rxrpc: Improve management and caching of client connection objects

Improve the management and caching of client rxrpc connection objects.
From this point, client connections will be managed separately from service
connections because AF_RXRPC controls the creation and re-use of client
connections but doesn't have that luxury with service connections.

Further, there will be limits on the numbers of client connections that may
be live on a machine.  No direct restriction will be placed on the number
of client calls, excepting that each client connection can support a
maximum of four concurrent calls.

Note that, for a number of reasons, we don't want to simply discard a
client connection as soon as the last call is apparently finished:

 (1) Security is negotiated per-connection and the context is then shared
     between all calls on that connection.  The context can be negotiated
     again if the connection lapses, but that involves holding up calls
     whilst at least two packets are exchanged and various crypto bits are
     performed - so we'd ideally like to cache it for a little while at
     least.

 (2) If a packet goes astray, we will need to retransmit a final ACK or
     ABORT packet.  To make this work, we need to keep around the
     connection details for a little while.

 (3) The locally held structures represent some amount of setup time, to be
     weighed against their occupation of memory when idle.


To this end, the client connection cache is managed by a state machine on
each connection.  There are five states:

 (1) INACTIVE - The connection is not held in any list and may not have
     been exposed to the world.  If it has been previously exposed, it was
     discarded from the idle list after expiring.

 (2) WAITING - The connection is waiting for the number of client conns to
     drop below the maximum capacity.  Calls may be in progress upon it
     from when it was active and got culled.

     The connection is on the rxrpc_waiting_client_conns list which is kept
     in to-be-granted order.  Culled conns with waiters go to the back of
     the queue just like new conns.

 (3) ACTIVE - The connection has at least one call in progress upon it, it
     may freely grant available channels to new calls and calls may be
     waiting on it for channels to become available.

     The connection is on the rxrpc_active_client_conns list which is kept
     in activation order for culling purposes.

 (4) CULLED - The connection got summarily culled to try and free up
     capacity.  Calls currently in progress on the connection are allowed
     to continue, but new calls will have to wait.  There can be no waiters
     in this state - the conn would have to go to the WAITING state
     instead.

 (5) IDLE - The connection has no calls in progress upon it and must have
     been exposed to the world (ie. the EXPOSED flag must be set).  When it
     expires, the EXPOSED flag is cleared and the connection transitions to
     the INACTIVE state.

     The connection is on the rxrpc_idle_client_conns list which is kept in
     order of how soon they'll expire.

A connection in the ACTIVE or CULLED state must have at least one active
call upon it; if in the WAITING state it may have active calls upon it;
other states may not have active calls.

As long as a connection remains active and doesn't get culled, it may
continue to process calls - even if there are connections on the wait
queue.  This simplifies things a bit and reduces the amount of checking we
need do.


There are a couple flags of relevance to the cache:

 (1) EXPOSED - The connection ID got exposed to the world.  If this flag is
     set, an extra ref is added to the connection preventing it from being
     reaped when it has no calls outstanding.  This flag is cleared and the
     ref dropped when a conn is discarded from the idle list.

 (2) DONT_REUSE - The connection should be discarded as soon as possible and
     should not be reused.


This commit also provides a number of new settings:

 (*) /proc/net/rxrpc/max_client_conns

     The maximum number of live client connections.  Above this number, new
     connections get added to the wait list and must wait for an active
     conn to be culled.  Culled connections can be reused, but they will go
     to the back of the wait list and have to wait.

 (*) /proc/net/rxrpc/reap_client_conns

     If the number of desired connections exceeds the maximum above, the
     active connection list will be culled until there are only this many
     left in it.

 (*) /proc/net/rxrpc/idle_conn_expiry

     The normal expiry time for a client connection, provided there are
     fewer than reap_client_conns of them around.

 (*) /proc/net/rxrpc/idle_conn_fast_expiry

     The expedited expiry time, used when there are more than
     reap_client_conns of them around.


Note that I combined the Tx wait queue with the channel grant wait queue to
save space as only one of these should be in use at once.

Note also that, for the moment, the service connection cache still uses the
old connection management code.

Signed-off-by: David Howells <dhowells@redhat.com>

1 files changed, 46 insertions, 10 deletions

diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 2efbfba87cbe..c761124961cc 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -255,6 +255,9 @@ enum rxrpc_conn_flag {
 	RXRPC_CONN_HAS_IDR,		/* Has a client conn ID assigned */
 	RXRPC_CONN_IN_SERVICE_CONNS,	/* Conn is in peer->service_conns */
 	RXRPC_CONN_IN_CLIENT_CONNS,	/* Conn is in local->client_conns */
+	RXRPC_CONN_EXPOSED,		/* Conn has extra ref for exposure */
+	RXRPC_CONN_DONT_REUSE,		/* Don't reuse this connection */
+	RXRPC_CONN_COUNTED,		/* Counted by rxrpc_nr_client_conns */
 };
 
 /*
@@ -265,6 +268,17 @@ enum rxrpc_conn_event {
 };
 
 /*
+ * The connection cache state.
+ */
+enum rxrpc_conn_cache_state {
+	RXRPC_CONN_CLIENT_INACTIVE,	/* Conn is not yet listed */
+	RXRPC_CONN_CLIENT_WAITING,	/* Conn is on wait list, waiting for capacity */
+	RXRPC_CONN_CLIENT_ACTIVE,	/* Conn is on active list, doing calls */
+	RXRPC_CONN_CLIENT_CULLED,	/* Conn is culled and delisted, doing calls */
+	RXRPC_CONN_CLIENT_IDLE,		/* Conn is on idle list, doing mostly nothing */
+};
+
+/*
  * The connection protocol state.
  */
 enum rxrpc_conn_proto_state {
@@ -276,6 +290,7 @@ enum rxrpc_conn_proto_state {
 	RXRPC_CONN_REMOTELY_ABORTED,	/* Conn aborted by peer */
 	RXRPC_CONN_LOCALLY_ABORTED,	/* Conn aborted locally */
 	RXRPC_CONN_NETWORK_ERROR,	/* Conn terminated by network error */
+	RXRPC_CONN_LOCAL_ERROR,		/* Conn terminated by local error */
 	RXRPC_CONN__NR_STATES
 };
 
@@ -288,8 +303,14 @@ struct rxrpc_connection {
 	struct rxrpc_conn_proto	proto;
 	struct rxrpc_conn_parameters params;
 
-	spinlock_t		channel_lock;
+	atomic_t		usage;
+	struct rcu_head		rcu;
+	struct list_head	cache_link;
 
+	spinlock_t		channel_lock;
+	unsigned char		active_chans;	/* Mask of active channels */
+#define RXRPC_ACTIVE_CHANS_MASK	((1 << RXRPC_MAXCALLS) - 1)
+	struct list_head	waiting_calls;	/* Calls waiting for channels */
 	struct rxrpc_channel {
 		struct rxrpc_call __rcu	*call;		/* Active call */
 		u32			call_id;	/* ID of current call */
@@ -302,9 +323,7 @@ struct rxrpc_connection {
 			u32		last_abort;
 		};
 	} channels[RXRPC_MAXCALLS];
-	wait_queue_head_t	channel_wq;	/* queue to wait for channel to become available */
 
-	struct rcu_head		rcu;
 	struct work_struct	processor;	/* connection event processor */
 	union {
 		struct rb_node	client_node;	/* Node in local->client_conns */
@@ -321,7 +340,7 @@ struct rxrpc_connection {
 	unsigned long		events;
 	unsigned long		idle_timestamp;	/* Time at which last became idle */
 	spinlock_t		state_lock;	/* state-change lock */
-	atomic_t		usage;
+	enum rxrpc_conn_cache_state cache_state : 8;
 	enum rxrpc_conn_proto_state state : 8;	/* current state of connection */
 	u32			local_abort;	/* local abort code */
 	u32			remote_abort;	/* remote abort code */
@@ -329,7 +348,6 @@ struct rxrpc_connection {
 	int			debug_id;	/* debug ID for printks */
 	atomic_t		serial;		/* packet serial number counter */
 	unsigned int		hi_serial;	/* highest serial number received */
-	atomic_t		avail_chans;	/* number of channels available */
 	u8			size_align;	/* data size alignment (for security) */
 	u8			header_size;	/* rxrpc + security header size */
 	u8			security_size;	/* security header size */
@@ -351,6 +369,7 @@ enum rxrpc_call_flag {
 	RXRPC_CALL_HAS_USERID,		/* has a user ID attached */
 	RXRPC_CALL_EXPECT_OOS,		/* expect out of sequence packets */
 	RXRPC_CALL_IS_SERVICE,		/* Call is service call */
+	RXRPC_CALL_EXPOSED,		/* The call was exposed to the world */
 };
 
 /*
@@ -417,13 +436,14 @@ struct rxrpc_call {
 	struct work_struct	destroyer;	/* call destroyer */
 	struct work_struct	processor;	/* packet processor and ACK generator */
 	struct list_head	link;		/* link in master call list */
+	struct list_head	chan_wait_link;	/* Link in conn->waiting_calls */
 	struct hlist_node	error_link;	/* link in error distribution list */
 	struct list_head	accept_link;	/* calls awaiting acceptance */
 	struct rb_node		sock_node;	/* node in socket call tree */
 	struct sk_buff_head	rx_queue;	/* received packets */
 	struct sk_buff_head	rx_oos_queue;	/* packets received out of sequence */
 	struct sk_buff		*tx_pending;	/* Tx socket buffer being filled */
-	wait_queue_head_t	tx_waitq;	/* wait for Tx window space to become available */
+	wait_queue_head_t	waitq;		/* Wait queue for channel or Tx */
 	__be32			crypto_buf[2];	/* Temporary packet crypto buffer */
 	unsigned long		user_call_ID;	/* user-defined call ID */
 	unsigned long		creation_jif;	/* time of call creation */
@@ -546,12 +566,19 @@ static inline bool rxrpc_is_client_call(const struct rxrpc_call *call)
 /*
  * conn_client.c
  */
+extern unsigned int rxrpc_max_client_connections;
+extern unsigned int rxrpc_reap_client_connections;
+extern unsigned int rxrpc_conn_idle_client_expiry;
+extern unsigned int rxrpc_conn_idle_client_fast_expiry;
 extern struct idr rxrpc_client_conn_ids;
 
 void rxrpc_destroy_client_conn_ids(void);
 int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *,
 		       struct sockaddr_rxrpc *, gfp_t);
-void rxrpc_unpublish_client_conn(struct rxrpc_connection *);
+void rxrpc_expose_client_call(struct rxrpc_call *);
+void rxrpc_disconnect_client_call(struct rxrpc_call *);
+void rxrpc_put_client_conn(struct rxrpc_connection *);
+void __exit rxrpc_destroy_all_client_connections(void);
 
 /*
  * conn_event.c
@@ -572,8 +599,9 @@ int rxrpc_extract_addr_from_skb(struct sockaddr_rxrpc *, struct sk_buff *);
 struct rxrpc_connection *rxrpc_alloc_connection(gfp_t);
 struct rxrpc_connection *rxrpc_find_connection_rcu(struct rxrpc_local *,
 						   struct sk_buff *);
-void __rxrpc_disconnect_call(struct rxrpc_call *);
+void __rxrpc_disconnect_call(struct rxrpc_connection *, struct rxrpc_call *);
 void rxrpc_disconnect_call(struct rxrpc_call *);
+void rxrpc_kill_connection(struct rxrpc_connection *);
 void __rxrpc_put_connection(struct rxrpc_connection *);
 void __exit rxrpc_destroy_all_connections(void);
 
@@ -600,8 +628,16 @@ struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *con
 
 static inline void rxrpc_put_connection(struct rxrpc_connection *conn)
 {
-	if (conn && atomic_dec_return(&conn->usage) == 1)
-		__rxrpc_put_connection(conn);
+	if (!conn)
+		return;
+
+	if (rxrpc_conn_is_client(conn)) {
+		if (atomic_dec_and_test(&conn->usage))
+			rxrpc_put_client_conn(conn);
+	} else {
+		if (atomic_dec_return(&conn->usage) == 1)
+			__rxrpc_put_connection(conn);
+	}
 }