Merge branch 'init_rework'

[pes-rpp/rpp-lib.git] / rpp / src / rpp / eth.c

/**
 * Copyright (c) 2001-2004 Swedish Institute of Computer Science.
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without modification,
 * are permitted provided that the following conditions are met:
 *
 * 1. Redistributions of source code must retain the above copyright notice,
 *    this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright notice,
 *    this list of conditions and the following disclaimer in the documentation
 *    and/or other materials provided with the distribution.
 * 3. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT
 * SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT
 * OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
 * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
 * OF SUCH DAMAGE.
 *
 * This file is part of the lwIP TCP/IP stack.
 *
 * Author: Adam Dunkels <adam@sics.se>
 *
 */

/**
 * Copyright (c) 2010 Texas Instruments Incorporated
 *
 * This file is dervied from the "ethernetif.c" skeleton Ethernet network
 * interface driver for lwIP.
 *
 */
/* Copyright (C) 2013-2014 Czech Technical University in Prague
 *
 * Authors:
 *     - Carlos Jenkins <carlos@jenkins.co.cr>
 *     - Rostislav Lisový
 *     - Jan Doležal <pm.jenik@gmail.com>
 *
 * This document contains proprietary information belonging to Czech
 * Technical University in Prague. Passing on and copying of this
 * document, and communication of its contents is not permitted
 * without prior written authorization.
 *
 * File : eth.c
 * Abstract:
 *     Ethernet Communication RPP API implementation file. Using one PHY.
 *
 * References:
 *     eth.h
 *     RPP API documentation.
 */

#include "rpp/rpp.h"

#ifndef FREERTOS_POSIX

/* lwIP headers */
#include "lwip/init.h"
#include "lwip/timers.h" /* for DHCP binding in NO_SYS mode */
#include "lwip/sys.h" /* includes - lwip/opt.h, lwip/err.h, arch/sys_arch.h */
#include "lwip/tcpip.h" /* includes - lwip/opt.h, lwip/api_msg.h, lwip/netifapi.h, lwip/pbuf.h, lwip/api.h, lwip/sys.h, lwip/timers.h, lwip/netif.h */
#include "lwip/stats.h" /* includes - lwip/mem.h, lwip/memp.h, lwip/opt.h */
#include "netif/etharp.h" /* includes - lwip/ip.h, lwip/netif.h, lwip/ip_addr.h, lwip/pbuf.h */
#include "lwip/def.h"
#include "lwip/snmp.h"
#include "lwip/dhcp.h"
#include "lwip/autoip.h"
#include "netif/ppp_oe.h"
/* end - lwIP headers */

#include "hal/hal.h"
#include "sys/sys.h" /* includes - sys/phy_dp83848h.h */
#include "drv/emac.h"
#include "os/os.h"
#include "types.h"


/* options */

/* Number of EMAC Instances */
#define MAX_EMAC_INSTANCE         1

#define DEFAULT_PHY_ADDR          0x1
#define FIND_FIRST_PHY_ALIVE      1 /* or use default (phy_address: 1) */
#define NUM_OF_PHYs               32
#define SIZE_OF_DESC              16 /* in bytes */
#define CHANNEL                   0 /* channel number for everything - for rx, tx, unicast, broadcast or damaged frames ; there are different channels for rx and tx operations */

#define MAX_TRANSFER_UNIT         1500 /* take in account oversized frames */
#define PBUF_LEN_MAX              MAX_TRANSFER_UNIT
#define MAX_RX_PBUF_ALLOC         10
#define MIN_PKT_LEN               60

/* Define those to better describe the network interface. */
#define IFNAME0                   'e'
#define IFNAME1                   'n'


/**
 * Time to wait for autonegotiation in ticks.
 */
#define TICKS_PHY_AUTONEG         4000

/**
 * NOT IMPLEMENTED
 * When cable is connected (link status goes up) autonegotiation and/or dhcp is started.
 */
#define PHY_LINK_MONITOR_INT      0 /* turn on handling interrupt - not implemented now */
/* end options */

/* Defining interface for all the emac instances */
static struct hdkif hdkif_data[MAX_EMAC_INSTANCE];

/*
** The lwIP network interface structure for the HDK Ethernet MAC.
*/
static struct netif hdkNetIF[MAX_EMAC_INSTANCE];

/* rpp startup init indicator */
static boolean_t initialized = FALSE, postInitialized = FALSE;

boolean_t sendDone = TRUE;

/* free pbuf notification */
#if !NO_SYS
xSemaphoreHandle pbufFreed;
#endif

/**********************************************testing functions**********************************************/

boolean_t isPostInitialized()
{
        return postInitialized;
}

uint32_t rpp_eth_phylinkstat(uint32_t instNum)
{
        struct hdkif *hdkif = &hdkif_data[instNum];
        return PHY_link_status_get(hdkif->mdio_base, hdkif->phy_addr, 1);
}

#define BYTE_BUFF_SIZE 2
#define OFFSET_MAC_LETTERS (MAC_BIG_LETTERS ? 'A' : 'a')
/* puts string of MAC address assigned to EMAC instance reffered by instNum into *mac */
void rpp_eth_get_macAddrStr(uint32_t instNum, uint8_t *macStr)
{
    uint8_t index, outindex = 0;
    char buff[BYTE_BUFF_SIZE];
    struct hdkif *hdkif = &hdkif_data[instNum];
        for(index = 0; index < MAC_ADDR_LEN; index++) {
                if(index)macStr[outindex++] = ':';
        buff[0] = (hdkif->mac_addr[(MAC_ADDR_LEN - 1) - index] >> 4);
        buff[1] = (hdkif->mac_addr[(MAC_ADDR_LEN - 1) - index] & 0xf);
                macStr[outindex++] = (buff[0]<10) ? (buff[0] + '0') : (buff[0] - '\012' + OFFSET_MAC_LETTERS);
                macStr[outindex++] = (buff[1]<10) ? (buff[1] + '0') : (buff[1] - '\012' + OFFSET_MAC_LETTERS);
        }
        macStr[outindex] = '\0';
}

/* @param ip will be filled accroding to content of ipstr */
err_t rpp_eth_stringToIP(ip_addr_t * ip, uint8_t * ipstr)
{
        uint8_t charProccessed, index = 0, dotindex = 0xff, tmp = 0, dots = 0, fldEdit = 0;
        uint32_t ipaddr = 0;
        for(charProccessed = ipstr[index]; (charProccessed >= '0' && charProccessed <= '9') || charProccessed == '.' ; charProccessed = ipstr[++index])
        {
                if(charProccessed == '.')
                {
                        if(++dotindex == index)
                        {
                                dots = 0;
                                break;
                        }
                        dotindex = index;

                        ipaddr = (ipaddr << 8) + tmp;

                        dots++;
                        if(dots > 3)break;

                        tmp = 0;
                        fldEdit = FALSE;
                }
                else
                {
                        fldEdit = TRUE;
                        tmp = tmp*10 + charProccessed - '0';
                }
        }
        if(dots != 3 || !fldEdit)
        {
                /* if unsuccesful, don't modify previous content */
                return FAILURE;
        }
        ipaddr = (ipaddr << 8) + tmp;
        ip->addr = ipaddr;
        return SUCCESS;
}

/* returns number in range 0-65535 where 0 is error */
uint16_t rpp_eth_portStrToInt(uint8_t *string)
{
        uint8_t index;
        uint16_t portNO = 0;
        for(index = 0;string[index] != '\0';index++)
        {
                if(string[index] < '0' || string[index] > '9')
                {
                        portNO = 0;
                        break;
                }
                portNO = portNO * 10 + string[index] - '0';
        }
        return portNO;
}

struct netif *rpp_eth_get_netif(uint32_t instNum)
{
        return &hdkNetIF[instNum];
}

/********************************************** forward declares **********************************************/

/*
 * interface initializes
 */
err_t rpp_eth_lwip_init(struct netif *netif);

/*
 * Initializes hw such as PHY, MDIO, EMAC, EMAC control module
 *
 * @return SUCCESS if initialization successful.\n
 *         FAILURE if initialization not successful.
 */
err_t rpp_eth_hw_init(struct hdkif *hdkif);

/*
 * Initializes hw, after lwIP was initialized and
 * OS was initialized in case OS is used.
 */
err_t rpp_eth_hw_init_postInit(struct netif *netif);

/********************************************** utility functions **********************************************/

/*
* Function to set the MAC address to the interface
* @param   inst_num the instance number
*
* @note    mac_addr[0] is considered MSB
*/
void
hdkif_macaddrset(u32_t inst_num, u8_t *mac_addr) {
  struct hdkif *hdkif;
  u32_t temp;

  hdkif = &hdkif_data[inst_num];

  /* set MAC hardware address */
  for(temp = 0; temp < MAC_ADDR_LEN; temp++) {
    hdkif->mac_addr[temp] = mac_addr[(MAC_ADDR_LEN - 1) - temp];
  }
#ifdef DEBUG
  uint8_t macStr[18];
  rpp_eth_get_macAddrStr(inst_num, macStr);
  rpp_debug_printf("Setting MAC... %s\r\n", macStr);
#endif
}

/**
* Function to setup the instance parameters inside the interface
* @param   hdkif
* @return  none.
*/
static void
hdkif_inst_config(struct hdkif *hdkif) {
  if(hdkif->inst_num == 0)
  {
    hdkif->emac_base = EMAC_BASE_m(0);
    hdkif->emac_ctrl_base = EMAC_CTRL_BASE_m(0);
    hdkif->emac_ctrl_ram = EMAC_CTRL_RAM_BASE_m(0);
    hdkif->mdio_base = MDIO_BASE_m(0);
    hdkif->phy_addr = DEFAULT_PHY_ADDR; /* Default address of PHY on "MDIO bus" (depends on PHY hw configuration) */
    hdkif->phy_autoneg = PHY_auto_negotiate;
    hdkif->phy_autoneg_start = PHY_start_auto_negotiate;
    hdkif->phy_autoneg_is_done = PHY_is_done_auto_negotiate;
    hdkif->phy_partnerability = PHY_partner_ability_get;
  }
}

/********************************************** initializing functions **********************************************/

int8_t rpp_eth_init()
{
    uint8_t instNum;
        int8_t retVal = SUCCESS;
    if(initialized) {
        return FAILURE;
    }

    /* config of EMAC instances and other stuff */
    for(instNum = 0; instNum < MAX_EMAC_INSTANCE; instNum++)
    {
        hdkif_data[instNum].inst_num = instNum;
        hdkif_inst_config(&hdkif_data[instNum]); /* hdkif_inst_config must contain instNum specific settings */
        if(retVal = rpp_eth_hw_init(&hdkif_data[instNum]))
        {
#ifdef DEBUG
                rpp_sci_printk("rpp_eth_hw_init: %d", retVal);
#endif
                retVal = FAILURE;
        }
    }

    /* when one of instances was not initialized correctly return FAILURE without setting initialized variable to TRUE */
    if(retVal)return FAILURE;

#if NO_SYS

#else /* NO_SYS */

#endif /* NO_SYS */

    initialized = TRUE;

    return retVal;
}

int8_t rpp_eth_init_postInit(uint32_t instNum, uint8_t *macArray)
{
    if(postInitialized) {
        return FAILURE;
    }
    int8_t retVal = SUCCESS;
    struct netif *netif = &hdkNetIF[instNum];
    u8_t mac_addr[MAC_ADDR_LEN] = RPP_MAC_ADDR;

    /* ----- lwIP stuff ----- */
    struct ip_addr ip_addr;
    struct ip_addr net_mask;
    struct ip_addr gw_addr;
    volatile unsigned int dhcpBindWait = 0x3FFFFFFF;

    if(macArray == NULL)
    {
        macArray = mac_addr; /* use default MAC */
    }
    hdkif_macaddrset(instNum, macArray);

    struct hdkif *hdkif;
#if NO_SYS
    lwip_init();
#else /* NO_SYS */
    /* this can be called only within post OS init */
    tcpip_init(NULL,NULL); /* calls lwip_init() implicitly, starts lwIP task (thread), when started function given to tcpip_init is executed first */
#endif /* NO_SYS */

#if STATIC_IP_ADDRESS
    ip_addr.addr = htonl(RPP_IP_ADDR);
    net_mask.addr = htonl(RPP_NETMASK);
    gw_addr.addr = htonl(RPP_GW);
#else /* STATIC_IP_ADDRESS */
    ip_addr.addr = 0;
    net_mask.addr = 0;
    gw_addr.addr = 0;
#endif /* STATIC_IP_ADDRESS */

    /* init and add new netif */
    /* add new network interface to lwIP list of ifaces and initialize netif with specific function */
#if NO_SYS
    if( netif_add(netif, &ip_addr, &net_mask, &gw_addr, (void *) instNum, rpp_eth_lwip_init, ethernet_input) == NULL )
#else /* NO_SYS */
    if( netif_add(netif, &ip_addr, &net_mask, &gw_addr, (void *) instNum, rpp_eth_lwip_init, tcpip_input) == NULL )
#endif /* NO_SYS */
    {
        return NETIF_ADD_ERR;
    }

    netif_set_default(netif);

    hdkif = (struct hdkif *) netif->state;
#if !NO_SYS
    /* ----- freeRTOS elements ----- */
    /* semaphore blocking receive task (rpp_eth_recv_raw) from receive operation, till RX interrupt occurs and notify it */
    vSemaphoreCreateBinary(hdkif->goRX);
    vSemaphoreCreateBinary(hdkif->goTX);
        /* create semaphore notificating that pbuf was freed */
        vSemaphoreCreateBinary(pbufFreed);
    /* run task rpp_eth_recv_raw */
    xTaskCreate( rpp_eth_recv_raw, "RXHandler", 200, netif, 0, NULL );
    /* ----- end - freeRTOS ----- */
#endif /* !NO_SYS */

    /* if we don't use link int change, then it must be done here */
#if !PHY_LINK_MONITOR_INT
    if(rpp_eth_phylinkstat(hdkif->inst_num))
    {
        rpp_debug_printf((const char *)"cable connected ... setting IP params\r\n");
#if STATIC_IP_ADDRESS
        netif_set_up(netif);
#elif LWIP_DHCP /* STATIC_IP_ADDRESS-LWIP_DHCP */
        if(dhcp_start(netif) != ERR_OK)
        {
                rpp_debug_printf("dhcp mem err\r\n");
                return DHCP_MEM_ERR;
        }
        rpp_debug_printf("binding DHCP\r");
        while((netif->dhcp->state != DHCP_BOUND) && (dhcpBindWait--))/*sys_check_timeouts()*/;
        if(!dhcpBindWait)
                rpp_debug_printf("dhcp binding timeout...\r\n");
#else /* LWIP_DHCP-LWIP_AUTOIP FIXME: there should be some kind of waiting till ip is assigned */
        autoip_start(netif);
#endif /* STATIC_IP_ADDRESS-LWIP_DHCP-LWIP_AUTOIP */
#ifdef DEBUG
        uint8_t ipString[16];
        rpp_eth_getIPDecimalStr(netif->ip_addr, ipString);
        rpp_debug_printf("Address: %s\r\n", ipString);
        rpp_eth_getIPDecimalStr(netif->netmask, ipString);
        rpp_debug_printf("Netmask: %s\r\n", ipString);
        rpp_eth_getIPDecimalStr(netif->gw, ipString);
        rpp_debug_printf("Gateway: %s\r\n", ipString);
#endif
    }
    else
    {
        rpp_debug_printf((const char *)"cable not connected\r\n");
        retVal = PHY_LINK_DOWN;
    }

#else /* !PHY_LINK_MONITOR_INT */
    /* now when we established environment needed for phy link status change we can allow it */
    /* set PHY number which is monitored for link changes in MDIO and enable interrupt */
    HWREG(hdkif->mdio_base + MDIO_USERPHYSEL0) = ((hdkif->phy_addr && 0x1f) | MDIO_USERPHYSEL0_LINKINTENB);

    /* enable MISC interrupt - link monitoring in EMAC controle module */
    HWREG(hdkif->emac_ctrl_base + EMAC_CTRL_CnMISCEN(0)) |= EMAC_CTRL_MISC_LINKINT0ENB;

#endif /* !PHY_LINK_MONITOR_INT */
    /* ----- end - lwIP stuff ----- */

    postInitialized = TRUE;
        return retVal;
}

err_t rpp_eth_lwip_init(struct netif *netif)
{
#if LWIP_NETIF_HOSTNAME
    /* Initialize interface hostname */
    netif->hostname = "rpp";
#endif /* LWIP_NETIF_HOSTNAME */

    uint32_t instNum = (uint32_t)netif->state;
    /*netif->num = instNum; - auto-initiated when netif_add is called */

    /* netif->state contained instNum, we replace it with corresponding hdkif */
    struct hdkif *hdkif = &hdkif_data[instNum];
    netif->state = hdkif;

    netif->name[0] = IFNAME0;
    netif->name[1] = IFNAME1;

#if !NO_SYS
    hdkif->waitTicksForPHYAneg = TICKS_PHY_AUTONEG;
#endif

    /*
     * Initialize the snmp variables and counters inside the struct netif.
     * The last argument should be replaced with your link speed, in units
     * of bits per second.
    */
    NETIF_INIT_SNMP(netif, snmp_ifType_ethernet_csmacd, 10000000);

    /* We directly use etharp_output() here to save a function call.
     * You can instead declare yo_SKIP_TO_HWur own function an call etharp_output()
     * from it if you have to do some checks before sending (e.g. if link
     * is available...)
     */
    netif->output = etharp_output;
    netif->linkoutput = rpp_eth_send;
    return rpp_eth_hw_init_postInit(netif);
}

#define INIT_ONLY_AFTER_RESET 1
err_t rpp_eth_hw_init(struct hdkif *hdkif)
{
        uint8_t index;
    uint16_t regContent;
    uint32_t physAlive;

    /* Deactivate reset pin of PHY */
    /* for hw reset of PHY, it is necessary that PIN_NAME_ETHRST is 1us logical low state, before putting it to logical high */
#if !INIT_ONLY_AFTER_RESET
#if NO_SYS
        uint8_t index = 80; /* Initially used to hw reset of PHY connected to GIO pin. This means 1us - for 80MHz clock. */
#else /* NO_SYS */
    uint8_t index = configCPU_CLOCK_HZ/1000000; /* Initially used to hw reset of PHY connected to GIO pin 'rpp project only'; 1us - according to PHY specification. */
#endif /* NO_SYS */
    hal_gpio_pin_set_value(*hal_gpio_pin_get_dsc(PIN_NAME_ETHRST,-1),0);
    while(index--);
#endif /* !INIT_ONLY_AFTER_RESET */
    /* we have pull-down resistor, so after reset, we only need to put ETHRST pin to log. high */
    hal_gpio_pin_set_value(*hal_gpio_pin_get_dsc(PIN_NAME_ETHRST,-1),1);

    /* initializes EMAC control module and EMAC module */
    EMACInit(hdkif->emac_ctrl_base, hdkif->emac_base);
    /* initializes MDIO module (reset) */
    MDIOInit(hdkif->mdio_base, 0x0, 0x0);

    /* enable used receive channel if necessary */


    /* try to read random register from defaultPhy to make MDIO fill alive bit for this one if it returned
       ACK bit in msg response - this must be done, because MDIO has not set alive bit of PHY after that
       short time after MDIOInit() */
        MDIOPhyRegRead(hdkif->mdio_base, hdkif->phy_addr, PHY_BMSR, &regContent);

    /* find first alive PHY - or use default if alive */
    physAlive = MDIOPhyAliveStatusGet(hdkif->mdio_base);
    if(!(physAlive & (1 << hdkif->phy_addr)))
    {
#if FIND_FIRST_PHY_ALIVE
        for(index = 0;index < NUM_OF_PHYs;index++) /* i..PHY tested */
        {
            if(physAlive && (1 << index))
            {
                hdkif->phy_addr = index;
                break;
            }
            else
            {
                MDIOPhyRegRead(hdkif->mdio_base, index, PHY_BMCR, &regContent); /* try to 'wake up' PHY on 'index' address by reading random register, making MDIO set alive bit for current PHY */
                physAlive = MDIOPhyAliveStatusGet(hdkif->mdio_base); /* get updated register */
                if(physAlive && (1 << index))
                {
                    hdkif->phy_addr = index;
                    break;
                }
            }
        }
        if(!physAlive)
        {
            rpp_debug_printf("no phy found, phys: %d\r\n", physAlive);
                return NO_PHY_ALIVE;
        }
#else
        rpp_debug_printf("default phy not alive\r\n");
        return DFLT_PHY_NOT_ALIVE;
#endif
    }

    /* start autonegotiation and check on completion later or when complete link register will be updated */
    hdkif->phy_autoneg_start(hdkif->mdio_base, hdkif->phy_addr, PHY_100BASETXDUPL_m | PHY_100BASETX_m | PHY_10BASETDUPL_m | PHY_10BASET_m);

    /* TODO: you can implement init of receive flow control somewhere here if desired - set RXBUFFERFLOWEN in MACCONTROL */


    /* acknowledge EMAC control module RX, TX and MISC interrupts */
        EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_RX);
        EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_TX);
        EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_MISC);

    /* sets which channel will receive broadcasts */
        EMACRxBroadCastEnable(hdkif->emac_base, CHANNEL);
        /* sets channel where all frames will be copied to - either with MAC address different from local device address, either packets with error will be copied; appropriate error will be set in the frame EOP buffer descriptor */
        EMACRxPromiscEnable(hdkif->emac_base, CHANNEL);
        /* enables unicast */
        EMACRxUnicastSet(hdkif->emac_base, CHANNEL);

    /* enable TX and RX interrupts in both EMAC module and EMAC control module */
        EMACTxIntPulseEnable(hdkif->emac_base, hdkif->emac_ctrl_base, 0, CHANNEL);
        EMACRxIntPulseEnable(hdkif->emac_base, hdkif->emac_ctrl_base, 0, CHANNEL);

        /* switch of send bd handling and when sending, just wait till send was done and then return (making the driver simple) */
        //vim_mask_clr(TXinterruptVectorNumber);

        return SUCCESS;
}

err_t rpp_eth_hw_init_postInit(struct netif *netif)
{
        volatile unsigned int autonegFinishWait = 0x3FFFFFFF; /* 0x3FFFFFFF is for 80MHz aproximately 13s */
    uint16_t regContent;
    uint32_t num_bd, pbuf_cnt = 0;
    volatile struct emac_tx_bd *curr_txbd, *last_txbd;
    volatile struct emac_rx_bd *curr_rxbd, *last_rxbd;
    struct pbuf *p, *q;
    struct rxch *rxch;
    struct txch *txch;

    struct hdkif *hdkif = (struct hdkif *) netif->state;

    rxch = &(hdkif->rxch);
    txch = &(hdkif->txch);

    rpp_debug_printf("autoneg started - check on cable if it's connected!\r\n");


        /*
        * Initialize the Descriptor Memory For TX and RX
        * Only Channel 0 is supported for both TX and RX
        */
        txch->free_head = ((volatile struct emac_tx_bd *) hdkif->emac_ctrl_ram);
        txch->next_bd_to_process = txch->free_head;
        txch->active_tail = NULL;

        /* Set the number of descriptors for the channel */
        num_bd = (SIZE_EMAC_CTRL_RAM >> 1) / sizeof(struct emac_tx_bd); /* take half of CPPI ram for TX BDs */

        curr_txbd = txch->free_head;

        /* Initialize all the TX buffer Descriptors */
        while(num_bd--) {
            curr_txbd->next = curr_txbd + 1;
            curr_txbd->flags_pktlen = 0;
            last_txbd = curr_txbd;
            curr_txbd = curr_txbd->next;
        }
        last_txbd->next = txch->free_head;

        /* Initialize the descriptors for the RX channel */
        rxch->active_head = (volatile struct emac_rx_bd*)(curr_txbd);
        rxch->free_head = NULL;
        rxch->freed_pbuf_len = 0;
        num_bd = ((SIZE_EMAC_CTRL_RAM >> 1) / sizeof(struct emac_rx_bd)); /* when using 20B emac_bd structs, u can allocate one more ;) ( 8192/20 = 409 together rx and tx and there is still 12 bytes remaining) */
        curr_rxbd = rxch->active_head;
        last_rxbd = curr_rxbd;

        /*
        ** Allocate the pbufs for the maximum count permitted or till the
        ** number of buffer desceriptors expire, which ever is earlier.
        */
        while(pbuf_cnt < MAX_RX_PBUF_ALLOC) {
            p = pbuf_alloc(PBUF_RAW, PBUF_LEN_MAX, PBUF_POOL);

            if(p != NULL) {
                /* write the descriptors if there are enough numbers to hold the pbuf*/
                if(((uint32_t)pbuf_clen(p)) <= num_bd) {
                    for(q = p; q != NULL; q = q->next) {
                        curr_rxbd->bufptr = (uint8_t *)(q->payload);
                        curr_rxbd->bufoff_len = q->len;
                        curr_rxbd->next = curr_rxbd + 1;
                        curr_rxbd->flags_pktlen = EMAC_DSC_FLAG_OWNER;

                        /* Save the pbuf */
                        curr_rxbd->pbuf = q;
                        last_rxbd = curr_rxbd;
                        curr_rxbd = curr_rxbd->next;
                        num_bd--;
                    }
                }

                /* free the allocated pbuf if no free descriptors are left */
                else {
                    pbuf_free(p);
                    break;
                }
            }
            else {
                break;
            }
            pbuf_cnt++;
        }

        if(!pbuf_cnt)rpp_debug_printf("no pbufs attached to rx buffer descriptors during init\n");

        last_rxbd->next = NULL;
        rxch->active_tail = last_rxbd;

        num_bd = ( ((uint32_t)rxch->active_tail - (uint32_t)rxch->active_head) / sizeof(struct emac_rx_bd) + 1 );
        rpp_debug_printf("%d pbuf chains allocated for %d rx buffer descriptors\n", pbuf_cnt, num_bd);
        /* for flow control or QoS - there could be set an interrupt when we reach preset amount of receive descriptors remaining - see RXBUFFERFLOWEN */
        /* EMACNumFreeBufSet(hdkif->emac_base, CHANNEL, num_bd); */ /* not used */

        /* set header descriptor pointers - this shows EMAC which descriptor is beginning one for writing received frames/packets  */
        EMACRxHdrDescPtrWrite(hdkif->emac_base, (uint32_t)rxch->active_head, CHANNEL);

        /* lwIP stuff */

        /* set length of MAC address */
        netif->hwaddr_len = MAC_ADDR_LEN;

        /* set MAC address */
        for(regContent = 0; regContent < MAC_ADDR_LEN; regContent++) {
            netif->hwaddr[regContent] = hdkif->mac_addr[(MAC_ADDR_LEN - 1) - regContent];
        }

        /* maximum transfer unit */
        netif->mtu = MAX_TRANSFER_UNIT;

        /* device capabilities */
        /* don't set NETIF_FLAG_ETHARP if this device is not an ethernet one */
        netif->flags = NETIF_FLAG_BROADCAST | NETIF_FLAG_ETHARP | NETIF_FLAG_LINK_UP;

        /* lwIP stuff - end */

    /* for flow control frames */
    EMACMACSrcAddrSet(hdkif->emac_base, hdkif->mac_addr);

    /*  Be sure to program all eight MAC address registers -
     *  whether the receive channel is to be enabled or not.
     */
    for (regContent = 0; regContent < 8; regContent++) /* i..channel_number */
        EMACMACAddrSet(hdkif->emac_base, regContent, hdkif->mac_addr, EMAC_MACADDR_NO_MATCH_NO_FILTER);

#if !PHY_LINK_MONITOR_INT /* in case we don't use interrupt when phy link status changes, we need to try to finish autoneg in here */
    /* wait for autonegotiation to be done or continue, when delay was reached */
#if !NO_SYS
    uint32_t timeToWake = hdkif->waitTicksForPHYAneg + sys_jiffies();
    while(hdkif->phy_autoneg_is_done(hdkif->mdio_base, hdkif->phy_addr) == FALSE && timeToWake > sys_jiffies())vTaskDelay(20); /* XXX: if init is not done at the startup, but couple days later, this might cause troubles */
#else
    while(hdkif->phy_autoneg_is_done(hdkif->mdio_base, hdkif->phy_addr) == FALSE && autonegFinishWait--); /* wait till aneg done */
#endif

    if(hdkif->phy_autoneg_is_done(hdkif->mdio_base, hdkif->phy_addr) != FALSE)
        rpp_debug_printf("aneg finished \r\n");
    else
        rpp_debug_printf("aneg timeout \r\n");

    /* provide informations retrieved from autoneg to EMAC module */
    hdkif->phy_partnerability(hdkif->mdio_base, hdkif->phy_addr, &regContent);
    if (regContent & (PHY_100BASETXDUPL_m | PHY_10BASETDUPL_m)) {
        EMACDuplexSet(hdkif->emac_base, EMAC_DUPLEX_FULL);
        /* this is right place to implement transmit flow control if desired - set TXFLOWEN in MACCONTROL */
    } else if (regContent & (PHY_100BASETX_m | PHY_10BASET_m)) {
        EMACDuplexSet(hdkif->emac_base, EMAC_DUPLEX_HALF);
    } else {
        rpp_debug_printf("Unknown duplex mode\r\n");
        return UNKN_DUPLEX_MODE;
    }
#endif /* !PHY_LINK_MONITOR_INT */

    /* enable hostpend interrupts in emac module */
    HWREG(hdkif->emac_base + EMAC_MACINTMASKSET) |= EMAC_MACINTMASKSET_HOSTMASK;

    /* enable hostpend interrupts in emac control module */
    HWREG(hdkif->emac_ctrl_base + EMAC_CTRL_CnMISCEN(0)) |= EMAC_CTRL_MISC_HOSTPENDENB;

        /* enable EMAC's Media Independent Interface TX and RX */
        EMACMIIEnable(hdkif->emac_base);
        /* enable EMAC transmit */
        EMACTxEnable(hdkif->emac_base);
        /* enable EMAC receive */
        EMACRxEnable(hdkif->emac_base);

        return SUCCESS;
}














/********************************************** send and receive functions / ISRs **********************************************/

err_t rpp_eth_send(struct netif * netif, struct pbuf *p)
{
        err_t retVal = SUCCESS;
        SYS_ARCH_DECL_PROTECT(lev);

        /**
         * This entire function must run within a "critical section" to preserve
         * the integrity of the transmit pbuf queue.
         *
         */
        SYS_ARCH_PROTECT(lev);

        /* protect from access receive acces from interrupt */
#if !SYS_LIGHTWEIGHT_PROT
        uint32_t prevProt = (uint32_t) _get_CPSR() & 0x80;
        _disable_IRQ();
#endif


        /* adjust the packet length if less than minimum required */
        if(p->tot_len < MIN_PKT_LEN) {
           p->tot_len = MIN_PKT_LEN;
           p->len = MIN_PKT_LEN;
        }

        /* call the actual transmit function */
        retVal = rpp_eth_send_raw(netif, p);

        /* Return to prior interrupt state and return. */
        SYS_ARCH_UNPROTECT(lev);

#if !SYS_LIGHTWEIGHT_PROT
    if(!prevProt)_enable_IRQ();
#endif

        return retVal;
}

/**
 * send packet, blocks till it is sent
 * every send uses bds on the beginning of CPPI RAM
 */
err_t rpp_eth_send_raw(struct netif *netif, struct pbuf *p)
{
        struct pbuf *q;
        volatile struct emac_tx_bd *curr_bd, *bd_end;
        struct hdkif *hdkif;

    hdkif = (struct hdkif *) netif->state;
        curr_bd = (struct emac_tx_bd *) hdkif->emac_ctrl_ram;

        /* destroy all flags previously set and update the total packet length */
        curr_bd->flags_pktlen = p->tot_len;

        /* Indicate the start of the packet */
        curr_bd->flags_pktlen |= (EMAC_DSC_FLAG_SOP | EMAC_DSC_FLAG_OWNER);

        /* Copy pbuf information into TX buffer descriptors */
        for(q = p; q != NULL; q = q->next) {

            /* Intialize the buffer pointer and length */
            curr_bd->bufptr = (uint8_t *)(q->payload);
            curr_bd->bufoff_len = (q->len) & 0xFFFF;
            bd_end = curr_bd;
            curr_bd->pbuf = p;
            curr_bd = curr_bd->next;
            curr_bd->flags_pktlen = 0x0; /* destroy all previously set flags */
        }

        /* Indicate the end of the packet */
        bd_end->next = NULL;
        bd_end->flags_pktlen |= EMAC_DSC_FLAG_EOP;

    /* don't write to hdp till it turns zero */
    while (0 != *((volatile uint32_t *) (hdkif->emac_base + EMAC_TXHDP(CHANNEL)) ));
    /* notify hw where to start transmission from */
    HWREG(hdkif->emac_base + EMAC_TXHDP(CHANNEL)) = hdkif->emac_ctrl_ram;
        //EMACTxHdrDescPtrWrite(hdkif->emac_base, hdkif->emac_ctrl_ram, CHANNEL);

        /* wait till queued pkt is sent */
    vim_mask_set(TXinterruptVectorNumber);
        xSemaphoreTake(hdkif->goTX, portMAX_DELAY);
        while (!(bd_end->flags_pktlen & EMAC_DSC_FLAG_EOQ) || (((struct emac_tx_bd *)(hdkif->emac_ctrl_ram))->flags_pktlen & EMAC_DSC_FLAG_OWNER))rpp_sci_printk("!");

        EMACTxCPWrite(hdkif->emac_base, CHANNEL, (uint32_t) bd_end);

    EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_TX);

        /* restore bds chain */
        bd_end->next = curr_bd;

    LINK_STATS_INC(link.xmit);
        return SUCCESS;
}

void rpp_eth_recv_raw(void *arg)
{
    volatile struct emac_rx_bd *curr_bd, *processed_bd, *curr_tail, *last_bd;
    volatile struct pbuf *pbuf, *q, *new_pbuf;
    uint32_t ex_len, len_to_alloc;
    uint16_t tot_len;
        struct netif *netif = (struct netif *) arg;
        struct hdkif *hdkif = (struct hdkif *) netif->state;
        struct rxch *rxch;

        rxch = &(hdkif->rxch);


#if !NO_SYS && !ONE_DRIVER_TASK
        SYS_ARCH_DECL_PROTECT(lev);
    for(;;)
    {
        vim_mask_set(RXinterruptVectorNumber);
        sys_arch_sem_wait(&(hdkif->goRX), 0);
        SYS_ARCH_PROTECT(lev);
#endif


            /* Get the bd which contains the earliest filled data */
            curr_bd = rxch->active_head;
            last_bd = rxch->active_tail;

            if(curr_bd->flags_pktlen & EMAC_DSC_FLAG_SOP != EMAC_DSC_FLAG_SOP)rpp_debug_printf("ERR: Active head is not SOP!\n");

            /**
             * Process the descriptors as long as data is available
             * when the DMA is receiving data, SOP flag will be set
             */
            while(curr_bd->flags_pktlen & EMAC_DSC_FLAG_SOP)
        {
                ex_len = 0;
                len_to_alloc = 0;
                /* Start processing once the packet is loaded */
                if((curr_bd->flags_pktlen & EMAC_DSC_FLAG_OWNER) != EMAC_DSC_FLAG_OWNER)
            {
                        if(rxch->free_head == NULL)
                        {
                                /* this bd chain will be freed after processing */
                            rxch->free_head = curr_bd;
                        }

                        /* Get the total length of the packet. curr_bd points to the start
                     * of the packet.
                     */
                    tot_len = (curr_bd->flags_pktlen) & 0xFFFF;

                    /* Get the start of the pbuf queue */
                    q = curr_bd->pbuf;

                    do {
                        /* Get the pbuf pointer which is associated with the current bd */
                        pbuf = curr_bd->pbuf;

                        /* If the earlier pbuf ended, update the chain */
                        if(pbuf->next == NULL)
                        {
                            pbuf->next = (struct pbuf*)(curr_bd->next)->pbuf;
                        }

                        len_to_alloc += pbuf->len;
                        /* Update the len and tot_len fields for the pbuf in the chain*/
                        pbuf->len = (curr_bd->bufoff_len) & 0xFFFF;
                        pbuf->tot_len = tot_len - ex_len ;
                        processed_bd = curr_bd;
                        ex_len += pbuf->len;
                        curr_bd = curr_bd->next;
#ifdef DEBUG
                        if(curr_bd == NULL)rpp_debug_printf("rx curr_bd set to zero!\n");
#endif
                        /* FIXME: curr_bd here could be NULL - (currently solved by having always allocated enough pbufs -> waiting for pbuf_free() function) a solution could be: we need to keep at least TCP_MSS bytes, so allocate at least that if not possible block */
                    } while((processed_bd->flags_pktlen & EMAC_DSC_FLAG_EOP) != EMAC_DSC_FLAG_EOP);

                    /**
                     * Close the chain for this pbuf. A full packet is received in
                     * this pbuf chain. Now this pbuf can be given to upper layers for
                     * processing. The start of the pbuf chain is now 'q'.
                    */
                    pbuf->next = NULL;

                    /* Adjust the link statistics */
                    LINK_STATS_INC(link.recv);

                    /* Process the packet - this function is specified while adding netif using netif_add() function */
                    if(netif->input((struct pbuf *)q, netif) != ERR_OK)
                    {
                        /* Adjust the link statistics */
                        LINK_STATS_INC(link.memerr);
                        LINK_STATS_INC(link.drop);
                    }

                    /* Acknowledge that this packet is processed */
                    EMACRxCPWrite(hdkif->emac_base, CHANNEL, (unsigned int)processed_bd);

                    if(curr_bd != NULL)
                    {
                        rxch->active_head = curr_bd;
                    }
                    else
                    {
                        rxch->active_head = processed_bd + 1; //todo:reconsider
                    }

                    /**
                     * The earlier pbuf chain is freed from the upper layer. So, we need to
                     * allocate a new pbuf chain and update the descriptors with the pbuf info.
                     * To support chaining, the total length freed by the upper layer is tracked.
                     * Care should be taken even if the allocation fails.
                     */
                    /**
                     * now len_to_alloc will contain the length of the pbuf which was freed
                     * from the upper layer
                     */
                    rxch->freed_pbuf_len += len_to_alloc;
#if !NO_SYS
                    /* if there is not enough pbufs predefined, we need application to process them and return them back to pool -> YIELD */
                    /* XXX: this is not fixed in NO_SYS version - problem is when curr_bd is set to NULL (see above), and we don't obtain new_pbuf
                     * from pool, then in *bd pointing to NULL might be such values (EMAC_DSC_FLAG_SOP == 1 &| EMAC_DSC_FLAG_OWNER == 0), that we
                     * stay cycling around */
                    do
                    {
#endif
                            new_pbuf = pbuf_alloc(PBUF_RAW, (rxch->freed_pbuf_len), PBUF_POOL);
#if !NO_SYS
                            if(new_pbuf != NULL)
                                break;
                            else
                                xSemaphoreTake(pbufFreed, 0);
                                xSemaphoreTake(pbufFreed, portMAX_DELAY);
                    }while(1);
#endif

                    /* Write the descriptors with the pbuf info till either of them expires */
                    if(new_pbuf != NULL)
                    {
                        curr_bd = rxch->free_head;

                        for(q = new_pbuf; (q != NULL) && (curr_bd != rxch->active_head); q = q->next)
                        {
                            curr_bd->bufptr = (uint8_t *)(q->payload);

                            /* no support for buf_offset. RXBUFFEROFFSET register is 0 */
                            curr_bd->bufoff_len = (q->len) & 0xFFFF;
                            curr_bd->flags_pktlen = EMAC_DSC_FLAG_OWNER;

                            rxch->freed_pbuf_len -= q->len;

                            /* Save the pbuf */
                            curr_bd->pbuf = q;
                            last_bd = curr_bd;
                            curr_bd = curr_bd->next;
                        }

                        /**
                         * At this point either pbuf expired or no rxbd to allocate. If
                         * there are no, enough rx bds to allocate all pbufs in the chain,
                         * free the rest of the pbuf
                         */
                        if(q != NULL)
                        {
                            pbuf_free((struct pbuf *)q);
                        }

                        curr_tail = rxch->active_tail;
                        last_bd->next = NULL;

                        curr_tail->next = rxch->free_head;

                        /**
                         * Check if the reception has ended. If the EOQ flag is set, the NULL
                         * Pointer is taken by the DMA engine. So we need to write the RX HDP
                         * with the next descriptor.
                         */
                        if(curr_tail->flags_pktlen & EMAC_DSC_FLAG_EOQ)
                        {
                            EMACRxHdrDescPtrWrite(hdkif->emac_base, (u32_t)(rxch->free_head), CHANNEL);
                        }

                        rxch->free_head  = curr_bd;
                        rxch->active_tail = last_bd;
                    }
            }
                curr_bd = rxch->active_head;
            }
                EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_RX);

#if !NO_SYS && !ONE_DRIVER_TASK
                SYS_ARCH_UNPROTECT(lev);
    }
#endif
}

void RxIntHandler(u32_t instNum)
{
#if !NO_SYS
        vim_mask_clr(RXinterruptVectorNumber);
        /*
        static portBASE_TYPE xHigherPriorityTaskWoken;
        xHigherPriorityTaskWoken = pdFALSE;
        */
        /* 'Give' the semaphore to unblock the task. */
        xSemaphoreGiveFromISR(hdkif_data[instNum].goRX, NULL); // whenever there is time to process received packets, wake up task processing them
        /*
    if( xHigherPriorityTaskWoken == pdTRUE )
    {
    */
    /* Giving the semaphore unblocked a task, and the priority of the
    unblocked task is higher than or equal to the currently running task
    - force a context switch to ensure that the interrupt returns directly
    to the unblocked (higher priority) task.
    NOTE: The actual macro to use to force a context switch from an
    ISR is dependent on the port. This is the correct macro for the
    Open Watcom DOS port. Other ports may require different syntax.
        Refer to the examples provided for the port being used to determine
    the syntax required. */
        /*
    portSWITCH_CONTEXT();
    }
        */
#else
        rpp_eth_recv_raw((void *) &hdkNetIF[instNum]);
#endif
}

void TxIntHandler(u32_t instNum)
{
        //struct hdkif *hdkif = &hdkif_data[instNum];
#if !NO_SYS
        vim_mask_clr(TXinterruptVectorNumber); /* see sys_startup.c */
        xSemaphoreGiveFromISR(hdkif_data[instNum].goTX, NULL);
#else
        //sendDone = TRUE;
        //rpp_eth_send_bd_handler((void *) &hdkif_data[instNum]);
#endif
}

/**
 * TX err codes:
 *
 * 0 No error
 * 1h SOP error; the buffer is the first buffer in a packet, but the SOP bit is not set in software.
 * 2h Ownership bit not set in SOP buffer
 * 3h Zero next buffer descriptor pointer without EOP
 * 4h Zero buffer pointer
 * 5h Zero buffer length
 * 6h Packet length error (sum of buffers is less than packet length)
 *
 *
 * RX err codes:
 *
 * 0 No error
 * 1h Reserved
 * 2h Ownership bit not set in SOP buffer
 * 3h Reserved
 * 4h Zero buffer pointer
 */
boolean_t HostPendErrHandler(void)
{
        uint8_t index = MAX_EMAC_INSTANCE, errFound = FALSE;
        uint32_t reg;
        struct hdkif *hdkif;

        while(index){
                hdkif = &hdkif_data[--index];
            if(EMACIntVectorGet(hdkif->emac_base) & EMAC_MACINVECTOR_HOSTPEND)errFound = TRUE;
        }
        if(!errFound)return FALSE; /* this is not the cause of the interrupt */
        rpp_sci_printk("HOSTPEND err\n");
        reg = HWREG(hdkif->emac_base + EMAC_MACSTATUS);
        rpp_sci_printk("TXCHERR: %d at CH: %d\n", ((reg >> EMAC_MACSTATUS_TXERRCODE_SHIFT) & 0x7), ((reg >> EMAC_MACSTATUS_TXERRCH_SHIFT) & 0x7));
        rpp_sci_printk("RXCHERR: %d at CH: %d\n", ((reg >> EMAC_MACSTATUS_RXERRCODE_SHIFT) & 0x7), ((reg >> EMAC_MACSTATUS_RXERRCH_SHIFT) & 0x7));
        /* no acknowledge - emac module has to be restarted */
        return TRUE; /* this was the reason of interrupt */
}

#if PHY_LINK_MONITOR_INT
boolean_t LinkIntHandler(void)
{
        uint8_t index = MAX_EMAC_INSTANCE, phyFound = FALSE;
        struct hdkif *hdkif;
        uint16_t regContent;
        volatile unsigned int autonegFinishWait = 0xFFFFFFF;
    volatile unsigned int dhcpBindWait = 0x3FFFFFFF;

        /* check each instance, whether this interrupt was meant for this function, if not return FALSE so other function may be tried */
        while(index){
        hdkif = &hdkif_data[--index];
            if( hdkif->phy_addr == (HWREG(hdkif->mdio_base + MDIO_USERPHYSEL0) & 0x1f) && EMACIntVectorGet(hdkif->emac_base) & EMAC_MACINVECTOR_LINKINT0 )phyFound = TRUE;
        }
    if(!phyFound)return FALSE;
    struct netif *netif = &hdkNetIF[hdkif->inst_num];

    /* we handles here connection of cable after startup, not changes of linkstatus */

    /* wait for autonegotiation to be done */
#if ONCE_LINK_SETUP
    while(hdkif->phy_autoneg_is_done(hdkif->mdio_base, hdkif->phy_addr) == FALSE);
#else /* ONCE_LINK_SETUP */
    while(hdkif->phy_autoneg_is_done(hdkif->mdio_base, hdkif->phy_addr) == FALSE & autonegFinishWait--);
#endif /* ONCE_LINK_SETUP */

    /* provide informations retrieved from autoneg to EMAC module */
    hdkif->phy_partnerability(hdkif->mdio_base, hdkif->phy_addr, &regContent);
    if (regContent & (PHY_100BASETXDUPL_m | PHY_10BASETDUPL_m)) {
        EMACDuplexSet(hdkif->emac_base, EMAC_DUPLEX_FULL);
        /* this is right place to implement transmit flow control if desired - set TXFLOWEN in MACCONTROL */
    } else if (regContent & (PHY_100BASETX_m | PHY_10BASET_m)) {
        EMACDuplexSet(hdkif->emac_base, EMAC_DUPLEX_HALF);
    } else {
        /* acknowledge MDIO module */
        HWREG(hdkif->mdio_base + MDIO_LINKINTRAW) = MDIO_LINKINTMASKED_USERPHY0;
        HWREG(hdkif->mdio_base + MDIO_LINKINTMASKED) = MDIO_LINKINTMASKED_USERPHY0;

        /* acknowledge EMAC control module by writing appropriate key to MACEOIVECTOR */
        EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_MISC);
        return FALSE;
    }
    /* if link is up configure lwip struct netif */
    if(rpp_eth_phylinkstat(hdkif->inst_num))
    {
#if STATIC_IP_ADDRESS
        netif_set_up(netif);
#elif LWIP_DHCP /* STATIC_IP_ADDRESS-LWIP_DHCP */
        if(dhcp_start(netif) != ERR_OK) /* XXX: can't be used from ISR (mem_malloc()) */
        {
                return DHCP_MEM_ERR;
        }
#if ONCE_LINK_SETUP
        while(netif->dhcp->state != DHCP_BOUND);
#else
        while(netif->dhcp->state != DHCP_BOUND & dhcpBindWait--);
        if(!dhcpBindWait)
        {
                /* acknowledge MDIO module */
                HWREG(hdkif->mdio_base + MDIO_LINKINTRAW) = MDIO_LINKINTMASKED_USERPHY0;
                HWREG(hdkif->mdio_base + MDIO_LINKINTMASKED) = MDIO_LINKINTMASKED_USERPHY0;

                /* acknowledge EMAC control module by writing appropriate key to MACEOIVECTOR */
                EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_MISC);
                return FALSE;
        }
#endif
#else /* LWIP_DHCP-LWIP_AUTOIP FIXME: there should be some kind of waiting till ip is assigned */
        autoip_start(netif);
#endif /* STATIC_IP_ADDRESS-LWIP_DHCP-LWIP_AUTOIP */
#if ONCE_LINK_SETUP
        /* turn this interrupt off */
        HWREG(hdkif->emac_ctrl_base + EMAC_CTRL_CnMISCEN(0)) &= (~EMAC_CTRL_MISC_LINKINT0ENB & 0xf);
#endif /* ONCE_LINK_SETUP */
    }
    else
    {
#if ONCE_LINK_SETUP
        /* acknowledge MDIO module */
        HWREG(hdkif->mdio_base + MDIO_LINKINTRAW) = MDIO_LINKINTMASKED_USERPHY0;
        HWREG(hdkif->mdio_base + MDIO_LINKINTMASKED) = MDIO_LINKINTMASKED_USERPHY0;

        /* acknowledge EMAC control module by writing appropriate key to MACEOIVECTOR */
        EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_MISC);
        return FALSE;
#else
#if STATIC_IP_ADDRESS
        netif_set_down(netif);
#elif LWIP_DHCP /* STATIC_IP_ADDRESS-LWIP_DHCP */
        dhcp_stop(netif);
#endif /* STATIC_IP_ADDRESS-LWIP_DHCP-LWIP_AUTOIP */
#endif
    }

        /* acknowledge MDIO module */
        HWREG(hdkif->mdio_base + MDIO_LINKINTRAW) = MDIO_LINKINTMASKED_USERPHY0;
        HWREG(hdkif->mdio_base + MDIO_LINKINTMASKED) = MDIO_LINKINTMASKED_USERPHY0;

        /* acknowledge EMAC control module by writing appropriate key to MACEOIVECTOR */
        EMACCoreIntAck(hdkif->emac_base, EMAC_INT_CORE0_MISC);

        return TRUE;
}
#endif /* PHY_LINK_MONITOR_INT */
#endif /* FREERTOS_POSIX */