Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from David Miller:

 1) Support ipv6 checksum offload in sunvnet driver, from Shannon
    Nelson.

 2) Move to RB-tree instead of custom AVL code in inetpeer, from Eric
    Dumazet.

 3) Allow generic XDP to work on virtual devices, from John Fastabend.

 4) Add bpf device maps and XDP_REDIRECT, which can be used to build
    arbitrary switching frameworks using XDP. From John Fastabend.

 5) Remove UFO offloads from the tree, gave us little other than bugs.

 6) Remove the IPSEC flow cache, from Florian Westphal.

 7) Support ipv6 route offload in mlxsw driver.

 8) Support VF representors in bnxt_en, from Sathya Perla.

 9) Add support for forward error correction modes to ethtool, from
    Vidya Sagar Ravipati.

10) Add time filter for packet scheduler action dumping, from Jamal Hadi
    Salim.

11) Extend the zerocopy sendmsg() used by virtio and tap to regular
    sockets via MSG_ZEROCOPY. From Willem de Bruijn.

12) Significantly rework value tracking in the BPF verifier, from Edward
    Cree.

13) Add new jump instructions to eBPF, from Daniel Borkmann.

14) Rework rtnetlink plumbing so that operations can be run without
    taking the RTNL semaphore. From Florian Westphal.

15) Support XDP in tap driver, from Jason Wang.

16) Add 32-bit eBPF JIT for ARM, from Shubham Bansal.

17) Add Huawei hinic ethernet driver.

18) Allow to report MD5 keys in TCP inet_diag dumps, from Ivan
    Delalande.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1780 commits)
  i40e: point wb_desc at the nvm_wb_desc during i40e_read_nvm_aq
  i40e: avoid NVM acquire deadlock during NVM update
  drivers: net: xgene: Remove return statement from void function
  drivers: net: xgene: Configure tx/rx delay for ACPI
  drivers: net: xgene: Read tx/rx delay for ACPI
  rocker: fix kcalloc parameter order
  rds: Fix non-atomic operation on shared flag variable
  net: sched: don't use GFP_KERNEL under spin lock
  vhost_net: correctly check tx avail during rx busy polling
  net: mdio-mux: add mdio_mux parameter to mdio_mux_init()
  rxrpc: Make service connection lookup always check for retry
  net: stmmac: Delete dead code for MDIO registration
  gianfar: Fix Tx flow control deactivation
  cxgb4: Ignore MPS_TX_INT_CAUSE[Bubble] for T6
  cxgb4: Fix pause frame count in t4_get_port_stats
  cxgb4: fix memory leak
  tun: rename generic_xdp to skb_xdp
  tun: reserve extra headroom only when XDP is set
  net: dsa: bcm_sf2: Configure IMP port TC2QOS mapping
  net: dsa: bcm_sf2: Advertise number of egress queues
  ...

1 files changed, 989 insertions, 0 deletions

diff --git a/drivers/staging/irda/drivers/au1k_ir.c b/drivers/staging/irda/drivers/au1k_ir.c
new file mode 100644
index 000000000000..be4ea6aa57a9
--- /dev/null
+++ b/drivers/staging/irda/drivers/au1k_ir.c
@@ -0,0 +1,989 @@
+/*
+ * Alchemy Semi Au1000 IrDA driver
+ *
+ * Copyright 2001 MontaVista Software Inc.
+ * Author: MontaVista Software, Inc.
+ *         	ppopov@mvista.com or source@mvista.com
+ *
+ *  This program is free software; you can distribute it and/or modify it
+ *  under the terms of the GNU General Public License (Version 2) as
+ *  published by the Free Software Foundation.
+ *
+ *  This program is distributed in the hope it will be useful, but WITHOUT
+ *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+ *  for more details.
+ *
+ *  You should have received a copy of the GNU General Public License along
+ *  with this program; if not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/clk.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+#include <net/irda/irda.h>
+#include <net/irda/irmod.h>
+#include <net/irda/wrapper.h>
+#include <net/irda/irda_device.h>
+#include <asm/mach-au1x00/au1000.h>
+
+/* registers */
+#define IR_RING_PTR_STATUS	0x00
+#define IR_RING_BASE_ADDR_H	0x04
+#define IR_RING_BASE_ADDR_L	0x08
+#define IR_RING_SIZE		0x0C
+#define IR_RING_PROMPT		0x10
+#define IR_RING_ADDR_CMPR	0x14
+#define IR_INT_CLEAR		0x18
+#define IR_CONFIG_1		0x20
+#define IR_SIR_FLAGS		0x24
+#define IR_STATUS		0x28
+#define IR_READ_PHY_CONFIG	0x2C
+#define IR_WRITE_PHY_CONFIG	0x30
+#define IR_MAX_PKT_LEN		0x34
+#define IR_RX_BYTE_CNT		0x38
+#define IR_CONFIG_2		0x3C
+#define IR_ENABLE		0x40
+
+/* Config1 */
+#define IR_RX_INVERT_LED	(1 << 0)
+#define IR_TX_INVERT_LED	(1 << 1)
+#define IR_ST			(1 << 2)
+#define IR_SF			(1 << 3)
+#define IR_SIR			(1 << 4)
+#define IR_MIR			(1 << 5)
+#define IR_FIR			(1 << 6)
+#define IR_16CRC		(1 << 7)
+#define IR_TD			(1 << 8)
+#define IR_RX_ALL		(1 << 9)
+#define IR_DMA_ENABLE		(1 << 10)
+#define IR_RX_ENABLE		(1 << 11)
+#define IR_TX_ENABLE		(1 << 12)
+#define IR_LOOPBACK		(1 << 14)
+#define IR_SIR_MODE		(IR_SIR | IR_DMA_ENABLE | \
+				 IR_RX_ALL | IR_RX_ENABLE | IR_SF | \
+				 IR_16CRC)
+
+/* ir_status */
+#define IR_RX_STATUS		(1 << 9)
+#define IR_TX_STATUS		(1 << 10)
+#define IR_PHYEN		(1 << 15)
+
+/* ir_write_phy_config */
+#define IR_BR(x)		(((x) & 0x3f) << 10)	/* baud rate */
+#define IR_PW(x)		(((x) & 0x1f) << 5)	/* pulse width */
+#define IR_P(x)			((x) & 0x1f)		/* preamble bits */
+
+/* Config2 */
+#define IR_MODE_INV		(1 << 0)
+#define IR_ONE_PIN		(1 << 1)
+#define IR_PHYCLK_40MHZ		(0 << 2)
+#define IR_PHYCLK_48MHZ		(1 << 2)
+#define IR_PHYCLK_56MHZ		(2 << 2)
+#define IR_PHYCLK_64MHZ		(3 << 2)
+#define IR_DP			(1 << 4)
+#define IR_DA			(1 << 5)
+#define IR_FLT_HIGH		(0 << 6)
+#define IR_FLT_MEDHI		(1 << 6)
+#define IR_FLT_MEDLO		(2 << 6)
+#define IR_FLT_LO		(3 << 6)
+#define IR_IEN			(1 << 8)
+
+/* ir_enable */
+#define IR_HC			(1 << 3)	/* divide SBUS clock by 2 */
+#define IR_CE			(1 << 2)	/* clock enable */
+#define IR_C			(1 << 1)	/* coherency bit */
+#define IR_BE			(1 << 0)	/* set in big endian mode */
+
+#define NUM_IR_DESC	64
+#define RING_SIZE_4	0x0
+#define RING_SIZE_16	0x3
+#define RING_SIZE_64	0xF
+#define MAX_NUM_IR_DESC	64
+#define MAX_BUF_SIZE	2048
+
+/* Ring descriptor flags */
+#define AU_OWN		(1 << 7) /* tx,rx */
+#define IR_DIS_CRC	(1 << 6) /* tx */
+#define IR_BAD_CRC	(1 << 5) /* tx */
+#define IR_NEED_PULSE	(1 << 4) /* tx */
+#define IR_FORCE_UNDER	(1 << 3) /* tx */
+#define IR_DISABLE_TX	(1 << 2) /* tx */
+#define IR_HW_UNDER	(1 << 0) /* tx */
+#define IR_TX_ERROR	(IR_DIS_CRC | IR_BAD_CRC | IR_HW_UNDER)
+
+#define IR_PHY_ERROR	(1 << 6) /* rx */
+#define IR_CRC_ERROR	(1 << 5) /* rx */
+#define IR_MAX_LEN	(1 << 4) /* rx */
+#define IR_FIFO_OVER	(1 << 3) /* rx */
+#define IR_SIR_ERROR	(1 << 2) /* rx */
+#define IR_RX_ERROR	(IR_PHY_ERROR | IR_CRC_ERROR | \
+			 IR_MAX_LEN | IR_FIFO_OVER | IR_SIR_ERROR)
+
+struct db_dest {
+	struct db_dest *pnext;
+	volatile u32 *vaddr;
+	dma_addr_t dma_addr;
+};
+
+struct ring_dest {
+	u8 count_0;	/* 7:0  */
+	u8 count_1;	/* 12:8 */
+	u8 reserved;
+	u8 flags;
+	u8 addr_0;	/* 7:0   */
+	u8 addr_1;	/* 15:8  */
+	u8 addr_2;	/* 23:16 */
+	u8 addr_3;	/* 31:24 */
+};
+
+/* Private data for each instance */
+struct au1k_private {
+	void __iomem *iobase;
+	int irq_rx, irq_tx;
+
+	struct db_dest *pDBfree;
+	struct db_dest db[2 * NUM_IR_DESC];
+	volatile struct ring_dest *rx_ring[NUM_IR_DESC];
+	volatile struct ring_dest *tx_ring[NUM_IR_DESC];
+	struct db_dest *rx_db_inuse[NUM_IR_DESC];
+	struct db_dest *tx_db_inuse[NUM_IR_DESC];
+	u32 rx_head;
+	u32 tx_head;
+	u32 tx_tail;
+	u32 tx_full;
+
+	iobuff_t rx_buff;
+
+	struct net_device *netdev;
+	struct qos_info qos;
+	struct irlap_cb *irlap;
+
+	u8 open;
+	u32 speed;
+	u32 newspeed;
+
+	struct resource *ioarea;
+	struct au1k_irda_platform_data *platdata;
+	struct clk *irda_clk;
+};
+
+static int qos_mtt_bits = 0x07;  /* 1 ms or more */
+
+static void au1k_irda_plat_set_phy_mode(struct au1k_private *p, int mode)
+{
+	if (p->platdata && p->platdata->set_phy_mode)
+		p->platdata->set_phy_mode(mode);
+}
+
+static inline unsigned long irda_read(struct au1k_private *p,
+				      unsigned long ofs)
+{
+	/*
+	* IrDA peripheral bug. You have to read the register
+	* twice to get the right value.
+	*/
+	(void)__raw_readl(p->iobase + ofs);
+	return __raw_readl(p->iobase + ofs);
+}
+
+static inline void irda_write(struct au1k_private *p, unsigned long ofs,
+			      unsigned long val)
+{
+	__raw_writel(val, p->iobase + ofs);
+	wmb();
+}
+
+/*
+ * Buffer allocation/deallocation routines. The buffer descriptor returned
+ * has the virtual and dma address of a buffer suitable for
+ * both, receive and transmit operations.
+ */
+static struct db_dest *GetFreeDB(struct au1k_private *aup)
+{
+	struct db_dest *db;
+	db = aup->pDBfree;
+
+	if (db)
+		aup->pDBfree = db->pnext;
+	return db;
+}
+
+/*
+  DMA memory allocation, derived from pci_alloc_consistent.
+  However, the Au1000 data cache is coherent (when programmed
+  so), therefore we return KSEG0 address, not KSEG1.
+*/
+static void *dma_alloc(size_t size, dma_addr_t *dma_handle)
+{
+	void *ret;
+	int gfp = GFP_ATOMIC | GFP_DMA;
+
+	ret = (void *)__get_free_pages(gfp, get_order(size));
+
+	if (ret != NULL) {
+		memset(ret, 0, size);
+		*dma_handle = virt_to_bus(ret);
+		ret = (void *)KSEG0ADDR(ret);
+	}
+	return ret;
+}
+
+static void dma_free(void *vaddr, size_t size)
+{
+	vaddr = (void *)KSEG0ADDR(vaddr);
+	free_pages((unsigned long) vaddr, get_order(size));
+}
+
+
+static void setup_hw_rings(struct au1k_private *aup, u32 rx_base, u32 tx_base)
+{
+	int i;
+	for (i = 0; i < NUM_IR_DESC; i++) {
+		aup->rx_ring[i] = (volatile struct ring_dest *)
+			(rx_base + sizeof(struct ring_dest) * i);
+	}
+	for (i = 0; i < NUM_IR_DESC; i++) {
+		aup->tx_ring[i] = (volatile struct ring_dest *)
+			(tx_base + sizeof(struct ring_dest) * i);
+	}
+}
+
+static int au1k_irda_init_iobuf(iobuff_t *io, int size)
+{
+	io->head = kmalloc(size, GFP_KERNEL);
+	if (io->head != NULL) {
+		io->truesize	= size;
+		io->in_frame	= FALSE;
+		io->state	= OUTSIDE_FRAME;
+		io->data	= io->head;
+	}
+	return io->head ? 0 : -ENOMEM;
+}
+
+/*
+ * Set the IrDA communications speed.
+ */
+static int au1k_irda_set_speed(struct net_device *dev, int speed)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	volatile struct ring_dest *ptxd;
+	unsigned long control;
+	int ret = 0, timeout = 10, i;
+
+	if (speed == aup->speed)
+		return ret;
+
+	/* disable PHY first */
+	au1k_irda_plat_set_phy_mode(aup, AU1000_IRDA_PHY_MODE_OFF);
+	irda_write(aup, IR_STATUS, irda_read(aup, IR_STATUS) & ~IR_PHYEN);
+
+	/* disable RX/TX */
+	irda_write(aup, IR_CONFIG_1,
+	    irda_read(aup, IR_CONFIG_1) & ~(IR_RX_ENABLE | IR_TX_ENABLE));
+	msleep(20);
+	while (irda_read(aup, IR_STATUS) & (IR_RX_STATUS | IR_TX_STATUS)) {
+		msleep(20);
+		if (!timeout--) {
+			printk(KERN_ERR "%s: rx/tx disable timeout\n",
+					dev->name);
+			break;
+		}
+	}
+
+	/* disable DMA */
+	irda_write(aup, IR_CONFIG_1,
+		   irda_read(aup, IR_CONFIG_1) & ~IR_DMA_ENABLE);
+	msleep(20);
+
+	/* After we disable tx/rx. the index pointers go back to zero. */
+	aup->tx_head = aup->tx_tail = aup->rx_head = 0;
+	for (i = 0; i < NUM_IR_DESC; i++) {
+		ptxd = aup->tx_ring[i];
+		ptxd->flags = 0;
+		ptxd->count_0 = 0;
+		ptxd->count_1 = 0;
+	}
+
+	for (i = 0; i < NUM_IR_DESC; i++) {
+		ptxd = aup->rx_ring[i];
+		ptxd->count_0 = 0;
+		ptxd->count_1 = 0;
+		ptxd->flags = AU_OWN;
+	}
+
+	if (speed == 4000000)
+		au1k_irda_plat_set_phy_mode(aup, AU1000_IRDA_PHY_MODE_FIR);
+	else
+		au1k_irda_plat_set_phy_mode(aup, AU1000_IRDA_PHY_MODE_SIR);
+
+	switch (speed) {
+	case 9600:
+		irda_write(aup, IR_WRITE_PHY_CONFIG, IR_BR(11) | IR_PW(12));
+		irda_write(aup, IR_CONFIG_1, IR_SIR_MODE);
+		break;
+	case 19200:
+		irda_write(aup, IR_WRITE_PHY_CONFIG, IR_BR(5) | IR_PW(12));
+		irda_write(aup, IR_CONFIG_1, IR_SIR_MODE);
+		break;
+	case 38400:
+		irda_write(aup, IR_WRITE_PHY_CONFIG, IR_BR(2) | IR_PW(12));
+		irda_write(aup, IR_CONFIG_1, IR_SIR_MODE);
+		break;
+	case 57600:
+		irda_write(aup, IR_WRITE_PHY_CONFIG, IR_BR(1) | IR_PW(12));
+		irda_write(aup, IR_CONFIG_1, IR_SIR_MODE);
+		break;
+	case 115200:
+		irda_write(aup, IR_WRITE_PHY_CONFIG, IR_PW(12));
+		irda_write(aup, IR_CONFIG_1, IR_SIR_MODE);
+		break;
+	case 4000000:
+		irda_write(aup, IR_WRITE_PHY_CONFIG, IR_P(15));
+		irda_write(aup, IR_CONFIG_1, IR_FIR | IR_DMA_ENABLE |
+				IR_RX_ENABLE);
+		break;
+	default:
+		printk(KERN_ERR "%s unsupported speed %x\n", dev->name, speed);
+		ret = -EINVAL;
+		break;
+	}
+
+	aup->speed = speed;
+	irda_write(aup, IR_STATUS, irda_read(aup, IR_STATUS) | IR_PHYEN);
+
+	control = irda_read(aup, IR_STATUS);
+	irda_write(aup, IR_RING_PROMPT, 0);
+
+	if (control & (1 << 14)) {
+		printk(KERN_ERR "%s: configuration error\n", dev->name);
+	} else {
+		if (control & (1 << 11))
+			printk(KERN_DEBUG "%s Valid SIR config\n", dev->name);
+		if (control & (1 << 12))
+			printk(KERN_DEBUG "%s Valid MIR config\n", dev->name);
+		if (control & (1 << 13))
+			printk(KERN_DEBUG "%s Valid FIR config\n", dev->name);
+		if (control & (1 << 10))
+			printk(KERN_DEBUG "%s TX enabled\n", dev->name);
+		if (control & (1 << 9))
+			printk(KERN_DEBUG "%s RX enabled\n", dev->name);
+	}
+
+	return ret;
+}
+
+static void update_rx_stats(struct net_device *dev, u32 status, u32 count)
+{
+	struct net_device_stats *ps = &dev->stats;
+
+	ps->rx_packets++;
+
+	if (status & IR_RX_ERROR) {
+		ps->rx_errors++;
+		if (status & (IR_PHY_ERROR | IR_FIFO_OVER))
+			ps->rx_missed_errors++;
+		if (status & IR_MAX_LEN)
+			ps->rx_length_errors++;
+		if (status & IR_CRC_ERROR)
+			ps->rx_crc_errors++;
+	} else
+		ps->rx_bytes += count;
+}
+
+static void update_tx_stats(struct net_device *dev, u32 status, u32 pkt_len)
+{
+	struct net_device_stats *ps = &dev->stats;
+
+	ps->tx_packets++;
+	ps->tx_bytes += pkt_len;
+
+	if (status & IR_TX_ERROR) {
+		ps->tx_errors++;
+		ps->tx_aborted_errors++;
+	}
+}
+
+static void au1k_tx_ack(struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	volatile struct ring_dest *ptxd;
+
+	ptxd = aup->tx_ring[aup->tx_tail];
+	while (!(ptxd->flags & AU_OWN) && (aup->tx_tail != aup->tx_head)) {
+		update_tx_stats(dev, ptxd->flags,
+				(ptxd->count_1 << 8) | ptxd->count_0);
+		ptxd->count_0 = 0;
+		ptxd->count_1 = 0;
+		wmb();
+		aup->tx_tail = (aup->tx_tail + 1) & (NUM_IR_DESC - 1);
+		ptxd = aup->tx_ring[aup->tx_tail];
+
+		if (aup->tx_full) {
+			aup->tx_full = 0;
+			netif_wake_queue(dev);
+		}
+	}
+
+	if (aup->tx_tail == aup->tx_head) {
+		if (aup->newspeed) {
+			au1k_irda_set_speed(dev, aup->newspeed);
+			aup->newspeed = 0;
+		} else {
+			irda_write(aup, IR_CONFIG_1,
+			    irda_read(aup, IR_CONFIG_1) & ~IR_TX_ENABLE);
+			irda_write(aup, IR_CONFIG_1,
+			    irda_read(aup, IR_CONFIG_1) | IR_RX_ENABLE);
+			irda_write(aup, IR_RING_PROMPT, 0);
+		}
+	}
+}
+
+static int au1k_irda_rx(struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	volatile struct ring_dest *prxd;
+	struct sk_buff *skb;
+	struct db_dest *pDB;
+	u32 flags, count;
+
+	prxd = aup->rx_ring[aup->rx_head];
+	flags = prxd->flags;
+
+	while (!(flags & AU_OWN))  {
+		pDB = aup->rx_db_inuse[aup->rx_head];
+		count = (prxd->count_1 << 8) | prxd->count_0;
+		if (!(flags & IR_RX_ERROR)) {
+			/* good frame */
+			update_rx_stats(dev, flags, count);
+			skb = alloc_skb(count + 1, GFP_ATOMIC);
+			if (skb == NULL) {
+				dev->stats.rx_dropped++;
+				continue;
+			}
+			skb_reserve(skb, 1);
+			if (aup->speed == 4000000)
+				skb_put(skb, count);
+			else
+				skb_put(skb, count - 2);
+			skb_copy_to_linear_data(skb, (void *)pDB->vaddr,
+						count - 2);
+			skb->dev = dev;
+			skb_reset_mac_header(skb);
+			skb->protocol = htons(ETH_P_IRDA);
+			netif_rx(skb);
+			prxd->count_0 = 0;
+			prxd->count_1 = 0;
+		}
+		prxd->flags |= AU_OWN;
+		aup->rx_head = (aup->rx_head + 1) & (NUM_IR_DESC - 1);
+		irda_write(aup, IR_RING_PROMPT, 0);
+
+		/* next descriptor */
+		prxd = aup->rx_ring[aup->rx_head];
+		flags = prxd->flags;
+
+	}
+	return 0;
+}
+
+static irqreturn_t au1k_irda_interrupt(int dummy, void *dev_id)
+{
+	struct net_device *dev = dev_id;
+	struct au1k_private *aup = netdev_priv(dev);
+
+	irda_write(aup, IR_INT_CLEAR, 0); /* ack irda interrupts */
+
+	au1k_irda_rx(dev);
+	au1k_tx_ack(dev);
+
+	return IRQ_HANDLED;
+}
+
+static int au1k_init(struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	u32 enable, ring_address, phyck;
+	struct clk *c;
+	int i;
+
+	c = clk_get(NULL, "irda_clk");
+	if (IS_ERR(c))
+		return PTR_ERR(c);
+	i = clk_prepare_enable(c);
+	if (i) {
+		clk_put(c);
+		return i;
+	}
+
+	switch (clk_get_rate(c)) {
+	case 40000000:
+		phyck = IR_PHYCLK_40MHZ;
+		break;
+	case 48000000:
+		phyck = IR_PHYCLK_48MHZ;
+		break;
+	case 56000000:
+		phyck = IR_PHYCLK_56MHZ;
+		break;
+	case 64000000:
+		phyck = IR_PHYCLK_64MHZ;
+		break;
+	default:
+		clk_disable_unprepare(c);
+		clk_put(c);
+		return -EINVAL;
+	}
+	aup->irda_clk = c;
+
+	enable = IR_HC | IR_CE | IR_C;
+#ifndef CONFIG_CPU_LITTLE_ENDIAN
+	enable |= IR_BE;
+#endif
+	aup->tx_head = 0;
+	aup->tx_tail = 0;
+	aup->rx_head = 0;
+
+	for (i = 0; i < NUM_IR_DESC; i++)
+		aup->rx_ring[i]->flags = AU_OWN;
+
+	irda_write(aup, IR_ENABLE, enable);
+	msleep(20);
+
+	/* disable PHY */
+	au1k_irda_plat_set_phy_mode(aup, AU1000_IRDA_PHY_MODE_OFF);
+	irda_write(aup, IR_STATUS, irda_read(aup, IR_STATUS) & ~IR_PHYEN);
+	msleep(20);
+
+	irda_write(aup, IR_MAX_PKT_LEN, MAX_BUF_SIZE);
+
+	ring_address = (u32)virt_to_phys((void *)aup->rx_ring[0]);
+	irda_write(aup, IR_RING_BASE_ADDR_H, ring_address >> 26);
+	irda_write(aup, IR_RING_BASE_ADDR_L, (ring_address >> 10) & 0xffff);
+
+	irda_write(aup, IR_RING_SIZE,
+				(RING_SIZE_64 << 8) | (RING_SIZE_64 << 12));
+
+	irda_write(aup, IR_CONFIG_2, phyck | IR_ONE_PIN);
+	irda_write(aup, IR_RING_ADDR_CMPR, 0);
+
+	au1k_irda_set_speed(dev, 9600);
+	return 0;
+}
+
+static int au1k_irda_start(struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	char hwname[32];
+	int retval;
+
+	retval = au1k_init(dev);
+	if (retval) {
+		printk(KERN_ERR "%s: error in au1k_init\n", dev->name);
+		return retval;
+	}
+
+	retval = request_irq(aup->irq_tx, &au1k_irda_interrupt, 0,
+			     dev->name, dev);
+	if (retval) {
+		printk(KERN_ERR "%s: unable to get IRQ %d\n",
+				dev->name, dev->irq);
+		return retval;
+	}
+	retval = request_irq(aup->irq_rx, &au1k_irda_interrupt, 0,
+			     dev->name, dev);
+	if (retval) {
+		free_irq(aup->irq_tx, dev);
+		printk(KERN_ERR "%s: unable to get IRQ %d\n",
+				dev->name, dev->irq);
+		return retval;
+	}
+
+	/* Give self a hardware name */
+	sprintf(hwname, "Au1000 SIR/FIR");
+	aup->irlap = irlap_open(dev, &aup->qos, hwname);
+	netif_start_queue(dev);
+
+	/* int enable */
+	irda_write(aup, IR_CONFIG_2, irda_read(aup, IR_CONFIG_2) | IR_IEN);
+
+	/* power up */
+	au1k_irda_plat_set_phy_mode(aup, AU1000_IRDA_PHY_MODE_SIR);
+
+	return 0;
+}
+
+static int au1k_irda_stop(struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+
+	au1k_irda_plat_set_phy_mode(aup, AU1000_IRDA_PHY_MODE_OFF);
+
+	/* disable interrupts */
+	irda_write(aup, IR_CONFIG_2, irda_read(aup, IR_CONFIG_2) & ~IR_IEN);
+	irda_write(aup, IR_CONFIG_1, 0);
+	irda_write(aup, IR_ENABLE, 0); /* disable clock */
+
+	if (aup->irlap) {
+		irlap_close(aup->irlap);
+		aup->irlap = NULL;
+	}
+
+	netif_stop_queue(dev);
+
+	/* disable the interrupt */
+	free_irq(aup->irq_tx, dev);
+	free_irq(aup->irq_rx, dev);
+
+	clk_disable_unprepare(aup->irda_clk);
+	clk_put(aup->irda_clk);
+
+	return 0;
+}
+
+/*
+ * Au1000 transmit routine.
+ */
+static int au1k_irda_hard_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	int speed = irda_get_next_speed(skb);
+	volatile struct ring_dest *ptxd;
+	struct db_dest *pDB;
+	u32 len, flags;
+
+	if (speed != aup->speed && speed != -1)
+		aup->newspeed = speed;
+
+	if ((skb->len == 0) && (aup->newspeed)) {
+		if (aup->tx_tail == aup->tx_head) {
+			au1k_irda_set_speed(dev, speed);
+			aup->newspeed = 0;
+		}
+		dev_kfree_skb(skb);
+		return NETDEV_TX_OK;
+	}
+
+	ptxd = aup->tx_ring[aup->tx_head];
+	flags = ptxd->flags;
+
+	if (flags & AU_OWN) {
+		printk(KERN_DEBUG "%s: tx_full\n", dev->name);
+		netif_stop_queue(dev);
+		aup->tx_full = 1;
+		return 1;
+	} else if (((aup->tx_head + 1) & (NUM_IR_DESC - 1)) == aup->tx_tail) {
+		printk(KERN_DEBUG "%s: tx_full\n", dev->name);
+		netif_stop_queue(dev);
+		aup->tx_full = 1;
+		return 1;
+	}
+
+	pDB = aup->tx_db_inuse[aup->tx_head];
+
+#if 0
+	if (irda_read(aup, IR_RX_BYTE_CNT) != 0) {
+		printk(KERN_DEBUG "tx warning: rx byte cnt %x\n",
+				irda_read(aup, IR_RX_BYTE_CNT));
+	}
+#endif
+
+	if (aup->speed == 4000000) {
+		/* FIR */
+		skb_copy_from_linear_data(skb, (void *)pDB->vaddr, skb->len);
+		ptxd->count_0 = skb->len & 0xff;
+		ptxd->count_1 = (skb->len >> 8) & 0xff;
+	} else {
+		/* SIR */
+		len = async_wrap_skb(skb, (u8 *)pDB->vaddr, MAX_BUF_SIZE);
+		ptxd->count_0 = len & 0xff;
+		ptxd->count_1 = (len >> 8) & 0xff;
+		ptxd->flags |= IR_DIS_CRC;
+	}
+	ptxd->flags |= AU_OWN;
+	wmb();
+
+	irda_write(aup, IR_CONFIG_1,
+		   irda_read(aup, IR_CONFIG_1) | IR_TX_ENABLE);
+	irda_write(aup, IR_RING_PROMPT, 0);
+
+	dev_kfree_skb(skb);
+	aup->tx_head = (aup->tx_head + 1) & (NUM_IR_DESC - 1);
+	return NETDEV_TX_OK;
+}
+
+/*
+ * The Tx ring has been full longer than the watchdog timeout
+ * value. The transmitter must be hung?
+ */
+static void au1k_tx_timeout(struct net_device *dev)
+{
+	u32 speed;
+	struct au1k_private *aup = netdev_priv(dev);
+
+	printk(KERN_ERR "%s: tx timeout\n", dev->name);
+	speed = aup->speed;
+	aup->speed = 0;
+	au1k_irda_set_speed(dev, speed);
+	aup->tx_full = 0;
+	netif_wake_queue(dev);
+}
+
+static int au1k_irda_ioctl(struct net_device *dev, struct ifreq *ifreq, int cmd)
+{
+	struct if_irda_req *rq = (struct if_irda_req *)ifreq;
+	struct au1k_private *aup = netdev_priv(dev);
+	int ret = -EOPNOTSUPP;
+
+	switch (cmd) {
+	case SIOCSBANDWIDTH:
+		if (capable(CAP_NET_ADMIN)) {
+			/*
+			 * We are unable to set the speed if the
+			 * device is not running.
+			 */
+			if (aup->open)
+				ret = au1k_irda_set_speed(dev,
+						rq->ifr_baudrate);
+			else {
+				printk(KERN_ERR "%s ioctl: !netif_running\n",
+						dev->name);
+				ret = 0;
+			}
+		}
+		break;
+
+	case SIOCSMEDIABUSY:
+		ret = -EPERM;
+		if (capable(CAP_NET_ADMIN)) {
+			irda_device_set_media_busy(dev, TRUE);
+			ret = 0;
+		}
+		break;
+
+	case SIOCGRECEIVING:
+		rq->ifr_receiving = 0;
+		break;
+	default:
+		break;
+	}
+	return ret;
+}
+
+static const struct net_device_ops au1k_irda_netdev_ops = {
+	.ndo_open		= au1k_irda_start,
+	.ndo_stop		= au1k_irda_stop,
+	.ndo_start_xmit		= au1k_irda_hard_xmit,
+	.ndo_tx_timeout		= au1k_tx_timeout,
+	.ndo_do_ioctl		= au1k_irda_ioctl,
+};
+
+static int au1k_irda_net_init(struct net_device *dev)
+{
+	struct au1k_private *aup = netdev_priv(dev);
+	struct db_dest *pDB, *pDBfree;
+	int i, err, retval = 0;
+	dma_addr_t temp;
+
+	err = au1k_irda_init_iobuf(&aup->rx_buff, 14384);
+	if (err)
+		goto out1;
+
+	dev->netdev_ops = &au1k_irda_netdev_ops;
+
+	irda_init_max_qos_capabilies(&aup->qos);
+
+	/* The only value we must override it the baudrate */
+	aup->qos.baud_rate.bits = IR_9600 | IR_19200 | IR_38400 |
+		IR_57600 | IR_115200 | IR_576000 | (IR_4000000 << 8);
+
+	aup->qos.min_turn_time.bits = qos_mtt_bits;
+	irda_qos_bits_to_value(&aup->qos);
+
+	retval = -ENOMEM;
+
+	/* Tx ring follows rx ring + 512 bytes */
+	/* we need a 1k aligned buffer */
+	aup->rx_ring[0] = (struct ring_dest *)
+		dma_alloc(2 * MAX_NUM_IR_DESC * (sizeof(struct ring_dest)),
+			  &temp);
+	if (!aup->rx_ring[0])
+		goto out2;
+
+	/* allocate the data buffers */
+	aup->db[0].vaddr =
+		dma_alloc(MAX_BUF_SIZE * 2 * NUM_IR_DESC, &temp);
+	if (!aup->db[0].vaddr)
+		goto out3;
+
+	setup_hw_rings(aup, (u32)aup->rx_ring[0], (u32)aup->rx_ring[0] + 512);
+
+	pDBfree = NULL;
+	pDB = aup->db;
+	for (i = 0; i < (2 * NUM_IR_DESC); i++) {
+		pDB->pnext = pDBfree;
+		pDBfree = pDB;
+		pDB->vaddr =
+		       (u32 *)((unsigned)aup->db[0].vaddr + (MAX_BUF_SIZE * i));
+		pDB->dma_addr = (dma_addr_t)virt_to_bus(pDB->vaddr);
+		pDB++;
+	}
+	aup->pDBfree = pDBfree;
+
+	/* attach a data buffer to each descriptor */
+	for (i = 0; i < NUM_IR_DESC; i++) {
+		pDB = GetFreeDB(aup);
+		if (!pDB)
+			goto out3;
+		aup->rx_ring[i]->addr_0 = (u8)(pDB->dma_addr & 0xff);
+		aup->rx_ring[i]->addr_1 = (u8)((pDB->dma_addr >>  8) & 0xff);
+		aup->rx_ring[i]->addr_2 = (u8)((pDB->dma_addr >> 16) & 0xff);
+		aup->rx_ring[i]->addr_3 = (u8)((pDB->dma_addr >> 24) & 0xff);
+		aup->rx_db_inuse[i] = pDB;
+	}
+	for (i = 0; i < NUM_IR_DESC; i++) {
+		pDB = GetFreeDB(aup);
+		if (!pDB)
+			goto out3;
+		aup->tx_ring[i]->addr_0 = (u8)(pDB->dma_addr & 0xff);
+		aup->tx_ring[i]->addr_1 = (u8)((pDB->dma_addr >>  8) & 0xff);
+		aup->tx_ring[i]->addr_2 = (u8)((pDB->dma_addr >> 16) & 0xff);
+		aup->tx_ring[i]->addr_3 = (u8)((pDB->dma_addr >> 24) & 0xff);
+		aup->tx_ring[i]->count_0 = 0;
+		aup->tx_ring[i]->count_1 = 0;
+		aup->tx_ring[i]->flags = 0;
+		aup->tx_db_inuse[i] = pDB;
+	}
+
+	return 0;
+
+out3:
+	dma_free((void *)aup->rx_ring[0],
+		2 * MAX_NUM_IR_DESC * (sizeof(struct ring_dest)));
+out2:
+	kfree(aup->rx_buff.head);
+out1:
+	printk(KERN_ERR "au1k_irda_net_init() failed.  Returns %d\n", retval);
+	return retval;
+}
+
+static int au1k_irda_probe(struct platform_device *pdev)
+{
+	struct au1k_private *aup;
+	struct net_device *dev;
+	struct resource *r;
+	struct clk *c;
+	int err;
+
+	dev = alloc_irdadev(sizeof(struct au1k_private));
+	if (!dev)
+		return -ENOMEM;
+
+	aup = netdev_priv(dev);
+
+	aup->platdata = pdev->dev.platform_data;
+
+	err = -EINVAL;
+	r = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!r)
+		goto out;
+
+	aup->irq_tx = r->start;
+
+	r = platform_get_resource(pdev, IORESOURCE_IRQ, 1);
+	if (!r)
+		goto out;
+
+	aup->irq_rx = r->start;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!r)
+		goto out;
+
+	err = -EBUSY;
+	aup->ioarea = request_mem_region(r->start, resource_size(r),
+					 pdev->name);
+	if (!aup->ioarea)
+		goto out;
+
+	/* bail out early if clock doesn't exist */
+	c = clk_get(NULL, "irda_clk");
+	if (IS_ERR(c)) {
+		err = PTR_ERR(c);
+		goto out;
+	}
+	clk_put(c);
+
+	aup->iobase = ioremap_nocache(r->start, resource_size(r));
+	if (!aup->iobase)
+		goto out2;
+
+	dev->irq = aup->irq_rx;
+
+	err = au1k_irda_net_init(dev);
+	if (err)
+		goto out3;
+	err = register_netdev(dev);
+	if (err)
+		goto out4;
+
+	platform_set_drvdata(pdev, dev);
+
+	printk(KERN_INFO "IrDA: Registered device %s\n", dev->name);
+	return 0;
+
+out4:
+	dma_free((void *)aup->db[0].vaddr,
+		MAX_BUF_SIZE * 2 * NUM_IR_DESC);
+	dma_free((void *)aup->rx_ring[0],
+		2 * MAX_NUM_IR_DESC * (sizeof(struct ring_dest)));
+	kfree(aup->rx_buff.head);
+out3:
+	iounmap(aup->iobase);
+out2:
+	release_resource(aup->ioarea);
+	kfree(aup->ioarea);
+out:
+	free_netdev(dev);
+	return err;
+}
+
+static int au1k_irda_remove(struct platform_device *pdev)
+{
+	struct net_device *dev = platform_get_drvdata(pdev);
+	struct au1k_private *aup = netdev_priv(dev);
+
+	unregister_netdev(dev);
+
+	dma_free((void *)aup->db[0].vaddr,
+		MAX_BUF_SIZE * 2 * NUM_IR_DESC);
+	dma_free((void *)aup->rx_ring[0],
+		2 * MAX_NUM_IR_DESC * (sizeof(struct ring_dest)));
+	kfree(aup->rx_buff.head);
+
+	iounmap(aup->iobase);
+	release_resource(aup->ioarea);
+	kfree(aup->ioarea);
+
+	free_netdev(dev);
+
+	return 0;
+}
+
+static struct platform_driver au1k_irda_driver = {
+	.driver	= {
+		.name	= "au1000-irda",
+	},
+	.probe		= au1k_irda_probe,
+	.remove		= au1k_irda_remove,
+};
+
+module_platform_driver(au1k_irda_driver);
+
+MODULE_AUTHOR("Pete Popov <ppopov@mvista.com>");
+MODULE_DESCRIPTION("Au1000 IrDA Device Driver");