[mcs814x] add Moschip MSC814x support

This target currently only supports Moschip's MCS8140 SoC, but support
for other chips in the same family (MCS8142, MCS8144) will be easy to add.

Target support is entirely using Device Tree for probing peripherals.
Drivers support include:
- PCI
- USB 1 & 2
- watchdog
- random number generator
- UART
- timer
- internal Ethernet PHY
- Ethernet MAC core

Support for the following boards is included using Device Tree
- Devolo dLAN USB Extender
- Tigal RBT-832

git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32462 3c298f89-4303-0410-b956-a3cf2f4a3e73

4 files changed, 1137 insertions, 0 deletions

diff --git a/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/Kconfig b/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/Kconfig
new file mode 100644
index 000000000..8fa38a409
--- /dev/null
+++ b/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/Kconfig
@@ -0,0 +1,4 @@
+config  NUPORT_ETHERNET_DRIVER
+	tristate "MCS8140 Ethernet driver"
+	depends on ETHERNET && ARCH_MCS814X
+          help
diff --git a/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/Makefile b/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/Makefile
new file mode 100644
index 000000000..9719c51b7
--- /dev/null
+++ b/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_NUPORT_ETHERNET_DRIVER)   += mcs8140.o
+
+mcs8140-objs  := nuport_mac.o 
diff --git a/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/nuport_mac.c b/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/nuport_mac.c
new file mode 100644
index 000000000..030811c3c
--- /dev/null
+++ b/target/linux/mcs814x/files-3.3/drivers/net/ethernet/mcs8140/nuport_mac.c
@@ -0,0 +1,1066 @@
+/*
+ * Moschip MCS8140 Ethernet MAC driver
+ *
+ * Copyright (C) 2003, Moschip Semiconductors
+ * Copyright (C) 2012, Florian Fainelli <florian@openwrt.org>
+ *
+ * Licensed under GPLv2
+ */
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/delay.h>
+#include <linux/ethtool.h>
+#include <linux/mii.h>
+#include <linux/io.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+#include <linux/irq.h>
+#include <linux/err.h>
+#include <linux/phy.h>
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+
+#include <asm/unaligned.h>
+#include <asm/sizes.h>
+#include <mach/hardware.h>
+
+/* Hardware registers */
+#define MAC_BASE_ADDR		((priv->mac_base))
+
+#define CTRL_REG		(MAC_BASE_ADDR)
+#define MII_BUSY		0x00000001
+#define MII_WRITE		0x00000002
+#define MAC_ADDR_HIGH_REG	(MAC_BASE_ADDR + 0x04)
+#define MAC_ADDR_LOW_REG	(MAC_BASE_ADDR + 0x08)
+#define MII_ADDR_REG		(MAC_BASE_ADDR + 0x14)
+#define MII_DATA_REG		(MAC_BASE_ADDR + 0x18)
+/* Link interrupt registers */
+#define LINK_INT_CSR		(MAC_BASE_ADDR + 0xD0)
+#define LINK_INT_POLL_TIME	(MAC_BASE_ADDR + 0xD4)
+
+#define DMA_CHAN_WIDTH		32
+#define DMA_RX_CHAN		0
+#define DMA_TX_CHAN		2
+
+/* Receive DMA registers */
+#define RX_DMA_BASE		((priv->dma_base) + \
+				(DMA_CHAN_WIDTH * DMA_RX_CHAN))
+#define RX_BUFFER_ADDR		(RX_DMA_BASE + 0x00)
+#define RX_MAX_BYTES		(RX_DMA_BASE + 0x04)
+#define RX_ACT_BYTES		(RX_DMA_BASE + 0x08)
+#define RX_START_DMA		(RX_DMA_BASE + 0x0C)
+#define RX_DMA_ENH		(RX_DMA_BASE + 0x14)
+
+/* Transmit DMA registers */
+#define TX_DMA_BASE		((priv->dma_base) + \
+				(DMA_CHAN_WIDTH * DMA_TX_CHAN))
+#define TX_BUFFER_ADDR		(TX_DMA_BASE + 0x00)
+#define TX_PKT_BYTES		(TX_DMA_BASE + 0x04)
+#define TX_BYTES_SENT		(TX_DMA_BASE + 0x08)
+#define TX_START_DMA		(TX_DMA_BASE + 0x0C)
+#define TX_DMA_STATUS		(TX_DMA_BASE + 0x10)
+#define TX_DMA_ENH		(TX_DMA_BASE + 0x14)
+
+#define RX_ALLOC_SIZE		SZ_2K
+#define MAX_ETH_FRAME_SIZE	1536
+#define RX_SKB_TAILROOM		128
+#define RX_SKB_HEADROOM		(RX_ALLOC_SIZE  - \
+				(MAX_ETH_FRAME_SIZE + RX_SKB_TAILROOM) + 0)
+
+			/* WDT     Late COL    Lenght     COL      Type */
+#define ERROR_FILTER_MASK ((1<<14) | (1<<15) | (1<<16) | (1<<17) | (0<<18) | \
+			/* MII    Dribbling    CRC    Len/type   Control */\
+			(1<<19) | (1<<20) | (1<<21) | (0<<24) | (1<<25) | \
+			/* Unsup   Missed */\
+			(1<<26) | (0<<31))
+#define  TX_RING_SIZE  30
+#define  RX_RING_SIZE  30
+
+static inline u32 nuport_mac_readl(void __iomem *reg)
+{
+	return __raw_readl(reg);
+}
+
+static inline u8 nuport_mac_readb(void __iomem *reg)
+{
+	return __raw_readb(reg);
+}
+
+static inline void nuport_mac_writel(u32 value, void __iomem *reg)
+{
+	__raw_writel(value, reg);
+}
+
+static inline void nuport_mac_writeb(u8 value, void __iomem *reg)
+{
+	__raw_writel(value, reg);
+}
+
+/* MAC private data */
+struct nuport_mac_priv {
+	spinlock_t lock;
+
+	void __iomem	*mac_base;
+	void __iomem	*dma_base;
+
+	int		rx_irq;
+	int		tx_irq;
+	int		link_irq;
+	struct clk	*emac_clk;
+	struct clk	*ephy_clk;
+
+	/* Transmit buffers */
+	struct sk_buff *tx_skb[TX_RING_SIZE];
+	unsigned int valid_txskb[TX_RING_SIZE];
+	unsigned int cur_tx;
+	unsigned int dma_tx;
+	unsigned int tx_full;
+
+	/* Receive buffers */
+	struct sk_buff *rx_skb[RX_RING_SIZE];
+	unsigned int irq_rxskb[RX_RING_SIZE];
+	int pkt_len[RX_RING_SIZE];
+	unsigned int cur_rx;
+	unsigned int dma_rx;
+	unsigned int rx_full;
+
+	unsigned int first_pkt;
+
+	/* Private data */
+	struct napi_struct napi;
+	struct net_device	*dev;
+	struct platform_device	*pdev;
+	struct mii_bus		*mii_bus;
+	struct phy_device	*phydev;
+	int			old_link;
+	int			old_duplex;
+};
+
+void dcache_invalidate_only(unsigned long start, unsigned long end)
+{
+	asm("\n"
+	    "       bic     r0, r0, #31\n"
+	    "1:     mcr     p15, 0, r0, c7, c6, 1\n"
+	    "       add     r0, r0, #32\n"
+	    "       cmp     r0, r1\n" "       blo     1b\n");
+}
+
+void dcache_clean_range(unsigned long start, unsigned long end)
+{
+	asm("\n"
+	    "       bic     r0, r0, #31\n"
+	    "1:     mcr     p15, 0, r0, c7, c10, 1 @ clean D entry\n"
+	    "       add     r0, r0, #32\n"
+	    "       cmp     r0, r1\n"
+	    "       blo     1b\n" \
+	    "       mcr     p15, 0, r0, c7, c10, 4 @ drain WB\n");
+}
+
+static inline int nuport_mac_mii_busy_wait(struct nuport_mac_priv *priv)
+{
+	unsigned long curr;
+	unsigned long finish = jiffies + 3 * HZ;
+
+	do {
+		curr = jiffies;
+		if (!(nuport_mac_readl(MII_ADDR_REG) & MII_BUSY))
+			return 0;
+		cpu_relax();
+	} while (!time_after_eq(curr, finish));
+
+	return -EBUSY;
+}
+
+/* Read from PHY registers */
+static int nuport_mac_mii_read(struct mii_bus *bus,
+				int mii_id, int regnum)
+{
+	struct net_device *dev = bus->priv;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	int ret;
+	u32 val = 0;
+
+	ret = nuport_mac_mii_busy_wait(priv);
+	if (ret)
+		return ret;
+
+	val |= (mii_id << 11) | (regnum << 6) | MII_BUSY;
+	nuport_mac_writel(val, MII_ADDR_REG);
+	ret = nuport_mac_mii_busy_wait(priv);
+	if (ret)
+		return ret;
+
+	return nuport_mac_readl(MII_DATA_REG);
+}
+
+static int nuport_mac_mii_write(struct mii_bus *bus, int mii_id,
+				int regnum, u16 value)
+{
+	struct net_device *dev = bus->priv;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	int ret;
+	u32 val = 0;
+
+	ret = nuport_mac_mii_busy_wait(priv);
+	if (ret)
+		return ret;
+
+	val |= (mii_id << 11) | (regnum << 6) | MII_BUSY | MII_WRITE;
+	nuport_mac_writel(value, MII_DATA_REG);
+	nuport_mac_writel(val, MII_ADDR_REG);
+
+	return nuport_mac_mii_busy_wait(priv);
+}
+
+static int nuport_mac_mii_reset(struct mii_bus *bus)
+{
+	return 0;
+}
+
+static int nuport_mac_start_tx_dma(struct nuport_mac_priv *priv,
+					struct sk_buff *skb)
+{
+	dma_addr_t p;
+	u32 reg;
+	unsigned int timeout = 2048;
+
+	while (timeout--) {
+		reg = nuport_mac_readl(TX_START_DMA);
+		if (!(reg & 0x01))
+			break;
+		cpu_relax();
+	}
+
+	if (!timeout)
+		return -EBUSY;
+
+	p = dma_map_single(&priv->pdev->dev, skb->data,
+			skb->len, DMA_TO_DEVICE);
+
+	/* enable enhanced mode */
+	nuport_mac_writel(0x03, TX_DMA_ENH);
+	nuport_mac_writel(p, TX_BUFFER_ADDR);
+	nuport_mac_writel((skb->len) - 1, TX_PKT_BYTES);
+	wmb();
+	nuport_mac_writel(0x0D, TX_START_DMA);
+
+	return 0;
+}
+
+static void nuport_mac_reset_tx_dma(struct nuport_mac_priv *priv)
+{
+	u32 reg;
+
+	reg = nuport_mac_readl(TX_START_DMA);
+	reg |= (1 << 24);
+	nuport_mac_writel(reg, TX_START_DMA);
+}
+
+static void nuport_mac_start_rx_dma(struct nuport_mac_priv *priv,
+					struct sk_buff *skb)
+{
+	dma_addr_t p;
+
+	p = dma_map_single(&priv->pdev->dev, skb->data,
+				RX_ALLOC_SIZE, DMA_FROM_DEVICE);
+
+	nuport_mac_writel(p, RX_BUFFER_ADDR);
+	wmb();
+	nuport_mac_writel(0x01, RX_START_DMA);
+}
+
+static void nuport_mac_reset_rx_dma(struct nuport_mac_priv *priv)
+{
+	u32 reg;
+
+	reg = nuport_mac_readl(RX_START_DMA);
+	reg |= (1 << 1);
+	nuport_mac_writel(reg, RX_START_DMA);
+}
+
+/* I suppose this might do something, but I am not sure actually */
+static void nuport_mac_disable_rx_dma(struct nuport_mac_priv *priv)
+{
+	u32 reg;
+
+	reg = nuport_mac_readl(RX_DMA_ENH);
+	reg &= ~(1 << 1);
+	nuport_mac_writel(reg, RX_DMA_ENH);
+}
+
+static void nuport_mac_enable_rx_dma(struct nuport_mac_priv *priv)
+{
+	u32 reg;
+
+	reg = nuport_mac_readl(RX_DMA_ENH);
+	reg |= (1 << 1);
+	nuport_mac_writel(reg, RX_DMA_ENH);
+}
+
+/* Add packets to the transmit queue */
+static int nuport_mac_start_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	unsigned long flags;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	int ret;
+
+	dcache_clean_range((u32) skb->data, (u32)(skb->data + skb->len));
+	spin_lock_irqsave(&priv->lock, flags);
+	if (priv->first_pkt) {
+		ret = nuport_mac_start_tx_dma(priv, skb);
+		if (ret) {
+			spin_unlock_irqrestore(&priv->lock, flags);
+			netdev_err(dev, "transmit path busy\n");
+			return NETDEV_TX_BUSY;
+		}
+		priv->first_pkt = 0;
+	}
+
+	priv->tx_skb[priv->cur_tx] = skb;
+	dev->stats.tx_bytes += skb->len;
+	dev->stats.tx_packets++;
+	priv->valid_txskb[priv->cur_tx] = 1;
+	priv->cur_tx++;
+	dev->trans_start = jiffies;
+
+	if (priv->cur_tx >= TX_RING_SIZE)
+		priv->cur_tx = 0;
+
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	if (priv->valid_txskb[priv->cur_tx]) {
+		priv->tx_full = 1;
+		netif_stop_queue(dev);
+	}
+
+	return NETDEV_TX_OK;
+}
+
+static void nuport_mac_adjust_link(struct net_device *dev)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	struct phy_device *phydev = priv->phydev;
+	unsigned int status_changed = 0;
+	u32 reg;
+
+	BUG_ON(!phydev);
+
+	if (priv->old_link != phydev->link) {
+		status_changed = 1;
+		priv->old_link = phydev->link;
+	}
+
+	if (phydev->link & (priv->old_duplex != phydev->duplex)) {
+		reg = nuport_mac_readl(CTRL_REG);
+		if (phydev->duplex == DUPLEX_FULL)
+			reg |= (1 << 20);
+		else
+			reg &= ~(1 << 20);
+		nuport_mac_writel(reg, CTRL_REG);
+
+		status_changed = 1;
+		priv->old_duplex = phydev->duplex;
+	}
+
+	if (!status_changed)
+		return;
+
+	pr_info("%s: link %s", dev->name, phydev->link ?
+		"UP" : "DOWN");
+	if (phydev->link) {
+		pr_cont(" - %d/%s", phydev->speed,
+		phydev->duplex == DUPLEX_FULL ? "full" : "half");
+	}
+	pr_cont("\n");
+}
+
+static irqreturn_t nuport_mac_link_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	u32 reg;
+	u8 phy_addr;
+
+	reg = nuport_mac_readl(LINK_INT_CSR);
+	phy_addr = (reg >> 1) & 0x0f;
+
+	if (phy_addr != priv->phydev->addr) {
+		netdev_err(dev, "spurious PHY irq (phy: %d)\n", phy_addr);
+		return IRQ_NONE;
+	}
+
+	priv->phydev->link = (reg & (1 << 16));
+	nuport_mac_adjust_link(dev);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t nuport_mac_tx_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = (struct net_device *)dev_id;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	struct sk_buff *skb;
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	skb = priv->tx_skb[priv->dma_tx];
+	priv->tx_skb[priv->dma_tx] = NULL;
+	priv->valid_txskb[priv->dma_tx] = 0;
+	dev_kfree_skb_irq(skb);
+
+	priv->dma_tx++;
+	if (priv->dma_tx >= TX_RING_SIZE)
+		priv->dma_tx = 0;
+
+	if (!priv->valid_txskb[priv->dma_tx])
+		priv->first_pkt = 1;
+	else {
+		ret = nuport_mac_start_tx_dma(priv, priv->tx_skb[priv->dma_tx]);
+		if (ret)
+			netdev_err(dev, "failed to restart TX dma\n");
+	}
+
+	if (priv->tx_full) {
+		netdev_err(dev, "restarting transmit queue\n");
+		netif_wake_queue(dev);
+		priv->tx_full = 0;
+	}
+
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+static unsigned int nuport_mac_has_work(struct nuport_mac_priv *priv)
+{
+	unsigned int i;
+
+	for (i = 0; i < RX_RING_SIZE; i++)
+		if (priv->irq_rxskb[i])
+			return 1;
+
+	return 0;
+}
+
+static irqreturn_t nuport_mac_rx_interrupt(int irq, void *dev_id)
+{
+	struct net_device *dev = (struct net_device *)dev_id;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	priv->pkt_len[priv->dma_rx] = nuport_mac_readl(RX_ACT_BYTES) - 4;
+	priv->irq_rxskb[priv->dma_rx] = 0;
+	priv->dma_rx++;
+
+	if (priv->dma_rx >= RX_RING_SIZE)
+		priv->dma_rx = 0;
+
+	if (priv->irq_rxskb[priv->dma_rx] == 1)
+		nuport_mac_start_rx_dma(priv, priv->rx_skb[priv->dma_rx]);
+
+	if (likely(nuport_mac_has_work(priv))) {
+		/* find a way to disable DMA rx irq */
+		nuport_mac_disable_rx_dma(priv);
+		napi_schedule(&priv->napi);
+	}
+
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	return IRQ_HANDLED;
+}
+
+/* Process received packets in tasklet */
+static int nuport_mac_rx(struct net_device *dev, int limit)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	struct sk_buff *skb;
+	int len, status;
+	int count = 0;
+
+	while (count < limit && !priv->irq_rxskb[priv->cur_rx]) {
+		skb = priv->rx_skb[priv->cur_rx];
+		len = priv->pkt_len[priv->cur_rx];
+		dcache_invalidate_only(((u32) skb->data),
+				       ((u32) (skb->data + len + 4)));
+
+		/* Remove 2 bytes added by RX buffer shifting */
+		len = len - 2;
+		skb->data = skb->data + 2;
+
+		/* Get packet status */
+		status = get_unaligned((u32 *) (skb->data + len));
+		skb->dev = dev;
+
+		/* packet filter failed */
+		if (!(status & (1 << 30))) {
+			dev_kfree_skb_irq(skb);
+			goto exit;
+		}
+
+		/* missed frame */
+		if (status & (1 << 31)) {
+			dev->stats.rx_missed_errors++;
+			dev_kfree_skb_irq(skb);
+			goto exit;
+		}
+
+		/* Not ethernet type */
+		if ((!(status & (1 << 18))) || (status & ERROR_FILTER_MASK))
+			dev->stats.rx_errors++;
+
+		if (len > MAX_ETH_FRAME_SIZE) {
+			dev_kfree_skb_irq(skb);
+			goto exit;
+		} else
+			skb_put(skb, len);
+
+		skb->protocol = eth_type_trans(skb, dev);
+		dev->stats.rx_packets++;
+
+		if (status & (1 << 29))
+			skb->pkt_type = PACKET_OTHERHOST;
+		if (status & (1 << 27))
+			skb->pkt_type = PACKET_MULTICAST;
+		if (status & (1 << 28))
+			skb->pkt_type = PACKET_BROADCAST;
+
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		/* Pass the received packet to network layer */
+		netif_receive_skb(skb);
+
+		if (status != NET_RX_DROP)
+			dev->stats.rx_bytes += len - 4;	/* Without CRC */
+		else
+			dev->stats.rx_dropped++;
+
+		dev->last_rx = jiffies;
+
+exit:
+		skb = netdev_alloc_skb(dev, RX_ALLOC_SIZE);
+		skb_reserve(skb, RX_SKB_HEADROOM);
+		priv->rx_skb[priv->cur_rx] = skb;
+		priv->irq_rxskb[priv->cur_rx] = 1;
+		priv->cur_rx++;
+
+		if (priv->cur_rx >= RX_RING_SIZE)
+			priv->cur_rx = 0;
+		count++;
+	}
+
+	return count;
+}
+
+static int nuport_mac_poll(struct napi_struct *napi, int budget)
+{
+	struct nuport_mac_priv *priv =
+		container_of(napi, struct nuport_mac_priv, napi);
+	struct net_device *dev = priv->dev;
+	int work_done;
+
+	work_done = nuport_mac_rx(dev, budget);
+
+	if (work_done < budget) {
+		napi_complete(napi);
+		nuport_mac_enable_rx_dma(priv);
+	}
+
+	return work_done;
+}
+
+static void nuport_mac_init_tx_ring(struct nuport_mac_priv *priv)
+{
+	int i;
+
+	priv->cur_tx = priv->dma_tx = priv->tx_full = 0;
+	for (i = 0; i < TX_RING_SIZE; i++) {
+		priv->tx_skb[i] = NULL;
+		priv->valid_txskb[i] = 0;
+	}
+	priv->first_pkt = 1;
+}
+
+static int nuport_mac_init_rx_ring(struct net_device *dev)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	struct sk_buff *skb;
+	int i;
+
+	priv->cur_rx = priv->dma_rx = priv->rx_full = 0;
+
+	for (i = 0; i < RX_RING_SIZE; i++) {
+		skb = netdev_alloc_skb(dev, RX_ALLOC_SIZE);
+		if (!skb)
+			return -ENOMEM;
+		skb_reserve(skb, RX_SKB_HEADROOM);
+		priv->rx_skb[i] = skb;
+		priv->irq_rxskb[i] = 1;
+	}
+
+	return 0;
+}
+
+static void nuport_mac_free_rx_ring(struct nuport_mac_priv *priv)
+{
+	int i;
+
+	for (i = 0; i < RX_RING_SIZE; i++) {
+		if (!priv->rx_skb[i])
+			continue;
+
+		dev_kfree_skb(priv->rx_skb[i]);
+		priv->rx_skb[i] = NULL;
+	}
+}
+
+static void nuport_mac_read_mac_address(struct net_device *dev)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	int i;
+
+	for (i = 0; i < 4; i++)
+		dev->dev_addr[i] = nuport_mac_readb(MAC_ADDR_LOW_REG + i);
+	dev->dev_addr[4] = nuport_mac_readb(MAC_ADDR_HIGH_REG);
+	dev->dev_addr[5] = nuport_mac_readb(MAC_ADDR_HIGH_REG + 1);
+
+	if (!is_valid_ether_addr(dev->dev_addr)) {
+		dev_info(&priv->pdev->dev, "using random address\n");
+		random_ether_addr(dev->dev_addr);
+	}
+}
+
+static int nuport_mac_change_mac_address(struct net_device *dev, void *mac_addr)
+{
+	struct sockaddr *addr = mac_addr;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	unsigned long *temp = (unsigned long *)dev->dev_addr;
+	u32 high, low;
+
+	if (netif_running(dev))
+		return -EBUSY;
+
+	memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN);
+
+	spin_lock_irq(&priv->lock);
+
+	nuport_mac_writel(*temp, MAC_ADDR_LOW_REG);
+	temp = (unsigned long *)(dev->dev_addr + 4);
+	nuport_mac_writel(*temp, MAC_ADDR_HIGH_REG);
+
+	low = nuport_mac_readl(MAC_ADDR_LOW_REG);
+	high = nuport_mac_readl(MAC_ADDR_HIGH_REG);
+
+	spin_unlock_irq(&priv->lock);
+
+	return 0;
+}
+
+static int nuport_mac_open(struct net_device *dev)
+{
+	int ret;
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	unsigned long flags;
+	u32 reg;
+	u8 tmp;
+
+	/* Enable hardware filters */
+	reg = nuport_mac_readl((void __iomem *)_CONFADDR_DBGLED);
+	reg |= 0x80;
+	nuport_mac_writel(reg, (void __iomem *)_CONFADDR_DBGLED);
+
+	/* Set LEDs to Link act and RX/TX mode */
+	reg = nuport_mac_readl((void __iomem *)(_CONFADDR_SYSDBG + 0x04));
+	reg |= 0x01;
+	nuport_mac_writel(reg, (void __iomem *)(_CONFADDR_SYSDBG + 0x04));
+
+	ret = clk_enable(priv->emac_clk);
+	if (ret) {
+		netdev_err(dev, "failed to enable EMAC clock\n");
+		return ret;
+	}
+
+	/* Set MAC into full duplex mode by default */
+	nuport_mac_writel(0x1010052C, CTRL_REG);
+
+	/* set mac address in hardware in case it was not already */
+	nuport_mac_change_mac_address(dev, dev->dev_addr);
+
+	ret = request_irq(priv->link_irq, &nuport_mac_link_interrupt,
+				0, dev->name, dev);
+	if (ret) {
+		netdev_err(dev, "unable to request link interrupt\n");
+		goto out_emac_clk;
+	}
+
+	phy_start(priv->phydev);
+
+	/* Enable link interrupt monitoring */
+	spin_lock_irqsave(&priv->lock, flags);
+	nuport_mac_writel(0x1041 | (priv->phydev->addr << 1), LINK_INT_CSR);
+	nuport_mac_writel(0xFFFFF, LINK_INT_POLL_TIME);
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	ret = request_irq(priv->tx_irq, &nuport_mac_tx_interrupt,
+				0, dev->name, dev);
+	if (ret) {
+		netdev_err(dev, "unable to request rx interrupt\n");
+		goto out_link_irq;
+	}
+
+	napi_enable(&priv->napi);
+
+	ret = request_irq(priv->rx_irq, &nuport_mac_rx_interrupt,
+				0, dev->name, dev);
+	if (ret) {
+		netdev_err(dev, "unable to request tx interrupt\n");
+		goto out_tx_irq;
+	}
+
+	/* Enable buffer shifting in RX */
+	tmp = nuport_mac_readb((void __iomem *)(_CONFADDR_SYSDBG + 0x1D));
+	tmp |= 0x01;
+	nuport_mac_writeb(tmp, (void __iomem *)(_CONFADDR_SYSDBG + 0x1D));
+
+	netif_start_queue(dev);
+
+	nuport_mac_init_tx_ring(priv);
+
+	ret = nuport_mac_init_rx_ring(dev);
+	if (ret) {
+		netdev_err(dev, "rx ring init failed\n");
+		goto out_rx_skb;
+	}
+
+	nuport_mac_reset_tx_dma(priv);
+	nuport_mac_reset_rx_dma(priv);
+
+	/* Start RX DMA */
+	nuport_mac_start_rx_dma(priv, priv->rx_skb[0]);
+
+	return 0;
+
+out_rx_skb:
+	nuport_mac_free_rx_ring(priv);
+	free_irq(priv->rx_irq, dev);
+out_tx_irq:
+	free_irq(priv->tx_irq, dev);
+out_link_irq:
+	free_irq(priv->link_irq, dev);
+out_emac_clk:
+	clk_disable(priv->emac_clk);
+	return ret;
+}
+
+static int nuport_mac_close(struct net_device *dev)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+
+	spin_lock_irq(&priv->lock);
+	napi_disable(&priv->napi);
+	netif_stop_queue(dev);
+
+	free_irq(priv->link_irq, dev);
+	nuport_mac_writel(0x00, LINK_INT_CSR);
+	nuport_mac_writel(0x00, LINK_INT_POLL_TIME);
+	phy_stop(priv->phydev);
+
+	free_irq(priv->tx_irq, dev);
+	free_irq(priv->rx_irq, dev);
+	spin_unlock_irq(&priv->lock);
+
+	nuport_mac_free_rx_ring(priv);
+
+	clk_disable(priv->emac_clk);
+
+	return 0;
+}
+
+static int nuport_mac_mii_probe(struct net_device *dev)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+	struct phy_device *phydev = NULL;
+	int ret;
+
+	ret = clk_enable(priv->ephy_clk);
+	if (ret) {
+		netdev_err(dev, "unable to enable ePHY clk\n");
+		return ret;
+	}
+
+	phydev = phy_find_first(priv->mii_bus);
+	if (!phydev) {
+		netdev_err(dev, "no PHYs found\n");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	phydev = phy_connect(dev, dev_name(&phydev->dev),
+			nuport_mac_adjust_link, 0,
+			PHY_INTERFACE_MODE_MII);
+	if (IS_ERR(phydev)) {
+		netdev_err(dev, "could not attach PHY\n");
+		ret = PTR_ERR(phydev);
+		goto out;
+	}
+
+	phydev->supported &= PHY_BASIC_FEATURES;
+	phydev->advertising = phydev->supported;
+	priv->phydev = phydev;
+	priv->old_link = 0;
+	priv->old_duplex = -1;
+
+	dev_info(&priv->pdev->dev, "attached PHY driver [%s] "
+		"(mii_bus:phy_addr=%d)\n",
+		phydev->drv->name, phydev->addr);
+
+	return 0;
+
+out:
+	/* disable the Ethernet PHY clock for the moment */
+	clk_disable(priv->ephy_clk);
+
+	return ret;
+}
+
+static void nuport_mac_ethtool_drvinfo(struct net_device *dev,
+					struct ethtool_drvinfo *info)
+{
+	strncpy(info->driver, "nuport-mac", sizeof(info->driver));
+	strncpy(info->version, "0.1", sizeof(info->version));
+	strncpy(info->fw_version, "N/A", sizeof(info->fw_version));
+	strncpy(info->bus_info, "internal", sizeof(info->bus_info));
+	info->n_stats = 0;
+	info->testinfo_len = 0;
+	info->regdump_len = 0;
+	info->eedump_len = 0;
+}
+
+static int nuport_mac_ethtool_get_settings(struct net_device *dev,
+					struct ethtool_cmd *cmd)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+
+	if (priv->phydev)
+		return phy_ethtool_gset(priv->phydev, cmd);
+
+	return -EINVAL;
+}
+
+static int nuport_mac_ethtool_set_settings(struct net_device *dev,
+					struct ethtool_cmd *cmd)
+{
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+
+	if (priv->phydev)
+		return phy_ethtool_sset(priv->phydev, cmd);
+
+	return -EINVAL;
+}
+
+static const struct ethtool_ops nuport_mac_ethtool_ops = {
+	.get_drvinfo		= nuport_mac_ethtool_drvinfo,
+	.get_link		= ethtool_op_get_link,
+	.get_settings		= nuport_mac_ethtool_get_settings,
+	.set_settings		= nuport_mac_ethtool_set_settings,
+};
+
+static const struct net_device_ops nuport_mac_ops = {
+	.ndo_open		= nuport_mac_open,
+	.ndo_stop		= nuport_mac_close,
+	.ndo_start_xmit		= nuport_mac_start_xmit,
+	.ndo_change_mtu		= eth_change_mtu,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_set_mac_address	= nuport_mac_change_mac_address,
+};
+
+static int __init nuport_mac_probe(struct platform_device *pdev)
+{
+	struct net_device *dev;
+	struct nuport_mac_priv *priv = NULL;
+	struct resource *regs, *dma;
+	int ret = 0;
+	int rx_irq, tx_irq, link_irq;
+	int i;
+
+	dev = alloc_etherdev(sizeof(struct nuport_mac_priv));
+	if (!dev) {
+		dev_err(&pdev->dev, "no memory for net_device\n");
+		return -ENOMEM;
+	}
+
+	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	dma = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	if (!regs || !dma) {
+		dev_err(&pdev->dev, "failed to get regs resources\n");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	rx_irq = platform_get_irq(pdev, 0);
+	tx_irq = platform_get_irq(pdev, 1);
+	link_irq = platform_get_irq(pdev, 2);
+	if (rx_irq < 0 || tx_irq < 0 || link_irq < 0) {
+		ret = -ENODEV;
+		goto out;
+	}
+
+	platform_set_drvdata(pdev, dev);
+	SET_NETDEV_DEV(dev, &pdev->dev);
+	priv = netdev_priv(dev);
+	priv->pdev = pdev;
+	priv->dev = dev;
+	spin_lock_init(&priv->lock);
+
+	priv->mac_base = devm_ioremap(&pdev->dev,
+				regs->start, resource_size(regs));
+	if (!priv->mac_base) {
+		dev_err(&pdev->dev, "failed to remap regs\n");
+		ret = -ENOMEM;
+		goto out_platform;
+	}
+
+	priv->dma_base = devm_ioremap(&pdev->dev,
+				dma->start, resource_size(dma));
+	if (!priv->dma_base) {
+		dev_err(&pdev->dev, "failed to remap dma-regs\n");
+		ret = -ENOMEM;
+		goto out_platform;
+	}
+
+	priv->emac_clk = clk_get(&pdev->dev, "emac");
+	if (IS_ERR_OR_NULL(priv->emac_clk)) {
+		dev_err(&pdev->dev, "failed to get emac clk\n");
+		ret = PTR_ERR(priv->emac_clk);
+		goto out_platform;
+	}
+
+	priv->ephy_clk = clk_get(&pdev->dev, "ephy");
+	if (IS_ERR_OR_NULL(priv->ephy_clk)) {
+		dev_err(&pdev->dev, "failed to get ephy clk\n");
+		ret = PTR_ERR(priv->ephy_clk);
+		goto out_platform;
+	}
+
+	priv->link_irq = link_irq;
+	priv->rx_irq = rx_irq;
+	priv->tx_irq = tx_irq;
+	dev->netdev_ops = &nuport_mac_ops;
+	dev->ethtool_ops = &nuport_mac_ethtool_ops;
+	dev->watchdog_timeo = HZ;
+	dev->flags = IFF_BROADCAST;	/* Supports Broadcast */
+
+	netif_napi_add(dev, &priv->napi, nuport_mac_poll, 64);
+
+	priv->mii_bus = mdiobus_alloc();
+	if (!priv->mii_bus) {
+		dev_err(&pdev->dev, "mii bus allocation failed\n");
+		goto out;
+	}
+
+	priv->mii_bus->priv = dev;
+	priv->mii_bus->read = nuport_mac_mii_read;
+	priv->mii_bus->write = nuport_mac_mii_write;
+	priv->mii_bus->reset = nuport_mac_mii_reset;
+	priv->mii_bus->name = "nuport-mac-mii";
+	priv->mii_bus->phy_mask = (1 << 0);
+	snprintf(priv->mii_bus->id, MII_BUS_ID_SIZE, "%s", pdev->name);
+	priv->mii_bus->irq = kzalloc(PHY_MAX_ADDR * sizeof(int), GFP_KERNEL);
+	if (!priv->mii_bus->irq) {
+		dev_err(&pdev->dev, "failed to allocate mii_bus irqs\n");
+		ret = -ENOMEM;
+		goto out_mdio;
+	}
+
+	/* We support PHY interrupts routed back to the MAC */
+	for (i = 0; i < PHY_MAX_ADDR; i++)
+		priv->mii_bus->irq[i] = PHY_IGNORE_INTERRUPT;
+
+	ret = mdiobus_register(priv->mii_bus);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register mii_bus\n");
+		goto out_mdio_irq;
+	}
+
+	ret = nuport_mac_mii_probe(dev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to probe MII bus\n");
+		goto out_mdio_unregister;
+	}
+
+	ret = register_netdev(dev);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to register net_device\n");
+		goto out_mdio_probe;
+	}
+
+	/* read existing mac address */
+	nuport_mac_read_mac_address(dev);
+
+	dev_info(&pdev->dev, "registered (MAC: %pM)\n", dev->dev_addr);
+
+	return ret;
+
+out_mdio_probe:
+	phy_disconnect(priv->phydev);
+out_mdio_unregister:
+	mdiobus_unregister(priv->mii_bus);
+out_mdio_irq:
+	kfree(priv->mii_bus->irq);
+out_mdio:
+	mdiobus_free(priv->mii_bus);
+out_platform:
+	platform_set_drvdata(pdev, NULL);
+out:
+	clk_put(priv->ephy_clk);
+	clk_put(priv->emac_clk);
+	free_netdev(dev);
+	platform_set_drvdata(pdev, NULL);
+	return ret;
+}
+
+static int nuport_mac_remove(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct nuport_mac_priv *priv = netdev_priv(dev);
+
+	unregister_netdev(dev);
+	phy_disconnect(priv->phydev);
+	mdiobus_unregister(priv->mii_bus);
+	kfree(priv->mii_bus->irq);
+	mdiobus_free(priv->mii_bus);
+	clk_put(priv->ephy_clk);
+	clk_put(priv->emac_clk);
+	free_netdev(dev);
+
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static struct of_device_id nuport_eth_ids[] __initdata = {
+	{.compatible = "moschip,nuport-mac",},
+	{ /* sentinel */ },
+};
+
+static struct platform_driver nuport_eth_driver = {
+	.driver = {
+		.name = "nuport-mac",
+		.owner = THIS_MODULE,
+		.of_match_table = nuport_eth_ids,
+	},
+	.probe	= nuport_mac_probe,
+	.remove	= __devexit_p(nuport_mac_remove),
+};
+
+module_platform_driver(nuport_eth_driver);
+
+MODULE_AUTHOR("Moschip Semiconductors Ltd.");
+MODULE_DESCRIPTION("Moschip MCS8140 Ethernet MAC driver");
+MODULE_LICENSE("GPL");
diff --git a/target/linux/mcs814x/files-3.3/drivers/net/phy/mcs814x.c b/target/linux/mcs814x/files-3.3/drivers/net/phy/mcs814x.c
new file mode 100644
index 000000000..e92176e3b
--- /dev/null
+++ b/target/linux/mcs814x/files-3.3/drivers/net/phy/mcs814x.c
@@ -0,0 +1,64 @@
+/*
+ * Driver for Moschip MCS814x internal PHY
+ *
+ * Copyright (c) 2012 Florian Fainelli <florian@openwrt.org>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/unistd.h>
+#include <linux/interrupt.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/skbuff.h>
+#include <linux/spinlock.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/mii.h>
+#include <linux/ethtool.h>
+#include <linux/phy.h>
+
+MODULE_DESCRIPTION("Moschip MCS814x PHY driver");
+MODULE_AUTHOR("Florian Fainelli <florian@openwrt.org>");
+MODULE_LICENSE("GPL");
+
+/* Nothing special about this PHY but its OUI (O) */
+static struct phy_driver mcs8140_driver = {
+	.phy_id		= 0,
+	.name		= "Moschip MCS8140",
+	.phy_id_mask	= 0x02,
+	.features	= PHY_BASIC_FEATURES,
+	.config_aneg	= &genphy_config_aneg,
+	.read_status	= &genphy_read_status,
+	.suspend	= genphy_suspend,
+	.resume		= genphy_resume,
+	.driver		= { .owner = THIS_MODULE,},
+};
+
+static int __init mcs814x_phy_init(void)
+{
+	return phy_driver_register(&mcs8140_driver);
+}
+
+static void __exit mcs814x_phy_exit(void)
+{
+	phy_driver_unregister(&mcs8140_driver);
+}
+
+module_init(mcs814x_phy_init);
+module_exit(mcs814x_phy_exit);
+
+static struct mdio_device_id __maybe_unused mcs814x_phy_tbl[] = {
+	{ 0x0, 0x0ffffff0 },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(mdio, mcs814x_phy_tbl);