* [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs
@ 2016-06-22 9:49 David Howells
2016-06-22 9:49 ` [PATCH net-next 01/14] rxrpc: checking for IS_ERR() instead of NULL David Howells
` (14 more replies)
0 siblings, 15 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:49 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Here's the next part of the AF_RXRPC rewrite. The primary purpose of this
set is to get rid of the rxrpc_conn_bundle and rxrpc_transport structs.
This simplifies things for future development of the connection handling.
To this end, the following significant changes are made:
(1) The rxrpc_connection struct is given pointers to the local and peer
endpoints, inside the rxrpc_conn_parameters struct. Pointers to the
transport's copy of these pointers are then redirected to the
connection struct.
(2) Exclusive connection handling is fixed. Exclusive connections should
do just one call and then be retired. They are used in security
negotiations and, I believe, the idea is to avoid reuse of negotiated
security contexts.
The current code is doing a single connection per socket and doing all
the calls over that. With this change it gets a new connection for
each call made.
(3) A new sendmsg() control message marker is added to make individual
calls operate over exclusive connections. This should be used in
future in preference to the sockopt that marks a socket as "exclusive
connection".
(4) IDs for client connections initiated by a machine are now allocated
from a global pool using the IDR facility and are unique across all
client connections, no matter their destination. The IDR facility is
then used to look up a connection on the connection ID alone. Other
parameters are then verified afterwards.
Note that the IDR facility may use a lot of memory if the IDs it holds
are widely scattered. Given this, in a future commit, client
connections will be retired if they are more than a certain distance
from the last ID allocated.
The client epoch is advanced by 1 each time the client ID counter
wraps. Connections outside the current epoch will also be retired in
a future commit.
(5) The connection bundle concept is removed and the client connection
tree is moved into the local endpoint. The queue for waiting for a
call channel is moved to the rxrpc_connection struct as there can only
be one connection for any particular key going to any particular peer
now.
(6) The rxrpc_transport struct is removed and the service connection tree
is moved into the peer struct.
The patches can be found here also:
http://git.kernel.org/cgit/linux/kernel/git/dhowells/linux-fs.git/log/?h=rxrpc-rewrite
Tagged thusly:
git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git
rxrpc-rewrite-20160622
David
---
Arnd Bergmann (1):
rxrpc: fix uninitialized variable use
Dan Carpenter (1):
rxrpc: checking for IS_ERR() instead of NULL
David Howells (12):
rxrpc: Use structs to hold connection params and protocol info
rxrpc: Replace conn->trans->{local,peer} with conn->params.{local,peer}
rxrpc: Fix exclusive connection handling
rxrpc: Pass sk_buff * rather than rxrpc_host_header * to functions
rxrpc: rxrpc_connection_lock shouldn't be a BH lock, but conn_lock is
rxrpc: Use IDR to allocate client conn IDs on a machine-wide basis
rxrpc: Validate the net address given to rxrpc_kernel_begin_call()
rxrpc: Calls displayed in /proc may in future lack a connection
rxrpc: Make rxrpc_send_packet() take a connection not a transport
rxrpc: Provide more refcount helper functions
rxrpc: Kill the client connection bundle concept
rxrpc: Kill off the rxrpc_transport struct
include/linux/rxrpc.h | 3
net/rxrpc/Makefile | 2
net/rxrpc/af_rxrpc.c | 87 +----
net/rxrpc/ar-internal.h | 173 +++++----
net/rxrpc/call_accept.c | 17 -
net/rxrpc/call_event.c | 20 +
net/rxrpc/call_object.c | 206 ++++++-----
net/rxrpc/conn_client.c | 99 +++++
net/rxrpc/conn_event.c | 16 -
net/rxrpc/conn_object.c | 860 +++++++++++++++++-----------------------------
net/rxrpc/input.c | 23 -
net/rxrpc/key.c | 2
net/rxrpc/local_object.c | 5
net/rxrpc/output.c | 109 +++---
net/rxrpc/peer_object.c | 5
net/rxrpc/proc.c | 39 +-
net/rxrpc/recvmsg.c | 6
net/rxrpc/rxkad.c | 74 ++--
net/rxrpc/security.c | 8
net/rxrpc/sysctl.c | 8
net/rxrpc/transport.c | 269 --------------
21 files changed, 780 insertions(+), 1251 deletions(-)
create mode 100644 net/rxrpc/conn_client.c
delete mode 100644 net/rxrpc/transport.c
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH net-next 01/14] rxrpc: checking for IS_ERR() instead of NULL
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
@ 2016-06-22 9:49 ` David Howells
2016-06-22 9:49 ` [PATCH net-next 02/14] rxrpc: fix uninitialized variable use David Howells
` (13 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:49 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, Dan Carpenter, linux-kernel
From: Dan Carpenter <dan.carpenter@oracle.com>
rxrpc_lookup_peer_rcu() and rxrpc_lookup_peer() return NULL on error, never
error pointers, so IS_ERR() can't be used.
Fix three callers of those functions.
Fixes: be6e6707f6ee ('rxrpc: Rework peer object handling to use hash table and RCU')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/af_rxrpc.c | 4 ++--
net/rxrpc/call_accept.c | 2 +-
net/rxrpc/input.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index c83c3c75d665..9dd160bb16d2 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -247,8 +247,8 @@ struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx,
/* find a remote transport endpoint from the local one */
peer = rxrpc_lookup_peer(rx->local, srx, gfp);
- if (IS_ERR(peer))
- return ERR_CAST(peer);
+ if (!peer)
+ return ERR_PTR(-ENOMEM);
/* find a transport */
trans = rxrpc_get_transport(rx->local, peer, gfp);
diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c
index 50136c76ebd1..553b67c144e5 100644
--- a/net/rxrpc/call_accept.c
+++ b/net/rxrpc/call_accept.c
@@ -96,7 +96,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local,
notification->mark = RXRPC_SKB_MARK_NEW_CALL;
peer = rxrpc_lookup_peer(local, srx, GFP_NOIO);
- if (IS_ERR(peer)) {
+ if (!peer) {
_debug("no peer");
ret = -EBUSY;
goto error;
diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index 47fb167af3e4..e11e4d785127 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -639,7 +639,7 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
rxrpc_get_addr_from_skb(local, skb, &srx);
rcu_read_lock();
peer = rxrpc_lookup_peer_rcu(local, &srx);
- if (IS_ERR(peer))
+ if (!peer)
goto cant_find_peer;
trans = rxrpc_find_transport(local, peer);
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 02/14] rxrpc: fix uninitialized variable use
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
2016-06-22 9:49 ` [PATCH net-next 01/14] rxrpc: checking for IS_ERR() instead of NULL David Howells
@ 2016-06-22 9:49 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 03/14] rxrpc: Use structs to hold connection params and protocol info David Howells
` (12 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:49 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, Arnd Bergmann, linux-kernel
From: Arnd Bergmann <arnd@arndb.de>
Hashing the peer key was introduced for AF_INET, but gcc
warns about the rxrpc_peer_hash_key function returning uninitialized
data for any other value of srx->transport.family:
net/rxrpc/peer_object.c: In function 'rxrpc_peer_hash_key':
net/rxrpc/peer_object.c:57:15: error: 'p' may be used uninitialized in this function [-Werror=maybe-uninitialized]
Assuming that nothing else can be set here, this changes the
function to just return zero in case of an unknown address
family.
Fixes: be6e6707f6ee ("rxrpc: Rework peer object handling to use hash table and RCU")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/peer_object.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c
index faf222c21698..6baad708f3b1 100644
--- a/net/rxrpc/peer_object.c
+++ b/net/rxrpc/peer_object.c
@@ -50,6 +50,9 @@ static unsigned long rxrpc_peer_hash_key(struct rxrpc_local *local,
size = sizeof(srx->transport.sin.sin_addr);
p = (u16 *)&srx->transport.sin.sin_addr;
break;
+ default:
+ WARN(1, "AF_RXRPC: Unsupported transport address family\n");
+ return 0;
}
/* Step through the peer address in 16-bit portions for speed */
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 03/14] rxrpc: Use structs to hold connection params and protocol info
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
2016-06-22 9:49 ` [PATCH net-next 01/14] rxrpc: checking for IS_ERR() instead of NULL David Howells
2016-06-22 9:49 ` [PATCH net-next 02/14] rxrpc: fix uninitialized variable use David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 04/14] rxrpc: Replace conn->trans->{local, peer} with conn->params.{local, peer} David Howells
` (11 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Define and use a structure to hold connection parameters. This makes it
easier to pass multiple connection parameters around.
Define and use a structure to hold protocol information used to hash a
connection for lookup on incoming packet. Most of these fields will be
disposed of eventually, including the duplicate local pointer.
Whilst we're at it rename "proto" to "family" when referring to a protocol
family.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/af_rxrpc.c | 55 ++++++++++++++-----------
net/rxrpc/ar-internal.h | 61 ++++++++++++++++++++++------
net/rxrpc/call_event.c | 6 +--
net/rxrpc/call_object.c | 44 ++++++++++----------
net/rxrpc/conn_event.c | 8 ++--
net/rxrpc/conn_object.c | 103 +++++++++++++++++++++++++----------------------
net/rxrpc/input.c | 4 +-
net/rxrpc/key.c | 2 -
net/rxrpc/output.c | 24 ++++++++---
net/rxrpc/proc.c | 12 +++--
net/rxrpc/recvmsg.c | 2 -
net/rxrpc/rxkad.c | 62 ++++++++++++++--------------
net/rxrpc/security.c | 6 +--
13 files changed, 226 insertions(+), 163 deletions(-)
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 9dd160bb16d2..48b45a0280c0 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -97,7 +97,7 @@ static int rxrpc_validate_address(struct rxrpc_sock *rx,
srx->transport_len > len)
return -EINVAL;
- if (srx->transport.family != rx->proto)
+ if (srx->transport.family != rx->family)
return -EAFNOSUPPORT;
switch (srx->transport.family) {
@@ -227,32 +227,30 @@ static int rxrpc_listen(struct socket *sock, int backlog)
/*
* find a transport by address
*/
-struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx,
- struct sockaddr *addr,
- int addr_len, int flags,
- gfp_t gfp)
+struct rxrpc_transport *
+rxrpc_name_to_transport(struct rxrpc_conn_parameters *cp,
+ struct sockaddr *addr,
+ int addr_len,
+ gfp_t gfp)
{
struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr;
struct rxrpc_transport *trans;
- struct rxrpc_peer *peer;
- _enter("%p,%p,%d,%d", rx, addr, addr_len, flags);
-
- ASSERT(rx->local != NULL);
+ _enter("%p,%d", addr, addr_len);
- if (rx->srx.transport_type != srx->transport_type)
+ if (cp->local->srx.transport_type != srx->transport_type)
return ERR_PTR(-ESOCKTNOSUPPORT);
- if (rx->srx.transport.family != srx->transport.family)
+ if (cp->local->srx.transport.family != srx->transport.family)
return ERR_PTR(-EAFNOSUPPORT);
/* find a remote transport endpoint from the local one */
- peer = rxrpc_lookup_peer(rx->local, srx, gfp);
- if (!peer)
+ cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp);
+ if (!cp->peer)
return ERR_PTR(-ENOMEM);
/* find a transport */
- trans = rxrpc_get_transport(rx->local, peer, gfp);
- rxrpc_put_peer(peer);
+ trans = rxrpc_get_transport(cp->local, cp->peer, gfp);
+ rxrpc_put_peer(cp->peer);
_leave(" = %p", trans);
return trans;
}
@@ -277,6 +275,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
unsigned long user_call_ID,
gfp_t gfp)
{
+ struct rxrpc_conn_parameters cp;
struct rxrpc_conn_bundle *bundle;
struct rxrpc_transport *trans;
struct rxrpc_call *call;
@@ -286,18 +285,26 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
lock_sock(&rx->sk);
- trans = rxrpc_name_to_transport(rx, (struct sockaddr *)srx,
- sizeof(*srx), 0, gfp);
+ if (!key)
+ key = rx->key;
+ if (key && !key->payload.data[0])
+ key = NULL; /* a no-security key */
+
+ memset(&cp, 0, sizeof(cp));
+ cp.local = rx->local;
+ cp.key = key;
+ cp.security_level = 0;
+ cp.exclusive = false;
+ cp.service_id = srx->srx_service;
+
+ trans = rxrpc_name_to_transport(&cp, (struct sockaddr *)srx,
+ sizeof(*srx), gfp);
if (IS_ERR(trans)) {
call = ERR_CAST(trans);
trans = NULL;
goto out_notrans;
}
-
- if (!key)
- key = rx->key;
- if (key && !key->payload.data[0])
- key = NULL; /* a no-security key */
+ cp.peer = trans->peer;
bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, gfp);
if (IS_ERR(bundle)) {
@@ -305,7 +312,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
goto out;
}
- call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, gfp);
+ call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID, gfp);
rxrpc_put_bundle(trans, bundle);
out:
rxrpc_put_transport(trans);
@@ -600,7 +607,7 @@ static int rxrpc_create(struct net *net, struct socket *sock, int protocol,
sk->sk_destruct = rxrpc_sock_destructor;
rx = rxrpc_sk(sk);
- rx->proto = protocol;
+ rx->family = protocol;
rx->calls = RB_ROOT;
INIT_LIST_HEAD(&rx->listen_link);
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index c168268467cd..efe6673deb28 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -72,7 +72,7 @@ struct rxrpc_sock {
#define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT
struct sockaddr_rxrpc srx; /* local address */
struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */
- sa_family_t proto; /* protocol created with */
+ sa_family_t family; /* protocol family created with */
};
#define rxrpc_sk(__sk) container_of((__sk), struct rxrpc_sock, sk)
@@ -262,6 +262,34 @@ struct rxrpc_conn_bundle {
};
/*
+ * Keys for matching a connection.
+ */
+struct rxrpc_conn_proto {
+ unsigned long hash_key;
+ struct rxrpc_local *local; /* Representation of local endpoint */
+ u32 epoch; /* epoch of this connection */
+ u32 cid; /* connection ID */
+ u8 in_clientflag; /* RXRPC_CLIENT_INITIATED if we are server */
+ u8 addr_size; /* Size of the address */
+ sa_family_t family; /* Transport protocol */
+ __be16 port; /* Peer UDP/UDP6 port */
+ union { /* Peer address */
+ struct in_addr ipv4_addr;
+ struct in6_addr ipv6_addr;
+ u32 raw_addr[0];
+ };
+};
+
+struct rxrpc_conn_parameters {
+ struct rxrpc_local *local; /* Representation of local endpoint */
+ struct rxrpc_peer *peer; /* Remote endpoint */
+ struct key *key; /* Security details */
+ bool exclusive; /* T if conn is exclusive */
+ u16 service_id; /* Service ID for this connection */
+ u32 security_level; /* Security level selected */
+};
+
+/*
* RxRPC connection definition
* - matched by { transport, service_id, conn_id, direction, key }
* - each connection can only handle four simultaneous calls
@@ -269,6 +297,9 @@ struct rxrpc_conn_bundle {
struct rxrpc_connection {
struct rxrpc_transport *trans; /* transport session */
struct rxrpc_conn_bundle *bundle; /* connection bundle (client) */
+ struct rxrpc_conn_proto proto;
+ struct rxrpc_conn_parameters params;
+
struct work_struct processor; /* connection event processor */
struct rb_node node; /* node in transport's lookup tree */
struct list_head link; /* link in master connection list */
@@ -277,7 +308,6 @@ struct rxrpc_connection {
struct sk_buff_head rx_queue; /* received conn-level packets */
struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* channels (active calls) */
const struct rxrpc_security *security; /* applied security module */
- struct key *key; /* security for this connection (client) */
struct key *server_key; /* security for this service */
struct crypto_skcipher *cipher; /* encryption handle */
struct rxrpc_crypt csum_iv; /* packet checksum base */
@@ -308,13 +338,8 @@ struct rxrpc_connection {
u8 size_align; /* data size alignment (for security) */
u8 header_size; /* rxrpc + security header size */
u8 security_size; /* security header size */
- u32 security_level; /* security level negotiated */
u32 security_nonce; /* response re-use preventer */
- u32 epoch; /* epoch of this connection */
- u32 cid; /* connection ID */
- u16 service_id; /* service ID for this connection */
u8 security_ix; /* security type */
- u8 in_clientflag; /* RXRPC_CLIENT_INITIATED if we are server */
u8 out_clientflag; /* RXRPC_CLIENT_INITIATED if we are client */
};
@@ -448,7 +473,7 @@ struct rxrpc_call {
unsigned long hash_key; /* Full hash key */
u8 in_clientflag; /* Copy of conn->in_clientflag for hashing */
struct rxrpc_local *local; /* Local endpoint. Used for hashing. */
- sa_family_t proto; /* Frame protocol */
+ sa_family_t family; /* Frame protocol */
u32 call_id; /* call ID on connection */
u32 cid; /* connection ID plus channel index */
u32 epoch; /* epoch of this connection */
@@ -481,9 +506,9 @@ extern u32 rxrpc_epoch;
extern atomic_t rxrpc_debug_id;
extern struct workqueue_struct *rxrpc_workqueue;
-extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *,
+extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_conn_parameters *,
struct sockaddr *,
- int, int, gfp_t);
+ int, gfp_t);
/*
* call_accept.c
@@ -512,6 +537,7 @@ struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *,
void *, sa_family_t, const void *);
struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long);
struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *,
+ struct rxrpc_conn_parameters *,
struct rxrpc_transport *,
struct rxrpc_conn_bundle *,
unsigned long, gfp_t);
@@ -541,8 +567,9 @@ struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *,
struct rxrpc_transport *,
struct key *, u16, gfp_t);
void rxrpc_put_bundle(struct rxrpc_transport *, struct rxrpc_conn_bundle *);
-int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_transport *,
- struct rxrpc_conn_bundle *, struct rxrpc_call *, gfp_t);
+int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *,
+ struct rxrpc_transport *, struct rxrpc_conn_bundle *,
+ struct rxrpc_call *, gfp_t);
void rxrpc_put_connection(struct rxrpc_connection *);
void __exit rxrpc_destroy_all_connections(void);
struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *,
@@ -550,6 +577,16 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *,
extern struct rxrpc_connection *
rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_host_header *);
+static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn)
+{
+ return conn->out_clientflag;
+}
+
+static inline bool rxrpc_conn_is_service(const struct rxrpc_connection *conn)
+{
+ return conn->proto.in_clientflag;
+}
+
/*
* input.c
*/
diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c
index e610b106c913..1571dfb95aa3 100644
--- a/net/rxrpc/call_event.c
+++ b/net/rxrpc/call_event.c
@@ -842,7 +842,7 @@ void rxrpc_process_call(struct work_struct *work)
msg.msg_controllen = 0;
msg.msg_flags = 0;
- whdr.epoch = htonl(call->conn->epoch);
+ whdr.epoch = htonl(call->conn->proto.epoch);
whdr.cid = htonl(call->cid);
whdr.callNumber = htonl(call->call_id);
whdr.seq = 0;
@@ -1264,7 +1264,7 @@ maybe_reschedule:
if (call->state >= RXRPC_CALL_COMPLETE &&
!list_empty(&call->accept_link)) {
_debug("X unlinking once-pending call %p { e=%lx f=%lx c=%x }",
- call, call->events, call->flags, call->conn->cid);
+ call, call->events, call->flags, call->conn->proto.cid);
read_lock_bh(&call->state_lock);
if (!test_bit(RXRPC_CALL_RELEASED, &call->flags) &&
@@ -1282,7 +1282,7 @@ error:
* this means there's a race between clearing the flag and setting the
* work pending bit and the work item being processed again */
if (call->events && !work_pending(&call->processor)) {
- _debug("jumpstart %x", call->conn->cid);
+ _debug("jumpstart %x", call->conn->proto.cid);
rxrpc_queue_call(call);
}
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index 8b4d47b3ccac..b7c6011c71bb 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -71,7 +71,7 @@ static unsigned long rxrpc_call_hashfunc(
u32 call_id,
u32 epoch,
u16 service_id,
- sa_family_t proto,
+ sa_family_t family,
void *localptr,
unsigned int addr_size,
const u8 *peer_addr)
@@ -92,7 +92,7 @@ static unsigned long rxrpc_call_hashfunc(
key += (cid & RXRPC_CIDMASK) >> RXRPC_CIDSHIFT;
key += cid & RXRPC_CHANNELMASK;
key += in_clientflag;
- key += proto;
+ key += family;
/* Step through the peer address in 16-bit portions for speed */
for (i = 0, p = (const u16 *)peer_addr; i < addr_size >> 1; i++, p++)
key += *p;
@@ -109,7 +109,7 @@ static void rxrpc_call_hash_add(struct rxrpc_call *call)
unsigned int addr_size = 0;
_enter("");
- switch (call->proto) {
+ switch (call->family) {
case AF_INET:
addr_size = sizeof(call->peer_ip.ipv4_addr);
break;
@@ -121,7 +121,7 @@ static void rxrpc_call_hash_add(struct rxrpc_call *call)
}
key = rxrpc_call_hashfunc(call->in_clientflag, call->cid,
call->call_id, call->epoch,
- call->service_id, call->proto,
+ call->service_id, call->family,
call->conn->trans->local, addr_size,
call->peer_ip.ipv6_addr);
/* Store the full key in the call */
@@ -151,7 +151,7 @@ static void rxrpc_call_hash_del(struct rxrpc_call *call)
struct rxrpc_call *rxrpc_find_call_hash(
struct rxrpc_host_header *hdr,
void *localptr,
- sa_family_t proto,
+ sa_family_t family,
const void *peer_addr)
{
unsigned long key;
@@ -161,7 +161,7 @@ struct rxrpc_call *rxrpc_find_call_hash(
u8 in_clientflag = hdr->flags & RXRPC_CLIENT_INITIATED;
_enter("");
- switch (proto) {
+ switch (family) {
case AF_INET:
addr_size = sizeof(call->peer_ip.ipv4_addr);
break;
@@ -174,7 +174,7 @@ struct rxrpc_call *rxrpc_find_call_hash(
key = rxrpc_call_hashfunc(in_clientflag, hdr->cid, hdr->callNumber,
hdr->epoch, hdr->serviceId,
- proto, localptr, addr_size,
+ family, localptr, addr_size,
peer_addr);
hash_for_each_possible_rcu(rxrpc_call_hash, call, hash_node, key) {
if (call->hash_key == key &&
@@ -182,7 +182,7 @@ struct rxrpc_call *rxrpc_find_call_hash(
call->cid == hdr->cid &&
call->in_clientflag == in_clientflag &&
call->service_id == hdr->serviceId &&
- call->proto == proto &&
+ call->family == family &&
call->local == localptr &&
memcmp(call->peer_ip.ipv6_addr, peer_addr,
addr_size) == 0 &&
@@ -286,6 +286,7 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp)
*/
static struct rxrpc_call *rxrpc_alloc_client_call(
struct rxrpc_sock *rx,
+ struct rxrpc_conn_parameters *cp,
struct rxrpc_transport *trans,
struct rxrpc_conn_bundle *bundle,
gfp_t gfp)
@@ -307,16 +308,16 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
call->socket = rx;
call->rx_data_post = 1;
- ret = rxrpc_connect_call(rx, trans, bundle, call, gfp);
+ ret = rxrpc_connect_call(rx, cp, trans, bundle, call, gfp);
if (ret < 0) {
kmem_cache_free(rxrpc_call_jar, call);
return ERR_PTR(ret);
}
/* Record copies of information for hashtable lookup */
- call->proto = rx->proto;
- call->local = trans->local;
- switch (call->proto) {
+ call->family = rx->family;
+ call->local = call->conn->params.local;
+ switch (call->family) {
case AF_INET:
call->peer_ip.ipv4_addr =
trans->peer->srx.transport.sin.sin_addr.s_addr;
@@ -327,9 +328,9 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
sizeof(call->peer_ip.ipv6_addr));
break;
}
- call->epoch = call->conn->epoch;
- call->service_id = call->conn->service_id;
- call->in_clientflag = call->conn->in_clientflag;
+ call->epoch = call->conn->proto.epoch;
+ call->service_id = call->conn->params.service_id;
+ call->in_clientflag = call->conn->proto.in_clientflag;
/* Add the new call to the hashtable */
rxrpc_call_hash_add(call);
@@ -349,6 +350,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
* - called in process context with IRQs enabled
*/
struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
+ struct rxrpc_conn_parameters *cp,
struct rxrpc_transport *trans,
struct rxrpc_conn_bundle *bundle,
unsigned long user_call_ID,
@@ -361,7 +363,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
rx, trans->debug_id, bundle ? bundle->debug_id : -1,
user_call_ID);
- call = rxrpc_alloc_client_call(rx, trans, bundle, gfp);
+ call = rxrpc_alloc_client_call(rx, cp, trans, bundle, gfp);
if (IS_ERR(call)) {
_leave(" = %ld", PTR_ERR(call));
return call;
@@ -524,9 +526,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
write_unlock_bh(&rxrpc_call_lock);
/* Record copies of information for hashtable lookup */
- call->proto = rx->proto;
+ call->family = rx->family;
call->local = conn->trans->local;
- switch (call->proto) {
+ switch (call->family) {
case AF_INET:
call->peer_ip.ipv4_addr =
conn->trans->peer->srx.transport.sin.sin_addr.s_addr;
@@ -539,9 +541,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
default:
break;
}
- call->epoch = conn->epoch;
- call->service_id = conn->service_id;
- call->in_clientflag = conn->in_clientflag;
+ call->epoch = conn->proto.epoch;
+ call->service_id = conn->params.service_id;
+ call->in_clientflag = conn->proto.in_clientflag;
/* Add the new call to the hashtable */
rxrpc_call_hash_add(call);
diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c
index 00c92b614485..51e280c662e0 100644
--- a/net/rxrpc/conn_event.c
+++ b/net/rxrpc/conn_event.c
@@ -94,8 +94,8 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn,
msg.msg_controllen = 0;
msg.msg_flags = 0;
- whdr.epoch = htonl(conn->epoch);
- whdr.cid = htonl(conn->cid);
+ whdr.epoch = htonl(conn->proto.epoch);
+ whdr.cid = htonl(conn->proto.cid);
whdr.callNumber = 0;
whdr.seq = 0;
whdr.type = RXRPC_PACKET_TYPE_ABORT;
@@ -103,7 +103,7 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn,
whdr.userStatus = 0;
whdr.securityIndex = conn->security_ix;
whdr._rsvd = 0;
- whdr.serviceId = htons(conn->service_id);
+ whdr.serviceId = htons(conn->params.service_id);
word = htonl(conn->local_abort);
@@ -220,7 +220,7 @@ static void rxrpc_secure_connection(struct rxrpc_connection *conn)
ASSERT(conn->security_ix != 0);
- if (!conn->key) {
+ if (!conn->params.key) {
_debug("set up security");
ret = rxrpc_init_server_conn_security(conn);
switch (ret) {
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index 8ecde4b77b55..c6787b6f459f 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -220,7 +220,7 @@ static void rxrpc_assign_connection_id(struct rxrpc_connection *conn)
_enter("");
- epoch = conn->epoch;
+ epoch = conn->proto.epoch;
write_lock_bh(&conn->trans->conn_lock);
@@ -237,13 +237,13 @@ attempt_insertion:
parent = *p;
xconn = rb_entry(parent, struct rxrpc_connection, node);
- if (epoch < xconn->epoch)
+ if (epoch < xconn->proto.epoch)
p = &(*p)->rb_left;
- else if (epoch > xconn->epoch)
+ else if (epoch > xconn->proto.epoch)
p = &(*p)->rb_right;
- else if (cid < xconn->cid)
+ else if (cid < xconn->proto.cid)
p = &(*p)->rb_left;
- else if (cid > xconn->cid)
+ else if (cid > xconn->proto.cid)
p = &(*p)->rb_right;
else
goto id_exists;
@@ -254,7 +254,7 @@ attempt_insertion:
rb_link_node(&conn->node, parent, p);
rb_insert_color(&conn->node, &conn->trans->client_conns);
- conn->cid = cid;
+ conn->proto.cid = cid;
write_unlock_bh(&conn->trans->conn_lock);
_leave(" [CID %x]", cid);
return;
@@ -275,8 +275,8 @@ id_exists:
goto attempt_insertion;
xconn = rb_entry(parent, struct rxrpc_connection, node);
- if (epoch < xconn->epoch ||
- cid < xconn->cid)
+ if (epoch < xconn->proto.epoch ||
+ cid < xconn->proto.cid)
goto attempt_insertion;
}
}
@@ -318,8 +318,8 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn,
* connect a call on an exclusive connection
*/
static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
+ struct rxrpc_conn_parameters *cp,
struct rxrpc_transport *trans,
- u16 service_id,
struct rxrpc_call *call,
gfp_t gfp)
{
@@ -340,19 +340,21 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
conn->trans = trans;
conn->bundle = NULL;
- conn->service_id = service_id;
- conn->epoch = rxrpc_epoch;
- conn->in_clientflag = 0;
+ conn->params = *cp;
+ conn->proto.local = cp->local;
+ conn->proto.epoch = rxrpc_epoch;
+ conn->proto.cid = 0;
+ conn->proto.in_clientflag = 0;
+ conn->proto.family = cp->peer->srx.transport.family;
conn->out_clientflag = RXRPC_CLIENT_INITIATED;
- conn->cid = 0;
conn->state = RXRPC_CONN_CLIENT;
conn->avail_calls = RXRPC_MAXCALLS - 1;
- conn->security_level = rx->min_sec_level;
- conn->key = key_get(rx->key);
+
+ key_get(conn->params.key);
ret = rxrpc_init_client_conn_security(conn);
if (ret < 0) {
- key_put(conn->key);
+ key_put(conn->params.key);
kfree(conn);
_leave(" = %d [key]", ret);
return ret;
@@ -389,7 +391,7 @@ found_channel:
conn->channels[chan] = call;
call->conn = conn;
call->channel = chan;
- call->cid = conn->cid | chan;
+ call->cid = conn->proto.cid | chan;
call->call_id = ++conn->call_counter;
_net("CONNECT client on conn %d chan %d as call %x",
@@ -412,6 +414,7 @@ no_free_channels:
* - called in process context with IRQs enabled
*/
int rxrpc_connect_call(struct rxrpc_sock *rx,
+ struct rxrpc_conn_parameters *cp,
struct rxrpc_transport *trans,
struct rxrpc_conn_bundle *bundle,
struct rxrpc_call *call,
@@ -425,8 +428,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
_enter("%p,%lx,", rx, call->user_call_ID);
if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags))
- return rxrpc_connect_exclusive(rx, trans, bundle->service_id,
- call, gfp);
+ return rxrpc_connect_exclusive(rx, cp, trans, call, gfp);
spin_lock(&trans->client_lock);
for (;;) {
@@ -517,19 +519,21 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
candidate->trans = trans;
candidate->bundle = bundle;
- candidate->service_id = bundle->service_id;
- candidate->epoch = rxrpc_epoch;
- candidate->in_clientflag = 0;
+ candidate->params = *cp;
+ candidate->proto.local = cp->local;
+ candidate->proto.epoch = rxrpc_epoch;
+ candidate->proto.cid = 0;
+ candidate->proto.in_clientflag = 0;
+ candidate->proto.family = cp->peer->srx.transport.family;
candidate->out_clientflag = RXRPC_CLIENT_INITIATED;
- candidate->cid = 0;
candidate->state = RXRPC_CONN_CLIENT;
candidate->avail_calls = RXRPC_MAXCALLS;
- candidate->security_level = rx->min_sec_level;
- candidate->key = key_get(bundle->key);
+
+ key_get(candidate->params.key);
ret = rxrpc_init_client_conn_security(candidate);
if (ret < 0) {
- key_put(candidate->key);
+ key_put(candidate->params.key);
kfree(candidate);
_leave(" = %d [key]", ret);
return ret;
@@ -577,7 +581,7 @@ found_channel:
conn->channels[chan] = call;
call->conn = conn;
call->channel = chan;
- call->cid = conn->cid | chan;
+ call->cid = conn->proto.cid | chan;
call->call_id = ++conn->call_counter;
_net("CONNECT client on conn %d chan %d as call %x",
@@ -626,15 +630,15 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
while (p) {
conn = rb_entry(p, struct rxrpc_connection, node);
- _debug("maybe %x", conn->cid);
+ _debug("maybe %x", conn->proto.cid);
- if (epoch < conn->epoch)
+ if (epoch < conn->proto.epoch)
p = p->rb_left;
- else if (epoch > conn->epoch)
+ else if (epoch > conn->proto.epoch)
p = p->rb_right;
- else if (cid < conn->cid)
+ else if (cid < conn->proto.cid)
p = p->rb_left;
- else if (cid > conn->cid)
+ else if (cid > conn->proto.cid)
p = p->rb_right;
else
goto found_extant_connection;
@@ -650,14 +654,17 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
}
candidate->trans = trans;
- candidate->epoch = hdr->epoch;
- candidate->cid = hdr->cid & RXRPC_CIDMASK;
- candidate->service_id = hdr->serviceId;
+ candidate->proto.local = trans->local;
+ candidate->proto.epoch = hdr->epoch;
+ candidate->proto.cid = hdr->cid & RXRPC_CIDMASK;
+ candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED;
+ candidate->params.local = trans->local;
+ candidate->params.peer = trans->peer;
+ candidate->params.service_id = hdr->serviceId;
candidate->security_ix = hdr->securityIndex;
- candidate->in_clientflag = RXRPC_CLIENT_INITIATED;
candidate->out_clientflag = 0;
candidate->state = RXRPC_CONN_SERVER;
- if (candidate->service_id)
+ if (candidate->params.service_id)
candidate->state = RXRPC_CONN_SERVER_UNSECURED;
write_lock_bh(&trans->conn_lock);
@@ -668,13 +675,13 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
p = *pp;
conn = rb_entry(p, struct rxrpc_connection, node);
- if (epoch < conn->epoch)
+ if (epoch < conn->proto.epoch)
pp = &(*pp)->rb_left;
- else if (epoch > conn->epoch)
+ else if (epoch > conn->proto.epoch)
pp = &(*pp)->rb_right;
- else if (cid < conn->cid)
+ else if (cid < conn->proto.cid)
pp = &(*pp)->rb_left;
- else if (cid > conn->cid)
+ else if (cid > conn->proto.cid)
pp = &(*pp)->rb_right;
else
goto found_extant_second;
@@ -696,7 +703,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
new = "new";
success:
- _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->cid);
+ _net("CONNECTION %s %d {%x}", new, conn->debug_id, conn->proto.cid);
_leave(" = %p {u=%d}", conn, atomic_read(&conn->usage));
return conn;
@@ -754,15 +761,15 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
while (p) {
conn = rb_entry(p, struct rxrpc_connection, node);
- _debug("maybe %x", conn->cid);
+ _debug("maybe %x", conn->proto.cid);
- if (epoch < conn->epoch)
+ if (epoch < conn->proto.epoch)
p = p->rb_left;
- else if (epoch > conn->epoch)
+ else if (epoch > conn->proto.epoch)
p = p->rb_right;
- else if (cid < conn->cid)
+ else if (cid < conn->proto.cid)
p = p->rb_left;
- else if (cid > conn->cid)
+ else if (cid > conn->proto.cid)
p = p->rb_right;
else
goto found;
@@ -816,7 +823,7 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn)
rxrpc_purge_queue(&conn->rx_queue);
conn->security->clear(conn);
- key_put(conn->key);
+ key_put(conn->params.key);
key_put(conn->server_key);
rxrpc_put_transport(conn->trans);
diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index e11e4d785127..c030abd4d2d8 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -360,7 +360,7 @@ void rxrpc_fast_process_packet(struct rxrpc_call *call, struct sk_buff *skb)
case RXRPC_PACKET_TYPE_BUSY:
_proto("Rx BUSY %%%u", sp->hdr.serial);
- if (call->conn->out_clientflag)
+ if (rxrpc_conn_is_service(call->conn))
goto protocol_error;
write_lock_bh(&call->state_lock);
@@ -533,7 +533,7 @@ static void rxrpc_post_packet_to_call(struct rxrpc_call *call,
case RXRPC_CALL_COMPLETE:
case RXRPC_CALL_CLIENT_FINAL_ACK:
/* complete server call */
- if (call->conn->in_clientflag)
+ if (rxrpc_conn_is_service(call->conn))
goto dead_call;
/* resend last packet of a completed call */
_debug("final ack again");
diff --git a/net/rxrpc/key.c b/net/rxrpc/key.c
index 4ad56fafe3a7..18c737a61d80 100644
--- a/net/rxrpc/key.c
+++ b/net/rxrpc/key.c
@@ -987,7 +987,7 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *conn,
if (ret < 0)
goto error;
- conn->key = key;
+ conn->params.key = key;
_leave(" = 0 [%d]", key_serial(key));
return 0;
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index e6fb3863b0bc..8c51745cccea 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -133,6 +133,7 @@ static struct rxrpc_call *
rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
unsigned long user_call_ID)
{
+ struct rxrpc_conn_parameters cp;
struct rxrpc_conn_bundle *bundle;
struct rxrpc_transport *trans;
struct rxrpc_call *call;
@@ -146,23 +147,32 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
if (!msg->msg_name)
return ERR_PTR(-EDESTADDRREQ);
- trans = rxrpc_name_to_transport(rx, msg->msg_name, msg->msg_namelen, 0,
+ key = rx->key;
+ if (key && !rx->key->payload.data[0])
+ key = NULL;
+
+ memset(&cp, 0, sizeof(cp));
+ cp.local = rx->local;
+ cp.key = rx->key;
+ cp.security_level = rx->min_sec_level;
+ cp.exclusive = test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags);
+ cp.service_id = srx->srx_service;
+ trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen,
GFP_KERNEL);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto out;
}
+ cp.peer = trans->peer;
- key = rx->key;
- if (key && !rx->key->payload.data[0])
- key = NULL;
- bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, GFP_KERNEL);
+ bundle = rxrpc_get_bundle(rx, trans, cp.key, srx->srx_service,
+ GFP_KERNEL);
if (IS_ERR(bundle)) {
ret = PTR_ERR(bundle);
goto out_trans;
}
- call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID,
+ call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID,
GFP_KERNEL);
rxrpc_put_bundle(trans, bundle);
rxrpc_put_transport(trans);
@@ -664,7 +674,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
seq = atomic_inc_return(&call->sequence);
- sp->hdr.epoch = conn->epoch;
+ sp->hdr.epoch = conn->proto.epoch;
sp->hdr.cid = call->cid;
sp->hdr.callNumber = call->call_id;
sp->hdr.seq = seq;
diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c
index 225163bc658d..bbee05850801 100644
--- a/net/rxrpc/proc.c
+++ b/net/rxrpc/proc.c
@@ -74,10 +74,10 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
" %-8.8s %08x %lx\n",
lbuff,
rbuff,
- call->conn->service_id,
+ call->conn->params.service_id,
call->cid,
call->call_id,
- call->conn->in_clientflag ? "Svc" : "Clt",
+ rxrpc_conn_is_service(call->conn) ? "Svc" : "Clt",
atomic_read(&call->usage),
rxrpc_call_states[call->state],
call->remote_abort ?: call->local_abort,
@@ -157,13 +157,13 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v)
" %s %08x %08x %08x\n",
lbuff,
rbuff,
- conn->service_id,
- conn->cid,
+ conn->params.service_id,
+ conn->proto.cid,
conn->call_counter,
- conn->in_clientflag ? "Svc" : "Clt",
+ rxrpc_conn_is_service(conn) ? "Svc" : "Clt",
atomic_read(&conn->usage),
rxrpc_conn_states[conn->state],
- key_serial(conn->key),
+ key_serial(conn->params.key),
atomic_read(&conn->serial),
atomic_read(&conn->hi_serial));
diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c
index 59706b9f2f7a..c5bac4e0db71 100644
--- a/net/rxrpc/recvmsg.c
+++ b/net/rxrpc/recvmsg.c
@@ -205,7 +205,7 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
/* we transferred the whole data packet */
if (sp->hdr.flags & RXRPC_LAST_PACKET) {
_debug("last");
- if (call->conn->out_clientflag) {
+ if (rxrpc_conn_is_client(call->conn)) {
/* last byte of reply received */
ret = copied;
goto terminal_message;
diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index 36a634027d9d..134c2713ae23 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -58,9 +58,9 @@ static int rxkad_init_connection_security(struct rxrpc_connection *conn)
struct rxrpc_key_token *token;
int ret;
- _enter("{%d},{%x}", conn->debug_id, key_serial(conn->key));
+ _enter("{%d},{%x}", conn->debug_id, key_serial(conn->params.key));
- token = conn->key->payload.data[0];
+ token = conn->params.key->payload.data[0];
conn->security_ix = token->security_index;
ci = crypto_alloc_skcipher("pcbc(fcrypt)", 0, CRYPTO_ALG_ASYNC);
@@ -74,7 +74,7 @@ static int rxkad_init_connection_security(struct rxrpc_connection *conn)
sizeof(token->kad->session_key)) < 0)
BUG();
- switch (conn->security_level) {
+ switch (conn->params.security_level) {
case RXRPC_SECURITY_PLAIN:
break;
case RXRPC_SECURITY_AUTH:
@@ -115,14 +115,14 @@ static void rxkad_prime_packet_security(struct rxrpc_connection *conn)
_enter("");
- if (!conn->key)
+ if (!conn->params.key)
return;
- token = conn->key->payload.data[0];
+ token = conn->params.key->payload.data[0];
memcpy(&iv, token->kad->session_key, sizeof(iv));
- tmpbuf.x[0] = htonl(conn->epoch);
- tmpbuf.x[1] = htonl(conn->cid);
+ tmpbuf.x[0] = htonl(conn->proto.epoch);
+ tmpbuf.x[1] = htonl(conn->proto.cid);
tmpbuf.x[2] = 0;
tmpbuf.x[3] = htonl(conn->security_ix);
@@ -220,7 +220,7 @@ static int rxkad_secure_packet_encrypt(const struct rxrpc_call *call,
rxkhdr.checksum = 0;
/* encrypt from the session key */
- token = call->conn->key->payload.data[0];
+ token = call->conn->params.key->payload.data[0];
memcpy(&iv, token->kad->session_key, sizeof(iv));
sg_init_one(&sg[0], sechdr, sizeof(rxkhdr));
@@ -277,13 +277,13 @@ static int rxkad_secure_packet(const struct rxrpc_call *call,
sp = rxrpc_skb(skb);
_enter("{%d{%x}},{#%u},%zu,",
- call->debug_id, key_serial(call->conn->key), sp->hdr.seq,
- data_size);
+ call->debug_id, key_serial(call->conn->params.key),
+ sp->hdr.seq, data_size);
if (!call->conn->cipher)
return 0;
- ret = key_validate(call->conn->key);
+ ret = key_validate(call->conn->params.key);
if (ret < 0)
return ret;
@@ -312,7 +312,7 @@ static int rxkad_secure_packet(const struct rxrpc_call *call,
y = 1; /* zero checksums are not permitted */
sp->hdr.cksum = y;
- switch (call->conn->security_level) {
+ switch (call->conn->params.security_level) {
case RXRPC_SECURITY_PLAIN:
ret = 0;
break;
@@ -446,7 +446,7 @@ static int rxkad_verify_packet_encrypt(const struct rxrpc_call *call,
skb_to_sgvec(skb, sg, 0, skb->len);
/* decrypt from the session key */
- token = call->conn->key->payload.data[0];
+ token = call->conn->params.key->payload.data[0];
memcpy(&iv, token->kad->session_key, sizeof(iv));
skcipher_request_set_tfm(req, call->conn->cipher);
@@ -516,7 +516,7 @@ static int rxkad_verify_packet(const struct rxrpc_call *call,
sp = rxrpc_skb(skb);
_enter("{%d{%x}},{#%u}",
- call->debug_id, key_serial(call->conn->key), sp->hdr.seq);
+ call->debug_id, key_serial(call->conn->params.key), sp->hdr.seq);
if (!call->conn->cipher)
return 0;
@@ -557,7 +557,7 @@ static int rxkad_verify_packet(const struct rxrpc_call *call,
return -EPROTO;
}
- switch (call->conn->security_level) {
+ switch (call->conn->params.security_level) {
case RXRPC_SECURITY_PLAIN:
ret = 0;
break;
@@ -589,9 +589,9 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
u32 serial;
int ret;
- _enter("{%d,%x}", conn->debug_id, key_serial(conn->key));
+ _enter("{%d,%x}", conn->debug_id, key_serial(conn->params.key));
- ret = key_validate(conn->key);
+ ret = key_validate(conn->params.key);
if (ret < 0)
return ret;
@@ -608,8 +608,8 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
msg.msg_controllen = 0;
msg.msg_flags = 0;
- whdr.epoch = htonl(conn->epoch);
- whdr.cid = htonl(conn->cid);
+ whdr.epoch = htonl(conn->proto.epoch);
+ whdr.cid = htonl(conn->proto.cid);
whdr.callNumber = 0;
whdr.seq = 0;
whdr.type = RXRPC_PACKET_TYPE_CHALLENGE;
@@ -617,7 +617,7 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
whdr.userStatus = 0;
whdr.securityIndex = conn->security_ix;
whdr._rsvd = 0;
- whdr.serviceId = htons(conn->service_id);
+ whdr.serviceId = htons(conn->params.service_id);
iov[0].iov_base = &whdr;
iov[0].iov_len = sizeof(whdr);
@@ -771,14 +771,14 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn,
u32 version, nonce, min_level, abort_code;
int ret;
- _enter("{%d,%x}", conn->debug_id, key_serial(conn->key));
+ _enter("{%d,%x}", conn->debug_id, key_serial(conn->params.key));
- if (!conn->key) {
+ if (!conn->params.key) {
_leave(" = -EPROTO [no key]");
return -EPROTO;
}
- ret = key_validate(conn->key);
+ ret = key_validate(conn->params.key);
if (ret < 0) {
*_abort_code = RXKADEXPIRED;
return ret;
@@ -801,20 +801,20 @@ static int rxkad_respond_to_challenge(struct rxrpc_connection *conn,
goto protocol_error;
abort_code = RXKADLEVELFAIL;
- if (conn->security_level < min_level)
+ if (conn->params.security_level < min_level)
goto protocol_error;
- token = conn->key->payload.data[0];
+ token = conn->params.key->payload.data[0];
/* build the response packet */
memset(&resp, 0, sizeof(resp));
resp.version = htonl(RXKAD_VERSION);
- resp.encrypted.epoch = htonl(conn->epoch);
- resp.encrypted.cid = htonl(conn->cid);
+ resp.encrypted.epoch = htonl(conn->proto.epoch);
+ resp.encrypted.cid = htonl(conn->proto.cid);
resp.encrypted.securityIndex = htonl(conn->security_ix);
resp.encrypted.inc_nonce = htonl(nonce + 1);
- resp.encrypted.level = htonl(conn->security_level);
+ resp.encrypted.level = htonl(conn->params.security_level);
resp.kvno = htonl(token->kad->kvno);
resp.ticket_len = htonl(token->kad->ticket_len);
@@ -1096,9 +1096,9 @@ static int rxkad_verify_response(struct rxrpc_connection *conn,
rxkad_decrypt_response(conn, &response, &session_key);
abort_code = RXKADSEALEDINCON;
- if (ntohl(response.encrypted.epoch) != conn->epoch)
+ if (ntohl(response.encrypted.epoch) != conn->proto.epoch)
goto protocol_error_free;
- if (ntohl(response.encrypted.cid) != conn->cid)
+ if (ntohl(response.encrypted.cid) != conn->proto.cid)
goto protocol_error_free;
if (ntohl(response.encrypted.securityIndex) != conn->security_ix)
goto protocol_error_free;
@@ -1122,7 +1122,7 @@ static int rxkad_verify_response(struct rxrpc_connection *conn,
level = ntohl(response.encrypted.level);
if (level > RXRPC_SECURITY_ENCRYPT)
goto protocol_error_free;
- conn->security_level = level;
+ conn->params.security_level = level;
/* create a key to hold the security data and expiration time - after
* this the connection security can be handled in exactly the same way
diff --git a/net/rxrpc/security.c b/net/rxrpc/security.c
index d223253b22fa..40955d0f2693 100644
--- a/net/rxrpc/security.c
+++ b/net/rxrpc/security.c
@@ -76,7 +76,7 @@ int rxrpc_init_client_conn_security(struct rxrpc_connection *conn)
{
const struct rxrpc_security *sec;
struct rxrpc_key_token *token;
- struct key *key = conn->key;
+ struct key *key = conn->params.key;
int ret;
_enter("{%d},{%x}", conn->debug_id, key_serial(key));
@@ -121,7 +121,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn)
_enter("");
- sprintf(kdesc, "%u:%u", conn->service_id, conn->security_ix);
+ sprintf(kdesc, "%u:%u", conn->params.service_id, conn->security_ix);
sec = rxrpc_security_lookup(conn->security_ix);
if (!sec) {
@@ -132,7 +132,7 @@ int rxrpc_init_server_conn_security(struct rxrpc_connection *conn)
/* find the service */
read_lock_bh(&local->services_lock);
list_for_each_entry(rx, &local->services, listen_link) {
- if (rx->srx.srx_service == conn->service_id)
+ if (rx->srx.srx_service == conn->params.service_id)
goto found_service;
}
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 04/14] rxrpc: Replace conn->trans->{local, peer} with conn->params.{local, peer}
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (2 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 03/14] rxrpc: Use structs to hold connection params and protocol info David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 05/14] rxrpc: Fix exclusive connection handling David Howells
` (10 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Replace accesses of conn->trans->{local,peer} with
conn->params.{local,peer} thus making it easier for a future commit to
remove the rxrpc_transport struct.
This also reduces the number of memory accesses involved.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/call_event.c | 12 ++++++------
net/rxrpc/call_object.c | 28 ++++++++++++++--------------
net/rxrpc/conn_event.c | 6 +++---
net/rxrpc/input.c | 2 +-
net/rxrpc/output.c | 2 +-
net/rxrpc/proc.c | 22 ++++++++++------------
net/rxrpc/recvmsg.c | 4 ++--
net/rxrpc/rxkad.c | 12 ++++++------
net/rxrpc/security.c | 2 +-
9 files changed, 44 insertions(+), 46 deletions(-)
diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c
index 1571dfb95aa3..b43faf573ed3 100644
--- a/net/rxrpc/call_event.c
+++ b/net/rxrpc/call_event.c
@@ -545,7 +545,7 @@ static void rxrpc_extract_ackinfo(struct rxrpc_call *call, struct sk_buff *skb,
mtu = min(ntohl(ackinfo.rxMTU), ntohl(ackinfo.maxMTU));
- peer = call->conn->trans->peer;
+ peer = call->conn->params.peer;
if (mtu < peer->maxdata) {
spin_lock_bh(&peer->lock);
peer->maxdata = mtu;
@@ -836,8 +836,8 @@ void rxrpc_process_call(struct work_struct *work)
/* there's a good chance we're going to have to send a message, so set
* one up in advance */
- msg.msg_name = &call->conn->trans->peer->srx.transport;
- msg.msg_namelen = call->conn->trans->peer->srx.transport_len;
+ msg.msg_name = &call->conn->params.peer->srx.transport;
+ msg.msg_namelen = call->conn->params.peer->srx.transport_len;
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_flags = 0;
@@ -1151,8 +1151,8 @@ send_ACK_with_skew:
ack.maxSkew = htons(atomic_read(&call->conn->hi_serial) -
ntohl(ack.serial));
send_ACK:
- mtu = call->conn->trans->peer->if_mtu;
- mtu -= call->conn->trans->peer->hdrsize;
+ mtu = call->conn->params.peer->if_mtu;
+ mtu -= call->conn->params.peer->hdrsize;
ackinfo.maxMTU = htonl(mtu);
ackinfo.rwind = htonl(rxrpc_rx_window_size);
@@ -1206,7 +1206,7 @@ send_message_2:
len += iov[1].iov_len;
}
- ret = kernel_sendmsg(call->conn->trans->local->socket,
+ ret = kernel_sendmsg(call->conn->params.local->socket,
&msg, iov, ioc, len);
if (ret < 0) {
_debug("sendmsg failed: %d", ret);
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index b7c6011c71bb..5c2dceaf6a9c 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -122,7 +122,7 @@ static void rxrpc_call_hash_add(struct rxrpc_call *call)
key = rxrpc_call_hashfunc(call->in_clientflag, call->cid,
call->call_id, call->epoch,
call->service_id, call->family,
- call->conn->trans->local, addr_size,
+ call->conn->params.local, addr_size,
call->peer_ip.ipv6_addr);
/* Store the full key in the call */
call->hash_key = key;
@@ -320,11 +320,11 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
switch (call->family) {
case AF_INET:
call->peer_ip.ipv4_addr =
- trans->peer->srx.transport.sin.sin_addr.s_addr;
+ call->conn->params.peer->srx.transport.sin.sin_addr.s_addr;
break;
case AF_INET6:
memcpy(call->peer_ip.ipv6_addr,
- trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8,
+ call->conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8,
sizeof(call->peer_ip.ipv6_addr));
break;
}
@@ -334,9 +334,9 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
/* Add the new call to the hashtable */
rxrpc_call_hash_add(call);
- spin_lock(&call->conn->trans->peer->lock);
- hlist_add_head(&call->error_link, &call->conn->trans->peer->error_targets);
- spin_unlock(&call->conn->trans->peer->lock);
+ spin_lock(&call->conn->params.peer->lock);
+ hlist_add_head(&call->error_link, &call->conn->params.peer->error_targets);
+ spin_unlock(&call->conn->params.peer->lock);
call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime;
add_timer(&call->lifetimer);
@@ -517,9 +517,9 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
atomic_inc(&conn->usage);
write_unlock_bh(&conn->lock);
- spin_lock(&conn->trans->peer->lock);
- hlist_add_head(&call->error_link, &conn->trans->peer->error_targets);
- spin_unlock(&conn->trans->peer->lock);
+ spin_lock(&conn->params.peer->lock);
+ hlist_add_head(&call->error_link, &conn->params.peer->error_targets);
+ spin_unlock(&conn->params.peer->lock);
write_lock_bh(&rxrpc_call_lock);
list_add_tail(&call->link, &rxrpc_calls);
@@ -527,15 +527,15 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
/* Record copies of information for hashtable lookup */
call->family = rx->family;
- call->local = conn->trans->local;
+ call->local = conn->params.local;
switch (call->family) {
case AF_INET:
call->peer_ip.ipv4_addr =
- conn->trans->peer->srx.transport.sin.sin_addr.s_addr;
+ conn->params.peer->srx.transport.sin.sin_addr.s_addr;
break;
case AF_INET6:
memcpy(call->peer_ip.ipv6_addr,
- conn->trans->peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8,
+ conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8,
sizeof(call->peer_ip.ipv6_addr));
break;
default:
@@ -813,9 +813,9 @@ static void rxrpc_cleanup_call(struct rxrpc_call *call)
}
if (call->conn) {
- spin_lock(&call->conn->trans->peer->lock);
+ spin_lock(&call->conn->params.peer->lock);
hlist_del_init(&call->error_link);
- spin_unlock(&call->conn->trans->peer->lock);
+ spin_unlock(&call->conn->params.peer->lock);
write_lock_bh(&call->conn->lock);
rb_erase(&call->conn_node, &call->conn->calls);
diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c
index 51e280c662e0..a022439f6f5a 100644
--- a/net/rxrpc/conn_event.c
+++ b/net/rxrpc/conn_event.c
@@ -88,8 +88,8 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn,
rxrpc_abort_calls(conn, RXRPC_CALL_LOCALLY_ABORTED, abort_code);
- msg.msg_name = &conn->trans->peer->srx.transport;
- msg.msg_namelen = conn->trans->peer->srx.transport_len;
+ msg.msg_name = &conn->params.peer->srx.transport;
+ msg.msg_namelen = conn->params.peer->srx.transport_len;
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_flags = 0;
@@ -118,7 +118,7 @@ static int rxrpc_abort_connection(struct rxrpc_connection *conn,
whdr.serial = htonl(serial);
_proto("Tx CONN ABORT %%%u { %d }", serial, conn->local_abort);
- ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len);
+ ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len);
if (ret < 0) {
_debug("sendmsg failed: %d", ret);
return -EAGAIN;
diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index c030abd4d2d8..6af7f40c5030 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -560,7 +560,7 @@ static void rxrpc_post_packet_to_call(struct rxrpc_call *call,
dead_call:
if (sp->hdr.type != RXRPC_PACKET_TYPE_ABORT) {
skb->priority = RX_CALL_DEAD;
- rxrpc_reject_packet(call->conn->trans->local, skb);
+ rxrpc_reject_packet(call->conn->params.local, skb);
goto unlock;
}
free_unlock:
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index 8c51745cccea..becbaa7c0a7c 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -583,7 +583,7 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
goto maybe_error;
}
- max = call->conn->trans->peer->maxdata;
+ max = call->conn->params.peer->maxdata;
max -= call->conn->security_size;
max &= ~(call->conn->size_align - 1UL);
diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c
index bbee05850801..9863270691d7 100644
--- a/net/rxrpc/proc.c
+++ b/net/rxrpc/proc.c
@@ -46,7 +46,7 @@ static void rxrpc_call_seq_stop(struct seq_file *seq, void *v)
static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
{
- struct rxrpc_transport *trans;
+ struct rxrpc_connection *conn;
struct rxrpc_call *call;
char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1];
@@ -59,15 +59,15 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
}
call = list_entry(v, struct rxrpc_call, link);
- trans = call->conn->trans;
+ conn = call->conn;
sprintf(lbuff, "%pI4:%u",
- &trans->local->srx.transport.sin.sin_addr,
- ntohs(trans->local->srx.transport.sin.sin_port));
+ &conn->params.local->srx.transport.sin.sin_addr,
+ ntohs(conn->params.local->srx.transport.sin.sin_port));
sprintf(rbuff, "%pI4:%u",
- &trans->peer->srx.transport.sin.sin_addr,
- ntohs(trans->peer->srx.transport.sin.sin_port));
+ &conn->params.peer->srx.transport.sin.sin_addr,
+ ntohs(conn->params.peer->srx.transport.sin.sin_port));
seq_printf(seq,
"UDP %-22.22s %-22.22s %4x %08x %08x %s %3u"
@@ -129,7 +129,6 @@ static void rxrpc_connection_seq_stop(struct seq_file *seq, void *v)
static int rxrpc_connection_seq_show(struct seq_file *seq, void *v)
{
struct rxrpc_connection *conn;
- struct rxrpc_transport *trans;
char lbuff[4 + 4 + 4 + 4 + 5 + 1], rbuff[4 + 4 + 4 + 4 + 5 + 1];
if (v == &rxrpc_connections) {
@@ -142,15 +141,14 @@ static int rxrpc_connection_seq_show(struct seq_file *seq, void *v)
}
conn = list_entry(v, struct rxrpc_connection, link);
- trans = conn->trans;
sprintf(lbuff, "%pI4:%u",
- &trans->local->srx.transport.sin.sin_addr,
- ntohs(trans->local->srx.transport.sin.sin_port));
+ &conn->params.local->srx.transport.sin.sin_addr,
+ ntohs(conn->params.local->srx.transport.sin.sin_port));
sprintf(rbuff, "%pI4:%u",
- &trans->peer->srx.transport.sin.sin_addr,
- ntohs(trans->peer->srx.transport.sin.sin_port));
+ &conn->params.peer->srx.transport.sin.sin_addr,
+ ntohs(conn->params.peer->srx.transport.sin.sin_port));
seq_printf(seq,
"UDP %-22.22s %-22.22s %4x %08x %08x %s %3u"
diff --git a/net/rxrpc/recvmsg.c b/net/rxrpc/recvmsg.c
index c5bac4e0db71..a3fa2ed85d63 100644
--- a/net/rxrpc/recvmsg.c
+++ b/net/rxrpc/recvmsg.c
@@ -147,9 +147,9 @@ int rxrpc_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
if (!continue_call) {
if (msg->msg_name) {
size_t len =
- sizeof(call->conn->trans->peer->srx);
+ sizeof(call->conn->params.peer->srx);
memcpy(msg->msg_name,
- &call->conn->trans->peer->srx, len);
+ &call->conn->params.peer->srx, len);
msg->msg_namelen = len;
}
sock_recv_timestamp(msg, &rx->sk, skb);
diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index 134c2713ae23..23c05ec6fa28 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -602,8 +602,8 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
challenge.min_level = htonl(0);
challenge.__padding = 0;
- msg.msg_name = &conn->trans->peer->srx.transport.sin;
- msg.msg_namelen = sizeof(conn->trans->peer->srx.transport.sin);
+ msg.msg_name = &conn->params.peer->srx.transport.sin;
+ msg.msg_namelen = sizeof(conn->params.peer->srx.transport.sin);
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_flags = 0;
@@ -630,7 +630,7 @@ static int rxkad_issue_challenge(struct rxrpc_connection *conn)
whdr.serial = htonl(serial);
_proto("Tx CHALLENGE %%%u", serial);
- ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 2, len);
+ ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 2, len);
if (ret < 0) {
_debug("sendmsg failed: %d", ret);
return -EAGAIN;
@@ -657,8 +657,8 @@ static int rxkad_send_response(struct rxrpc_connection *conn,
_enter("");
- msg.msg_name = &conn->trans->peer->srx.transport.sin;
- msg.msg_namelen = sizeof(conn->trans->peer->srx.transport.sin);
+ msg.msg_name = &conn->params.peer->srx.transport.sin;
+ msg.msg_namelen = sizeof(conn->params.peer->srx.transport.sin);
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_flags = 0;
@@ -684,7 +684,7 @@ static int rxkad_send_response(struct rxrpc_connection *conn,
whdr.serial = htonl(serial);
_proto("Tx RESPONSE %%%u", serial);
- ret = kernel_sendmsg(conn->trans->local->socket, &msg, iov, 3, len);
+ ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 3, len);
if (ret < 0) {
_debug("sendmsg failed: %d", ret);
return -EAGAIN;
diff --git a/net/rxrpc/security.c b/net/rxrpc/security.c
index 40955d0f2693..814d285ff802 100644
--- a/net/rxrpc/security.c
+++ b/net/rxrpc/security.c
@@ -113,7 +113,7 @@ int rxrpc_init_client_conn_security(struct rxrpc_connection *conn)
int rxrpc_init_server_conn_security(struct rxrpc_connection *conn)
{
const struct rxrpc_security *sec;
- struct rxrpc_local *local = conn->trans->local;
+ struct rxrpc_local *local = conn->params.local;
struct rxrpc_sock *rx;
struct key *key;
key_ref_t kref;
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 05/14] rxrpc: Fix exclusive connection handling
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (3 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 04/14] rxrpc: Replace conn->trans->{local, peer} with conn->params.{local, peer} David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 06/14] rxrpc: Pass sk_buff * rather than rxrpc_host_header * to functions David Howells
` (9 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
"Exclusive connections" are meant to be used for a single client call and
then scrapped. The idea is to limit the use of the negotiated security
context. The current code, however, isn't doing this: it is instead
restricting the socket to a single virtual connection and doing all the
calls over that.
This is changed such that the socket no longer maintains a special virtual
connection over which it will do all the calls, but rather gets a new one
each time a new exclusive call is made.
Further, using a socket option for this is a poor choice. It should be
done on sendmsg with a control message marker instead so that calls can be
marked exclusive individually. To that end, add RXRPC_EXCLUSIVE_CALL
which, if passed to sendmsg() as a control message element, will cause the
call to be done on an single-use connection.
The socket option (RXRPC_EXCLUSIVE_CONNECTION) still exists and, if set,
will override any lack of RXRPC_EXCLUSIVE_CALL being specified so that
programs using the setsockopt() will appear to work the same.
Signed-off-by: David Howells <dhowells@redhat.com>
---
include/linux/rxrpc.h | 3 +
net/rxrpc/af_rxrpc.c | 7 ---
net/rxrpc/ar-internal.h | 7 ++-
net/rxrpc/conn_object.c | 97 +++++++++++++++++++----------------------------
net/rxrpc/output.c | 19 +++++++--
5 files changed, 60 insertions(+), 73 deletions(-)
diff --git a/include/linux/rxrpc.h b/include/linux/rxrpc.h
index 1e8f216e2cf1..c68307bc306f 100644
--- a/include/linux/rxrpc.h
+++ b/include/linux/rxrpc.h
@@ -35,7 +35,7 @@ struct sockaddr_rxrpc {
*/
#define RXRPC_SECURITY_KEY 1 /* [clnt] set client security key */
#define RXRPC_SECURITY_KEYRING 2 /* [srvr] set ring of server security keys */
-#define RXRPC_EXCLUSIVE_CONNECTION 3 /* [clnt] use exclusive RxRPC connection */
+#define RXRPC_EXCLUSIVE_CONNECTION 3 /* Deprecated; use RXRPC_EXCLUSIVE_CALL instead */
#define RXRPC_MIN_SECURITY_LEVEL 4 /* minimum security level */
/*
@@ -52,6 +52,7 @@ struct sockaddr_rxrpc {
#define RXRPC_LOCAL_ERROR 7 /* -r: local error generated [terminal] */
#define RXRPC_NEW_CALL 8 /* -r: [Service] new incoming call notification */
#define RXRPC_ACCEPT 9 /* s-: [Service] accept request */
+#define RXRPC_EXCLUSIVE_CALL 10 /* s-: Call should be on exclusive connection */
/*
* RxRPC security levels
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 48b45a0280c0..73f5c553eef4 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -494,7 +494,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname,
ret = -EISCONN;
if (rx->sk.sk_state != RXRPC_UNBOUND)
goto error;
- set_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags);
+ rx->exclusive = true;
goto success;
case RXRPC_SECURITY_KEY:
@@ -669,11 +669,6 @@ static int rxrpc_release_sock(struct sock *sk)
flush_workqueue(rxrpc_workqueue);
rxrpc_purge_queue(&sk->sk_receive_queue);
- if (rx->conn) {
- rxrpc_put_connection(rx->conn);
- rx->conn = NULL;
- }
-
if (rx->local) {
rxrpc_put_local(rx->local);
rx->local = NULL;
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index efe6673deb28..4ca99445e0b7 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -37,6 +37,8 @@ struct rxrpc_crypt {
#define rxrpc_queue_call(CALL) rxrpc_queue_work(&(CALL)->processor)
#define rxrpc_queue_conn(CONN) rxrpc_queue_work(&(CONN)->processor)
+struct rxrpc_connection;
+
/*
* sk_state for RxRPC sockets
*/
@@ -57,7 +59,6 @@ struct rxrpc_sock {
struct sock sk;
rxrpc_interceptor_t interceptor; /* kernel service Rx interceptor function */
struct rxrpc_local *local; /* local endpoint */
- struct rxrpc_connection *conn; /* exclusive virtual connection */
struct list_head listen_link; /* link in the local endpoint's listen list */
struct list_head secureq; /* calls awaiting connection security clearance */
struct list_head acceptq; /* calls awaiting acceptance */
@@ -66,13 +67,13 @@ struct rxrpc_sock {
struct rb_root calls; /* outstanding calls on this socket */
unsigned long flags;
#define RXRPC_SOCK_CONNECTED 0 /* connect_srx is set */
-#define RXRPC_SOCK_EXCLUSIVE_CONN 1 /* exclusive connection for a client socket */
rwlock_t call_lock; /* lock for calls */
u32 min_sec_level; /* minimum security level */
#define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT
+ bool exclusive; /* Exclusive connection for a client socket */
+ sa_family_t family; /* Protocol family created with */
struct sockaddr_rxrpc srx; /* local address */
struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */
- sa_family_t family; /* protocol family created with */
};
#define rxrpc_sk(__sk) container_of((__sk), struct rxrpc_sock, sk)
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index c6787b6f459f..6164373d6ce3 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -328,71 +328,57 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
_enter("");
- conn = rx->conn;
+ conn = rxrpc_alloc_connection(gfp);
if (!conn) {
- /* not yet present - create a candidate for a new connection
- * and then redo the check */
- conn = rxrpc_alloc_connection(gfp);
- if (!conn) {
- _leave(" = -ENOMEM");
- return -ENOMEM;
- }
+ _leave(" = -ENOMEM");
+ return -ENOMEM;
+ }
- conn->trans = trans;
- conn->bundle = NULL;
- conn->params = *cp;
- conn->proto.local = cp->local;
- conn->proto.epoch = rxrpc_epoch;
- conn->proto.cid = 0;
- conn->proto.in_clientflag = 0;
- conn->proto.family = cp->peer->srx.transport.family;
- conn->out_clientflag = RXRPC_CLIENT_INITIATED;
- conn->state = RXRPC_CONN_CLIENT;
- conn->avail_calls = RXRPC_MAXCALLS - 1;
-
- key_get(conn->params.key);
-
- ret = rxrpc_init_client_conn_security(conn);
- if (ret < 0) {
- key_put(conn->params.key);
- kfree(conn);
- _leave(" = %d [key]", ret);
- return ret;
- }
+ conn->trans = trans;
+ conn->bundle = NULL;
+ conn->params = *cp;
+ conn->proto.local = cp->local;
+ conn->proto.epoch = rxrpc_epoch;
+ conn->proto.cid = 0;
+ conn->proto.in_clientflag = 0;
+ conn->proto.family = cp->peer->srx.transport.family;
+ conn->out_clientflag = RXRPC_CLIENT_INITIATED;
+ conn->state = RXRPC_CONN_CLIENT;
+ conn->avail_calls = RXRPC_MAXCALLS - 1;
+
+ key_get(conn->params.key);
+
+ ret = rxrpc_init_client_conn_security(conn);
+ if (ret < 0) {
+ key_put(conn->params.key);
+ kfree(conn);
+ _leave(" = %d [key]", ret);
+ return ret;
+ }
- write_lock_bh(&rxrpc_connection_lock);
- list_add_tail(&conn->link, &rxrpc_connections);
- write_unlock_bh(&rxrpc_connection_lock);
+ write_lock_bh(&rxrpc_connection_lock);
+ list_add_tail(&conn->link, &rxrpc_connections);
+ write_unlock_bh(&rxrpc_connection_lock);
- spin_lock(&trans->client_lock);
- atomic_inc(&trans->usage);
+ spin_lock(&trans->client_lock);
+ atomic_inc(&trans->usage);
- _net("CONNECT EXCL new %d on TRANS %d",
- conn->debug_id, conn->trans->debug_id);
+ _net("CONNECT EXCL new %d on TRANS %d",
+ conn->debug_id, conn->trans->debug_id);
- rxrpc_assign_connection_id(conn);
- rx->conn = conn;
- } else {
- spin_lock(&trans->client_lock);
- }
+ rxrpc_assign_connection_id(conn);
- /* we've got a connection with a free channel and we can now attach the
- * call to it
- * - we're holding the transport's client lock
- * - we're holding a reference on the connection
+ /* Since no one else can use the connection, we just use the first
+ * channel.
*/
- for (chan = 0; chan < RXRPC_MAXCALLS; chan++)
- if (!conn->channels[chan])
- goto found_channel;
- goto no_free_channels;
-
-found_channel:
+ chan = 0;
atomic_inc(&conn->usage);
conn->channels[chan] = call;
+ conn->call_counter = 1;
call->conn = conn;
call->channel = chan;
call->cid = conn->proto.cid | chan;
- call->call_id = ++conn->call_counter;
+ call->call_id = 1;
_net("CONNECT client on conn %d chan %d as call %x",
conn->debug_id, chan, call->call_id);
@@ -402,11 +388,6 @@ found_channel:
rxrpc_add_call_ID_to_conn(conn, call);
_leave(" = 0");
return 0;
-
-no_free_channels:
- spin_unlock(&trans->client_lock);
- _leave(" = -ENOSR");
- return -ENOSR;
}
/*
@@ -427,7 +408,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
_enter("%p,%lx,", rx, call->user_call_ID);
- if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags))
+ if (cp->exclusive)
return rxrpc_connect_exclusive(rx, cp, trans, call, gfp);
spin_lock(&trans->client_lock);
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index becbaa7c0a7c..6f8ab0ef839f 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -35,7 +35,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx,
static int rxrpc_sendmsg_cmsg(struct msghdr *msg,
unsigned long *user_call_ID,
enum rxrpc_command *command,
- u32 *abort_code)
+ u32 *abort_code,
+ bool *_exclusive)
{
struct cmsghdr *cmsg;
bool got_user_ID = false;
@@ -93,6 +94,11 @@ static int rxrpc_sendmsg_cmsg(struct msghdr *msg,
return -EINVAL;
break;
+ case RXRPC_EXCLUSIVE_CALL:
+ *_exclusive = true;
+ if (len != 0)
+ return -EINVAL;
+ break;
default:
return -EINVAL;
}
@@ -131,7 +137,7 @@ static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code)
*/
static struct rxrpc_call *
rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
- unsigned long user_call_ID)
+ unsigned long user_call_ID, bool exclusive)
{
struct rxrpc_conn_parameters cp;
struct rxrpc_conn_bundle *bundle;
@@ -155,7 +161,7 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
cp.local = rx->local;
cp.key = rx->key;
cp.security_level = rx->min_sec_level;
- cp.exclusive = test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags);
+ cp.exclusive = rx->exclusive | exclusive;
cp.service_id = srx->srx_service;
trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen,
GFP_KERNEL);
@@ -201,12 +207,14 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len)
enum rxrpc_command cmd;
struct rxrpc_call *call;
unsigned long user_call_ID = 0;
+ bool exclusive = false;
u32 abort_code = 0;
int ret;
_enter("");
- ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code);
+ ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code,
+ &exclusive);
if (ret < 0)
return ret;
@@ -224,7 +232,8 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len)
if (!call) {
if (cmd != RXRPC_CMD_SEND_DATA)
return -EBADSLT;
- call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID);
+ call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID,
+ exclusive);
if (IS_ERR(call))
return PTR_ERR(call);
}
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 06/14] rxrpc: Pass sk_buff * rather than rxrpc_host_header * to functions
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (4 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 05/14] rxrpc: Fix exclusive connection handling David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 07/14] rxrpc: rxrpc_connection_lock shouldn't be a BH lock, but conn_lock is David Howells
` (8 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Pass a pointer to struct sk_buff rather than struct rxrpc_host_header to
functions so that they can in the future get at transport protocol parameters
rather than just RxRPC parameters.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/ar-internal.h | 6 +++---
net/rxrpc/call_accept.c | 4 ++--
net/rxrpc/call_object.c | 21 +++++++++++---------
net/rxrpc/conn_object.c | 49 ++++++++++++++++++++++++-----------------------
net/rxrpc/input.c | 7 +++----
5 files changed, 44 insertions(+), 43 deletions(-)
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 4ca99445e0b7..60ba22f56957 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -544,7 +544,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *,
unsigned long, gfp_t);
struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *,
struct rxrpc_connection *,
- struct rxrpc_host_header *);
+ struct sk_buff *);
void rxrpc_release_call(struct rxrpc_call *);
void rxrpc_release_calls_on_socket(struct rxrpc_sock *);
void __rxrpc_put_call(struct rxrpc_call *);
@@ -574,9 +574,9 @@ int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *,
void rxrpc_put_connection(struct rxrpc_connection *);
void __exit rxrpc_destroy_all_connections(void);
struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *,
- struct rxrpc_host_header *);
+ struct sk_buff *);
extern struct rxrpc_connection *
-rxrpc_incoming_connection(struct rxrpc_transport *, struct rxrpc_host_header *);
+rxrpc_incoming_connection(struct rxrpc_transport *, struct sk_buff *);
static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn)
{
diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c
index 553b67c144e5..5a70dc4e28c6 100644
--- a/net/rxrpc/call_accept.c
+++ b/net/rxrpc/call_accept.c
@@ -110,7 +110,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local,
goto error;
}
- conn = rxrpc_incoming_connection(trans, &sp->hdr);
+ conn = rxrpc_incoming_connection(trans, skb);
rxrpc_put_transport(trans);
if (IS_ERR(conn)) {
_debug("no conn");
@@ -118,7 +118,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local,
goto error;
}
- call = rxrpc_incoming_call(rx, conn, &sp->hdr);
+ call = rxrpc_incoming_call(rx, conn, skb);
rxrpc_put_connection(conn);
if (IS_ERR(call)) {
_debug("no call");
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index 5c2dceaf6a9c..d83f2cbb80a9 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -421,8 +421,9 @@ found_user_ID_now_present:
*/
struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
struct rxrpc_connection *conn,
- struct rxrpc_host_header *hdr)
+ struct sk_buff *skb)
{
+ struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
struct rxrpc_call *call, *candidate;
struct rb_node **p, *parent;
u32 call_id;
@@ -435,13 +436,13 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
if (!candidate)
return ERR_PTR(-EBUSY);
- candidate->socket = rx;
- candidate->conn = conn;
- candidate->cid = hdr->cid;
- candidate->call_id = hdr->callNumber;
- candidate->channel = hdr->cid & RXRPC_CHANNELMASK;
- candidate->rx_data_post = 0;
- candidate->state = RXRPC_CALL_SERVER_ACCEPTING;
+ candidate->socket = rx;
+ candidate->conn = conn;
+ candidate->cid = sp->hdr.cid;
+ candidate->call_id = sp->hdr.callNumber;
+ candidate->channel = sp->hdr.cid & RXRPC_CHANNELMASK;
+ candidate->rx_data_post = 0;
+ candidate->state = RXRPC_CALL_SERVER_ACCEPTING;
if (conn->security_ix > 0)
candidate->state = RXRPC_CALL_SERVER_SECURING;
@@ -450,7 +451,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
/* set the channel for this call */
call = conn->channels[candidate->channel];
_debug("channel[%u] is %p", candidate->channel, call);
- if (call && call->call_id == hdr->callNumber) {
+ if (call && call->call_id == sp->hdr.callNumber) {
/* already set; must've been a duplicate packet */
_debug("extant call [%d]", call->state);
ASSERTCMP(call->conn, ==, conn);
@@ -488,7 +489,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
/* check the call number isn't duplicate */
_debug("check dup");
- call_id = hdr->callNumber;
+ call_id = sp->hdr.callNumber;
p = &conn->calls.rb_node;
parent = NULL;
while (*p) {
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index 6164373d6ce3..3b42fc4f4fe3 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -588,10 +588,10 @@ interrupted:
* get a record of an incoming connection
*/
struct rxrpc_connection *
-rxrpc_incoming_connection(struct rxrpc_transport *trans,
- struct rxrpc_host_header *hdr)
+rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb)
{
struct rxrpc_connection *conn, *candidate = NULL;
+ struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
struct rb_node *p, **pp;
const char *new = "old";
__be32 epoch;
@@ -599,10 +599,10 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
_enter("");
- ASSERT(hdr->flags & RXRPC_CLIENT_INITIATED);
+ ASSERT(sp->hdr.flags & RXRPC_CLIENT_INITIATED);
- epoch = hdr->epoch;
- cid = hdr->cid & RXRPC_CIDMASK;
+ epoch = sp->hdr.epoch;
+ cid = sp->hdr.cid & RXRPC_CIDMASK;
/* search the connection list first */
read_lock_bh(&trans->conn_lock);
@@ -634,19 +634,19 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans,
return ERR_PTR(-ENOMEM);
}
- candidate->trans = trans;
- candidate->proto.local = trans->local;
- candidate->proto.epoch = hdr->epoch;
- candidate->proto.cid = hdr->cid & RXRPC_CIDMASK;
- candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED;
- candidate->params.local = trans->local;
- candidate->params.peer = trans->peer;
- candidate->params.service_id = hdr->serviceId;
- candidate->security_ix = hdr->securityIndex;
- candidate->out_clientflag = 0;
- candidate->state = RXRPC_CONN_SERVER;
+ candidate->trans = trans;
+ candidate->proto.local = trans->local;
+ candidate->proto.epoch = sp->hdr.epoch;
+ candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK;
+ candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED;
+ candidate->params.local = trans->local;
+ candidate->params.peer = trans->peer;
+ candidate->params.service_id = sp->hdr.serviceId;
+ candidate->security_ix = sp->hdr.securityIndex;
+ candidate->out_clientflag = 0;
+ candidate->state = RXRPC_CONN_SERVER;
if (candidate->params.service_id)
- candidate->state = RXRPC_CONN_SERVER_UNSECURED;
+ candidate->state = RXRPC_CONN_SERVER_UNSECURED;
write_lock_bh(&trans->conn_lock);
@@ -691,7 +691,7 @@ success:
/* we found the connection in the list immediately */
found_extant_connection:
- if (hdr->securityIndex != conn->security_ix) {
+ if (sp->hdr.securityIndex != conn->security_ix) {
read_unlock_bh(&trans->conn_lock);
goto security_mismatch;
}
@@ -701,7 +701,7 @@ found_extant_connection:
/* we found the connection on the second time through the list */
found_extant_second:
- if (hdr->securityIndex != conn->security_ix) {
+ if (sp->hdr.securityIndex != conn->security_ix) {
write_unlock_bh(&trans->conn_lock);
goto security_mismatch;
}
@@ -721,20 +721,21 @@ security_mismatch:
* packet
*/
struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
- struct rxrpc_host_header *hdr)
+ struct sk_buff *skb)
{
struct rxrpc_connection *conn;
+ struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
struct rb_node *p;
u32 epoch, cid;
- _enter(",{%x,%x}", hdr->cid, hdr->flags);
+ _enter(",{%x,%x}", sp->hdr.cid, sp->hdr.flags);
read_lock_bh(&trans->conn_lock);
- cid = hdr->cid & RXRPC_CIDMASK;
- epoch = hdr->epoch;
+ cid = sp->hdr.cid & RXRPC_CIDMASK;
+ epoch = sp->hdr.epoch;
- if (hdr->flags & RXRPC_CLIENT_INITIATED)
+ if (sp->hdr.flags & RXRPC_CLIENT_INITIATED)
p = trans->server_conns.rb_node;
else
p = trans->client_conns.rb_node;
diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index 6af7f40c5030..cf540efa9c17 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -628,8 +628,7 @@ int rxrpc_extract_header(struct rxrpc_skb_priv *sp, struct sk_buff *skb)
}
static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
- struct sk_buff *skb,
- struct rxrpc_skb_priv *sp)
+ struct sk_buff *skb)
{
struct rxrpc_peer *peer;
struct rxrpc_transport *trans;
@@ -647,7 +646,7 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
if (!trans)
goto cant_find_conn;
- conn = rxrpc_find_connection(trans, &sp->hdr);
+ conn = rxrpc_find_connection(trans, skb);
rxrpc_put_transport(trans);
if (!conn)
goto cant_find_conn;
@@ -739,7 +738,7 @@ void rxrpc_data_ready(struct sock *sk)
* old-fashioned way doesn't really hurt */
struct rxrpc_connection *conn;
- conn = rxrpc_conn_from_local(local, skb, sp);
+ conn = rxrpc_conn_from_local(local, skb);
if (!conn)
goto cant_route_call;
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 07/14] rxrpc: rxrpc_connection_lock shouldn't be a BH lock, but conn_lock is
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (5 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 06/14] rxrpc: Pass sk_buff * rather than rxrpc_host_header * to functions David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 08/14] rxrpc: Use IDR to allocate client conn IDs on a machine-wide basis David Howells
` (7 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
rxrpc_connection_lock shouldn't be accessed as a BH-excluding lock. It's
only accessed in a few places and none of those are in BH-context.
rxrpc_transport::conn_lock, however, *is* a BH-excluding lock and should be
accessed so consistently.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/conn_object.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index 3b42fc4f4fe3..cab2f6dbc5a1 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -356,9 +356,9 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
return ret;
}
- write_lock_bh(&rxrpc_connection_lock);
+ write_lock(&rxrpc_connection_lock);
list_add_tail(&conn->link, &rxrpc_connections);
- write_unlock_bh(&rxrpc_connection_lock);
+ write_unlock(&rxrpc_connection_lock);
spin_lock(&trans->client_lock);
atomic_inc(&trans->usage);
@@ -677,9 +677,9 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb)
write_unlock_bh(&trans->conn_lock);
- write_lock_bh(&rxrpc_connection_lock);
+ write_lock(&rxrpc_connection_lock);
list_add_tail(&conn->link, &rxrpc_connections);
- write_unlock_bh(&rxrpc_connection_lock);
+ write_unlock(&rxrpc_connection_lock);
new = "new";
@@ -828,7 +828,7 @@ static void rxrpc_connection_reaper(struct work_struct *work)
now = ktime_get_seconds();
earliest = ULONG_MAX;
- write_lock_bh(&rxrpc_connection_lock);
+ write_lock(&rxrpc_connection_lock);
list_for_each_entry_safe(conn, _p, &rxrpc_connections, link) {
_debug("reap CONN %d { u=%d,t=%ld }",
conn->debug_id, atomic_read(&conn->usage),
@@ -838,7 +838,7 @@ static void rxrpc_connection_reaper(struct work_struct *work)
continue;
spin_lock(&conn->trans->client_lock);
- write_lock(&conn->trans->conn_lock);
+ write_lock_bh(&conn->trans->conn_lock);
reap_time = conn->put_time + rxrpc_connection_expiry;
if (atomic_read(&conn->usage) > 0) {
@@ -860,10 +860,10 @@ static void rxrpc_connection_reaper(struct work_struct *work)
earliest = reap_time;
}
- write_unlock(&conn->trans->conn_lock);
+ write_unlock_bh(&conn->trans->conn_lock);
spin_unlock(&conn->trans->client_lock);
}
- write_unlock_bh(&rxrpc_connection_lock);
+ write_unlock(&rxrpc_connection_lock);
if (earliest != ULONG_MAX) {
_debug("reschedule reaper %ld", (long) earliest - now);
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 08/14] rxrpc: Use IDR to allocate client conn IDs on a machine-wide basis
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (6 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 07/14] rxrpc: rxrpc_connection_lock shouldn't be a BH lock, but conn_lock is David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 09/14] rxrpc: Validate the net address given to rxrpc_kernel_begin_call() David Howells
` (6 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Use the IDR facility to allocate client connection IDs on a machine-wide
basis so that each client connection has a unique identifier. When the
connection ID space wraps, we advance the epoch by 1, thereby effectively
having a 62-bit ID space. The IDR facility is then used to look up client
connections during incoming packet routing instead of using an rbtree
rooted on the transport.
This change allows for the removal of the transport in the future and also
means that client connections can be looked up directly in the data-ready
handler by connection ID.
The ID management code is placed in a new file, conn-client.c, to which all
the client connection-specific code will eventually move.
Note that the IDR tree gets very expensive on memory if the connection IDs
are widely scattered throughout the number space, so we shall need to
retire connections that have, say, an ID more than four times the maximum
number of client conns away from the current allocation point to try and
keep the IDs concentrated. We will also need to retire connections from an
old epoch.
Also note that, for the moment, a pointer to the transport has to be passed
through into the ID allocation function so that we can take a BH lock to
prevent a locking issue against in-BH lookup of client connections. This
will go away later when RCU is used for server connections also.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/Makefile | 1
net/rxrpc/af_rxrpc.c | 2
net/rxrpc/ar-internal.h | 13 ++-
net/rxrpc/conn_client.c | 99 ++++++++++++++++++++
net/rxrpc/conn_object.c | 231 +++++++++++++++++------------------------------
net/rxrpc/transport.c | 2
6 files changed, 196 insertions(+), 152 deletions(-)
create mode 100644 net/rxrpc/conn_client.c
diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile
index b005027f80cf..cfa221536f33 100644
--- a/net/rxrpc/Makefile
+++ b/net/rxrpc/Makefile
@@ -7,6 +7,7 @@ af-rxrpc-y := \
call_accept.o \
call_event.o \
call_object.o \
+ conn_client.o \
conn_event.o \
conn_object.o \
input.o \
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 73f5c553eef4..408bd024125b 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -858,6 +858,8 @@ static void __exit af_rxrpc_exit(void)
_debug("synchronise RCU");
rcu_barrier();
_debug("destroy locals");
+ ASSERT(idr_is_empty(&rxrpc_client_conn_ids));
+ idr_destroy(&rxrpc_client_conn_ids);
rxrpc_destroy_all_locals();
remove_proc_entry("rxrpc_conns", init_net.proc_net);
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 60ba22f56957..89966508b26c 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -233,7 +233,6 @@ struct rxrpc_transport {
struct rxrpc_local *local; /* local transport endpoint */
struct rxrpc_peer *peer; /* remote transport endpoint */
struct rb_root bundles; /* client connection bundles on this transport */
- struct rb_root client_conns; /* client connections on this transport */
struct rb_root server_conns; /* server connections on this transport */
struct list_head link; /* link in master session list */
unsigned long put_time; /* time at which to reap */
@@ -241,7 +240,6 @@ struct rxrpc_transport {
rwlock_t conn_lock; /* lock for active/dead connections */
atomic_t usage;
int debug_id; /* debug ID for printks */
- unsigned int conn_idcounter; /* connection ID counter (client) */
};
/*
@@ -312,6 +310,8 @@ struct rxrpc_connection {
struct key *server_key; /* security for this service */
struct crypto_skcipher *cipher; /* encryption handle */
struct rxrpc_crypt csum_iv; /* packet checksum base */
+ unsigned long flags;
+#define RXRPC_CONN_HAS_IDR 0 /* - Has a client conn ID assigned */
unsigned long events;
#define RXRPC_CONN_CHALLENGE 0 /* send challenge packet */
unsigned long put_time; /* time at which to reap */
@@ -551,6 +551,15 @@ void __rxrpc_put_call(struct rxrpc_call *);
void __exit rxrpc_destroy_all_calls(void);
/*
+ * conn_client.c
+ */
+extern struct idr rxrpc_client_conn_ids;
+
+int rxrpc_get_client_connection_id(struct rxrpc_connection *,
+ struct rxrpc_transport *, gfp_t);
+void rxrpc_put_client_connection_id(struct rxrpc_connection *);
+
+/*
* conn_event.c
*/
void rxrpc_process_connection(struct work_struct *);
diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c
new file mode 100644
index 000000000000..2cccb4be289d
--- /dev/null
+++ b/net/rxrpc/conn_client.c
@@ -0,0 +1,99 @@
+/* Client connection-specific management code.
+ *
+ * Copyright (C) 2016 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public Licence
+ * as published by the Free Software Foundation; either version
+ * 2 of the Licence, or (at your option) any later version.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/slab.h>
+#include <linux/idr.h>
+#include <linux/timer.h>
+#include "ar-internal.h"
+
+/*
+ * We use machine-unique IDs for our client connections.
+ */
+DEFINE_IDR(rxrpc_client_conn_ids);
+static DEFINE_SPINLOCK(rxrpc_conn_id_lock);
+
+/*
+ * Get a connection ID and epoch for a client connection from the global pool.
+ * The connection struct pointer is then recorded in the idr radix tree. The
+ * epoch is changed if this wraps.
+ *
+ * TODO: The IDR tree gets very expensive on memory if the connection IDs are
+ * widely scattered throughout the number space, so we shall need to retire
+ * connections that have, say, an ID more than four times the maximum number of
+ * client conns away from the current allocation point to try and keep the IDs
+ * concentrated. We will also need to retire connections from an old epoch.
+ */
+int rxrpc_get_client_connection_id(struct rxrpc_connection *conn,
+ struct rxrpc_transport *trans,
+ gfp_t gfp)
+{
+ u32 epoch;
+ int id;
+
+ _enter("");
+
+ idr_preload(gfp);
+ write_lock_bh(&trans->conn_lock);
+ spin_lock(&rxrpc_conn_id_lock);
+
+ epoch = rxrpc_epoch;
+
+ /* We could use idr_alloc_cyclic() here, but we really need to know
+ * when the thing wraps so that we can advance the epoch.
+ */
+ if (rxrpc_client_conn_ids.cur == 0)
+ rxrpc_client_conn_ids.cur = 1;
+ id = idr_alloc(&rxrpc_client_conn_ids, conn,
+ rxrpc_client_conn_ids.cur, 0x40000000, GFP_NOWAIT);
+ if (id < 0) {
+ if (id != -ENOSPC)
+ goto error;
+ id = idr_alloc(&rxrpc_client_conn_ids, conn,
+ 1, 0x40000000, GFP_NOWAIT);
+ if (id < 0)
+ goto error;
+ epoch++;
+ rxrpc_epoch = epoch;
+ }
+ rxrpc_client_conn_ids.cur = id + 1;
+
+ spin_unlock(&rxrpc_conn_id_lock);
+ write_unlock_bh(&trans->conn_lock);
+ idr_preload_end();
+
+ conn->proto.epoch = epoch;
+ conn->proto.cid = id << RXRPC_CIDSHIFT;
+ set_bit(RXRPC_CONN_HAS_IDR, &conn->flags);
+ _leave(" [CID %x:%x]", epoch, conn->proto.cid);
+ return 0;
+
+error:
+ spin_unlock(&rxrpc_conn_id_lock);
+ write_unlock_bh(&trans->conn_lock);
+ idr_preload_end();
+ _leave(" = %d", id);
+ return id;
+}
+
+/*
+ * Release a connection ID for a client connection from the global pool.
+ */
+void rxrpc_put_client_connection_id(struct rxrpc_connection *conn)
+{
+ if (test_bit(RXRPC_CONN_HAS_IDR, &conn->flags)) {
+ spin_lock(&rxrpc_conn_id_lock);
+ idr_remove(&rxrpc_client_conn_ids,
+ conn->proto.cid >> RXRPC_CIDSHIFT);
+ spin_unlock(&rxrpc_conn_id_lock);
+ }
+}
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index cab2f6dbc5a1..312b75091d29 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -207,81 +207,6 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp)
}
/*
- * assign a connection ID to a connection and add it to the transport's
- * connection lookup tree
- * - called with transport client lock held
- */
-static void rxrpc_assign_connection_id(struct rxrpc_connection *conn)
-{
- struct rxrpc_connection *xconn;
- struct rb_node *parent, **p;
- __be32 epoch;
- u32 cid;
-
- _enter("");
-
- epoch = conn->proto.epoch;
-
- write_lock_bh(&conn->trans->conn_lock);
-
- conn->trans->conn_idcounter += RXRPC_CID_INC;
- if (conn->trans->conn_idcounter < RXRPC_CID_INC)
- conn->trans->conn_idcounter = RXRPC_CID_INC;
- cid = conn->trans->conn_idcounter;
-
-attempt_insertion:
- parent = NULL;
- p = &conn->trans->client_conns.rb_node;
-
- while (*p) {
- parent = *p;
- xconn = rb_entry(parent, struct rxrpc_connection, node);
-
- if (epoch < xconn->proto.epoch)
- p = &(*p)->rb_left;
- else if (epoch > xconn->proto.epoch)
- p = &(*p)->rb_right;
- else if (cid < xconn->proto.cid)
- p = &(*p)->rb_left;
- else if (cid > xconn->proto.cid)
- p = &(*p)->rb_right;
- else
- goto id_exists;
- }
-
- /* we've found a suitable hole - arrange for this connection to occupy
- * it */
- rb_link_node(&conn->node, parent, p);
- rb_insert_color(&conn->node, &conn->trans->client_conns);
-
- conn->proto.cid = cid;
- write_unlock_bh(&conn->trans->conn_lock);
- _leave(" [CID %x]", cid);
- return;
-
- /* we found a connection with the proposed ID - walk the tree from that
- * point looking for the next unused ID */
-id_exists:
- for (;;) {
- cid += RXRPC_CID_INC;
- if (cid < RXRPC_CID_INC) {
- cid = RXRPC_CID_INC;
- conn->trans->conn_idcounter = cid;
- goto attempt_insertion;
- }
-
- parent = rb_next(parent);
- if (!parent)
- goto attempt_insertion;
-
- xconn = rb_entry(parent, struct rxrpc_connection, node);
- if (epoch < xconn->proto.epoch ||
- cid < xconn->proto.cid)
- goto attempt_insertion;
- }
-}
-
-/*
* add a call to a connection's call-by-ID tree
*/
static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn,
@@ -315,27 +240,24 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn,
}
/*
- * connect a call on an exclusive connection
+ * Allocate a client connection.
*/
-static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
- struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
- struct rxrpc_call *call,
- gfp_t gfp)
+static struct rxrpc_connection *
+rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp,
+ struct rxrpc_transport *trans,
+ gfp_t gfp)
{
struct rxrpc_connection *conn;
- int chan, ret;
+ int ret;
_enter("");
conn = rxrpc_alloc_connection(gfp);
if (!conn) {
_leave(" = -ENOMEM");
- return -ENOMEM;
+ return ERR_PTR(-ENOMEM);
}
- conn->trans = trans;
- conn->bundle = NULL;
conn->params = *cp;
conn->proto.local = cp->local;
conn->proto.epoch = rxrpc_epoch;
@@ -344,35 +266,75 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
conn->proto.family = cp->peer->srx.transport.family;
conn->out_clientflag = RXRPC_CLIENT_INITIATED;
conn->state = RXRPC_CONN_CLIENT;
- conn->avail_calls = RXRPC_MAXCALLS - 1;
- key_get(conn->params.key);
+ switch (conn->proto.family) {
+ case AF_INET:
+ conn->proto.addr_size = sizeof(conn->proto.ipv4_addr);
+ conn->proto.ipv4_addr = cp->peer->srx.transport.sin.sin_addr;
+ conn->proto.port = cp->peer->srx.transport.sin.sin_port;
+ break;
+ }
+
+ ret = rxrpc_get_client_connection_id(conn, trans, gfp);
+ if (ret < 0)
+ goto error_0;
ret = rxrpc_init_client_conn_security(conn);
- if (ret < 0) {
- key_put(conn->params.key);
- kfree(conn);
- _leave(" = %d [key]", ret);
- return ret;
- }
+ if (ret < 0)
+ goto error_1;
+
+ conn->security->prime_packet_security(conn);
write_lock(&rxrpc_connection_lock);
list_add_tail(&conn->link, &rxrpc_connections);
write_unlock(&rxrpc_connection_lock);
- spin_lock(&trans->client_lock);
+ key_get(conn->params.key);
+
+ _leave(" = %p", conn);
+ return conn;
+
+error_1:
+ rxrpc_put_client_connection_id(conn);
+error_0:
+ kfree(conn);
+ _leave(" = %d", ret);
+ return ERR_PTR(ret);
+}
+
+/*
+ * connect a call on an exclusive connection
+ */
+static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
+ struct rxrpc_conn_parameters *cp,
+ struct rxrpc_transport *trans,
+ struct rxrpc_call *call,
+ gfp_t gfp)
+{
+ struct rxrpc_connection *conn;
+ int chan;
+
+ _enter("");
+
+ conn = rxrpc_alloc_client_connection(cp, trans, gfp);
+ if (IS_ERR(conn)) {
+ _leave(" = %ld", PTR_ERR(conn));
+ return PTR_ERR(conn);
+ }
+
atomic_inc(&trans->usage);
+ conn->trans = trans;
+ conn->bundle = NULL;
_net("CONNECT EXCL new %d on TRANS %d",
conn->debug_id, conn->trans->debug_id);
- rxrpc_assign_connection_id(conn);
-
/* Since no one else can use the connection, we just use the first
* channel.
*/
chan = 0;
atomic_inc(&conn->usage);
+ conn->avail_calls = RXRPC_MAXCALLS - 1;
conn->channels[chan] = call;
conn->call_counter = 1;
call->conn = conn;
@@ -383,8 +345,6 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
_net("CONNECT client on conn %d chan %d as call %x",
conn->debug_id, chan, call->call_id);
- spin_unlock(&trans->client_lock);
-
rxrpc_add_call_ID_to_conn(conn, call);
_leave(" = 0");
return 0;
@@ -402,7 +362,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
gfp_t gfp)
{
struct rxrpc_connection *conn, *candidate;
- int chan, ret;
+ int chan;
DECLARE_WAITQUEUE(myself, current);
@@ -492,51 +452,25 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
/* not yet present - create a candidate for a new connection and then
* redo the check */
- candidate = rxrpc_alloc_connection(gfp);
+ candidate = rxrpc_alloc_client_connection(cp, trans, gfp);
if (!candidate) {
_leave(" = -ENOMEM");
return -ENOMEM;
}
+ atomic_inc(&bundle->usage);
+ atomic_inc(&trans->usage);
candidate->trans = trans;
candidate->bundle = bundle;
- candidate->params = *cp;
- candidate->proto.local = cp->local;
- candidate->proto.epoch = rxrpc_epoch;
- candidate->proto.cid = 0;
- candidate->proto.in_clientflag = 0;
- candidate->proto.family = cp->peer->srx.transport.family;
- candidate->out_clientflag = RXRPC_CLIENT_INITIATED;
- candidate->state = RXRPC_CONN_CLIENT;
- candidate->avail_calls = RXRPC_MAXCALLS;
-
- key_get(candidate->params.key);
-
- ret = rxrpc_init_client_conn_security(candidate);
- if (ret < 0) {
- key_put(candidate->params.key);
- kfree(candidate);
- _leave(" = %d [key]", ret);
- return ret;
- }
-
- write_lock_bh(&rxrpc_connection_lock);
- list_add_tail(&candidate->link, &rxrpc_connections);
- write_unlock_bh(&rxrpc_connection_lock);
spin_lock(&trans->client_lock);
list_add(&candidate->bundle_link, &bundle->unused_conns);
bundle->num_conns++;
- atomic_inc(&bundle->usage);
- atomic_inc(&trans->usage);
_net("CONNECT new %d on TRANS %d",
candidate->debug_id, candidate->trans->debug_id);
- rxrpc_assign_connection_id(candidate);
- candidate->security->prime_packet_security(candidate);
-
/* leave the candidate lurking in zombie mode attached to the
* bundle until we're ready for it */
rxrpc_put_connection(candidate);
@@ -735,25 +669,27 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
cid = sp->hdr.cid & RXRPC_CIDMASK;
epoch = sp->hdr.epoch;
- if (sp->hdr.flags & RXRPC_CLIENT_INITIATED)
+ if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) {
p = trans->server_conns.rb_node;
- else
- p = trans->client_conns.rb_node;
-
- while (p) {
- conn = rb_entry(p, struct rxrpc_connection, node);
-
- _debug("maybe %x", conn->proto.cid);
-
- if (epoch < conn->proto.epoch)
- p = p->rb_left;
- else if (epoch > conn->proto.epoch)
- p = p->rb_right;
- else if (cid < conn->proto.cid)
- p = p->rb_left;
- else if (cid > conn->proto.cid)
- p = p->rb_right;
- else
+ while (p) {
+ conn = rb_entry(p, struct rxrpc_connection, node);
+
+ _debug("maybe %x", conn->proto.cid);
+
+ if (epoch < conn->proto.epoch)
+ p = p->rb_left;
+ else if (epoch > conn->proto.epoch)
+ p = p->rb_right;
+ else if (cid < conn->proto.cid)
+ p = p->rb_left;
+ else if (cid > conn->proto.cid)
+ p = p->rb_right;
+ else
+ goto found;
+ }
+ } else {
+ conn = idr_find(&rxrpc_client_conn_ids, cid >> RXRPC_CIDSHIFT);
+ if (conn && conn->proto.epoch == epoch)
goto found;
}
@@ -846,8 +782,7 @@ static void rxrpc_connection_reaper(struct work_struct *work)
} else if (reap_time <= now) {
list_move_tail(&conn->link, &graveyard);
if (conn->out_clientflag)
- rb_erase(&conn->node,
- &conn->trans->client_conns);
+ rxrpc_put_client_connection_id(conn);
else
rb_erase(&conn->node,
&conn->trans->server_conns);
diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c
index 24c71218a6f8..140628d94bb0 100644
--- a/net/rxrpc/transport.c
+++ b/net/rxrpc/transport.c
@@ -47,12 +47,10 @@ static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local,
trans->peer = peer;
INIT_LIST_HEAD(&trans->link);
trans->bundles = RB_ROOT;
- trans->client_conns = RB_ROOT;
trans->server_conns = RB_ROOT;
spin_lock_init(&trans->client_lock);
rwlock_init(&trans->conn_lock);
atomic_set(&trans->usage, 1);
- trans->conn_idcounter = peer->srx.srx_service << 16;
trans->debug_id = atomic_inc_return(&rxrpc_debug_id);
}
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 09/14] rxrpc: Validate the net address given to rxrpc_kernel_begin_call()
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (7 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 08/14] rxrpc: Use IDR to allocate client conn IDs on a machine-wide basis David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:50 ` [PATCH net-next 10/14] rxrpc: Calls displayed in /proc may in future lack a connection David Howells
` (5 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Validate the net address given to rxrpc_kernel_begin_call() before using
it.
Whilst this should be mostly unnecessary for in-kernel users, it does clear
the tail of the address struct in case we want to hash or compare the whole
thing.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/af_rxrpc.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 408bd024125b..b29bb50af5de 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -280,9 +280,14 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
struct rxrpc_transport *trans;
struct rxrpc_call *call;
struct rxrpc_sock *rx = rxrpc_sk(sock->sk);
+ int ret;
_enter(",,%x,%lx", key_serial(key), user_call_ID);
+ ret = rxrpc_validate_address(rx, srx, sizeof(*srx));
+ if (ret < 0)
+ return ERR_PTR(ret);
+
lock_sock(&rx->sk);
if (!key)
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 10/14] rxrpc: Calls displayed in /proc may in future lack a connection
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (8 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 09/14] rxrpc: Validate the net address given to rxrpc_kernel_begin_call() David Howells
@ 2016-06-22 9:50 ` David Howells
2016-06-22 9:51 ` [PATCH net-next 11/14] rxrpc: Make rxrpc_send_packet() take a connection not a transport David Howells
` (4 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:50 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Allocated rxrpc calls displayed in /proc/net/rxrpc_calls may in future be
on the proc list before they're connected or after they've been
disconnected - in which case they may not have a pointer to a connection
struct that can be used to get data from there.
Deal with this by using stuff from the call struct in preference where
possible and printing "no_connection" rather than a peer address if no
connection is assigned.
This change also has the added bonus that the service ID is now taken from
the call rather the connection which will allow per-call service upgrades
to be shown - something required for AuriStor server compatibility.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/proc.c | 19 +++++++++++--------
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/net/rxrpc/proc.c b/net/rxrpc/proc.c
index 9863270691d7..500cdcdc843c 100644
--- a/net/rxrpc/proc.c
+++ b/net/rxrpc/proc.c
@@ -59,25 +59,28 @@ static int rxrpc_call_seq_show(struct seq_file *seq, void *v)
}
call = list_entry(v, struct rxrpc_call, link);
- conn = call->conn;
sprintf(lbuff, "%pI4:%u",
- &conn->params.local->srx.transport.sin.sin_addr,
- ntohs(conn->params.local->srx.transport.sin.sin_port));
+ &call->local->srx.transport.sin.sin_addr,
+ ntohs(call->local->srx.transport.sin.sin_port));
- sprintf(rbuff, "%pI4:%u",
- &conn->params.peer->srx.transport.sin.sin_addr,
- ntohs(conn->params.peer->srx.transport.sin.sin_port));
+ conn = call->conn;
+ if (conn)
+ sprintf(rbuff, "%pI4:%u",
+ &conn->params.peer->srx.transport.sin.sin_addr,
+ ntohs(conn->params.peer->srx.transport.sin.sin_port));
+ else
+ strcpy(rbuff, "no_connection");
seq_printf(seq,
"UDP %-22.22s %-22.22s %4x %08x %08x %s %3u"
" %-8.8s %08x %lx\n",
lbuff,
rbuff,
- call->conn->params.service_id,
+ call->service_id,
call->cid,
call->call_id,
- rxrpc_conn_is_service(call->conn) ? "Svc" : "Clt",
+ call->in_clientflag ? "Svc" : "Clt",
atomic_read(&call->usage),
rxrpc_call_states[call->state],
call->remote_abort ?: call->local_abort,
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 11/14] rxrpc: Make rxrpc_send_packet() take a connection not a transport
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (9 preceding siblings ...)
2016-06-22 9:50 ` [PATCH net-next 10/14] rxrpc: Calls displayed in /proc may in future lack a connection David Howells
@ 2016-06-22 9:51 ` David Howells
2016-06-22 9:51 ` [PATCH net-next 12/14] rxrpc: Provide more refcount helper functions David Howells
` (3 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:51 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Make rxrpc_send_packet() take a connection not a transport as part of the
phasing out of the rxrpc_transport struct.
Whilst we're at it, rename the function to rxrpc_send_data_packet() to
differentiate it from the other packet sending functions.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/ar-internal.h | 2 +-
net/rxrpc/call_event.c | 2 +-
net/rxrpc/output.c | 51 +++++++++++++++++++++++++++--------------------
3 files changed, 31 insertions(+), 24 deletions(-)
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 89966508b26c..cfbd028aa551 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -670,7 +670,7 @@ extern const char *rxrpc_acks(u8 reason);
*/
extern unsigned int rxrpc_resend_timeout;
-int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *);
+int rxrpc_send_data_packet(struct rxrpc_connection *, struct sk_buff *);
int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t);
/*
diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c
index b43faf573ed3..0ba84295f913 100644
--- a/net/rxrpc/call_event.c
+++ b/net/rxrpc/call_event.c
@@ -187,7 +187,7 @@ static void rxrpc_resend(struct rxrpc_call *call)
_proto("Tx DATA %%%u { #%d }",
sp->hdr.serial, sp->hdr.seq);
- if (rxrpc_send_packet(call->conn->trans, txb) < 0) {
+ if (rxrpc_send_data_packet(call->conn, txb) < 0) {
stop = true;
sp->resend_at = jiffies + 3;
} else {
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index 6f8ab0ef839f..db3933cf6b97 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -338,7 +338,7 @@ EXPORT_SYMBOL(rxrpc_kernel_abort_call);
/*
* send a packet through the transport endpoint
*/
-int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb)
+int rxrpc_send_data_packet(struct rxrpc_connection *conn, struct sk_buff *skb)
{
struct kvec iov[1];
struct msghdr msg;
@@ -349,30 +349,30 @@ int rxrpc_send_packet(struct rxrpc_transport *trans, struct sk_buff *skb)
iov[0].iov_base = skb->head;
iov[0].iov_len = skb->len;
- msg.msg_name = &trans->peer->srx.transport.sin;
- msg.msg_namelen = sizeof(trans->peer->srx.transport.sin);
+ msg.msg_name = &conn->params.peer->srx.transport;
+ msg.msg_namelen = conn->params.peer->srx.transport_len;
msg.msg_control = NULL;
msg.msg_controllen = 0;
msg.msg_flags = 0;
/* send the packet with the don't fragment bit set if we currently
* think it's small enough */
- if (skb->len - sizeof(struct rxrpc_wire_header) < trans->peer->maxdata) {
- down_read(&trans->local->defrag_sem);
+ if (skb->len - sizeof(struct rxrpc_wire_header) < conn->params.peer->maxdata) {
+ down_read(&conn->params.local->defrag_sem);
/* send the packet by UDP
* - returns -EMSGSIZE if UDP would have to fragment the packet
* to go out of the interface
* - in which case, we'll have processed the ICMP error
* message and update the peer record
*/
- ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1,
+ ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 1,
iov[0].iov_len);
- up_read(&trans->local->defrag_sem);
+ up_read(&conn->params.local->defrag_sem);
if (ret == -EMSGSIZE)
goto send_fragmentable;
- _leave(" = %d [%u]", ret, trans->peer->maxdata);
+ _leave(" = %d [%u]", ret, conn->params.peer->maxdata);
return ret;
}
@@ -380,21 +380,28 @@ send_fragmentable:
/* attempt to send this message with fragmentation enabled */
_debug("send fragment");
- down_write(&trans->local->defrag_sem);
- opt = IP_PMTUDISC_DONT;
- ret = kernel_setsockopt(trans->local->socket, SOL_IP, IP_MTU_DISCOVER,
- (char *) &opt, sizeof(opt));
- if (ret == 0) {
- ret = kernel_sendmsg(trans->local->socket, &msg, iov, 1,
- iov[0].iov_len);
-
- opt = IP_PMTUDISC_DO;
- kernel_setsockopt(trans->local->socket, SOL_IP,
- IP_MTU_DISCOVER, (char *) &opt, sizeof(opt));
+ down_write(&conn->params.local->defrag_sem);
+
+ switch (conn->params.local->srx.transport.family) {
+ case AF_INET:
+ opt = IP_PMTUDISC_DONT;
+ ret = kernel_setsockopt(conn->params.local->socket,
+ SOL_IP, IP_MTU_DISCOVER,
+ (char *)&opt, sizeof(opt));
+ if (ret == 0) {
+ ret = kernel_sendmsg(conn->params.local->socket, &msg, iov, 1,
+ iov[0].iov_len);
+
+ opt = IP_PMTUDISC_DO;
+ kernel_setsockopt(conn->params.local->socket, SOL_IP,
+ IP_MTU_DISCOVER,
+ (char *)&opt, sizeof(opt));
+ }
+ break;
}
- up_write(&trans->local->defrag_sem);
- _leave(" = %d [frag %u]", ret, trans->peer->maxdata);
+ up_write(&conn->params.local->defrag_sem);
+ _leave(" = %d [frag %u]", ret, conn->params.peer->maxdata);
return ret;
}
@@ -506,7 +513,7 @@ static void rxrpc_queue_packet(struct rxrpc_call *call, struct sk_buff *skb,
if (try_to_del_timer_sync(&call->ack_timer) >= 0) {
/* the packet may be freed by rxrpc_process_call() before this
* returns */
- ret = rxrpc_send_packet(call->conn->trans, skb);
+ ret = rxrpc_send_data_packet(call->conn, skb);
_net("sent skb %p", skb);
} else {
_debug("failed to delete ACK timer");
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 12/14] rxrpc: Provide more refcount helper functions
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (10 preceding siblings ...)
2016-06-22 9:51 ` [PATCH net-next 11/14] rxrpc: Make rxrpc_send_packet() take a connection not a transport David Howells
@ 2016-06-22 9:51 ` David Howells
2016-06-22 9:51 ` [PATCH net-next 13/14] rxrpc: Kill the client connection bundle concept David Howells
` (2 subsequent siblings)
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:51 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Provide refcount helper functions for connections so that the code doesn't
touch local or connection usage counts directly.
Also make it such that local and peer put functions can take a NULL
pointer.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/af_rxrpc.c | 7 ++-----
net/rxrpc/ar-internal.h | 15 +++++++++++++--
net/rxrpc/call_accept.c | 2 +-
net/rxrpc/call_object.c | 2 +-
net/rxrpc/conn_event.c | 2 +-
net/rxrpc/conn_object.c | 12 ++++++------
net/rxrpc/input.c | 2 +-
net/rxrpc/local_object.c | 2 +-
8 files changed, 26 insertions(+), 18 deletions(-)
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index b29bb50af5de..57dcbfc061e4 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -674,11 +674,8 @@ static int rxrpc_release_sock(struct sock *sk)
flush_workqueue(rxrpc_workqueue);
rxrpc_purge_queue(&sk->sk_receive_queue);
- if (rx->local) {
- rxrpc_put_local(rx->local);
- rx->local = NULL;
- }
-
+ rxrpc_put_local(rx->local);
+ rx->local = NULL;
key_put(rx->key);
rx->key = NULL;
key_put(rx->securities);
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index cfbd028aa551..c0ed5e7f22ef 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -597,6 +597,17 @@ static inline bool rxrpc_conn_is_service(const struct rxrpc_connection *conn)
return conn->proto.in_clientflag;
}
+static inline void rxrpc_get_connection(struct rxrpc_connection *conn)
+{
+ atomic_inc(&conn->usage);
+}
+
+static inline
+struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *conn)
+{
+ return atomic_inc_not_zero(&conn->usage) ? conn : NULL;
+}
+
/*
* input.c
*/
@@ -645,7 +656,7 @@ struct rxrpc_local *rxrpc_get_local_maybe(struct rxrpc_local *local)
static inline void rxrpc_put_local(struct rxrpc_local *local)
{
- if (atomic_dec_and_test(&local->usage))
+ if (local && atomic_dec_and_test(&local->usage))
__rxrpc_put_local(local);
}
@@ -702,7 +713,7 @@ struct rxrpc_peer *rxrpc_get_peer_maybe(struct rxrpc_peer *peer)
extern void __rxrpc_put_peer(struct rxrpc_peer *peer);
static inline void rxrpc_put_peer(struct rxrpc_peer *peer)
{
- if (atomic_dec_and_test(&peer->usage))
+ if (peer && atomic_dec_and_test(&peer->usage))
__rxrpc_put_peer(peer);
}
diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c
index 5a70dc4e28c6..833ad0622b61 100644
--- a/net/rxrpc/call_accept.c
+++ b/net/rxrpc/call_accept.c
@@ -141,7 +141,7 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local,
_debug("await conn sec");
list_add_tail(&call->accept_link, &rx->secureq);
call->conn->state = RXRPC_CONN_SERVER_CHALLENGING;
- atomic_inc(&call->conn->usage);
+ rxrpc_get_connection(call->conn);
set_bit(RXRPC_CONN_CHALLENGE, &call->conn->events);
rxrpc_queue_conn(call->conn);
} else {
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index d83f2cbb80a9..45849a66bc56 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -515,7 +515,7 @@ struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *rx,
rb_insert_color(&call->conn_node, &conn->calls);
conn->channels[call->channel] = call;
sock_hold(&rx->sk);
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
write_unlock_bh(&conn->lock);
spin_lock(&conn->params.peer->lock);
diff --git a/net/rxrpc/conn_event.c b/net/rxrpc/conn_event.c
index a022439f6f5a..bf6971555eac 100644
--- a/net/rxrpc/conn_event.c
+++ b/net/rxrpc/conn_event.c
@@ -263,7 +263,7 @@ void rxrpc_process_connection(struct work_struct *work)
_enter("{%d}", conn->debug_id);
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
if (test_and_clear_bit(RXRPC_CONN_CHALLENGE, &conn->events)) {
rxrpc_secure_connection(conn);
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index 312b75091d29..1754f2e2e16b 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -333,7 +333,7 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
* channel.
*/
chan = 0;
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
conn->avail_calls = RXRPC_MAXCALLS - 1;
conn->channels[chan] = call;
conn->call_counter = 1;
@@ -392,7 +392,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
conn->channels[1] == NULL ||
conn->channels[2] == NULL ||
conn->channels[3] == NULL);
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
break;
}
@@ -412,7 +412,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx,
conn->channels[1] == NULL &&
conn->channels[2] == NULL &&
conn->channels[3] == NULL);
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
list_move(&conn->bundle_link, &bundle->avail_conns);
break;
}
@@ -629,7 +629,7 @@ found_extant_connection:
read_unlock_bh(&trans->conn_lock);
goto security_mismatch;
}
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
read_unlock_bh(&trans->conn_lock);
goto success;
@@ -639,7 +639,7 @@ found_extant_second:
write_unlock_bh(&trans->conn_lock);
goto security_mismatch;
}
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
write_unlock_bh(&trans->conn_lock);
kfree(candidate);
goto success;
@@ -698,7 +698,7 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
return NULL;
found:
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
read_unlock_bh(&trans->conn_lock);
_leave(" = %p", conn);
return conn;
diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index cf540efa9c17..799aec18aa7b 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -580,7 +580,7 @@ static void rxrpc_post_packet_to_conn(struct rxrpc_connection *conn,
{
_enter("%p,%p", conn, skb);
- atomic_inc(&conn->usage);
+ rxrpc_get_connection(conn);
skb_queue_tail(&conn->rx_queue, skb);
rxrpc_queue_conn(conn);
}
diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c
index 009b321712bc..5703b0d18ed4 100644
--- a/net/rxrpc/local_object.c
+++ b/net/rxrpc/local_object.c
@@ -209,7 +209,7 @@ struct rxrpc_local *rxrpc_lookup_local(const struct sockaddr_rxrpc *srx)
* bind the transport socket may still fail if we're attempting
* to use a local address that the dying object is still using.
*/
- if (!atomic_inc_not_zero(&local->usage)) {
+ if (!rxrpc_get_local_maybe(local)) {
cursor = cursor->next;
list_del_init(&local->link);
break;
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 13/14] rxrpc: Kill the client connection bundle concept
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (11 preceding siblings ...)
2016-06-22 9:51 ` [PATCH net-next 12/14] rxrpc: Provide more refcount helper functions David Howells
@ 2016-06-22 9:51 ` David Howells
2016-06-22 9:51 ` [PATCH net-next 14/14] rxrpc: Kill off the rxrpc_transport struct David Howells
2016-06-22 10:03 ` [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:51 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
Kill off the concept of maintaining a bundle of connections to a particular
target service to increase the number of call slots available for any
beyond four for that service (there are four call slots per connection).
This will make cleaning up the connection handling code easier and
facilitate removal of the rxrpc_transport struct. Bundling can be
reintroduced later if necessary.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/af_rxrpc.c | 11 -
net/rxrpc/ar-internal.h | 54 ++---
net/rxrpc/call_object.c | 124 ++++++-----
net/rxrpc/conn_object.c | 518 +++++++++++++++-------------------------------
net/rxrpc/local_object.c | 3
net/rxrpc/output.c | 11 -
net/rxrpc/transport.c | 2
7 files changed, 260 insertions(+), 463 deletions(-)
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 57dcbfc061e4..f3b6ed8196c3 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -276,7 +276,6 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
gfp_t gfp)
{
struct rxrpc_conn_parameters cp;
- struct rxrpc_conn_bundle *bundle;
struct rxrpc_transport *trans;
struct rxrpc_call *call;
struct rxrpc_sock *rx = rxrpc_sk(sock->sk);
@@ -311,15 +310,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
}
cp.peer = trans->peer;
- bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, gfp);
- if (IS_ERR(bundle)) {
- call = ERR_CAST(bundle);
- goto out;
- }
-
- call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID, gfp);
- rxrpc_put_bundle(trans, bundle);
-out:
+ call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID, gfp);
rxrpc_put_transport(trans);
out_notrans:
release_sock(&rx->sk);
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index c0ed5e7f22ef..9a0bd153c71b 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -186,7 +186,8 @@ struct rxrpc_local {
struct sk_buff_head accept_queue; /* incoming calls awaiting acceptance */
struct sk_buff_head reject_queue; /* packets awaiting rejection */
struct sk_buff_head event_queue; /* endpoint event packets awaiting processing */
- struct mutex conn_lock; /* Client connection creation lock */
+ struct rb_root client_conns; /* Client connections by socket params */
+ spinlock_t client_conns_lock; /* Lock for client_conns */
spinlock_t lock; /* access lock */
rwlock_t services_lock; /* lock for services list */
int debug_id; /* debug ID for printks */
@@ -232,35 +233,15 @@ struct rxrpc_peer {
struct rxrpc_transport {
struct rxrpc_local *local; /* local transport endpoint */
struct rxrpc_peer *peer; /* remote transport endpoint */
- struct rb_root bundles; /* client connection bundles on this transport */
struct rb_root server_conns; /* server connections on this transport */
struct list_head link; /* link in master session list */
unsigned long put_time; /* time at which to reap */
- spinlock_t client_lock; /* client connection allocation lock */
rwlock_t conn_lock; /* lock for active/dead connections */
atomic_t usage;
int debug_id; /* debug ID for printks */
};
/*
- * RxRPC client connection bundle
- * - matched by { transport, service_id, key }
- */
-struct rxrpc_conn_bundle {
- struct rb_node node; /* node in transport's lookup tree */
- struct list_head unused_conns; /* unused connections in this bundle */
- struct list_head avail_conns; /* available connections in this bundle */
- struct list_head busy_conns; /* busy connections in this bundle */
- struct key *key; /* security for this bundle */
- wait_queue_head_t chanwait; /* wait for channel to become available */
- atomic_t usage;
- int debug_id; /* debug ID for printks */
- unsigned short num_conns; /* number of connections in this bundle */
- u16 service_id; /* Service ID for this bundle */
- u8 security_ix; /* security type */
-};
-
-/*
* Keys for matching a connection.
*/
struct rxrpc_conn_proto {
@@ -295,17 +276,21 @@ struct rxrpc_conn_parameters {
*/
struct rxrpc_connection {
struct rxrpc_transport *trans; /* transport session */
- struct rxrpc_conn_bundle *bundle; /* connection bundle (client) */
struct rxrpc_conn_proto proto;
struct rxrpc_conn_parameters params;
+ spinlock_t channel_lock;
+ struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* active calls */
+ wait_queue_head_t channel_wq; /* queue to wait for channel to become available */
+
struct work_struct processor; /* connection event processor */
- struct rb_node node; /* node in transport's lookup tree */
+ union {
+ struct rb_node client_node; /* Node in local->client_conns */
+ struct rb_node service_node; /* Node in trans->server_conns */
+ };
struct list_head link; /* link in master connection list */
- struct list_head bundle_link; /* link in bundle */
struct rb_root calls; /* calls on this connection */
struct sk_buff_head rx_queue; /* received conn-level packets */
- struct rxrpc_call *channels[RXRPC_MAXCALLS]; /* channels (active calls) */
const struct rxrpc_security *security; /* applied security module */
struct key *server_key; /* security for this service */
struct crypto_skcipher *cipher; /* encryption handle */
@@ -314,7 +299,7 @@ struct rxrpc_connection {
#define RXRPC_CONN_HAS_IDR 0 /* - Has a client conn ID assigned */
unsigned long events;
#define RXRPC_CONN_CHALLENGE 0 /* send challenge packet */
- unsigned long put_time; /* time at which to reap */
+ unsigned long put_time; /* Time at which last put */
rwlock_t lock; /* access lock */
spinlock_t state_lock; /* state-change lock */
atomic_t usage;
@@ -335,7 +320,7 @@ struct rxrpc_connection {
unsigned int call_counter; /* call ID counter */
atomic_t serial; /* packet serial number counter */
atomic_t hi_serial; /* highest serial number received */
- u8 avail_calls; /* number of calls available */
+ 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 */
@@ -386,6 +371,8 @@ enum rxrpc_call_event {
* The states that a call can be in.
*/
enum rxrpc_call_state {
+ RXRPC_CALL_UNINITIALISED,
+ RXRPC_CALL_CLIENT_AWAIT_CONN, /* - client waiting for connection to become available */
RXRPC_CALL_CLIENT_SEND_REQUEST, /* - client sending request phase */
RXRPC_CALL_CLIENT_AWAIT_REPLY, /* - client awaiting reply */
RXRPC_CALL_CLIENT_RECV_REPLY, /* - client receiving reply phase */
@@ -540,7 +527,7 @@ struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long
struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *,
struct rxrpc_conn_parameters *,
struct rxrpc_transport *,
- struct rxrpc_conn_bundle *,
+ struct sockaddr_rxrpc *,
unsigned long, gfp_t);
struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *,
struct rxrpc_connection *,
@@ -573,13 +560,10 @@ extern unsigned int rxrpc_connection_expiry;
extern struct list_head rxrpc_connections;
extern rwlock_t rxrpc_connection_lock;
-struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *,
- struct rxrpc_transport *,
- struct key *, u16, gfp_t);
-void rxrpc_put_bundle(struct rxrpc_transport *, struct rxrpc_conn_bundle *);
-int rxrpc_connect_call(struct rxrpc_sock *, struct rxrpc_conn_parameters *,
- struct rxrpc_transport *, struct rxrpc_conn_bundle *,
- struct rxrpc_call *, gfp_t);
+int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *,
+ struct rxrpc_transport *,
+ struct sockaddr_rxrpc *, gfp_t);
+void rxrpc_disconnect_call(struct rxrpc_call *);
void rxrpc_put_connection(struct rxrpc_connection *);
void __exit rxrpc_destroy_all_connections(void);
struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *,
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index 45849a66bc56..9b3b48abe12f 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -31,6 +31,8 @@ unsigned int rxrpc_max_call_lifetime = 60 * HZ;
unsigned int rxrpc_dead_call_expiry = 2 * HZ;
const char *const rxrpc_call_states[NR__RXRPC_CALL_STATES] = {
+ [RXRPC_CALL_UNINITIALISED] = "Uninit",
+ [RXRPC_CALL_CLIENT_AWAIT_CONN] = "ClWtConn",
[RXRPC_CALL_CLIENT_SEND_REQUEST] = "ClSndReq",
[RXRPC_CALL_CLIENT_AWAIT_REPLY] = "ClAwtRpl",
[RXRPC_CALL_CLIENT_RECV_REPLY] = "ClRcvRpl",
@@ -261,6 +263,7 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp)
(unsigned long) call);
INIT_WORK(&call->destroyer, &rxrpc_destroy_call);
INIT_WORK(&call->processor, &rxrpc_process_call);
+ INIT_LIST_HEAD(&call->link);
INIT_LIST_HEAD(&call->accept_link);
skb_queue_head_init(&call->rx_queue);
skb_queue_head_init(&call->rx_oos_queue);
@@ -269,7 +272,6 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp)
rwlock_init(&call->state_lock);
atomic_set(&call->usage, 1);
call->debug_id = atomic_inc_return(&rxrpc_debug_id);
- call->state = RXRPC_CALL_CLIENT_SEND_REQUEST;
memset(&call->sock_node, 0xed, sizeof(call->sock_node));
@@ -282,55 +284,70 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp)
}
/*
- * allocate a new client call and attempt to get a connection slot for it
+ * Allocate a new client call.
*/
static struct rxrpc_call *rxrpc_alloc_client_call(
struct rxrpc_sock *rx,
struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
- struct rxrpc_conn_bundle *bundle,
+ struct sockaddr_rxrpc *srx,
gfp_t gfp)
{
struct rxrpc_call *call;
- int ret;
_enter("");
- ASSERT(rx != NULL);
- ASSERT(trans != NULL);
- ASSERT(bundle != NULL);
+ ASSERT(rx->local != NULL);
call = rxrpc_alloc_call(gfp);
if (!call)
return ERR_PTR(-ENOMEM);
+ call->state = RXRPC_CALL_CLIENT_AWAIT_CONN;
sock_hold(&rx->sk);
call->socket = rx;
call->rx_data_post = 1;
- ret = rxrpc_connect_call(rx, cp, trans, bundle, call, gfp);
- if (ret < 0) {
- kmem_cache_free(rxrpc_call_jar, call);
- return ERR_PTR(ret);
- }
-
/* Record copies of information for hashtable lookup */
call->family = rx->family;
- call->local = call->conn->params.local;
+ call->local = rx->local;
switch (call->family) {
case AF_INET:
- call->peer_ip.ipv4_addr =
- call->conn->params.peer->srx.transport.sin.sin_addr.s_addr;
+ call->peer_ip.ipv4_addr = srx->transport.sin.sin_addr.s_addr;
break;
case AF_INET6:
memcpy(call->peer_ip.ipv6_addr,
- call->conn->params.peer->srx.transport.sin6.sin6_addr.in6_u.u6_addr8,
+ srx->transport.sin6.sin6_addr.in6_u.u6_addr8,
sizeof(call->peer_ip.ipv6_addr));
break;
}
- call->epoch = call->conn->proto.epoch;
- call->service_id = call->conn->params.service_id;
- call->in_clientflag = call->conn->proto.in_clientflag;
+
+ call->service_id = srx->srx_service;
+ call->in_clientflag = 0;
+
+ _leave(" = %p", call);
+ return call;
+}
+
+/*
+ * Begin client call.
+ */
+static int rxrpc_begin_client_call(struct rxrpc_call *call,
+ struct rxrpc_conn_parameters *cp,
+ struct rxrpc_transport *trans,
+ struct sockaddr_rxrpc *srx,
+ gfp_t gfp)
+{
+ int ret;
+
+ /* Set up or get a connection record and set the protocol parameters,
+ * including channel number and call ID.
+ */
+ ret = rxrpc_connect_call(call, cp, trans, srx, gfp);
+ if (ret < 0)
+ return ret;
+
+ call->state = RXRPC_CALL_CLIENT_SEND_REQUEST;
+
/* Add the new call to the hashtable */
rxrpc_call_hash_add(call);
@@ -340,9 +357,7 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
call->lifetimer.expires = jiffies + rxrpc_max_call_lifetime;
add_timer(&call->lifetimer);
-
- _leave(" = %p", call);
- return call;
+ return 0;
}
/*
@@ -352,23 +367,23 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
struct rxrpc_conn_parameters *cp,
struct rxrpc_transport *trans,
- struct rxrpc_conn_bundle *bundle,
+ struct sockaddr_rxrpc *srx,
unsigned long user_call_ID,
gfp_t gfp)
{
struct rxrpc_call *call, *xcall;
struct rb_node *parent, **pp;
+ int ret;
- _enter("%p,%d,%d,%lx",
- rx, trans->debug_id, bundle ? bundle->debug_id : -1,
- user_call_ID);
+ _enter("%p,%lx", rx, user_call_ID);
- call = rxrpc_alloc_client_call(rx, cp, trans, bundle, gfp);
+ call = rxrpc_alloc_client_call(rx, cp, srx, gfp);
if (IS_ERR(call)) {
_leave(" = %ld", PTR_ERR(call));
return call;
}
+ /* Publish the call, even though it is incompletely set up as yet */
call->user_call_ID = user_call_ID;
__set_bit(RXRPC_CALL_HAS_USERID, &call->flags);
@@ -398,11 +413,29 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
list_add_tail(&call->link, &rxrpc_calls);
write_unlock_bh(&rxrpc_call_lock);
+ ret = rxrpc_begin_client_call(call, cp, trans, srx, gfp);
+ if (ret < 0)
+ goto error;
+
_net("CALL new %d on CONN %d", call->debug_id, call->conn->debug_id);
_leave(" = %p [new]", call);
return call;
+error:
+ write_lock(&rx->call_lock);
+ rb_erase(&call->sock_node, &rx->calls);
+ write_unlock(&rx->call_lock);
+ rxrpc_put_call(call);
+
+ write_lock_bh(&rxrpc_call_lock);
+ list_del(&call->link);
+ write_unlock_bh(&rxrpc_call_lock);
+
+ rxrpc_put_call(call);
+ _leave(" = %d", ret);
+ return ERR_PTR(ret);
+
/* We unexpectedly found the user ID in the list after taking
* the call_lock. This shouldn't happen unless the user races
* with itself and tries to add the same user ID twice at the
@@ -612,40 +645,13 @@ void rxrpc_release_call(struct rxrpc_call *call)
write_unlock_bh(&rx->call_lock);
/* free up the channel for reuse */
- spin_lock(&conn->trans->client_lock);
+ spin_lock(&conn->channel_lock);
write_lock_bh(&conn->lock);
write_lock(&call->state_lock);
- if (conn->channels[call->channel] == call)
- conn->channels[call->channel] = NULL;
-
- if (conn->out_clientflag && conn->bundle) {
- conn->avail_calls++;
- switch (conn->avail_calls) {
- case 1:
- list_move_tail(&conn->bundle_link,
- &conn->bundle->avail_conns);
- case 2 ... RXRPC_MAXCALLS - 1:
- ASSERT(conn->channels[0] == NULL ||
- conn->channels[1] == NULL ||
- conn->channels[2] == NULL ||
- conn->channels[3] == NULL);
- break;
- case RXRPC_MAXCALLS:
- list_move_tail(&conn->bundle_link,
- &conn->bundle->unused_conns);
- ASSERT(conn->channels[0] == NULL &&
- conn->channels[1] == NULL &&
- conn->channels[2] == NULL &&
- conn->channels[3] == NULL);
- break;
- default:
- pr_err("conn->avail_calls=%d\n", conn->avail_calls);
- BUG();
- }
- }
+ rxrpc_disconnect_call(call);
- spin_unlock(&conn->trans->client_lock);
+ spin_unlock(&conn->channel_lock);
if (call->state < RXRPC_CALL_COMPLETE &&
call->state != RXRPC_CALL_CLIENT_FINAL_ACK) {
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index 1754f2e2e16b..5b27241ebea6 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -32,152 +32,6 @@ DEFINE_RWLOCK(rxrpc_connection_lock);
static DECLARE_DELAYED_WORK(rxrpc_connection_reap, rxrpc_connection_reaper);
/*
- * allocate a new client connection bundle
- */
-static struct rxrpc_conn_bundle *rxrpc_alloc_bundle(gfp_t gfp)
-{
- struct rxrpc_conn_bundle *bundle;
-
- _enter("");
-
- bundle = kzalloc(sizeof(struct rxrpc_conn_bundle), gfp);
- if (bundle) {
- INIT_LIST_HEAD(&bundle->unused_conns);
- INIT_LIST_HEAD(&bundle->avail_conns);
- INIT_LIST_HEAD(&bundle->busy_conns);
- init_waitqueue_head(&bundle->chanwait);
- atomic_set(&bundle->usage, 1);
- }
-
- _leave(" = %p", bundle);
- return bundle;
-}
-
-/*
- * compare bundle parameters with what we're looking for
- * - return -ve, 0 or +ve
- */
-static inline
-int rxrpc_cmp_bundle(const struct rxrpc_conn_bundle *bundle,
- struct key *key, u16 service_id)
-{
- return (bundle->service_id - service_id) ?:
- ((unsigned long)bundle->key - (unsigned long)key);
-}
-
-/*
- * get bundle of client connections that a client socket can make use of
- */
-struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx,
- struct rxrpc_transport *trans,
- struct key *key,
- u16 service_id,
- gfp_t gfp)
-{
- struct rxrpc_conn_bundle *bundle, *candidate;
- struct rb_node *p, *parent, **pp;
-
- _enter("%p{%x},%x,%hx,",
- rx, key_serial(key), trans->debug_id, service_id);
-
- /* search the extant bundles first for one that matches the specified
- * user ID */
- spin_lock(&trans->client_lock);
-
- p = trans->bundles.rb_node;
- while (p) {
- bundle = rb_entry(p, struct rxrpc_conn_bundle, node);
-
- if (rxrpc_cmp_bundle(bundle, key, service_id) < 0)
- p = p->rb_left;
- else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0)
- p = p->rb_right;
- else
- goto found_extant_bundle;
- }
-
- spin_unlock(&trans->client_lock);
-
- /* not yet present - create a candidate for a new record and then
- * redo the search */
- candidate = rxrpc_alloc_bundle(gfp);
- if (!candidate) {
- _leave(" = -ENOMEM");
- return ERR_PTR(-ENOMEM);
- }
-
- candidate->key = key_get(key);
- candidate->service_id = service_id;
-
- spin_lock(&trans->client_lock);
-
- pp = &trans->bundles.rb_node;
- parent = NULL;
- while (*pp) {
- parent = *pp;
- bundle = rb_entry(parent, struct rxrpc_conn_bundle, node);
-
- if (rxrpc_cmp_bundle(bundle, key, service_id) < 0)
- pp = &(*pp)->rb_left;
- else if (rxrpc_cmp_bundle(bundle, key, service_id) > 0)
- pp = &(*pp)->rb_right;
- else
- goto found_extant_second;
- }
-
- /* second search also failed; add the new bundle */
- bundle = candidate;
- candidate = NULL;
-
- rb_link_node(&bundle->node, parent, pp);
- rb_insert_color(&bundle->node, &trans->bundles);
- spin_unlock(&trans->client_lock);
- _net("BUNDLE new on trans %d", trans->debug_id);
- _leave(" = %p [new]", bundle);
- return bundle;
-
- /* we found the bundle in the list immediately */
-found_extant_bundle:
- atomic_inc(&bundle->usage);
- spin_unlock(&trans->client_lock);
- _net("BUNDLE old on trans %d", trans->debug_id);
- _leave(" = %p [extant %d]", bundle, atomic_read(&bundle->usage));
- return bundle;
-
- /* we found the bundle on the second time through the list */
-found_extant_second:
- atomic_inc(&bundle->usage);
- spin_unlock(&trans->client_lock);
- kfree(candidate);
- _net("BUNDLE old2 on trans %d", trans->debug_id);
- _leave(" = %p [second %d]", bundle, atomic_read(&bundle->usage));
- return bundle;
-}
-
-/*
- * release a bundle
- */
-void rxrpc_put_bundle(struct rxrpc_transport *trans,
- struct rxrpc_conn_bundle *bundle)
-{
- _enter("%p,%p{%d}",trans, bundle, atomic_read(&bundle->usage));
-
- if (atomic_dec_and_lock(&bundle->usage, &trans->client_lock)) {
- _debug("Destroy bundle");
- rb_erase(&bundle->node, &trans->bundles);
- spin_unlock(&trans->client_lock);
- ASSERT(list_empty(&bundle->unused_conns));
- ASSERT(list_empty(&bundle->avail_conns));
- ASSERT(list_empty(&bundle->busy_conns));
- ASSERTCMP(bundle->num_conns, ==, 0);
- key_put(bundle->key);
- kfree(bundle);
- }
-
- _leave("");
-}
-
-/*
* allocate a new connection
*/
static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp)
@@ -188,8 +42,10 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp)
conn = kzalloc(sizeof(struct rxrpc_connection), gfp);
if (conn) {
+ spin_lock_init(&conn->channel_lock);
+ init_waitqueue_head(&conn->channel_wq);
INIT_WORK(&conn->processor, &rxrpc_process_connection);
- INIT_LIST_HEAD(&conn->bundle_link);
+ INIT_LIST_HEAD(&conn->link);
conn->calls = RB_ROOT;
skb_queue_head_init(&conn->rx_queue);
conn->security = &rxrpc_no_security;
@@ -197,7 +53,7 @@ static struct rxrpc_connection *rxrpc_alloc_connection(gfp_t gfp)
spin_lock_init(&conn->state_lock);
atomic_set(&conn->usage, 1);
conn->debug_id = atomic_inc_return(&rxrpc_debug_id);
- conn->avail_calls = RXRPC_MAXCALLS;
+ atomic_set(&conn->avail_chans, RXRPC_MAXCALLS);
conn->size_align = 4;
conn->header_size = sizeof(struct rxrpc_wire_header);
}
@@ -240,7 +96,8 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn,
}
/*
- * Allocate a client connection.
+ * Allocate a client connection. The caller must take care to clear any
+ * padding bytes in *cp.
*/
static struct rxrpc_connection *
rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp,
@@ -290,6 +147,8 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp,
write_unlock(&rxrpc_connection_lock);
key_get(conn->params.key);
+ conn->trans = trans;
+ atomic_inc(&trans->usage);
_leave(" = %p", conn);
return conn;
@@ -303,217 +162,172 @@ error_0:
}
/*
- * connect a call on an exclusive connection
- */
-static int rxrpc_connect_exclusive(struct rxrpc_sock *rx,
- struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
- struct rxrpc_call *call,
- gfp_t gfp)
-{
- struct rxrpc_connection *conn;
- int chan;
-
- _enter("");
-
- conn = rxrpc_alloc_client_connection(cp, trans, gfp);
- if (IS_ERR(conn)) {
- _leave(" = %ld", PTR_ERR(conn));
- return PTR_ERR(conn);
- }
-
- atomic_inc(&trans->usage);
- conn->trans = trans;
- conn->bundle = NULL;
-
- _net("CONNECT EXCL new %d on TRANS %d",
- conn->debug_id, conn->trans->debug_id);
-
- /* Since no one else can use the connection, we just use the first
- * channel.
- */
- chan = 0;
- rxrpc_get_connection(conn);
- conn->avail_calls = RXRPC_MAXCALLS - 1;
- conn->channels[chan] = call;
- conn->call_counter = 1;
- call->conn = conn;
- call->channel = chan;
- call->cid = conn->proto.cid | chan;
- call->call_id = 1;
-
- _net("CONNECT client on conn %d chan %d as call %x",
- conn->debug_id, chan, call->call_id);
-
- rxrpc_add_call_ID_to_conn(conn, call);
- _leave(" = 0");
- return 0;
-}
-
-/*
* find a connection for a call
* - called in process context with IRQs enabled
*/
-int rxrpc_connect_call(struct rxrpc_sock *rx,
+int rxrpc_connect_call(struct rxrpc_call *call,
struct rxrpc_conn_parameters *cp,
struct rxrpc_transport *trans,
- struct rxrpc_conn_bundle *bundle,
- struct rxrpc_call *call,
+ struct sockaddr_rxrpc *srx,
gfp_t gfp)
{
- struct rxrpc_connection *conn, *candidate;
+ struct rxrpc_connection *conn, *candidate = NULL;
+ struct rxrpc_local *local = cp->local;
+ struct rb_node *p, **pp, *parent;
+ long diff;
int chan;
DECLARE_WAITQUEUE(myself, current);
- _enter("%p,%lx,", rx, call->user_call_ID);
-
- if (cp->exclusive)
- return rxrpc_connect_exclusive(rx, cp, trans, call, gfp);
-
- spin_lock(&trans->client_lock);
- for (;;) {
- /* see if the bundle has a call slot available */
- if (!list_empty(&bundle->avail_conns)) {
- _debug("avail");
- conn = list_entry(bundle->avail_conns.next,
- struct rxrpc_connection,
- bundle_link);
- if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) {
- list_del_init(&conn->bundle_link);
- bundle->num_conns--;
- continue;
- }
- if (--conn->avail_calls == 0)
- list_move(&conn->bundle_link,
- &bundle->busy_conns);
- ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS);
- ASSERT(conn->channels[0] == NULL ||
- conn->channels[1] == NULL ||
- conn->channels[2] == NULL ||
- conn->channels[3] == NULL);
- rxrpc_get_connection(conn);
- break;
- }
+ _enter("{%d,%lx},", call->debug_id, call->user_call_ID);
- if (!list_empty(&bundle->unused_conns)) {
- _debug("unused");
- conn = list_entry(bundle->unused_conns.next,
- struct rxrpc_connection,
- bundle_link);
- if (conn->state >= RXRPC_CONN_REMOTELY_ABORTED) {
- list_del_init(&conn->bundle_link);
- bundle->num_conns--;
- continue;
- }
- ASSERTCMP(conn->avail_calls, ==, RXRPC_MAXCALLS);
- conn->avail_calls = RXRPC_MAXCALLS - 1;
- ASSERT(conn->channels[0] == NULL &&
- conn->channels[1] == NULL &&
- conn->channels[2] == NULL &&
- conn->channels[3] == NULL);
- rxrpc_get_connection(conn);
- list_move(&conn->bundle_link, &bundle->avail_conns);
- break;
- }
+ cp->peer = trans->peer;
+ rxrpc_get_peer(cp->peer);
- /* need to allocate a new connection */
- _debug("get new conn [%d]", bundle->num_conns);
-
- spin_unlock(&trans->client_lock);
-
- if (signal_pending(current))
- goto interrupted;
-
- if (bundle->num_conns >= 20) {
- _debug("too many conns");
-
- if (!gfpflags_allow_blocking(gfp)) {
- _leave(" = -EAGAIN");
- return -EAGAIN;
- }
-
- add_wait_queue(&bundle->chanwait, &myself);
- for (;;) {
- set_current_state(TASK_INTERRUPTIBLE);
- if (bundle->num_conns < 20 ||
- !list_empty(&bundle->unused_conns) ||
- !list_empty(&bundle->avail_conns))
- break;
- if (signal_pending(current))
- goto interrupted_dequeue;
- schedule();
- }
- remove_wait_queue(&bundle->chanwait, &myself);
- __set_current_state(TASK_RUNNING);
- spin_lock(&trans->client_lock);
- continue;
- }
+ if (!cp->exclusive) {
+ /* Search for a existing client connection unless this is going
+ * to be a connection that's used exclusively for a single call.
+ */
+ _debug("search 1");
+ spin_lock(&local->client_conns_lock);
+ p = local->client_conns.rb_node;
+ while (p) {
+ conn = rb_entry(p, struct rxrpc_connection, client_node);
- /* not yet present - create a candidate for a new connection and then
- * redo the check */
- candidate = rxrpc_alloc_client_connection(cp, trans, gfp);
- if (!candidate) {
- _leave(" = -ENOMEM");
- return -ENOMEM;
+#define cmp(X) ((long)conn->params.X - (long)cp->X)
+ diff = (cmp(peer) ?:
+ cmp(key) ?:
+ cmp(security_level));
+ if (diff < 0)
+ p = p->rb_left;
+ else if (diff > 0)
+ p = p->rb_right;
+ else
+ goto found_extant_conn;
}
+ spin_unlock(&local->client_conns_lock);
+ }
- atomic_inc(&bundle->usage);
- atomic_inc(&trans->usage);
- candidate->trans = trans;
- candidate->bundle = bundle;
+ /* We didn't find a connection or we want an exclusive one. */
+ _debug("get new conn");
+ candidate = rxrpc_alloc_client_connection(cp, trans, gfp);
+ if (!candidate) {
+ _leave(" = -ENOMEM");
+ return -ENOMEM;
+ }
- spin_lock(&trans->client_lock);
+ if (cp->exclusive) {
+ /* Assign the call on an exclusive connection to channel 0 and
+ * don't add the connection to the endpoint's shareable conn
+ * lookup tree.
+ */
+ _debug("exclusive chan 0");
+ atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1);
+ spin_lock(&conn->channel_lock);
+ chan = 0;
+ goto found_channel;
+ }
- list_add(&candidate->bundle_link, &bundle->unused_conns);
- bundle->num_conns++;
+ /* We need to redo the search before attempting to add a new connection
+ * lest we race with someone else adding a conflicting instance.
+ */
+ _debug("search 2");
+ spin_lock(&local->client_conns_lock);
- _net("CONNECT new %d on TRANS %d",
- candidate->debug_id, candidate->trans->debug_id);
+ pp = &local->client_conns.rb_node;
+ parent = NULL;
+ while (*pp) {
+ parent = *pp;
+ conn = rb_entry(parent, struct rxrpc_connection, client_node);
- /* leave the candidate lurking in zombie mode attached to the
- * bundle until we're ready for it */
- rxrpc_put_connection(candidate);
- candidate = NULL;
+ diff = (cmp(peer) ?:
+ cmp(key) ?:
+ cmp(security_level));
+ if (diff < 0)
+ pp = &(*pp)->rb_left;
+ else if (diff > 0)
+ pp = &(*pp)->rb_right;
+ else
+ goto found_extant_conn;
}
- /* we've got a connection with a free channel and we can now attach the
- * call to it
- * - we're holding the transport's client lock
- * - we're holding a reference on the connection
- * - we're holding a reference on the bundle
+ /* The second search also failed; simply add the new connection with
+ * the new call in channel 0. Note that we need to take the channel
+ * lock before dropping the client conn lock.
*/
- for (chan = 0; chan < RXRPC_MAXCALLS; chan++)
- if (!conn->channels[chan])
- goto found_channel;
- ASSERT(conn->channels[0] == NULL ||
- conn->channels[1] == NULL ||
- conn->channels[2] == NULL ||
- conn->channels[3] == NULL);
- BUG();
+ _debug("new conn");
+ conn = candidate;
+ candidate = NULL;
-found_channel:
- conn->channels[chan] = call;
- call->conn = conn;
- call->channel = chan;
- call->cid = conn->proto.cid | chan;
- call->call_id = ++conn->call_counter;
+ rb_link_node(&conn->client_node, parent, pp);
+ rb_insert_color(&conn->client_node, &local->client_conns);
+
+ atomic_set(&conn->avail_chans, RXRPC_MAXCALLS - 1);
+ spin_lock(&conn->channel_lock);
+ spin_unlock(&local->client_conns_lock);
+ chan = 0;
- _net("CONNECT client on conn %d chan %d as call %x",
- conn->debug_id, chan, call->call_id);
+found_channel:
+ _debug("found chan");
+ call->conn = conn;
+ call->channel = chan;
+ call->epoch = conn->proto.epoch;
+ call->cid = conn->proto.cid | chan;
+ call->call_id = ++conn->call_counter;
+ rcu_assign_pointer(conn->channels[chan], call);
- ASSERTCMP(conn->avail_calls, <, RXRPC_MAXCALLS);
- spin_unlock(&trans->client_lock);
+ _net("CONNECT call %d on conn %d", call->debug_id, conn->debug_id);
rxrpc_add_call_ID_to_conn(conn, call);
-
- _leave(" = 0");
+ spin_unlock(&conn->channel_lock);
+ _leave(" = %p {u=%d}", conn, atomic_read(&conn->usage));
return 0;
-interrupted_dequeue:
- remove_wait_queue(&bundle->chanwait, &myself);
- __set_current_state(TASK_RUNNING);
+ /* We found a suitable connection already in existence. Discard any
+ * candidate we may have allocated, and try to get a channel on this
+ * one.
+ */
+found_extant_conn:
+ _debug("found conn");
+ rxrpc_get_connection(conn);
+ spin_unlock(&local->client_conns_lock);
+
+ rxrpc_put_connection(candidate);
+
+ if (!atomic_add_unless(&conn->avail_chans, -1, 0)) {
+ if (!gfpflags_allow_blocking(gfp)) {
+ rxrpc_put_connection(conn);
+ _leave(" = -EAGAIN");
+ return -EAGAIN;
+ }
+
+ add_wait_queue(&conn->channel_wq, &myself);
+ for (;;) {
+ set_current_state(TASK_INTERRUPTIBLE);
+ if (atomic_add_unless(&conn->avail_chans, -1, 0))
+ break;
+ if (signal_pending(current))
+ goto interrupted;
+ schedule();
+ }
+ remove_wait_queue(&conn->channel_wq, &myself);
+ __set_current_state(TASK_RUNNING);
+ }
+
+ /* The connection allegedly now has a free channel and we can now
+ * attach the call to it.
+ */
+ spin_lock(&conn->channel_lock);
+
+ for (chan = 0; chan < RXRPC_MAXCALLS; chan++)
+ if (!conn->channels[chan])
+ goto found_channel;
+ BUG();
+
interrupted:
+ remove_wait_queue(&conn->channel_wq, &myself);
+ __set_current_state(TASK_RUNNING);
+ rxrpc_put_connection(conn);
_leave(" = -ERESTARTSYS");
return -ERESTARTSYS;
}
@@ -521,8 +335,8 @@ interrupted:
/*
* get a record of an incoming connection
*/
-struct rxrpc_connection *
-rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb)
+struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans,
+ struct sk_buff *skb)
{
struct rxrpc_connection *conn, *candidate = NULL;
struct rxrpc_skb_priv *sp = rxrpc_skb(skb);
@@ -543,7 +357,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb)
p = trans->server_conns.rb_node;
while (p) {
- conn = rb_entry(p, struct rxrpc_connection, node);
+ conn = rb_entry(p, struct rxrpc_connection, service_node);
_debug("maybe %x", conn->proto.cid);
@@ -588,7 +402,7 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb)
p = NULL;
while (*pp) {
p = *pp;
- conn = rb_entry(p, struct rxrpc_connection, node);
+ conn = rb_entry(p, struct rxrpc_connection, service_node);
if (epoch < conn->proto.epoch)
pp = &(*pp)->rb_left;
@@ -605,8 +419,8 @@ rxrpc_incoming_connection(struct rxrpc_transport *trans, struct sk_buff *skb)
/* we can now add the new candidate to the list */
conn = candidate;
candidate = NULL;
- rb_link_node(&conn->node, p, pp);
- rb_insert_color(&conn->node, &trans->server_conns);
+ rb_link_node(&conn->service_node, p, pp);
+ rb_insert_color(&conn->service_node, &trans->server_conns);
atomic_inc(&conn->trans->usage);
write_unlock_bh(&trans->conn_lock);
@@ -672,7 +486,7 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) {
p = trans->server_conns.rb_node;
while (p) {
- conn = rb_entry(p, struct rxrpc_connection, node);
+ conn = rb_entry(p, struct rxrpc_connection, service_node);
_debug("maybe %x", conn->proto.cid);
@@ -705,10 +519,31 @@ found:
}
/*
+ * Disconnect a call and clear any channel it occupies when that call
+ * terminates.
+ */
+void rxrpc_disconnect_call(struct rxrpc_call *call)
+{
+ struct rxrpc_connection *conn = call->conn;
+ unsigned chan = call->channel;
+
+ _enter("%d,%d", conn->debug_id, call->channel);
+
+ if (conn->channels[chan] == call) {
+ rcu_assign_pointer(conn->channels[chan], NULL);
+ atomic_inc(&conn->avail_chans);
+ wake_up(&conn->channel_wq);
+ }
+}
+
+/*
* release a virtual connection
*/
void rxrpc_put_connection(struct rxrpc_connection *conn)
{
+ if (!conn)
+ return;
+
_enter("%p{u=%d,d=%d}",
conn, atomic_read(&conn->usage), conn->debug_id);
@@ -734,9 +569,6 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn)
_net("DESTROY CONN %d", conn->debug_id);
- if (conn->bundle)
- rxrpc_put_bundle(conn->trans, conn->bundle);
-
ASSERT(RB_EMPTY_ROOT(&conn->calls));
rxrpc_purge_queue(&conn->rx_queue);
@@ -773,7 +605,6 @@ static void rxrpc_connection_reaper(struct work_struct *work)
if (likely(atomic_read(&conn->usage) > 0))
continue;
- spin_lock(&conn->trans->client_lock);
write_lock_bh(&conn->trans->conn_lock);
reap_time = conn->put_time + rxrpc_connection_expiry;
@@ -784,19 +615,14 @@ static void rxrpc_connection_reaper(struct work_struct *work)
if (conn->out_clientflag)
rxrpc_put_client_connection_id(conn);
else
- rb_erase(&conn->node,
+ rb_erase(&conn->service_node,
&conn->trans->server_conns);
- if (conn->bundle) {
- list_del_init(&conn->bundle_link);
- conn->bundle->num_conns--;
- }
} else if (reap_time < earliest) {
earliest = reap_time;
}
write_unlock_bh(&conn->trans->conn_lock);
- spin_unlock(&conn->trans->client_lock);
}
write_unlock(&rxrpc_connection_lock);
diff --git a/net/rxrpc/local_object.c b/net/rxrpc/local_object.c
index 5703b0d18ed4..284a14c30b9b 100644
--- a/net/rxrpc/local_object.c
+++ b/net/rxrpc/local_object.c
@@ -80,7 +80,8 @@ static struct rxrpc_local *rxrpc_alloc_local(const struct sockaddr_rxrpc *srx)
skb_queue_head_init(&local->accept_queue);
skb_queue_head_init(&local->reject_queue);
skb_queue_head_init(&local->event_queue);
- mutex_init(&local->conn_lock);
+ local->client_conns = RB_ROOT;
+ spin_lock_init(&local->client_conns_lock);
spin_lock_init(&local->lock);
rwlock_init(&local->services_lock);
local->debug_id = atomic_inc_return(&rxrpc_debug_id);
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index db3933cf6b97..8e24939aeac8 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -140,7 +140,6 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
unsigned long user_call_ID, bool exclusive)
{
struct rxrpc_conn_parameters cp;
- struct rxrpc_conn_bundle *bundle;
struct rxrpc_transport *trans;
struct rxrpc_call *call;
struct key *key;
@@ -171,16 +170,8 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
}
cp.peer = trans->peer;
- bundle = rxrpc_get_bundle(rx, trans, cp.key, srx->srx_service,
- GFP_KERNEL);
- if (IS_ERR(bundle)) {
- ret = PTR_ERR(bundle);
- goto out_trans;
- }
-
- call = rxrpc_new_client_call(rx, &cp, trans, bundle, user_call_ID,
+ call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID,
GFP_KERNEL);
- rxrpc_put_bundle(trans, bundle);
rxrpc_put_transport(trans);
if (IS_ERR(call)) {
ret = PTR_ERR(call);
diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c
index 140628d94bb0..71947402d071 100644
--- a/net/rxrpc/transport.c
+++ b/net/rxrpc/transport.c
@@ -46,9 +46,7 @@ static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local,
trans->local = local;
trans->peer = peer;
INIT_LIST_HEAD(&trans->link);
- trans->bundles = RB_ROOT;
trans->server_conns = RB_ROOT;
- spin_lock_init(&trans->client_lock);
rwlock_init(&trans->conn_lock);
atomic_set(&trans->usage, 1);
trans->debug_id = atomic_inc_return(&rxrpc_debug_id);
^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH net-next 14/14] rxrpc: Kill off the rxrpc_transport struct
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (12 preceding siblings ...)
2016-06-22 9:51 ` [PATCH net-next 13/14] rxrpc: Kill the client connection bundle concept David Howells
@ 2016-06-22 9:51 ` David Howells
2016-06-22 10:03 ` [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 9:51 UTC (permalink / raw)
To: davem; +Cc: dhowells, netdev, linux-afs, linux-kernel
The rxrpc_transport struct is now redundant, given that the rxrpc_peer
struct is now per peer port rather than per peer host, so get rid of it.
Service connection lists are transferred to the rxrpc_peer struct, as is
the conn_lock. Previous patches moved the client connection handling out
of the rxrpc_transport struct and discarded the connection bundling code.
Signed-off-by: David Howells <dhowells@redhat.com>
---
net/rxrpc/Makefile | 1
net/rxrpc/af_rxrpc.c | 46 --------
net/rxrpc/ar-internal.h | 59 ++--------
net/rxrpc/call_accept.c | 11 --
net/rxrpc/call_object.c | 17 +--
net/rxrpc/conn_client.c | 8 +
net/rxrpc/conn_object.c | 80 ++++++++------
net/rxrpc/input.c | 8 -
net/rxrpc/output.c | 24 ----
net/rxrpc/peer_object.c | 2
net/rxrpc/sysctl.c | 8 -
net/rxrpc/transport.c | 265 -----------------------------------------------
12 files changed, 71 insertions(+), 458 deletions(-)
delete mode 100644 net/rxrpc/transport.c
diff --git a/net/rxrpc/Makefile b/net/rxrpc/Makefile
index cfa221536f33..6522e50fb750 100644
--- a/net/rxrpc/Makefile
+++ b/net/rxrpc/Makefile
@@ -22,7 +22,6 @@ af-rxrpc-y := \
recvmsg.o \
security.o \
skbuff.o \
- transport.o \
utils.o
af-rxrpc-$(CONFIG_PROC_FS) += proc.o
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index f3b6ed8196c3..5d3e795a7c48 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -224,37 +224,6 @@ static int rxrpc_listen(struct socket *sock, int backlog)
return ret;
}
-/*
- * find a transport by address
- */
-struct rxrpc_transport *
-rxrpc_name_to_transport(struct rxrpc_conn_parameters *cp,
- struct sockaddr *addr,
- int addr_len,
- gfp_t gfp)
-{
- struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr;
- struct rxrpc_transport *trans;
-
- _enter("%p,%d", addr, addr_len);
-
- if (cp->local->srx.transport_type != srx->transport_type)
- return ERR_PTR(-ESOCKTNOSUPPORT);
- if (cp->local->srx.transport.family != srx->transport.family)
- return ERR_PTR(-EAFNOSUPPORT);
-
- /* find a remote transport endpoint from the local one */
- cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp);
- if (!cp->peer)
- return ERR_PTR(-ENOMEM);
-
- /* find a transport */
- trans = rxrpc_get_transport(cp->local, cp->peer, gfp);
- rxrpc_put_peer(cp->peer);
- _leave(" = %p", trans);
- return trans;
-}
-
/**
* rxrpc_kernel_begin_call - Allow a kernel service to begin a call
* @sock: The socket on which to make the call
@@ -276,7 +245,6 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
gfp_t gfp)
{
struct rxrpc_conn_parameters cp;
- struct rxrpc_transport *trans;
struct rxrpc_call *call;
struct rxrpc_sock *rx = rxrpc_sk(sock->sk);
int ret;
@@ -300,19 +268,8 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock,
cp.security_level = 0;
cp.exclusive = false;
cp.service_id = srx->srx_service;
+ call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, gfp);
- trans = rxrpc_name_to_transport(&cp, (struct sockaddr *)srx,
- sizeof(*srx), gfp);
- if (IS_ERR(trans)) {
- call = ERR_CAST(trans);
- trans = NULL;
- goto out_notrans;
- }
- cp.peer = trans->peer;
-
- call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID, gfp);
- rxrpc_put_transport(trans);
-out_notrans:
release_sock(&rx->sk);
_leave(" = %p", call);
return call;
@@ -831,7 +788,6 @@ static void __exit af_rxrpc_exit(void)
proto_unregister(&rxrpc_proto);
rxrpc_destroy_all_calls();
rxrpc_destroy_all_connections();
- rxrpc_destroy_all_transports();
ASSERTCMP(atomic_read(&rxrpc_n_skbs), ==, 0);
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 9a0bd153c71b..a8a293bacaea 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -207,6 +207,8 @@ struct rxrpc_peer {
struct rxrpc_local *local;
struct hlist_head error_targets; /* targets for net error distribution */
struct work_struct error_distributor;
+ struct rb_root service_conns; /* Service connections */
+ rwlock_t conn_lock;
spinlock_t lock; /* access lock */
unsigned int if_mtu; /* interface MTU for this peer */
unsigned int mtu; /* network MTU for this peer */
@@ -226,22 +228,6 @@ struct rxrpc_peer {
};
/*
- * RxRPC point-to-point transport / connection manager definition
- * - handles a bundle of connections between two endpoints
- * - matched by { local, peer }
- */
-struct rxrpc_transport {
- struct rxrpc_local *local; /* local transport endpoint */
- struct rxrpc_peer *peer; /* remote transport endpoint */
- struct rb_root server_conns; /* server connections on this transport */
- struct list_head link; /* link in master session list */
- unsigned long put_time; /* time at which to reap */
- rwlock_t conn_lock; /* lock for active/dead connections */
- atomic_t usage;
- int debug_id; /* debug ID for printks */
-};
-
-/*
* Keys for matching a connection.
*/
struct rxrpc_conn_proto {
@@ -271,11 +257,10 @@ struct rxrpc_conn_parameters {
/*
* RxRPC connection definition
- * - matched by { transport, service_id, conn_id, direction, key }
+ * - matched by { local, peer, epoch, conn_id, direction }
* - each connection can only handle four simultaneous calls
*/
struct rxrpc_connection {
- struct rxrpc_transport *trans; /* transport session */
struct rxrpc_conn_proto proto;
struct rxrpc_conn_parameters params;
@@ -286,7 +271,7 @@ struct rxrpc_connection {
struct work_struct processor; /* connection event processor */
union {
struct rb_node client_node; /* Node in local->client_conns */
- struct rb_node service_node; /* Node in trans->server_conns */
+ struct rb_node service_node; /* Node in peer->service_conns */
};
struct list_head link; /* link in master connection list */
struct rb_root calls; /* calls on this connection */
@@ -494,10 +479,6 @@ extern u32 rxrpc_epoch;
extern atomic_t rxrpc_debug_id;
extern struct workqueue_struct *rxrpc_workqueue;
-extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_conn_parameters *,
- struct sockaddr *,
- int, gfp_t);
-
/*
* call_accept.c
*/
@@ -526,7 +507,6 @@ struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *,
struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long);
struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *,
struct rxrpc_conn_parameters *,
- struct rxrpc_transport *,
struct sockaddr_rxrpc *,
unsigned long, gfp_t);
struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *,
@@ -543,7 +523,7 @@ void __exit rxrpc_destroy_all_calls(void);
extern struct idr rxrpc_client_conn_ids;
int rxrpc_get_client_connection_id(struct rxrpc_connection *,
- struct rxrpc_transport *, gfp_t);
+ struct rxrpc_peer *, gfp_t);
void rxrpc_put_client_connection_id(struct rxrpc_connection *);
/*
@@ -561,15 +541,16 @@ extern struct list_head rxrpc_connections;
extern rwlock_t rxrpc_connection_lock;
int rxrpc_connect_call(struct rxrpc_call *, struct rxrpc_conn_parameters *,
- struct rxrpc_transport *,
struct sockaddr_rxrpc *, gfp_t);
+struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *,
+ struct rxrpc_peer *,
+ struct sk_buff *);
void rxrpc_disconnect_call(struct rxrpc_call *);
void rxrpc_put_connection(struct rxrpc_connection *);
void __exit rxrpc_destroy_all_connections(void);
-struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *,
- struct sk_buff *);
-extern struct rxrpc_connection *
-rxrpc_incoming_connection(struct rxrpc_transport *, struct sk_buff *);
+struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *,
+ struct rxrpc_peer *,
+ struct sk_buff *);
static inline bool rxrpc_conn_is_client(const struct rxrpc_connection *conn)
{
@@ -586,12 +567,6 @@ static inline void rxrpc_get_connection(struct rxrpc_connection *conn)
atomic_inc(&conn->usage);
}
-static inline
-struct rxrpc_connection *rxrpc_get_connection_maybe(struct rxrpc_connection *conn)
-{
- return atomic_inc_not_zero(&conn->usage) ? conn : NULL;
-}
-
/*
* input.c
*/
@@ -746,18 +721,6 @@ static inline void rxrpc_sysctl_exit(void) {}
#endif
/*
- * transport.c
- */
-extern unsigned int rxrpc_transport_expiry;
-
-struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *,
- struct rxrpc_peer *, gfp_t);
-void rxrpc_put_transport(struct rxrpc_transport *);
-void __exit rxrpc_destroy_all_transports(void);
-struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *,
- struct rxrpc_peer *);
-
-/*
* utils.c
*/
void rxrpc_get_addr_from_skb(struct rxrpc_local *, const struct sk_buff *,
diff --git a/net/rxrpc/call_accept.c b/net/rxrpc/call_accept.c
index 833ad0622b61..202e053a3c6d 100644
--- a/net/rxrpc/call_accept.c
+++ b/net/rxrpc/call_accept.c
@@ -74,7 +74,6 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local,
struct sockaddr_rxrpc *srx)
{
struct rxrpc_connection *conn;
- struct rxrpc_transport *trans;
struct rxrpc_skb_priv *sp, *nsp;
struct rxrpc_peer *peer;
struct rxrpc_call *call;
@@ -102,16 +101,8 @@ static int rxrpc_accept_incoming_call(struct rxrpc_local *local,
goto error;
}
- trans = rxrpc_get_transport(local, peer, GFP_NOIO);
+ conn = rxrpc_incoming_connection(local, peer, skb);
rxrpc_put_peer(peer);
- if (IS_ERR(trans)) {
- _debug("no trans");
- ret = -EBUSY;
- goto error;
- }
-
- conn = rxrpc_incoming_connection(trans, skb);
- rxrpc_put_transport(trans);
if (IS_ERR(conn)) {
_debug("no conn");
ret = PTR_ERR(conn);
diff --git a/net/rxrpc/call_object.c b/net/rxrpc/call_object.c
index 9b3b48abe12f..5c9ccb5f0bc7 100644
--- a/net/rxrpc/call_object.c
+++ b/net/rxrpc/call_object.c
@@ -286,11 +286,9 @@ static struct rxrpc_call *rxrpc_alloc_call(gfp_t gfp)
/*
* Allocate a new client call.
*/
-static struct rxrpc_call *rxrpc_alloc_client_call(
- struct rxrpc_sock *rx,
- struct rxrpc_conn_parameters *cp,
- struct sockaddr_rxrpc *srx,
- gfp_t gfp)
+static struct rxrpc_call *rxrpc_alloc_client_call(struct rxrpc_sock *rx,
+ struct sockaddr_rxrpc *srx,
+ gfp_t gfp)
{
struct rxrpc_call *call;
@@ -320,7 +318,6 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
sizeof(call->peer_ip.ipv6_addr));
break;
}
-
call->service_id = srx->srx_service;
call->in_clientflag = 0;
@@ -333,7 +330,6 @@ static struct rxrpc_call *rxrpc_alloc_client_call(
*/
static int rxrpc_begin_client_call(struct rxrpc_call *call,
struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
struct sockaddr_rxrpc *srx,
gfp_t gfp)
{
@@ -342,7 +338,7 @@ static int rxrpc_begin_client_call(struct rxrpc_call *call,
/* Set up or get a connection record and set the protocol parameters,
* including channel number and call ID.
*/
- ret = rxrpc_connect_call(call, cp, trans, srx, gfp);
+ ret = rxrpc_connect_call(call, cp, srx, gfp);
if (ret < 0)
return ret;
@@ -366,7 +362,6 @@ static int rxrpc_begin_client_call(struct rxrpc_call *call,
*/
struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
struct sockaddr_rxrpc *srx,
unsigned long user_call_ID,
gfp_t gfp)
@@ -377,7 +372,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
_enter("%p,%lx", rx, user_call_ID);
- call = rxrpc_alloc_client_call(rx, cp, srx, gfp);
+ call = rxrpc_alloc_client_call(rx, srx, gfp);
if (IS_ERR(call)) {
_leave(" = %ld", PTR_ERR(call));
return call;
@@ -413,7 +408,7 @@ struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx,
list_add_tail(&call->link, &rxrpc_calls);
write_unlock_bh(&rxrpc_call_lock);
- ret = rxrpc_begin_client_call(call, cp, trans, srx, gfp);
+ ret = rxrpc_begin_client_call(call, cp, srx, gfp);
if (ret < 0)
goto error;
diff --git a/net/rxrpc/conn_client.c b/net/rxrpc/conn_client.c
index 2cccb4be289d..7a4fab97ad27 100644
--- a/net/rxrpc/conn_client.c
+++ b/net/rxrpc/conn_client.c
@@ -34,7 +34,7 @@ static DEFINE_SPINLOCK(rxrpc_conn_id_lock);
* concentrated. We will also need to retire connections from an old epoch.
*/
int rxrpc_get_client_connection_id(struct rxrpc_connection *conn,
- struct rxrpc_transport *trans,
+ struct rxrpc_peer *peer,
gfp_t gfp)
{
u32 epoch;
@@ -43,7 +43,7 @@ int rxrpc_get_client_connection_id(struct rxrpc_connection *conn,
_enter("");
idr_preload(gfp);
- write_lock_bh(&trans->conn_lock);
+ write_lock_bh(&peer->conn_lock);
spin_lock(&rxrpc_conn_id_lock);
epoch = rxrpc_epoch;
@@ -68,7 +68,7 @@ int rxrpc_get_client_connection_id(struct rxrpc_connection *conn,
rxrpc_client_conn_ids.cur = id + 1;
spin_unlock(&rxrpc_conn_id_lock);
- write_unlock_bh(&trans->conn_lock);
+ write_unlock_bh(&peer->conn_lock);
idr_preload_end();
conn->proto.epoch = epoch;
@@ -79,7 +79,7 @@ int rxrpc_get_client_connection_id(struct rxrpc_connection *conn,
error:
spin_unlock(&rxrpc_conn_id_lock);
- write_unlock_bh(&trans->conn_lock);
+ write_unlock_bh(&peer->conn_lock);
idr_preload_end();
_leave(" = %d", id);
return id;
diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c
index 5b27241ebea6..64936884b168 100644
--- a/net/rxrpc/conn_object.c
+++ b/net/rxrpc/conn_object.c
@@ -100,9 +100,7 @@ static void rxrpc_add_call_ID_to_conn(struct rxrpc_connection *conn,
* padding bytes in *cp.
*/
static struct rxrpc_connection *
-rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
- gfp_t gfp)
+rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp, gfp_t gfp)
{
struct rxrpc_connection *conn;
int ret;
@@ -132,7 +130,7 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp,
break;
}
- ret = rxrpc_get_client_connection_id(conn, trans, gfp);
+ ret = rxrpc_get_client_connection_id(conn, cp->peer, gfp);
if (ret < 0)
goto error_0;
@@ -146,9 +144,10 @@ rxrpc_alloc_client_connection(struct rxrpc_conn_parameters *cp,
list_add_tail(&conn->link, &rxrpc_connections);
write_unlock(&rxrpc_connection_lock);
+ /* We steal the caller's peer ref. */
+ cp->peer = NULL;
+ rxrpc_get_local(conn->params.local);
key_get(conn->params.key);
- conn->trans = trans;
- atomic_inc(&trans->usage);
_leave(" = %p", conn);
return conn;
@@ -167,7 +166,6 @@ error_0:
*/
int rxrpc_connect_call(struct rxrpc_call *call,
struct rxrpc_conn_parameters *cp,
- struct rxrpc_transport *trans,
struct sockaddr_rxrpc *srx,
gfp_t gfp)
{
@@ -181,8 +179,9 @@ int rxrpc_connect_call(struct rxrpc_call *call,
_enter("{%d,%lx},", call->debug_id, call->user_call_ID);
- cp->peer = trans->peer;
- rxrpc_get_peer(cp->peer);
+ cp->peer = rxrpc_lookup_peer(cp->local, srx, gfp);
+ if (!cp->peer)
+ return -ENOMEM;
if (!cp->exclusive) {
/* Search for a existing client connection unless this is going
@@ -210,7 +209,7 @@ int rxrpc_connect_call(struct rxrpc_call *call,
/* We didn't find a connection or we want an exclusive one. */
_debug("get new conn");
- candidate = rxrpc_alloc_client_connection(cp, trans, gfp);
+ candidate = rxrpc_alloc_client_connection(cp, gfp);
if (!candidate) {
_leave(" = -ENOMEM");
return -ENOMEM;
@@ -280,6 +279,8 @@ found_channel:
rxrpc_add_call_ID_to_conn(conn, call);
spin_unlock(&conn->channel_lock);
+ rxrpc_put_peer(cp->peer);
+ cp->peer = NULL;
_leave(" = %p {u=%d}", conn, atomic_read(&conn->usage));
return 0;
@@ -328,6 +329,8 @@ interrupted:
remove_wait_queue(&conn->channel_wq, &myself);
__set_current_state(TASK_RUNNING);
rxrpc_put_connection(conn);
+ rxrpc_put_peer(cp->peer);
+ cp->peer = NULL;
_leave(" = -ERESTARTSYS");
return -ERESTARTSYS;
}
@@ -335,7 +338,8 @@ interrupted:
/*
* get a record of an incoming connection
*/
-struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans,
+struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_local *local,
+ struct rxrpc_peer *peer,
struct sk_buff *skb)
{
struct rxrpc_connection *conn, *candidate = NULL;
@@ -353,9 +357,9 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans
cid = sp->hdr.cid & RXRPC_CIDMASK;
/* search the connection list first */
- read_lock_bh(&trans->conn_lock);
+ read_lock_bh(&peer->conn_lock);
- p = trans->server_conns.rb_node;
+ p = peer->service_conns.rb_node;
while (p) {
conn = rb_entry(p, struct rxrpc_connection, service_node);
@@ -372,7 +376,7 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans
else
goto found_extant_connection;
}
- read_unlock_bh(&trans->conn_lock);
+ read_unlock_bh(&peer->conn_lock);
/* not yet present - create a candidate for a new record and then
* redo the search */
@@ -382,13 +386,12 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans
return ERR_PTR(-ENOMEM);
}
- candidate->trans = trans;
- candidate->proto.local = trans->local;
+ candidate->proto.local = local;
candidate->proto.epoch = sp->hdr.epoch;
candidate->proto.cid = sp->hdr.cid & RXRPC_CIDMASK;
candidate->proto.in_clientflag = RXRPC_CLIENT_INITIATED;
- candidate->params.local = trans->local;
- candidate->params.peer = trans->peer;
+ candidate->params.local = local;
+ candidate->params.peer = peer;
candidate->params.service_id = sp->hdr.serviceId;
candidate->security_ix = sp->hdr.securityIndex;
candidate->out_clientflag = 0;
@@ -396,9 +399,9 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans
if (candidate->params.service_id)
candidate->state = RXRPC_CONN_SERVER_UNSECURED;
- write_lock_bh(&trans->conn_lock);
+ write_lock_bh(&peer->conn_lock);
- pp = &trans->server_conns.rb_node;
+ pp = &peer->service_conns.rb_node;
p = NULL;
while (*pp) {
p = *pp;
@@ -420,10 +423,11 @@ struct rxrpc_connection *rxrpc_incoming_connection(struct rxrpc_transport *trans
conn = candidate;
candidate = NULL;
rb_link_node(&conn->service_node, p, pp);
- rb_insert_color(&conn->service_node, &trans->server_conns);
- atomic_inc(&conn->trans->usage);
+ rb_insert_color(&conn->service_node, &peer->service_conns);
+ rxrpc_get_peer(peer);
+ rxrpc_get_local(local);
- write_unlock_bh(&trans->conn_lock);
+ write_unlock_bh(&peer->conn_lock);
write_lock(&rxrpc_connection_lock);
list_add_tail(&conn->link, &rxrpc_connections);
@@ -440,21 +444,21 @@ success:
/* we found the connection in the list immediately */
found_extant_connection:
if (sp->hdr.securityIndex != conn->security_ix) {
- read_unlock_bh(&trans->conn_lock);
+ read_unlock_bh(&peer->conn_lock);
goto security_mismatch;
}
rxrpc_get_connection(conn);
- read_unlock_bh(&trans->conn_lock);
+ read_unlock_bh(&peer->conn_lock);
goto success;
/* we found the connection on the second time through the list */
found_extant_second:
if (sp->hdr.securityIndex != conn->security_ix) {
- write_unlock_bh(&trans->conn_lock);
+ write_unlock_bh(&peer->conn_lock);
goto security_mismatch;
}
rxrpc_get_connection(conn);
- write_unlock_bh(&trans->conn_lock);
+ write_unlock_bh(&peer->conn_lock);
kfree(candidate);
goto success;
@@ -468,7 +472,8 @@ security_mismatch:
* find a connection based on transport and RxRPC connection ID for an incoming
* packet
*/
-struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
+struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_local *local,
+ struct rxrpc_peer *peer,
struct sk_buff *skb)
{
struct rxrpc_connection *conn;
@@ -478,13 +483,13 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
_enter(",{%x,%x}", sp->hdr.cid, sp->hdr.flags);
- read_lock_bh(&trans->conn_lock);
+ read_lock_bh(&peer->conn_lock);
cid = sp->hdr.cid & RXRPC_CIDMASK;
epoch = sp->hdr.epoch;
if (sp->hdr.flags & RXRPC_CLIENT_INITIATED) {
- p = trans->server_conns.rb_node;
+ p = peer->service_conns.rb_node;
while (p) {
conn = rb_entry(p, struct rxrpc_connection, service_node);
@@ -507,13 +512,13 @@ struct rxrpc_connection *rxrpc_find_connection(struct rxrpc_transport *trans,
goto found;
}
- read_unlock_bh(&trans->conn_lock);
+ read_unlock_bh(&peer->conn_lock);
_leave(" = NULL");
return NULL;
found:
rxrpc_get_connection(conn);
- read_unlock_bh(&trans->conn_lock);
+ read_unlock_bh(&peer->conn_lock);
_leave(" = %p", conn);
return conn;
}
@@ -575,8 +580,9 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn)
conn->security->clear(conn);
key_put(conn->params.key);
key_put(conn->server_key);
+ rxrpc_put_peer(conn->params.peer);
+ rxrpc_put_local(conn->params.local);
- rxrpc_put_transport(conn->trans);
kfree(conn);
_leave("");
}
@@ -587,6 +593,7 @@ static void rxrpc_destroy_connection(struct rxrpc_connection *conn)
static void rxrpc_connection_reaper(struct work_struct *work)
{
struct rxrpc_connection *conn, *_p;
+ struct rxrpc_peer *peer;
unsigned long now, earliest, reap_time;
LIST_HEAD(graveyard);
@@ -605,7 +612,8 @@ static void rxrpc_connection_reaper(struct work_struct *work)
if (likely(atomic_read(&conn->usage) > 0))
continue;
- write_lock_bh(&conn->trans->conn_lock);
+ peer = conn->params.peer;
+ write_lock_bh(&peer->conn_lock);
reap_time = conn->put_time + rxrpc_connection_expiry;
if (atomic_read(&conn->usage) > 0) {
@@ -616,13 +624,13 @@ static void rxrpc_connection_reaper(struct work_struct *work)
rxrpc_put_client_connection_id(conn);
else
rb_erase(&conn->service_node,
- &conn->trans->server_conns);
+ &peer->service_conns);
} else if (reap_time < earliest) {
earliest = reap_time;
}
- write_unlock_bh(&conn->trans->conn_lock);
+ write_unlock_bh(&peer->conn_lock);
}
write_unlock(&rxrpc_connection_lock);
diff --git a/net/rxrpc/input.c b/net/rxrpc/input.c
index 799aec18aa7b..f4bd57b77b93 100644
--- a/net/rxrpc/input.c
+++ b/net/rxrpc/input.c
@@ -631,7 +631,6 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
struct sk_buff *skb)
{
struct rxrpc_peer *peer;
- struct rxrpc_transport *trans;
struct rxrpc_connection *conn;
struct sockaddr_rxrpc srx;
@@ -641,13 +640,8 @@ static struct rxrpc_connection *rxrpc_conn_from_local(struct rxrpc_local *local,
if (!peer)
goto cant_find_peer;
- trans = rxrpc_find_transport(local, peer);
+ conn = rxrpc_find_connection(local, peer, skb);
rcu_read_unlock();
- if (!trans)
- goto cant_find_conn;
-
- conn = rxrpc_find_connection(trans, skb);
- rxrpc_put_transport(trans);
if (!conn)
goto cant_find_conn;
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index 8e24939aeac8..f4bda06b7d2d 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -140,10 +140,8 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
unsigned long user_call_ID, bool exclusive)
{
struct rxrpc_conn_parameters cp;
- struct rxrpc_transport *trans;
struct rxrpc_call *call;
struct key *key;
- long ret;
DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name);
@@ -162,30 +160,10 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg,
cp.security_level = rx->min_sec_level;
cp.exclusive = rx->exclusive | exclusive;
cp.service_id = srx->srx_service;
- trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen,
- GFP_KERNEL);
- if (IS_ERR(trans)) {
- ret = PTR_ERR(trans);
- goto out;
- }
- cp.peer = trans->peer;
-
- call = rxrpc_new_client_call(rx, &cp, trans, srx, user_call_ID,
- GFP_KERNEL);
- rxrpc_put_transport(trans);
- if (IS_ERR(call)) {
- ret = PTR_ERR(call);
- goto out_trans;
- }
+ call = rxrpc_new_client_call(rx, &cp, srx, user_call_ID, GFP_KERNEL);
_leave(" = %p\n", call);
return call;
-
-out_trans:
- rxrpc_put_transport(trans);
-out:
- _leave(" = %ld", ret);
- return ERR_PTR(ret);
}
/*
diff --git a/net/rxrpc/peer_object.c b/net/rxrpc/peer_object.c
index 6baad708f3b1..01d4930a11f7 100644
--- a/net/rxrpc/peer_object.c
+++ b/net/rxrpc/peer_object.c
@@ -188,6 +188,8 @@ struct rxrpc_peer *rxrpc_alloc_peer(struct rxrpc_local *local, gfp_t gfp)
INIT_HLIST_HEAD(&peer->error_targets);
INIT_WORK(&peer->error_distributor,
&rxrpc_peer_error_distributor);
+ peer->service_conns = RB_ROOT;
+ rwlock_init(&peer->conn_lock);
spin_lock_init(&peer->lock);
peer->debug_id = atomic_inc_return(&rxrpc_debug_id);
}
diff --git a/net/rxrpc/sysctl.c b/net/rxrpc/sysctl.c
index a99690a8a3da..03ad08774d4e 100644
--- a/net/rxrpc/sysctl.c
+++ b/net/rxrpc/sysctl.c
@@ -90,14 +90,6 @@ static struct ctl_table rxrpc_sysctl_table[] = {
.proc_handler = proc_dointvec_minmax,
.extra1 = (void *)&one,
},
- {
- .procname = "transport_expiry",
- .data = &rxrpc_transport_expiry,
- .maxlen = sizeof(unsigned int),
- .mode = 0644,
- .proc_handler = proc_dointvec_minmax,
- .extra1 = (void *)&one,
- },
/* Non-time values */
{
diff --git a/net/rxrpc/transport.c b/net/rxrpc/transport.c
deleted file mode 100644
index 71947402d071..000000000000
--- a/net/rxrpc/transport.c
+++ /dev/null
@@ -1,265 +0,0 @@
-/* RxRPC point-to-point transport session management
- *
- * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * 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 the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
-
-#include <linux/module.h>
-#include <linux/net.h>
-#include <linux/skbuff.h>
-#include <linux/slab.h>
-#include <net/sock.h>
-#include <net/af_rxrpc.h>
-#include "ar-internal.h"
-
-/*
- * Time after last use at which transport record is cleaned up.
- */
-unsigned int rxrpc_transport_expiry = 3600 * 24;
-
-static void rxrpc_transport_reaper(struct work_struct *work);
-
-static LIST_HEAD(rxrpc_transports);
-static DEFINE_RWLOCK(rxrpc_transport_lock);
-static DECLARE_DELAYED_WORK(rxrpc_transport_reap, rxrpc_transport_reaper);
-
-/*
- * allocate a new transport session manager
- */
-static struct rxrpc_transport *rxrpc_alloc_transport(struct rxrpc_local *local,
- struct rxrpc_peer *peer,
- gfp_t gfp)
-{
- struct rxrpc_transport *trans;
-
- _enter("");
-
- trans = kzalloc(sizeof(struct rxrpc_transport), gfp);
- if (trans) {
- trans->local = local;
- trans->peer = peer;
- INIT_LIST_HEAD(&trans->link);
- trans->server_conns = RB_ROOT;
- rwlock_init(&trans->conn_lock);
- atomic_set(&trans->usage, 1);
- trans->debug_id = atomic_inc_return(&rxrpc_debug_id);
- }
-
- _leave(" = %p", trans);
- return trans;
-}
-
-/*
- * obtain a transport session for the nominated endpoints
- */
-struct rxrpc_transport *rxrpc_get_transport(struct rxrpc_local *local,
- struct rxrpc_peer *peer,
- gfp_t gfp)
-{
- struct rxrpc_transport *trans, *candidate;
- const char *new = "old";
- int usage;
-
- _enter("{%pI4+%hu},{%pI4+%hu},",
- &local->srx.transport.sin.sin_addr,
- ntohs(local->srx.transport.sin.sin_port),
- &peer->srx.transport.sin.sin_addr,
- ntohs(peer->srx.transport.sin.sin_port));
-
- /* search the transport list first */
- read_lock_bh(&rxrpc_transport_lock);
- list_for_each_entry(trans, &rxrpc_transports, link) {
- if (trans->local == local && trans->peer == peer)
- goto found_extant_transport;
- }
- read_unlock_bh(&rxrpc_transport_lock);
-
- /* not yet present - create a candidate for a new record and then
- * redo the search */
- candidate = rxrpc_alloc_transport(local, peer, gfp);
- if (!candidate) {
- _leave(" = -ENOMEM");
- return ERR_PTR(-ENOMEM);
- }
-
- write_lock_bh(&rxrpc_transport_lock);
-
- list_for_each_entry(trans, &rxrpc_transports, link) {
- if (trans->local == local && trans->peer == peer)
- goto found_extant_second;
- }
-
- /* we can now add the new candidate to the list */
- trans = candidate;
- candidate = NULL;
- usage = atomic_read(&trans->usage);
-
- rxrpc_get_local(trans->local);
- rxrpc_get_peer(trans->peer);
- list_add_tail(&trans->link, &rxrpc_transports);
- write_unlock_bh(&rxrpc_transport_lock);
- new = "new";
-
-success:
- _net("TRANSPORT %s %d local %d -> peer %d",
- new,
- trans->debug_id,
- trans->local->debug_id,
- trans->peer->debug_id);
-
- _leave(" = %p {u=%d}", trans, usage);
- return trans;
-
- /* we found the transport in the list immediately */
-found_extant_transport:
- usage = atomic_inc_return(&trans->usage);
- read_unlock_bh(&rxrpc_transport_lock);
- goto success;
-
- /* we found the transport on the second time through the list */
-found_extant_second:
- usage = atomic_inc_return(&trans->usage);
- write_unlock_bh(&rxrpc_transport_lock);
- kfree(candidate);
- goto success;
-}
-
-/*
- * find the transport connecting two endpoints
- */
-struct rxrpc_transport *rxrpc_find_transport(struct rxrpc_local *local,
- struct rxrpc_peer *peer)
-{
- struct rxrpc_transport *trans;
-
- _enter("{%pI4+%hu},{%pI4+%hu},",
- &local->srx.transport.sin.sin_addr,
- ntohs(local->srx.transport.sin.sin_port),
- &peer->srx.transport.sin.sin_addr,
- ntohs(peer->srx.transport.sin.sin_port));
-
- /* search the transport list */
- read_lock_bh(&rxrpc_transport_lock);
-
- list_for_each_entry(trans, &rxrpc_transports, link) {
- if (trans->local == local && trans->peer == peer)
- goto found_extant_transport;
- }
-
- read_unlock_bh(&rxrpc_transport_lock);
- _leave(" = NULL");
- return NULL;
-
-found_extant_transport:
- atomic_inc(&trans->usage);
- read_unlock_bh(&rxrpc_transport_lock);
- _leave(" = %p", trans);
- return trans;
-}
-
-/*
- * release a transport session
- */
-void rxrpc_put_transport(struct rxrpc_transport *trans)
-{
- _enter("%p{u=%d}", trans, atomic_read(&trans->usage));
-
- ASSERTCMP(atomic_read(&trans->usage), >, 0);
-
- trans->put_time = ktime_get_seconds();
- if (unlikely(atomic_dec_and_test(&trans->usage))) {
- _debug("zombie");
- /* let the reaper determine the timeout to avoid a race with
- * overextending the timeout if the reaper is running at the
- * same time */
- rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0);
- }
- _leave("");
-}
-
-/*
- * clean up a transport session
- */
-static void rxrpc_cleanup_transport(struct rxrpc_transport *trans)
-{
- _net("DESTROY TRANS %d", trans->debug_id);
-
- rxrpc_put_local(trans->local);
- rxrpc_put_peer(trans->peer);
- kfree(trans);
-}
-
-/*
- * reap dead transports that have passed their expiry date
- */
-static void rxrpc_transport_reaper(struct work_struct *work)
-{
- struct rxrpc_transport *trans, *_p;
- unsigned long now, earliest, reap_time;
-
- LIST_HEAD(graveyard);
-
- _enter("");
-
- now = ktime_get_seconds();
- earliest = ULONG_MAX;
-
- /* extract all the transports that have been dead too long */
- write_lock_bh(&rxrpc_transport_lock);
- list_for_each_entry_safe(trans, _p, &rxrpc_transports, link) {
- _debug("reap TRANS %d { u=%d t=%ld }",
- trans->debug_id, atomic_read(&trans->usage),
- (long) now - (long) trans->put_time);
-
- if (likely(atomic_read(&trans->usage) > 0))
- continue;
-
- reap_time = trans->put_time + rxrpc_transport_expiry;
- if (reap_time <= now)
- list_move_tail(&trans->link, &graveyard);
- else if (reap_time < earliest)
- earliest = reap_time;
- }
- write_unlock_bh(&rxrpc_transport_lock);
-
- if (earliest != ULONG_MAX) {
- _debug("reschedule reaper %ld", (long) earliest - now);
- ASSERTCMP(earliest, >, now);
- rxrpc_queue_delayed_work(&rxrpc_transport_reap,
- (earliest - now) * HZ);
- }
-
- /* then destroy all those pulled out */
- while (!list_empty(&graveyard)) {
- trans = list_entry(graveyard.next, struct rxrpc_transport,
- link);
- list_del_init(&trans->link);
-
- ASSERTCMP(atomic_read(&trans->usage), ==, 0);
- rxrpc_cleanup_transport(trans);
- }
-
- _leave("");
-}
-
-/*
- * preemptively destroy all the transport session records rather than waiting
- * for them to time out
- */
-void __exit rxrpc_destroy_all_transports(void)
-{
- _enter("");
-
- rxrpc_transport_expiry = 0;
- cancel_delayed_work(&rxrpc_transport_reap);
- rxrpc_queue_delayed_work(&rxrpc_transport_reap, 0);
-
- _leave("");
-}
^ permalink raw reply related [flat|nested] 16+ messages in thread
* Re: [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
` (13 preceding siblings ...)
2016-06-22 9:51 ` [PATCH net-next 14/14] rxrpc: Kill off the rxrpc_transport struct David Howells
@ 2016-06-22 10:03 ` David Howells
14 siblings, 0 replies; 16+ messages in thread
From: David Howells @ 2016-06-22 10:03 UTC (permalink / raw)
Cc: dhowells, davem, netdev, linux-afs, linux-kernel
Ignore this version please. Fengguang's magic box of tricks spotted a bug in
it.
David
^ permalink raw reply [flat|nested] 16+ messages in thread
end of thread, other threads:[~2016-06-22 10:04 UTC | newest]
Thread overview: 16+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-06-22 9:49 [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
2016-06-22 9:49 ` [PATCH net-next 01/14] rxrpc: checking for IS_ERR() instead of NULL David Howells
2016-06-22 9:49 ` [PATCH net-next 02/14] rxrpc: fix uninitialized variable use David Howells
2016-06-22 9:50 ` [PATCH net-next 03/14] rxrpc: Use structs to hold connection params and protocol info David Howells
2016-06-22 9:50 ` [PATCH net-next 04/14] rxrpc: Replace conn->trans->{local, peer} with conn->params.{local, peer} David Howells
2016-06-22 9:50 ` [PATCH net-next 05/14] rxrpc: Fix exclusive connection handling David Howells
2016-06-22 9:50 ` [PATCH net-next 06/14] rxrpc: Pass sk_buff * rather than rxrpc_host_header * to functions David Howells
2016-06-22 9:50 ` [PATCH net-next 07/14] rxrpc: rxrpc_connection_lock shouldn't be a BH lock, but conn_lock is David Howells
2016-06-22 9:50 ` [PATCH net-next 08/14] rxrpc: Use IDR to allocate client conn IDs on a machine-wide basis David Howells
2016-06-22 9:50 ` [PATCH net-next 09/14] rxrpc: Validate the net address given to rxrpc_kernel_begin_call() David Howells
2016-06-22 9:50 ` [PATCH net-next 10/14] rxrpc: Calls displayed in /proc may in future lack a connection David Howells
2016-06-22 9:51 ` [PATCH net-next 11/14] rxrpc: Make rxrpc_send_packet() take a connection not a transport David Howells
2016-06-22 9:51 ` [PATCH net-next 12/14] rxrpc: Provide more refcount helper functions David Howells
2016-06-22 9:51 ` [PATCH net-next 13/14] rxrpc: Kill the client connection bundle concept David Howells
2016-06-22 9:51 ` [PATCH net-next 14/14] rxrpc: Kill off the rxrpc_transport struct David Howells
2016-06-22 10:03 ` [PATCH net-next 00/14] rxrpc: Get rid of conn bundle and transport structs David Howells
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).