Merge branch 'net-core-fix-a-lockdep-splat-in-the-dev_addr_list'

Taehee Yoo says:

====================
net: core: fix a lockdep splat in the dev_addr_list.

This patchset is to avoid lockdep splat.

When a stacked interface graph is changed, netif_addr_lock() is called
recursively and it internally calls spin_lock_nested().
The parameter of spin_lock_nested() is 'dev->lower_level',
this is called subclass.
The problem of 'dev->lower_level' is that while 'dev->lower_level' is
being used as a subclass of spin_lock_nested(), its value can be changed.
So, spin_lock_nested() would be called recursively with the same
subclass value, the lockdep understands a deadlock.
In order to avoid this, a new variable is needed and it is going to be
used as a parameter of spin_lock_nested().
The first and second patch is a preparation patch for the third patch.
In the third patch, the problem will be fixed.

The first patch is to add __netdev_upper_dev_unlink().
An existed netdev_upper_dev_unlink() is renamed to
__netdev_upper_dev_unlink(). and netdev_upper_dev_unlink()
is added as an wrapper of this function.

The second patch is to add the netdev_nested_priv structure.
netdev_walk_all_{ upper | lower }_dev() pass both private functions
and "data" pointer to handle their own things.
At this point, the data pointer type is void *.
In order to make it easier to expand common variables and functions,
this new netdev_nested_priv structure is added.

The third patch is to add a new variable 'nested_level'
into the net_device structure.
This variable will be used as a parameter of spin_lock_nested() of
dev->addr_list_lock.
Due to this variable, it can avoid lockdep splat.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 54 insertions, 14 deletions

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 7bd4fcdd0738..9fdb3ebef306 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1955,6 +1955,7 @@ struct net_device {
 	unsigned short		type;
 	unsigned short		hard_header_len;
 	unsigned char		min_header_len;
+	unsigned char		name_assign_type;
 
 	unsigned short		needed_headroom;
 	unsigned short		needed_tailroom;
@@ -1965,12 +1966,12 @@ struct net_device {
 	unsigned char		addr_len;
 	unsigned char		upper_level;
 	unsigned char		lower_level;
+
 	unsigned short		neigh_priv_len;
 	unsigned short          dev_id;
 	unsigned short          dev_port;
 	spinlock_t		addr_list_lock;
-	unsigned char		name_assign_type;
-	bool			uc_promisc;
+
 	struct netdev_hw_addr_list	uc;
 	struct netdev_hw_addr_list	mc;
 	struct netdev_hw_addr_list	dev_addrs;
@@ -1978,8 +1979,15 @@ struct net_device {
 #ifdef CONFIG_SYSFS
 	struct kset		*queues_kset;
 #endif
+#ifdef CONFIG_LOCKDEP
+	struct list_head	unlink_list;
+#endif
 	unsigned int		promiscuity;
 	unsigned int		allmulti;
+	bool			uc_promisc;
+#ifdef CONFIG_LOCKDEP
+	unsigned char		nested_level;
+#endif
 
 
 	/* Protocol-specific pointers */
@@ -4260,17 +4268,23 @@ static inline void netif_tx_disable(struct net_device *dev)
 
 static inline void netif_addr_lock(struct net_device *dev)
 {
-	spin_lock(&dev->addr_list_lock);
-}
+	unsigned char nest_level = 0;
 
-static inline void netif_addr_lock_nested(struct net_device *dev)
-{
-	spin_lock_nested(&dev->addr_list_lock, dev->lower_level);
+#ifdef CONFIG_LOCKDEP
+	nest_level = dev->nested_level;
+#endif
+	spin_lock_nested(&dev->addr_list_lock, nest_level);
 }
 
 static inline void netif_addr_lock_bh(struct net_device *dev)
 {
-	spin_lock_bh(&dev->addr_list_lock);
+	unsigned char nest_level = 0;
+
+#ifdef CONFIG_LOCKDEP
+	nest_level = dev->nested_level;
+#endif
+	local_bh_disable();
+	spin_lock_nested(&dev->addr_list_lock, nest_level);
 }
 
 static inline void netif_addr_unlock(struct net_device *dev)
@@ -4455,12 +4469,38 @@ extern int		dev_rx_weight;
 extern int		dev_tx_weight;
 extern int		gro_normal_batch;
 
+enum {
+	NESTED_SYNC_IMM_BIT,
+	NESTED_SYNC_TODO_BIT,
+};
+
+#define __NESTED_SYNC_BIT(bit)	((u32)1 << (bit))
+#define __NESTED_SYNC(name)	__NESTED_SYNC_BIT(NESTED_SYNC_ ## name ## _BIT)
+
+#define NESTED_SYNC_IMM		__NESTED_SYNC(IMM)
+#define NESTED_SYNC_TODO	__NESTED_SYNC(TODO)
+
+struct netdev_nested_priv {
+	unsigned char flags;
+	void *data;
+};
+
 bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
 struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
 						     struct list_head **iter);
 struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
 						     struct list_head **iter);
 
+#ifdef CONFIG_LOCKDEP
+static LIST_HEAD(net_unlink_list);
+
+static inline void net_unlink_todo(struct net_device *dev)
+{
+	if (list_empty(&dev->unlink_list))
+		list_add_tail(&dev->unlink_list, &net_unlink_list);
+}
+#endif
+
 /* iterate through upper list, must be called under RCU read lock */
 #define netdev_for_each_upper_dev_rcu(dev, updev, iter) \
 	for (iter = &(dev)->adj_list.upper, \
@@ -4470,8 +4510,8 @@ struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
 
 int netdev_walk_all_upper_dev_rcu(struct net_device *dev,
 				  int (*fn)(struct net_device *upper_dev,
-					    void *data),
-				  void *data);
+					    struct netdev_nested_priv *priv),
+				  struct netdev_nested_priv *priv);
 
 bool netdev_has_upper_dev_all_rcu(struct net_device *dev,
 				  struct net_device *upper_dev);
@@ -4508,12 +4548,12 @@ struct net_device *netdev_next_lower_dev_rcu(struct net_device *dev,
 					     struct list_head **iter);
 int netdev_walk_all_lower_dev(struct net_device *dev,
 			      int (*fn)(struct net_device *lower_dev,
-					void *data),
-			      void *data);
+					struct netdev_nested_priv *priv),
+			      struct netdev_nested_priv *priv);
 int netdev_walk_all_lower_dev_rcu(struct net_device *dev,
 				  int (*fn)(struct net_device *lower_dev,
-					    void *data),
-				  void *data);
+					    struct netdev_nested_priv *priv),
+				  struct netdev_nested_priv *priv);
 
 void *netdev_adjacent_get_private(struct list_head *adj_list);
 void *netdev_lower_get_first_private_rcu(struct net_device *dev);