Merge cantab.net:/home/src/bklinux-2.6

into cantab.net:/home/src/ntfs-2.6

5 files changed, 107 insertions, 73 deletions

diff --git a/include/net/dst.h b/include/net/dst.h
index ed2504c6b4e9..543ff945bb2f 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -183,6 +183,12 @@ static inline void dst_free(struct dst_entry * dst)
 	__dst_free(dst);
 }
 
+static inline void dst_rcu_free(struct rcu_head *head)
+{
+	struct dst_entry *dst = container_of(head, struct dst_entry, rcu_head);
+	dst_free(dst);
+}
+
 static inline void dst_confirm(struct dst_entry *dst)
 {
 	if (dst)
diff --git a/include/net/protocol.h b/include/net/protocol.h
index f67327486a9c..357691f6a45f 100644
--- a/include/net/protocol.h
+++ b/include/net/protocol.h
@@ -34,8 +34,7 @@
 
 
 /* This is used to register protocols. */
-struct inet_protocol 
-{
+struct net_protocol {
 	int			(*handler)(struct sk_buff *skb);
 	void			(*err_handler)(struct sk_buff *skb, u32 info);
 	int			no_policy;
@@ -78,15 +77,15 @@ struct inet_protosw {
 #define INET_PROTOSW_REUSE 0x01	     /* Are ports automatically reusable? */
 #define INET_PROTOSW_PERMANENT 0x02  /* Permanent protocols are unremovable. */
 
-extern struct inet_protocol *inet_protocol_base;
-extern struct inet_protocol *inet_protos[MAX_INET_PROTOS];
+extern struct net_protocol *inet_protocol_base;
+extern struct net_protocol *inet_protos[MAX_INET_PROTOS];
 
 #if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
 extern struct inet6_protocol *inet6_protos[MAX_INET_PROTOS];
 #endif
 
-extern int	inet_add_protocol(struct inet_protocol *prot, unsigned char num);
-extern int	inet_del_protocol(struct inet_protocol *prot, unsigned char num);
+extern int	inet_add_protocol(struct net_protocol *prot, unsigned char num);
+extern int	inet_del_protocol(struct net_protocol *prot, unsigned char num);
 extern void	inet_register_protosw(struct inet_protosw *p);
 extern void	inet_unregister_protosw(struct inet_protosw *p);
 
diff --git a/include/net/route.h b/include/net/route.h
index 0e7210e6d194..a5e9c575ea3e 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -129,7 +129,7 @@ extern void		ip_rt_send_redirect(struct sk_buff *skb);
 
 extern unsigned		inet_addr_type(u32 addr);
 extern void		ip_rt_multicast_event(struct in_device *);
-extern int		ip_rt_ioctl(unsigned int cmd, void *arg);
+extern int		ip_rt_ioctl(unsigned int cmd, void __user *arg);
 extern void		ip_rt_get_source(u8 *src, struct rtable *rt);
 extern int		ip_rt_dump(struct sk_buff *skb,  struct netlink_callback *cb);
 
diff --git a/include/net/sock.h b/include/net/sock.h
index 0398823e18ed..9da91b9be903 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -167,6 +167,8 @@ struct sock_common {
   *	@sk_socket - Identd and reporting IO signals
   *	@sk_user_data - RPC layer private data
   *	@sk_owner - module that owns this socket
+  *	@sk_sndmsg_page - cached page for sendmsg
+  *	@sk_sndmsg_off - cached offset for sendmsg
   *	@sk_send_head - front of stuff to transmit
   *	@sk_write_pending - a write to stream socket waits to start
   *	@sk_queue_shrunk - write queue has been shrunk recently
@@ -249,8 +251,10 @@ struct sock {
 	struct timeval		sk_stamp;
 	struct socket		*sk_socket;
 	void			*sk_user_data;
-	struct sk_buff		*sk_send_head;
 	struct module		*sk_owner;
+	struct page		*sk_sndmsg_page;
+	__u32			sk_sndmsg_off;
+	struct sk_buff		*sk_send_head;
 	int			sk_write_pending;
 	void			*sk_security;
 	__u8			sk_queue_shrunk;
@@ -487,10 +491,11 @@ extern int sk_stream_wait_connect(struct sock *sk, long *timeo_p);
 extern int sk_stream_wait_memory(struct sock *sk, long *timeo_p);
 extern void sk_stream_wait_close(struct sock *sk, long timeo_p);
 extern int sk_stream_error(struct sock *sk, int flags, int err);
+extern void sk_stream_kill_queues(struct sock *sk);
 
 extern int sk_wait_data(struct sock *sk, long *timeo);
 
-/* IP protocol blocks we attach to sockets.
+/* Networking protocol blocks we attach to sockets.
  * socket layer -> transport layer interface
  * transport -> network interface is defined by struct inet_proto
  */
@@ -534,6 +539,22 @@ struct proto {
 	void			(*unhash)(struct sock *sk);
 	int			(*get_port)(struct sock *sk, unsigned short snum);
 
+	/* Memory pressure */
+	void			(*enter_memory_pressure)(void);
+	atomic_t		*memory_allocated;	/* Current allocated memory. */
+	atomic_t		*sockets_allocated;	/* Current number of sockets. */
+	/*
+	 * Pressure flag: try to collapse.
+	 * Technical note: it is used by multiple contexts non atomically.
+	 * All the sk_stream_mem_schedule() is of this nature: accounting
+	 * is strict, actions are advisory and have some latency.
+	 */
+	int			*memory_pressure;
+	int			*sysctl_mem;
+	int			*sysctl_wmem;
+	int			*sysctl_rmem;
+	int			max_header;
+
 	char			name[32];
 
 	struct {
@@ -624,6 +645,37 @@ static inline struct inode *SOCK_INODE(struct socket *socket)
 	return &container_of(socket, struct socket_alloc, socket)->vfs_inode;
 }
 
+extern void __sk_stream_mem_reclaim(struct sock *sk);
+extern int sk_stream_mem_schedule(struct sock *sk, int size, int kind);
+
+#define SK_STREAM_MEM_QUANTUM ((int)PAGE_SIZE)
+
+static inline int sk_stream_pages(int amt)
+{
+	return (amt + SK_STREAM_MEM_QUANTUM - 1) / SK_STREAM_MEM_QUANTUM;
+}
+
+static inline void sk_stream_mem_reclaim(struct sock *sk)
+{
+	if (sk->sk_forward_alloc >= SK_STREAM_MEM_QUANTUM)
+		__sk_stream_mem_reclaim(sk);
+}
+
+static inline void sk_stream_writequeue_purge(struct sock *sk)
+{
+	struct sk_buff *skb;
+
+	while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL)
+		sk_stream_free_skb(sk, skb);
+	sk_stream_mem_reclaim(sk);
+}
+
+static inline int sk_stream_rmem_schedule(struct sock *sk, struct sk_buff *skb)
+{
+	return (int)skb->truesize <= sk->sk_forward_alloc ||
+		sk_stream_mem_schedule(sk, skb->truesize, 1);
+}
+
 /* Used by processes to "lock" a socket state, so that
  * interrupts and bottom half handlers won't change it
  * from under us. It essentially blocks any incoming
@@ -1105,6 +1157,46 @@ static inline void sk_stream_moderate_sndbuf(struct sock *sk)
 	}
 }
 
+static inline struct sk_buff *sk_stream_alloc_pskb(struct sock *sk,
+						   int size, int mem, int gfp)
+{
+	struct sk_buff *skb = alloc_skb(size + sk->sk_prot->max_header, gfp);
+
+	if (skb) {
+		skb->truesize += mem;
+		if (sk->sk_forward_alloc >= (int)skb->truesize ||
+		    sk_stream_mem_schedule(sk, skb->truesize, 0)) {
+			skb_reserve(skb, sk->sk_prot->max_header);
+			return skb;
+		}
+		__kfree_skb(skb);
+	} else {
+		sk->sk_prot->enter_memory_pressure();
+		sk_stream_moderate_sndbuf(sk);
+	}
+	return NULL;
+}
+
+static inline struct sk_buff *sk_stream_alloc_skb(struct sock *sk,
+						  int size, int gfp)
+{
+	return sk_stream_alloc_pskb(sk, size, 0, gfp);
+}
+
+static inline struct page *sk_stream_alloc_page(struct sock *sk)
+{
+	struct page *page = NULL;
+
+	if (sk->sk_forward_alloc >= (int)PAGE_SIZE ||
+	    sk_stream_mem_schedule(sk, PAGE_SIZE, 0))
+		page = alloc_pages(sk->sk_allocation, 0);
+	else {
+		sk->sk_prot->enter_memory_pressure();
+		sk_stream_moderate_sndbuf(sk);
+	}
+	return page;
+}
+
 #define sk_stream_for_retrans_queue(skb, sk)				\
 		for (skb = (sk)->sk_write_queue.next;			\
 		     (skb != (sk)->sk_send_head) &&			\
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 22a62d58c24d..3f4c524de3e2 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -1867,70 +1867,7 @@ static __inline__ void tcp_openreq_init(struct open_request *req,
 	req->rmt_port = skb->h.th->source;
 }
 
-#define TCP_MEM_QUANTUM	((int)PAGE_SIZE)
-
-extern void __tcp_mem_reclaim(struct sock *sk);
-extern int tcp_mem_schedule(struct sock *sk, int size, int kind);
-
-static inline void tcp_mem_reclaim(struct sock *sk)
-{
-	if (sk->sk_forward_alloc >= TCP_MEM_QUANTUM)
-		__tcp_mem_reclaim(sk);
-}
-
-static inline void tcp_enter_memory_pressure(void)
-{
-	if (!tcp_memory_pressure) {
-		NET_INC_STATS(TCPMemoryPressures);
-		tcp_memory_pressure = 1;
-	}
-}
-
-static inline struct sk_buff *tcp_alloc_pskb(struct sock *sk, int size, int mem, int gfp)
-{
-	struct sk_buff *skb = alloc_skb(size+MAX_TCP_HEADER, gfp);
-
-	if (skb) {
-		skb->truesize += mem;
-		if (sk->sk_forward_alloc >= (int)skb->truesize ||
-		    tcp_mem_schedule(sk, skb->truesize, 0)) {
-			skb_reserve(skb, MAX_TCP_HEADER);
-			return skb;
-		}
-		__kfree_skb(skb);
-	} else {
-		tcp_enter_memory_pressure();
-		sk_stream_moderate_sndbuf(sk);
-	}
-	return NULL;
-}
-
-static inline struct sk_buff *tcp_alloc_skb(struct sock *sk, int size, int gfp)
-{
-	return tcp_alloc_pskb(sk, size, 0, gfp);
-}
-
-static inline struct page * tcp_alloc_page(struct sock *sk)
-{
-	if (sk->sk_forward_alloc >= (int)PAGE_SIZE ||
-	    tcp_mem_schedule(sk, PAGE_SIZE, 0)) {
-		struct page *page = alloc_pages(sk->sk_allocation, 0);
-		if (page)
-			return page;
-	}
-	tcp_enter_memory_pressure();
-	sk_stream_moderate_sndbuf(sk);
-	return NULL;
-}
-
-static inline void tcp_writequeue_purge(struct sock *sk)
-{
-	struct sk_buff *skb;
-
-	while ((skb = __skb_dequeue(&sk->sk_write_queue)) != NULL)
-		sk_stream_free_skb(sk, skb);
-	tcp_mem_reclaim(sk);
-}
+extern void tcp_enter_memory_pressure(void);
 
 extern void tcp_listen_wlock(void);