/************************************************************************
 * s2io.c: A Linux PCI-X Ethernet driver for Neterion 10GbE Server NIC
 * Copyright(c) 2002-2010 Exar Corp.
 *
 * This software may be used and distributed according to the terms of
 * the GNU General Public License (GPL), incorporated herein by reference.
 * Drivers based on or derived from this code fall under the GPL and must
 * retain the authorship, copyright and license notice.  This file is not
 * a complete program and may only be used when the entire operating
 * system is licensed under the GPL.
 * See the file COPYING in this distribution for more information.
 *
 * Credits:
 * Jeff Garzik          : For pointing out the improper error condition
 *                        check in the s2io_xmit routine and also some
 *                        issues in the Tx watch dog function. Also for
 *                        patiently answering all those innumerable
 *                        questions regaring the 2.6 porting issues.
 * Stephen Hemminger    : Providing proper 2.6 porting mechanism for some
 *                        macros available only in 2.6 Kernel.
 * Francois Romieu      : For pointing out all code part that were
 *                        deprecated and also styling related comments.
 * Grant Grundler       : For helping me get rid of some Architecture
 *                        dependent code.
 * Christopher Hellwig  : Some more 2.6 specific issues in the driver.
 *
 * The module loadable parameters that are supported by the driver and a brief
 * explanation of all the variables.
 *
 * rx_ring_num : This can be used to program the number of receive rings used
 * in the driver.
 * rx_ring_sz: This defines the number of receive blocks each ring can have.
 *     This is also an array of size 8.
 * rx_ring_mode: This defines the operation mode of all 8 rings. The valid
 *              values are 1, 2.
 * tx_fifo_num: This defines the number of Tx FIFOs thats used int the driver.
 * tx_fifo_len: This too is an array of 8. Each element defines the number of
 * Tx descriptors that can be associated with each corresponding FIFO.
 * intr_type: This defines the type of interrupt. The values can be 0(INTA),
 *     2(MSI_X). Default value is '2(MSI_X)'
 * lro_max_pkts: This parameter defines maximum number of packets can be
 *     aggregated as a single large packet
 * napi: This parameter used to enable/disable NAPI (polling Rx)
 *     Possible values '1' for enable and '0' for disable. Default is '1'
 * ufo: This parameter used to enable/disable UDP Fragmentation Offload(UFO)
 *      Possible values '1' for enable and '0' for disable. Default is '0'
 * vlan_tag_strip: This can be used to enable or disable vlan stripping.
 *                 Possible values '1' for enable , '0' for disable.
 *                 Default is '2' - which means disable in promisc mode
 *                 and enable in non-promiscuous mode.
 * multiq: This parameter used to enable/disable MULTIQUEUE support.
 *      Possible values '1' for enable and '0' for disable. Default is '0'
 ************************************************************************/

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/module.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/ioport.h>
#include <linux/pci.h>
#include <linux/dma-mapping.h>
#include <linux/kernel.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/mdio.h>
#include <linux/skbuff.h>
#include <linux/init.h>
#include <linux/delay.h>
#include <linux/stddef.h>
#include <linux/ioctl.h>
#include <linux/timex.h>
#include <linux/ethtool.h>
#include <linux/workqueue.h>
#include <linux/if_vlan.h>
#include <linux/ip.h>
#include <linux/tcp.h>
#include <linux/uaccess.h>
#include <linux/io.h>
#include <linux/slab.h>
#include <net/tcp.h>

#include <asm/system.h>
#include <asm/div64.h>
#include <asm/irq.h>

/* local include */
#include "s2io.h"
#include "s2io-regs.h"

#define DRV_VERSION "2.0.26.28"

/* S2io Driver name & version. */
static const char s2io_driver_name[] = "Neterion";
static const char s2io_driver_version[] = DRV_VERSION;

static const int rxd_size[2] = {32, 48};
static const int rxd_count[2] = {127, 85};

static inline int RXD_IS_UP2DT(struct RxD_t *rxdp)
{
        int ret;

        ret = ((!(rxdp->Control_1 & RXD_OWN_XENA)) &&
               (GET_RXD_MARKER(rxdp->Control_2) != THE_RXD_MARK));

        return ret;
}

/*
 * Cards with following subsystem_id have a link state indication
 * problem, 600B, 600C, 600D, 640B, 640C and 640D.
 * macro below identifies these cards given the subsystem_id.
 */
#define CARDS_WITH_FAULTY_LINK_INDICATORS(dev_type, subid)              \
        (dev_type == XFRAME_I_DEVICE) ?                                 \
        ((((subid >= 0x600B) && (subid <= 0x600D)) ||                   \
          ((subid >= 0x640B) && (subid <= 0x640D))) ? 1 : 0) : 0

#define LINK_IS_UP(val64) (!(val64 & (ADAPTER_STATUS_RMAC_REMOTE_FAULT | \
                                      ADAPTER_STATUS_RMAC_LOCAL_FAULT)))

static inline int is_s2io_card_up(const struct s2io_nic *sp)
{
        return test_bit(__S2IO_STATE_CARD_UP, &sp->state);
}

/* Ethtool related variables and Macros. */
static const char s2io_gstrings[][ETH_GSTRING_LEN] = {
        "Register test\t(offline)",
        "Eeprom test\t(offline)",
        "Link test\t(online)",
        "RLDRAM test\t(offline)",
        "BIST Test\t(offline)"
};

static const char ethtool_xena_stats_keys[][ETH_GSTRING_LEN] = {
        {"tmac_frms"},
        {"tmac_data_octets"},
        {"tmac_drop_frms"},
        {"tmac_mcst_frms"},
        {"tmac_bcst_frms"},
        {"tmac_pause_ctrl_frms"},
        {"tmac_ttl_octets"},
        {"tmac_ucst_frms"},
        {"tmac_nucst_frms"},
        {"tmac_any_err_frms"},
        {"tmac_ttl_less_fb_octets"},
        {"tmac_vld_ip_octets"},
        {"tmac_vld_ip"},
        {"tmac_drop_ip"},
        {"tmac_icmp"},
        {"tmac_rst_tcp"},
        {"tmac_tcp"},
        {"tmac_udp"},
        {"rmac_vld_frms"},
        {"rmac_data_octets"},
        {"rmac_fcs_err_frms"},
        {"rmac_drop_frms"},
        {"rmac_vld_mcst_frms"},
        {"rmac_vld_bcst_frms"},
        {"rmac_in_rng_len_err_frms"},
        {"rmac_out_rng_len_err_frms"},
        {"rmac_long_frms"},
        {"rmac_pause_ctrl_frms"},
        {"rmac_unsup_ctrl_frms"},
        {"rmac_ttl_octets"},
        {"rmac_accepted_ucst_frms"},
        {"rmac_accepted_nucst_frms"},
        {"rmac_discarded_frms"},
        {"rmac_drop_events"},
        {"rmac_ttl_less_fb_octets"},
        {"rmac_ttl_frms"},
        {"rmac_usized_frms"},
        {"rmac_osized_frms"},
        {"rmac_frag_frms"},
        {"rmac_jabber_frms"},
        {"rmac_ttl_64_frms"},
        {"rmac_ttl_65_127_frms"},
        {"rmac_ttl_128_255_frms"},
        {"rmac_ttl_256_511_frms"},
        {"rmac_ttl_512_1023_frms"},
        {"rmac_ttl_1024_1518_frms"},
        {"rmac_ip"},
        {"rmac_ip_octets"},
        {"rmac_hdr_err_ip"},
        {"rmac_drop_ip"},
        {"rmac_icmp"},
        {"rmac_tcp"},
        {"rmac_udp"},
        {"rmac_err_drp_udp"},
        {"rmac_xgmii_err_sym"},
        {"rmac_frms_q0"},
        {"rmac_frms_q1"},
        {"rmac_frms_q2"},
        {"rmac_frms_q3"},
        {"rmac_frms_q4"},
        {"rmac_frms_q5"},
        {"rmac_frms_q6"},
        {"rmac_frms_q7"},
        {"rmac_full_q0"},
        {"rmac_full_q1"},
        {"rmac_full_q2"},
        {"rmac_full_q3"},
        {"rmac_full_q4"},
        {"rmac_full_q5"},
        {"rmac_full_q6"},
        {"rmac_full_q7"},
        {"rmac_pause_cnt"},
        {"rmac_xgmii_data_err_cnt"},
        {"rmac_xgmii_ctrl_err_cnt"},
        {"rmac_accepted_ip"},
        {"rmac_err_tcp"},
        {"rd_req_cnt"},
        {"new_rd_req_cnt"},
        {"new_rd_req_rtry_cnt"},
        {"rd_rtry_cnt"},
        {"wr_rtry_rd_ack_cnt"},
        {"wr_req_cnt"},
        {"new_wr_req_cnt"},
        {"new_wr_req_rtry_cnt"},
        {"wr_rtry_cnt"},
        {"wr_disc_cnt"},
        {"rd_rtry_wr_ack_cnt"},
        {"txp_wr_cnt"},
        {"txd_rd_cnt"},
        {"txd_wr_cnt"},
        {"rxd_rd_cnt"},
        {"rxd_wr_cnt"},
        {"txf_rd_cnt"},
        {"rxf_wr_cnt"}
};

static const char ethtool_enhanced_stats_keys[][ETH_GSTRING_LEN] = {
        {"rmac_ttl_1519_4095_frms"},
        {"rmac_ttl_4096_8191_frms"},
        {"rmac_ttl_8192_max_frms"},
        {"rmac_ttl_gt_max_frms"},
        {"rmac_osized_alt_frms"},
        {"rmac_jabber_alt_frms"},
        {"rmac_gt_max_alt_frms"},
        {"rmac_vlan_frms"},
        {"rmac_len_discard"},
        {"rmac_fcs_discard"},
        {"rmac_pf_discard"},
        {"rmac_da_discard"},
        {"rmac_red_discard"},
        {"rmac_rts_discard"},
        {"rmac_ingm_full_discard"},
        {"link_fault_cnt"}
};

static const char ethtool_driver_stats_keys[][ETH_GSTRING_LEN] = {
        {"\n DRIVER STATISTICS"},
        {"single_bit_ecc_errs"},
        {"double_bit_ecc_errs"},
        {"parity_err_cnt"},
        {"serious_err_cnt"},
        {"soft_reset_cnt"},
        {"fifo_full_cnt"},
        {"ring_0_full_cnt"},
        {"ring_1_full_cnt"},
        {"ring_2_full_cnt"},
        {"ring_3_full_cnt"},
        {"ring_4_full_cnt"},
        {"ring_5_full_cnt"},
        {"ring_6_full_cnt"},
        {"ring_7_full_cnt"},
        {"alarm_transceiver_temp_high"},
        {"alarm_transceiver_temp_low"},
        {"alarm_laser_bias_current_high"},
        {"alarm_laser_bias_current_low"},
        {"alarm_laser_output_power_high"},
        {"alarm_laser_output_power_low"},
        {"warn_transceiver_temp_high"},
        {"warn_transceiver_temp_low"},
        {"warn_laser_bias_current_high"},
        {"warn_laser_bias_current_low"},
        {"warn_laser_output_power_high"},
        {"warn_laser_output_power_low"},
        {"lro_aggregated_pkts"},
        {"lro_flush_both_count"},
        {"lro_out_of_sequence_pkts"},
        {"lro_flush_due_to_max_pkts"},
        {"lro_avg_aggr_pkts"},
        {"mem_alloc_fail_cnt"},
        {"pci_map_fail_cnt"},
        {"watchdog_timer_cnt"},
        {"mem_allocated"},
        {"mem_freed"},
        {"link_up_cnt"},
        {"link_down_cnt"},
        {"link_up_time"},
        {"link_down_time"},
        {"tx_tcode_buf_abort_cnt"},
        {"tx_tcode_desc_abort_cnt"},
        {"tx_tcode_parity_err_cnt"},
        {"tx_tcode_link_loss_cnt"},
        {"tx_tcode_list_proc_err_cnt"},
        {"rx_tcode_parity_err_cnt"},
        {"rx_tcode_abort_cnt"},
        {"rx_tcode_parity_abort_cnt"},
        {"rx_tcode_rda_fail_cnt"},
        {"rx_tcode_unkn_prot_cnt"},
        {"rx_tcode_fcs_err_cnt"},
        {"rx_tcode_buf_size_err_cnt"},
        {"rx_tcode_rxd_corrupt_cnt"},
        {"rx_tcode_unkn_err_cnt"},
        {"tda_err_cnt"},
        {"pfc_err_cnt"},
        {"pcc_err_cnt"},
        {"tti_err_cnt"},
        {"tpa_err_cnt"},
        {"sm_err_cnt"},
        {"lso_err_cnt"},
        {"mac_tmac_err_cnt"},
        {"mac_rmac_err_cnt"},
        {"xgxs_txgxs_err_cnt"},
        {"xgxs_rxgxs_err_cnt"},
        {"rc_err_cnt"},
        {"prc_pcix_err_cnt"},
        {"rpa_err_cnt"},
        {"rda_err_cnt"},
        {"rti_err_cnt"},
        {"mc_err_cnt"}
};

#define S2IO_XENA_STAT_LEN      ARRAY_SIZE(ethtool_xena_stats_keys)
#define S2IO_ENHANCED_STAT_LEN  ARRAY_SIZE(ethtool_enhanced_stats_keys)
#define S2IO_DRIVER_STAT_LEN    ARRAY_SIZE(ethtool_driver_stats_keys)

#define XFRAME_I_STAT_LEN (S2IO_XENA_STAT_LEN + S2IO_DRIVER_STAT_LEN)
#define XFRAME_II_STAT_LEN (XFRAME_I_STAT_LEN + S2IO_ENHANCED_STAT_LEN)

#define XFRAME_I_STAT_STRINGS_LEN (XFRAME_I_STAT_LEN * ETH_GSTRING_LEN)
#define XFRAME_II_STAT_STRINGS_LEN (XFRAME_II_STAT_LEN * ETH_GSTRING_LEN)

#define S2IO_TEST_LEN   ARRAY_SIZE(s2io_gstrings)
#define S2IO_STRINGS_LEN        (S2IO_TEST_LEN * ETH_GSTRING_LEN)

#define S2IO_TIMER_CONF(timer, handle, arg, exp)        \
        init_timer(&timer);                             \
        timer.function = handle;                        \
        timer.data = (unsigned long)arg;                \
        mod_timer(&timer, (jiffies + exp))              \

/* copy mac addr to def_mac_addr array */
static void do_s2io_copy_mac_addr(struct s2io_nic *sp, int offset, u64 mac_addr)
{
        sp->def_mac_addr[offset].mac_addr[5] = (u8) (mac_addr);
        sp->def_mac_addr[offset].mac_addr[4] = (u8) (mac_addr >> 8);
        sp->def_mac_addr[offset].mac_addr[3] = (u8) (mac_addr >> 16);
        sp->def_mac_addr[offset].mac_addr[2] = (u8) (mac_addr >> 24);
        sp->def_mac_addr[offset].mac_addr[1] = (u8) (mac_addr >> 32);
        sp->def_mac_addr[offset].mac_addr[0] = (u8) (mac_addr >> 40);
}

/* Add the vlan */
static void s2io_vlan_rx_register(struct net_device *dev,
                                  struct vlan_group *grp)
{
        int i;
        struct s2io_nic *nic = netdev_priv(dev);
        unsigned long flags[MAX_TX_FIFOS];
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;

        for (i = 0; i < config->tx_fifo_num; i++) {
                struct fifo_info *fifo = &mac_control->fifos[i];

                spin_lock_irqsave(&fifo->tx_lock, flags[i]);
        }

        nic->vlgrp = grp;

        for (i = config->tx_fifo_num - 1; i >= 0; i--) {
                struct fifo_info *fifo = &mac_control->fifos[i];

                spin_unlock_irqrestore(&fifo->tx_lock, flags[i]);
        }
}

/* Unregister the vlan */
static void s2io_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
{
        int i;
        struct s2io_nic *nic = netdev_priv(dev);
        unsigned long flags[MAX_TX_FIFOS];
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;

        for (i = 0; i < config->tx_fifo_num; i++) {
                struct fifo_info *fifo = &mac_control->fifos[i];

                spin_lock_irqsave(&fifo->tx_lock, flags[i]);
        }

        if (nic->vlgrp)
                vlan_group_set_device(nic->vlgrp, vid, NULL);

        for (i = config->tx_fifo_num - 1; i >= 0; i--) {
                struct fifo_info *fifo = &mac_control->fifos[i];

                spin_unlock_irqrestore(&fifo->tx_lock, flags[i]);
        }
}

/*
 * Constants to be programmed into the Xena's registers, to configure
 * the XAUI.
 */

#define END_SIGN        0x0
static const u64 herc_act_dtx_cfg[] = {
        /* Set address */
        0x8000051536750000ULL, 0x80000515367500E0ULL,
        /* Write data */
        0x8000051536750004ULL, 0x80000515367500E4ULL,
        /* Set address */
        0x80010515003F0000ULL, 0x80010515003F00E0ULL,
        /* Write data */
        0x80010515003F0004ULL, 0x80010515003F00E4ULL,
        /* Set address */
        0x801205150D440000ULL, 0x801205150D4400E0ULL,
        /* Write data */
        0x801205150D440004ULL, 0x801205150D4400E4ULL,
        /* Set address */
        0x80020515F2100000ULL, 0x80020515F21000E0ULL,
        /* Write data */
        0x80020515F2100004ULL, 0x80020515F21000E4ULL,
        /* Done */
        END_SIGN
};

static const u64 xena_dtx_cfg[] = {
        /* Set address */
        0x8000051500000000ULL, 0x80000515000000E0ULL,
        /* Write data */
        0x80000515D9350004ULL, 0x80000515D93500E4ULL,
        /* Set address */
        0x8001051500000000ULL, 0x80010515000000E0ULL,
        /* Write data */
        0x80010515001E0004ULL, 0x80010515001E00E4ULL,
        /* Set address */
        0x8002051500000000ULL, 0x80020515000000E0ULL,
        /* Write data */
        0x80020515F2100004ULL, 0x80020515F21000E4ULL,
        END_SIGN
};

/*
 * Constants for Fixing the MacAddress problem seen mostly on
 * Alpha machines.
 */
static const u64 fix_mac[] = {
        0x0060000000000000ULL, 0x0060600000000000ULL,
        0x0040600000000000ULL, 0x0000600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0060600000000000ULL,
        0x0020600000000000ULL, 0x0000600000000000ULL,
        0x0040600000000000ULL, 0x0060600000000000ULL,
        END_SIGN
};

MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_VERSION);


/* Module Loadable parameters. */
S2IO_PARM_INT(tx_fifo_num, FIFO_DEFAULT_NUM);
S2IO_PARM_INT(rx_ring_num, 1);
S2IO_PARM_INT(multiq, 0);
S2IO_PARM_INT(rx_ring_mode, 1);
S2IO_PARM_INT(use_continuous_tx_intrs, 1);
S2IO_PARM_INT(rmac_pause_time, 0x100);
S2IO_PARM_INT(mc_pause_threshold_q0q3, 187);
S2IO_PARM_INT(mc_pause_threshold_q4q7, 187);
S2IO_PARM_INT(shared_splits, 0);
S2IO_PARM_INT(tmac_util_period, 5);
S2IO_PARM_INT(rmac_util_period, 5);
S2IO_PARM_INT(l3l4hdr_size, 128);
/* 0 is no steering, 1 is Priority steering, 2 is Default steering */
S2IO_PARM_INT(tx_steering_type, TX_DEFAULT_STEERING);
/* Frequency of Rx desc syncs expressed as power of 2 */
S2IO_PARM_INT(rxsync_frequency, 3);
/* Interrupt type. Values can be 0(INTA), 2(MSI_X) */
S2IO_PARM_INT(intr_type, 2);
/* Large receive offload feature */

/* Max pkts to be aggregated by LRO at one time. If not specified,
 * aggregation happens until we hit max IP pkt size(64K)
 */
S2IO_PARM_INT(lro_max_pkts, 0xFFFF);
S2IO_PARM_INT(indicate_max_pkts, 0);

S2IO_PARM_INT(napi, 1);
S2IO_PARM_INT(ufo, 0);
S2IO_PARM_INT(vlan_tag_strip, NO_STRIP_IN_PROMISC);

static unsigned int tx_fifo_len[MAX_TX_FIFOS] =
{DEFAULT_FIFO_0_LEN, [1 ...(MAX_TX_FIFOS - 1)] = DEFAULT_FIFO_1_7_LEN};
static unsigned int rx_ring_sz[MAX_RX_RINGS] =
{[0 ...(MAX_RX_RINGS - 1)] = SMALL_BLK_CNT};
static unsigned int rts_frm_len[MAX_RX_RINGS] =
{[0 ...(MAX_RX_RINGS - 1)] = 0 };

module_param_array(tx_fifo_len, uint, NULL, 0);
module_param_array(rx_ring_sz, uint, NULL, 0);
module_param_array(rts_frm_len, uint, NULL, 0);

/*
 * S2IO device table.
 * This table lists all the devices that this driver supports.
 */
static DEFINE_PCI_DEVICE_TABLE(s2io_tbl) = {
        {PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_S2IO_WIN,
         PCI_ANY_ID, PCI_ANY_ID},
        {PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_S2IO_UNI,
         PCI_ANY_ID, PCI_ANY_ID},
        {PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_HERC_WIN,
         PCI_ANY_ID, PCI_ANY_ID},
        {PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_HERC_UNI,
         PCI_ANY_ID, PCI_ANY_ID},
        {0,}
};

MODULE_DEVICE_TABLE(pci, s2io_tbl);

static struct pci_error_handlers s2io_err_handler = {
        .error_detected = s2io_io_error_detected,
        .slot_reset = s2io_io_slot_reset,
        .resume = s2io_io_resume,
};

static struct pci_driver s2io_driver = {
        .name = "S2IO",
        .id_table = s2io_tbl,
        .probe = s2io_init_nic,
        .remove = __devexit_p(s2io_rem_nic),
        .err_handler = &s2io_err_handler,
};

/* A simplifier macro used both by init and free shared_mem Fns(). */
#define TXD_MEM_PAGE_CNT(len, per_each) ((len+per_each - 1) / per_each)

/* netqueue manipulation helper functions */
static inline void s2io_stop_all_tx_queue(struct s2io_nic *sp)
{
        if (!sp->config.multiq) {
                int i;

                for (i = 0; i < sp->config.tx_fifo_num; i++)
                        sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_STOP;
        }
        netif_tx_stop_all_queues(sp->dev);
}

static inline void s2io_stop_tx_queue(struct s2io_nic *sp, int fifo_no)
{
        if (!sp->config.multiq)
                sp->mac_control.fifos[fifo_no].queue_state =
                        FIFO_QUEUE_STOP;

        netif_tx_stop_all_queues(sp->dev);
}

static inline void s2io_start_all_tx_queue(struct s2io_nic *sp)
{
        if (!sp->config.multiq) {
                int i;

                for (i = 0; i < sp->config.tx_fifo_num; i++)
                        sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
        }
        netif_tx_start_all_queues(sp->dev);
}

static inline void s2io_start_tx_queue(struct s2io_nic *sp, int fifo_no)
{
        if (!sp->config.multiq)
                sp->mac_control.fifos[fifo_no].queue_state =
                        FIFO_QUEUE_START;

        netif_tx_start_all_queues(sp->dev);
}

static inline void s2io_wake_all_tx_queue(struct s2io_nic *sp)
{
        if (!sp->config.multiq) {
                int i;

                for (i = 0; i < sp->config.tx_fifo_num; i++)
                        sp->mac_control.fifos[i].queue_state = FIFO_QUEUE_START;
        }
        netif_tx_wake_all_queues(sp->dev);
}

static inline void s2io_wake_tx_queue(
        struct fifo_info *fifo, int cnt, u8 multiq)
{

        if (multiq) {
                if (cnt && __netif_subqueue_stopped(fifo->dev, fifo->fifo_no))
                        netif_wake_subqueue(fifo->dev, fifo->fifo_no);
        } else if (cnt && (fifo->queue_state == FIFO_QUEUE_STOP)) {
                if (netif_queue_stopped(fifo->dev)) {
                        fifo->queue_state = FIFO_QUEUE_START;
                        netif_wake_queue(fifo->dev);
                }
        }
}

/**
 * init_shared_mem - Allocation and Initialization of Memory
 * @nic: Device private variable.
 * Description: The function allocates all the memory areas shared
 * between the NIC and the driver. This includes Tx descriptors,
 * Rx descriptors and the statistics block.
 */

static int init_shared_mem(struct s2io_nic *nic)
{
        u32 size;
        void *tmp_v_addr, *tmp_v_addr_next;
        dma_addr_t tmp_p_addr, tmp_p_addr_next;
        struct RxD_block *pre_rxd_blk = NULL;
        int i, j, blk_cnt;
        int lst_size, lst_per_page;
        struct net_device *dev = nic->dev;
        unsigned long tmp;
        struct buffAdd *ba;
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;
        unsigned long long mem_allocated = 0;

        /* Allocation and initialization of TXDLs in FIFOs */
        size = 0;
        for (i = 0; i < config->tx_fifo_num; i++) {
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                size += tx_cfg->fifo_len;
        }
        if (size > MAX_AVAILABLE_TXDS) {
                DBG_PRINT(ERR_DBG,
                          "Too many TxDs requested: %d, max supported: %d\n",
                          size, MAX_AVAILABLE_TXDS);
                return -EINVAL;
        }

        size = 0;
        for (i = 0; i < config->tx_fifo_num; i++) {
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                size = tx_cfg->fifo_len;
                /*
                 * Legal values are from 2 to 8192
                 */
                if (size < 2) {
                        DBG_PRINT(ERR_DBG, "Fifo %d: Invalid length (%d) - "
                                  "Valid lengths are 2 through 8192\n",
                                  i, size);
                        return -EINVAL;
                }
        }

        lst_size = (sizeof(struct TxD) * config->max_txds);
        lst_per_page = PAGE_SIZE / lst_size;

        for (i = 0; i < config->tx_fifo_num; i++) {
                struct fifo_info *fifo = &mac_control->fifos[i];
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];
                int fifo_len = tx_cfg->fifo_len;
                int list_holder_size = fifo_len * sizeof(struct list_info_hold);

                fifo->list_info = kzalloc(list_holder_size, GFP_KERNEL);
                if (!fifo->list_info) {
                        DBG_PRINT(INFO_DBG, "Malloc failed for list_info\n");
                        return -ENOMEM;
                }
                mem_allocated += list_holder_size;
        }
        for (i = 0; i < config->tx_fifo_num; i++) {
                int page_num = TXD_MEM_PAGE_CNT(config->tx_cfg[i].fifo_len,
                                                lst_per_page);
                struct fifo_info *fifo = &mac_control->fifos[i];
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                fifo->tx_curr_put_info.offset = 0;
                fifo->tx_curr_put_info.fifo_len = tx_cfg->fifo_len - 1;
                fifo->tx_curr_get_info.offset = 0;
                fifo->tx_curr_get_info.fifo_len = tx_cfg->fifo_len - 1;
                fifo->fifo_no = i;
                fifo->nic = nic;
                fifo->max_txds = MAX_SKB_FRAGS + 2;
                fifo->dev = dev;

                for (j = 0; j < page_num; j++) {
                        int k = 0;
                        dma_addr_t tmp_p;
                        void *tmp_v;
                        tmp_v = pci_alloc_consistent(nic->pdev,
                                                     PAGE_SIZE, &tmp_p);
                        if (!tmp_v) {
                                DBG_PRINT(INFO_DBG,
                                          "pci_alloc_consistent failed for TxDL\n");
                                return -ENOMEM;
                        }
                        /* If we got a zero DMA address(can happen on
                         * certain platforms like PPC), reallocate.
                         * Store virtual address of page we don't want,
                         * to be freed later.
                         */
                        if (!tmp_p) {
                                mac_control->zerodma_virt_addr = tmp_v;
                                DBG_PRINT(INIT_DBG,
                                          "%s: Zero DMA address for TxDL. "
                                          "Virtual address %p\n",
                                          dev->name, tmp_v);
                                tmp_v = pci_alloc_consistent(nic->pdev,
                                                             PAGE_SIZE, &tmp_p);
                                if (!tmp_v) {
                                        DBG_PRINT(INFO_DBG,
                                                  "pci_alloc_consistent failed for TxDL\n");
                                        return -ENOMEM;
                                }
                                mem_allocated += PAGE_SIZE;
                        }
                        while (k < lst_per_page) {
                                int l = (j * lst_per_page) + k;
                                if (l == tx_cfg->fifo_len)
                                        break;
                                fifo->list_info[l].list_virt_addr =
                                        tmp_v + (k * lst_size);
                                fifo->list_info[l].list_phy_addr =
                                        tmp_p + (k * lst_size);
                                k++;
                        }
                }
        }

        for (i = 0; i < config->tx_fifo_num; i++) {
                struct fifo_info *fifo = &mac_control->fifos[i];
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                size = tx_cfg->fifo_len;
                fifo->ufo_in_band_v = kcalloc(size, sizeof(u64), GFP_KERNEL);
                if (!fifo->ufo_in_band_v)
                        return -ENOMEM;
                mem_allocated += (size * sizeof(u64));
        }

        /* Allocation and initialization of RXDs in Rings */
        size = 0;
        for (i = 0; i < config->rx_ring_num; i++) {
                struct rx_ring_config *rx_cfg = &config->rx_cfg[i];
                struct ring_info *ring = &mac_control->rings[i];

                if (rx_cfg->num_rxd % (rxd_count[nic->rxd_mode] + 1)) {
                        DBG_PRINT(ERR_DBG, "%s: Ring%d RxD count is not a "
                                  "multiple of RxDs per Block\n",
                                  dev->name, i);
                        return FAILURE;
                }
                size += rx_cfg->num_rxd;
                ring->block_count = rx_cfg->num_rxd /
                        (rxd_count[nic->rxd_mode] + 1);
                ring->pkt_cnt = rx_cfg->num_rxd - ring->block_count;
        }
        if (nic->rxd_mode == RXD_MODE_1)
                size = (size * (sizeof(struct RxD1)));
        else
                size = (size * (sizeof(struct RxD3)));

        for (i = 0; i < config->rx_ring_num; i++) {
                struct rx_ring_config *rx_cfg = &config->rx_cfg[i];
                struct ring_info *ring = &mac_control->rings[i];

                ring->rx_curr_get_info.block_index = 0;
                ring->rx_curr_get_info.offset = 0;
                ring->rx_curr_get_info.ring_len = rx_cfg->num_rxd - 1;
                ring->rx_curr_put_info.block_index = 0;
                ring->rx_curr_put_info.offset = 0;
                ring->rx_curr_put_info.ring_len = rx_cfg->num_rxd - 1;
                ring->nic = nic;
                ring->ring_no = i;

                blk_cnt = rx_cfg->num_rxd / (rxd_count[nic->rxd_mode] + 1);
                /*  Allocating all the Rx blocks */
                for (j = 0; j < blk_cnt; j++) {
                        struct rx_block_info *rx_blocks;
                        int l;

                        rx_blocks = &ring->rx_blocks[j];
                        size = SIZE_OF_BLOCK;   /* size is always page size */
                        tmp_v_addr = pci_alloc_consistent(nic->pdev, size,
                                                          &tmp_p_addr);
                        if (tmp_v_addr == NULL) {
                                /*
                                 * In case of failure, free_shared_mem()
                                 * is called, which should free any
                                 * memory that was alloced till the
                                 * failure happened.
                                 */
                                rx_blocks->block_virt_addr = tmp_v_addr;
                                return -ENOMEM;
                        }
                        mem_allocated += size;
                        memset(tmp_v_addr, 0, size);

                        size = sizeof(struct rxd_info) *
                                rxd_count[nic->rxd_mode];
                        rx_blocks->block_virt_addr = tmp_v_addr;
                        rx_blocks->block_dma_addr = tmp_p_addr;
                        rx_blocks->rxds = kmalloc(size,  GFP_KERNEL);
                        if (!rx_blocks->rxds)
                                return -ENOMEM;
                        mem_allocated += size;
                        for (l = 0; l < rxd_count[nic->rxd_mode]; l++) {
                                rx_blocks->rxds[l].virt_addr =
                                        rx_blocks->block_virt_addr +
                                        (rxd_size[nic->rxd_mode] * l);
                                rx_blocks->rxds[l].dma_addr =
                                        rx_blocks->block_dma_addr +
                                        (rxd_size[nic->rxd_mode] * l);
                        }
                }
                /* Interlinking all Rx Blocks */
                for (j = 0; j < blk_cnt; j++) {
                        int next = (j + 1) % blk_cnt;
                        tmp_v_addr = ring->rx_blocks[j].block_virt_addr;
                        tmp_v_addr_next = ring->rx_blocks[next].block_virt_addr;
                        tmp_p_addr = ring->rx_blocks[j].block_dma_addr;
                        tmp_p_addr_next = ring->rx_blocks[next].block_dma_addr;

                        pre_rxd_blk = (struct RxD_block *)tmp_v_addr;
                        pre_rxd_blk->reserved_2_pNext_RxD_block =
                                (unsigned long)tmp_v_addr_next;
                        pre_rxd_blk->pNext_RxD_Blk_physical =
                                (u64)tmp_p_addr_next;
                }
        }
        if (nic->rxd_mode == RXD_MODE_3B) {
                /*
                 * Allocation of Storages for buffer addresses in 2BUFF mode
                 * and the buffers as well.
                 */
                for (i = 0; i < config->rx_ring_num; i++) {
                        struct rx_ring_config *rx_cfg = &config->rx_cfg[i];
                        struct ring_info *ring = &mac_control->rings[i];

                        blk_cnt = rx_cfg->num_rxd /
                                (rxd_count[nic->rxd_mode] + 1);
                        size = sizeof(struct buffAdd *) * blk_cnt;
                        ring->ba = kmalloc(size, GFP_KERNEL);
                        if (!ring->ba)
                                return -ENOMEM;
                        mem_allocated += size;
                        for (j = 0; j < blk_cnt; j++) {
                                int k = 0;

                                size = sizeof(struct buffAdd) *
                                        (rxd_count[nic->rxd_mode] + 1);
                                ring->ba[j] = kmalloc(size, GFP_KERNEL);
                                if (!ring->ba[j])
                                        return -ENOMEM;
                                mem_allocated += size;
                                while (k != rxd_count[nic->rxd_mode]) {
                                        ba = &ring->ba[j][k];
                                        size = BUF0_LEN + ALIGN_SIZE;
                                        ba->ba_0_org = kmalloc(size, GFP_KERNEL);
                                        if (!ba->ba_0_org)
                                                return -ENOMEM;
                                        mem_allocated += size;
                                        tmp = (unsigned long)ba->ba_0_org;
                                        tmp += ALIGN_SIZE;
                                        tmp &= ~((unsigned long)ALIGN_SIZE);
                                        ba->ba_0 = (void *)tmp;

                                        size = BUF1_LEN + ALIGN_SIZE;
                                        ba->ba_1_org = kmalloc(size, GFP_KERNEL);
                                        if (!ba->ba_1_org)
                                                return -ENOMEM;
                                        mem_allocated += size;
                                        tmp = (unsigned long)ba->ba_1_org;
                                        tmp += ALIGN_SIZE;
                                        tmp &= ~((unsigned long)ALIGN_SIZE);
                                        ba->ba_1 = (void *)tmp;
                                        k++;
                                }
                        }
                }
        }

        /* Allocation and initialization of Statistics block */
        size = sizeof(struct stat_block);
        mac_control->stats_mem =
                pci_alloc_consistent(nic->pdev, size,
                                     &mac_control->stats_mem_phy);

        if (!mac_control->stats_mem) {
                /*
                 * In case of failure, free_shared_mem() is called, which
                 * should free any memory that was alloced till the
                 * failure happened.
                 */
                return -ENOMEM;
        }
        mem_allocated += size;
        mac_control->stats_mem_sz = size;

        tmp_v_addr = mac_control->stats_mem;
        mac_control->stats_info = (struct stat_block *)tmp_v_addr;
        memset(tmp_v_addr, 0, size);
        DBG_PRINT(INIT_DBG, "%s: Ring Mem PHY: 0x%llx\n",
                dev_name(&nic->pdev->dev), (unsigned long long)tmp_p_addr);
        mac_control->stats_info->sw_stat.mem_allocated += mem_allocated;
        return SUCCESS;
}

/**
 * free_shared_mem - Free the allocated Memory
 * @nic:  Device private variable.
 * Description: This function is to free all memory locations allocated by
 * the init_shared_mem() function and return it to the kernel.
 */

static void free_shared_mem(struct s2io_nic *nic)
{
        int i, j, blk_cnt, size;
        void *tmp_v_addr;
        dma_addr_t tmp_p_addr;
        int lst_size, lst_per_page;
        struct net_device *dev;
        int page_num = 0;
        struct config_param *config;
        struct mac_info *mac_control;
        struct stat_block *stats;
        struct swStat *swstats;

        if (!nic)
                return;

        dev = nic->dev;

        config = &nic->config;
        mac_control = &nic->mac_control;
        stats = mac_control->stats_info;
        swstats = &stats->sw_stat;

        lst_size = sizeof(struct TxD) * config->max_txds;
        lst_per_page = PAGE_SIZE / lst_size;

        for (i = 0; i < config->tx_fifo_num; i++) {
                struct fifo_info *fifo = &mac_control->fifos[i];
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                page_num = TXD_MEM_PAGE_CNT(tx_cfg->fifo_len, lst_per_page);
                for (j = 0; j < page_num; j++) {
                        int mem_blks = (j * lst_per_page);
                        struct list_info_hold *fli;

                        if (!fifo->list_info)
                                return;

                        fli = &fifo->list_info[mem_blks];
                        if (!fli->list_virt_addr)
                                break;
                        pci_free_consistent(nic->pdev, PAGE_SIZE,
                                            fli->list_virt_addr,
                                            fli->list_phy_addr);
                        swstats->mem_freed += PAGE_SIZE;
                }
                /* If we got a zero DMA address during allocation,
                 * free the page now
                 */
                if (mac_control->zerodma_virt_addr) {
                        pci_free_consistent(nic->pdev, PAGE_SIZE,
                                            mac_control->zerodma_virt_addr,
                                            (dma_addr_t)0);
                        DBG_PRINT(INIT_DBG,
                                  "%s: Freeing TxDL with zero DMA address. "
                                  "Virtual address %p\n",
                                  dev->name, mac_control->zerodma_virt_addr);
                        swstats->mem_freed += PAGE_SIZE;
                }
                kfree(fifo->list_info);
                swstats->mem_freed += tx_cfg->fifo_len *
                        sizeof(struct list_info_hold);
        }

        size = SIZE_OF_BLOCK;
        for (i = 0; i < config->rx_ring_num; i++) {

                struct ring_info *ring = &mac_control->rings[i];

                blk_cnt = ring->block_count;
                for (j = 0; j < blk_cnt; j++) {
                        tmp_v_addr = ring->rx_blocks[j].block_virt_addr;
                        tmp_p_addr = ring->rx_blocks[j].block_dma_addr;
                        if (tmp_v_addr == NULL)
                                break;
                        pci_free_consistent(nic->pdev, size,
                                            tmp_v_addr, tmp_p_addr);
                        swstats->mem_freed += size;
                        kfree(ring->rx_blocks[j].rxds);
                        swstats->mem_freed += sizeof(struct rxd_info) *
                                rxd_count[nic->rxd_mode];
                }
        }

        if (nic->rxd_mode == RXD_MODE_3B) {
                /* Freeing buffer storage addresses in 2BUFF mode. */
                for (i = 0; i < config->rx_ring_num; i++) {
                        struct rx_ring_config *rx_cfg = &config->rx_cfg[i];
                        struct ring_info *ring = &mac_control->rings[i];

                        blk_cnt = rx_cfg->num_rxd /
                                (rxd_count[nic->rxd_mode] + 1);
                        for (j = 0; j < blk_cnt; j++) {
                                int k = 0;
                                if (!ring->ba[j])
                                        continue;
                                while (k != rxd_count[nic->rxd_mode]) {
                                        struct buffAdd *ba = &ring->ba[j][k];
                                        kfree(ba->ba_0_org);
                                        swstats->mem_freed +=
                                                BUF0_LEN + ALIGN_SIZE;
                                        kfree(ba->ba_1_org);
                                        swstats->mem_freed +=
                                                BUF1_LEN + ALIGN_SIZE;
                                        k++;
                                }
                                kfree(ring->ba[j]);
                                swstats->mem_freed += sizeof(struct buffAdd) *
                                        (rxd_count[nic->rxd_mode] + 1);
                        }
                        kfree(ring->ba);
                        swstats->mem_freed += sizeof(struct buffAdd *) *
                                blk_cnt;
                }
        }

        for (i = 0; i < nic->config.tx_fifo_num; i++) {
                struct fifo_info *fifo = &mac_control->fifos[i];
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                if (fifo->ufo_in_band_v) {
                        swstats->mem_freed += tx_cfg->fifo_len *
                                sizeof(u64);
                        kfree(fifo->ufo_in_band_v);
                }
        }

        if (mac_control->stats_mem) {
                swstats->mem_freed += mac_control->stats_mem_sz;
                pci_free_consistent(nic->pdev,
                                    mac_control->stats_mem_sz,
                                    mac_control->stats_mem,
                                    mac_control->stats_mem_phy);
        }
}

/**
 * s2io_verify_pci_mode -
 */

static int s2io_verify_pci_mode(struct s2io_nic *nic)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        register u64 val64 = 0;
        int     mode;

        val64 = readq(&bar0->pci_mode);
        mode = (u8)GET_PCI_MODE(val64);

        if (val64 & PCI_MODE_UNKNOWN_MODE)
                return -1;      /* Unknown PCI mode */
        return mode;
}

#define NEC_VENID   0x1033
#define NEC_DEVID   0x0125
static int s2io_on_nec_bridge(struct pci_dev *s2io_pdev)
{
        struct pci_dev *tdev = NULL;
        while ((tdev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, tdev)) != NULL) {
                if (tdev->vendor == NEC_VENID && tdev->device == NEC_DEVID) {
                        if (tdev->bus == s2io_pdev->bus->parent) {
                                pci_dev_put(tdev);
                                return 1;
                        }
                }
        }
        return 0;
}

static int bus_speed[8] = {33, 133, 133, 200, 266, 133, 200, 266};
/**
 * s2io_print_pci_mode -
 */
static int s2io_print_pci_mode(struct s2io_nic *nic)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        register u64 val64 = 0;
        int     mode;
        struct config_param *config = &nic->config;
        const char *pcimode;

        val64 = readq(&bar0->pci_mode);
        mode = (u8)GET_PCI_MODE(val64);

        if (val64 & PCI_MODE_UNKNOWN_MODE)
                return -1;      /* Unknown PCI mode */

        config->bus_speed = bus_speed[mode];

        if (s2io_on_nec_bridge(nic->pdev)) {
                DBG_PRINT(ERR_DBG, "%s: Device is on PCI-E bus\n",
                          nic->dev->name);
                return mode;
        }

        switch (mode) {
        case PCI_MODE_PCI_33:
                pcimode = "33MHz PCI bus";
                break;
        case PCI_MODE_PCI_66:
                pcimode = "66MHz PCI bus";
                break;
        case PCI_MODE_PCIX_M1_66:
                pcimode = "66MHz PCIX(M1) bus";
                break;
        case PCI_MODE_PCIX_M1_100:
                pcimode = "100MHz PCIX(M1) bus";
                break;
        case PCI_MODE_PCIX_M1_133:
                pcimode = "133MHz PCIX(M1) bus";
                break;
        case PCI_MODE_PCIX_M2_66:
                pcimode = "133MHz PCIX(M2) bus";
                break;
        case PCI_MODE_PCIX_M2_100:
                pcimode = "200MHz PCIX(M2) bus";
                break;
        case PCI_MODE_PCIX_M2_133:
                pcimode = "266MHz PCIX(M2) bus";
                break;
        default:
                pcimode = "unsupported bus!";
                mode = -1;
        }

        DBG_PRINT(ERR_DBG, "%s: Device is on %d bit %s\n",
                  nic->dev->name, val64 & PCI_MODE_32_BITS ? 32 : 64, pcimode);

        return mode;
}

/**
 *  init_tti - Initialization transmit traffic interrupt scheme
 *  @nic: device private variable
 *  @link: link status (UP/DOWN) used to enable/disable continuous
 *  transmit interrupts
 *  Description: The function configures transmit traffic interrupts
 *  Return Value:  SUCCESS on success and
 *  '-1' on failure
 */

static int init_tti(struct s2io_nic *nic, int link)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        register u64 val64 = 0;
        int i;
        struct config_param *config = &nic->config;

        for (i = 0; i < config->tx_fifo_num; i++) {
                /*
                 * TTI Initialization. Default Tx timer gets us about
                 * 250 interrupts per sec. Continuous interrupts are enabled
                 * by default.
                 */
                if (nic->device_type == XFRAME_II_DEVICE) {
                        int count = (nic->config.bus_speed * 125)/2;
                        val64 = TTI_DATA1_MEM_TX_TIMER_VAL(count);
                } else
                        val64 = TTI_DATA1_MEM_TX_TIMER_VAL(0x2078);

                val64 |= TTI_DATA1_MEM_TX_URNG_A(0xA) |
                        TTI_DATA1_MEM_TX_URNG_B(0x10) |
                        TTI_DATA1_MEM_TX_URNG_C(0x30) |
                        TTI_DATA1_MEM_TX_TIMER_AC_EN;
                if (i == 0)
                        if (use_continuous_tx_intrs && (link == LINK_UP))
                                val64 |= TTI_DATA1_MEM_TX_TIMER_CI_EN;
                writeq(val64, &bar0->tti_data1_mem);

                if (nic->config.intr_type == MSI_X) {
                        val64 = TTI_DATA2_MEM_TX_UFC_A(0x10) |
                                TTI_DATA2_MEM_TX_UFC_B(0x100) |
                                TTI_DATA2_MEM_TX_UFC_C(0x200) |
                                TTI_DATA2_MEM_TX_UFC_D(0x300);
                } else {
                        if ((nic->config.tx_steering_type ==
                             TX_DEFAULT_STEERING) &&
                            (config->tx_fifo_num > 1) &&
                            (i >= nic->udp_fifo_idx) &&
                            (i < (nic->udp_fifo_idx +
                                  nic->total_udp_fifos)))
                                val64 = TTI_DATA2_MEM_TX_UFC_A(0x50) |
                                        TTI_DATA2_MEM_TX_UFC_B(0x80) |
                                        TTI_DATA2_MEM_TX_UFC_C(0x100) |
                                        TTI_DATA2_MEM_TX_UFC_D(0x120);
                        else
                                val64 = TTI_DATA2_MEM_TX_UFC_A(0x10) |
                                        TTI_DATA2_MEM_TX_UFC_B(0x20) |
                                        TTI_DATA2_MEM_TX_UFC_C(0x40) |
                                        TTI_DATA2_MEM_TX_UFC_D(0x80);
                }

                writeq(val64, &bar0->tti_data2_mem);

                val64 = TTI_CMD_MEM_WE |
                        TTI_CMD_MEM_STROBE_NEW_CMD |
                        TTI_CMD_MEM_OFFSET(i);
                writeq(val64, &bar0->tti_command_mem);

                if (wait_for_cmd_complete(&bar0->tti_command_mem,
                                          TTI_CMD_MEM_STROBE_NEW_CMD,
                                          S2IO_BIT_RESET) != SUCCESS)
                        return FAILURE;
        }

        return SUCCESS;
}

/**
 *  init_nic - Initialization of hardware
 *  @nic: device private variable
 *  Description: The function sequentially configures every block
 *  of the H/W from their reset values.
 *  Return Value:  SUCCESS on success and
 *  '-1' on failure (endian settings incorrect).
 */

static int init_nic(struct s2io_nic *nic)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        struct net_device *dev = nic->dev;
        register u64 val64 = 0;
        void __iomem *add;
        u32 time;
        int i, j;
        int dtx_cnt = 0;
        unsigned long long mem_share;
        int mem_size;
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;

        /* to set the swapper controle on the card */
        if (s2io_set_swapper(nic)) {
                DBG_PRINT(ERR_DBG, "ERROR: Setting Swapper failed\n");
                return -EIO;
        }

        /*
         * Herc requires EOI to be removed from reset before XGXS, so..
         */
        if (nic->device_type & XFRAME_II_DEVICE) {
                val64 = 0xA500000000ULL;
                writeq(val64, &bar0->sw_reset);
                msleep(500);
                val64 = readq(&bar0->sw_reset);
        }

        /* Remove XGXS from reset state */
        val64 = 0;
        writeq(val64, &bar0->sw_reset);
        msleep(500);
        val64 = readq(&bar0->sw_reset);

        /* Ensure that it's safe to access registers by checking
         * RIC_RUNNING bit is reset. Check is valid only for XframeII.
         */
        if (nic->device_type == XFRAME_II_DEVICE) {
                for (i = 0; i < 50; i++) {
                        val64 = readq(&bar0->adapter_status);
                        if (!(val64 & ADAPTER_STATUS_RIC_RUNNING))
                                break;
                        msleep(10);
                }
                if (i == 50)
                        return -ENODEV;
        }

        /*  Enable Receiving broadcasts */
        add = &bar0->mac_cfg;
        val64 = readq(&bar0->mac_cfg);
        val64 |= MAC_RMAC_BCAST_ENABLE;
        writeq(RMAC_CFG_KEY(0x4C0D), &bar0->rmac_cfg_key);
        writel((u32)val64, add);
        writeq(RMAC_CFG_KEY(0x4C0D), &bar0->rmac_cfg_key);
        writel((u32) (val64 >> 32), (add + 4));

        /* Read registers in all blocks */
        val64 = readq(&bar0->mac_int_mask);
        val64 = readq(&bar0->mc_int_mask);
        val64 = readq(&bar0->xgxs_int_mask);

        /*  Set MTU */
        val64 = dev->mtu;
        writeq(vBIT(val64, 2, 14), &bar0->rmac_max_pyld_len);

        if (nic->device_type & XFRAME_II_DEVICE) {
                while (herc_act_dtx_cfg[dtx_cnt] != END_SIGN) {
                        SPECIAL_REG_WRITE(herc_act_dtx_cfg[dtx_cnt],
                                          &bar0->dtx_control, UF);
                        if (dtx_cnt & 0x1)
                                msleep(1); /* Necessary!! */
                        dtx_cnt++;
                }
        } else {
                while (xena_dtx_cfg[dtx_cnt] != END_SIGN) {
                        SPECIAL_REG_WRITE(xena_dtx_cfg[dtx_cnt],
                                          &bar0->dtx_control, UF);
                        val64 = readq(&bar0->dtx_control);
                        dtx_cnt++;
                }
        }

        /*  Tx DMA Initialization */
        val64 = 0;
        writeq(val64, &bar0->tx_fifo_partition_0);
        writeq(val64, &bar0->tx_fifo_partition_1);
        writeq(val64, &bar0->tx_fifo_partition_2);
        writeq(val64, &bar0->tx_fifo_partition_3);

        for (i = 0, j = 0; i < config->tx_fifo_num; i++) {
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];

                val64 |= vBIT(tx_cfg->fifo_len - 1, ((j * 32) + 19), 13) |
                        vBIT(tx_cfg->fifo_priority, ((j * 32) + 5), 3);

                if (i == (config->tx_fifo_num - 1)) {
                        if (i % 2 == 0)
                                i++;
                }

                switch (i) {
                case 1:
                        writeq(val64, &bar0->tx_fifo_partition_0);
                        val64 = 0;
                        j = 0;
                        break;
                case 3:
                        writeq(val64, &bar0->tx_fifo_partition_1);
                        val64 = 0;
                        j = 0;
                        break;
                case 5:
                        writeq(val64, &bar0->tx_fifo_partition_2);
                        val64 = 0;
                        j = 0;
                        break;
                case 7:
                        writeq(val64, &bar0->tx_fifo_partition_3);
                        val64 = 0;
                        j = 0;
                        break;
                default:
                        j++;
                        break;
                }
        }

        /*
         * Disable 4 PCCs for Xena1, 2 and 3 as per H/W bug
         * SXE-008 TRANSMIT DMA ARBITRATION ISSUE.
         */
        if ((nic->device_type == XFRAME_I_DEVICE) && (nic->pdev->revision < 4))
                writeq(PCC_ENABLE_FOUR, &bar0->pcc_enable);

        val64 = readq(&bar0->tx_fifo_partition_0);
        DBG_PRINT(INIT_DBG, "Fifo partition at: 0x%p is: 0x%llx\n",
                  &bar0->tx_fifo_partition_0, (unsigned long long)val64);

        /*
         * Initialization of Tx_PA_CONFIG register to ignore packet
         * integrity checking.
         */
        val64 = readq(&bar0->tx_pa_cfg);
        val64 |= TX_PA_CFG_IGNORE_FRM_ERR |
                TX_PA_CFG_IGNORE_SNAP_OUI |
                TX_PA_CFG_IGNORE_LLC_CTRL |
                TX_PA_CFG_IGNORE_L2_ERR;
        writeq(val64, &bar0->tx_pa_cfg);

        /* Rx DMA intialization. */
        val64 = 0;
        for (i = 0; i < config->rx_ring_num; i++) {
                struct rx_ring_config *rx_cfg = &config->rx_cfg[i];

                val64 |= vBIT(rx_cfg->ring_priority, (5 + (i * 8)), 3);
        }
        writeq(val64, &bar0->rx_queue_priority);

        /*
         * Allocating equal share of memory to all the
         * configured Rings.
         */
        val64 = 0;
        if (nic->device_type & XFRAME_II_DEVICE)
                mem_size = 32;
        else
                mem_size = 64;

        for (i = 0; i < config->rx_ring_num; i++) {
                switch (i) {
                case 0:
                        mem_share = (mem_size / config->rx_ring_num +
                                     mem_size % config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q0_SZ(mem_share);
                        continue;
                case 1:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q1_SZ(mem_share);
                        continue;
                case 2:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q2_SZ(mem_share);
                        continue;
                case 3:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q3_SZ(mem_share);
                        continue;
                case 4:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q4_SZ(mem_share);
                        continue;
                case 5:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q5_SZ(mem_share);
                        continue;
                case 6:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q6_SZ(mem_share);
                        continue;
                case 7:
                        mem_share = (mem_size / config->rx_ring_num);
                        val64 |= RX_QUEUE_CFG_Q7_SZ(mem_share);
                        continue;
                }
        }
        writeq(val64, &bar0->rx_queue_cfg);

        /*
         * Filling Tx round robin registers
         * as per the number of FIFOs for equal scheduling priority
         */
        switch (config->tx_fifo_num) {
        case 1:
                val64 = 0x0;
                writeq(val64, &bar0->tx_w_round_robin_0);
                writeq(val64, &bar0->tx_w_round_robin_1);
                writeq(val64, &bar0->tx_w_round_robin_2);
                writeq(val64, &bar0->tx_w_round_robin_3);
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 2:
                val64 = 0x0001000100010001ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                writeq(val64, &bar0->tx_w_round_robin_1);
                writeq(val64, &bar0->tx_w_round_robin_2);
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0001000100000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 3:
                val64 = 0x0001020001020001ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                val64 = 0x0200010200010200ULL;
                writeq(val64, &bar0->tx_w_round_robin_1);
                val64 = 0x0102000102000102ULL;
                writeq(val64, &bar0->tx_w_round_robin_2);
                val64 = 0x0001020001020001ULL;
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0200010200000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 4:
                val64 = 0x0001020300010203ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                writeq(val64, &bar0->tx_w_round_robin_1);
                writeq(val64, &bar0->tx_w_round_robin_2);
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0001020300000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 5:
                val64 = 0x0001020304000102ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                val64 = 0x0304000102030400ULL;
                writeq(val64, &bar0->tx_w_round_robin_1);
                val64 = 0x0102030400010203ULL;
                writeq(val64, &bar0->tx_w_round_robin_2);
                val64 = 0x0400010203040001ULL;
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0203040000000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 6:
                val64 = 0x0001020304050001ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                val64 = 0x0203040500010203ULL;
                writeq(val64, &bar0->tx_w_round_robin_1);
                val64 = 0x0405000102030405ULL;
                writeq(val64, &bar0->tx_w_round_robin_2);
                val64 = 0x0001020304050001ULL;
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0203040500000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 7:
                val64 = 0x0001020304050600ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                val64 = 0x0102030405060001ULL;
                writeq(val64, &bar0->tx_w_round_robin_1);
                val64 = 0x0203040506000102ULL;
                writeq(val64, &bar0->tx_w_round_robin_2);
                val64 = 0x0304050600010203ULL;
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0405060000000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        case 8:
                val64 = 0x0001020304050607ULL;
                writeq(val64, &bar0->tx_w_round_robin_0);
                writeq(val64, &bar0->tx_w_round_robin_1);
                writeq(val64, &bar0->tx_w_round_robin_2);
                writeq(val64, &bar0->tx_w_round_robin_3);
                val64 = 0x0001020300000000ULL;
                writeq(val64, &bar0->tx_w_round_robin_4);
                break;
        }

        /* Enable all configured Tx FIFO partitions */
        val64 = readq(&bar0->tx_fifo_partition_0);
        val64 |= (TX_FIFO_PARTITION_EN);
        writeq(val64, &bar0->tx_fifo_partition_0);

        /* Filling the Rx round robin registers as per the
         * number of Rings and steering based on QoS with
         * equal priority.
         */
        switch (config->rx_ring_num) {
        case 1:
                val64 = 0x0;
                writeq(val64, &bar0->rx_w_round_robin_0);
                writeq(val64, &bar0->rx_w_round_robin_1);
                writeq(val64, &bar0->rx_w_round_robin_2);
                writeq(val64, &bar0->rx_w_round_robin_3);
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080808080808080ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 2:
                val64 = 0x0001000100010001ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                writeq(val64, &bar0->rx_w_round_robin_1);
                writeq(val64, &bar0->rx_w_round_robin_2);
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0001000100000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080808040404040ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 3:
                val64 = 0x0001020001020001ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                val64 = 0x0200010200010200ULL;
                writeq(val64, &bar0->rx_w_round_robin_1);
                val64 = 0x0102000102000102ULL;
                writeq(val64, &bar0->rx_w_round_robin_2);
                val64 = 0x0001020001020001ULL;
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0200010200000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080804040402020ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 4:
                val64 = 0x0001020300010203ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                writeq(val64, &bar0->rx_w_round_robin_1);
                writeq(val64, &bar0->rx_w_round_robin_2);
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0001020300000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080404020201010ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 5:
                val64 = 0x0001020304000102ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                val64 = 0x0304000102030400ULL;
                writeq(val64, &bar0->rx_w_round_robin_1);
                val64 = 0x0102030400010203ULL;
                writeq(val64, &bar0->rx_w_round_robin_2);
                val64 = 0x0400010203040001ULL;
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0203040000000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080404020201008ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 6:
                val64 = 0x0001020304050001ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                val64 = 0x0203040500010203ULL;
                writeq(val64, &bar0->rx_w_round_robin_1);
                val64 = 0x0405000102030405ULL;
                writeq(val64, &bar0->rx_w_round_robin_2);
                val64 = 0x0001020304050001ULL;
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0203040500000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080404020100804ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 7:
                val64 = 0x0001020304050600ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                val64 = 0x0102030405060001ULL;
                writeq(val64, &bar0->rx_w_round_robin_1);
                val64 = 0x0203040506000102ULL;
                writeq(val64, &bar0->rx_w_round_robin_2);
                val64 = 0x0304050600010203ULL;
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0405060000000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8080402010080402ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        case 8:
                val64 = 0x0001020304050607ULL;
                writeq(val64, &bar0->rx_w_round_robin_0);
                writeq(val64, &bar0->rx_w_round_robin_1);
                writeq(val64, &bar0->rx_w_round_robin_2);
                writeq(val64, &bar0->rx_w_round_robin_3);
                val64 = 0x0001020300000000ULL;
                writeq(val64, &bar0->rx_w_round_robin_4);

                val64 = 0x8040201008040201ULL;
                writeq(val64, &bar0->rts_qos_steering);
                break;
        }

        /* UDP Fix */
        val64 = 0;
        for (i = 0; i < 8; i++)
                writeq(val64, &bar0->rts_frm_len_n[i]);

        /* Set the default rts frame length for the rings configured */
        val64 = MAC_RTS_FRM_LEN_SET(dev->mtu+22);
        for (i = 0 ; i < config->rx_ring_num ; i++)
                writeq(val64, &bar0->rts_frm_len_n[i]);

        /* Set the frame length for the configured rings
         * desired by the user
         */
        for (i = 0; i < config->rx_ring_num; i++) {
                /* If rts_frm_len[i] == 0 then it is assumed that user not
                 * specified frame length steering.
                 * If the user provides the frame length then program
                 * the rts_frm_len register for those values or else
                 * leave it as it is.
                 */
                if (rts_frm_len[i] != 0) {
                        writeq(MAC_RTS_FRM_LEN_SET(rts_frm_len[i]),
                               &bar0->rts_frm_len_n[i]);
                }
        }

        /* Disable differentiated services steering logic */
        for (i = 0; i < 64; i++) {
                if (rts_ds_steer(nic, i, 0) == FAILURE) {
                        DBG_PRINT(ERR_DBG,
                                  "%s: rts_ds_steer failed on codepoint %d\n",
                                  dev->name, i);
                        return -ENODEV;
                }
        }

        /* Program statistics memory */
        writeq(mac_control->stats_mem_phy, &bar0->stat_addr);

        if (nic->device_type == XFRAME_II_DEVICE) {
                val64 = STAT_BC(0x320);
                writeq(val64, &bar0->stat_byte_cnt);
        }

        /*
         * Initializing the sampling rate for the device to calculate the
         * bandwidth utilization.
         */
        val64 = MAC_TX_LINK_UTIL_VAL(tmac_util_period) |
                MAC_RX_LINK_UTIL_VAL(rmac_util_period);
        writeq(val64, &bar0->mac_link_util);

        /*
         * Initializing the Transmit and Receive Traffic Interrupt
         * Scheme.
         */

        /* Initialize TTI */
        if (SUCCESS != init_tti(nic, nic->last_link_state))
                return -ENODEV;

        /* RTI Initialization */
        if (nic->device_type == XFRAME_II_DEVICE) {
                /*
                 * Programmed to generate Apprx 500 Intrs per
                 * second
                 */
                int count = (nic->config.bus_speed * 125)/4;
                val64 = RTI_DATA1_MEM_RX_TIMER_VAL(count);
        } else
                val64 = RTI_DATA1_MEM_RX_TIMER_VAL(0xFFF);
        val64 |= RTI_DATA1_MEM_RX_URNG_A(0xA) |
                RTI_DATA1_MEM_RX_URNG_B(0x10) |
                RTI_DATA1_MEM_RX_URNG_C(0x30) |
                RTI_DATA1_MEM_RX_TIMER_AC_EN;

        writeq(val64, &bar0->rti_data1_mem);

        val64 = RTI_DATA2_MEM_RX_UFC_A(0x1) |
                RTI_DATA2_MEM_RX_UFC_B(0x2) ;
        if (nic->config.intr_type == MSI_X)
                val64 |= (RTI_DATA2_MEM_RX_UFC_C(0x20) |
                          RTI_DATA2_MEM_RX_UFC_D(0x40));
        else
                val64 |= (RTI_DATA2_MEM_RX_UFC_C(0x40) |
                          RTI_DATA2_MEM_RX_UFC_D(0x80));
        writeq(val64, &bar0->rti_data2_mem);

        for (i = 0; i < config->rx_ring_num; i++) {
                val64 = RTI_CMD_MEM_WE |
                        RTI_CMD_MEM_STROBE_NEW_CMD |
                        RTI_CMD_MEM_OFFSET(i);
                writeq(val64, &bar0->rti_command_mem);

                /*
                 * Once the operation completes, the Strobe bit of the
                 * command register will be reset. We poll for this
                 * particular condition. We wait for a maximum of 500ms
                 * for the operation to complete, if it's not complete
                 * by then we return error.
                 */
                time = 0;
                while (true) {
                        val64 = readq(&bar0->rti_command_mem);
                        if (!(val64 & RTI_CMD_MEM_STROBE_NEW_CMD))
                                break;

                        if (time > 10) {
                                DBG_PRINT(ERR_DBG, "%s: RTI init failed\n",
                                          dev->name);
                                return -ENODEV;
                        }
                        time++;
                        msleep(50);
                }
        }

        /*
         * Initializing proper values as Pause threshold into all
         * the 8 Queues on Rx side.
         */
        writeq(0xffbbffbbffbbffbbULL, &bar0->mc_pause_thresh_q0q3);
        writeq(0xffbbffbbffbbffbbULL, &bar0->mc_pause_thresh_q4q7);

        /* Disable RMAC PAD STRIPPING */
        add = &bar0->mac_cfg;
        val64 = readq(&bar0->mac_cfg);
        val64 &= ~(MAC_CFG_RMAC_STRIP_PAD);
        writeq(RMAC_CFG_KEY(0x4C0D), &bar0->rmac_cfg_key);
        writel((u32) (val64), add);
        writeq(RMAC_CFG_KEY(0x4C0D), &bar0->rmac_cfg_key);
        writel((u32) (val64 >> 32), (add + 4));
        val64 = readq(&bar0->mac_cfg);

        /* Enable FCS stripping by adapter */
        add = &bar0->mac_cfg;
        val64 = readq(&bar0->mac_cfg);
        val64 |= MAC_CFG_RMAC_STRIP_FCS;
        if (nic->device_type == XFRAME_II_DEVICE)
                writeq(val64, &bar0->mac_cfg);
        else {
                writeq(RMAC_CFG_KEY(0x4C0D), &bar0->rmac_cfg_key);
                writel((u32) (val64), add);
                writeq(RMAC_CFG_KEY(0x4C0D), &bar0->rmac_cfg_key);
                writel((u32) (val64 >> 32), (add + 4));
        }

        /*
         * Set the time value to be inserted in the pause frame
         * generated by xena.
         */
        val64 = readq(&bar0->rmac_pause_cfg);
        val64 &= ~(RMAC_PAUSE_HG_PTIME(0xffff));
        val64 |= RMAC_PAUSE_HG_PTIME(nic->mac_control.rmac_pause_time);
        writeq(val64, &bar0->rmac_pause_cfg);

        /*
         * Set the Threshold Limit for Generating the pause frame
         * If the amount of data in any Queue exceeds ratio of
         * (mac_control.mc_pause_threshold_q0q3 or q4q7)/256
         * pause frame is generated
         */
        val64 = 0;
        for (i = 0; i < 4; i++) {
                val64 |= (((u64)0xFF00 |
                           nic->mac_control.mc_pause_threshold_q0q3)
                          << (i * 2 * 8));
        }
        writeq(val64, &bar0->mc_pause_thresh_q0q3);

        val64 = 0;
        for (i = 0; i < 4; i++) {
                val64 |= (((u64)0xFF00 |
                           nic->mac_control.mc_pause_threshold_q4q7)
                          << (i * 2 * 8));
        }
        writeq(val64, &bar0->mc_pause_thresh_q4q7);

        /*
         * TxDMA will stop Read request if the number of read split has
         * exceeded the limit pointed by shared_splits
         */
        val64 = readq(&bar0->pic_control);
        val64 |= PIC_CNTL_SHARED_SPLITS(shared_splits);
        writeq(val64, &bar0->pic_control);

        if (nic->config.bus_speed == 266) {
                writeq(TXREQTO_VAL(0x7f) | TXREQTO_EN, &bar0->txreqtimeout);
                writeq(0x0, &bar0->read_retry_delay);
                writeq(0x0, &bar0->write_retry_delay);
        }

        /*
         * Programming the Herc to split every write transaction
         * that does not start on an ADB to reduce disconnects.
         */
        if (nic->device_type == XFRAME_II_DEVICE) {
                val64 = FAULT_BEHAVIOUR | EXT_REQ_EN |
                        MISC_LINK_STABILITY_PRD(3);
                writeq(val64, &bar0->misc_control);
                val64 = readq(&bar0->pic_control2);
                val64 &= ~(s2BIT(13)|s2BIT(14)|s2BIT(15));
                writeq(val64, &bar0->pic_control2);
        }
        if (strstr(nic->product_name, "CX4")) {
                val64 = TMAC_AVG_IPG(0x17);
                writeq(val64, &bar0->tmac_avg_ipg);
        }

        return SUCCESS;
}
#define LINK_UP_DOWN_INTERRUPT          1
#define MAC_RMAC_ERR_TIMER              2

static int s2io_link_fault_indication(struct s2io_nic *nic)
{
        if (nic->device_type == XFRAME_II_DEVICE)
                return LINK_UP_DOWN_INTERRUPT;
        else
                return MAC_RMAC_ERR_TIMER;
}

/**
 *  do_s2io_write_bits -  update alarm bits in alarm register
 *  @value: alarm bits
 *  @flag: interrupt status
 *  @addr: address value
 *  Description: update alarm bits in alarm register
 *  Return Value:
 *  NONE.
 */
static void do_s2io_write_bits(u64 value, int flag, void __iomem *addr)
{
        u64 temp64;

        temp64 = readq(addr);

        if (flag == ENABLE_INTRS)
                temp64 &= ~((u64)value);
        else
                temp64 |= ((u64)value);
        writeq(temp64, addr);
}

static void en_dis_err_alarms(struct s2io_nic *nic, u16 mask, int flag)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        register u64 gen_int_mask = 0;
        u64 interruptible;

        writeq(DISABLE_ALL_INTRS, &bar0->general_int_mask);
        if (mask & TX_DMA_INTR) {
                gen_int_mask |= TXDMA_INT_M;

                do_s2io_write_bits(TXDMA_TDA_INT | TXDMA_PFC_INT |
                                   TXDMA_PCC_INT | TXDMA_TTI_INT |
                                   TXDMA_LSO_INT | TXDMA_TPA_INT |
                                   TXDMA_SM_INT, flag, &bar0->txdma_int_mask);

                do_s2io_write_bits(PFC_ECC_DB_ERR | PFC_SM_ERR_ALARM |
                                   PFC_MISC_0_ERR | PFC_MISC_1_ERR |
                                   PFC_PCIX_ERR | PFC_ECC_SG_ERR, flag,
                                   &bar0->pfc_err_mask);

                do_s2io_write_bits(TDA_Fn_ECC_DB_ERR | TDA_SM0_ERR_ALARM |
                                   TDA_SM1_ERR_ALARM | TDA_Fn_ECC_SG_ERR |
                                   TDA_PCIX_ERR, flag, &bar0->tda_err_mask);

                do_s2io_write_bits(PCC_FB_ECC_DB_ERR | PCC_TXB_ECC_DB_ERR |
                                   PCC_SM_ERR_ALARM | PCC_WR_ERR_ALARM |
                                   PCC_N_SERR | PCC_6_COF_OV_ERR |
                                   PCC_7_COF_OV_ERR | PCC_6_LSO_OV_ERR |
                                   PCC_7_LSO_OV_ERR | PCC_FB_ECC_SG_ERR |
                                   PCC_TXB_ECC_SG_ERR,
                                   flag, &bar0->pcc_err_mask);

                do_s2io_write_bits(TTI_SM_ERR_ALARM | TTI_ECC_SG_ERR |
                                   TTI_ECC_DB_ERR, flag, &bar0->tti_err_mask);

                do_s2io_write_bits(LSO6_ABORT | LSO7_ABORT |
                                   LSO6_SM_ERR_ALARM | LSO7_SM_ERR_ALARM |
                                   LSO6_SEND_OFLOW | LSO7_SEND_OFLOW,
                                   flag, &bar0->lso_err_mask);

                do_s2io_write_bits(TPA_SM_ERR_ALARM | TPA_TX_FRM_DROP,
                                   flag, &bar0->tpa_err_mask);

                do_s2io_write_bits(SM_SM_ERR_ALARM, flag, &bar0->sm_err_mask);
        }

        if (mask & TX_MAC_INTR) {
                gen_int_mask |= TXMAC_INT_M;
                do_s2io_write_bits(MAC_INT_STATUS_TMAC_INT, flag,
                                   &bar0->mac_int_mask);
                do_s2io_write_bits(TMAC_TX_BUF_OVRN | TMAC_TX_SM_ERR |
                                   TMAC_ECC_SG_ERR | TMAC_ECC_DB_ERR |
                                   TMAC_DESC_ECC_SG_ERR | TMAC_DESC_ECC_DB_ERR,
                                   flag, &bar0->mac_tmac_err_mask);
        }

        if (mask & TX_XGXS_INTR) {
                gen_int_mask |= TXXGXS_INT_M;
                do_s2io_write_bits(XGXS_INT_STATUS_TXGXS, flag,
                                   &bar0->xgxs_int_mask);
                do_s2io_write_bits(TXGXS_ESTORE_UFLOW | TXGXS_TX_SM_ERR |
                                   TXGXS_ECC_SG_ERR | TXGXS_ECC_DB_ERR,
                                   flag, &bar0->xgxs_txgxs_err_mask);
        }

        if (mask & RX_DMA_INTR) {
                gen_int_mask |= RXDMA_INT_M;
                do_s2io_write_bits(RXDMA_INT_RC_INT_M | RXDMA_INT_RPA_INT_M |
                                   RXDMA_INT_RDA_INT_M | RXDMA_INT_RTI_INT_M,
                                   flag, &bar0->rxdma_int_mask);
                do_s2io_write_bits(RC_PRCn_ECC_DB_ERR | RC_FTC_ECC_DB_ERR |
                                   RC_PRCn_SM_ERR_ALARM | RC_FTC_SM_ERR_ALARM |
                                   RC_PRCn_ECC_SG_ERR | RC_FTC_ECC_SG_ERR |
                                   RC_RDA_FAIL_WR_Rn, flag, &bar0->rc_err_mask);
                do_s2io_write_bits(PRC_PCI_AB_RD_Rn | PRC_PCI_AB_WR_Rn |
                                   PRC_PCI_AB_F_WR_Rn | PRC_PCI_DP_RD_Rn |
                                   PRC_PCI_DP_WR_Rn | PRC_PCI_DP_F_WR_Rn, flag,
                                   &bar0->prc_pcix_err_mask);
                do_s2io_write_bits(RPA_SM_ERR_ALARM | RPA_CREDIT_ERR |
                                   RPA_ECC_SG_ERR | RPA_ECC_DB_ERR, flag,
                                   &bar0->rpa_err_mask);
                do_s2io_write_bits(RDA_RXDn_ECC_DB_ERR | RDA_FRM_ECC_DB_N_AERR |
                                   RDA_SM1_ERR_ALARM | RDA_SM0_ERR_ALARM |
                                   RDA_RXD_ECC_DB_SERR | RDA_RXDn_ECC_SG_ERR |
                                   RDA_FRM_ECC_SG_ERR |
                                   RDA_MISC_ERR|RDA_PCIX_ERR,

                                   flag, &bar0->rda_err_mask);
                do_s2io_write_bits(RTI_SM_ERR_ALARM |
                                   RTI_ECC_SG_ERR | RTI_ECC_DB_ERR,
                                   flag, &bar0->rti_err_mask);
        }

        if (mask & RX_MAC_INTR) {
                gen_int_mask |= RXMAC_INT_M;
                do_s2io_write_bits(MAC_INT_STATUS_RMAC_INT, flag,
                                   &bar0->mac_int_mask);
                interruptible = (RMAC_RX_BUFF_OVRN | RMAC_RX_SM_ERR |
                                 RMAC_UNUSED_INT | RMAC_SINGLE_ECC_ERR |
                                 RMAC_DOUBLE_ECC_ERR);
                if (s2io_link_fault_indication(nic) == MAC_RMAC_ERR_TIMER)
                        interruptible |= RMAC_LINK_STATE_CHANGE_INT;
                do_s2io_write_bits(interruptible,
                                   flag, &bar0->mac_rmac_err_mask);
        }

        if (mask & RX_XGXS_INTR) {
                gen_int_mask |= RXXGXS_INT_M;
                do_s2io_write_bits(XGXS_INT_STATUS_RXGXS, flag,
                                   &bar0->xgxs_int_mask);
                do_s2io_write_bits(RXGXS_ESTORE_OFLOW | RXGXS_RX_SM_ERR, flag,
                                   &bar0->xgxs_rxgxs_err_mask);
        }

        if (mask & MC_INTR) {
                gen_int_mask |= MC_INT_M;
                do_s2io_write_bits(MC_INT_MASK_MC_INT,
                                   flag, &bar0->mc_int_mask);
                do_s2io_write_bits(MC_ERR_REG_SM_ERR | MC_ERR_REG_ECC_ALL_SNG |
                                   MC_ERR_REG_ECC_ALL_DBL | PLL_LOCK_N, flag,
                                   &bar0->mc_err_mask);
        }
        nic->general_int_mask = gen_int_mask;

        /* Remove this line when alarm interrupts are enabled */
        nic->general_int_mask = 0;
}

/**
 *  en_dis_able_nic_intrs - Enable or Disable the interrupts
 *  @nic: device private variable,
 *  @mask: A mask indicating which Intr block must be modified and,
 *  @flag: A flag indicating whether to enable or disable the Intrs.
 *  Description: This function will either disable or enable the interrupts
 *  depending on the flag argument. The mask argument can be used to
 *  enable/disable any Intr block.
 *  Return Value: NONE.
 */

static void en_dis_able_nic_intrs(struct s2io_nic *nic, u16 mask, int flag)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        register u64 temp64 = 0, intr_mask = 0;

        intr_mask = nic->general_int_mask;

        /*  Top level interrupt classification */
        /*  PIC Interrupts */
        if (mask & TX_PIC_INTR) {
                /*  Enable PIC Intrs in the general intr mask register */
                intr_mask |= TXPIC_INT_M;
                if (flag == ENABLE_INTRS) {
                        /*
                         * If Hercules adapter enable GPIO otherwise
                         * disable all PCIX, Flash, MDIO, IIC and GPIO
                         * interrupts for now.
                         * TODO
                         */
                        if (s2io_link_fault_indication(nic) ==
                            LINK_UP_DOWN_INTERRUPT) {
                                do_s2io_write_bits(PIC_INT_GPIO, flag,
                                                   &bar0->pic_int_mask);
                                do_s2io_write_bits(GPIO_INT_MASK_LINK_UP, flag,
                                                   &bar0->gpio_int_mask);
                        } else
                                writeq(DISABLE_ALL_INTRS, &bar0->pic_int_mask);
                } else if (flag == DISABLE_INTRS) {
                        /*
                         * Disable PIC Intrs in the general
                         * intr mask register
                         */
                        writeq(DISABLE_ALL_INTRS, &bar0->pic_int_mask);
                }
        }

        /*  Tx traffic interrupts */
        if (mask & TX_TRAFFIC_INTR) {
                intr_mask |= TXTRAFFIC_INT_M;
                if (flag == ENABLE_INTRS) {
                        /*
                         * Enable all the Tx side interrupts
                         * writing 0 Enables all 64 TX interrupt levels
                         */
                        writeq(0x0, &bar0->tx_traffic_mask);
                } else if (flag == DISABLE_INTRS) {
                        /*
                         * Disable Tx Traffic Intrs in the general intr mask
                         * register.
                         */
                        writeq(DISABLE_ALL_INTRS, &bar0->tx_traffic_mask);
                }
        }

        /*  Rx traffic interrupts */
        if (mask & RX_TRAFFIC_INTR) {
                intr_mask |= RXTRAFFIC_INT_M;
                if (flag == ENABLE_INTRS) {
                        /* writing 0 Enables all 8 RX interrupt levels */
                        writeq(0x0, &bar0->rx_traffic_mask);
                } else if (flag == DISABLE_INTRS) {
                        /*
                         * Disable Rx Traffic Intrs in the general intr mask
                         * register.
                         */
                        writeq(DISABLE_ALL_INTRS, &bar0->rx_traffic_mask);
                }
        }

        temp64 = readq(&bar0->general_int_mask);
        if (flag == ENABLE_INTRS)
                temp64 &= ~((u64)intr_mask);
        else
                temp64 = DISABLE_ALL_INTRS;
        writeq(temp64, &bar0->general_int_mask);

        nic->general_int_mask = readq(&bar0->general_int_mask);
}

/**
 *  verify_pcc_quiescent- Checks for PCC quiescent state
 *  Return: 1 If PCC is quiescence
 *          0 If PCC is not quiescence
 */
static int verify_pcc_quiescent(struct s2io_nic *sp, int flag)
{
        int ret = 0, herc;
        struct XENA_dev_config __iomem *bar0 = sp->bar0;
        u64 val64 = readq(&bar0->adapter_status);

        herc = (sp->device_type == XFRAME_II_DEVICE);

        if (flag == false) {
                if ((!herc && (sp->pdev->revision >= 4)) || herc) {
                        if (!(val64 & ADAPTER_STATUS_RMAC_PCC_IDLE))
                                ret = 1;
                } else {
                        if (!(val64 & ADAPTER_STATUS_RMAC_PCC_FOUR_IDLE))
                                ret = 1;
                }
        } else {
                if ((!herc && (sp->pdev->revision >= 4)) || herc) {
                        if (((val64 & ADAPTER_STATUS_RMAC_PCC_IDLE) ==
                             ADAPTER_STATUS_RMAC_PCC_IDLE))
                                ret = 1;
                } else {
                        if (((val64 & ADAPTER_STATUS_RMAC_PCC_FOUR_IDLE) ==
                             ADAPTER_STATUS_RMAC_PCC_FOUR_IDLE))
                                ret = 1;
                }
        }

        return ret;
}
/**
 *  verify_xena_quiescence - Checks whether the H/W is ready
 *  Description: Returns whether the H/W is ready to go or not. Depending
 *  on whether adapter enable bit was written or not the comparison
 *  differs and the calling function passes the input argument flag to
 *  indicate this.
 *  Return: 1 If xena is quiescence
 *          0 If Xena is not quiescence
 */

static int verify_xena_quiescence(struct s2io_nic *sp)
{
        int  mode;
        struct XENA_dev_config __iomem *bar0 = sp->bar0;
        u64 val64 = readq(&bar0->adapter_status);
        mode = s2io_verify_pci_mode(sp);

        if (!(val64 & ADAPTER_STATUS_TDMA_READY)) {
                DBG_PRINT(ERR_DBG, "TDMA is not ready!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_RDMA_READY)) {
                DBG_PRINT(ERR_DBG, "RDMA is not ready!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_PFC_READY)) {
                DBG_PRINT(ERR_DBG, "PFC is not ready!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_TMAC_BUF_EMPTY)) {
                DBG_PRINT(ERR_DBG, "TMAC BUF is not empty!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_PIC_QUIESCENT)) {
                DBG_PRINT(ERR_DBG, "PIC is not QUIESCENT!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_MC_DRAM_READY)) {
                DBG_PRINT(ERR_DBG, "MC_DRAM is not ready!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_MC_QUEUES_READY)) {
                DBG_PRINT(ERR_DBG, "MC_QUEUES is not ready!\n");
                return 0;
        }
        if (!(val64 & ADAPTER_STATUS_M_PLL_LOCK)) {
                DBG_PRINT(ERR_DBG, "M_PLL is not locked!\n");
                return 0;
        }

        /*
         * In PCI 33 mode, the P_PLL is not used, and therefore,
         * the the P_PLL_LOCK bit in the adapter_status register will
         * not be asserted.
         */
        if (!(val64 & ADAPTER_STATUS_P_PLL_LOCK) &&
            sp->device_type == XFRAME_II_DEVICE &&
            mode != PCI_MODE_PCI_33) {
                DBG_PRINT(ERR_DBG, "P_PLL is not locked!\n");
                return 0;
        }
        if (!((val64 & ADAPTER_STATUS_RC_PRC_QUIESCENT) ==
              ADAPTER_STATUS_RC_PRC_QUIESCENT)) {
                DBG_PRINT(ERR_DBG, "RC_PRC is not QUIESCENT!\n");
                return 0;
        }
        return 1;
}

/**
 * fix_mac_address -  Fix for Mac addr problem on Alpha platforms
 * @sp: Pointer to device specifc structure
 * Description :
 * New procedure to clear mac address reading  problems on Alpha platforms
 *
 */

static void fix_mac_address(struct s2io_nic *sp)
{
        struct XENA_dev_config __iomem *bar0 = sp->bar0;
        u64 val64;
        int i = 0;

        while (fix_mac[i] != END_SIGN) {
                writeq(fix_mac[i++], &bar0->gpio_control);
                udelay(10);
                val64 = readq(&bar0->gpio_control);
        }
}

/**
 *  start_nic - Turns the device on
 *  @nic : device private variable.
 *  Description:
 *  This function actually turns the device on. Before this  function is
 *  called,all Registers are configured from their reset states
 *  and shared memory is allocated but the NIC is still quiescent. On
 *  calling this function, the device interrupts are cleared and the NIC is
 *  literally switched on by writing into the adapter control register.
 *  Return Value:
 *  SUCCESS on success and -1 on failure.
 */

static int start_nic(struct s2io_nic *nic)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        struct net_device *dev = nic->dev;
        register u64 val64 = 0;
        u16 subid, i;
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;

        /*  PRC Initialization and configuration */
        for (i = 0; i < config->rx_ring_num; i++) {
                struct ring_info *ring = &mac_control->rings[i];

                writeq((u64)ring->rx_blocks[0].block_dma_addr,
                       &bar0->prc_rxd0_n[i]);

                val64 = readq(&bar0->prc_ctrl_n[i]);
                if (nic->rxd_mode == RXD_MODE_1)
                        val64 |= PRC_CTRL_RC_ENABLED;
                else
                        val64 |= PRC_CTRL_RC_ENABLED | PRC_CTRL_RING_MODE_3;
                if (nic->device_type == XFRAME_II_DEVICE)
                        val64 |= PRC_CTRL_GROUP_READS;
                val64 &= ~PRC_CTRL_RXD_BACKOFF_INTERVAL(0xFFFFFF);
                val64 |= PRC_CTRL_RXD_BACKOFF_INTERVAL(0x1000);
                writeq(val64, &bar0->prc_ctrl_n[i]);
        }

        if (nic->rxd_mode == RXD_MODE_3B) {
                /* Enabling 2 buffer mode by writing into Rx_pa_cfg reg. */
                val64 = readq(&bar0->rx_pa_cfg);
                val64 |= RX_PA_CFG_IGNORE_L2_ERR;
                writeq(val64, &bar0->rx_pa_cfg);
        }

        if (vlan_tag_strip == 0) {
                val64 = readq(&bar0->rx_pa_cfg);
                val64 &= ~RX_PA_CFG_STRIP_VLAN_TAG;
                writeq(val64, &bar0->rx_pa_cfg);
                nic->vlan_strip_flag = 0;
        }

        /*
         * Enabling MC-RLDRAM. After enabling the device, we timeout
         * for around 100ms, which is approximately the time required
         * for the device to be ready for operation.
         */
        val64 = readq(&bar0->mc_rldram_mrs);
        val64 |= MC_RLDRAM_QUEUE_SIZE_ENABLE | MC_RLDRAM_MRS_ENABLE;
        SPECIAL_REG_WRITE(val64, &bar0->mc_rldram_mrs, UF);
        val64 = readq(&bar0->mc_rldram_mrs);

        msleep(100);    /* Delay by around 100 ms. */

        /* Enabling ECC Protection. */
        val64 = readq(&bar0->adapter_control);
        val64 &= ~ADAPTER_ECC_EN;
        writeq(val64, &bar0->adapter_control);

        /*
         * Verify if the device is ready to be enabled, if so enable
         * it.
         */
        val64 = readq(&bar0->adapter_status);
        if (!verify_xena_quiescence(nic)) {
                DBG_PRINT(ERR_DBG, "%s: device is not ready, "
                          "Adapter status reads: 0x%llx\n",
                          dev->name, (unsigned long long)val64);
                return FAILURE;
        }

        /*
         * With some switches, link might be already up at this point.
         * Because of this weird behavior, when we enable laser,
         * we may not get link. We need to handle this. We cannot
         * figure out which switch is misbehaving. So we are forced to
         * make a global change.
         */

        /* Enabling Laser. */
        val64 = readq(&bar0->adapter_control);
        val64 |= ADAPTER_EOI_TX_ON;
        writeq(val64, &bar0->adapter_control);

        if (s2io_link_fault_indication(nic) == MAC_RMAC_ERR_TIMER) {
                /*
                 * Dont see link state interrupts initially on some switches,
                 * so directly scheduling the link state task here.
                 */
                schedule_work(&nic->set_link_task);
        }
        /* SXE-002: Initialize link and activity LED */
        subid = nic->pdev->subsystem_device;
        if (((subid & 0xFF) >= 0x07) &&
            (nic->device_type == XFRAME_I_DEVICE)) {
                val64 = readq(&bar0->gpio_control);
                val64 |= 0x0000800000000000ULL;
                writeq(val64, &bar0->gpio_control);
                val64 = 0x0411040400000000ULL;
                writeq(val64, (void __iomem *)bar0 + 0x2700);
        }

        return SUCCESS;
}
/**
 * s2io_txdl_getskb - Get the skb from txdl, unmap and return skb
 */
static struct sk_buff *s2io_txdl_getskb(struct fifo_info *fifo_data,
                                        struct TxD *txdlp, int get_off)
{
        struct s2io_nic *nic = fifo_data->nic;
        struct sk_buff *skb;
        struct TxD *txds;
        u16 j, frg_cnt;

        txds = txdlp;
        if (txds->Host_Control == (u64)(long)fifo_data->ufo_in_band_v) {
                pci_unmap_single(nic->pdev, (dma_addr_t)txds->Buffer_Pointer,
                                 sizeof(u64), PCI_DMA_TODEVICE);
                txds++;
        }

        skb = (struct sk_buff *)((unsigned long)txds->Host_Control);
        if (!skb) {
                memset(txdlp, 0, (sizeof(struct TxD) * fifo_data->max_txds));
                return NULL;
        }
        pci_unmap_single(nic->pdev, (dma_addr_t)txds->Buffer_Pointer,
                         skb_headlen(skb), PCI_DMA_TODEVICE);
        frg_cnt = skb_shinfo(skb)->nr_frags;
        if (frg_cnt) {
                txds++;
                for (j = 0; j < frg_cnt; j++, txds++) {
                        skb_frag_t *frag = &skb_shinfo(skb)->frags[j];
                        if (!txds->Buffer_Pointer)
                                break;
                        pci_unmap_page(nic->pdev,
                                       (dma_addr_t)txds->Buffer_Pointer,
                                       frag->size, PCI_DMA_TODEVICE);
                }
        }
        memset(txdlp, 0, (sizeof(struct TxD) * fifo_data->max_txds));
        return skb;
}

/**
 *  free_tx_buffers - Free all queued Tx buffers
 *  @nic : device private variable.
 *  Description:
 *  Free all queued Tx buffers.
 *  Return Value: void
 */

static void free_tx_buffers(struct s2io_nic *nic)
{
        struct net_device *dev = nic->dev;
        struct sk_buff *skb;
        struct TxD *txdp;
        int i, j;
        int cnt = 0;
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;
        struct stat_block *stats = mac_control->stats_info;
        struct swStat *swstats = &stats->sw_stat;

        for (i = 0; i < config->tx_fifo_num; i++) {
                struct tx_fifo_config *tx_cfg = &config->tx_cfg[i];
                struct fifo_info *fifo = &mac_control->fifos[i];
                unsigned long flags;

                spin_lock_irqsave(&fifo->tx_lock, flags);
                for (j = 0; j < tx_cfg->fifo_len; j++) {
                        txdp = (struct TxD *)fifo->list_info[j].list_virt_addr;
                        skb = s2io_txdl_getskb(&mac_control->fifos[i], txdp, j);
                        if (skb) {
                                swstats->mem_freed += skb->truesize;
                                dev_kfree_skb(skb);
                                cnt++;
                        }
                }
                DBG_PRINT(INTR_DBG,
                          "%s: forcibly freeing %d skbs on FIFO%d\n",
                          dev->name, cnt, i);
                fifo->tx_curr_get_info.offset = 0;
                fifo->tx_curr_put_info.offset = 0;
                spin_unlock_irqrestore(&fifo->tx_lock, flags);
        }
}

/**
 *   stop_nic -  To stop the nic
 *   @nic ; device private variable.
 *   Description:
 *   This function does exactly the opposite of what the start_nic()
 *   function does. This function is called to stop the device.
 *   Return Value:
 *   void.
 */

static void stop_nic(struct s2io_nic *nic)
{
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        register u64 val64 = 0;
        u16 interruptible;

        /*  Disable all interrupts */
        en_dis_err_alarms(nic, ENA_ALL_INTRS, DISABLE_INTRS);
        interruptible = TX_TRAFFIC_INTR | RX_TRAFFIC_INTR;
        interruptible |= TX_PIC_INTR;
        en_dis_able_nic_intrs(nic, interruptible, DISABLE_INTRS);

        /* Clearing Adapter_En bit of ADAPTER_CONTROL Register */
        val64 = readq(&bar0->adapter_control);
        val64 &= ~(ADAPTER_CNTL_EN);
        writeq(val64, &bar0->adapter_control);
}

/**
 *  fill_rx_buffers - Allocates the Rx side skbs
 *  @ring_info: per ring structure
 *  @from_card_up: If this is true, we will map the buffer to get
 *     the dma address for buf0 and buf1 to give it to the card.
 *     Else we will sync the already mapped buffer to give it to the card.
 *  Description:
 *  The function allocates Rx side skbs and puts the physical
 *  address of these buffers into the RxD buffer pointers, so that the NIC
 *  can DMA the received frame into these locations.
 *  The NIC supports 3 receive modes, viz
 *  1. single buffer,
 *  2. three buffer and
 *  3. Five buffer modes.
 *  Each mode defines how many fragments the received frame will be split
 *  up into by the NIC. The frame is split into L3 header, L4 Header,
 *  L4 payload in three buffer mode and in 5 buffer mode, L4 payload itself
 *  is split into 3 fragments. As of now only single buffer mode is
 *  supported.
 *   Return Value:
 *  SUCCESS on success or an appropriate -ve value on failure.
 */
static int fill_rx_buffers(struct s2io_nic *nic, struct ring_info *ring,
                           int from_card_up)
{
        struct sk_buff *skb;
        struct RxD_t *rxdp;
        int off, size, block_no, block_no1;
        u32 alloc_tab = 0;
        u32 alloc_cnt;
        u64 tmp;
        struct buffAdd *ba;
        struct RxD_t *first_rxdp = NULL;
        u64 Buffer0_ptr = 0, Buffer1_ptr = 0;
        int rxd_index = 0;
        struct RxD1 *rxdp1;
        struct RxD3 *rxdp3;
        struct swStat *swstats = &ring->nic->mac_control.stats_info->sw_stat;

        alloc_cnt = ring->pkt_cnt - ring->rx_bufs_left;

        block_no1 = ring->rx_curr_get_info.block_index;
        while (alloc_tab < alloc_cnt) {
                block_no = ring->rx_curr_put_info.block_index;

                off = ring->rx_curr_put_info.offset;

                rxdp = ring->rx_blocks[block_no].rxds[off].virt_addr;

                rxd_index = off + 1;
                if (block_no)
                        rxd_index += (block_no * ring->rxd_count);

                if ((block_no == block_no1) &&
                    (off == ring->rx_curr_get_info.offset) &&
                    (rxdp->Host_Control)) {
                        DBG_PRINT(INTR_DBG, "%s: Get and Put info equated\n",
                                  ring->dev->name);
                        goto end;
                }
                if (off && (off == ring->rxd_count)) {
                        ring->rx_curr_put_info.block_index++;
                        if (ring->rx_curr_put_info.block_index ==
                            ring->block_count)
                                ring->rx_curr_put_info.block_index = 0;
                        block_no = ring->rx_curr_put_info.block_index;
                        off = 0;
                        ring->rx_curr_put_info.offset = off;
                        rxdp = ring->rx_blocks[block_no].block_virt_addr;
                        DBG_PRINT(INTR_DBG, "%s: Next block at: %p\n",
                                  ring->dev->name, rxdp);

                }

                if ((rxdp->Control_1 & RXD_OWN_XENA) &&
                    ((ring->rxd_mode == RXD_MODE_3B) &&
                     (rxdp->Control_2 & s2BIT(0)))) {
                        ring->rx_curr_put_info.offset = off;
                        goto end;
                }
                /* calculate size of skb based on ring mode */
                size = ring->mtu +
                        HEADER_ETHERNET_II_802_3_SIZE +
                        HEADER_802_2_SIZE + HEADER_SNAP_SIZE;
                if (ring->rxd_mode == RXD_MODE_1)
                        size += NET_IP_ALIGN;
                else
                        size = ring->mtu + ALIGN_SIZE + BUF0_LEN + 4;

                /* allocate skb */
                skb = dev_alloc_skb(size);
                if (!skb) {
                        DBG_PRINT(INFO_DBG, "%s: Could not allocate skb\n",
                                  ring->dev->name);
                        if (first_rxdp) {
                                wmb();
                                first_rxdp->Control_1 |= RXD_OWN_XENA;
                        }
                        swstats->mem_alloc_fail_cnt++;

                        return -ENOMEM ;
                }
                swstats->mem_allocated += skb->truesize;

                if (ring->rxd_mode == RXD_MODE_1) {
                        /* 1 buffer mode - normal operation mode */
                        rxdp1 = (struct RxD1 *)rxdp;
                        memset(rxdp, 0, sizeof(struct RxD1));
                        skb_reserve(skb, NET_IP_ALIGN);
                        rxdp1->Buffer0_ptr =
                                pci_map_single(ring->pdev, skb->data,
                                               size - NET_IP_ALIGN,
                                               PCI_DMA_FROMDEVICE);
                        if (pci_dma_mapping_error(nic->pdev,
                                                  rxdp1->Buffer0_ptr))
                                goto pci_map_failed;

                        rxdp->Control_2 =
                                SET_BUFFER0_SIZE_1(size - NET_IP_ALIGN);
                        rxdp->Host_Control = (unsigned long)skb;
                } else if (ring->rxd_mode == RXD_MODE_3B) {
                        /*
                         * 2 buffer mode -
                         * 2 buffer mode provides 128
                         * byte aligned receive buffers.
                         */

                        rxdp3 = (struct RxD3 *)rxdp;
                        /* save buffer pointers to avoid frequent dma mapping */
                        Buffer0_ptr = rxdp3->Buffer0_ptr;
                        Buffer1_ptr = rxdp3->Buffer1_ptr;
                        memset(rxdp, 0, sizeof(struct RxD3));
                        /* restore the buffer pointers for dma sync*/
                        rxdp3->Buffer0_ptr = Buffer0_ptr;
                        rxdp3->Buffer1_ptr = Buffer1_ptr;

                        ba = &ring->ba[block_no][off];
                        skb_reserve(skb, BUF0_LEN);
                        tmp = (u64)(unsigned long)skb->data;
                        tmp += ALIGN_SIZE;
                        tmp &= ~ALIGN_SIZE;
                        skb->data = (void *) (unsigned long)tmp;
                        skb_reset_tail_pointer(skb);

                        if (from_card_up) {
                                rxdp3->Buffer0_ptr =
                                        pci_map_single(ring->pdev, ba->ba_0,
                                                       BUF0_LEN,
                                                       PCI_DMA_FROMDEVICE);
                                if (pci_dma_mapping_error(nic->pdev,
                                                          rxdp3->Buffer0_ptr))
                                        goto pci_map_failed;
                        } else
                                pci_dma_sync_single_for_device(ring->pdev,
                                                               (dma_addr_t)rxdp3->Buffer0_ptr,
                                                               BUF0_LEN,
                                                               PCI_DMA_FROMDEVICE);

                        rxdp->Control_2 = SET_BUFFER0_SIZE_3(BUF0_LEN);
                        if (ring->rxd_mode == RXD_MODE_3B) {
                                /* Two buffer mode */

                                /*
                                 * Buffer2 will have L3/L4 header plus
                                 * L4 payload
                                 */
                                rxdp3->Buffer2_ptr = pci_map_single(ring->pdev,
                                                                    skb->data,
                                                                    ring->mtu + 4,
                                                                    PCI_DMA_FROMDEVICE);

                                if (pci_dma_mapping_error(nic->pdev,
                                                          rxdp3->Buffer2_ptr))
                                        goto pci_map_failed;

                                if (from_card_up) {
                                        rxdp3->Buffer1_ptr =
                                                pci_map_single(ring->pdev,
                                                               ba->ba_1,
                                                               BUF1_LEN,
                                                               PCI_DMA_FROMDEVICE);

                                        if (pci_dma_mapping_error(nic->pdev,
                                                                  rxdp3->Buffer1_ptr)) {
                                                pci_unmap_single(ring->pdev,
                                                                 (dma_addr_t)(unsigned long)
                                                                 skb->data,
                                                                 ring->mtu + 4,
                                                                 PCI_DMA_FROMDEVICE);
                                                goto pci_map_failed;
                                        }
                                }
                                rxdp->Control_2 |= SET_BUFFER1_SIZE_3(1);
                                rxdp->Control_2 |= SET_BUFFER2_SIZE_3
                                        (ring->mtu + 4);
                        }
                        rxdp->Control_2 |= s2BIT(0);
                        rxdp->Host_Control = (unsigned long) (skb);
                }
                if (alloc_tab & ((1 << rxsync_frequency) - 1))
                        rxdp->Control_1 |= RXD_OWN_XENA;
                off++;
                if (off == (ring->rxd_count + 1))
                        off = 0;
                ring->rx_curr_put_info.offset = off;

                rxdp->Control_2 |= SET_RXD_MARKER;
                if (!(alloc_tab & ((1 << rxsync_frequency) - 1))) {
                        if (first_rxdp) {
                                wmb();
                                first_rxdp->Control_1 |= RXD_OWN_XENA;
                        }
                        first_rxdp = rxdp;
                }
                ring->rx_bufs_left += 1;
                alloc_tab++;
        }

end:
        /* Transfer ownership of first descriptor to adapter just before
         * exiting. Before that, use memory barrier so that ownership
         * and other fields are seen by adapter correctly.
         */
        if (first_rxdp) {
                wmb();
                first_rxdp->Control_1 |= RXD_OWN_XENA;
        }

        return SUCCESS;

pci_map_failed:
        swstats->pci_map_fail_cnt++;
        swstats->mem_freed += skb->truesize;
        dev_kfree_skb_irq(skb);
        return -ENOMEM;
}

static void free_rxd_blk(struct s2io_nic *sp, int ring_no, int blk)
{
        struct net_device *dev = sp->dev;
        int j;
        struct sk_buff *skb;
        struct RxD_t *rxdp;
        struct buffAdd *ba;
        struct RxD1 *rxdp1;
        struct RxD3 *rxdp3;
        struct mac_info *mac_control = &sp->mac_control;
        struct stat_block *stats = mac_control->stats_info;
        struct swStat *swstats = &stats->sw_stat;

        for (j = 0 ; j < rxd_count[sp->rxd_mode]; j++) {
                rxdp = mac_control->rings[ring_no].
                        rx_blocks[blk].rxds[j].virt_addr;
                skb = (struct sk_buff *)((unsigned long)rxdp->Host_Control);
                if (!skb)
                        continue;
                if (sp->rxd_mode == RXD_MODE_1) {
                        rxdp1 = (struct RxD1 *)rxdp;
                        pci_unmap_single(sp->pdev,
                                         (dma_addr_t)rxdp1->Buffer0_ptr,
                                         dev->mtu +
                                         HEADER_ETHERNET_II_802_3_SIZE +
                                         HEADER_802_2_SIZE + HEADER_SNAP_SIZE,
                                         PCI_DMA_FROMDEVICE);
                        memset(rxdp, 0, sizeof(struct RxD1));
                } else if (sp->rxd_mode == RXD_MODE_3B) {
                        rxdp3 = (struct RxD3 *)rxdp;
                        ba = &mac_control->rings[ring_no].ba[blk][j];
                        pci_unmap_single(sp->pdev,
                                         (dma_addr_t)rxdp3->Buffer0_ptr,
                                         BUF0_LEN,
                                         PCI_DMA_FROMDEVICE);
                        pci_unmap_single(sp->pdev,
                                         (dma_addr_t)rxdp3->Buffer1_ptr,
                                         BUF1_LEN,
                                         PCI_DMA_FROMDEVICE);
                        pci_unmap_single(sp->pdev,
                                         (dma_addr_t)rxdp3->Buffer2_ptr,
                                         dev->mtu + 4,
                                         PCI_DMA_FROMDEVICE);
                        memset(rxdp, 0, sizeof(struct RxD3));
                }
                swstats->mem_freed += skb->truesize;
                dev_kfree_skb(skb);
                mac_control->rings[ring_no].rx_bufs_left -= 1;
        }
}

/**
 *  free_rx_buffers - Frees all Rx buffers
 *  @sp: device private variable.
 *  Description:
 *  This function will free all Rx buffers allocated by host.
 *  Return Value:
 *  NONE.
 */

static void free_rx_buffers(struct s2io_nic *sp)
{
        struct net_device *dev = sp->dev;
        int i, blk = 0, buf_cnt = 0;
        struct config_param *config = &sp->config;
        struct mac_info *mac_control = &sp->mac_control;

        for (i = 0; i < config->rx_ring_num; i++) {
                struct ring_info *ring = &mac_control->rings[i];

                for (blk = 0; blk < rx_ring_sz[i]; blk++)
                        free_rxd_blk(sp, i, blk);

                ring->rx_curr_put_info.block_index = 0;
                ring->rx_curr_get_info.block_index = 0;
                ring->rx_curr_put_info.offset = 0;
                ring->rx_curr_get_info.offset = 0;
                ring->rx_bufs_left = 0;
                DBG_PRINT(INIT_DBG, "%s: Freed 0x%x Rx Buffers on ring%d\n",
                          dev->name, buf_cnt, i);
        }
}

static int s2io_chk_rx_buffers(struct s2io_nic *nic, struct ring_info *ring)
{
        if (fill_rx_buffers(nic, ring, 0) == -ENOMEM) {
                DBG_PRINT(INFO_DBG, "%s: Out of memory in Rx Intr!!\n",
                          ring->dev->name);
        }
        return 0;
}

/**
 * s2io_poll - Rx interrupt handler for NAPI support
 * @napi : pointer to the napi structure.
 * @budget : The number of packets that were budgeted to be processed
 * during  one pass through the 'Poll" function.
 * Description:
 * Comes into picture only if NAPI support has been incorporated. It does
 * the same thing that rx_intr_handler does, but not in a interrupt context
 * also It will process only a given number of packets.
 * Return value:
 * 0 on success and 1 if there are No Rx packets to be processed.
 */

static int s2io_poll_msix(struct napi_struct *napi, int budget)
{
        struct ring_info *ring = container_of(napi, struct ring_info, napi);
        struct net_device *dev = ring->dev;
        int pkts_processed = 0;
        u8 __iomem *addr = NULL;
        u8 val8 = 0;
        struct s2io_nic *nic = netdev_priv(dev);
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        int budget_org = budget;

        if (unlikely(!is_s2io_card_up(nic)))
                return 0;

        pkts_processed = rx_intr_handler(ring, budget);
        s2io_chk_rx_buffers(nic, ring);

        if (pkts_processed < budget_org) {
                napi_complete(napi);
                /*Re Enable MSI-Rx Vector*/
                addr = (u8 __iomem *)&bar0->xmsi_mask_reg;
                addr += 7 - ring->ring_no;
                val8 = (ring->ring_no == 0) ? 0x3f : 0xbf;
                writeb(val8, addr);
                val8 = readb(addr);
        }
        return pkts_processed;
}

static int s2io_poll_inta(struct napi_struct *napi, int budget)
{
        struct s2io_nic *nic = container_of(napi, struct s2io_nic, napi);
        int pkts_processed = 0;
        int ring_pkts_processed, i;
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        int budget_org = budget;
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;

        if (unlikely(!is_s2io_card_up(nic)))
                return 0;

        for (i = 0; i < config->rx_ring_num; i++) {
                struct ring_info *ring = &mac_control->rings[i];
                ring_pkts_processed = rx_intr_handler(ring, budget);
                s2io_chk_rx_buffers(nic, ring);
                pkts_processed += ring_pkts_processed;
                budget -= ring_pkts_processed;
                if (budget <= 0)
                        break;
        }
        if (pkts_processed < budget_org) {
                napi_complete(napi);
                /* Re enable the Rx interrupts for the ring */
                writeq(0, &bar0->rx_traffic_mask);
                readl(&bar0->rx_traffic_mask);
        }
        return pkts_processed;
}

#ifdef CONFIG_NET_POLL_CONTROLLER
/**
 * s2io_netpoll - netpoll event handler entry point
 * @dev : pointer to the device structure.
 * Description:
 *      This function will be called by upper layer to check for events on the
 * interface in situations where interrupts are disabled. It is used for
 * specific in-kernel networking tasks, such as remote consoles and kernel
 * debugging over the network (example netdump in RedHat).
 */
static void s2io_netpoll(struct net_device *dev)
{
        struct s2io_nic *nic = netdev_priv(dev);
        struct XENA_dev_config __iomem *bar0 = nic->bar0;
        u64 val64 = 0xFFFFFFFFFFFFFFFFULL;
        int i;
        struct config_param *config = &nic->config;
        struct mac_info *mac_control = &nic->mac_control;

        if (pci_channel_offline(nic->pdev))
                return;

        disable_irq(dev->irq);

        writeq(val64, &bar0->rx_traffic_int);
        writeq(val64, &bar0->tx_traffic_int);

        /* we need to free up the transmitted skbufs or else netpoll will
         * run out of skbs and will fail and eventually netpoll application such
         * as netdump will fail.
         */
        for (i = 0; i < config->tx_fifo_num; i++)
                tx_intr_handler(&mac_control->fifos[i]);

        /* check for received packet and indicate up to network */
        for (i = 0; i < config->rx_ring_num; i++) {
                struct ring_info *ring = &mac_control->rings[i];

                rx_intr_handler(ring, 0);
        }

        for (i = 0; i < config->rx_ring_num; i++) {
                struct ring_info *ring = &mac_control->rings[i];

                if (fill_rx_buffers(nic, ring, 0) == -ENOMEM) {
                        DBG_PRINT(INFO_DBG,
                                  "%s: Out of memory in Rx Netpoll!!\n",
                                  dev->name);
                        break;
                }
        }
        enable_irq(dev->irq);
}
#endif

/**
 *  rx_intr_handler - Rx interrupt handler
 *  @ring_info: per ring structure.
 *  @budget: budget for napi processing.
 *  Description:
 *  If the interrupt is because of a received frame or if the
 *  receive ring contains fresh as yet un-processed frames,this function is
 *  called. It picks out the RxD at which place the last Rx processing had
 *  stopped and sends the skb to the OSM's Rx handler and then increments
 *  the offset.
 *  Return Value:
 *  No. of napi packets processed.
 */
static int rx_intr_handler(struct ring_info *ring_data, int budget)
{
        int get_block, put_block;
        struct rx_curr_get_info get_info, put_info;
        struct RxD_t *rxdp;
        struct sk_buff *skb;
        int pkt_cnt = 0, napi_pkts = 0;
        int i;
        struct RxD1 *rxdp1;
        struct RxD3 *rxdp3;

        get_info = ring_data->rx_curr_get_info;
        get_block = get_info.block_index;
        memcpy(&put_info, &ring_data->rx_curr_put_info, sizeof(put_info));
        put_block = put_info.block_index;
        rxdp = ring_data->rx_blocks[get_block].rxds[get_info.offset].virt_addr;

        while (RXD_IS_UP2DT(rxdp)) {
                /*
                 * If your are next to put index then it's
                 * FIFO full condition
                 */
                if ((get_block == put_block) &&
                    (get_info.offset + 1) == put_info.offset) {
                        DBG_PRINT(INTR_DBG, "%s: Ring Full\n",
                                  ring_data->dev->name);
                        break;
                }
                skb = (struct sk_buff *)((unsigned long)rxdp->Host_Control);
                if (skb == NULL) {
                        DBG_PRINT(ERR_DBG, "%s: NULL skb in Rx Intr\n",
                                  ring_data->dev->name);
                        return 0;
                }
                if (ring_data->rxd_mode == RXD_MODE_1) {
                        rxdp1 = (struct RxD1 *)rxdp;
                        pci_unmap_single(ring_data->pdev, (dma_addr_t)
                                         rxdp1->Buffer0_ptr,
                                         ring_data->mtu +
                                         HEADER_ETHERNET_II_802_3_SIZE +
                                         HEADER_802_2_SIZE +
                                         HEADER_SNAP_SIZE,
                                         PCI_DMA_FROMDEVICE);
                } else if (ring_data->rxd_mode == RXD_MODE_3B) {