Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next-2.6: (1244 commits)
  pkt_sched: Rename PSCHED_US2NS and PSCHED_NS2US
  ipv4: Fix fib_trie rebalancing
  Bluetooth: Fix issue with uninitialized nsh.type in DTL-1 driver
  Bluetooth: Fix Kconfig issue with RFKILL integration
  PIM-SM: namespace changes
  ipv4: update ARPD help text
  net: use a deferred timer in rt_check_expire
  ieee802154: fix kconfig bool/tristate muckup
  bonding: initialization rework
  bonding: use is_zero_ether_addr
  bonding: network device names are case sensative
  bonding: elminate bad refcount code
  bonding: fix style issues
  bonding: fix destructor
  bonding: remove bonding read/write semaphore
  bonding: initialize before registration
  bonding: bond_create always called with default parameters
  x_tables: Convert printk to pr_err
  netfilter: conntrack: optional reliable conntrack event delivery
  list_nulls: add hlist_nulls_add_head and hlist_nulls_del
  ...

1 files changed, 130 insertions, 7 deletions

diff --git a/net/core/drop_monitor.c b/net/core/drop_monitor.c
index b75b6cea49da..9d66fa953ab7 100644
--- a/net/core/drop_monitor.c
+++ b/net/core/drop_monitor.c
@@ -22,8 +22,10 @@
 #include <linux/timer.h>
 #include <linux/bitops.h>
 #include <net/genetlink.h>
+#include <net/netevent.h>
 
 #include <trace/events/skb.h>
+#include <trace/events/napi.h>
 
 #include <asm/unaligned.h>
 
@@ -38,7 +40,8 @@ static void send_dm_alert(struct work_struct *unused);
  * and the work handle that will send up
  * netlink alerts
  */
-struct sock *dm_sock;
+static int trace_state = TRACE_OFF;
+static spinlock_t trace_state_lock = SPIN_LOCK_UNLOCKED;
 
 struct per_cpu_dm_data {
         struct work_struct dm_alert_work;
@@ -47,11 +50,18 @@ struct per_cpu_dm_data {
         struct timer_list send_timer;
 };
 
+struct dm_hw_stat_delta {
+        struct net_device *dev;
+        struct list_head list;
+        struct rcu_head rcu;
+        unsigned long last_drop_val;
+};
+
 static struct genl_family net_drop_monitor_family = {
         .id             = GENL_ID_GENERATE,
         .hdrsize        = 0,
         .name           = "NET_DM",
-        .version        = 1,
+        .version        = 2,
         .maxattr        = NET_DM_CMD_MAX,
 };
 
@@ -59,19 +69,24 @@ static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_cpu_data);
 
 static int dm_hit_limit = 64;
 static int dm_delay = 1;
-
+static unsigned long dm_hw_check_delta = 2*HZ;
+static LIST_HEAD(hw_stats_list);
 
 static void reset_per_cpu_data(struct per_cpu_dm_data *data)
 {
         size_t al;
         struct net_dm_alert_msg *msg;
+        struct nlattr *nla;
 
         al = sizeof(struct net_dm_alert_msg);
         al += dm_hit_limit * sizeof(struct net_dm_drop_point);
+        al += sizeof(struct nlattr);
+
         data->skb = genlmsg_new(al, GFP_KERNEL);
         genlmsg_put(data->skb, 0, 0, &net_drop_monitor_family,
                         0, NET_DM_CMD_ALERT);
-        msg = __nla_reserve_nohdr(data->skb, sizeof(struct net_dm_alert_msg));
+        nla = nla_reserve(data->skb, NLA_UNSPEC, sizeof(struct net_dm_alert_msg));
+        msg = nla_data(nla);
         memset(msg, 0, al);
         atomic_set(&data->dm_hit_count, dm_hit_limit);
 }
@@ -111,10 +126,11 @@ static void sched_send_work(unsigned long unused)
         schedule_work(&data->dm_alert_work);
 }
 
-static void trace_kfree_skb_hit(struct sk_buff *skb, void *location)
+static void trace_drop_common(struct sk_buff *skb, void *location)
 {
         struct net_dm_alert_msg *msg;
         struct nlmsghdr *nlh;
+        struct nlattr *nla;
         int i;
         struct per_cpu_dm_data *data = &__get_cpu_var(dm_cpu_data);
 
@@ -127,7 +143,8 @@ static void trace_kfree_skb_hit(struct sk_buff *skb, void *location)
         }
 
         nlh = (struct nlmsghdr *)data->skb->data;
-        msg = genlmsg_data(nlmsg_data(nlh));
+        nla = genlmsg_data(nlmsg_data(nlh));
+        msg = nla_data(nla);
         for (i = 0; i < msg->entries; i++) {
                 if (!memcmp(&location, msg->points[i].pc, sizeof(void *))) {
                         msg->points[i].count++;
@@ -139,6 +156,7 @@ static void trace_kfree_skb_hit(struct sk_buff *skb, void *location)
          * We need to create a new entry
          */
         __nla_reserve_nohdr(data->skb, sizeof(struct net_dm_drop_point));
+        nla->nla_len += NLA_ALIGN(sizeof(struct net_dm_drop_point));
         memcpy(msg->points[msg->entries].pc, &location, sizeof(void *));
         msg->points[msg->entries].count = 1;
         msg->entries++;
@@ -152,24 +170,80 @@ out:
         return;
 }
 
+static void trace_kfree_skb_hit(struct sk_buff *skb, void *location)
+{
+        trace_drop_common(skb, location);
+}
+
+static void trace_napi_poll_hit(struct napi_struct *napi)
+{
+        struct dm_hw_stat_delta *new_stat;
+
+        /*
+         * Ratelimit our check time to dm_hw_check_delta jiffies
+         */
+        if (!time_after(jiffies, napi->dev->last_rx + dm_hw_check_delta))
+                return;
+
+        rcu_read_lock();
+        list_for_each_entry_rcu(new_stat, &hw_stats_list, list) {
+                if ((new_stat->dev == napi->dev)  &&
+                    (napi->dev->stats.rx_dropped != new_stat->last_drop_val)) {
+                        trace_drop_common(NULL, NULL);
+                        new_stat->last_drop_val = napi->dev->stats.rx_dropped;
+                        break;
+                }
+        }
+        rcu_read_unlock();
+}
+
+
+static void free_dm_hw_stat(struct rcu_head *head)
+{
+        struct dm_hw_stat_delta *n;
+        n = container_of(head, struct dm_hw_stat_delta, rcu);
+        kfree(n);
+}
+
 static int set_all_monitor_traces(int state)
 {
         int rc = 0;
+        struct dm_hw_stat_delta *new_stat = NULL;
+        struct dm_hw_stat_delta *temp;
+
+        spin_lock(&trace_state_lock);
 
         switch (state) {
         case TRACE_ON:
                 rc |= register_trace_kfree_skb(trace_kfree_skb_hit);
+                rc |= register_trace_napi_poll(trace_napi_poll_hit);
                 break;
         case TRACE_OFF:
                 rc |= unregister_trace_kfree_skb(trace_kfree_skb_hit);
+                rc |= unregister_trace_napi_poll(trace_napi_poll_hit);
 
                 tracepoint_synchronize_unregister();
+
+                /*
+                 * Clean the device list
+                 */
+                list_for_each_entry_safe(new_stat, temp, &hw_stats_list, list) {
+                        if (new_stat->dev == NULL) {
+                                list_del_rcu(&new_stat->list);
+                                call_rcu(&new_stat->rcu, free_dm_hw_stat);
+                        }
+                }
                 break;
         default:
                 rc = 1;
                 break;
         }
 
+        if (!rc)
+                trace_state = state;
+
+        spin_unlock(&trace_state_lock);
+
         if (rc)
                 return -EINPROGRESS;
         return rc;
@@ -197,6 +271,44 @@ static int net_dm_cmd_trace(struct sk_buff *skb,
         return -ENOTSUPP;
 }
 
+static int dropmon_net_event(struct notifier_block *ev_block,
+                        unsigned long event, void *ptr)
+{
+        struct net_device *dev = ptr;
+        struct dm_hw_stat_delta *new_stat = NULL;
+        struct dm_hw_stat_delta *tmp;
+
+        switch (event) {
+        case NETDEV_REGISTER:
+                new_stat = kzalloc(sizeof(struct dm_hw_stat_delta), GFP_KERNEL);
+
+                if (!new_stat)
+                        goto out;
+
+                new_stat->dev = dev;
+                INIT_RCU_HEAD(&new_stat->rcu);
+                spin_lock(&trace_state_lock);
+                list_add_rcu(&new_stat->list, &hw_stats_list);
+                spin_unlock(&trace_state_lock);
+                break;
+        case NETDEV_UNREGISTER:
+                spin_lock(&trace_state_lock);
+                list_for_each_entry_safe(new_stat, tmp, &hw_stats_list, list) {
+                        if (new_stat->dev == dev) {
+                                new_stat->dev = NULL;
+                                if (trace_state == TRACE_OFF) {
+                                        list_del_rcu(&new_stat->list);
+                                        call_rcu(&new_stat->rcu, free_dm_hw_stat);
+                                        break;
+                                }
+                        }
+                }
+                spin_unlock(&trace_state_lock);
+                break;
+        }
+out:
+        return NOTIFY_DONE;
+}
 
 static struct genl_ops dropmon_ops[] = {
         {
@@ -213,6 +325,10 @@ static struct genl_ops dropmon_ops[] = {
         },
 };
 
+static struct notifier_block dropmon_net_notifier = {
+        .notifier_call = dropmon_net_event
+};
+
 static int __init init_net_drop_monitor(void)
 {
         int cpu;
@@ -236,12 +352,18 @@ static int __init init_net_drop_monitor(void)
                 ret = genl_register_ops(&net_drop_monitor_family,
                                         &dropmon_ops[i]);
                 if (ret) {
-                        printk(KERN_CRIT "failed to register operation %d\n",
+                        printk(KERN_CRIT "Failed to register operation %d\n",
                                 dropmon_ops[i].cmd);
                         goto out_unreg;
                 }
         }
 
+        rc = register_netdevice_notifier(&dropmon_net_notifier);
+        if (rc < 0) {
+                printk(KERN_CRIT "Failed to register netdevice notifier\n");
+                goto out_unreg;
+        }
+
         rc = 0;
 
         for_each_present_cpu(cpu) {
@@ -252,6 +374,7 @@ static int __init init_net_drop_monitor(void)
                 data->send_timer.data = cpu;
                 data->send_timer.function = sched_send_work;
         }
+
         goto out;
 
 out_unreg: