb_headlen(skb),
                                         PCI_DMA_TODEVICE);

                        last = tx_buf->nr_frags;
                        j += 2;
                        for (k = 0; k < last; k++, j++) {
                                int ring_idx = j & bp->tx_ring_mask;
                                skb_frag_t *frag = &skb_shinfo(skb)->frags[k];

                                tx_buf = &txr->tx_buf_ring[ring_idx];
                                dma_unmap_page(
                                        &pdev->dev,
                                        dma_unmap_addr(tx_buf, mapping),
                                        skb_frag_size(frag), PCI_DMA_TODEVICE);
                        }
                        dev_kfree_skb(skb);
                }
                netdev_tx_reset_queue(netdev_get_tx_queue(bp->dev, i));
        }
}

static void bnxt_free_rx_skbs(struct bnxt *bp)
{
        int i, max_idx, max_agg_idx;
        struct pci_dev *pdev = bp->pdev;

        if (!bp->rx_ring)
                return;

        max_idx = bp->rx_nr_pages * RX_DESC_CNT;
        max_agg_idx = bp->rx_agg_nr_pages * RX_DESC_CNT;
        for (i = 0; i < bp->rx_nr_rings; i++) {
                struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
                int j;

                if (rxr->rx_tpa) {
                        for (j = 0; j < MAX_TPA; j++) {
                                struct bnxt_tpa_info *tpa_info =
                                                        &rxr->rx_tpa[j];
                                u8 *data = tpa_info->data;

                                if (!data)
                                        continue;

                                dma_unmap_single(
                                        &pdev->dev,
                                        dma_unmap_addr(tpa_info, mapping),
                                        bp->rx_buf_use_size,
                                        PCI_DMA_FROMDEVICE);

                                tpa_info->data = NULL;

                                kfree(data);
                        }
                }

                for (j = 0; j < max_idx; j++) {
                        struct bnxt_sw_rx_bd *rx_buf = &rxr->rx_buf_ring[j];
                        u8 *data = rx_buf->data;

                        if (!data)
                                continue;

                        dma_unmap_single(&pdev->dev,
                                         dma_unmap_addr(rx_buf, mapping),
                                         bp->rx_buf_use_size,
                                         PCI_DMA_FROMDEVICE);

                        rx_buf->data = NULL;

                        kfree(data);
                }

                for (j = 0; j < max_agg_idx; j++) {
                        struct bnxt_sw_rx_agg_bd *rx_agg_buf =
                                &rxr->rx_agg_ring[j];
                        struct page *page = rx_agg_buf->page;

                        if (!page)
                                continue;

                        dma_unmap_page(&pdev->dev,
                                       dma_unmap_addr(rx_agg_buf, mapping),
                                       BNXT_RX_PAGE_SIZE, PCI_DMA_FROMDEVICE);

                        rx_agg_buf->page = NULL;
                        __clear_bit(j, rxr->rx_agg_bmap);

                        __free_page(page);
                }
                if (rxr->rx_page) {
                        __free_page(rxr->rx_page);
                        rxr->rx_page = NULL;
                }
        }
}

static void bnxt_free_skbs(struct bnxt *bp)
{
        bnxt_free_tx_skbs(bp);
        bnxt_free_rx_skbs(bp);
}

static void bnxt_free_ring(struct bnxt *bp, struct bnxt_ring_struct *ring)
{
        struct pci_dev *pdev = bp->pdev;
        int i;

        for (i = 0; i < ring->nr_pages; i++) {
                if (!ring->pg_arr[i])
                        continue;

                dma_free_coherent(&pdev->dev, ring->page_size,
                                  ring->pg_arr[i], ring->dma_arr[i]);

                ring->pg_arr[i] = NULL;
        }
        if (ring->pg_tbl) {
                dma_free_coherent(&pdev->dev, ring->nr_pages * 8,
                                  ring->pg_tbl, ring->pg_tbl_map);
                ring->pg_tbl = NULL;
        }
        if (ring->vmem_size && *ring->vmem) {
                vfree(*ring->vmem);
                *ring->vmem = NULL;
        }
}

static int bnxt_alloc_ring(struct bnxt *bp, struct bnxt_ring_struct *ring)
{
        int i;
        struct pci_dev *pdev = bp->pdev;

        if (ring->nr_pages > 1) {
                ring->pg_tbl = dma_alloc_coherent(&pdev->dev,
                                                  ring->nr_pages * 8,
                                                  &ring->pg_tbl_map,
                                                  GFP_KERNEL);
                if (!ring->pg_tbl)
                        return -ENOMEM;
        }

        for (i = 0; i < ring->nr_pages; i++) {
                ring->pg_arr[i] = dma_alloc_coherent(&pdev->dev,
                                                     ring->page_size,
                                                     &ring->dma_arr[i],
                                                     GFP_KERNEL);
                if (!ring->pg_arr[i])
                        return -ENOMEM;

                if (ring->nr_pages > 1)
                        ring->pg_tbl[i] = cpu_to_le64(ring->dma_arr[i]);
        }

        if (ring->vmem_size) {
                *ring->vmem = vzalloc(ring->vmem_size);
                if (!(*ring->vmem))
                        return -ENOMEM;
        }
        return 0;
}

static void bnxt_free_rx_rings(struct bnxt *bp)
{
        int i;

        if (!bp->rx_ring)
                return;

        for (i = 0; i < bp->rx_nr_rings; i++) {
                struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
                struct bnxt_ring_struct *ring;

                kfree(rxr->rx_tpa);
                rxr->rx_tpa = NULL;

                kfree(rxr->rx_agg_bmap);
                rxr->rx_agg_bmap = NULL;

                ring = &rxr->rx_ring_struct;
                bnxt_free_ring(bp, ring);

                ring = &rxr->rx_agg_ring_struct;
                bnxt_free_ring(bp, ring);
        }
}

static int bnxt_alloc_rx_rings(struct bnxt *bp)
{
        int i, rc, agg_rings = 0, tpa_rings = 0;

        if (!bp->rx_ring)
                return -ENOMEM;

        if (bp->flags & BNXT_FLAG_AGG_RINGS)
                agg_rings = 1;

        if (bp->flags & BNXT_FLAG_TPA)
                tpa_rings = 1;

        for (i = 0; i < bp->rx_nr_rings; i++) {
                struct bnxt_rx_ring_info *rxr = &bp->rx_ring[i];
                struct bnxt_ring_struct *ring;

                ring = &rxr->rx_ring_struct;

                rc = bnxt_alloc_ring(bp, ring);
                if (rc)
                        return rc;

                if (agg_rings) {
                        u16 mem_size;

                        ring = &rxr->rx_agg_ring_struct;
                        rc = bnxt_alloc_ring(bp, ring);
                        if (rc)
                                return rc;

                        rxr->rx_agg_bmap_size = bp->rx_agg_ring_mask + 1;
                        mem_size = rxr->rx_agg_bmap_size / 8;
                        rxr->rx_agg_bmap = kzalloc(mem_size, GFP_KERNEL);
                        if (!rxr->rx_agg_bmap)
                                return -ENOMEM;

                        if (tpa_rings) {
                                rxr->rx_tpa = kcalloc(MAX_TPA,
                                                sizeof(struct bnxt_tpa_info),
                                                GFP_KERNEL);
                                if (!rxr->rx_tpa)
                                        return -ENOMEM;
                        }
                }
        }
        return 0;
}

static void bnxt_free_tx_rings(struct bnxt *bp)
{
        int i;
        struct pci_dev *pdev = bp->pdev;

        if (!bp->tx_ring)
                return;

        for (i = 0; i < bp->tx_nr_rings; i++) {
                struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
                struct bnxt_ring_struct *ring;

                if (txr->tx_push) {
                        dma_free_coherent(&pdev->dev, bp->tx_push_size,
                                          txr->tx_push, txr->tx_push_mapping);
                        txr->tx_push = NULL;
                }

                ring = &txr->tx_ring_struct;

                bnxt_free_ring(bp, ring);
        }
}

static int bnxt_alloc_tx_rings(struct bnxt *bp)
{
        int i, j, rc;
        struct pci_dev *pdev = bp->pdev;

        bp->tx_push_size = 0;
        if (bp->tx_push_thresh) {
                int push_size;

                push_size  = L1_CACHE_ALIGN(sizeof(struct tx_push_bd) +
                                        bp->tx_push_thresh);

                if (push_size > 256) {
                        push_size = 0;
                        bp->tx_push_thresh = 0;
                }

                bp->tx_push_size = push_size;
        }

        for (i = 0, j = 0; i < bp->tx_nr_rings; i++) {
                struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
                struct bnxt_ring_struct *ring;

                ring = &txr->tx_ring_struct;

                rc = bnxt_alloc_ring(bp, ring);
                if (rc)
                        return rc;

                if (bp->tx_push_size) {
                        dma_addr_t mapping;

                        /* One pre-allocated DMA buffer to backup
                         * TX push operation
                         */
                        txr->tx_push = dma_alloc_coherent(&pdev->dev,
                                                bp->tx_push_size,
                                                &txr->tx_push_mapping,
                                                GFP_KERNEL);

                        if (!txr->tx_push)
                                return -ENOMEM;

                        mapping = txr->tx_push_mapping +
                                sizeof(struct tx_push_bd);
                        txr->data_mapping = cpu_to_le64(mapping);

                        memset(txr->tx_push, 0, sizeof(struct tx_push_bd));
                }
                ring->queue_id = bp->q_info[j].queue_id;
                if (i % bp->tx_nr_rings_per_tc == (bp->tx_nr_rings_per_tc - 1))
                        j++;
        }
        return 0;
}

static void bnxt_free_cp_rings(struct bnxt *bp)
{
        int i;

        if (!bp->bnapi)
                return;

        for (i = 0; i < bp->cp_nr_rings; i++) {
                struct bnxt_napi *bnapi = bp->bnapi[i];
                struct bnxt_cp_ring_info *cpr;
                struct bnxt_ring_struct *ring;

                if (!bnapi)
                        continue;

                cpr = &bnapi->cp_ring;
                ring = &cpr->cp_ring_struct;

                bnxt_free_ring(bp, ring);
        }
}

static int bnxt_alloc_cp_rings(struct bnxt *bp)
{
        int i, rc;

        for (i = 0; i < bp->cp_nr_rings; i++) {
                struct bnxt_napi *bnapi = bp->bnapi[i];
                struct bnxt_cp_ring_info *cpr;
                struct bnxt_ring_struct *ring;

                if (!bnapi)
                        continue;

                cpr = &bnapi->cp_ring;
                ring = &cpr->cp_ring_struct;

                rc = bnxt_alloc_ring(bp, ring);
                if (rc)
                        return rc;
        }
        return 0;
}

static void bnxt_init_ring_struct(struct bnxt *bp)
{
        int i;

        for (i = 0; i < bp->cp_nr_rings; i++) {
                struct bnxt_napi *bnapi = bp->bnapi[i];
                struct bnxt_cp_ring_info *cpr;
                struct bnxt_rx_ring_info *rxr;
                struct bnxt_tx_ring_info *txr;
                struct bnxt_ring_struct *ring;

                if (!bnapi)
                        continue;

                cpr = &bnapi->cp_ring;
                ring = &cpr->cp_ring_struct;
                ring->nr_pages = bp->cp_nr_pages;
                ring->page_size = HW_CMPD_RING_SIZE;
                ring->pg_arr = (void **)cpr->cp_desc_ring;
                ring->dma_arr = cpr->cp_desc_mapping;
                ring->vmem_size = 0;

                rxr = bnapi->rx_ring;
                if (!rxr)
                        goto skip_rx;

                ring = &rxr->rx_ring_struct;
                ring->nr_pages = bp->rx_nr_pages;
                ring->page_size = HW_RXBD_RING_SIZE;
                ring->pg_arr = (void **)rxr->rx_desc_ring;
                ring->dma_arr = rxr->rx_desc_mapping;
                ring->vmem_size = SW_RXBD_RING_SIZE * bp->rx_nr_pages;
                ring->vmem = (void **)&rxr->rx_buf_ring;

                ring = &rxr->rx_agg_ring_struct;
                ring->nr_pages = bp->rx_agg_nr_pages;
                ring->page_size = HW_RXBD_RING_SIZE;
                ring->pg_arr = (void **)rxr->rx_agg_desc_ring;
                ring->dma_arr = rxr->rx_agg_desc_mapping;
                ring->vmem_size = SW_RXBD_AGG_RING_SIZE * bp->rx_agg_nr_pages;
                ring->vmem = (void **)&rxr->rx_agg_ring;

skip_rx:
                txr = bnapi->tx_ring;
                if (!txr)
                        continue;

                ring = &txr->tx_ring_struct;
                ring->nr_pages = bp->tx_nr_pages;
                ring->page_size = HW_RXBD_RING_SIZE;

                ring->pg_arr = (void **)txr->tx_desc_ring;
                ring->dma_arr = txr->tx_desc_mapping;
                ring->vmem_size = SW_TXBD_RING_SIZE * bp->tx_nr_pages;
                ring->vmem = (void **)&txr->tx_buf_ring;
        }
}

static void bnxt_init_rxbd_pages(struct bnxt_ring_struct *ring, u32 type)
{
        int i;
        u32 prod;
        struct rx_bd **rx_buf_ring;

        rx_buf_ring = (struct rx_bd **)ring->pg_arr;
        for (i = 0, prod = 0; i < ring->nr_pages; i++) {
                int j;
                struct rx_bd *rxbd;

                rxbd = rx_buf_ring[i];
                if (!rxbd)
                        continue;

                for (j = 0; j < RX_DESC_CNT; j++, rxbd++, prod++) {
                        rxbd->rx_bd_len_flags_type = cpu_to_le32(type);
                        rxbd->rx_bd_opaque = prod;
                }
        }
}

static int bnxt_init_one_rx_ring(struct bnxt *bp, int ring_nr)
{
        struct net_device *dev = bp->dev;
        struct bnxt_rx_ring_info *rxr;
        struct bnxt_ring_struct *ring;
        u32 prod, type;
        int i;

        type = (bp->rx_buf_use_size << RX_BD_LEN_SHIFT) |
                RX_BD_TYPE_RX_PACKET_BD | RX_BD_FLAGS_EOP;

        if (NET_IP_ALIGN == 2)
                type |= RX_BD_FLAGS_SOP;

        rxr = &bp->rx_ring[ring_nr];
        ring = &rxr->rx_ring_struct;
        bnxt_init_rxbd_pages(ring, type);

        prod = rxr->rx_prod;
        for (i = 0; i < bp->rx_ring_size; i++) {
                if (bnxt_alloc_rx_data(bp, rxr, prod, GFP_KERNEL) != 0) {
                        netdev_warn(dev, "init'ed rx ring %d with %d/%d skbs only\n",
                                    ring_nr, i, bp->rx_ring_size);
                        break;
                }
                prod = NEXT_RX(prod);
        }
        rxr->rx_prod = prod;
        ring->fw_ring_id = INVALID_HW_RING_ID;

        ring = &rxr->rx_agg_ring_struct;
        ring->fw_ring_id = INVALID_HW_RING_ID;

        if (!(bp->flags & BNXT_FLAG_AGG_RINGS))
                return 0;

        type = ((u32)BNXT_RX_PAGE_SIZE << RX_BD_LEN_SHIFT) |
                RX_BD_TYPE_RX_AGG_BD | RX_BD_FLAGS_SOP;

        bnxt_init_rxbd_pages(ring, type);

        prod = rxr->rx_agg_prod;
        for (i = 0; i < bp->rx_agg_ring_size; i++) {
                if (bnxt_alloc_rx_page(bp, rxr, prod, GFP_KERNEL) != 0) {
                        netdev_warn(dev, "init'ed rx ring %d with %d/%d pages only\n",
                                    ring_nr, i, bp->rx_ring_size);
                        break;
                }
                prod = NEXT_RX_AGG(prod);
        }
        rxr->rx_agg_prod = prod;

        if (bp->flags & BNXT_FLAG_TPA) {
                if (rxr->rx_tpa) {
                        u8 *data;
                        dma_addr_t mapping;

                        for (i = 0; i < MAX_TPA; i++) {
                                data = __bnxt_alloc_rx_data(bp, &mapping,
                                                            GFP_KERNEL);
                                if (!data)
                                        return -ENOMEM;

                                rxr->rx_tpa[i].data = data;
                                rxr->rx_tpa[i].mapping = mapping;
                        }
                } else {
                        netdev_err(bp->dev, "No resource allocated for LRO/GRO\n");
                        return -ENOMEM;
                }
        }

        return 0;
}

static int bnxt_init_rx_rings(struct bnxt *bp)
{
        int i, rc = 0;

        for (i = 0; i < bp->rx_nr_rings; i++) {
                rc = bnxt_init_one_rx_ring(bp, i);
                if (rc)
                        break;
        }

        return rc;
}

static int bnxt_init_tx_rings(struct bnxt *bp)
{
        u16 i;

        bp->tx_wake_thresh = max_t(int, bp->tx_ring_size / 2,
                                   MAX_SKB_FRAGS + 1);

        for (i = 0; i < bp->tx_nr_rings; i++) {
                struct bnxt_tx_ring_info *txr = &bp->tx_ring[i];
                struct bnxt_ring_struct *ring = &txr->tx_ring_struct;

                ring->fw_ring_id = INVALID_HW_RING_ID;
        }

        return 0;
}

static void bnxt_free_ring_grps(struct bnxt *bp)
{
        kfree(bp->grp_info);
        bp->grp_info = NULL;
}

static int bnxt_init_ring_grps(struct bnxt *bp, bool irq_re_init)
{
        int i;

        if (irq_re_init) {
                bp->grp_info = kcalloc(bp->cp_nr_rings,
                                       sizeof(struct bnxt_ring_grp_info),
                                       GFP_KERNEL);
                if (!bp->grp_info)
                        return -ENOMEM;
        }
        for (i = 0; i < bp->cp_nr_rings; i++) {
                if (irq_re_init)
                        bp->grp_info[i].fw_stats_ctx = INVALID_HW_RING_ID;
                bp->grp_info[i].fw_grp_id = INVALID_HW_RING_ID;
                bp->grp_info[i].rx_fw_ring_id = INVALID_HW_RING_ID;
                bp->grp_info[i].agg_fw_ring_id = INVALID_HW_RING_ID;
                bp->grp_info[i].cp_fw_ring_id = INVALID_HW_RING_ID;
        }
        return 0;
}

static void bnxt_free_vnics(struct bnxt *bp)
{
        kfree(bp->vnic_info);
        bp->vnic_info = NULL;
        bp->nr_vnics = 0;
}

static int bnxt_alloc_vnics(struct bnxt *bp)
{
        int num_vnics = 1;

#ifdef CONFIG_RFS_ACCEL
        if (bp->flags & BNXT_FLAG_RFS)
                num_vnics += bp->rx_nr_rings;
#endif

        bp->vnic_info = kcalloc(num_vnics, sizeof(struct bnxt_vnic_info),
                                GFP_KERNEL);
        if (!bp->vnic_info)
                return -ENOMEM;

        bp->nr_vnics = num_vnics;
        return 0;
}

static void bnxt_init_vnics(struct bnxt *bp)
{
        int i;

        for (i = 0; i < bp->nr_vnics; i++) {
                struct bnxt_vnic_info *vnic = &bp->vnic_info[i];

                vnic->fw_vnic_id = INVALID_HW_RING_ID;
                vnic->fw_rss_cos_lb_ctx = INVALID_HW_RING_ID;
                vnic->fw_l2_ctx_id = INVALID_HW_RING_ID;

                if (bp->vnic_info[i].rss_hash_key) {
                        if (i == 0)
                                prandom_bytes(vnic->rss_hash_key,
                                              HW_HASH_KEY_SIZE);
                        else
                                memcpy(vnic->rss_hash_key,
                                       bp->vnic_info[0].rss_hash_key,
                                       HW_HASH_KEY_SIZE);
                }
        }
}

static int bnxt_calc_nr_ring_pages(u32 ring_size, int desc_per_pg)
{
        int pages;

        pages = ring_size / desc_per_pg;

        if (!pages)
                return 1;

        pages++;

        while (pages & (pages - 1))
                pages++;

        return pages;
}

static void bnxt_set_tpa_flags(struct bnxt *bp)
{
        bp->flags &= ~BNXT_FLAG_TPA;
        if (bp->dev->features & NETIF_F_LRO)
                bp->flags |= BNXT_FLAG_LRO;
        if ((bp->dev->features & NETIF_F_GRO) && (bp->pdev->revision > 0))
                bp->flags |= BNXT_FLAG_GRO;
}

/* bp->rx_ring_size, bp->tx_ring_size, dev->mtu, BNXT_FLAG_{G|L}RO flags must
 * be set on entry.
 */
void bnxt_set_ring_params(struct bnxt *bp)
{
        u32 ring_size, rx_size, rx_space;
        u32 agg_factor = 0, agg_ring_size = 0;

        /* 8 for CRC and VLAN */
        rx_size = SKB_DATA_ALIGN(bp->dev->mtu + ETH_HLEN + NET_IP_ALIGN + 8);

        rx_space = rx_size + NET_SKB_PAD +
                SKB_DATA_ALIGN(sizeof(struct skb_shared_info));

        bp->rx_copy_thresh = BNXT_RX_COPY_THRESH;
        ring_size = bp->rx_ring_size;
        bp->rx_agg_ring_size = 0;
        bp->rx_agg_nr_pages = 0;

        if (bp->flags & BNXT_FLAG_TPA)
                agg_factor = min_t(u32, 4, 65536 / BNXT_RX_PAGE_SIZE);

        bp->flags &= ~BNXT_FLAG_JUMBO;
        if (rx_space > PAGE_SIZE) {
                u32 jumbo_factor;

                bp->flags |= BNXT_FLAG_JUMBO;
                jumbo_factor = PAGE_ALIGN(bp->dev->mtu - 40) >> PAGE_SHIFT;
                if (jumbo_factor > agg_factor)
                        agg_factor = jumbo_factor;
        }
        agg_ring_size = ring_size * agg_factor;

        if (agg_ring_size) {
                bp->rx_agg_nr_pages = bnxt_calc_nr_ring_pages(agg_ring_size,
                                                        RX_DESC_CNT);
                if (bp->rx_agg_nr_pages > MAX_RX_AGG_PAGES) {
                        u32 tmp = agg_ring_size;

                        bp->rx_agg_nr_pages = MAX_RX_AGG_PAGES;
                        agg_ring_size = MAX_RX_AGG_PAGES * RX_DESC_CNT - 1;
                        netdev_warn(bp->dev, "rx agg ring size %d reduced to %d.\n",
                                    tmp, agg_ring_size);
                }
                bp->rx_agg_ring_size = agg_ring_size;
                bp->rx_agg_ring_mask = (bp->rx_agg_nr_pages * RX_DESC_CNT) - 1;
                rx_size = SKB_DATA_ALIGN(BNXT_RX_COPY_THRESH + NET_IP_ALIGN);
                rx_space = rx_size + NET_SKB_PAD +
                        SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
        }

        bp->rx_buf_use_size = rx_size;
        bp->rx_buf_size = rx_space;

        bp->rx_nr_pages = bnxt_calc_nr_ring_pages(ring_size, RX_DESC_CNT);
        bp->rx_ring_mask = (bp->rx_nr_pages * RX_DESC_CNT) - 1;

        ring_size = bp->tx_ring_size;
        bp->tx_nr_pages = bnxt_calc_nr_ring_pages(ring_size, TX_DESC_CNT);
        bp->tx_ring_mask = (bp->tx_nr_pages * TX_DESC_CNT) - 1;

        ring_size = bp->rx_ring_size * (2 + agg_factor) + bp->tx_ring_size;
        bp->cp_ring_size = ring_size;

        bp->cp_nr_pages = bnxt_calc_nr_ring_pages(ring_size, CP_DESC_CNT);
        if (bp->cp_nr_pages > MAX_CP_PAGES) {
                bp->cp_nr_pages = MAX_CP_PAGES;
                bp->cp_ring_size = MAX_CP_PAGES * CP_DESC_CNT - 1;
                netdev_warn(bp->dev, "completion ring size %d reduced to %d.\n",
                            ring_size, bp->cp_ring_size);
        }
        bp->cp_bit = bp->cp_nr_pages * CP_DESC_CNT;
        bp->cp_ring_mask = bp->cp_bit - 1;
}

static void bnxt_free_vnic_attributes(struct bnxt *bp)
{
        int i;
        struct bnxt_vnic_info *vnic;
        struct pci_dev *pdev = bp->pdev;

        if (!bp->vnic_info)
                return;

        for (i = 0; i < bp->nr_vnics; i++) {
                vnic = &bp->vnic_info[i];

                kfree(vnic->fw_grp_ids);
                vnic->fw_grp_ids = NULL;

                kfree(vnic->uc_list);
                vnic->uc_list = NULL;

                if (vnic->mc_list) {
                        dma_free_coherent(&pdev->dev, vnic->mc_list_size,
                                          vnic->mc_list, vnic->mc_list_mapping);
                        vnic->mc_list = NULL;
                }

                if (vnic->rss_table) {
                        dma_free_coherent(&pdev->dev, PAGE_SIZE,
                                          vnic->rss_table,
                                          vnic->rss_table_dma_addr);
                        vnic->rss_table = NULL;
                }

                vnic->rss_hash_key = NULL;
                vnic->flags = 0;
        }
}

static int bnxt_alloc_vnic_attributes(struct bnxt *bp)
{
        int i, rc = 0, size;
        struct bnxt_vnic_info *vnic;
        struct pci_dev *pdev = bp->pdev;
        int max_rings;

        for (i = 0; i < bp->nr_vnics; i++) {
                vnic = &bp->vnic_info[i];

                if (vnic->flags & BNXT_VNIC_UCAST_FLAG) {
                        int mem_size = (BNXT_MAX_UC_ADDRS - 1) * ETH_ALEN;

                        if (mem_size > 0) {
                                vnic->uc_list = kmalloc(mem_size, GFP_KERNEL);
                                if (!vnic->uc_list) {
                                        rc = -ENOMEM;
                                        goto out;
                                }
                        }
                }

                if (vnic->flags & BNXT_VNIC_MCAST_FLAG) {
                        vnic->mc_list_size = BNXT_MAX_MC_ADDRS * ETH_ALEN;
                        vnic->mc_list =
                                dma_alloc_coherent(&pdev->dev,
                                                   vnic->mc_list_size,
                                                   &vnic->mc_list_mapping,
                                                   GFP_KERNEL);
                        if (!vnic->mc_list) {
                                rc = -ENOMEM;
                                goto out;
                        }
                }

                if (vnic->flags & BNXT_VNIC_RSS_FLAG)
                        max_rings = bp->rx_nr_rings;
                else
                        max_rings = 1;

                vnic->fw_grp_ids = kcalloc(max_rings, sizeof(u16), GFP_KERNEL);
                if (!vnic->fw_grp_ids) {
                        rc = -ENOMEM;
                        goto out;
                }

                /* Allocate rss table and hash key */
                vnic->rss_table = dma_alloc_coherent(&pdev->dev, PAGE_SIZE,
                                                     &vnic->rss_table_dma_addr,
                                                     GFP_KERNEL);
                if (!vnic->rss_table) {
                        rc = -ENOMEM;
                        goto out;
                }

                size = L1_CACHE_ALIGN(HW_HASH_INDEX_SIZE * sizeof(u16));

                vnic->rss_hash_key = ((void *)vnic->rss_table) + size;
                vnic->rss_hash_key_dma_addr = vnic->rss_table_dma_addr + size;
        }
        return 0;

out:
        return rc;
}

static void bnxt_free_hwrm_resources(struct bnxt *bp)
{
        struct pci_dev *pdev = bp->pdev;

        dma_free_coherent(&pdev->dev, PAGE_SIZE, bp->hwrm_cmd_resp_addr,
                          bp->hwrm_cmd_resp_dma_addr);

        bp->hwrm_cmd_resp_addr = NULL;
        if (bp->hwrm_dbg_resp_addr) {
                dma_free_coherent(&pdev->dev, HWRM_DBG_REG_BUF_SIZE,
                                  bp->hwrm_dbg_resp_addr,
                                  bp->hwrm_dbg_resp_dma_addr);

                bp->hwrm_dbg_resp_addr = NULL;
        }
}

static int bnxt_alloc_hwrm_resources(struct bnxt *bp)
{
        struct pci_dev *pdev = bp->pdev;

        bp->hwrm_cmd_resp_addr = dma_alloc_coherent(&pdev->dev, PAGE_SIZE,
                                                   &bp->hwrm_cmd_resp_dma_addr,
                                                   GFP_KERNEL);
        if (!bp->hwrm_cmd_resp_addr)
                return -ENOMEM;
        bp->hwrm_dbg_resp_addr = dma_alloc_coherent(&pdev->dev,
                                                    HWRM_DBG_REG_BUF_SIZE,
                                                    &bp->hwrm_dbg_resp_dma_addr,
                                                    GFP_KERNEL);
        if (!bp->hwrm_dbg_resp_addr)
                netdev_warn(bp->dev, "fail to alloc debug register dma mem\n");

        return 0;
}

static void bnxt_free_stats(struct bnxt *bp)
{
        u32 size, i;
        struct pci_dev *pdev = bp->pdev;

        if (bp->hw_rx_port_stats) {
                dma_free_coherent(&pdev->dev, bp->hw_port_stats_size,
                                  bp->hw_rx_port_stats,
                                  bp->hw_rx_port_stats_map);
                bp->hw_rx_port_stats = NULL;
                bp->flags &= ~BNXT_FLAG_PORT_STATS;
        }

        if (!bp->bnapi)
                return;

        size = sizeof(struct ctx_hw_stats);

        for (i = 0; i < bp->cp_nr_rings; i++) {
                struct bnxt_napi *bnapi = bp->bnapi[i];
                struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;

                if (cpr->hw_stats) {
                        dma_free_coherent(&pdev->dev, size, cpr->hw_stats,
                                          cpr->hw_stats_map);
                        cpr->hw_stats = NULL;
                }
        }
}

static int bnxt_alloc_stats(struct bnxt *bp)
{
        u32 size, i;
        struct pci_dev *pdev = bp->pdev;

        size = sizeof(struct ctx_hw_stats);

        for (i = 0; i < bp->cp_nr_rings; i++) {
                struct bnxt_napi *bnapi = bp->bnapi[i];
                struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;

                cpr->hw_stats = dma_alloc_coherent(&pdev->dev, size,
                                                   &cpr->hw_stats_map,
                                                   GFP_KERNEL);
                if (!cpr->hw_stats)
                        return -ENOMEM;

                cpr->hw_stats_ctx_id = INVALID_STATS_CTX_ID;
        }

        if (BNXT_PF(bp)) {
                bp->hw_port_stats_size = sizeof(struct rx_port_stats) +
                                         sizeof(struct tx_port_stats) + 1024;

                bp->hw_rx_port_stats =
                        dma_alloc_coherent(&pdev->dev, bp->hw_port_stats_size,
                                           &bp->hw_rx_port_stats_map,
                                           GFP_KERNEL);
                if (!bp->hw_rx_port_stats)
                        return -ENOMEM;

                bp->hw_tx_port_stats = (void *)(bp->hw_rx_port_stats + 1) +
                                       512;
                bp->hw_tx_port_stats_map = bp->hw_rx_port_stats_map +
                                           sizeof(struct rx_port_stats) + 512;
                bp->flags |= BNXT_FLAG_PORT_STATS;
        }
        return 0;
}

static void bnxt_clear_ring_indices(struct bnxt *bp)
{
        int i;

        if (!bp->bnapi)
                return;

        for (i = 0; i < bp->cp_nr_rings; i++) {
                struct bnxt_napi *bnapi = bp->bnapi[i];
                struct bnxt_cp_ring_info *cpr;
                struct bnxt_rx_ring_info *rxr;
                struct bnxt_tx_ring_info *txr;

                if (!bnapi)
                        continue;

                cpr = &bnapi->cp_ring;
                cpr->cp_raw_cons = 0;

                txr = bnapi->tx_ring;
                if (txr) {
                        txr->tx_prod = 0;
                        txr->tx_cons = 0;
                }

                rxr = bnapi->rx_ring;
                if (rxr) {
                        rxr->rx_prod = 0;
                        rxr->rx_agg_prod = 0;
                        rxr->rx_sw_agg_prod = 0;
                        rxr->rx_next_cons = 0;
                }
        }
}

static void bnxt_free_ntp_fltrs(struct bnxt *bp, bool irq_reinit)
{
#ifdef CONFIG_RFS_ACCEL
        int i;

        /* Under rtnl_lock and all our NAPIs have been disabled.  It's
         * safe to delete the hash table.
         */
        for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++) {
                struct hlist_head *head;
                struct hlist_node *tmp;
                struct bnxt_ntuple_filter *fltr;

                head = &bp->ntp_fltr_hash_tbl[i];
                hlist_for_each_entry_safe(fltr, tmp, head, hash) {
                        hlist_del(&fltr->hash);
                        kfree(fltr);
                }
        }
        if (irq_reinit) {
                kfree(bp->ntp_fltr_bmap);
                bp->ntp_fltr_bmap = NULL;
        }
        bp->ntp_fltr_count = 0;
#endif
}

static int bnxt_alloc_ntp_fltrs(struct bnxt *bp)
{
#ifdef CONFIG_RFS_ACCEL
        int i, rc = 0;

        if (!(bp->flags & BNXT_FLAG_RFS))
                return 0;

        for (i = 0; i < BNXT_NTP_FLTR_HASH_SIZE; i++)
                INIT_HLIST_HEAD(&bp->ntp_fltr_hash_tbl[i]);

        bp->ntp_fltr_count = 0;
        bp->ntp_fltr_bmap = kzalloc(BITS_TO_LONGS(BNXT_NTP_FLTR_MAX_FLTR),
                                    GFP_KERNEL);

        if (!bp->ntp_fltr_bmap)
                rc = -ENOMEM;

        return rc;
#else
        return 0;
#endif
}

static void bnxt_free_mem(struct bnxt *bp, bool irq_re_init)
{
        bnxt_free_vnic_attributes(bp);
        bnxt_free_tx_rings(bp);
        bnxt_free_rx_rings(bp);
        bnxt_free_cp_rings(bp);
        bnxt_free_ntp_fltrs(bp, irq_re_init);
        if (irq_re_init) {
                bnxt_free_stats(bp);
                bnxt_free_ring_grps(bp);
                bnxt_free_vnics(bp);
                kfree(bp->tx_ring);
                bp->tx_ring = NULL;
                kfree(bp->rx_ring);
                bp->rx_ring = NULL;
                kfree(bp->bnapi);
                bp->bnapi = NULL;
        } else {
                bnxt_clear_ring_indices(bp);
        }
}

static int bnxt_alloc_mem(struct bnxt *bp, bool irq_re_init)
{
        int i, j, rc, size, arr_size;
        void *bnapi;

        if (irq_re_init) {
                /* Allocate bnapi mem pointer array and mem block for
                 * all queues
                 */
                arr_size = L1_CACHE_ALIGN(sizeof(struct bnxt_napi *) *
                                bp->cp_nr_rings);
                size = L1_CACHE_ALIGN(sizeof(struct bnxt_napi));
                bnapi = kzalloc(arr_size + size * bp->cp_nr_rings, GFP_KERNEL);
                if (!bnapi)
                        return -ENOMEM;

                bp->bnapi = bnapi;
                bnapi += arr_size;
                for (i = 0; i < bp->cp_nr_rings; i++, bnapi += size) {
                        bp->bnapi[i] = bnapi;
                        bp->bnapi[i]->index = i;
                        bp->bnapi[i]->bp = bp;
                }

                bp->rx_ring = kcalloc(bp->rx_nr_rings,
                                      sizeof(struct bnxt_rx_ring_info),
                                      GFP_KERNEL);
                if (!bp->rx_ring)
                        return -ENOMEM;

                for (i = 0; i < bp->rx_nr_rings; i++) {
                        bp->rx_ring[i].bnapi = bp->bnapi[i];
                        bp->bnapi[i]->rx_ring = &bp->rx_ring[i];
                }

                bp->tx_ring = kcalloc(bp->tx_nr_rings,
                                      sizeof(struct bnxt_tx_ring_info),
                                      GFP_KERNEL);
                if (!bp->tx_ring)
                        return -ENOMEM;

                if (bp->flags & BNXT_FLAG_SHARED_RINGS)
                        j = 0;
                else
                        j = bp->rx_nr_rings;

                for (i = 0; i < bp->tx_nr_rings; i++, j++) {
                        bp->tx_ring[i].bnapi = bp->bnapi[j];
                        bp->bnapi[j]->tx_ring = &bp->tx_ring[i];
                }

                rc = bnxt_alloc_stats(bp);
                if (rc)
                        goto alloc_mem_err;

                rc = bnxt_alloc_ntp_fltrs(bp);
                if (rc)
                        goto alloc_mem_err;

                rc = bnxt_alloc_vnics(bp);
                if (rc)
                        goto alloc_mem_err;
        }

        bnxt_init_ring_struct(bp);

        rc = bnxt_alloc_rx_rings(bp);
        if (rc)
                goto alloc_mem_err;

        rc = bnxt_alloc_tx_rings(bp);
        if (rc)
                goto alloc_mem_err;

        rc = bnxt_alloc_cp_rings(bp);
        if (rc)
                goto alloc_mem_err;

        bp->vnic_info[0].flags |= BNXT_VNIC_RSS_FLAG | BNXT_VNIC_MCAST_FLAG |
                                  BNXT_VNIC_UCAST_FLAG;
        rc = bnxt_alloc_vnic_attributes(bp);
        if (rc)
                goto alloc_mem_err;
        return 0;

alloc_mem_err:
        bnxt_free_mem(bp, true);
        return rc;
}

void bnxt_hwrm_cmd_hdr_init(struct bnxt *bp, void *request, u16 req_type,
                            u16 cmpl_ring, u16 target_id)
{
        struct input *req = request;

        req->req_type = cpu_to_le16(req_type);
        req->cmpl_ring = cpu_to_le16(cmpl_ring);
        req->target_id = cpu_to_le16(target_id);
        req->resp_addr = cpu_to_le64(bp->hwrm_cmd_resp_dma_addr);
}

static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len,
                                 int timeout, bool silent)
{
        int i, intr_process, rc;
        struct input *req = msg;
        u32 *data = msg;
        __le32 *resp_len, *valid;
        u16 cp_ring_id, len = 0;
        struct hwrm_err_output *resp = bp->hwrm_cmd_resp_addr;

        req->seq_id = cpu_to_le16(bp->hwrm_cmd_seq++);
        memset(resp, 0, PAGE_SIZE);
        cp_ring_id = le16_to_cpu(req->cmpl_ring);
        intr_process = (cp_ring_id == INVALID_HW_RING_ID) ? 0 : 1;

        /* Write request msg to hwrm channel */
        __iowrite32_copy(bp->bar0, data, msg_len / 4);

        for (i = msg_len; i < BNXT_HWRM_MAX_REQ_LEN; i += 4)
                writel(0, bp->bar0 + i);

        /* currently supports only one outstanding message */
        if (intr_process)
                bp->hwrm_intr_seq_id = le16_to_cpu(req->seq_id);

        /* Ring channel doorbell */
        writel(1, bp->bar0 + 0x100);

        if (!timeout)
                timeout = DFLT_HWRM_CMD_TIMEOUT;

        i = 0;
        if (intr_process) {
                /* Wait until hwrm response cmpl interrupt is processed */
                while (bp->hwrm_intr_seq_id != HWRM_SEQ_ID_INVALID &&
                       i++ < timeout) {
                        usleep_range(600, 800);
                }

                if (bp->hwrm_intr_seq_id != HWRM_SEQ_ID_INVALID) {
                        netdev_err(bp->dev, "Resp cmpl intr err msg: 0x%x\n",
                                   le16_to_cpu(req->req_type));
                        return -1;
                }
        } else {
                /* Check if response len is updated */
                resp_len = bp->hwrm_cmd_resp_addr + HWRM_RESP_LEN_OFFSET;
                for (i = 0; i < timeout; i++) {
                        len = (le32_to_cpu(*resp_len) & HWRM_RESP_LEN_MASK) >>
                              HWRM_RESP_LEN_SFT;
                        if (len)
                                break;
                        usleep_range(600, 800);
                }

                if (i >= timeout) {
                        netdev_err(bp->dev, "Error (timeout: %d) msg {0x%x 0x%x} len:%d\n",
                                   timeout, le16_to_cpu(req->req_type),
                                   le16_to_cpu(req->seq_id), *resp_len);
                        return -1;
                }

                /* Last word of resp contains valid bit */
                valid = bp->hwrm_cmd_resp_addr + len - 4;
                for (i = 0; i < timeout; i++) {
                        if (le32_to_cpu(*valid) & HWRM_RESP_VALID_MASK)
                                break;
                        usleep_range(600, 800);
                }

                if (i >= timeout) {
                        netdev_err(bp->dev, "Error (timeout: %d) msg {0x%x 0x%x} len:%d v:%d\n",
                                   timeout, le16_to_cpu(req->req_type),
                                   le16_to_cpu(req->seq_id), len, *valid);
                        return -1;
                }
        }

        rc = le16_to_cpu(resp->error_code);
        if (rc && !silent)
                netdev_err(bp->dev, "hwrm req_type 0x%x seq id 0x%x error 0x%x\n",
                           le16_to_cpu(resp->req_type),
                           le16_to_cpu(resp->seq_id), rc);
        return rc;
}

int _hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
{
        return bnxt_hwrm_do_send_msg(bp, msg, msg_len, timeout, false);
}

int hwrm_send_message(struct bnxt *bp, void *msg, u32 msg_len, int timeout)
{
        int rc;

        mutex_lock(&bp->hwrm_cmd_lock);
        rc = _hwrm_send_message(bp, msg, msg_len, timeout);
        mutex_unlock(&bp->hwrm_cmd_lock);
        return rc;
}

int hwrm_send_message_silent(struct bnxt *bp, void *msg, u32 msg_len,
                             int timeout)
{
        int rc;

        mutex_lock(&bp->hwrm_cmd_lock);
        rc = bnxt_hwrm_do_send_msg(bp, msg, msg_len, timeout, true);
        mutex_unlock(&bp->hwrm_cmd_lock);
        return rc;
}

static int bnxt_hwrm_func_drv_rgtr(struct bnxt *bp)
{
        struct hwrm_func_drv_rgtr_input req = {0};
        int i;

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_RGTR, -1, -1);

        req.enables =
                cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_OS_TYPE |
                            FUNC_DRV_RGTR_REQ_ENABLES_VER |
                            FUNC_DRV_RGTR_REQ_ENABLES_ASYNC_EVENT_FWD);

        /* TODO: current async event fwd bits are not defined and the firmware
         * only checks if it is non-zero to enable async event forwarding
         */
        req.async_event_fwd[0] |= cpu_to_le32(1);
        req.os_type = cpu_to_le16(1);
        req.ver_maj = DRV_VER_MAJ;
        req.ver_min = DRV_VER_MIN;
        req.ver_upd = DRV_VER_UPD;

        if (BNXT_PF(bp)) {
                DECLARE_BITMAP(vf_req_snif_bmap, 256);
                u32 *data = (u32 *)vf_req_snif_bmap;

                memset(vf_req_snif_bmap, 0, sizeof(vf_req_snif_bmap));
                for (i = 0; i < ARRAY_SIZE(bnxt_vf_req_snif); i++)
                        __set_bit(bnxt_vf_req_snif[i], vf_req_snif_bmap);

                for (i = 0; i < 8; i++)
                        req.vf_req_fwd[i] = cpu_to_le32(data[i]);

                req.enables |=
                        cpu_to_le32(FUNC_DRV_RGTR_REQ_ENABLES_VF_REQ_FWD);
        }

        return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
}

static int bnxt_hwrm_func_drv_unrgtr(struct bnxt *bp)
{
        struct hwrm_func_drv_unrgtr_input req = {0};

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FUNC_DRV_UNRGTR, -1, -1);
        return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
}

static int bnxt_hwrm_tunnel_dst_port_free(struct bnxt *bp, u8 tunnel_type)
{
        u32 rc = 0;
        struct hwrm_tunnel_dst_port_free_input req = {0};

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_TUNNEL_DST_PORT_FREE, -1, -1);
        req.tunnel_type = tunnel_type;

        switch (tunnel_type) {
        case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_VXLAN:
                req.tunnel_dst_port_id = bp->vxlan_fw_dst_port_id;
                break;
        case TUNNEL_DST_PORT_FREE_REQ_TUNNEL_TYPE_GENEVE:
                req.tunnel_dst_port_id = bp->nge_fw_dst_port_id;
                break;
        default:
                break;
        }

        rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
        if (rc)
                netdev_err(bp->dev, "hwrm_tunnel_dst_port_free failed. rc:%d\n",
                           rc);
        return rc;
}

static int bnxt_hwrm_tunnel_dst_port_alloc(struct bnxt *bp, __be16 port,
                                           u8 tunnel_type)
{
        u32 rc = 0;
        struct hwrm_tunnel_dst_port_alloc_input req = {0};
        struct hwrm_tunnel_dst_port_alloc_output *resp = bp->hwrm_cmd_resp_addr;

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_TUNNEL_DST_PORT_ALLOC, -1, -1);

        req.tunnel_type = tunnel_type;
        req.tunnel_dst_port_val = port;

        mutex_lock(&bp->hwrm_cmd_lock);
        rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
        if (rc) {
                netdev_err(bp->dev, "hwrm_tunnel_dst_port_alloc failed. rc:%d\n",
                           rc);
                goto err_out;
        }

        if (tunnel_type & TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_VXLAN)
                bp->vxlan_fw_dst_port_id = resp->tunnel_dst_port_id;

        else if (tunnel_type & TUNNEL_DST_PORT_ALLOC_REQ_TUNNEL_TYPE_GENEVE)
                bp->nge_fw_dst_port_id = resp->tunnel_dst_port_id;
err_out:
        mutex_unlock(&bp->hwrm_cmd_lock);
        return rc;
}

static int bnxt_hwrm_cfa_l2_set_rx_mask(struct bnxt *bp, u16 vnic_id)
{
        struct hwrm_cfa_l2_set_rx_mask_input req = {0};
        struct bnxt_vnic_info *vnic = &bp->vnic_info[vnic_id];

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_L2_SET_RX_MASK, -1, -1);
        req.vnic_id = cpu_to_le32(vnic->fw_vnic_id);

        req.num_mc_entries = cpu_to_le32(vnic->mc_list_count);
        req.mc_tbl_addr = cpu_to_le64(vnic->mc_list_mapping);
        req.mask = cpu_to_le32(vnic->rx_mask);
        return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
}

#ifdef CONFIG_RFS_ACCEL
static int bnxt_hwrm_cfa_ntuple_filter_free(struct bnxt *bp,
                                            struct bnxt_ntuple_filter *fltr)
{
        struct hwrm_cfa_ntuple_filter_free_input req = {0};

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_NTUPLE_FILTER_FREE, -1, -1);
        req.ntuple_filter_id = fltr->filter_id;
        return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
}

#define BNXT_NTP_FLTR_FLAGS                                     \
        (CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_L2_FILTER_ID |     \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_ETHERTYPE |        \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_MACADDR |      \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_IPADDR_TYPE |      \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR |       \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_IPADDR_MASK |  \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_IPADDR |       \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_IPADDR_MASK |  \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_IP_PROTOCOL |      \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_PORT |         \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_SRC_PORT_MASK |    \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_PORT |         \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_PORT_MASK |    \
         CFA_NTUPLE_FILTER_ALLOC_REQ_ENABLES_DST_ID)

static int bnxt_hwrm_cfa_ntuple_filter_alloc(struct bnxt *bp,
                                             struct bnxt_ntuple_filter *fltr)
{
        int rc = 0;
        struct hwrm_cfa_ntuple_filter_alloc_input req = {0};
        struct hwrm_cfa_ntuple_filter_alloc_output *resp =
                bp->hwrm_cmd_resp_addr;
        struct flow_keys *keys = &fltr->fkeys;
        struct bnxt_vnic_info *vnic = &bp->vnic_info[fltr->rxq + 1];

        bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_CFA_NTUPLE_FILTER_ALLOC, -1, -1);
        req.l2_filter_id = bp->vnic_info[0].fw_l2_filter_id[0];

        req.enables = cpu_to_le32(BNXT_NTP_FLTR_FLAGS);

        req.ethertype = htons(ETH_P_IP);