net: Distributed Switch Architecture protocol support

Distributed Switch Architecture is a protocol for managing hardware
switch chips.  It consists of a set of MII management registers and
commands to configure the switch, and an ethernet header format to
signal which of the ports of the switch a packet was received from
or is intended to be sent to.

The switches that this driver supports are typically embedded in
access points and routers, and a typical setup with a DSA switch
looks something like this:

	+-----------+       +-----------+
	|           | RGMII |           |
	|           +-------+           +------ 1000baseT MDI ("WAN")
	|           |       |  6-port   +------ 1000baseT MDI ("LAN1")
	|    CPU    |       |  ethernet +------ 1000baseT MDI ("LAN2")
	|           |MIImgmt|  switch   +------ 1000baseT MDI ("LAN3")
	|           +-------+  w/5 PHYs +------ 1000baseT MDI ("LAN4")
	|           |       |           |
	+-----------+       +-----------+

The switch driver presents each port on the switch as a separate
network interface to Linux, polls the switch to maintain software
link state of those ports, forwards MII management interface
accesses to those network interfaces (e.g. as done by ethtool) to
the switch, and exposes the switch's hardware statistics counters
via the appropriate Linux kernel interfaces.

This initial patch supports the MII management interface register
layout of the Marvell 88E6123, 88E6161 and 88E6165 switch chips, and
supports the "Ethertype DSA" packet tagging format.

(There is no officially registered ethertype for the Ethertype DSA
packet format, so we just grab a random one.  The ethertype to use
is programmed into the switch, and the switch driver uses the value
of ETH_P_EDSA for this, so this define can be changed at any time in
the future if the one we chose is allocated to another protocol or
if Ethertype DSA gets its own officially registered ethertype, and
everything will continue to work.)
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Tested-by: Nicolas Pitre <nico@marvell.com>
Tested-by: Byron Bradley <byron.bbradley@gmail.com>
Tested-by: Tim Ellis <tim.ellis@mac.com>
Tested-by: Peter van Valderen <linux@ddcrew.com>
Tested-by: Dirk Teurlings <dirk@upexia.nl>
Signed-off-by: David S. Miller <davem@davemloft.net>

include/linux/if_ether.h

@@ -77,6 +77,7 @@
 #define ETH_P_PAE	0x888E		/* Port Access Entity (IEEE 802.1X) */
 #define ETH_P_AOE	0x88A2		/* ATA over Ethernet		*/
 #define ETH_P_TIPC	0x88CA		/* TIPC 			*/
+#define ETH_P_EDSA	0xDADA		/* Ethertype DSA [ NOT AN OFFICIALLY REGISTERED ID ] */
 
 /*
  *	Non DIX types. Won't clash for 1500 types.

include/linux/netdevice.h

@@ -607,6 +607,9 @@ struct net_device
 
 	/* Protocol specific pointers */
 	
+#ifdef CONFIG_NET_DSA
+	void			*dsa_ptr;	/* dsa specific data */
+#endif
 	void 			*atalk_ptr;	/* AppleTalk link 	*/
 	void			*ip_ptr;	/* IPv4 specific data	*/  
 	void                    *dn_ptr;        /* DECnet specific data */

include/net/dsa.h

+/*
+ * include/net/dsa.h - Driver for Distributed Switch Architecture switch chips
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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.
+ */
+
+#ifndef __LINUX_NET_DSA_H
+#define __LINUX_NET_DSA_H
+
+#define DSA_MAX_PORTS	12
+
+struct dsa_platform_data {
+	/*
+	 * Reference to a Linux network interface that connects
+	 * to the switch chip.
+	 */
+	struct device	*netdev;
+
+	/*
+	 * How to access the switch configuration registers, and
+	 * the names of the switch ports (use "cpu" to designate
+	 * the switch port that the cpu is connected to).
+	 */
+	struct device	*mii_bus;
+	int		sw_addr;
+	char		*port_names[DSA_MAX_PORTS];
+};
+
+
+#endif

net/Kconfig

@@ -180,6 +180,7 @@ source "net/tipc/Kconfig"
 source "net/atm/Kconfig"
 source "net/802/Kconfig"
 source "net/bridge/Kconfig"
+source "net/dsa/Kconfig"
 source "net/8021q/Kconfig"
 source "net/decnet/Kconfig"
 source "net/llc/Kconfig"

net/Makefile

@@ -26,6 +26,7 @@ obj-$(CONFIG_PACKET)		+= packet/
 obj-$(CONFIG_NET_KEY)		+= key/
 obj-$(CONFIG_NET_SCHED)		+= sched/
 obj-$(CONFIG_BRIDGE)		+= bridge/
+obj-$(CONFIG_NET_DSA)		+= dsa/
 obj-$(CONFIG_IPX)		+= ipx/
 obj-$(CONFIG_ATALK)		+= appletalk/
 obj-$(CONFIG_WAN_ROUTER)	+= wanrouter/

net/dsa/Kconfig

+menuconfig NET_DSA
+	bool "Distributed Switch Architecture support"
+	default n
+	depends on EXPERIMENTAL
+	---help---
+	  This allows you to use hardware switch chips that use
+	  the Distributed Switch Architecture.
+
+
+if NET_DSA
+
+# tagging formats
+config NET_DSA_TAG_EDSA
+	bool
+	default n
+
+
+# switch drivers
+config NET_DSA_MV88E6XXX
+	bool
+	default n
+
+config NET_DSA_MV88E6123_61_65
+	bool "Marvell 88E6123/6161/6165 ethernet switch chip support"
+	select NET_DSA_MV88E6XXX
+	select NET_DSA_TAG_EDSA
+	---help---
+	  This enables support for the Marvell 88E6123/6161/6165
+	  ethernet switch chips.
+
+endif

net/dsa/Makefile

+# tagging formats
+obj-$(CONFIG_NET_DSA_TAG_EDSA) += tag_edsa.o
+
+# switch drivers
+obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx.o
+obj-$(CONFIG_NET_DSA_MV88E6123_61_65) += mv88e6123_61_65.o
+
+# the core
+obj-$(CONFIG_NET_DSA) += dsa.o slave.o

net/dsa/dsa.c

+/*
+ * net/dsa/dsa.c - Hardware switch handling
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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/list.h>
+#include <linux/netdevice.h>
+#include <linux/platform_device.h>
+#include <net/dsa.h>
+#include "dsa_priv.h"
+
+char dsa_driver_version[] = "0.1";
+
+
+/* switch driver registration ***********************************************/
+static DEFINE_MUTEX(dsa_switch_drivers_mutex);
+static LIST_HEAD(dsa_switch_drivers);
+
+void register_switch_driver(struct dsa_switch_driver *drv)
+{
+	mutex_lock(&dsa_switch_drivers_mutex);
+	list_add_tail(&drv->list, &dsa_switch_drivers);
+	mutex_unlock(&dsa_switch_drivers_mutex);
+}
+
+void unregister_switch_driver(struct dsa_switch_driver *drv)
+{
+	mutex_lock(&dsa_switch_drivers_mutex);
+	list_del_init(&drv->list);
+	mutex_unlock(&dsa_switch_drivers_mutex);
+}
+
+static struct dsa_switch_driver *
+dsa_switch_probe(struct mii_bus *bus, int sw_addr, char **_name)
+{
+	struct dsa_switch_driver *ret;
+	struct list_head *list;
+	char *name;
+
+	ret = NULL;
+	name = NULL;
+
+	mutex_lock(&dsa_switch_drivers_mutex);
+	list_for_each(list, &dsa_switch_drivers) {
+		struct dsa_switch_driver *drv;
+
+		drv = list_entry(list, struct dsa_switch_driver, list);
+
+		name = drv->probe(bus, sw_addr);
+		if (name != NULL) {
+			ret = drv;
+			break;
+		}
+	}
+	mutex_unlock(&dsa_switch_drivers_mutex);
+
+	*_name = name;
+
+	return ret;
+}
+
+
+/* basic switch operations **************************************************/
+static struct dsa_switch *
+dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
+		 struct mii_bus *bus, struct net_device *dev)
+{
+	struct dsa_switch *ds;
+	int ret;
+	struct dsa_switch_driver *drv;
+	char *name;
+	int i;
+
+	/*
+	 * Probe for switch model.
+	 */
+	drv = dsa_switch_probe(bus, pd->sw_addr, &name);
+	if (drv == NULL) {
+		printk(KERN_ERR "%s: could not detect attached switch\n",
+		       dev->name);
+		return ERR_PTR(-EINVAL);
+	}
+	printk(KERN_INFO "%s: detected a %s switch\n", dev->name, name);
+
+
+	/*
+	 * Allocate and initialise switch state.
+	 */
+	ds = kzalloc(sizeof(*ds) + drv->priv_size, GFP_KERNEL);
+	if (ds == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	ds->pd = pd;
+	ds->master_netdev = dev;
+	ds->master_mii_bus = bus;
+
+	ds->drv = drv;
+	ds->tag_protocol = drv->tag_protocol;
+
+
+	/*
+	 * Validate supplied switch configuration.
+	 */
+	ds->cpu_port = -1;
+	for (i = 0; i < DSA_MAX_PORTS; i++) {
+		char *name;
+
+		name = pd->port_names[i];
+		if (name == NULL)
+			continue;
+
+		if (!strcmp(name, "cpu")) {
+			if (ds->cpu_port != -1) {
+				printk(KERN_ERR "multiple cpu ports?!\n");
+				ret = -EINVAL;
+				goto out;
+			}
+			ds->cpu_port = i;
+		} else {
+			ds->valid_port_mask |= 1 << i;
+		}
+	}
+
+	if (ds->cpu_port == -1) {
+		printk(KERN_ERR "no cpu port?!\n");
+		ret = -EINVAL;
+		goto out;
+	}
+
+
+	/*
+	 * If we use a tagging format that doesn't have an ethertype
+	 * field, make sure that all packets from this point on get
+	 * sent to the tag format's receive function.  (Which will
+	 * discard received packets until we set ds->ports[] below.)
+	 */
+	wmb();
+	dev->dsa_ptr = (void *)ds;
+
+
+	/*
+	 * Do basic register setup.
+	 */
+	ret = drv->setup(ds);
+	if (ret < 0)
+		goto out;
+
+	ret = drv->set_addr(ds, dev->dev_addr);
+	if (ret < 0)
+		goto out;
+
+	ds->slave_mii_bus = mdiobus_alloc();
+	if (ds->slave_mii_bus == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+	dsa_slave_mii_bus_init(ds);
+
+	ret = mdiobus_register(ds->slave_mii_bus);
+	if (ret < 0)
+		goto out_free;
+
+
+	/*
+	 * Create network devices for physical switch ports.
+	 */
+	wmb();
+	for (i = 0; i < DSA_MAX_PORTS; i++) {
+		struct net_device *slave_dev;
+
+		if (!(ds->valid_port_mask & (1 << i)))
+			continue;
+
+		slave_dev = dsa_slave_create(ds, parent, i, pd->port_names[i]);
+		if (slave_dev == NULL) {
+			printk(KERN_ERR "%s: can't create dsa slave "
+			       "device for port %d(%s)\n",
+			       dev->name, i, pd->port_names[i]);
+			continue;
+		}
+
+		ds->ports[i] = slave_dev;
+	}
+
+	return ds;
+
+out_free:
+	mdiobus_free(ds->slave_mii_bus);
+out:
+	dev->dsa_ptr = NULL;
+	kfree(ds);
+	return ERR_PTR(ret);
+}
+
+static void dsa_switch_destroy(struct dsa_switch *ds)
+{
+}
+
+
+/* link polling *************************************************************/
+static void dsa_link_poll_work(struct work_struct *ugly)
+{
+	struct dsa_switch *ds;
+
+	ds = container_of(ugly, struct dsa_switch, link_poll_work);
+
+	ds->drv->poll_link(ds);
+	mod_timer(&ds->link_poll_timer, round_jiffies(jiffies + HZ));
+}
+
+static void dsa_link_poll_timer(unsigned long _ds)
+{
+	struct dsa_switch *ds = (void *)_ds;
+
+	schedule_work(&ds->link_poll_work);
+}
+
+
+/* platform driver init and cleanup *****************************************/
+static int dev_is_class(struct device *dev, void *class)
+{
+	if (dev->class != NULL && !strcmp(dev->class->name, class))
+		return 1;
+
+	return 0;
+}
+
+static struct device *dev_find_class(struct device *parent, char *class)
+{
+	if (dev_is_class(parent, class)) {
+		get_device(parent);
+		return parent;
+	}
+
+	return device_find_child(parent, class, dev_is_class);
+}
+
+static struct mii_bus *dev_to_mii_bus(struct device *dev)
+{
+	struct device *d;
+
+	d = dev_find_class(dev, "mdio_bus");
+	if (d != NULL) {
+		struct mii_bus *bus;
+
+		bus = to_mii_bus(d);
+		put_device(d);
+
+		return bus;
+	}
+
+	return NULL;
+}
+
+static struct net_device *dev_to_net_device(struct device *dev)
+{
+	struct device *d;
+
+	d = dev_find_class(dev, "net");
+	if (d != NULL) {
+		struct net_device *nd;
+
+		nd = to_net_dev(d);
+		dev_hold(nd);
+		put_device(d);
+
+		return nd;
+	}
+
+	return NULL;
+}
+
+static int dsa_probe(struct platform_device *pdev)
+{
+	static int dsa_version_printed;
+	struct dsa_platform_data *pd = pdev->dev.platform_data;
+	struct net_device *dev;
+	struct mii_bus *bus;
+	struct dsa_switch *ds;
+
+	if (!dsa_version_printed++)
+		printk(KERN_NOTICE "Distributed Switch Architecture "
+			"driver version %s\n", dsa_driver_version);
+
+	if (pd == NULL || pd->mii_bus == NULL || pd->netdev == NULL)
+		return -EINVAL;
+
+	bus = dev_to_mii_bus(pd->mii_bus);
+	if (bus == NULL)
+		return -EINVAL;
+
+	dev = dev_to_net_device(pd->netdev);
+	if (dev == NULL)
+		return -EINVAL;
+
+	if (dev->dsa_ptr != NULL) {
+		dev_put(dev);
+		return -EEXIST;
+	}
+
+	ds = dsa_switch_setup(&pdev->dev, pd, bus, dev);
+	if (IS_ERR(ds)) {
+		dev_put(dev);
+		return PTR_ERR(ds);
+	}
+
+	if (ds->drv->poll_link != NULL) {
+		INIT_WORK(&ds->link_poll_work, dsa_link_poll_work);
+		init_timer(&ds->link_poll_timer);
+		ds->link_poll_timer.data = (unsigned long)ds;
+		ds->link_poll_timer.function = dsa_link_poll_timer;
+		ds->link_poll_timer.expires = round_jiffies(jiffies + HZ);
+		add_timer(&ds->link_poll_timer);
+	}
+
+	platform_set_drvdata(pdev, ds);
+
+	return 0;
+}
+
+static int dsa_remove(struct platform_device *pdev)
+{
+	struct dsa_switch *ds = platform_get_drvdata(pdev);
+
+	if (ds->drv->poll_link != NULL)
+		del_timer_sync(&ds->link_poll_timer);
+
+	flush_scheduled_work();
+
+	dsa_switch_destroy(ds);
+
+	return 0;
+}
+
+static void dsa_shutdown(struct platform_device *pdev)
+{
+}
+
+static struct platform_driver dsa_driver = {
+	.probe		= dsa_probe,
+	.remove		= dsa_remove,
+	.shutdown	= dsa_shutdown,
+	.driver = {
+		.name	= "dsa",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init dsa_init_module(void)
+{
+	return platform_driver_register(&dsa_driver);
+}
+module_init(dsa_init_module);
+
+static void __exit dsa_cleanup_module(void)
+{
+	platform_driver_unregister(&dsa_driver);
+}
+module_exit(dsa_cleanup_module);
+
+MODULE_AUTHOR("Lennert Buytenhek <buytenh@wantstofly.org>")
+MODULE_DESCRIPTION("Driver for Distributed Switch Architecture switch chips");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:dsa");

net/dsa/dsa_priv.h

+/*
+ * net/dsa/dsa_priv.h - Hardware switch handling
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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.
+ */
+
+#ifndef __DSA_PRIV_H
+#define __DSA_PRIV_H
+
+#include <linux/list.h>
+#include <linux/phy.h>
+#include <linux/timer.h>
+#include <linux/workqueue.h>
+#include <net/dsa.h>
+
+struct dsa_switch {
+	/*
+	 * Configuration data for the platform device that owns
+	 * this dsa switch instance.
+	 */
+	struct dsa_platform_data	*pd;
+
+	/*
+	 * References to network device and mii bus to use.
+	 */
+	struct net_device		*master_netdev;
+	struct mii_bus			*master_mii_bus;
+
+	/*
+	 * The used switch driver and frame tagging type.
+	 */
+	struct dsa_switch_driver	*drv;
+	__be16				tag_protocol;
+
+	/*
+	 * Slave mii_bus and devices for the individual ports.
+	 */
+	int				cpu_port;
+	u32				valid_port_mask;
+	struct mii_bus			*slave_mii_bus;
+	struct net_device		*ports[DSA_MAX_PORTS];
+
+	/*
+	 * Link state polling.
+	 */
+	struct work_struct		link_poll_work;
+	struct timer_list		link_poll_timer;
+};
+
+struct dsa_slave_priv {
+	struct net_device	*dev;
+	struct dsa_switch	*parent;
+	int			port;
+	struct phy_device	*phy;
+};
+
+struct dsa_switch_driver {
+	struct list_head	list;
+
+	__be16			tag_protocol;
+	int			priv_size;
+
+	/*
+	 * Probing and setup.
+	 */
+	char	*(*probe)(struct mii_bus *bus, int sw_addr);
+	int	(*setup)(struct dsa_switch *ds);
+	int	(*set_addr)(struct dsa_switch *ds, u8 *addr);
+
+	/*
+	 * Access to the switch's PHY registers.
+	 */
+	int	(*phy_read)(struct dsa_switch *ds, int port, int regnum);
+	int	(*phy_write)(struct dsa_switch *ds, int port,
+			     int regnum, u16 val);
+
+	/*
+	 * Link state polling and IRQ handling.
+	 */
+	void	(*poll_link)(struct dsa_switch *ds);
+
+	/*
+	 * ethtool hardware statistics.
+	 */
+	void	(*get_strings)(struct dsa_switch *ds, int port, uint8_t *data);
+	void	(*get_ethtool_stats)(struct dsa_switch *ds,
+				     int port, uint64_t *data);
+	int	(*get_sset_count)(struct dsa_switch *ds);
+};
+
+/* dsa.c */
+extern char dsa_driver_version[];
+void register_switch_driver(struct dsa_switch_driver *type);
+void unregister_switch_driver(struct dsa_switch_driver *type);
+
+/* slave.c */
+void dsa_slave_mii_bus_init(struct dsa_switch *ds);
+struct net_device *dsa_slave_create(struct dsa_switch *ds,
+				    struct device *parent,
+				    int port, char *name);
+
+/* tag_edsa.c */
+int edsa_xmit(struct sk_buff *skb, struct net_device *dev);
+
+
+#endif

net/dsa/mv88e6123_61_65.c

+/*
+ * net/dsa/mv88e6123_61_65.c - Marvell 88e6123/6161/6165 switch chip support
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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/list.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include "dsa_priv.h"
+#include "mv88e6xxx.h"
+
+static char *mv88e6123_61_65_probe(struct mii_bus *bus, int sw_addr)
+{
+	int ret;
+
+	ret = __mv88e6xxx_reg_read(bus, sw_addr, REG_PORT(0), 0x03);
+	if (ret >= 0) {
+		ret &= 0xfff0;
+		if (ret == 0x1210)
+			return "Marvell 88E6123";
+		if (ret == 0x1610)
+			return "Marvell 88E6161";
+		if (ret == 0x1650)
+			return "Marvell 88E6165";
+	}
+
+	return NULL;
+}
+
+static int mv88e6123_61_65_switch_reset(struct dsa_switch *ds)
+{
+	int i;
+	int ret;
+
+	/*
+	 * Set all ports to the disabled state.
+	 */
+	for (i = 0; i < 8; i++) {
+		ret = REG_READ(REG_PORT(i), 0x04);
+		REG_WRITE(REG_PORT(i), 0x04, ret & 0xfffc);
+	}
+
+	/*
+	 * Wait for transmit queues to drain.
+	 */
+	msleep(2);
+
+	/*
+	 * Reset the switch.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x04, 0xc400);
+
+	/*
+	 * Wait up to one second for reset to complete.
+	 */
+	for (i = 0; i < 1000; i++) {
+		ret = REG_READ(REG_GLOBAL, 0x00);
+		if ((ret & 0xc800) == 0xc800)
+			break;
+
+		msleep(1);
+	}
+	if (i == 1000)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+static int mv88e6123_61_65_setup_global(struct dsa_switch *ds)
+{
+	int ret;
+	int i;
+
+	/*
+	 * Disable the PHY polling unit (since there won't be any
+	 * external PHYs to poll), don't discard packets with
+	 * excessive collisions, and mask all interrupt sources.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x04, 0x0000);
+
+	/*
+	 * Set the default address aging time to 5 minutes, and
+	 * enable address learn messages to be sent to all message
+	 * ports.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x0a, 0x0148);
+
+	/*
+	 * Configure the priority mapping registers.
+	 */
+	ret = mv88e6xxx_config_prio(ds);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Configure the cpu port, and configure the cpu port as the
+	 * port to which ingress and egress monitor frames are to be
+	 * sent.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x1a, (ds->cpu_port * 0x1110));
+
+	/*
+	 * Disable remote management for now, and set the switch's
+	 * DSA device number to zero.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x1c, 0x0000);
+
+	/*
+	 * Send all frames with destination addresses matching
+	 * 01:80:c2:00:00:2x to the CPU port.
+	 */
+	REG_WRITE(REG_GLOBAL2, 0x02, 0xffff);
+
+	/*
+	 * Send all frames with destination addresses matching
+	 * 01:80:c2:00:00:0x to the CPU port.
+	 */
+	REG_WRITE(REG_GLOBAL2, 0x03, 0xffff);
+
+	/*
+	 * Disable the loopback filter, disable flow control
+	 * messages, disable flood broadcast override, disable
+	 * removing of provider tags, disable ATU age violation
+	 * interrupts, disable tag flow control, force flow
+	 * control priority to the highest, and send all special
+	 * multicast frames to the CPU at the highest priority.
+	 */
+	REG_WRITE(REG_GLOBAL2, 0x05, 0x00ff);
+
+	/*
+	 * Map all DSA device IDs to the CPU port.
+	 */
+	for (i = 0; i < 32; i++)
+		REG_WRITE(REG_GLOBAL2, 0x06, 0x8000 | (i << 8) | ds->cpu_port);
+
+	/*
+	 * Clear all trunk masks.
+	 */
+	for (i = 0; i < 8; i++)
+		REG_WRITE(REG_GLOBAL2, 0x07, 0x8000 | (i << 12) | 0xff);
+
+	/*
+	 * Clear all trunk mappings.
+	 */
+	for (i = 0; i < 16; i++)
+		REG_WRITE(REG_GLOBAL2, 0x08, 0x8000 | (i << 11));
+
+	/*
+	 * Disable ingress rate limiting by resetting all ingress
+	 * rate limit registers to their initial state.
+	 */
+	for (i = 0; i < 6; i++)
+		REG_WRITE(REG_GLOBAL2, 0x09, 0x9000 | (i << 8));
+
+	/*
+	 * Initialise cross-chip port VLAN table to reset defaults.
+	 */
+	REG_WRITE(REG_GLOBAL2, 0x0b, 0x9000);
+
+	/*
+	 * Clear the priority override table.
+	 */
+	for (i = 0; i < 16; i++)
+		REG_WRITE(REG_GLOBAL2, 0x0f, 0x8000 | (i << 8));
+
+	/* @@@ initialise AVB (22/23) watchdog (27) sdet (29) registers */
+
+	return 0;
+}
+
+static int mv88e6123_61_65_setup_port(struct dsa_switch *ds, int p)
+{
+	int addr = REG_PORT(p);
+
+	/*
+	 * MAC Forcing register: don't force link, speed, duplex
+	 * or flow control state to any particular values.
+	 */
+	REG_WRITE(addr, 0x01, 0x0003);
+
+	/*
+	 * Do not limit the period of time that this port can be
+	 * paused for by the remote end or the period of time that
+	 * this port can pause the remote end.
+	 */
+	REG_WRITE(addr, 0x02, 0x0000);
+
+	/*
+	 * Port Control: disable Drop-on-Unlock, disable Drop-on-Lock,
+	 * configure the EDSA tagging mode if this is the CPU port,
+	 * disable Header mode, enable IGMP/MLD snooping, disable VLAN
+	 * tunneling, determine priority by looking at 802.1p and IP
+	 * priority fields (IP prio has precedence), and set STP state
+	 * to Forwarding.  Finally, if this is the CPU port, additionally
+	 * enable forwarding of unknown unicast and multicast addresses.
+	 */
+	REG_WRITE(addr, 0x04,
+			(p == ds->cpu_port) ? 0x373f : 0x0433);
+
+	/*
+	 * Port Control 1: disable trunking.  Also, if this is the
+	 * CPU port, enable learn messages to be sent to this port.
+	 */
+	REG_WRITE(addr, 0x05, (p == ds->cpu_port) ? 0x8000 : 0x0000);
+
+	/*
+	 * Port based VLAN map: give each port its own address
+	 * database, allow the CPU port to talk to each of the 'real'
+	 * ports, and allow each of the 'real' ports to only talk to
+	 * the CPU port.
+	 */
+	REG_WRITE(addr, 0x06,
+			((p & 0xf) << 12) |
+			 ((p == ds->cpu_port) ?
+				ds->valid_port_mask :
+				(1 << ds->cpu_port)));
+
+	/*
+	 * Default VLAN ID and priority: don't set a default VLAN
+	 * ID, and set the default packet priority to zero.
+	 */
+	REG_WRITE(addr, 0x07, 0x0000);
+
+	/*
+	 * Port Control 2: don't force a good FCS, set the maximum
+	 * frame size to 10240 bytes, don't let the switch add or
+	 * strip 802.1q tags, don't discard tagged or untagged frames
+	 * on this port, do a destination address lookup on all
+	 * received packets as usual, disable ARP mirroring and don't
+	 * send a copy of all transmitted/received frames on this port
+	 * to the CPU.
+	 */
+	REG_WRITE(addr, 0x08, 0x2080);
+
+	/*
+	 * Egress rate control: disable egress rate control.
+	 */
+	REG_WRITE(addr, 0x09, 0x0001);
+
+	/*
+	 * Egress rate control 2: disable egress rate control.
+	 */
+	REG_WRITE(addr, 0x0a, 0x0000);
+
+	/*
+	 * Port Association Vector: when learning source addresses
+	 * of packets, add the address to the address database using
+	 * a port bitmap that has only the bit for this port set and
+	 * the other bits clear.
+	 */
+	REG_WRITE(addr, 0x0b, 1 << p);
+
+	/*
+	 * Port ATU control: disable limiting the number of address
+	 * database entries that this port is allowed to use.
+	 */
+	REG_WRITE(addr, 0x0c, 0x0000);
+
+	/*
+	 * Priorit Override: disable DA, SA and VTU priority override.
+	 */
+	REG_WRITE(addr, 0x0d, 0x0000);
+
+	/*
+	 * Port Ethertype: use the Ethertype DSA Ethertype value.
+	 */
+	REG_WRITE(addr, 0x0f, ETH_P_EDSA);
+
+	/*
+	 * Tag Remap: use an identity 802.1p prio -> switch prio
+	 * mapping.
+	 */
+	REG_WRITE(addr, 0x18, 0x3210);
+
+	/*
+	 * Tag Remap 2: use an identity 802.1p prio -> switch prio
+	 * mapping.
+	 */
+	REG_WRITE(addr, 0x19, 0x7654);
+
+	return 0;
+}
+
+static int mv88e6123_61_65_setup(struct dsa_switch *ds)
+{
+	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	int i;
+	int ret;
+
+	mutex_init(&ps->smi_mutex);
+	mutex_init(&ps->stats_mutex);
+
+	ret = mv88e6123_61_65_switch_reset(ds);
+	if (ret < 0)
+		return ret;
+
+	/* @@@ initialise vtu and atu */
+
+	ret = mv88e6123_61_65_setup_global(ds);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < 6; i++) {
+		ret = mv88e6123_61_65_setup_port(ds, i);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int mv88e6123_61_65_port_to_phy_addr(int port)
+{
+	if (port >= 0 && port <= 4)
+		return port;
+	return -1;
+}
+
+static int
+mv88e6123_61_65_phy_read(struct dsa_switch *ds, int port, int regnum)
+{
+	int addr = mv88e6123_61_65_port_to_phy_addr(port);
+	return mv88e6xxx_phy_read(ds, addr, regnum);
+}
+
+static int
+mv88e6123_61_65_phy_write(struct dsa_switch *ds,
+			      int port, int regnum, u16 val)
+{
+	int addr = mv88e6123_61_65_port_to_phy_addr(port);
+	return mv88e6xxx_phy_write(ds, addr, regnum, val);
+}
+
+static struct mv88e6xxx_hw_stat mv88e6123_61_65_hw_stats[] = {
+	{ "in_good_octets", 8, 0x00, },
+	{ "in_bad_octets", 4, 0x02, },
+	{ "in_unicast", 4, 0x04, },
+	{ "in_broadcasts", 4, 0x06, },
+	{ "in_multicasts", 4, 0x07, },
+	{ "in_pause", 4, 0x16, },
+	{ "in_undersize", 4, 0x18, },
+	{ "in_fragments", 4, 0x19, },
+	{ "in_oversize", 4, 0x1a, },
+	{ "in_jabber", 4, 0x1b, },
+	{ "in_rx_error", 4, 0x1c, },
+	{ "in_fcs_error", 4, 0x1d, },
+	{ "out_octets", 8, 0x0e, },
+	{ "out_unicast", 4, 0x10, },
+	{ "out_broadcasts", 4, 0x13, },
+	{ "out_multicasts", 4, 0x12, },
+	{ "out_pause", 4, 0x15, },
+	{ "excessive", 4, 0x11, },
+	{ "collisions", 4, 0x1e, },
+	{ "deferred", 4, 0x05, },
+	{ "single", 4, 0x14, },
+	{ "multiple", 4, 0x17, },
+	{ "out_fcs_error", 4, 0x03, },
+	{ "late", 4, 0x1f, },
+	{ "hist_64bytes", 4, 0x08, },
+	{ "hist_65_127bytes", 4, 0x09, },
+	{ "hist_128_255bytes", 4, 0x0a, },
+	{ "hist_256_511bytes", 4, 0x0b, },
+	{ "hist_512_1023bytes", 4, 0x0c, },
+	{ "hist_1024_max_bytes", 4, 0x0d, },
+};
+
+static void
+mv88e6123_61_65_get_strings(struct dsa_switch *ds, int port, uint8_t *data)
+{
+	mv88e6xxx_get_strings(ds, ARRAY_SIZE(mv88e6123_61_65_hw_stats),
+			      mv88e6123_61_65_hw_stats, port, data);
+}
+
+static void
+mv88e6123_61_65_get_ethtool_stats(struct dsa_switch *ds,
+				  int port, uint64_t *data)
+{
+	mv88e6xxx_get_ethtool_stats(ds, ARRAY_SIZE(mv88e6123_61_65_hw_stats),
+				    mv88e6123_61_65_hw_stats, port, data);
+}
+
+static int mv88e6123_61_65_get_sset_count(struct dsa_switch *ds)
+{
+	return ARRAY_SIZE(mv88e6123_61_65_hw_stats);
+}
+
+static struct dsa_switch_driver mv88e6123_61_65_switch_driver = {
+	.tag_protocol		= __constant_htons(ETH_P_EDSA),
+	.priv_size		= sizeof(struct mv88e6xxx_priv_state),
+	.probe			= mv88e6123_61_65_probe,
+	.setup			= mv88e6123_61_65_setup,
+	.set_addr		= mv88e6xxx_set_addr_indirect,
+	.phy_read		= mv88e6123_61_65_phy_read,
+	.phy_write		= mv88e6123_61_65_phy_write,
+	.poll_link		= mv88e6xxx_poll_link,
+	.get_strings		= mv88e6123_61_65_get_strings,
+	.get_ethtool_stats	= mv88e6123_61_65_get_ethtool_stats,
+	.get_sset_count		= mv88e6123_61_65_get_sset_count,
+};
+
+int __init mv88e6123_61_65_init(void)
+{
+	register_switch_driver(&mv88e6123_61_65_switch_driver);
+	return 0;
+}
+module_init(mv88e6123_61_65_init);
+
+void __exit mv88e6123_61_65_cleanup(void)
+{
+	unregister_switch_driver(&mv88e6123_61_65_switch_driver);
+}
+module_exit(mv88e6123_61_65_cleanup);

net/dsa/mv88e6xxx.c

+/*
+ * net/dsa/mv88e6xxx.c - Marvell 88e6xxx switch chip support
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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/list.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include "dsa_priv.h"
+#include "mv88e6xxx.h"
+
+/*
+ * If the switch's ADDR[4:0] strap pins are strapped to zero, it will
+ * use all 32 SMI bus addresses on its SMI bus, and all switch registers
+ * will be directly accessible on some {device address,register address}
+ * pair.  If the ADDR[4:0] pins are not strapped to zero, the switch
+ * will only respond to SMI transactions to that specific address, and
+ * an indirect addressing mechanism needs to be used to access its
+ * registers.
+ */
+static int mv88e6xxx_reg_wait_ready(struct mii_bus *bus, int sw_addr)
+{
+	int ret;
+	int i;
+
+	for (i = 0; i < 16; i++) {
+		ret = mdiobus_read(bus, sw_addr, 0);
+		if (ret < 0)
+			return ret;
+
+		if ((ret & 0x8000) == 0)
+			return 0;
+	}
+
+	return -ETIMEDOUT;
+}
+
+int __mv88e6xxx_reg_read(struct mii_bus *bus, int sw_addr, int addr, int reg)
+{
+	int ret;
+
+	if (sw_addr == 0)
+		return mdiobus_read(bus, addr, reg);
+
+	/*
+	 * Wait for the bus to become free.
+	 */
+	ret = mv88e6xxx_reg_wait_ready(bus, sw_addr);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Transmit the read command.
+	 */
+	ret = mdiobus_write(bus, sw_addr, 0, 0x9800 | (addr << 5) | reg);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Wait for the read command to complete.
+	 */
+	ret = mv88e6xxx_reg_wait_ready(bus, sw_addr);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Read the data.
+	 */
+	ret = mdiobus_read(bus, sw_addr, 1);
+	if (ret < 0)
+		return ret;
+
+	return ret & 0xffff;
+}
+
+int mv88e6xxx_reg_read(struct dsa_switch *ds, int addr, int reg)
+{
+	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	int ret;
+
+	mutex_lock(&ps->smi_mutex);
+	ret = __mv88e6xxx_reg_read(ds->master_mii_bus,
+				   ds->pd->sw_addr, addr, reg);
+	mutex_unlock(&ps->smi_mutex);
+
+	return ret;
+}
+
+int __mv88e6xxx_reg_write(struct mii_bus *bus, int sw_addr, int addr,
+			  int reg, u16 val)
+{
+	int ret;
+
+	if (sw_addr == 0)
+		return mdiobus_write(bus, addr, reg, val);
+
+	/*
+	 * Wait for the bus to become free.
+	 */
+	ret = mv88e6xxx_reg_wait_ready(bus, sw_addr);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Transmit the data to write.
+	 */
+	ret = mdiobus_write(bus, sw_addr, 1, val);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Transmit the write command.
+	 */
+	ret = mdiobus_write(bus, sw_addr, 0, 0x9400 | (addr << 5) | reg);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Wait for the write command to complete.
+	 */
+	ret = mv88e6xxx_reg_wait_ready(bus, sw_addr);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+int mv88e6xxx_reg_write(struct dsa_switch *ds, int addr, int reg, u16 val)
+{
+	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	int ret;
+
+	mutex_lock(&ps->smi_mutex);
+	ret = __mv88e6xxx_reg_write(ds->master_mii_bus,
+				    ds->pd->sw_addr, addr, reg, val);
+	mutex_unlock(&ps->smi_mutex);
+
+	return ret;
+}
+
+int mv88e6xxx_config_prio(struct dsa_switch *ds)
+{
+	/*
+	 * Configure the IP ToS mapping registers.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x10, 0x0000);
+	REG_WRITE(REG_GLOBAL, 0x11, 0x0000);
+	REG_WRITE(REG_GLOBAL, 0x12, 0x5555);
+	REG_WRITE(REG_GLOBAL, 0x13, 0x5555);
+	REG_WRITE(REG_GLOBAL, 0x14, 0xaaaa);
+	REG_WRITE(REG_GLOBAL, 0x15, 0xaaaa);
+	REG_WRITE(REG_GLOBAL, 0x16, 0xffff);
+	REG_WRITE(REG_GLOBAL, 0x17, 0xffff);
+
+	/*
+	 * Configure the IEEE 802.1p priority mapping register.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x18, 0xfa41);
+
+	return 0;
+}
+
+int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr)
+{
+	int i;
+	int ret;
+
+	for (i = 0; i < 6; i++) {
+		int j;
+
+		/*
+		 * Write the MAC address byte.
+		 */
+		REG_WRITE(REG_GLOBAL2, 0x0d, 0x8000 | (i << 8) | addr[i]);
+
+		/*
+		 * Wait for the write to complete.
+		 */
+		for (j = 0; j < 16; j++) {
+			ret = REG_READ(REG_GLOBAL2, 0x0d);
+			if ((ret & 0x8000) == 0)
+				break;
+		}
+		if (j == 16)
+			return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
+int mv88e6xxx_phy_read(struct dsa_switch *ds, int addr, int regnum)
+{
+	if (addr >= 0)
+		return mv88e6xxx_reg_read(ds, addr, regnum);
+	return 0xffff;
+}
+
+int mv88e6xxx_phy_write(struct dsa_switch *ds, int addr, int regnum, u16 val)
+{
+	if (addr >= 0)
+		return mv88e6xxx_reg_write(ds, addr, regnum, val);
+	return 0;
+}
+
+void mv88e6xxx_poll_link(struct dsa_switch *ds)
+{
+	int i;
+
+	for (i = 0; i < DSA_MAX_PORTS; i++) {
+		struct net_device *dev;
+		int port_status;
+		int link;
+		int speed;
+		int duplex;
+		int fc;
+
+		dev = ds->ports[i];
+		if (dev == NULL)
+			continue;
+
+		link = 0;
+		if (dev->flags & IFF_UP) {
+			port_status = mv88e6xxx_reg_read(ds, REG_PORT(i), 0x00);
+			if (port_status < 0)
+				continue;
+
+			link = !!(port_status & 0x0800);
+		}
+
+		if (!link) {
+			if (netif_carrier_ok(dev)) {
+				printk(KERN_INFO "%s: link down\n", dev->name);
+				netif_carrier_off(dev);
+			}
+			continue;
+		}
+
+		switch (port_status & 0x0300) {
+		case 0x0000:
+			speed = 10;
+			break;
+		case 0x0100:
+			speed = 100;
+			break;
+		case 0x0200:
+			speed = 1000;
+			break;
+		default:
+			speed = -1;
+			break;
+		}
+		duplex = (port_status & 0x0400) ? 1 : 0;
+		fc = (port_status & 0x8000) ? 1 : 0;
+
+		if (!netif_carrier_ok(dev)) {
+			printk(KERN_INFO "%s: link up, %d Mb/s, %s duplex, "
+					 "flow control %sabled\n", dev->name,
+					 speed, duplex ? "full" : "half",
+					 fc ? "en" : "dis");
+			netif_carrier_on(dev);
+		}
+	}
+}
+
+static int mv88e6xxx_stats_wait(struct dsa_switch *ds)
+{
+	int ret;
+	int i;
+
+	for (i = 0; i < 10; i++) {
+		ret = REG_READ(REG_GLOBAL2, 0x1d);
+		if ((ret & 0x8000) == 0)
+			return 0;
+	}
+
+	return -ETIMEDOUT;
+}
+
+static int mv88e6xxx_stats_snapshot(struct dsa_switch *ds, int port)
+{
+	int ret;
+
+	/*
+	 * Snapshot the hardware statistics counters for this port.
+	 */
+	REG_WRITE(REG_GLOBAL, 0x1d, 0xdc00 | port);
+
+	/*
+	 * Wait for the snapshotting to complete.
+	 */
+	ret = mv88e6xxx_stats_wait(ds);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static void mv88e6xxx_stats_read(struct dsa_switch *ds, int stat, u32 *val)
+{
+	u32 _val;
+	int ret;
+
+	*val = 0;
+
+	ret = mv88e6xxx_reg_write(ds, REG_GLOBAL, 0x1d, 0xcc00 | stat);
+	if (ret < 0)
+		return;
+
+	ret = mv88e6xxx_stats_wait(ds);
+	if (ret < 0)
+		return;
+
+	ret = mv88e6xxx_reg_read(ds, REG_GLOBAL, 0x1e);
+	if (ret < 0)
+		return;
+
+	_val = ret << 16;
+
+	ret = mv88e6xxx_reg_read(ds, REG_GLOBAL, 0x1f);
+	if (ret < 0)
+		return;
+
+	*val = _val | ret;
+}
+
+void mv88e6xxx_get_strings(struct dsa_switch *ds,
+			   int nr_stats, struct mv88e6xxx_hw_stat *stats,
+			   int port, uint8_t *data)
+{
+	int i;
+
+	for (i = 0; i < nr_stats; i++) {
+		memcpy(data + i * ETH_GSTRING_LEN,
+		       stats[i].string, ETH_GSTRING_LEN);
+	}
+}
+
+void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds,
+				 int nr_stats, struct mv88e6xxx_hw_stat *stats,
+				 int port, uint64_t *data)
+{
+	struct mv88e6xxx_priv_state *ps = (void *)(ds + 1);
+	int ret;
+	int i;
+
+	mutex_lock(&ps->stats_mutex);
+
+	ret = mv88e6xxx_stats_snapshot(ds, port);
+	if (ret < 0) {
+		mutex_unlock(&ps->stats_mutex);
+		return;
+	}
+
+	/*
+	 * Read each of the counters.
+	 */
+	for (i = 0; i < nr_stats; i++) {
+		struct mv88e6xxx_hw_stat *s = stats + i;
+		u32 low;
+		u32 high;
+
+		mv88e6xxx_stats_read(ds, s->reg, &low);
+		if (s->sizeof_stat == 8)
+			mv88e6xxx_stats_read(ds, s->reg + 1, &high);
+		else
+			high = 0;
+
+		data[i] = (((u64)high) << 32) | low;
+	}
+
+	mutex_unlock(&ps->stats_mutex);
+}

net/dsa/mv88e6xxx.h

+/*
+ * net/dsa/mv88e6xxx.h - Marvell 88e6xxx switch chip support
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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.
+ */
+
+#ifndef __MV88E6XXX_H
+#define __MV88E6XXX_H
+
+#define REG_PORT(p)		(0x10 + (p))
+#define REG_GLOBAL		0x1b
+#define REG_GLOBAL2		0x1c
+
+struct mv88e6xxx_priv_state {
+	/*
+	 * When using multi-chip addressing, this mutex protects
+	 * access to the indirect access registers.  (In single-chip
+	 * mode, this mutex is effectively useless.)
+	 */
+	struct mutex	smi_mutex;
+
+	/*
+	 * This mutex serialises access to the statistics unit.
+	 * Hold this mutex over snapshot + dump sequences.
+	 */
+	struct mutex	stats_mutex;
+};
+
+struct mv88e6xxx_hw_stat {
+	char string[ETH_GSTRING_LEN];
+	int sizeof_stat;
+	int reg;
+};
+
+int __mv88e6xxx_reg_read(struct mii_bus *bus, int sw_addr, int addr, int reg);
+int mv88e6xxx_reg_read(struct dsa_switch *ds, int addr, int reg);
+int __mv88e6xxx_reg_write(struct mii_bus *bus, int sw_addr, int addr,
+                          int reg, u16 val);
+int mv88e6xxx_reg_write(struct dsa_switch *ds, int addr, int reg, u16 val);
+int mv88e6xxx_config_prio(struct dsa_switch *ds);
+int mv88e6xxx_set_addr_indirect(struct dsa_switch *ds, u8 *addr);
+int mv88e6xxx_phy_read(struct dsa_switch *ds, int addr, int regnum);
+int mv88e6xxx_phy_write(struct dsa_switch *ds, int addr, int regnum, u16 val);
+void mv88e6xxx_poll_link(struct dsa_switch *ds);
+void mv88e6xxx_get_strings(struct dsa_switch *ds,
+			   int nr_stats, struct mv88e6xxx_hw_stat *stats,
+			   int port, uint8_t *data);
+void mv88e6xxx_get_ethtool_stats(struct dsa_switch *ds,
+				 int nr_stats, struct mv88e6xxx_hw_stat *stats,
+				 int port, uint64_t *data);
+
+#define REG_READ(addr, reg)						\
+	({								\
+		int __ret;						\
+									\
+		__ret = mv88e6xxx_reg_read(ds, addr, reg);		\
+		if (__ret < 0)						\
+			return __ret;					\
+		__ret;							\
+	})
+
+#define REG_WRITE(addr, reg, val)					\
+	({								\
+		int __ret;						\
+									\
+		__ret = mv88e6xxx_reg_write(ds, addr, reg, val);	\
+		if (__ret < 0)						\
+			return __ret;					\
+	})
+
+
+
+#endif

net/dsa/slave.c

+/*
+ * net/dsa/slave.c - Slave device handling
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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/list.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include "dsa_priv.h"
+
+/* slave mii_bus handling ***************************************************/
+static int dsa_slave_phy_read(struct mii_bus *bus, int addr, int reg)
+{
+	struct dsa_switch *ds = bus->priv;
+
+	if (ds->valid_port_mask & (1 << addr))
+		return ds->drv->phy_read(ds, addr, reg);
+
+	return 0xffff;
+}
+
+static int dsa_slave_phy_write(struct mii_bus *bus, int addr, int reg, u16 val)
+{
+	struct dsa_switch *ds = bus->priv;
+
+	if (ds->valid_port_mask & (1 << addr))
+		return ds->drv->phy_write(ds, addr, reg, val);
+
+	return 0;
+}
+
+void dsa_slave_mii_bus_init(struct dsa_switch *ds)
+{
+	ds->slave_mii_bus->priv = (void *)ds;
+	ds->slave_mii_bus->name = "dsa slave smi";
+	ds->slave_mii_bus->read = dsa_slave_phy_read;
+	ds->slave_mii_bus->write = dsa_slave_phy_write;
+	snprintf(ds->slave_mii_bus->id, MII_BUS_ID_SIZE, "%s:%.2x",
+			ds->master_mii_bus->id, ds->pd->sw_addr);
+	ds->slave_mii_bus->parent = &(ds->master_mii_bus->dev);
+}
+
+
+/* slave device handling ****************************************************/
+static int dsa_slave_open(struct net_device *dev)
+{
+	return 0;
+}
+
+static int dsa_slave_close(struct net_device *dev)
+{
+	return 0;
+}
+
+static void dsa_slave_change_rx_flags(struct net_device *dev, int change)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct net_device *master = p->parent->master_netdev;
+
+	if (change & IFF_ALLMULTI)
+		dev_set_allmulti(master, dev->flags & IFF_ALLMULTI ? 1 : -1);
+	if (change & IFF_PROMISC)
+		dev_set_promiscuity(master, dev->flags & IFF_PROMISC ? 1 : -1);
+}
+
+static void dsa_slave_set_rx_mode(struct net_device *dev)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct net_device *master = p->parent->master_netdev;
+
+	dev_mc_sync(master, dev);
+	dev_unicast_sync(master, dev);
+}
+
+static int dsa_slave_set_mac_address(struct net_device *dev, void *addr)
+{
+	memcpy(dev->dev_addr, addr + 2, 6);
+
+	return 0;
+}
+
+static int dsa_slave_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct mii_ioctl_data *mii_data = if_mii(ifr);
+
+	if (p->phy != NULL)
+		return phy_mii_ioctl(p->phy, mii_data, cmd);
+
+	return -EOPNOTSUPP;
+}
+
+
+/* ethtool operations *******************************************************/
+static int
+dsa_slave_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	int err;
+
+	err = -EOPNOTSUPP;
+	if (p->phy != NULL) {
+		err = phy_read_status(p->phy);
+		if (err == 0)
+			err = phy_ethtool_gset(p->phy, cmd);
+	}
+
+	return err;
+}
+
+static int
+dsa_slave_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+
+	if (p->phy != NULL)
+		return phy_ethtool_sset(p->phy, cmd);
+
+	return -EOPNOTSUPP;
+}
+
+static void dsa_slave_get_drvinfo(struct net_device *dev,
+				  struct ethtool_drvinfo *drvinfo)
+{
+	strncpy(drvinfo->driver, "dsa", 32);
+	strncpy(drvinfo->version, dsa_driver_version, 32);
+	strncpy(drvinfo->fw_version, "N/A", 32);
+	strncpy(drvinfo->bus_info, "platform", 32);
+}
+
+static int dsa_slave_nway_reset(struct net_device *dev)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+
+	if (p->phy != NULL)
+		return genphy_restart_aneg(p->phy);
+
+	return -EOPNOTSUPP;
+}
+
+static u32 dsa_slave_get_link(struct net_device *dev)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+
+	if (p->phy != NULL) {
+		genphy_update_link(p->phy);
+		return p->phy->link;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static void dsa_slave_get_strings(struct net_device *dev,
+				  uint32_t stringset, uint8_t *data)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct dsa_switch *ds = p->parent;
+
+	if (stringset == ETH_SS_STATS) {
+		int len = ETH_GSTRING_LEN;
+
+		strncpy(data, "tx_packets", len);
+		strncpy(data + len, "tx_bytes", len);
+		strncpy(data + 2 * len, "rx_packets", len);
+		strncpy(data + 3 * len, "rx_bytes", len);
+		if (ds->drv->get_strings != NULL)
+			ds->drv->get_strings(ds, p->port, data + 4 * len);
+	}
+}
+
+static void dsa_slave_get_ethtool_stats(struct net_device *dev,
+					struct ethtool_stats *stats,
+					uint64_t *data)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct dsa_switch *ds = p->parent;
+
+	data[0] = p->dev->stats.tx_packets;
+	data[1] = p->dev->stats.tx_bytes;
+	data[2] = p->dev->stats.rx_packets;
+	data[3] = p->dev->stats.rx_bytes;
+	if (ds->drv->get_ethtool_stats != NULL)
+		ds->drv->get_ethtool_stats(ds, p->port, data + 4);
+}
+
+static int dsa_slave_get_sset_count(struct net_device *dev, int sset)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	struct dsa_switch *ds = p->parent;
+
+	if (sset == ETH_SS_STATS) {
+		int count;
+
+		count = 4;
+		if (ds->drv->get_sset_count != NULL)
+			count += ds->drv->get_sset_count(ds);
+
+		return count;
+	}
+
+	return -EOPNOTSUPP;
+}
+
+static const struct ethtool_ops dsa_slave_ethtool_ops = {
+	.get_settings		= dsa_slave_get_settings,
+	.set_settings		= dsa_slave_set_settings,
+	.get_drvinfo		= dsa_slave_get_drvinfo,
+	.nway_reset		= dsa_slave_nway_reset,
+	.get_link		= dsa_slave_get_link,
+	.set_sg			= ethtool_op_set_sg,
+	.get_strings		= dsa_slave_get_strings,
+	.get_ethtool_stats	= dsa_slave_get_ethtool_stats,
+	.get_sset_count		= dsa_slave_get_sset_count,
+};
+
+
+/* slave device setup *******************************************************/
+struct net_device *
+dsa_slave_create(struct dsa_switch *ds, struct device *parent,
+		 int port, char *name)
+{
+	struct net_device *master = ds->master_netdev;
+	struct net_device *slave_dev;
+	struct dsa_slave_priv *p;
+	int ret;
+
+	slave_dev = alloc_netdev(sizeof(struct dsa_slave_priv),
+				 name, ether_setup);
+	if (slave_dev == NULL)
+		return slave_dev;
+
+	slave_dev->features = master->vlan_features;
+	SET_ETHTOOL_OPS(slave_dev, &dsa_slave_ethtool_ops);
+	memcpy(slave_dev->dev_addr, master->dev_addr, ETH_ALEN);
+	slave_dev->tx_queue_len = 0;
+	switch (ds->tag_protocol) {
+#ifdef CONFIG_NET_DSA_TAG_EDSA
+	case htons(ETH_P_EDSA):
+		slave_dev->hard_start_xmit = edsa_xmit;
+		break;
+#endif
+	default:
+		BUG();
+	}
+	slave_dev->open = dsa_slave_open;
+	slave_dev->stop = dsa_slave_close;
+	slave_dev->change_rx_flags = dsa_slave_change_rx_flags;
+	slave_dev->set_rx_mode = dsa_slave_set_rx_mode;
+	slave_dev->set_multicast_list = dsa_slave_set_rx_mode;
+	slave_dev->set_mac_address = dsa_slave_set_mac_address;
+	slave_dev->do_ioctl = dsa_slave_ioctl;
+	SET_NETDEV_DEV(slave_dev, parent);
+	slave_dev->vlan_features = master->vlan_features;
+
+	p = netdev_priv(slave_dev);
+	p->dev = slave_dev;
+	p->parent = ds;
+	p->port = port;
+	p->phy = ds->slave_mii_bus->phy_map[port];
+
+	ret = register_netdev(slave_dev);
+	if (ret) {
+		printk(KERN_ERR "%s: error %d registering interface %s\n",
+				master->name, ret, slave_dev->name);
+		free_netdev(slave_dev);
+		return NULL;
+	}
+
+	netif_carrier_off(slave_dev);
+
+	if (p->phy != NULL) {
+		phy_attach(slave_dev, p->phy->dev.bus_id,
+			   0, PHY_INTERFACE_MODE_GMII);
+
+		p->phy->autoneg = AUTONEG_ENABLE;
+		p->phy->speed = 0;
+		p->phy->duplex = 0;
+		p->phy->advertising = p->phy->supported | ADVERTISED_Autoneg;
+		phy_start_aneg(p->phy);
+	}
+
+	return slave_dev;
+}

net/dsa/tag_edsa.c

+/*
+ * net/dsa/tag_edsa.c - Ethertype DSA tagging
+ * Copyright (c) 2008 Marvell Semiconductor
+ *
+ * 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/etherdevice.h>
+#include <linux/list.h>
+#include <linux/netdevice.h>
+#include "dsa_priv.h"
+
+#define DSA_HLEN	4
+#define EDSA_HLEN	8
+
+int edsa_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct dsa_slave_priv *p = netdev_priv(dev);
+	u8 *edsa_header;
+
+	dev->stats.tx_packets++;
+	dev->stats.tx_bytes += skb->len;
+
+	/*
+	 * Convert the outermost 802.1q tag to a DSA tag and prepend
+	 * a DSA ethertype field is the packet is tagged, or insert
+	 * a DSA ethertype plus DSA tag between the addresses and the
+	 * current ethertype field if the packet is untagged.
+	 */
+	if (skb->protocol == htons(ETH_P_8021Q)) {
+		if (skb_cow_head(skb, DSA_HLEN) < 0)
+			goto out_free;
+		skb_push(skb, DSA_HLEN);
+
+		memmove(skb->data, skb->data + DSA_HLEN, 2 * ETH_ALEN);
+
+		/*
+		 * Construct tagged FROM_CPU DSA tag from 802.1q tag.
+		 */
+		edsa_header = skb->data + 2 * ETH_ALEN;
+		edsa_header[0] = (ETH_P_EDSA >> 8) & 0xff;
+		edsa_header[1] = ETH_P_EDSA & 0xff;
+		edsa_header[2] = 0x00;
+		edsa_header[3] = 0x00;
+		edsa_header[4] = 0x60;
+		edsa_header[5] = p->port << 3;
+
+		/*
+		 * Move CFI field from byte 6 to byte 5.
+		 */
+		if (edsa_header[6] & 0x10) {
+			edsa_header[5] |= 0x01;
+			edsa_header[6] &= ~0x10;
+		}
+	} else {
+		if (skb_cow_head(skb, EDSA_HLEN) < 0)
+			goto out_free;
+		skb_push(skb, EDSA_HLEN);
+
+		memmove(skb->data, skb->data + EDSA_HLEN, 2 * ETH_ALEN);
+
+		/*
+		 * Construct untagged FROM_CPU DSA tag.
+		 */
+		edsa_header = skb->data + 2 * ETH_ALEN;
+		edsa_header[0] = (ETH_P_EDSA >> 8) & 0xff;
+		edsa_header[1] = ETH_P_EDSA & 0xff;
+		edsa_header[2] = 0x00;
+		edsa_header[3] = 0x00;
+		edsa_header[4] = 0x40;
+		edsa_header[5] = p->port << 3;
+		edsa_header[6] = 0x00;
+		edsa_header[7] = 0x00;
+	}
+
+	skb->protocol = htons(ETH_P_EDSA);
+
+	skb->dev = p->parent->master_netdev;
+	dev_queue_xmit(skb);
+
+	return NETDEV_TX_OK;
+
+out_free:
+	kfree_skb(skb);
+	return NETDEV_TX_OK;
+}
+
+static int edsa_rcv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *pt, struct net_device *orig_dev)
+{
+	struct dsa_switch *ds = dev->dsa_ptr;
+	u8 *edsa_header;
+	int source_port;
+
+	if (unlikely(ds == NULL))
+		goto out_drop;
+
+	skb = skb_unshare(skb, GFP_ATOMIC);
+	if (skb == NULL)
+		goto out;
+
+	if (unlikely(!pskb_may_pull(skb, EDSA_HLEN)))
+		goto out_drop;
+
+	/*
+	 * Skip the two null bytes after the ethertype.
+	 */
+	edsa_header = skb->data + 2;
+
+	/*
+	 * Check that frame type is either TO_CPU or FORWARD, and
+	 * that the source device is zero.
+	 */
+	if ((edsa_header[0] & 0xdf) != 0x00 && (edsa_header[0] & 0xdf) != 0xc0)
+		goto out_drop;
+
+	/*
+	 * Check that the source port is a registered DSA port.
+	 */
+	source_port = (edsa_header[1] >> 3) & 0x1f;
+	if (source_port >= DSA_MAX_PORTS || ds->ports[source_port] == NULL)
+		goto out_drop;
+
+	/*
+	 * If the 'tagged' bit is set, convert the DSA tag to a 802.1q
+	 * tag and delete the ethertype part.  If the 'tagged' bit is
+	 * clear, delete the ethertype and the DSA tag parts.
+	 */
+	if (edsa_header[0] & 0x20) {
+		u8 new_header[4];
+
+		/*
+		 * Insert 802.1q ethertype and copy the VLAN-related
+		 * fields, but clear the bit that will hold CFI (since
+		 * DSA uses that bit location for another purpose).
+		 */
+		new_header[0] = (ETH_P_8021Q >> 8) & 0xff;
+		new_header[1] = ETH_P_8021Q & 0xff;
+		new_header[2] = edsa_header[2] & ~0x10;
+		new_header[3] = edsa_header[3];
+
+		/*
+		 * Move CFI bit from its place in the DSA header to
+		 * its 802.1q-designated place.
+		 */
+		if (edsa_header[1] & 0x01)
+			new_header[2] |= 0x10;
+
+		skb_pull_rcsum(skb, DSA_HLEN);
+
+		/*
+		 * Update packet checksum if skb is CHECKSUM_COMPLETE.
+		 */
+		if (skb->ip_summed == CHECKSUM_COMPLETE) {
+			__wsum c = skb->csum;
+			c = csum_add(c, csum_partial(new_header + 2, 2, 0));
+			c = csum_sub(c, csum_partial(edsa_header + 2, 2, 0));
+			skb->csum = c;
+		}
+
+		memcpy(edsa_header, new_header, DSA_HLEN);
+
+		memmove(skb->data - ETH_HLEN,
+			skb->data - ETH_HLEN - DSA_HLEN,
+			2 * ETH_ALEN);
+	} else {
+		/*
+		 * Remove DSA tag and update checksum.
+		 */
+		skb_pull_rcsum(skb, EDSA_HLEN);
+		memmove(skb->data - ETH_HLEN,
+			skb->data - ETH_HLEN - EDSA_HLEN,
+			2 * ETH_ALEN);
+	}
+
+	skb->dev = ds->ports[source_port];
+	skb_push(skb, ETH_HLEN);
+	skb->protocol = eth_type_trans(skb, skb->dev);
+
+	skb->dev->last_rx = jiffies;
+	skb->dev->stats.rx_packets++;
+	skb->dev->stats.rx_bytes += skb->len;
+
+	netif_receive_skb(skb);
+
+	return 0;
+
+out_drop:
+	kfree_skb(skb);
+out:
+	return 0;
+}
+
+static struct packet_type edsa_packet_type = {
+	.type	= __constant_htons(ETH_P_EDSA),
+	.func	= edsa_rcv,
+};
+
+static int __init edsa_init_module(void)
+{
+	dev_add_pack(&edsa_packet_type);
+	return 0;
+}
+module_init(edsa_init_module);
+
+static void __exit edsa_cleanup_module(void)
+{
+	dev_remove_pack(&edsa_packet_type);
+}
+module_exit(edsa_cleanup_module);