bfin_mac.c

/*
 * Blackfin On-Chip MAC Driver
 *
 * Copyright 2004-2007 Analog Devices Inc.
 *
 * Enter bugs at http://blackfin.uclinux.org/
 *
 * Licensed under the GPL-2 or later.
 */

#include <linux/init.h>
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/delay.h>
#include <linux/timer.h>
#include <linux/errno.h>
#include <linux/irq.h>
#include <linux/io.h>
#include <linux/ioport.h>
#include <linux/crc32.h>
#include <linux/device.h>
#include <linux/spinlock.h>
#include <linux/mii.h>
#include <linux/phy.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/ethtool.h>
#include <linux/skbuff.h>
#include <linux/platform_device.h>

#include <asm/dma.h>
#include <linux/dma-mapping.h>

#include <asm/div64.h>
#include <asm/dpmc.h>
#include <asm/blackfin.h>
#include <asm/cacheflush.h>
#include <asm/portmux.h>

#include "bfin_mac.h"

#define DRV_NAME	"bfin_mac"
#define DRV_VERSION	"1.1"
#define DRV_AUTHOR	"Bryan Wu, Luke Yang"
#define DRV_DESC	"Blackfin on-chip Ethernet MAC driver"

MODULE_AUTHOR(DRV_AUTHOR);
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION(DRV_DESC);
MODULE_ALIAS("platform:bfin_mac");

#if defined(CONFIG_BFIN_MAC_USE_L1)
# define bfin_mac_alloc(dma_handle, size)  l1_data_sram_zalloc(size)
# define bfin_mac_free(dma_handle, ptr)    l1_data_sram_free(ptr)
#else
# define bfin_mac_alloc(dma_handle, size) \
	dma_alloc_coherent(NULL, size, dma_handle, GFP_KERNEL)
# define bfin_mac_free(dma_handle, ptr) \
	dma_free_coherent(NULL, sizeof(*ptr), ptr, dma_handle)
#endif

#define PKT_BUF_SZ 1580

#define MAX_TIMEOUT_CNT	500

/* pointers to maintain transmit list */
static struct net_dma_desc_tx *tx_list_head;
static struct net_dma_desc_tx *tx_list_tail;
static struct net_dma_desc_rx *rx_list_head;
static struct net_dma_desc_rx *rx_list_tail;
static struct net_dma_desc_rx *current_rx_ptr;
static struct net_dma_desc_tx *current_tx_ptr;
static struct net_dma_desc_tx *tx_desc;
static struct net_dma_desc_rx *rx_desc;

#if defined(CONFIG_BFIN_MAC_RMII)
static u16 pin_req[] = P_RMII0;
#else
static u16 pin_req[] = P_MII0;
#endif

static void desc_list_free(void)
{
	struct net_dma_desc_rx *r;
	struct net_dma_desc_tx *t;
	int i;
#if !defined(CONFIG_BFIN_MAC_USE_L1)
	dma_addr_t dma_handle = 0;
#endif

	if (tx_desc) {
		t = tx_list_head;
		for (i = 0; i < CONFIG_BFIN_TX_DESC_NUM; i++) {
			if (t) {
				if (t->skb) {
					dev_kfree_skb(t->skb);
					t->skb = NULL;
				}
				t = t->next;
			}
		}
		bfin_mac_free(dma_handle, tx_desc);
	}

	if (rx_desc) {
		r = rx_list_head;
		for (i = 0; i < CONFIG_BFIN_RX_DESC_NUM; i++) {
			if (r) {
				if (r->skb) {
					dev_kfree_skb(r->skb);
					r->skb = NULL;
				}
				r = r->next;
			}
		}
		bfin_mac_free(dma_handle, rx_desc);
	}
}

static int desc_list_init(void)
{
	int i;
	struct sk_buff *new_skb;
#if !defined(CONFIG_BFIN_MAC_USE_L1)
	/*
	 * This dma_handle is useless in Blackfin dma_alloc_coherent().
	 * The real dma handler is the return value of dma_alloc_coherent().
	 */
	dma_addr_t dma_handle;
#endif

	tx_desc = bfin_mac_alloc(&dma_handle,
				sizeof(struct net_dma_desc_tx) *
				CONFIG_BFIN_TX_DESC_NUM);
	if (tx_desc == NULL)
		goto init_error;

	rx_desc = bfin_mac_alloc(&dma_handle,
				sizeof(struct net_dma_desc_rx) *
				CONFIG_BFIN_RX_DESC_NUM);
	if (rx_desc == NULL)
		goto init_error;

	/* init tx_list */
	tx_list_head = tx_list_tail = tx_desc;

	for (i = 0; i < CONFIG_BFIN_TX_DESC_NUM; i++) {
		struct net_dma_desc_tx *t = tx_desc + i;
		struct dma_descriptor *a = &(t->desc_a);
		struct dma_descriptor *b = &(t->desc_b);

		/*
		 * disable DMA
		 * read from memory WNR = 0
		 * wordsize is 32 bits
		 * 6 half words is desc size
		 * large desc flow
		 */
		a->config = WDSIZE_32 | NDSIZE_6 | DMAFLOW_LARGE;
		a->start_addr = (unsigned long)t->packet;
		a->x_count = 0;
		a->next_dma_desc = b;

		/*
		 * enabled DMA
		 * write to memory WNR = 1
		 * wordsize is 32 bits
		 * disable interrupt
		 * 6 half words is desc size
		 * large desc flow
		 */
		b->config = DMAEN | WNR | WDSIZE_32 | NDSIZE_6 | DMAFLOW_LARGE;
		b->start_addr = (unsigned long)(&(t->status));
		b->x_count = 0;

		t->skb = NULL;
		tx_list_tail->desc_b.next_dma_desc = a;
		tx_list_tail->next = t;
		tx_list_tail = t;
	}
	tx_list_tail->next = tx_list_head;	/* tx_list is a circle */
	tx_list_tail->desc_b.next_dma_desc = &(tx_list_head->desc_a);
	current_tx_ptr = tx_list_head;

	/* init rx_list */
	rx_list_head = rx_list_tail = rx_desc;

	for (i = 0; i < CONFIG_BFIN_RX_DESC_NUM; i++) {
		struct net_dma_desc_rx *r = rx_desc + i;
		struct dma_descriptor *a = &(r->desc_a);
		struct dma_descriptor *b = &(r->desc_b);

		/* allocate a new skb for next time receive */
		new_skb = dev_alloc_skb(PKT_BUF_SZ + NET_IP_ALIGN);
		if (!new_skb) {
			printk(KERN_NOTICE DRV_NAME
			       ": init: low on mem - packet dropped\n");
			goto init_error;
		}
		skb_reserve(new_skb, NET_IP_ALIGN);
		/* Invidate the data cache of skb->data range when it is write back
		 * cache. It will prevent overwritting the new data from DMA
		 */
		blackfin_dcache_invalidate_range((unsigned long)new_skb->head,
					 (unsigned long)new_skb->end);
		r->skb = new_skb;

		/*
		 * enabled DMA
		 * write to memory WNR = 1
		 * wordsize is 32 bits
		 * disable interrupt
		 * 6 half words is desc size
		 * large desc flow
		 */
		a->config = DMAEN | WNR | WDSIZE_32 | NDSIZE_6 | DMAFLOW_LARGE;
		/* since RXDWA is enabled */
		a->start_addr = (unsigned long)new_skb->data - 2;
		a->x_count = 0;
		a->next_dma_desc = b;

		/*
		 * enabled DMA
		 * write to memory WNR = 1
		 * wordsize is 32 bits
		 * enable interrupt
		 * 6 half words is desc size
		 * large desc flow
		 */
		b->config = DMAEN | WNR | WDSIZE_32 | DI_EN |
				NDSIZE_6 | DMAFLOW_LARGE;
		b->start_addr = (unsigned long)(&(r->status));
		b->x_count = 0;

		rx_list_tail->desc_b.next_dma_desc = a;
		rx_list_tail->next = r;
		rx_list_tail = r;
	}
	rx_list_tail->next = rx_list_head;	/* rx_list is a circle */
	rx_list_tail->desc_b.next_dma_desc = &(rx_list_head->desc_a);
	current_rx_ptr = rx_list_head;

	return 0;

init_error:
	desc_list_free();
	printk(KERN_ERR DRV_NAME ": kmalloc failed\n");
	return -ENOMEM;
}


/*---PHY CONTROL AND CONFIGURATION-----------------------------------------*/

/*
 * MII operations
 */
/* Wait until the previous MDC/MDIO transaction has completed */
static int bfin_mdio_poll(void)
{
	int timeout_cnt = MAX_TIMEOUT_CNT;

	/* poll the STABUSY bit */
	while ((bfin_read_EMAC_STAADD()) & STABUSY) {
		udelay(1);
		if (timeout_cnt-- < 0) {
			printk(KERN_ERR DRV_NAME
			": wait MDC/MDIO transaction to complete timeout\n");
			return -ETIMEDOUT;
		}
	}

	return 0;
}

/* Read an off-chip register in a PHY through the MDC/MDIO port */
static int bfin_mdiobus_read(struct mii_bus *bus, int phy_addr, int regnum)
{
	int ret;

	ret = bfin_mdio_poll();
	if (ret)
		return ret;

	/* read mode */
	bfin_write_EMAC_STAADD(SET_PHYAD((u16) phy_addr) |
				SET_REGAD((u16) regnum) |
				STABUSY);

	ret = bfin_mdio_poll();
	if (ret)
		return ret;

	return (int) bfin_read_EMAC_STADAT();
}

/* Write an off-chip register in a PHY through the MDC/MDIO port */
static int bfin_mdiobus_write(struct mii_bus *bus, int phy_addr, int regnum,
			      u16 value)
{
	int ret;

	ret = bfin_mdio_poll();
	if (ret)
		return ret;

	bfin_write_EMAC_STADAT((u32) value);

	/* write mode */
	bfin_write_EMAC_STAADD(SET_PHYAD((u16) phy_addr) |
				SET_REGAD((u16) regnum) |
				STAOP |
				STABUSY);

	return bfin_mdio_poll();
}

static int bfin_mdiobus_reset(struct mii_bus *bus)
{
	return 0;
}

static void bfin_mac_adjust_link(struct net_device *dev)
{
	struct bfin_mac_local *lp = netdev_priv(dev);
	struct phy_device *phydev = lp->phydev;
	unsigned long flags;
	int new_state = 0;

	spin_lock_irqsave(&lp->lock, flags);
	if (phydev->link) {
		/* Now we make sure that we can be in full duplex mode.
		 * If not, we operate in half-duplex mode. */
		if (phydev->duplex != lp->old_duplex) {
			u32 opmode = bfin_read_EMAC_OPMODE();
			new_state = 1;

			if (phydev->duplex)
				opmode |= FDMODE;
			else
				opmode &= ~(FDMODE);

			bfin_write_EMAC_OPMODE(opmode);
			lp->old_duplex = phydev->duplex;
		}

		if (phydev->speed != lp->old_speed) {
#if defined(CONFIG_BFIN_MAC_RMII)
			u32 opmode = bfin_read_EMAC_OPMODE();
			switch (phydev->speed) {
			case 10:
				opmode |= RMII_10;
				break;
			case 100:
				opmode &= ~(RMII_10);
				break;
			default:
				printk(KERN_WARNING
					"%s: Ack!  Speed (%d) is not 10/100!\n",
					DRV_NAME, phydev->speed);
				break;
			}
			bfin_write_EMAC_OPMODE(opmode);
#endif

			new_state = 1;
			lp->old_speed = phydev->speed;
		}

		if (!lp->old_link) {
			new_state = 1;
			lp->old_link = 1;
		}
	} else if (lp->old_link) {
		new_state = 1;
		lp->old_link = 0;
		lp->old_speed = 0;
		lp->old_duplex = -1;
	}

	if (new_state) {
		u32 opmode = bfin_read_EMAC_OPMODE();
		phy_print_status(phydev);
		pr_debug("EMAC_OPMODE = 0x%08x\n", opmode);
	}

	spin_unlock_irqrestore(&lp->lock, flags);
}

/* MDC  = 2.5 MHz */
#define MDC_CLK 2500000

static int mii_probe(struct net_device *dev)
{
	struct bfin_mac_local *lp = netdev_priv(dev);
	struct phy_device *phydev = NULL;
	unsigned short sysctl;
	int i;
	u32 sclk, mdc_div;

	/* Enable PHY output early */
	if (!(bfin_read_VR_CTL() & CLKBUFOE))
		bfin_write_VR_CTL(bfin_read_VR_CTL() | CLKBUFOE);

	sclk = get_sclk();
	mdc_div = ((sclk / MDC_CLK) / 2) - 1;

	sysctl = bfin_read_EMAC_SYSCTL();
	sysctl = (sysctl & ~MDCDIV) | SET_MDCDIV(mdc_div);
	bfin_write_EMAC_SYSCTL(sysctl);

	/* search for connect PHY device */
	for (i = 0; i < PHY_MAX_ADDR; i++) {
		struct phy_device *const tmp_phydev = lp->mii_bus->phy_map[i];

		if (!tmp_phydev)
			continue; /* no PHY here... */

		phydev = tmp_phydev;
		break; /* found it */
	}

	/* now we are supposed to have a proper phydev, to attach to... */
	if (!phydev) {
		printk(KERN_INFO "%s: Don't found any phy device at all\n",
			dev->name);
		return -ENODEV;
	}

#if defined(CONFIG_BFIN_MAC_RMII)
	phydev = phy_connect(dev, dev_name(&phydev->dev), &bfin_mac_adjust_link,
			0, PHY_INTERFACE_MODE_RMII);
#else
	phydev = phy_connect(dev, dev_name(&phydev->dev), &bfin_mac_adjust_link,
			0, PHY_INTERFACE_MODE_MII);
#endif

	if (IS_ERR(phydev)) {
		printk(KERN_ERR "%s: Could not attach to PHY\n", dev->name);
		return PTR_ERR(phydev);
	}

	/* mask with MAC supported features */
	phydev->supported &= (SUPPORTED_10baseT_Half
			      | SUPPORTED_10baseT_Full
			      | SUPPORTED_100baseT_Half
			      | SUPPORTED_100baseT_Full
			      | SUPPORTED_Autoneg
			      | SUPPORTED_Pause | SUPPORTED_Asym_Pause
			      | SUPPORTED_MII
			      | SUPPORTED_TP);

	phydev->advertising = phydev->supported;

	lp->old_link = 0;
	lp->old_speed = 0;
	lp->old_duplex = -1;
	lp->phydev = phydev;

	printk(KERN_INFO "%s: attached PHY driver [%s] "
	       "(mii_bus:phy_addr=%s, irq=%d, mdc_clk=%dHz(mdc_div=%d)"
	       "@sclk=%dMHz)\n",
	       DRV_NAME, phydev->drv->name, dev_name(&phydev->dev), phydev->irq,
	       MDC_CLK, mdc_div, sclk/1000000);

	return 0;
}

/*
 * Ethtool support
 */

/*
 * interrupt routine for magic packet wakeup
 */
static irqreturn_t bfin_mac_wake_interrupt(int irq, void *dev_id)
{
	return IRQ_HANDLED;
}

static int
bfin_mac_ethtool_getsettings(struct net_device *dev, struct ethtool_cmd *cmd)
{
	struct bfin_mac_local *lp = netdev_priv(dev);

	if (lp->phydev)
		return phy_ethtool_gset(lp->phydev, cmd);

	return -EINVAL;
}

static int
bfin_mac_ethtool_setsettings(struct net_device *dev, struct ethtool_cmd *cmd)
{
	struct bfin_mac_local *lp = netdev_priv(dev);

	if (!capable(CAP_NET_ADMIN))
		return -EPERM;

	if (lp->phydev)
		return phy_ethtool_sset(lp->phydev, cmd);

	return -EINVAL;
}

static void bfin_mac_ethtool_getdrvinfo(struct net_device *dev,
					struct ethtool_drvinfo *info)
{
	strcpy(info->driver, DRV_NAME);
	strcpy(info->version, DRV_VERSION);
	strcpy(info->fw_version, "N/A");
	strcpy(info->bus_info, dev_name(&dev->dev));
}

static void bfin_mac_ethtool_getwol(struct net_device *dev,
	struct ethtool_wolinfo *wolinfo)
{
	struct bfin_mac_local *lp = netdev_priv(dev);

	wolinfo->supported = WAKE_MAGIC;
	wolinfo->wolopts = lp->wol;
}

static int bfin_mac_ethtool_setwol(struct net_device *dev,
	struct ethtool_wolinfo *wolinfo)
{
	struct bfin_mac_local *lp = netdev_priv(dev);
	int rc;

	if (wolinfo->wolopts & (WAKE_MAGICSECURE |
				WAKE_UCAST |
				WAKE_MCAST |
				WAKE_BCAST |
				WAKE_ARP))
		return -EOPNOTSUPP;

	lp->wol = wolinfo->wolopts;

	if (lp->wol && !lp->irq_wake_requested) {
		/* register wake irq handler */
		rc = request_irq(IRQ_MAC_WAKEDET, bfin_mac_wake_interrupt,
				 IRQF_DISABLED, "EMAC_WAKE", dev);
		if (rc)
			return rc;
		lp->irq_wake_requested = true;
	}

	if (!lp->wol && lp->irq_wake_requested) {
		free_irq(IRQ_MAC_WAKEDET, dev);
		lp->irq_wake_requested = false;
	}

	/* Make sure the PHY driver doesn't suspend */
	device_init_wakeup(&dev->dev, lp->wol);

	return 0;
}

static const struct ethtool_ops bfin_mac_ethtool_ops = {
	.get_settings = bfin_mac_ethtool_getsettings,
	.set_settings = bfin_mac_ethtool_setsettings,
	.get_link = ethtool_op_get_link,
	.get_drvinfo = bfin_mac_ethtool_getdrvinfo,
	.get_wol = bfin_mac_ethtool_getwol,
	.set_wol = bfin_mac_ethtool_setwol,
};

/**************************************************************************/
void setup_system_regs(struct net_device *dev)
{
	unsigned short sysctl;

	/*
	 * Odd word alignment for Receive Frame DMA word
	 * Configure checksum support and rcve frame word alignment
	 */
	sysctl = bfin_read_EMAC_SYSCTL();
	sysctl |= RXDWA;
#if defined(BFIN_MAC_CSUM_OFFLOAD)
	sysctl |= RXCKS;
#else
	sysctl &= ~RXCKS;
#endif
	bfin_write_EMAC_SYSCTL(sysctl);

	bfin_write_EMAC_MMC_CTL(RSTC | CROLL);

	/* Initialize the TX DMA channel registers */
	bfin_write_DMA2_X_COUNT(0);
	bfin_write_DMA2_X_MODIFY(4);
	bfin_write_DMA2_Y_COUNT(0);
	bfin_write_DMA2_Y_MODIFY(0);

	/* Initialize the RX DMA channel registers */
	bfin_write_DMA1_X_COUNT(0);
	bfin_write_DMA1_X_MODIFY(4);
	bfin_write_DMA1_Y_COUNT(0);
	bfin_write_DMA1_Y_MODIFY(0);
}

static void setup_mac_addr(u8 *mac_addr)
{
	u32 addr_low = le32_to_cpu(*(__le32 *) & mac_addr[0]);
	u16 addr_hi = le16_to_cpu(*(__le16 *) & mac_addr[4]);

	/* this depends on a little-endian machine */
	bfin_write_EMAC_ADDRLO(addr_low);
	bfin_write_EMAC_ADDRHI(addr_hi);
}

static int bfin_mac_set_mac_address(struct net_device *dev, void *p)
{
	struct sockaddr *addr = p;
	if (netif_running(dev))
		return -EBUSY;
	memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
	setup_mac_addr(dev->dev_addr);
	return 0;
}

#ifdef CONFIG_BFIN_MAC_USE_HWSTAMP
#define bfin_mac_hwtstamp_is_none(cfg) ((cfg) == HWTSTAMP_FILTER_NONE)

static int bfin_mac_hwtstamp_ioctl(struct net_device *netdev,
		struct ifreq *ifr, int cmd)
{
	struct hwtstamp_config config;
	struct bfin_mac_local *lp = netdev_priv(netdev);
	u16 ptpctl;
	u32 ptpfv1, ptpfv2, ptpfv3, ptpfoff;

	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
		return -EFAULT;

	pr_debug("%s config flag:0x%x, tx_type:0x%x, rx_filter:0x%x\n",
			__func__, config.flags, config.tx_type, config.rx_filter);

	/* reserved for future extensions */
	if (config.flags)
		return -EINVAL;

	if ((config.tx_type != HWTSTAMP_TX_OFF) &&
			(config.tx_type != HWTSTAMP_TX_ON))
		return -ERANGE;

	ptpctl = bfin_read_EMAC_PTP_CTL();

	switch (config.rx_filter) {
	case HWTSTAMP_FILTER_NONE:
		/*
		 * Dont allow any timestamping
		 */
		ptpfv3 = 0xFFFFFFFF;
		bfin_write_EMAC_PTP_FV3(ptpfv3);
		break;
	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
		/*
		 * Clear the five comparison mask bits (bits[12:8]) in EMAC_PTP_CTL)
		 * to enable all the field matches.
		 */
		ptpctl &= ~0x1F00;
		bfin_write_EMAC_PTP_CTL(ptpctl);
		/*
		 * Keep the default values of the EMAC_PTP_FOFF register.
		 */
		ptpfoff = 0x4A24170C;
		bfin_write_EMAC_PTP_FOFF(ptpfoff);
		/*
		 * Keep the default values of the EMAC_PTP_FV1 and EMAC_PTP_FV2
		 * registers.
		 */
		ptpfv1 = 0x11040800;
		bfin_write_EMAC_PTP_FV1(ptpfv1);
		ptpfv2 = 0x0140013F;
		bfin_write_EMAC_PTP_FV2(ptpfv2);
		/*
		 * The default value (0xFFFC) allows the timestamping of both
		 * received Sync messages and Delay_Req messages.
		 */
		ptpfv3 = 0xFFFFFFFC;
		bfin_write_EMAC_PTP_FV3(ptpfv3);

		config.rx_filter = HWTSTAMP_FILTER_PTP_V1_L4_EVENT;
		break;
	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
		/* Clear all five comparison mask bits (bits[12:8]) in the
		 * EMAC_PTP_CTL register to enable all the field matches.
		 */
		ptpctl &= ~0x1F00;
		bfin_write_EMAC_PTP_CTL(ptpctl);
		/*
		 * Keep the default values of the EMAC_PTP_FOFF register, except set
		 * the PTPCOF field to 0x2A.
		 */
		ptpfoff = 0x2A24170C;
		bfin_write_EMAC_PTP_FOFF(ptpfoff);
		/*
		 * Keep the default values of the EMAC_PTP_FV1 and EMAC_PTP_FV2
		 * registers.
		 */
		ptpfv1 = 0x11040800;
		bfin_write_EMAC_PTP_FV1(ptpfv1);
		ptpfv2 = 0x0140013F;
		bfin_write_EMAC_PTP_FV2(ptpfv2);
		/*
		 * To allow the timestamping of Pdelay_Req and Pdelay_Resp, set
		 * the value to 0xFFF0.
		 */
		ptpfv3 = 0xFFFFFFF0;
		bfin_write_EMAC_PTP_FV3(ptpfv3);

		config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L4_EVENT;
		break;
	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
		/*
		 * Clear bits 8 and 12 of the EMAC_PTP_CTL register to enable only the
		 * EFTM and PTPCM field comparison.
		 */
		ptpctl &= ~0x1100;
		bfin_write_EMAC_PTP_CTL(ptpctl);
		/*
		 * Keep the default values of all the fields of the EMAC_PTP_FOFF
		 * register, except set the PTPCOF field to 0x0E.
		 */
		ptpfoff = 0x0E24170C;
		bfin_write_EMAC_PTP_FOFF(ptpfoff);
		/*
		 * Program bits [15:0] of the EMAC_PTP_FV1 register to 0x88F7, which
		 * corresponds to PTP messages on the MAC layer.
		 */
		ptpfv1 = 0x110488F7;
		bfin_write_EMAC_PTP_FV1(ptpfv1);
		ptpfv2 = 0x0140013F;
		bfin_write_EMAC_PTP_FV2(ptpfv2);
		/*
		 * To allow the timestamping of Pdelay_Req and Pdelay_Resp
		 * messages, set the value to 0xFFF0.
		 */
		ptpfv3 = 0xFFFFFFF0;
		bfin_write_EMAC_PTP_FV3(ptpfv3);

		config.rx_filter = HWTSTAMP_FILTER_PTP_V2_L2_EVENT;
		break;
	default:
		return -ERANGE;
	}

	if (config.tx_type == HWTSTAMP_TX_OFF &&
	    bfin_mac_hwtstamp_is_none(config.rx_filter)) {
		ptpctl &= ~PTP_EN;
		bfin_write_EMAC_PTP_CTL(ptpctl);

		SSYNC();
	} else {
		ptpctl |= PTP_EN;
		bfin_write_EMAC_PTP_CTL(ptpctl);

		/*
		 * clear any existing timestamp
		 */
		bfin_read_EMAC_PTP_RXSNAPLO();
		bfin_read_EMAC_PTP_RXSNAPHI();

		bfin_read_EMAC_PTP_TXSNAPLO();
		bfin_read_EMAC_PTP_TXSNAPHI();

		/*
		 * Set registers so that rollover occurs soon to test this.
		 */
		bfin_write_EMAC_PTP_TIMELO(0x00000000);
		bfin_write_EMAC_PTP_TIMEHI(0xFF800000);

		SSYNC();

		lp->compare.last_update = 0;
		timecounter_init(&lp->clock,
				&lp->cycles,
				ktime_to_ns(ktime_get_real()));
		timecompare_update(&lp->compare, 0);
	}

	lp->stamp_cfg = config;
	return copy_to_user(ifr->ifr_data, &config, sizeof(config)) ?
		-EFAULT : 0;
}

static void bfin_dump_hwtamp(char *s, ktime_t *hw, ktime_t *ts, struct timecompare *cmp)
{
	ktime_t sys = ktime_get_real();

	pr_debug("%s %s hardware:%d,%d transform system:%d,%d system:%d,%d, cmp:%lld, %lld\n",
			__func__, s, hw->tv.sec, hw->tv.nsec, ts->tv.sec, ts->tv.nsec, sys.tv.sec,
			sys.tv.nsec, cmp->offset, cmp->skew);
}

static void bfin_tx_hwtstamp(struct net_device *netdev, struct sk_buff *skb)
{
	struct bfin_mac_local *lp = netdev_priv(netdev);
	union skb_shared_tx *shtx = skb_tx(skb);

	if (shtx->hardware) {
		int timeout_cnt = MAX_TIMEOUT_CNT;

		/* When doing time stamping, keep the connection to the socket
		 * a while longer
		 */
		shtx->in_progress = 1;

		/*
		 * The timestamping is done at the EMAC module's MII/RMII interface
		 * when the module sees the Start of Frame of an event message packet. This
		 * interface is the closest possible place to the physical Ethernet transmission
		 * medium, providing the best timing accuracy.
		 */
		while ((!(bfin_read_EMAC_PTP_ISTAT() & TXTL)) && (--timeout_cnt))
			udelay(1);
		if (timeout_cnt == 0)
			printk(KERN_ERR DRV_NAME
					": fails to timestamp the TX packet\n");
		else {
			struct skb_shared_hwtstamps shhwtstamps;
			u64 ns;
			u64 regval;

			regval = bfin_read_EMAC_PTP_TXSNAPLO();
			regval |= (u64)bfin_read_EMAC_PTP_TXSNAPHI() << 32;
			memset(&shhwtstamps, 0, sizeof(shhwtstamps));
			ns = timecounter_cyc2time(&lp->clock,
					regval);
			timecompare_update(&lp->compare, ns);
			shhwtstamps.hwtstamp = ns_to_ktime(ns);
			shhwtstamps.syststamp =
				timecompare_transform(&lp->compare, ns);
			skb_tstamp_tx(skb, &shhwtstamps);

			bfin_dump_hwtamp("TX", &shhwtstamps.hwtstamp, &shhwtstamps.syststamp, &lp->compare);
		}
	}
}

static void bfin_rx_hwtstamp(struct net_device *netdev, struct sk_buff *skb)
{
	struct bfin_mac_local *lp = netdev_priv(netdev);
	u32 valid;
	u64 regval, ns;
	struct skb_shared_hwtstamps *shhwtstamps;

	if (bfin_mac_hwtstamp_is_none(lp->stamp_cfg.rx_filter))
		return;

	valid = bfin_read_EMAC_PTP_ISTAT() & RXEL;
	if (!valid)
		return;

	shhwtstamps = skb_hwtstamps(skb);

	regval = bfin_read_EMAC_PTP_RXSNAPLO();
	regval |= (u64)bfin_read_EMAC_PTP_RXSNAPHI() << 32;
	ns = timecounter_cyc2time(&lp->clock, regval);
	timecompare_update(&lp->compare, ns);
	memset(shhwtstamps, 0, sizeof(*shhwtstamps));
	shhwtstamps->hwtstamp = ns_to_ktime(ns);
	shhwtstamps->syststamp = timecompare_transform(&lp->compare, ns);

	bfin_dump_hwtamp("RX", &shhwtstamps->hwtstamp, &shhwtstamps->syststamp, &lp->compare);
}

/*
 * bfin_read_clock - read raw cycle counter (to be used by time counter)
 */
static cycle_t bfin_read_clock(const struct cyclecounter *tc)
{
	u64 stamp;

	stamp =  bfin_read_EMAC_PTP_TIMELO();
	stamp |= (u64)bfin_read_EMAC_PTP_TIMEHI() << 32ULL;

	return stamp;
}

#define PTP_CLK 25000000

static void bfin_mac_hwtstamp_init(struct net_device *netdev)
{
	struct bfin_mac_local *lp = netdev_priv(netdev);
	u64 append;

	/* Initialize hardware timer */
	append = PTP_CLK * (1ULL << 32);
	do_div(append, get_sclk());
	bfin_write_EMAC_PTP_ADDEND((u32)append);

	memset(&lp->cycles, 0, sizeof(lp->cycles));
	lp->cycles.read = bfin_read_clock;
	lp->cycles.mask = CLOCKSOURCE_MASK(64);
	lp->cycles.mult = 1000000000 / PTP_CLK;
	lp->cycles.shift = 0;

	/* Synchronize our NIC clock against system wall clock */
	memset(&lp->compare, 0, sizeof(lp->compare));
	lp->compare.source = &lp->clock;
	lp->compare.target = ktime_get_real;
	lp->compare.num_samples = 10;

	/* Initialize hwstamp config */
	lp->stamp_cfg.rx_filter = HWTSTAMP_FILTER_NONE;
	lp->stamp_cfg.tx_type = HWTSTAMP_TX_OFF;
}

#else
# define bfin_mac_hwtstamp_is_none(cfg) 0
# define bfin_mac_hwtstamp_init(dev)
# define bfin_mac_hwtstamp_ioctl(dev, ifr, cmd) (-EOPNOTSUPP)
# define bfin_rx_hwtstamp(dev, skb)
# define bfin_tx_hwtstamp(dev, skb)
#endif

static inline void _tx_reclaim_skb(void)
{
	do {
		tx_list_head->desc_a.config &= ~DMAEN;
		tx_list_head->status.status_word = 0;
		if (tx_list_head->skb) {
			dev_kfree_skb(tx_list_head->skb);
			tx_list_head->skb = NULL;
		}
		tx_list_head = tx_list_head->next;

	} while (tx_list_head->status.status_word != 0);
}

static void tx_reclaim_skb(struct bfin_mac_local *lp)
{
	int timeout_cnt = MAX_TIMEOUT_CNT;

	if (tx_list_head->status.status_word != 0)
		_tx_reclaim_skb();

	if (current_tx_ptr->next == tx_list_head) {
		while (tx_list_head->status.status_word == 0) {
			/* slow down polling to avoid too many queue stop. */
			udelay(10);
			/* reclaim skb if DMA is not running. */
			if (!(bfin_read_DMA2_IRQ_STATUS() & DMA_RUN))
				break;
			if (timeout_cnt-- < 0)
				break;
		}

		if (timeout_cnt >= 0)
			_tx_reclaim_skb();
		else
			netif_stop_queue(lp->ndev);
	}

	if (current_tx_ptr->next != tx_list_head &&
		netif_queue_stopped(lp->ndev))
		netif_wake_queue(lp->ndev);

	if (tx_list_head != current_tx_ptr) {
		/* shorten the timer interval if tx queue is stopped */
		if (netif_queue_stopped(lp->ndev))
			lp->tx_reclaim_timer.expires =
				jiffies + (TX_RECLAIM_JIFFIES >> 4);
		else
			lp->tx_reclaim_timer.expires =
				jiffies + TX_RECLAIM_JIFFIES;

		mod_timer(&lp->tx_reclaim_timer,
			lp->tx_reclaim_timer.expires);
	}

	return;
}

static void tx_reclaim_skb_timeout(unsigned long lp)
{
	tx_reclaim_skb((struct bfin_mac_local *)lp);
}

static int bfin_mac_hard_start_xmit(struct sk_buff *skb,
				struct net_device *dev)
{
	struct bfin_mac_local *lp = netdev_priv(dev);
	u16 *data;
	u32 data_align = (unsigned long)(skb->data) & 0x3;
	union skb_shared_tx *shtx = skb_tx(skb);

	current_tx_ptr->skb = skb;

	if (data_align == 0x2) {
		/* move skb->data to current_tx_ptr payload */
		data = (u16 *)(skb->data) - 1;