dsa: add switch chip cascading support

The initial version of the DSA driver only supported a single switch
chip per network interface, while DSA-capable switch chips can be
interconnected to form a tree of switch chips.  This patch adds support
for multiple switch chips on a network interface.

An example topology for a 16-port device with an embedded CPU is as
follows:

	+-----+          +--------+       +--------+
	|     |eth0    10| switch |9    10| switch |
	| CPU +----------+        +-------+        |
	|     |          | chip 0 |       | chip 1 |
	+-----+          +---++---+       +---++---+
	                     ||               ||
	                     ||               ||
	                     ||1000baseT      ||1000baseT
	                     ||ports 1-8      ||ports 9-16

This requires a couple of interdependent changes in the DSA layer:

- The dsa platform driver data needs to be extended: there is still
  only one netdevice per DSA driver instance (eth0 in the example
  above), but each of the switch chips in the tree needs its own
  mii_bus device pointer, MII management bus address, and port name
  array. (include/net/dsa.h)  The existing in-tree dsa users need
  some small changes to deal with this. (arch/arm)

- The DSA and Ethertype DSA tagging modules need to be extended to
  use the DSA device ID field on receive and demultiplex the packet
  accordingly, and fill in the DSA device ID field on transmit
  according to which switch chip the packet is heading to.
  (net/dsa/tag_{dsa,edsa}.c)

- The concept of "CPU port", which is the switch chip port that the
  CPU is connected to (port 10 on switch chip 0 in the example), needs
  to be extended with the concept of "upstream port", which is the
  port on the switch chip that will bring us one hop closer to the CPU
  (port 10 for both switch chips in the example above).

- The dsa platform data needs to specify which ports on which switch
  chips are links to other switch chips, so that we can enable DSA
  tagging mode on them.  (For inter-switch links, we always use
  non-EtherType DSA tagging, since it has lower overhead.  The CPU
  link uses dsa or edsa tagging depending on what the 'root' switch
  chip supports.)  This is done by specifying "dsa" for the given
  port in the port array.

- The dsa platform data needs to be extended with information on via
  which port to reach any given switch chip from any given switch chip.
  This info is specified via the per-switch chip data struct ->rtable[]
  array, which gives the nexthop ports for each of the other switches
  in the tree.

For the example topology above, the dsa platform data would look
something like this:

	static struct dsa_chip_data sw[2] = {
		{
			.mii_bus	= &foo,
			.sw_addr	= 1,
			.port_names[0]	= "p1",
			.port_names[1]	= "p2",
			.port_names[2]	= "p3",
			.port_names[3]	= "p4",
			.port_names[4]	= "p5",
			.port_names[5]	= "p6",
			.port_names[6]	= "p7",
			.port_names[7]	= "p8",
			.port_names[9]	= "dsa",
			.port_names[10]	= "cpu",
			.rtable		= (s8 []){ -1, 9, },
		}, {
			.mii_bus	= &foo,
			.sw_addr	= 2,
			.port_names[0]	= "p9",
			.port_names[1]	= "p10",
			.port_names[2]	= "p11",
			.port_names[3]	= "p12",
			.port_names[4]	= "p13",
			.port_names[5]	= "p14",
			.port_names[6]	= "p15",
			.port_names[7]	= "p16",
			.port_names[10]	= "dsa",
			.rtable		= (s8 []){ 10, -1, },
		},
	},

	static struct dsa_platform_data pd = {
		.netdev		= &foo,
		.nr_switches	= 2,
		.sw		= sw,
	};

Signed-off-by: Lennert Buytenhek <buytenh@marvell.com>
Tested-by: Gary Thomas <gary@mlbassoc.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 109 insertions, 68 deletions

diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 33e99462023a..71489f69a42c 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -1,6 +1,6 @@
 /*
  * net/dsa/dsa.c - Hardware switch handling
- * Copyright (c) 2008 Marvell Semiconductor
+ * Copyright (c) 2008-2009 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
@@ -67,12 +67,13 @@ dsa_switch_probe(struct mii_bus *bus, int sw_addr, char **_name)
 
 /* 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)
+dsa_switch_setup(struct dsa_switch_tree *dst, int index,
+                 struct device *parent, struct mii_bus *bus)
 {
+        struct dsa_chip_data *pd = dst->pd->chip + index;
+        struct dsa_switch_driver *drv;
         struct dsa_switch *ds;
         int ret;
-        struct dsa_switch_driver *drv;
         char *name;
         int i;
 
@@ -81,11 +82,12 @@ dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
          */
         drv = dsa_switch_probe(bus, pd->sw_addr, &name);
         if (drv == NULL) {
-                printk(KERN_ERR "%s: could not detect attached switch\n",
-                       dev->name);
+                printk(KERN_ERR "%s[%d]: could not detect attached switch\n",
+                       dst->master_netdev->name, index);
                 return ERR_PTR(-EINVAL);
         }
-        printk(KERN_INFO "%s: detected a %s switch\n", dev->name, name);
+        printk(KERN_INFO "%s[%d]: detected a %s switch\n",
+                dst->master_netdev->name, index, name);
 
 
         /*
@@ -95,18 +97,16 @@ dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
         if (ds == NULL)
                 return ERR_PTR(-ENOMEM);
 
-        ds->pd = pd;
-        ds->master_netdev = dev;
-        ds->master_mii_bus = bus;
-
+        ds->dst = dst;
+        ds->index = index;
+        ds->pd = dst->pd->chip + index;
         ds->drv = drv;
-        ds->tag_protocol = drv->tag_protocol;
+        ds->master_mii_bus = bus;
 
 
         /*
          * Validate supplied switch configuration.
          */
-        ds->cpu_port = -1;
         for (i = 0; i < DSA_MAX_PORTS; i++) {
                 char *name;
 
@@ -115,32 +115,28 @@ dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
                         continue;
 
                 if (!strcmp(name, "cpu")) {
-                        if (ds->cpu_port != -1) {
+                        if (dst->cpu_switch != -1) {
                                 printk(KERN_ERR "multiple cpu ports?!\n");
                                 ret = -EINVAL;
                                 goto out;
                         }
-                        ds->cpu_port = i;
+                        dst->cpu_switch = index;
+                        dst->cpu_port = i;
+                } else if (!strcmp(name, "dsa")) {
+                        ds->dsa_port_mask |= 1 << i;
                 } else {
-                        ds->valid_port_mask |= 1 << i;
+                        ds->phys_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.)
+         * If the CPU connects to this switch, set the switch tree
+         * tagging protocol to the preferred tagging format of this
+         * switch.
          */
-        wmb();
-        dev->dsa_ptr = (void *)ds;
+        if (ds->dst->cpu_switch == index)
+                ds->dst->tag_protocol = drv->tag_protocol;
 
 
         /*
@@ -150,7 +146,7 @@ dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
         if (ret < 0)
                 goto out;
 
-        ret = drv->set_addr(ds, dev->dev_addr);
+        ret = drv->set_addr(ds, dst->master_netdev->dev_addr);
         if (ret < 0)
                 goto out;
 
@@ -169,18 +165,18 @@ dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
         /*
          * 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)))
+                if (!(ds->phys_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]);
+                        printk(KERN_ERR "%s[%d]: can't create dsa "
+                               "slave device for port %d(%s)\n",
+                               dst->master_netdev->name,
+                               index, i, pd->port_names[i]);
                         continue;
                 }
 
@@ -192,7 +188,6 @@ dsa_switch_setup(struct device *parent, struct dsa_platform_data *pd,
 out_free:
         mdiobus_free(ds->slave_mii_bus);
 out:
-        dev->dsa_ptr = NULL;
         kfree(ds);
         return ERR_PTR(ret);
 }
@@ -212,35 +207,42 @@ static void dsa_switch_destroy(struct dsa_switch *ds)
  */
 bool dsa_uses_dsa_tags(void *dsa_ptr)
 {
-        struct dsa_switch *ds = dsa_ptr;
+        struct dsa_switch_tree *dst = dsa_ptr;
 
-        return !!(ds->tag_protocol == htons(ETH_P_DSA));
+        return !!(dst->tag_protocol == htons(ETH_P_DSA));
 }
 
 bool dsa_uses_trailer_tags(void *dsa_ptr)
 {
-        struct dsa_switch *ds = dsa_ptr;
+        struct dsa_switch_tree *dst = dsa_ptr;
 
-        return !!(ds->tag_protocol == htons(ETH_P_TRAILER));
+        return !!(dst->tag_protocol == htons(ETH_P_TRAILER));
 }
 
 
 /* link polling *************************************************************/
 static void dsa_link_poll_work(struct work_struct *ugly)
 {
-        struct dsa_switch *ds;
+        struct dsa_switch_tree *dst;
+        int i;
+
+        dst = container_of(ugly, struct dsa_switch_tree, link_poll_work);
 
-        ds = container_of(ugly, struct dsa_switch, link_poll_work);
+        for (i = 0; i < dst->pd->nr_chips; i++) {
+                struct dsa_switch *ds = dst->ds[i];
 
-        ds->drv->poll_link(ds);
-        mod_timer(&ds->link_poll_timer, round_jiffies(jiffies + HZ));
+                if (ds != NULL && ds->drv->poll_link != NULL)
+                        ds->drv->poll_link(ds);
+        }
+
+        mod_timer(&dst->link_poll_timer, round_jiffies(jiffies + HZ));
 }
 
-static void dsa_link_poll_timer(unsigned long _ds)
+static void dsa_link_poll_timer(unsigned long _dst)
 {
-        struct dsa_switch *ds = (void *)_ds;
+        struct dsa_switch_tree *dst = (void *)_dst;
 
-        schedule_work(&ds->link_poll_work);
+        schedule_work(&dst->link_poll_work);
 }
 
 
@@ -303,18 +305,14 @@ 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;
+        struct dsa_switch_tree *dst;
+        int i;
 
         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)
+        if (pd == NULL || pd->netdev == NULL)
                 return -EINVAL;
 
         dev = dev_to_net_device(pd->netdev);
@@ -326,36 +324,79 @@ static int dsa_probe(struct platform_device *pdev)
                 return -EEXIST;
         }
 
-        ds = dsa_switch_setup(&pdev->dev, pd, bus, dev);
-        if (IS_ERR(ds)) {
+        dst = kzalloc(sizeof(*dst), GFP_KERNEL);
+        if (dst == NULL) {
                 dev_put(dev);
-                return PTR_ERR(ds);
+                return -ENOMEM;
         }
 
-        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, dst);
+
+        dst->pd = pd;
+        dst->master_netdev = dev;
+        dst->cpu_switch = -1;
+        dst->cpu_port = -1;
+
+        for (i = 0; i < pd->nr_chips; i++) {
+                struct mii_bus *bus;
+                struct dsa_switch *ds;
+
+                bus = dev_to_mii_bus(pd->chip[i].mii_bus);
+                if (bus == NULL) {
+                        printk(KERN_ERR "%s[%d]: no mii bus found for "
+                                "dsa switch\n", dev->name, i);
+                        continue;
+                }
+
+                ds = dsa_switch_setup(dst, i, &pdev->dev, bus);
+                if (IS_ERR(ds)) {
+                        printk(KERN_ERR "%s[%d]: couldn't create dsa switch "
+                                "instance (error %ld)\n", dev->name, i,
+                                PTR_ERR(ds));
+                        continue;
+                }
+
+                dst->ds[i] = ds;
+                if (ds->drv->poll_link != NULL)
+                        dst->link_poll_needed = 1;
         }
 
-        platform_set_drvdata(pdev, ds);
+        /*
+         * 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.
+         */
+        wmb();
+        dev->dsa_ptr = (void *)dst;
+
+        if (dst->link_poll_needed) {
+                INIT_WORK(&dst->link_poll_work, dsa_link_poll_work);
+                init_timer(&dst->link_poll_timer);
+                dst->link_poll_timer.data = (unsigned long)dst;
+                dst->link_poll_timer.function = dsa_link_poll_timer;
+                dst->link_poll_timer.expires = round_jiffies(jiffies + HZ);
+                add_timer(&dst->link_poll_timer);
+        }
 
         return 0;
 }
 
 static int dsa_remove(struct platform_device *pdev)
 {
-        struct dsa_switch *ds = platform_get_drvdata(pdev);
+        struct dsa_switch_tree *dst = platform_get_drvdata(pdev);
+        int i;
 
-        if (ds->drv->poll_link != NULL)
-                del_timer_sync(&ds->link_poll_timer);
+        if (dst->link_poll_needed)
+                del_timer_sync(&dst->link_poll_timer);
 
         flush_scheduled_work();
 
-        dsa_switch_destroy(ds);
+        for (i = 0; i < dst->pd->nr_chips; i++) {
+                struct dsa_switch *ds = dst->ds[i];
+
+                if (ds != NULL)
+                        dsa_switch_destroy(ds);
+        }
 
         return 0;
 }