/**********************************************************************
 * Author: Cavium, Inc.
 *
 * Contact: support@cavium.com
 *          Please include "LiquidIO" in the subject.
 *
 * Copyright (c) 2003-2016 Cavium, Inc.
 *
 * This file is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License, Version 2, as
 * published by the Free Software Foundation.
 *
 * This file is distributed in the hope that it will be useful, but
 * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
 * NONINFRINGEMENT.  See the GNU General Public License for more details.
 ***********************************************************************/
#include <linux/module.h>
#include <linux/interrupt.h>
#include <linux/pci.h>
#include <linux/firmware.h>
#include <net/vxlan.h>
#include <linux/kthread.h>
#include "liquidio_common.h"
#include "octeon_droq.h"
#include "octeon_iq.h"
#include "response_manager.h"
#include "octeon_device.h"
#include "octeon_nic.h"
#include "octeon_main.h"
#include "octeon_network.h"
#include "cn66xx_regs.h"
#include "cn66xx_device.h"
#include "cn68xx_device.h"
#include "cn23xx_pf_device.h"
#include "liquidio_image.h"
#include "lio_vf_rep.h"

MODULE_AUTHOR("Cavium Networks, <support@cavium.com>");
MODULE_DESCRIPTION("Cavium LiquidIO Intelligent Server Adapter Driver");
MODULE_LICENSE("GPL");
MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210SV_NAME
                "_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_210NV_NAME
                "_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_410NV_NAME
                "_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);
MODULE_FIRMWARE(LIO_FW_DIR LIO_FW_BASE_NAME LIO_23XX_NAME
                "_" LIO_FW_NAME_TYPE_NIC LIO_FW_NAME_SUFFIX);

static int ddr_timeout = 10000;
module_param(ddr_timeout, int, 0644);
MODULE_PARM_DESC(ddr_timeout,
                 "Number of milliseconds to wait for DDR initialization. 0 waits for ddr_timeout to be set to non-zero value before starting to check");

#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)

static int debug = -1;
module_param(debug, int, 0644);
MODULE_PARM_DESC(debug, "NETIF_MSG debug bits");

static char fw_type[LIO_MAX_FW_TYPE_LEN] = LIO_FW_NAME_TYPE_AUTO;
module_param_string(fw_type, fw_type, sizeof(fw_type), 0444);
MODULE_PARM_DESC(fw_type, "Type of firmware to be loaded (default is \"auto\"), which uses firmware in flash, if present, else loads \"nic\".");

static u32 console_bitmask;
module_param(console_bitmask, int, 0644);
MODULE_PARM_DESC(console_bitmask,
                 "Bitmask indicating which consoles have debug output redirected to syslog.");

/**
 * octeon_console_debug_enabled - determines if a given console has debug enabled.
 * @console: console to check
 * Return:  1 = enabled. 0 otherwise
 */
static int octeon_console_debug_enabled(u32 console)
{
        return (console_bitmask >> (console)) & 0x1;
}

/* Polling interval for determining when NIC application is alive */
#define LIQUIDIO_STARTER_POLL_INTERVAL_MS 100

/* runtime link query interval */
#define LIQUIDIO_LINK_QUERY_INTERVAL_MS         1000
/* update localtime to octeon firmware every 60 seconds.
 * make firmware to use same time reference, so that it will be easy to
 * correlate firmware logged events/errors with host events, for debugging.
 */
#define LIO_SYNC_OCTEON_TIME_INTERVAL_MS 60000

/* time to wait for possible in-flight requests in milliseconds */
#define WAIT_INFLIGHT_REQUEST   msecs_to_jiffies(1000)

struct lio_trusted_vf_ctx {
        struct completion complete;
        int status;
};

struct oct_link_status_resp {
        u64 rh;
        struct oct_link_info link_info;
        u64 status;
};

struct oct_timestamp_resp {
        u64 rh;
        u64 timestamp;
        u64 status;
};

#define OCT_TIMESTAMP_RESP_SIZE (sizeof(struct oct_timestamp_resp))

union tx_info {
        u64 u64;
        struct {
#ifdef __BIG_ENDIAN_BITFIELD
                u16 gso_size;
                u16 gso_segs;
                u32 reserved;
#else
                u32 reserved;
                u16 gso_segs;
                u16 gso_size;
#endif
        } s;
};

/* Octeon device properties to be used by the NIC module.
 * Each octeon device in the system will be represented
 * by this structure in the NIC module.
 */

#define OCTNIC_GSO_MAX_HEADER_SIZE 128
#define OCTNIC_GSO_MAX_SIZE                                                    \
        (CN23XX_DEFAULT_INPUT_JABBER - OCTNIC_GSO_MAX_HEADER_SIZE)

struct handshake {
        struct completion init;
        struct completion started;
        struct pci_dev *pci_dev;
        int init_ok;
        int started_ok;
};

#ifdef CONFIG_PCI_IOV
static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs);
#endif

static int octeon_dbg_console_print(struct octeon_device *oct, u32 console_num,
                                    char *prefix, char *suffix);

static int octeon_device_init(struct octeon_device *);
static int liquidio_stop(struct net_device *netdev);
static void liquidio_remove(struct pci_dev *pdev);
static int liquidio_probe(struct pci_dev *pdev,
                          const struct pci_device_id *ent);
static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
                                      int linkstate);

static struct handshake handshake[MAX_OCTEON_DEVICES];
static struct completion first_stage;

static void octeon_droq_bh(struct tasklet_struct *t)
{
        int q_no;
        int reschedule = 0;
        struct octeon_device_priv *oct_priv = from_tasklet(oct_priv, t,
                                                          droq_tasklet);
        struct octeon_device *oct = oct_priv->dev;

        for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES(oct); q_no++) {
                if (!(oct->io_qmask.oq & BIT_ULL(q_no)))
                        continue;
                reschedule |= octeon_droq_process_packets(oct, oct->droq[q_no],
                                                          MAX_PACKET_BUDGET);
                lio_enable_irq(oct->droq[q_no], NULL);

                if (OCTEON_CN23XX_PF(oct) && oct->msix_on) {
                        /* set time and cnt interrupt thresholds for this DROQ
                         * for NAPI
                         */
                        int adjusted_q_no = q_no + oct->sriov_info.pf_srn;

                        octeon_write_csr64(
                            oct, CN23XX_SLI_OQ_PKT_INT_LEVELS(adjusted_q_no),
                            0x5700000040ULL);
                        octeon_write_csr64(
                            oct, CN23XX_SLI_OQ_PKTS_SENT(adjusted_q_no), 0);
                }
        }

        if (reschedule)
                tasklet_schedule(&oct_priv->droq_tasklet);
}

static int lio_wait_for_oq_pkts(struct octeon_device *oct)
{
        struct octeon_device_priv *oct_priv =
                (struct octeon_device_priv *)oct->priv;
        int retry = 100, pkt_cnt = 0, pending_pkts = 0;
        int i;

        do {
                pending_pkts = 0;

                for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
                        if (!(oct->io_qmask.oq & BIT_ULL(i)))
                                continue;
                        pkt_cnt += octeon_droq_check_hw_for_pkts(oct->droq[i]);
                }
                if (pkt_cnt > 0) {
                        pending_pkts += pkt_cnt;
                        tasklet_schedule(&oct_priv->droq_tasklet);
                }
                pkt_cnt = 0;
                schedule_timeout_uninterruptible(1);

        } while (retry-- && pending_pkts);

        return pkt_cnt;
}

/**
 * force_io_queues_off - Forces all IO queues off on a given device
 * @oct: Pointer to Octeon device
 */
static void force_io_queues_off(struct octeon_device *oct)
{
        if ((oct->chip_id == OCTEON_CN66XX) ||
            (oct->chip_id == OCTEON_CN68XX)) {
                /* Reset the Enable bits for Input Queues. */
                octeon_write_csr(oct, CN6XXX_SLI_PKT_INSTR_ENB, 0);

                /* Reset the Enable bits for Output Queues. */
                octeon_write_csr(oct, CN6XXX_SLI_PKT_OUT_ENB, 0);
        }
}

/**
 * pcierror_quiesce_device - Cause device to go quiet so it can be safely removed/reset/etc
 * @oct: Pointer to Octeon device
 */
static inline void pcierror_quiesce_device(struct octeon_device *oct)
{
        int i;

        /* Disable the input and output queues now. No more packets will
         * arrive from Octeon, but we should wait for all packet processing
         * to finish.
         */
        force_io_queues_off(oct);

        /* To allow for in-flight requests */
        schedule_timeout_uninterruptible(WAIT_INFLIGHT_REQUEST);

        if (wait_for_pending_requests(oct))
                dev_err(&oct->pci_dev->dev, "There were pending requests\n");

        /* Force all requests waiting to be fetched by OCTEON to complete. */
        for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
                struct octeon_instr_queue *iq;

                if (!(oct->io_qmask.iq & BIT_ULL(i)))
                        continue;
                iq = oct->instr_queue[i];

                if (atomic_read(&iq->instr_pending)) {
                        spin_lock_bh(&iq->lock);
                        iq->fill_cnt = 0;
                        iq->octeon_read_index = iq->host_write_index;
                        iq->stats.instr_processed +=
                                atomic_read(&iq->instr_pending);
                        lio_process_iq_request_list(oct, iq, 0);
                        spin_unlock_bh(&iq->lock);
                }
        }

        /* Force all pending ordered list requests to time out. */
        lio_process_ordered_list(oct, 1);

        /* We do not need to wait for output queue packets to be processed. */
}

/**
 * cleanup_aer_uncorrect_error_status - Cleanup PCI AER uncorrectable error status
 * @dev: Pointer to PCI device
 */
static void cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
{
        int pos = 0x100;
        u32 status, mask;

        pr_info("%s :\n", __func__);

        pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
        pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, &mask);
        if (dev->error_state == pci_channel_io_normal)
                status &= ~mask;        /* Clear corresponding nonfatal bits */
        else
                status &= mask;         /* Clear corresponding fatal bits */
        pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, status);
}

/**
 * stop_pci_io - Stop all PCI IO to a given device
 * @oct: Pointer to Octeon device
 */
static void stop_pci_io(struct octeon_device *oct)
{
        /* No more instructions will be forwarded. */
        atomic_set(&oct->status, OCT_DEV_IN_RESET);

        pci_disable_device(oct->pci_dev);

        /* Disable interrupts  */
        oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);

        pcierror_quiesce_device(oct);

        /* Release the interrupt line */
        free_irq(oct->pci_dev->irq, oct);

        if (oct->flags & LIO_FLAG_MSI_ENABLED)
                pci_disable_msi(oct->pci_dev);

        dev_dbg(&oct->pci_dev->dev, "Device state is now %s\n",
                lio_get_state_string(&oct->status));

        /* making it a common function for all OCTEON models */
        cleanup_aer_uncorrect_error_status(oct->pci_dev);
}

/**
 * liquidio_pcie_error_detected - called when PCI error is detected
 * @pdev: Pointer to PCI device
 * @state: The current pci connection state
 *
 * This function is called after a PCI bus error affecting
 * this device has been detected.
 */
static pci_ers_result_t liquidio_pcie_error_detected(struct pci_dev *pdev,
                                                     pci_channel_state_t state)
{
        struct octeon_device *oct = pci_get_drvdata(pdev);

        /* Non-correctable Non-fatal errors */
        if (state == pci_channel_io_normal) {
                dev_err(&oct->pci_dev->dev, "Non-correctable non-fatal error reported:\n");
                cleanup_aer_uncorrect_error_status(oct->pci_dev);
                return PCI_ERS_RESULT_CAN_RECOVER;
        }

        /* Non-correctable Fatal errors */
        dev_err(&oct->pci_dev->dev, "Non-correctable FATAL reported by PCI AER driver\n");
        stop_pci_io(oct);

        /* Always return a DISCONNECT. There is no support for recovery but only
         * for a clean shutdown.
         */
        return PCI_ERS_RESULT_DISCONNECT;
}

/**
 * liquidio_pcie_mmio_enabled - mmio handler
 * @pdev: Pointer to PCI device
 */
static pci_ers_result_t liquidio_pcie_mmio_enabled(struct pci_dev __maybe_unused *pdev)
{
        /* We should never hit this since we never ask for a reset for a Fatal
         * Error. We always return DISCONNECT in io_error above.
         * But play safe and return RECOVERED for now.
         */
        return PCI_ERS_RESULT_RECOVERED;
}

/**
 * liquidio_pcie_slot_reset - called after the pci bus has been reset.
 * @pdev: Pointer to PCI device
 *
 * Restart the card from scratch, as if from a cold-boot. Implementation
 * resembles the first-half of the octeon_resume routine.
 */
static pci_ers_result_t liquidio_pcie_slot_reset(struct pci_dev __maybe_unused *pdev)
{
        /* We should never hit this since we never ask for a reset for a Fatal
         * Error. We always return DISCONNECT in io_error above.
         * But play safe and return RECOVERED for now.
         */
        return PCI_ERS_RESULT_RECOVERED;
}

/**
 * liquidio_pcie_resume - called when traffic can start flowing again.
 * @pdev: Pointer to PCI device
 *
 * This callback is called when the error recovery driver tells us that
 * its OK to resume normal operation. Implementation resembles the
 * second-half of the octeon_resume routine.
 */
static void liquidio_pcie_resume(struct pci_dev __maybe_unused *pdev)
{
        /* Nothing to be done here. */
}

#define liquidio_suspend NULL
#define liquidio_resume NULL

/* For PCI-E Advanced Error Recovery (AER) Interface */
static const struct pci_error_handlers liquidio_err_handler = {
        .error_detected = liquidio_pcie_error_detected,
        .mmio_enabled   = liquidio_pcie_mmio_enabled,
        .slot_reset     = liquidio_pcie_slot_reset,
        .resume         = liquidio_pcie_resume,
};

static const struct pci_device_id liquidio_pci_tbl[] = {
        {       /* 68xx */
                PCI_VENDOR_ID_CAVIUM, 0x91, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0
        },
        {       /* 66xx */
                PCI_VENDOR_ID_CAVIUM, 0x92, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0
        },
        {       /* 23xx pf */
                PCI_VENDOR_ID_CAVIUM, 0x9702, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0
        },
        {
                0, 0, 0, 0, 0, 0, 0
        }
};
MODULE_DEVICE_TABLE(pci, liquidio_pci_tbl);

static SIMPLE_DEV_PM_OPS(liquidio_pm_ops, liquidio_suspend, liquidio_resume);

static struct pci_driver liquidio_pci_driver = {
        .name           = "LiquidIO",
        .id_table       = liquidio_pci_tbl,
        .probe          = liquidio_probe,
        .remove         = liquidio_remove,
        .err_handler    = &liquidio_err_handler,    /* For AER */
        .driver.pm      = &liquidio_pm_ops,
#ifdef CONFIG_PCI_IOV
        .sriov_configure = liquidio_enable_sriov,
#endif
};

/**
 * liquidio_init_pci - register PCI driver
 */
static int liquidio_init_pci(void)
{
        return pci_register_driver(&liquidio_pci_driver);
}

/**
 * liquidio_deinit_pci - unregister PCI driver
 */
static void liquidio_deinit_pci(void)
{
        pci_unregister_driver(&liquidio_pci_driver);
}

/**
 * check_txq_status - Check Tx queue status, and take appropriate action
 * @lio: per-network private data
 * Return: 0 if full, number of queues woken up otherwise
 */
static inline int check_txq_status(struct lio *lio)
{
        int numqs = lio->netdev->real_num_tx_queues;
        int ret_val = 0;
        int q, iq;

        /* check each sub-queue state */
        for (q = 0; q < numqs; q++) {
                iq = lio->linfo.txpciq[q %
                        lio->oct_dev->num_iqs].s.q_no;
                if (octnet_iq_is_full(lio->oct_dev, iq))
                        continue;
                if (__netif_subqueue_stopped(lio->netdev, q)) {
                        netif_wake_subqueue(lio->netdev, q);
                        INCR_INSTRQUEUE_PKT_COUNT(lio->oct_dev, iq,
                                                  tx_restart, 1);
                        ret_val++;
                }
        }

        return ret_val;
}

/**
 * print_link_info -  Print link information
 * @netdev: network device
 */
static void print_link_info(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);

        if (!ifstate_check(lio, LIO_IFSTATE_RESETTING) &&
            ifstate_check(lio, LIO_IFSTATE_REGISTERED)) {
                struct oct_link_info *linfo = &lio->linfo;

                if (linfo->link.s.link_up) {
                        netif_info(lio, link, lio->netdev, "%d Mbps %s Duplex UP\n",
                                   linfo->link.s.speed,
                                   (linfo->link.s.duplex) ? "Full" : "Half");
                } else {
                        netif_info(lio, link, lio->netdev, "Link Down\n");
                }
        }
}

/**
 * octnet_link_status_change - Routine to notify MTU change
 * @work: work_struct data structure
 */
static void octnet_link_status_change(struct work_struct *work)
{
        struct cavium_wk *wk = (struct cavium_wk *)work;
        struct lio *lio = (struct lio *)wk->ctxptr;

        /* lio->linfo.link.s.mtu always contains max MTU of the lio interface.
         * this API is invoked only when new max-MTU of the interface is
         * less than current MTU.
         */
        rtnl_lock();
        dev_set_mtu(lio->netdev, lio->linfo.link.s.mtu);
        rtnl_unlock();
}

/**
 * setup_link_status_change_wq - Sets up the mtu status change work
 * @netdev: network device
 */
static inline int setup_link_status_change_wq(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;

        lio->link_status_wq.wq = alloc_workqueue("link-status",
                                                 WQ_MEM_RECLAIM, 0);
        if (!lio->link_status_wq.wq) {
                dev_err(&oct->pci_dev->dev, "unable to create cavium link status wq\n");
                return -1;
        }
        INIT_DELAYED_WORK(&lio->link_status_wq.wk.work,
                          octnet_link_status_change);
        lio->link_status_wq.wk.ctxptr = lio;

        return 0;
}

static inline void cleanup_link_status_change_wq(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);

        if (lio->link_status_wq.wq) {
                cancel_delayed_work_sync(&lio->link_status_wq.wk.work);
                destroy_workqueue(lio->link_status_wq.wq);
        }
}

/**
 * update_link_status - Update link status
 * @netdev: network device
 * @ls: link status structure
 *
 * Called on receipt of a link status response from the core application to
 * update each interface's link status.
 */
static inline void update_link_status(struct net_device *netdev,
                                      union oct_link_status *ls)
{
        struct lio *lio = GET_LIO(netdev);
        int changed = (lio->linfo.link.u64 != ls->u64);
        int current_max_mtu = lio->linfo.link.s.mtu;
        struct octeon_device *oct = lio->oct_dev;

        dev_dbg(&oct->pci_dev->dev, "%s: lio->linfo.link.u64=%llx, ls->u64=%llx\n",
                __func__, lio->linfo.link.u64, ls->u64);
        lio->linfo.link.u64 = ls->u64;

        if ((lio->intf_open) && (changed)) {
                print_link_info(netdev);
                lio->link_changes++;

                if (lio->linfo.link.s.link_up) {
                        dev_dbg(&oct->pci_dev->dev, "%s: link_up", __func__);
                        netif_carrier_on(netdev);
                        wake_txqs(netdev);
                } else {
                        dev_dbg(&oct->pci_dev->dev, "%s: link_off", __func__);
                        netif_carrier_off(netdev);
                        stop_txqs(netdev);
                }
                if (lio->linfo.link.s.mtu != current_max_mtu) {
                        netif_info(lio, probe, lio->netdev, "Max MTU changed from %d to %d\n",
                                   current_max_mtu, lio->linfo.link.s.mtu);
                        netdev->max_mtu = lio->linfo.link.s.mtu;
                }
                if (lio->linfo.link.s.mtu < netdev->mtu) {
                        dev_warn(&oct->pci_dev->dev,
                                 "Current MTU is higher than new max MTU; Reducing the current mtu from %d to %d\n",
                                     netdev->mtu, lio->linfo.link.s.mtu);
                        queue_delayed_work(lio->link_status_wq.wq,
                                           &lio->link_status_wq.wk.work, 0);
                }
        }
}

/**
 * lio_sync_octeon_time - send latest localtime to octeon firmware so that
 * firmware will correct it's time, in case there is a time skew
 *
 * @work: work scheduled to send time update to octeon firmware
 **/
static void lio_sync_octeon_time(struct work_struct *work)
{
        struct cavium_wk *wk = (struct cavium_wk *)work;
        struct lio *lio = (struct lio *)wk->ctxptr;
        struct octeon_device *oct = lio->oct_dev;
        struct octeon_soft_command *sc;
        struct timespec64 ts;
        struct lio_time *lt;
        int ret;

        sc = octeon_alloc_soft_command(oct, sizeof(struct lio_time), 16, 0);
        if (!sc) {
                dev_err(&oct->pci_dev->dev,
                        "Failed to sync time to octeon: soft command allocation failed\n");
                return;
        }

        lt = (struct lio_time *)sc->virtdptr;

        /* Get time of the day */
        ktime_get_real_ts64(&ts);
        lt->sec = ts.tv_sec;
        lt->nsec = ts.tv_nsec;
        octeon_swap_8B_data((u64 *)lt, (sizeof(struct lio_time)) / 8);

        sc->iq_no = lio->linfo.txpciq[0].s.q_no;
        octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
                                    OPCODE_NIC_SYNC_OCTEON_TIME, 0, 0, 0);

        init_completion(&sc->complete);
        sc->sc_status = OCTEON_REQUEST_PENDING;

        ret = octeon_send_soft_command(oct, sc);
        if (ret == IQ_SEND_FAILED) {
                dev_err(&oct->pci_dev->dev,
                        "Failed to sync time to octeon: failed to send soft command\n");
                octeon_free_soft_command(oct, sc);
        } else {
                WRITE_ONCE(sc->caller_is_done, true);
        }

        queue_delayed_work(lio->sync_octeon_time_wq.wq,
                           &lio->sync_octeon_time_wq.wk.work,
                           msecs_to_jiffies(LIO_SYNC_OCTEON_TIME_INTERVAL_MS));
}

/**
 * setup_sync_octeon_time_wq - prepare work to periodically update local time to octeon firmware
 *
 * @netdev: network device which should send time update to firmware
 **/
static inline int setup_sync_octeon_time_wq(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;

        lio->sync_octeon_time_wq.wq =
                alloc_workqueue("update-octeon-time", WQ_MEM_RECLAIM, 0);
        if (!lio->sync_octeon_time_wq.wq) {
                dev_err(&oct->pci_dev->dev, "Unable to create wq to update octeon time\n");
                return -1;
        }
        INIT_DELAYED_WORK(&lio->sync_octeon_time_wq.wk.work,
                          lio_sync_octeon_time);
        lio->sync_octeon_time_wq.wk.ctxptr = lio;
        queue_delayed_work(lio->sync_octeon_time_wq.wq,
                           &lio->sync_octeon_time_wq.wk.work,
                           msecs_to_jiffies(LIO_SYNC_OCTEON_TIME_INTERVAL_MS));

        return 0;
}

/**
 * cleanup_sync_octeon_time_wq - destroy wq
 *
 * @netdev: network device which should send time update to firmware
 *
 * Stop scheduling and destroy the work created to periodically update local
 * time to octeon firmware.
 **/
static inline void cleanup_sync_octeon_time_wq(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct cavium_wq *time_wq = &lio->sync_octeon_time_wq;

        if (time_wq->wq) {
                cancel_delayed_work_sync(&time_wq->wk.work);
                destroy_workqueue(time_wq->wq);
        }
}

static struct octeon_device *get_other_octeon_device(struct octeon_device *oct)
{
        struct octeon_device *other_oct;

        other_oct = lio_get_device(oct->octeon_id + 1);

        if (other_oct && other_oct->pci_dev) {
                int oct_busnum, other_oct_busnum;

                oct_busnum = oct->pci_dev->bus->number;
                other_oct_busnum = other_oct->pci_dev->bus->number;

                if (oct_busnum == other_oct_busnum) {
                        int oct_slot, other_oct_slot;

                        oct_slot = PCI_SLOT(oct->pci_dev->devfn);
                        other_oct_slot = PCI_SLOT(other_oct->pci_dev->devfn);

                        if (oct_slot == other_oct_slot)
                                return other_oct;
                }
        }

        return NULL;
}

static void disable_all_vf_links(struct octeon_device *oct)
{
        struct net_device *netdev;
        int max_vfs, vf, i;

        if (!oct)
                return;

        max_vfs = oct->sriov_info.max_vfs;

        for (i = 0; i < oct->ifcount; i++) {
                netdev = oct->props[i].netdev;
                if (!netdev)
                        continue;

                for (vf = 0; vf < max_vfs; vf++)
                        liquidio_set_vf_link_state(netdev, vf,
                                                   IFLA_VF_LINK_STATE_DISABLE);
        }
}

static int liquidio_watchdog(void *param)
{
        bool err_msg_was_printed[LIO_MAX_CORES];
        u16 mask_of_crashed_or_stuck_cores = 0;
        bool all_vf_links_are_disabled = false;
        struct octeon_device *oct = param;
        struct octeon_device *other_oct;
#ifdef CONFIG_MODULE_UNLOAD
        long refcount, vfs_referencing_pf;
        u64 vfs_mask1, vfs_mask2;
#endif
        int core;

        memset(err_msg_was_printed, 0, sizeof(err_msg_was_printed));

        while (!kthread_should_stop()) {
                /* sleep for a couple of seconds so that we don't hog the CPU */
                set_current_state(TASK_INTERRUPTIBLE);
                schedule_timeout(msecs_to_jiffies(2000));

                mask_of_crashed_or_stuck_cores =
                    (u16)octeon_read_csr64(oct, CN23XX_SLI_SCRATCH2);

                if (!mask_of_crashed_or_stuck_cores)
                        continue;

                WRITE_ONCE(oct->cores_crashed, true);
                other_oct = get_other_octeon_device(oct);
                if (other_oct)
                        WRITE_ONCE(other_oct->cores_crashed, true);

                for (core = 0; core < LIO_MAX_CORES; core++) {
                        bool core_crashed_or_got_stuck;

                        core_crashed_or_got_stuck =
                                                (mask_of_crashed_or_stuck_cores
                                                 >> core) & 1;

                        if (core_crashed_or_got_stuck &&
                            !err_msg_was_printed[core]) {
                                dev_err(&oct->pci_dev->dev,
                                        "ERROR: Octeon core %d crashed or got stuck!  See oct-fwdump for details.\n",
                                        core);
                                err_msg_was_printed[core] = true;
                        }
                }

                if (all_vf_links_are_disabled)
                        continue;

                disable_all_vf_links(oct);
                disable_all_vf_links(other_oct);
                all_vf_links_are_disabled = true;

#ifdef CONFIG_MODULE_UNLOAD
                vfs_mask1 = READ_ONCE(oct->sriov_info.vf_drv_loaded_mask);
                vfs_mask2 = READ_ONCE(other_oct->sriov_info.vf_drv_loaded_mask);

                vfs_referencing_pf  = hweight64(vfs_mask1);
                vfs_referencing_pf += hweight64(vfs_mask2);

                refcount = module_refcount(THIS_MODULE);
                if (refcount >= vfs_referencing_pf) {
                        while (vfs_referencing_pf) {
                                module_put(THIS_MODULE);
                                vfs_referencing_pf--;
                        }
                }
#endif
        }

        return 0;
}

/**
 * liquidio_probe - PCI probe handler
 * @pdev: PCI device structure
 * @ent: unused
 */
static int
liquidio_probe(struct pci_dev *pdev, const struct pci_device_id __maybe_unused *ent)
{
        struct octeon_device *oct_dev = NULL;
        struct handshake *hs;

        oct_dev = octeon_allocate_device(pdev->device,
                                         sizeof(struct octeon_device_priv));
        if (!oct_dev) {
                dev_err(&pdev->dev, "Unable to allocate device\n");
                return -ENOMEM;
        }

        if (pdev->device == OCTEON_CN23XX_PF_VID)
                oct_dev->msix_on = LIO_FLAG_MSIX_ENABLED;

        /* Enable PTP for 6XXX Device */
        if (((pdev->device == OCTEON_CN66XX) ||
             (pdev->device == OCTEON_CN68XX)))
                oct_dev->ptp_enable = true;
        else
                oct_dev->ptp_enable = false;

        dev_info(&pdev->dev, "Initializing device %x:%x.\n",
                 (u32)pdev->vendor, (u32)pdev->device);

        /* Assign octeon_device for this device to the private data area. */
        pci_set_drvdata(pdev, oct_dev);

        /* set linux specific device pointer */
        oct_dev->pci_dev = (void *)pdev;

        oct_dev->subsystem_id = pdev->subsystem_vendor |
                (pdev->subsystem_device << 16);

        hs = &handshake[oct_dev->octeon_id];
        init_completion(&hs->init);
        init_completion(&hs->started);
        hs->pci_dev = pdev;

        if (oct_dev->octeon_id == 0)
                /* first LiquidIO NIC is detected */
                complete(&first_stage);

        if (octeon_device_init(oct_dev)) {
                complete(&hs->init);
                liquidio_remove(pdev);
                return -ENOMEM;
        }

        if (OCTEON_CN23XX_PF(oct_dev)) {
                u8 bus, device, function;

                if (atomic_read(oct_dev->adapter_refcount) == 1) {
                        /* Each NIC gets one watchdog kernel thread.  The first
                         * PF (of each NIC) that gets pci_driver->probe()'d
                         * creates that thread.
                         */
                        bus = pdev->bus->number;
                        device = PCI_SLOT(pdev->devfn);
                        function = PCI_FUNC(pdev->devfn);
                        oct_dev->watchdog_task = kthread_create(
                            liquidio_watchdog, oct_dev,
                            "liowd/%02hhx:%02hhx.%hhx", bus, device, function);
                        if (!IS_ERR(oct_dev->watchdog_task)) {
                                wake_up_process(oct_dev->watchdog_task);
                        } else {
                                oct_dev->watchdog_task = NULL;
                                dev_err(&oct_dev->pci_dev->dev,
                                        "failed to create kernel_thread\n");
                                liquidio_remove(pdev);
                                return -1;
                        }
                }
        }

        oct_dev->rx_pause = 1;
        oct_dev->tx_pause = 1;

        dev_dbg(&oct_dev->pci_dev->dev, "Device is ready\n");

        return 0;
}

static bool fw_type_is_auto(void)
{
        return strncmp(fw_type, LIO_FW_NAME_TYPE_AUTO,
                       sizeof(LIO_FW_NAME_TYPE_AUTO)) == 0;
}

/**
 * octeon_pci_flr - PCI FLR for each Octeon device.
 * @oct: octeon device
 */
static void octeon_pci_flr(struct octeon_device *oct)
{
        int rc;

        pci_save_state(oct->pci_dev);

        pci_cfg_access_lock(oct->pci_dev);

        /* Quiesce the device completely */
        pci_write_config_word(oct->pci_dev, PCI_COMMAND,
                              PCI_COMMAND_INTX_DISABLE);

        rc = __pci_reset_function_locked(oct->pci_dev);

        if (rc != 0)
                dev_err(&oct->pci_dev->dev, "Error %d resetting PCI function %d\n",
                        rc, oct->pf_num);

        pci_cfg_access_unlock(oct->pci_dev);

        pci_restore_state(oct->pci_dev);
}

/**
 * octeon_destroy_resources - Destroy resources associated with octeon device
 * @oct: octeon device
 */
static void octeon_destroy_resources(struct octeon_device *oct)
{
        int i, refcount;
        struct msix_entry *msix_entries;
        struct octeon_device_priv *oct_priv =
                (struct octeon_device_priv *)oct->priv;

        struct handshake *hs;

        switch (atomic_read(&oct->status)) {
        case OCT_DEV_RUNNING:
        case OCT_DEV_CORE_OK:

                /* No more instructions will be forwarded. */
                atomic_set(&oct->status, OCT_DEV_IN_RESET);

                oct->app_mode = CVM_DRV_INVALID_APP;
                dev_dbg(&oct->pci_dev->dev, "Device state is now %s\n",
                        lio_get_state_string(&oct->status));

                schedule_timeout_uninterruptible(HZ / 10);

                fallthrough;
        case OCT_DEV_HOST_OK:

        case OCT_DEV_CONSOLE_INIT_DONE:
                /* Remove any consoles */
                octeon_remove_consoles(oct);

                fallthrough;
        case OCT_DEV_IO_QUEUES_DONE:
                if (lio_wait_for_instr_fetch(oct))
                        dev_err(&oct->pci_dev->dev, "IQ had pending instructions\n");

                if (wait_for_pending_requests(oct))
                        dev_err(&oct->pci_dev->dev, "There were pending requests\n");

                /* Disable the input and output queues now. No more packets will
                 * arrive from Octeon, but we should wait for all packet
                 * processing to finish.
                 */
                oct->fn_list.disable_io_queues(oct);

                if (lio_wait_for_oq_pkts(oct))
                        dev_err(&oct->pci_dev->dev, "OQ had pending packets\n");

                /* Force all requests waiting to be fetched by OCTEON to
                 * complete.
                 */
                for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
                        struct octeon_instr_queue *iq;

                        if (!(oct->io_qmask.iq & BIT_ULL(i)))
                                continue;
                        iq = oct->instr_queue[i];

                        if (atomic_read(&iq->instr_pending)) {
                                spin_lock_bh(&iq->lock);
                                iq->fill_cnt = 0;
                                iq->octeon_read_index = iq->host_write_index;
                                iq->stats.instr_processed +=
                                        atomic_read(&iq->instr_pending);
                                lio_process_iq_request_list(oct, iq, 0);
                                spin_unlock_bh(&iq->lock);
                        }
                }

                lio_process_ordered_list(oct, 1);
                octeon_free_sc_done_list(oct);
                octeon_free_sc_zombie_list(oct);

                fallthrough;
        case OCT_DEV_INTR_SET_DONE:
                /* Disable interrupts  */
                oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);

                if (oct->msix_on) {
                        msix_entries = (struct msix_entry *)oct->msix_entries;
                        for (i = 0; i < oct->num_msix_irqs - 1; i++) {
                                if (oct->ioq_vector[i].vector) {
                                        /* clear the affinity_cpumask */
                                        irq_set_affinity_hint(
                                                        msix_entries[i].vector,
                                                        NULL);
                                        free_irq(msix_entries[i].vector,
                                                 &oct->ioq_vector[i]);
                                        oct->ioq_vector[i].vector = 0;
                                }
                        }
                        /* non-iov vector's argument is oct struct */
                        free_irq(msix_entries[i].vector, oct);

                        pci_disable_msix(oct->pci_dev);
                        kfree(oct->msix_entries);
                        oct->msix_entries = NULL;
                } else {
                        /* Release the interrupt line */
                        free_irq(oct->pci_dev->irq, oct);

                        if (oct->flags & LIO_FLAG_MSI_ENABLED)
                                pci_disable_msi(oct->pci_dev);
                }

                kfree(oct->irq_name_storage);
                oct->irq_name_storage = NULL;

                fallthrough;
        case OCT_DEV_MSIX_ALLOC_VECTOR_DONE:
                if (OCTEON_CN23XX_PF(oct))
                        octeon_free_ioq_vector(oct);

                fallthrough;
        case OCT_DEV_MBOX_SETUP_DONE:
                if (OCTEON_CN23XX_PF(oct))
                        oct->fn_list.free_mbox(oct);

                fallthrough;
        case OCT_DEV_IN_RESET:
        case OCT_DEV_DROQ_INIT_DONE:
                /* Wait for any pending operations */
                mdelay(100);
                for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
                        if (!(oct->io_qmask.oq & BIT_ULL(i)))
                                continue;
                        octeon_delete_droq(oct, i);
                }

                /* Force any pending handshakes to complete */
                for (i = 0; i < MAX_OCTEON_DEVICES; i++) {
                        hs = &handshake[i];

                        if (hs->pci_dev) {
                                handshake[oct->octeon_id].init_ok = 0;
                                complete(&handshake[oct->octeon_id].init);
                                handshake[oct->octeon_id].started_ok = 0;
                                complete(&handshake[oct->octeon_id].started);
                        }
                }

                fallthrough;
        case OCT_DEV_RESP_LIST_INIT_DONE:
                octeon_delete_response_list(oct);

                fallthrough;
        case OCT_DEV_INSTR_QUEUE_INIT_DONE:
                for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
                        if (!(oct->io_qmask.iq & BIT_ULL(i)))
                                continue;
                        octeon_delete_instr_queue(oct, i);
                }
#ifdef CONFIG_PCI_IOV
                if (oct->sriov_info.sriov_enabled)
                        pci_disable_sriov(oct->pci_dev);
#endif
                fallthrough;
        case OCT_DEV_SC_BUFF_POOL_INIT_DONE:
                octeon_free_sc_buffer_pool(oct);

                fallthrough;
        case OCT_DEV_DISPATCH_INIT_DONE:
                octeon_delete_dispatch_list(oct);
                cancel_delayed_work_sync(&oct->nic_poll_work.work);

                fallthrough;
        case OCT_DEV_PCI_MAP_DONE:
                refcount = octeon_deregister_device(oct);

                /* Soft reset the octeon device before exiting.
                 * However, if fw was loaded from card (i.e. autoboot),
                 * perform an FLR instead.
                 * Implementation note: only soft-reset the device
                 * if it is a CN6XXX OR the LAST CN23XX device.
                 */
                if (atomic_read(oct->adapter_fw_state) == FW_IS_PRELOADED)
                        octeon_pci_flr(oct);
                else if (OCTEON_CN6XXX(oct) || !refcount)
                        oct->fn_list.soft_reset(oct);

                octeon_unmap_pci_barx(oct, 0);
                octeon_unmap_pci_barx(oct, 1);

                fallthrough;
        case OCT_DEV_PCI_ENABLE_DONE:
                pci_clear_master(oct->pci_dev);
                /* Disable the device, releasing the PCI INT */
                pci_disable_device(oct->pci_dev);

                fallthrough;
        case OCT_DEV_BEGIN_STATE:
                /* Nothing to be done here either */
                break;
        }                       /* end switch (oct->status) */

        tasklet_kill(&oct_priv->droq_tasklet);
}

/**
 * send_rx_ctrl_cmd - Send Rx control command
 * @lio: per-network private data
 * @start_stop: whether to start or stop
 */
static int send_rx_ctrl_cmd(struct lio *lio, int start_stop)
{
        struct octeon_soft_command *sc;
        union octnet_cmd *ncmd;
        struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
        int retval;

        if (oct->props[lio->ifidx].rx_on == start_stop)
                return 0;

        sc = (struct octeon_soft_command *)
                octeon_alloc_soft_command(oct, OCTNET_CMD_SIZE,
                                          16, 0);
        if (!sc) {
                netif_info(lio, rx_err, lio->netdev,
                           "Failed to allocate octeon_soft_command struct\n");
                return -ENOMEM;
        }

        ncmd = (union octnet_cmd *)sc->virtdptr;

        ncmd->u64 = 0;
        ncmd->s.cmd = OCTNET_CMD_RX_CTL;
        ncmd->s.param1 = start_stop;

        octeon_swap_8B_data((u64 *)ncmd, (OCTNET_CMD_SIZE >> 3));

        sc->iq_no = lio->linfo.txpciq[0].s.q_no;

        octeon_prepare_soft_command(oct, sc, OPCODE_NIC,
                                    OPCODE_NIC_CMD, 0, 0, 0);

        init_completion(&sc->complete);
        sc->sc_status = OCTEON_REQUEST_PENDING;

        retval = octeon_send_soft_command(oct, sc);
        if (retval == IQ_SEND_FAILED) {
                netif_info(lio, rx_err, lio->netdev, "Failed to send RX Control message\n");
                octeon_free_soft_command(oct, sc);
        } else {
                /* Sleep on a wait queue till the cond flag indicates that the
                 * response arrived or timed-out.
                 */
                retval = wait_for_sc_completion_timeout(oct, sc, 0);
                if (retval)
                        return retval;

                oct->props[lio->ifidx].rx_on = start_stop;
                WRITE_ONCE(sc->caller_is_done, true);
        }

        return retval;
}

/**
 * liquidio_destroy_nic_device - Destroy NIC device interface
 * @oct: octeon device
 * @ifidx: which interface to destroy
 *
 * Cleanup associated with each interface for an Octeon device  when NIC
 * module is being unloaded or if initialization fails during load.
 */
static void liquidio_destroy_nic_device(struct octeon_device *oct, int ifidx)
{
        struct net_device *netdev = oct->props[ifidx].netdev;
        struct octeon_device_priv *oct_priv =
                (struct octeon_device_priv *)oct->priv;
        struct napi_struct *napi, *n;
        struct lio *lio;

        if (!netdev) {
                dev_err(&oct->pci_dev->dev, "%s No netdevice ptr for index %d\n",
                        __func__, ifidx);
                return;
        }

        lio = GET_LIO(netdev);

        dev_dbg(&oct->pci_dev->dev, "NIC device cleanup\n");

        if (atomic_read(&lio->ifstate) & LIO_IFSTATE_RUNNING)
                liquidio_stop(netdev);

        if (oct->props[lio->ifidx].napi_enabled == 1) {
                list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
                        napi_disable(napi);

                oct->props[lio->ifidx].napi_enabled = 0;

                if (OCTEON_CN23XX_PF(oct))
                        oct->droq[0]->ops.poll_mode = 0;
        }

        /* Delete NAPI */
        list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
                netif_napi_del(napi);

        tasklet_enable(&oct_priv->droq_tasklet);

        if (atomic_read(&lio->ifstate) & LIO_IFSTATE_REGISTERED)
                unregister_netdev(netdev);

        cleanup_sync_octeon_time_wq(netdev);
        cleanup_link_status_change_wq(netdev);

        cleanup_rx_oom_poll_fn(netdev);

        lio_delete_glists(lio);

        free_netdev(netdev);

        oct->props[ifidx].gmxport = -1;

        oct->props[ifidx].netdev = NULL;
}

/**
 * liquidio_stop_nic_module - Stop complete NIC functionality
 * @oct: octeon device
 */
static int liquidio_stop_nic_module(struct octeon_device *oct)
{
        int i, j;
        struct lio *lio;

        dev_dbg(&oct->pci_dev->dev, "Stopping network interfaces\n");
        if (!oct->ifcount) {
                dev_err(&oct->pci_dev->dev, "Init for Octeon was not completed\n");
                return 1;
        }

        spin_lock_bh(&oct->cmd_resp_wqlock);
        oct->cmd_resp_state = OCT_DRV_OFFLINE;
        spin_unlock_bh(&oct->cmd_resp_wqlock);

        lio_vf_rep_destroy(oct);

        for (i = 0; i < oct->ifcount; i++) {
                lio = GET_LIO(oct->props[i].netdev);
                for (j = 0; j < oct->num_oqs; j++)
                        octeon_unregister_droq_ops(oct,
                                                   lio->linfo.rxpciq[j].s.q_no);
        }

        for (i = 0; i < oct->ifcount; i++)
                liquidio_destroy_nic_device(oct, i);

        if (oct->devlink) {
                devlink_unregister(oct->devlink);
                devlink_free(oct->devlink);
                oct->devlink = NULL;
        }

        dev_dbg(&oct->pci_dev->dev, "Network interfaces stopped\n");
        return 0;
}

/**
 * liquidio_remove - Cleans up resources at unload time
 * @pdev: PCI device structure
 */
static void liquidio_remove(struct pci_dev *pdev)
{
        struct octeon_device *oct_dev = pci_get_drvdata(pdev);

        dev_dbg(&oct_dev->pci_dev->dev, "Stopping device\n");

        if (oct_dev->watchdog_task)
                kthread_stop(oct_dev->watchdog_task);

        if (!oct_dev->octeon_id &&
            oct_dev->fw_info.app_cap_flags & LIQUIDIO_SWITCHDEV_CAP)
                lio_vf_rep_modexit();

        if (oct_dev->app_mode && (oct_dev->app_mode == CVM_DRV_NIC_APP))
                liquidio_stop_nic_module(oct_dev);

        /* Reset the octeon device and cleanup all memory allocated for
         * the octeon device by driver.
         */
        octeon_destroy_resources(oct_dev);

        dev_info(&oct_dev->pci_dev->dev, "Device removed\n");

        /* This octeon device has been removed. Update the global
         * data structure to reflect this. Free the device structure.
         */
        octeon_free_device_mem(oct_dev);
}

/**
 * octeon_chip_specific_setup - Identify the Octeon device and to map the BAR address space
 * @oct: octeon device
 */
static int octeon_chip_specific_setup(struct octeon_device *oct)
{
        u32 dev_id, rev_id;
        int ret = 1;

        pci_read_config_dword(oct->pci_dev, 0, &dev_id);
        pci_read_config_dword(oct->pci_dev, 8, &rev_id);
        oct->rev_id = rev_id & 0xff;

        switch (dev_id) {
        case OCTEON_CN68XX_PCIID:
                oct->chip_id = OCTEON_CN68XX;
                ret = lio_setup_cn68xx_octeon_device(oct);
                break;

        case OCTEON_CN66XX_PCIID:
                oct->chip_id = OCTEON_CN66XX;
                ret = lio_setup_cn66xx_octeon_device(oct);
                break;

        case OCTEON_CN23XX_PCIID_PF:
                oct->chip_id = OCTEON_CN23XX_PF_VID;
                ret = setup_cn23xx_octeon_pf_device(oct);
                if (ret)
                        break;
#ifdef CONFIG_PCI_IOV
                if (!ret)
                        pci_sriov_set_totalvfs(oct->pci_dev,
                                               oct->sriov_info.max_vfs);
#endif
                break;

        default:
                dev_err(&oct->pci_dev->dev, "Unknown device found (dev_id: %x)\n",
                        dev_id);
        }

        return ret;
}

/**
 * octeon_pci_os_setup - PCI initialization for each Octeon device.
 * @oct: octeon device
 */
static int octeon_pci_os_setup(struct octeon_device *oct)
{
        /* setup PCI stuff first */
        if (pci_enable_device(oct->pci_dev)) {
                dev_err(&oct->pci_dev->dev, "pci_enable_device failed\n");
                return 1;
        }

        if (dma_set_mask_and_coherent(&oct->pci_dev->dev, DMA_BIT_MASK(64))) {
                dev_err(&oct->pci_dev->dev, "Unexpected DMA device capability\n");
                pci_disable_device(oct->pci_dev);
                return 1;
        }

        /* Enable PCI DMA Master. */
        pci_set_master(oct->pci_dev);

        return 0;
}

/**
 * free_netbuf - Unmap and free network buffer
 * @buf: buffer
 */
static void free_netbuf(void *buf)
{
        struct sk_buff *skb;
        struct octnet_buf_free_info *finfo;
        struct lio *lio;

        finfo = (struct octnet_buf_free_info *)buf;
        skb = finfo->skb;
        lio = finfo->lio;

        dma_unmap_single(&lio->oct_dev->pci_dev->dev, finfo->dptr, skb->len,
                         DMA_TO_DEVICE);

        tx_buffer_free(skb);
}

/**
 * free_netsgbuf - Unmap and free gather buffer
 * @buf: buffer
 */
static void free_netsgbuf(void *buf)
{
        struct octnet_buf_free_info *finfo;
        struct sk_buff *skb;
        struct lio *lio;
        struct octnic_gather *g;
        int i, frags, iq;

        finfo = (struct octnet_buf_free_info *)buf;
        skb = finfo->skb;
        lio = finfo->lio;
        g = finfo->g;
        frags = skb_shinfo(skb)->nr_frags;

        dma_unmap_single(&lio->oct_dev->pci_dev->dev,
                         g->sg[0].ptr[0], (skb->len - skb->data_len),
                         DMA_TO_DEVICE);

        i = 1;
        while (frags--) {
                skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];

                pci_unmap_page((lio->oct_dev)->pci_dev,
                               g->sg[(i >> 2)].ptr[(i & 3)],
                               skb_frag_size(frag), DMA_TO_DEVICE);
                i++;
        }

        iq = skb_iq(lio->oct_dev, skb);
        spin_lock(&lio->glist_lock[iq]);
        list_add_tail(&g->list, &lio->glist[iq]);
        spin_unlock(&lio->glist_lock[iq]);

        tx_buffer_free(skb);
}

/**
 * free_netsgbuf_with_resp - Unmap and free gather buffer with response
 * @buf: buffer
 */
static void free_netsgbuf_with_resp(void *buf)
{
        struct octeon_soft_command *sc;
        struct octnet_buf_free_info *finfo;
        struct sk_buff *skb;
        struct lio *lio;
        struct octnic_gather *g;
        int i, frags, iq;

        sc = (struct octeon_soft_command *)buf;
        skb = (struct sk_buff *)sc->callback_arg;
        finfo = (struct octnet_buf_free_info *)&skb->cb;

        lio = finfo->lio;
        g = finfo->g;
        frags = skb_shinfo(skb)->nr_frags;

        dma_unmap_single(&lio->oct_dev->pci_dev->dev,
                         g->sg[0].ptr[0], (skb->len - skb->data_len),
                         DMA_TO_DEVICE);

        i = 1;
        while (frags--) {
                skb_frag_t *frag = &skb_shinfo(skb)->frags[i - 1];

                pci_unmap_page((lio->oct_dev)->pci_dev,
                               g->sg[(i >> 2)].ptr[(i & 3)],
                               skb_frag_size(frag), DMA_TO_DEVICE);
                i++;
        }

        iq = skb_iq(lio->oct_dev, skb);

        spin_lock(&lio->glist_lock[iq]);
        list_add_tail(&g->list, &lio->glist[iq]);
        spin_unlock(&lio->glist_lock[iq]);

        /* Don't free the skb yet */
}

/**
 * liquidio_ptp_adjfreq - Adjust ptp frequency
 * @ptp: PTP clock info
 * @ppb: how much to adjust by, in parts-per-billion
 */
static int liquidio_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb)
{
        struct lio *lio = container_of(ptp, struct lio, ptp_info);
        struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;
        u64 comp, delta;
        unsigned long flags;
        bool neg_adj = false;

        if (ppb < 0) {
                neg_adj = true;
                ppb = -ppb;
        }

        /* The hardware adds the clock compensation value to the
         * PTP clock on every coprocessor clock cycle, so we
         * compute the delta in terms of coprocessor clocks.
         */
        delta = (u64)ppb << 32;
        do_div(delta, oct->coproc_clock_rate);

        spin_lock_irqsave(&lio->ptp_lock, flags);
        comp = lio_pci_readq(oct, CN6XXX_MIO_PTP_CLOCK_COMP);
        if (neg_adj)
                comp -= delta;
        else
                comp += delta;
        lio_pci_writeq(oct, comp, CN6XXX_MIO_PTP_CLOCK_COMP);
        spin_unlock_irqrestore(&lio->ptp_lock, flags);

        return 0;
}

/**
 * liquidio_ptp_adjtime - Adjust ptp time
 * @ptp: PTP clock info
 * @delta: how much to adjust by, in nanosecs
 */
static int liquidio_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
{
        unsigned long flags;
        struct lio *lio = container_of(ptp, struct lio, ptp_info);

        spin_lock_irqsave(&lio->ptp_lock, flags);
        lio->ptp_adjust += delta;
        spin_unlock_irqrestore(&lio->ptp_lock, flags);

        return 0;
}

/**
 * liquidio_ptp_gettime - Get hardware clock time, including any adjustment
 * @ptp: PTP clock info
 * @ts: timespec
 */
static int liquidio_ptp_gettime(struct ptp_clock_info *ptp,
                                struct timespec64 *ts)
{
        u64 ns;
        unsigned long flags;
        struct lio *lio = container_of(ptp, struct lio, ptp_info);
        struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;

        spin_lock_irqsave(&lio->ptp_lock, flags);
        ns = lio_pci_readq(oct, CN6XXX_MIO_PTP_CLOCK_HI);
        ns += lio->ptp_adjust;
        spin_unlock_irqrestore(&lio->ptp_lock, flags);

        *ts = ns_to_timespec64(ns);

        return 0;
}

/**
 * liquidio_ptp_settime - Set hardware clock time. Reset adjustment
 * @ptp: PTP clock info
 * @ts: timespec
 */
static int liquidio_ptp_settime(struct ptp_clock_info *ptp,
                                const struct timespec64 *ts)
{
        u64 ns;
        unsigned long flags;
        struct lio *lio = container_of(ptp, struct lio, ptp_info);
        struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;

        ns = timespec64_to_ns(ts);

        spin_lock_irqsave(&lio->ptp_lock, flags);
        lio_pci_writeq(oct, ns, CN6XXX_MIO_PTP_CLOCK_HI);
        lio->ptp_adjust = 0;
        spin_unlock_irqrestore(&lio->ptp_lock, flags);

        return 0;
}

/**
 * liquidio_ptp_enable - Check if PTP is enabled
 * @ptp: PTP clock info
 * @rq: request
 * @on: is it on
 */
static int
liquidio_ptp_enable(struct ptp_clock_info __maybe_unused *ptp,
                    struct ptp_clock_request __maybe_unused *rq,
                    int __maybe_unused on)
{
        return -EOPNOTSUPP;
}

/**
 * oct_ptp_open - Open PTP clock source
 * @netdev: network device
 */
static void oct_ptp_open(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = (struct octeon_device *)lio->oct_dev;

        spin_lock_init(&lio->ptp_lock);

        snprintf(lio->ptp_info.name, 16, "%s", netdev->name);
        lio->ptp_info.owner = THIS_MODULE;
        lio->ptp_info.max_adj = 250000000;
        lio->ptp_info.n_alarm = 0;
        lio->ptp_info.n_ext_ts = 0;
        lio->ptp_info.n_per_out = 0;
        lio->ptp_info.pps = 0;
        lio->ptp_info.adjfreq = liquidio_ptp_adjfreq;
        lio->ptp_info.adjtime = liquidio_ptp_adjtime;
        lio->ptp_info.gettime64 = liquidio_ptp_gettime;
        lio->ptp_info.settime64 = liquidio_ptp_settime;
        lio->ptp_info.enable = liquidio_ptp_enable;

        lio->ptp_adjust = 0;

        lio->ptp_clock = ptp_clock_register(&lio->ptp_info,
                                             &oct->pci_dev->dev);

        if (IS_ERR(lio->ptp_clock))
                lio->ptp_clock = NULL;
}

/**
 * liquidio_ptp_init - Init PTP clock
 * @oct: octeon device
 */
static void liquidio_ptp_init(struct octeon_device *oct)
{
        u64 clock_comp, cfg;

        clock_comp = (u64)NSEC_PER_SEC << 32;
        do_div(clock_comp, oct->coproc_clock_rate);
        lio_pci_writeq(oct, clock_comp, CN6XXX_MIO_PTP_CLOCK_COMP);

        /* Enable */
        cfg = lio_pci_readq(oct, CN6XXX_MIO_PTP_CLOCK_CFG);
        lio_pci_writeq(oct, cfg | 0x01, CN6XXX_MIO_PTP_CLOCK_CFG);
}

/**
 * load_firmware - Load firmware to device
 * @oct: octeon device
 *
 * Maps device to firmware filename, requests firmware, and downloads it
 */
static int load_firmware(struct octeon_device *oct)
{
        int ret = 0;
        const struct firmware *fw;
        char fw_name[LIO_MAX_FW_FILENAME_LEN];
        char *tmp_fw_type;

        if (fw_type_is_auto()) {
                tmp_fw_type = LIO_FW_NAME_TYPE_NIC;
                strncpy(fw_type, tmp_fw_type, sizeof(fw_type));
        } else {
                tmp_fw_type = fw_type;
        }

        sprintf(fw_name, "%s%s%s_%s%s", LIO_FW_DIR, LIO_FW_BASE_NAME,
                octeon_get_conf(oct)->card_name, tmp_fw_type,
                LIO_FW_NAME_SUFFIX);

        ret = request_firmware(&fw, fw_name, &oct->pci_dev->dev);
        if (ret) {
                dev_err(&oct->pci_dev->dev, "Request firmware failed. Could not find file %s.\n",
                        fw_name);
                release_firmware(fw);
                return ret;
        }

        ret = octeon_download_firmware(oct, fw->data, fw->size);

        release_firmware(fw);

        return ret;
}

/**
 * octnet_poll_check_txq_status - Poll routine for checking transmit queue status
 * @work: work_struct data structure
 */
static void octnet_poll_check_txq_status(struct work_struct *work)
{
        struct cavium_wk *wk = (struct cavium_wk *)work;
        struct lio *lio = (struct lio *)wk->ctxptr;

        if (!ifstate_check(lio, LIO_IFSTATE_RUNNING))
                return;

        check_txq_status(lio);
        queue_delayed_work(lio->txq_status_wq.wq,
                           &lio->txq_status_wq.wk.work, msecs_to_jiffies(1));
}

/**
 * setup_tx_poll_fn - Sets up the txq poll check
 * @netdev: network device
 */
static inline int setup_tx_poll_fn(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;

        lio->txq_status_wq.wq = alloc_workqueue("txq-status",
                                                WQ_MEM_RECLAIM, 0);
        if (!lio->txq_status_wq.wq) {
                dev_err(&oct->pci_dev->dev, "unable to create cavium txq status wq\n");
                return -1;
        }
        INIT_DELAYED_WORK(&lio->txq_status_wq.wk.work,
                          octnet_poll_check_txq_status);
        lio->txq_status_wq.wk.ctxptr = lio;
        queue_delayed_work(lio->txq_status_wq.wq,
                           &lio->txq_status_wq.wk.work, msecs_to_jiffies(1));
        return 0;
}

static inline void cleanup_tx_poll_fn(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);

        if (lio->txq_status_wq.wq) {
                cancel_delayed_work_sync(&lio->txq_status_wq.wk.work);
                destroy_workqueue(lio->txq_status_wq.wq);
        }
}

/**
 * liquidio_open - Net device open for LiquidIO
 * @netdev: network device
 */
static int liquidio_open(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;
        struct octeon_device_priv *oct_priv =
                (struct octeon_device_priv *)oct->priv;
        struct napi_struct *napi, *n;
        int ret = 0;

        if (oct->props[lio->ifidx].napi_enabled == 0) {
                tasklet_disable(&oct_priv->droq_tasklet);

                list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
                        napi_enable(napi);

                oct->props[lio->ifidx].napi_enabled = 1;

                if (OCTEON_CN23XX_PF(oct))
                        oct->droq[0]->ops.poll_mode = 1;
        }

        if (oct->ptp_enable)
                oct_ptp_open(netdev);

        ifstate_set(lio, LIO_IFSTATE_RUNNING);

        if (OCTEON_CN23XX_PF(oct)) {
                if (!oct->msix_on)
                        if (setup_tx_poll_fn(netdev))
                                return -1;
        } else {
                if (setup_tx_poll_fn(netdev))
                        return -1;
        }

        netif_tx_start_all_queues(netdev);

        /* Ready for link status updates */
        lio->intf_open = 1;

        netif_info(lio, ifup, lio->netdev, "Interface Open, ready for traffic\n");

        /* tell Octeon to start forwarding packets to host */
        ret = send_rx_ctrl_cmd(lio, 1);
        if (ret)
                return ret;

        /* start periodical statistics fetch */
        INIT_DELAYED_WORK(&lio->stats_wk.work, lio_fetch_stats);
        lio->stats_wk.ctxptr = lio;
        schedule_delayed_work(&lio->stats_wk.work, msecs_to_jiffies
                                        (LIQUIDIO_NDEV_STATS_POLL_TIME_MS));

        dev_info(&oct->pci_dev->dev, "%s interface is opened\n",
                 netdev->name);

        return ret;
}

/**
 * liquidio_stop - Net device stop for LiquidIO
 * @netdev: network device
 */
static int liquidio_stop(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;
        struct octeon_device_priv *oct_priv =
                (struct octeon_device_priv *)oct->priv;
        struct napi_struct *napi, *n;
        int ret = 0;

        ifstate_reset(lio, LIO_IFSTATE_RUNNING);

        /* Stop any link updates */
        lio->intf_open = 0;

        stop_txqs(netdev);

        /* Inform that netif carrier is down */
        netif_carrier_off(netdev);
        netif_tx_disable(netdev);

        lio->linfo.link.s.link_up = 0;
        lio->link_changes++;

        /* Tell Octeon that nic interface is down. */
        ret = send_rx_ctrl_cmd(lio, 0);
        if (ret)
                return ret;

        if (OCTEON_CN23XX_PF(oct)) {
                if (!oct->msix_on)
                        cleanup_tx_poll_fn(netdev);
        } else {
                cleanup_tx_poll_fn(netdev);
        }

        cancel_delayed_work_sync(&lio->stats_wk.work);

        if (lio->ptp_clock) {
                ptp_clock_unregister(lio->ptp_clock);
                lio->ptp_clock = NULL;
        }

        /* Wait for any pending Rx descriptors */
        if (lio_wait_for_clean_oq(oct))
                netif_info(lio, rx_err, lio->netdev,
                           "Proceeding with stop interface after partial RX desc processing\n");

        if (oct->props[lio->ifidx].napi_enabled == 1) {
                list_for_each_entry_safe(napi, n, &netdev->napi_list, dev_list)
                        napi_disable(napi);

                oct->props[lio->ifidx].napi_enabled = 0;

                if (OCTEON_CN23XX_PF(oct))
                        oct->droq[0]->ops.poll_mode = 0;

                tasklet_enable(&oct_priv->droq_tasklet);
        }

        dev_info(&oct->pci_dev->dev, "%s interface is stopped\n", netdev->name);

        return ret;
}

/**
 * get_new_flags - Converts a mask based on net device flags
 * @netdev: network device
 *
 * This routine generates a octnet_ifflags mask from the net device flags
 * received from the OS.
 */
static inline enum octnet_ifflags get_new_flags(struct net_device *netdev)
{
        enum octnet_ifflags f = OCTNET_IFFLAG_UNICAST;

        if (netdev->flags & IFF_PROMISC)
                f |= OCTNET_IFFLAG_PROMISC;

        if (netdev->flags & IFF_ALLMULTI)
                f |= OCTNET_IFFLAG_ALLMULTI;

        if (netdev->flags & IFF_MULTICAST) {
                f |= OCTNET_IFFLAG_MULTICAST;

                /* Accept all multicast addresses if there are more than we
                 * can handle
                 */
                if (netdev_mc_count(netdev) > MAX_OCTEON_MULTICAST_ADDR)
                        f |= OCTNET_IFFLAG_ALLMULTI;
        }

        if (netdev->flags & IFF_BROADCAST)
                f |= OCTNET_IFFLAG_BROADCAST;

        return f;
}

/**
 * liquidio_set_mcast_list - Net device set_multicast_list
 * @netdev: network device
 */
static void liquidio_set_mcast_list(struct net_device *netdev)
{
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;
        struct octnic_ctrl_pkt nctrl;
        struct netdev_hw_addr *ha;
        u64 *mc;
        int ret;
        int mc_count = min(netdev_mc_count(netdev), MAX_OCTEON_MULTICAST_ADDR);

        memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));

        /* Create a ctrl pkt command to be sent to core app. */
        nctrl.ncmd.u64 = 0;
        nctrl.ncmd.s.cmd = OCTNET_CMD_SET_MULTI_LIST;
        nctrl.ncmd.s.param1 = get_new_flags(netdev);
        nctrl.ncmd.s.param2 = mc_count;
        nctrl.ncmd.s.more = mc_count;
        nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
        nctrl.netpndev = (u64)netdev;
        nctrl.cb_fn = liquidio_link_ctrl_cmd_completion;

        /* copy all the addresses into the udd */
        mc = &nctrl.udd[0];
        netdev_for_each_mc_addr(ha, netdev) {
                *mc = 0;
                memcpy(((u8 *)mc) + 2, ha->addr, ETH_ALEN);
                /* no need to swap bytes */

                if (++mc > &nctrl.udd[mc_count])
                        break;
        }

        /* Apparently, any activity in this call from the kernel has to
         * be atomic. So we won't wait for response.
         */

        ret = octnet_send_nic_ctrl_pkt(lio->oct_dev, &nctrl);
        if (ret) {
                dev_err(&oct->pci_dev->dev, "DEVFLAGS change failed in core (ret: 0x%x)\n",
                        ret);
        }
}

/**
 * liquidio_set_mac - Net device set_mac_address
 * @netdev: network device
 * @p: pointer to sockaddr
 */
static int liquidio_set_mac(struct net_device *netdev, void *p)
{
        int ret = 0;
        struct lio *lio = GET_LIO(netdev);
        struct octeon_device *oct = lio->oct_dev;
        struct sockaddr *addr = (struct sockaddr *)p;
        struct octnic_ctrl_pkt nctrl;

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

        memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));