WSL2-Linux-Kernel/net/tipc/bcast.c

1007 строки
26 KiB
C
Исходник Обычный вид История

/*
* net/tipc/bcast.c: TIPC broadcast code
*
* Copyright (c) 2004-2006, 2014-2015, Ericsson AB
* Copyright (c) 2004, Intel Corporation.
* Copyright (c) 2005, 2010-2011, Wind River Systems
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the names of the copyright holders nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* Alternatively, this software may be distributed under the terms of the
* GNU General Public License ("GPL") version 2 as published by the Free
* Software Foundation.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "socket.h"
#include "msg.h"
#include "bcast.h"
#include "name_distr.h"
#include "core.h"
#define MAX_PKT_DEFAULT_MCAST 1500 /* bcast link max packet size (fixed) */
#define BCLINK_WIN_DEFAULT 20 /* bcast link window size (default) */
const char tipc_bclink_name[] = "broadcast-link";
static void tipc_nmap_diff(struct tipc_node_map *nm_a,
struct tipc_node_map *nm_b,
struct tipc_node_map *nm_diff);
static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node);
static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node);
static void tipc_bclink_lock(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
spin_lock_bh(&tn->bclink->lock);
}
static void tipc_bclink_unlock(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
spin_unlock_bh(&tn->bclink->lock);
}
void tipc_bclink_input(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
tipc_sk_mcast_rcv(net, &tn->bclink->arrvq, &tn->bclink->inputq);
}
uint tipc_bclink_get_mtu(void)
{
return MAX_PKT_DEFAULT_MCAST;
}
static u32 bcbuf_acks(struct sk_buff *buf)
{
return (u32)(unsigned long)TIPC_SKB_CB(buf)->handle;
}
static void bcbuf_set_acks(struct sk_buff *buf, u32 acks)
{
TIPC_SKB_CB(buf)->handle = (void *)(unsigned long)acks;
}
static void bcbuf_decr_acks(struct sk_buff *buf)
{
bcbuf_set_acks(buf, bcbuf_acks(buf) - 1);
}
void tipc_bclink_add_node(struct net *net, u32 addr)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
tipc_bclink_lock(net);
tipc_nmap_add(&tn->bclink->bcast_nodes, addr);
tipc_bclink_unlock(net);
}
void tipc_bclink_remove_node(struct net *net, u32 addr)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
tipc_bclink_lock(net);
tipc_nmap_remove(&tn->bclink->bcast_nodes, addr);
/* Last node? => reset backlog queue */
if (!tn->bclink->bcast_nodes.count)
tipc_link_purge_backlog(&tn->bclink->link);
tipc_bclink_unlock(net);
}
static void bclink_set_last_sent(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_link *bcl = tn->bcl;
bcl->silent_intv_cnt = mod(bcl->snd_nxt - 1);
}
u32 tipc_bclink_get_last_sent(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
return tn->bcl->silent_intv_cnt;
}
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
static void bclink_update_last_sent(struct tipc_node *node, u32 seqno)
{
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
node->bclink.last_sent = less_eq(node->bclink.last_sent, seqno) ?
seqno : node->bclink.last_sent;
}
/**
* tipc_bclink_retransmit_to - get most recent node to request retransmission
*
* Called with bclink_lock locked
*/
struct tipc_node *tipc_bclink_retransmit_to(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
return tn->bclink->retransmit_to;
}
/**
* bclink_retransmit_pkt - retransmit broadcast packets
* @after: sequence number of last packet to *not* retransmit
* @to: sequence number of last packet to retransmit
*
* Called with bclink_lock locked
*/
static void bclink_retransmit_pkt(struct tipc_net *tn, u32 after, u32 to)
{
struct sk_buff *skb;
struct tipc_link *bcl = tn->bcl;
skb_queue_walk(&bcl->transmq, skb) {
if (more(buf_seqno(skb), after)) {
tipc_link_retransmit(bcl, skb, mod(to - after));
break;
}
}
}
tipc: fix bug in multicast congestion handling One aim of commit 50100a5e39461b2a61d6040e73c384766c29975d ("tipc: use pseudo message to wake up sockets after link congestion") was to handle link congestion abatement in a uniform way for both unicast and multicast transmit. However, the latter doesn't work correctly, and has been broken since the referenced commit was applied. If a user now sends a burst of multicast messages that is big enough to cause broadcast link congestion, it will be put to sleep, and not be waked up when the congestion abates as it should be. This has two reasons. First, the flag that is used, TIPC_WAKEUP_USERS, is set correctly, but in the wrong field. Instead of setting it in the 'action_flags' field of the arrival node struct, it is by mistake set in the dummy node struct that is owned by the broadcast link, where it will never tested for. Second, we cannot use the same flag for waking up unicast and multicast users, since the function tipc_node_unlock() needs to pick the wakeup pseudo messages to deliver from different queues. It must hence be able to distinguish between the two cases. This commit solves this problem by adding a new flag TIPC_WAKEUP_BCAST_USERS, and a new function tipc_bclink_wakeup_user(). The latter is to be called by tipc_node_unlock() when the named flag, now set in the correct field, is encountered. v2: using explicit 'unsigned int' declaration instead of 'uint', as per comment from David Miller. Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-07 22:12:34 +04:00
/**
* tipc_bclink_wakeup_users - wake up pending users
*
* Called with no locks taken
*/
void tipc_bclink_wakeup_users(struct net *net)
tipc: fix bug in multicast congestion handling One aim of commit 50100a5e39461b2a61d6040e73c384766c29975d ("tipc: use pseudo message to wake up sockets after link congestion") was to handle link congestion abatement in a uniform way for both unicast and multicast transmit. However, the latter doesn't work correctly, and has been broken since the referenced commit was applied. If a user now sends a burst of multicast messages that is big enough to cause broadcast link congestion, it will be put to sleep, and not be waked up when the congestion abates as it should be. This has two reasons. First, the flag that is used, TIPC_WAKEUP_USERS, is set correctly, but in the wrong field. Instead of setting it in the 'action_flags' field of the arrival node struct, it is by mistake set in the dummy node struct that is owned by the broadcast link, where it will never tested for. Second, we cannot use the same flag for waking up unicast and multicast users, since the function tipc_node_unlock() needs to pick the wakeup pseudo messages to deliver from different queues. It must hence be able to distinguish between the two cases. This commit solves this problem by adding a new flag TIPC_WAKEUP_BCAST_USERS, and a new function tipc_bclink_wakeup_user(). The latter is to be called by tipc_node_unlock() when the named flag, now set in the correct field, is encountered. v2: using explicit 'unsigned int' declaration instead of 'uint', as per comment from David Miller. Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-07 22:12:34 +04:00
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 16:36:41 +03:00
tipc_sk_rcv(net, &tn->bclink->link.wakeupq);
tipc: fix bug in multicast congestion handling One aim of commit 50100a5e39461b2a61d6040e73c384766c29975d ("tipc: use pseudo message to wake up sockets after link congestion") was to handle link congestion abatement in a uniform way for both unicast and multicast transmit. However, the latter doesn't work correctly, and has been broken since the referenced commit was applied. If a user now sends a burst of multicast messages that is big enough to cause broadcast link congestion, it will be put to sleep, and not be waked up when the congestion abates as it should be. This has two reasons. First, the flag that is used, TIPC_WAKEUP_USERS, is set correctly, but in the wrong field. Instead of setting it in the 'action_flags' field of the arrival node struct, it is by mistake set in the dummy node struct that is owned by the broadcast link, where it will never tested for. Second, we cannot use the same flag for waking up unicast and multicast users, since the function tipc_node_unlock() needs to pick the wakeup pseudo messages to deliver from different queues. It must hence be able to distinguish between the two cases. This commit solves this problem by adding a new flag TIPC_WAKEUP_BCAST_USERS, and a new function tipc_bclink_wakeup_user(). The latter is to be called by tipc_node_unlock() when the named flag, now set in the correct field, is encountered. v2: using explicit 'unsigned int' declaration instead of 'uint', as per comment from David Miller. Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-07 22:12:34 +04:00
}
/**
* tipc_bclink_acknowledge - handle acknowledgement of broadcast packets
* @n_ptr: node that sent acknowledgement info
* @acked: broadcast sequence # that has been acknowledged
*
* Node is locked, bclink_lock unlocked.
*/
void tipc_bclink_acknowledge(struct tipc_node *n_ptr, u32 acked)
{
struct sk_buff *skb, *tmp;
unsigned int released = 0;
struct net *net = n_ptr->net;
struct tipc_net *tn = net_generic(net, tipc_net_id);
if (unlikely(!n_ptr->bclink.recv_permitted))
return;
tipc_bclink_lock(net);
/* Bail out if tx queue is empty (no clean up is required) */
skb = skb_peek(&tn->bcl->transmq);
if (!skb)
goto exit;
/* Determine which messages need to be acknowledged */
if (acked == INVALID_LINK_SEQ) {
/*
* Contact with specified node has been lost, so need to
* acknowledge sent messages only (if other nodes still exist)
* or both sent and unsent messages (otherwise)
*/
if (tn->bclink->bcast_nodes.count)
acked = tn->bcl->silent_intv_cnt;
else
acked = tn->bcl->snd_nxt;
} else {
/*
* Bail out if specified sequence number does not correspond
* to a message that has been sent and not yet acknowledged
*/
if (less(acked, buf_seqno(skb)) ||
less(tn->bcl->silent_intv_cnt, acked) ||
less_eq(acked, n_ptr->bclink.acked))
goto exit;
}
/* Skip over packets that node has previously acknowledged */
skb_queue_walk(&tn->bcl->transmq, skb) {
if (more(buf_seqno(skb), n_ptr->bclink.acked))
break;
}
/* Update packets that node is now acknowledging */
skb_queue_walk_from_safe(&tn->bcl->transmq, skb, tmp) {
if (more(buf_seqno(skb), acked))
break;
bcbuf_decr_acks(skb);
bclink_set_last_sent(net);
if (bcbuf_acks(skb) == 0) {
__skb_unlink(skb, &tn->bcl->transmq);
kfree_skb(skb);
released = 1;
}
}
n_ptr->bclink.acked = acked;
/* Try resolving broadcast link congestion, if necessary */
if (unlikely(skb_peek(&tn->bcl->backlogq))) {
tipc_link_push_packets(tn->bcl);
bclink_set_last_sent(net);
}
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 16:36:41 +03:00
if (unlikely(released && !skb_queue_empty(&tn->bcl->wakeupq)))
tipc: fix bug in multicast congestion handling One aim of commit 50100a5e39461b2a61d6040e73c384766c29975d ("tipc: use pseudo message to wake up sockets after link congestion") was to handle link congestion abatement in a uniform way for both unicast and multicast transmit. However, the latter doesn't work correctly, and has been broken since the referenced commit was applied. If a user now sends a burst of multicast messages that is big enough to cause broadcast link congestion, it will be put to sleep, and not be waked up when the congestion abates as it should be. This has two reasons. First, the flag that is used, TIPC_WAKEUP_USERS, is set correctly, but in the wrong field. Instead of setting it in the 'action_flags' field of the arrival node struct, it is by mistake set in the dummy node struct that is owned by the broadcast link, where it will never tested for. Second, we cannot use the same flag for waking up unicast and multicast users, since the function tipc_node_unlock() needs to pick the wakeup pseudo messages to deliver from different queues. It must hence be able to distinguish between the two cases. This commit solves this problem by adding a new flag TIPC_WAKEUP_BCAST_USERS, and a new function tipc_bclink_wakeup_user(). The latter is to be called by tipc_node_unlock() when the named flag, now set in the correct field, is encountered. v2: using explicit 'unsigned int' declaration instead of 'uint', as per comment from David Miller. Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-10-07 22:12:34 +04:00
n_ptr->action_flags |= TIPC_WAKEUP_BCAST_USERS;
exit:
tipc_bclink_unlock(net);
}
/**
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
* tipc_bclink_update_link_state - update broadcast link state
*
tipc: purge tipc_net_lock lock Now tipc routing hierarchy comprises the structures 'node', 'link'and 'bearer'. The whole hierarchy is protected by a big read/write lock, tipc_net_lock, to ensure that nothing is added or removed while code is accessing any of these structures. Obviously the locking policy makes node, link and bearer components closely bound together so that their relationship becomes unnecessarily complex. In the worst case, such locking policy not only has a negative influence on performance, but also it's prone to lead to deadlock occasionally. In order o decouple the complex relationship between bearer and node as well as link, the locking policy is adjusted as follows: - Bearer level RTNL lock is used on update side, and RCU is used on read side. Meanwhile, all bearer instances including broadcast bearer are saved into bearer_list array. - Node and link level All node instances are saved into two tipc_node_list and node_htable lists. The two lists are protected by node_list_lock on write side, and they are guarded with RCU lock on read side. All members in node structure including link instances are protected by node spin lock. - The relationship between bearer and node When link accesses bearer, it first needs to find the bearer with its bearer identity from the bearer_list array. When bearer accesses node, it can iterate the node_htable hash list with the node address to find the corresponding node. In the new locking policy, every component has its private locking solution and the relationship between bearer and node is very simple, that is, they can find each other with node address or bearer identity from node_htable hash list or bearer_list array. Until now above all changes have been done, so tipc_net_lock can be removed safely. Signed-off-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Tested-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-21 06:55:48 +04:00
* RCU and node lock set
*/
2015-02-05 16:36:36 +03:00
void tipc_bclink_update_link_state(struct tipc_node *n_ptr,
u32 last_sent)
{
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
struct sk_buff *buf;
2015-02-05 16:36:36 +03:00
struct net *net = n_ptr->net;
struct tipc_net *tn = net_generic(net, tipc_net_id);
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
/* Ignore "stale" link state info */
if (less_eq(last_sent, n_ptr->bclink.last_in))
return;
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
/* Update link synchronization state; quit if in sync */
bclink_update_last_sent(n_ptr, last_sent);
if (n_ptr->bclink.last_sent == n_ptr->bclink.last_in)
return;
/* Update out-of-sync state; quit if loss is still unconfirmed */
if ((++n_ptr->bclink.oos_state) == 1) {
if (n_ptr->bclink.deferred_size < (TIPC_MIN_LINK_WIN / 2))
return;
n_ptr->bclink.oos_state++;
}
/* Don't NACK if one has been recently sent (or seen) */
if (n_ptr->bclink.oos_state & 0x1)
return;
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
/* Send NACK */
buf = tipc_buf_acquire(INT_H_SIZE);
if (buf) {
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
struct tipc_msg *msg = buf_msg(buf);
struct sk_buff *skb = skb_peek(&n_ptr->bclink.deferdq);
u32 to = skb ? buf_seqno(skb) - 1 : n_ptr->bclink.last_sent;
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
2015-02-05 16:36:36 +03:00
tipc_msg_init(tn->own_addr, msg, BCAST_PROTOCOL, STATE_MSG,
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
INT_H_SIZE, n_ptr->addr);
msg_set_non_seq(msg, 1);
msg_set_mc_netid(msg, tn->net_id);
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
msg_set_bcast_ack(msg, n_ptr->bclink.last_in);
msg_set_bcgap_after(msg, n_ptr->bclink.last_in);
msg_set_bcgap_to(msg, to);
tipc_bclink_lock(net);
tipc_bearer_send(net, MAX_BEARERS, buf, NULL);
tn->bcl->stats.sent_nacks++;
tipc_bclink_unlock(net);
kfree_skb(buf);
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
n_ptr->bclink.oos_state++;
}
}
/**
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
* bclink_peek_nack - monitor retransmission requests sent by other nodes
*
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
* Delay any upcoming NACK by this node if another node has already
* requested the first message this node is going to ask for.
*/
static void bclink_peek_nack(struct net *net, struct tipc_msg *msg)
{
struct tipc_node *n_ptr = tipc_node_find(net, msg_destnode(msg));
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
if (unlikely(!n_ptr))
return;
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
tipc_node_lock(n_ptr);
if (n_ptr->bclink.recv_permitted &&
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
(n_ptr->bclink.last_in != n_ptr->bclink.last_sent) &&
(n_ptr->bclink.last_in == msg_bcgap_after(msg)))
n_ptr->bclink.oos_state = 2;
tipc_node_unlock(n_ptr);
tipc_node_put(n_ptr);
}
/* tipc_bclink_xmit - deliver buffer chain to all nodes in cluster
* and to identified node local sockets
* @net: the applicable net namespace
* @list: chain of buffers containing message
* Consumes the buffer chain, except when returning -ELINKCONG
* Returns 0 if success, otherwise errno: -ELINKCONG,-EHOSTUNREACH,-EMSGSIZE
*/
int tipc_bclink_xmit(struct net *net, struct sk_buff_head *list)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_link *bcl = tn->bcl;
struct tipc_bclink *bclink = tn->bclink;
int rc = 0;
int bc = 0;
struct sk_buff *skb;
struct sk_buff_head arrvq;
struct sk_buff_head inputq;
/* Prepare clone of message for local node */
skb = tipc_msg_reassemble(list);
if (unlikely(!skb))
return -EHOSTUNREACH;
/* Broadcast to all nodes */
if (likely(bclink)) {
tipc_bclink_lock(net);
if (likely(bclink->bcast_nodes.count)) {
rc = __tipc_link_xmit(net, bcl, list);
if (likely(!rc)) {
u32 len = skb_queue_len(&bcl->transmq);
bclink_set_last_sent(net);
bcl->stats.queue_sz_counts++;
bcl->stats.accu_queue_sz += len;
}
bc = 1;
}
tipc_bclink_unlock(net);
}
if (unlikely(!bc))
__skb_queue_purge(list);
if (unlikely(rc)) {
kfree_skb(skb);
return rc;
}
/* Deliver message clone */
__skb_queue_head_init(&arrvq);
skb_queue_head_init(&inputq);
__skb_queue_tail(&arrvq, skb);
tipc_sk_mcast_rcv(net, &arrvq, &inputq);
return rc;
}
/**
* bclink_accept_pkt - accept an incoming, in-sequence broadcast packet
*
* Called with both sending node's lock and bclink_lock taken.
*/
static void bclink_accept_pkt(struct tipc_node *node, u32 seqno)
{
struct tipc_net *tn = net_generic(node->net, tipc_net_id);
bclink_update_last_sent(node, seqno);
node->bclink.last_in = seqno;
node->bclink.oos_state = 0;
tn->bcl->stats.recv_info++;
/*
* Unicast an ACK periodically, ensuring that
* all nodes in the cluster don't ACK at the same time
*/
if (((seqno - tn->own_addr) % TIPC_MIN_LINK_WIN) == 0) {
tipc_link_proto_xmit(node_active_link(node, node->addr),
STATE_MSG, 0, 0, 0, 0);
tn->bcl->stats.sent_acks++;
}
}
/**
tipc: align tipc function names with common naming practice in the network Rename the following functions, which are shorter and more in line with common naming practice in the network subsystem. tipc_bclink_send_msg->tipc_bclink_xmit tipc_bclink_recv_pkt->tipc_bclink_rcv tipc_disc_recv_msg->tipc_disc_rcv tipc_link_send_proto_msg->tipc_link_proto_xmit link_recv_proto_msg->tipc_link_proto_rcv link_send_sections_long->tipc_link_iovec_long_xmit tipc_link_send_sections_fast->tipc_link_iovec_xmit_fast tipc_link_send_sync->tipc_link_sync_xmit tipc_link_recv_sync->tipc_link_sync_rcv tipc_link_send_buf->__tipc_link_xmit tipc_link_send->tipc_link_xmit tipc_link_send_names->tipc_link_names_xmit tipc_named_recv->tipc_named_rcv tipc_link_recv_bundle->tipc_link_bundle_rcv tipc_link_dup_send_queue->tipc_link_dup_queue_xmit link_send_long_buf->tipc_link_frag_xmit tipc_multicast->tipc_port_mcast_xmit tipc_port_recv_mcast->tipc_port_mcast_rcv tipc_port_reject_sections->tipc_port_iovec_reject tipc_port_recv_proto_msg->tipc_port_proto_rcv tipc_connect->tipc_port_connect __tipc_connect->__tipc_port_connect __tipc_disconnect->__tipc_port_disconnect tipc_disconnect->tipc_port_disconnect tipc_shutdown->tipc_port_shutdown tipc_port_recv_msg->tipc_port_rcv tipc_port_recv_sections->tipc_port_iovec_rcv release->tipc_release accept->tipc_accept bind->tipc_bind get_name->tipc_getname poll->tipc_poll send_msg->tipc_sendmsg send_packet->tipc_send_packet send_stream->tipc_send_stream recv_msg->tipc_recvmsg recv_stream->tipc_recv_stream connect->tipc_connect listen->tipc_listen shutdown->tipc_shutdown setsockopt->tipc_setsockopt getsockopt->tipc_getsockopt Above changes have no impact on current users of the functions. Signed-off-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-18 12:06:46 +04:00
* tipc_bclink_rcv - receive a broadcast packet, and deliver upwards
*
tipc: purge tipc_net_lock lock Now tipc routing hierarchy comprises the structures 'node', 'link'and 'bearer'. The whole hierarchy is protected by a big read/write lock, tipc_net_lock, to ensure that nothing is added or removed while code is accessing any of these structures. Obviously the locking policy makes node, link and bearer components closely bound together so that their relationship becomes unnecessarily complex. In the worst case, such locking policy not only has a negative influence on performance, but also it's prone to lead to deadlock occasionally. In order o decouple the complex relationship between bearer and node as well as link, the locking policy is adjusted as follows: - Bearer level RTNL lock is used on update side, and RCU is used on read side. Meanwhile, all bearer instances including broadcast bearer are saved into bearer_list array. - Node and link level All node instances are saved into two tipc_node_list and node_htable lists. The two lists are protected by node_list_lock on write side, and they are guarded with RCU lock on read side. All members in node structure including link instances are protected by node spin lock. - The relationship between bearer and node When link accesses bearer, it first needs to find the bearer with its bearer identity from the bearer_list array. When bearer accesses node, it can iterate the node_htable hash list with the node address to find the corresponding node. In the new locking policy, every component has its private locking solution and the relationship between bearer and node is very simple, that is, they can find each other with node address or bearer identity from node_htable hash list or bearer_list array. Until now above all changes have been done, so tipc_net_lock can be removed safely. Signed-off-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Tested-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-21 06:55:48 +04:00
* RCU is locked, no other locks set
*/
void tipc_bclink_rcv(struct net *net, struct sk_buff *buf)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_link *bcl = tn->bcl;
struct tipc_msg *msg = buf_msg(buf);
struct tipc_node *node;
u32 next_in;
u32 seqno;
int deferred = 0;
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 16:36:41 +03:00
int pos = 0;
struct sk_buff *iskb;
struct sk_buff_head *arrvq, *inputq;
/* Screen out unwanted broadcast messages */
if (msg_mc_netid(msg) != tn->net_id)
goto exit;
node = tipc_node_find(net, msg_prevnode(msg));
if (unlikely(!node))
goto exit;
tipc_node_lock(node);
if (unlikely(!node->bclink.recv_permitted))
goto unlock;
/* Handle broadcast protocol message */
if (unlikely(msg_user(msg) == BCAST_PROTOCOL)) {
if (msg_type(msg) != STATE_MSG)
goto unlock;
if (msg_destnode(msg) == tn->own_addr) {
tipc_bclink_acknowledge(node, msg_bcast_ack(msg));
tipc_bclink_lock(net);
bcl->stats.recv_nacks++;
tn->bclink->retransmit_to = node;
bclink_retransmit_pkt(tn, msg_bcgap_after(msg),
msg_bcgap_to(msg));
tipc_bclink_unlock(net);
tipc: fix potential deadlock when all links are reset [ 60.988363] ====================================================== [ 60.988754] [ INFO: possible circular locking dependency detected ] [ 60.989152] 3.19.0+ #194 Not tainted [ 60.989377] ------------------------------------------------------- [ 60.989781] swapper/3/0 is trying to acquire lock: [ 60.990079] (&(&n_ptr->lock)->rlock){+.-...}, at: [<ffffffffa0006dca>] tipc_link_retransmit+0x1aa/0x240 [tipc] [ 60.990743] [ 60.990743] but task is already holding lock: [ 60.991106] (&(&bclink->lock)->rlock){+.-...}, at: [<ffffffffa00004be>] tipc_bclink_lock+0x8e/0xa0 [tipc] [ 60.991738] [ 60.991738] which lock already depends on the new lock. [ 60.991738] [ 60.992174] [ 60.992174] the existing dependency chain (in reverse order) is: [ 60.992174] -> #1 (&(&bclink->lock)->rlock){+.-...}: [ 60.992174] [<ffffffff810a9c0c>] lock_acquire+0x9c/0x140 [ 60.992174] [<ffffffff8179c41f>] _raw_spin_lock_bh+0x3f/0x50 [ 60.992174] [<ffffffffa00004be>] tipc_bclink_lock+0x8e/0xa0 [tipc] [ 60.992174] [<ffffffffa0000f57>] tipc_bclink_add_node+0x97/0xf0 [tipc] [ 60.992174] [<ffffffffa0011815>] tipc_node_link_up+0xf5/0x110 [tipc] [ 60.992174] [<ffffffffa0007783>] link_state_event+0x2b3/0x4f0 [tipc] [ 60.992174] [<ffffffffa00193c0>] tipc_link_proto_rcv+0x24c/0x418 [tipc] [ 60.992174] [<ffffffffa0008857>] tipc_rcv+0x827/0xac0 [tipc] [ 60.992174] [<ffffffffa0002ca3>] tipc_l2_rcv_msg+0x73/0xd0 [tipc] [ 60.992174] [<ffffffff81646e66>] __netif_receive_skb_core+0x746/0x980 [ 60.992174] [<ffffffff816470c1>] __netif_receive_skb+0x21/0x70 [ 60.992174] [<ffffffff81647295>] netif_receive_skb_internal+0x35/0x130 [ 60.992174] [<ffffffff81648218>] napi_gro_receive+0x158/0x1d0 [ 60.992174] [<ffffffff81559e05>] e1000_clean_rx_irq+0x155/0x490 [ 60.992174] [<ffffffff8155c1b7>] e1000_clean+0x267/0x990 [ 60.992174] [<ffffffff81647b60>] net_rx_action+0x150/0x360 [ 60.992174] [<ffffffff8105ec43>] __do_softirq+0x123/0x360 [ 60.992174] [<ffffffff8105f12e>] irq_exit+0x8e/0xb0 [ 60.992174] [<ffffffff8179f9f5>] do_IRQ+0x65/0x110 [ 60.992174] [<ffffffff8179da6f>] ret_from_intr+0x0/0x13 [ 60.992174] [<ffffffff8100de9f>] arch_cpu_idle+0xf/0x20 [ 60.992174] [<ffffffff8109dfa6>] cpu_startup_entry+0x2f6/0x3f0 [ 60.992174] [<ffffffff81033cda>] start_secondary+0x13a/0x150 [ 60.992174] -> #0 (&(&n_ptr->lock)->rlock){+.-...}: [ 60.992174] [<ffffffff810a8f7d>] __lock_acquire+0x163d/0x1ca0 [ 60.992174] [<ffffffff810a9c0c>] lock_acquire+0x9c/0x140 [ 60.992174] [<ffffffff8179c41f>] _raw_spin_lock_bh+0x3f/0x50 [ 60.992174] [<ffffffffa0006dca>] tipc_link_retransmit+0x1aa/0x240 [tipc] [ 60.992174] [<ffffffffa0001e11>] tipc_bclink_rcv+0x611/0x640 [tipc] [ 60.992174] [<ffffffffa0008646>] tipc_rcv+0x616/0xac0 [tipc] [ 60.992174] [<ffffffffa0002ca3>] tipc_l2_rcv_msg+0x73/0xd0 [tipc] [ 60.992174] [<ffffffff81646e66>] __netif_receive_skb_core+0x746/0x980 [ 60.992174] [<ffffffff816470c1>] __netif_receive_skb+0x21/0x70 [ 60.992174] [<ffffffff81647295>] netif_receive_skb_internal+0x35/0x130 [ 60.992174] [<ffffffff81648218>] napi_gro_receive+0x158/0x1d0 [ 60.992174] [<ffffffff81559e05>] e1000_clean_rx_irq+0x155/0x490 [ 60.992174] [<ffffffff8155c1b7>] e1000_clean+0x267/0x990 [ 60.992174] [<ffffffff81647b60>] net_rx_action+0x150/0x360 [ 60.992174] [<ffffffff8105ec43>] __do_softirq+0x123/0x360 [ 60.992174] [<ffffffff8105f12e>] irq_exit+0x8e/0xb0 [ 60.992174] [<ffffffff8179f9f5>] do_IRQ+0x65/0x110 [ 60.992174] [<ffffffff8179da6f>] ret_from_intr+0x0/0x13 [ 60.992174] [<ffffffff8100de9f>] arch_cpu_idle+0xf/0x20 [ 60.992174] [<ffffffff8109dfa6>] cpu_startup_entry+0x2f6/0x3f0 [ 60.992174] [<ffffffff81033cda>] start_secondary+0x13a/0x150 [ 60.992174] [ 60.992174] other info that might help us debug this: [ 60.992174] [ 60.992174] Possible unsafe locking scenario: [ 60.992174] [ 60.992174] CPU0 CPU1 [ 60.992174] ---- ---- [ 60.992174] lock(&(&bclink->lock)->rlock); [ 60.992174] lock(&(&n_ptr->lock)->rlock); [ 60.992174] lock(&(&bclink->lock)->rlock); [ 60.992174] lock(&(&n_ptr->lock)->rlock); [ 60.992174] [ 60.992174] *** DEADLOCK *** [ 60.992174] [ 60.992174] 3 locks held by swapper/3/0: [ 60.992174] #0: (rcu_read_lock){......}, at: [<ffffffff81646791>] __netif_receive_skb_core+0x71/0x980 [ 60.992174] #1: (rcu_read_lock){......}, at: [<ffffffffa0002c35>] tipc_l2_rcv_msg+0x5/0xd0 [tipc] [ 60.992174] #2: (&(&bclink->lock)->rlock){+.-...}, at: [<ffffffffa00004be>] tipc_bclink_lock+0x8e/0xa0 [tipc] [ 60.992174] The correct the sequence of grabbing n_ptr->lock and bclink->lock should be that the former is first held and the latter is then taken, which exactly happened on CPU1. But especially when the retransmission of broadcast link is failed, bclink->lock is first held in tipc_bclink_rcv(), and n_ptr->lock is taken in link_retransmit_failure() called by tipc_link_retransmit() subsequently, which is demonstrated on CPU0. As a result, deadlock occurs. If the order of holding the two locks happening on CPU0 is reversed, the deadlock risk will be relieved. Therefore, the node lock taken in link_retransmit_failure() originally is moved to tipc_bclink_rcv() so that it's obtained before bclink lock. But the precondition of the adjustment of node lock is that responding to bclink reset event must be moved from tipc_bclink_unlock() to tipc_node_unlock(). Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-26 13:10:23 +03:00
tipc_node_unlock(node);
} else {
tipc_node_unlock(node);
bclink_peek_nack(net, msg);
}
tipc_node_put(node);
goto exit;
}
/* Handle in-sequence broadcast message */
seqno = msg_seqno(msg);
next_in = mod(node->bclink.last_in + 1);
arrvq = &tn->bclink->arrvq;
inputq = &tn->bclink->inputq;
if (likely(seqno == next_in)) {
receive:
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
/* Deliver message to destination */
if (likely(msg_isdata(msg))) {
tipc_bclink_lock(net);
bclink_accept_pkt(node, seqno);
spin_lock_bh(&inputq->lock);
__skb_queue_tail(arrvq, buf);
spin_unlock_bh(&inputq->lock);
node->action_flags |= TIPC_BCAST_MSG_EVT;
tipc_bclink_unlock(net);
tipc_node_unlock(node);
} else if (msg_user(msg) == MSG_BUNDLER) {
tipc_bclink_lock(net);
bclink_accept_pkt(node, seqno);
bcl->stats.recv_bundles++;
bcl->stats.recv_bundled += msg_msgcnt(msg);
pos = 0;
while (tipc_msg_extract(buf, &iskb, &pos)) {
spin_lock_bh(&inputq->lock);
__skb_queue_tail(arrvq, iskb);
spin_unlock_bh(&inputq->lock);
}
node->action_flags |= TIPC_BCAST_MSG_EVT;
tipc_bclink_unlock(net);
tipc_node_unlock(node);
} else if (msg_user(msg) == MSG_FRAGMENTER) {
tipc_bclink_lock(net);
bclink_accept_pkt(node, seqno);
tipc_buf_append(&node->bclink.reasm_buf, &buf);
if (unlikely(!buf && !node->bclink.reasm_buf)) {
tipc_bclink_unlock(net);
goto unlock;
}
bcl->stats.recv_fragments++;
if (buf) {
bcl->stats.recv_fragmented++;
tipc: message reassembly using fragment chain When the first fragment of a long data data message is received on a link, a reassembly buffer large enough to hold the data from this and all subsequent fragments of the message is allocated. The payload of each new fragment is copied into this buffer upon arrival. When the last fragment is received, the reassembled message is delivered upwards to the port/socket layer. Not only is this an inefficient approach, but it may also cause bursts of reassembly failures in low memory situations. since we may fail to allocate the necessary large buffer in the first place. Furthermore, after 100 subsequent such failures the link will be reset, something that in reality aggravates the situation. To remedy this problem, this patch introduces a different approach. Instead of allocating a big reassembly buffer, we now append the arriving fragments to a reassembly chain on the link, and deliver the whole chain up to the socket layer once the last fragment has been received. This is safe because the retransmission layer of a TIPC link always delivers packets in strict uninterrupted order, to the reassembly layer as to all other upper layers. Hence there can never be more than one fragment chain pending reassembly at any given time in a link, and we can trust (but still verify) that the fragments will be chained up in the correct order. Signed-off-by: Erik Hugne <erik.hugne@ericsson.com> Reviewed-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-06 12:28:06 +04:00
msg = buf_msg(buf);
tipc_bclink_unlock(net);
goto receive;
}
tipc_bclink_unlock(net);
tipc_node_unlock(node);
} else {
tipc_bclink_lock(net);
bclink_accept_pkt(node, seqno);
tipc_bclink_unlock(net);
tipc_node_unlock(node);
kfree_skb(buf);
}
buf = NULL;
/* Determine new synchronization state */
tipc_node_lock(node);
if (unlikely(!tipc_node_is_up(node)))
goto unlock;
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
if (node->bclink.last_in == node->bclink.last_sent)
goto unlock;
if (skb_queue_empty(&node->bclink.deferdq)) {
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
node->bclink.oos_state = 1;
goto unlock;
}
msg = buf_msg(skb_peek(&node->bclink.deferdq));
seqno = msg_seqno(msg);
next_in = mod(next_in + 1);
if (seqno != next_in)
goto unlock;
/* Take in-sequence message from deferred queue & deliver it */
buf = __skb_dequeue(&node->bclink.deferdq);
goto receive;
}
/* Handle out-of-sequence broadcast message */
if (less(next_in, seqno)) {
deferred = tipc_link_defer_pkt(&node->bclink.deferdq,
buf);
tipc: Major redesign of broadcast link ACK/NACK algorithms Completely redesigns broadcast link ACK and NACK mechanisms to prevent spurious retransmit requests in dual LAN networks, and to prevent the broadcast link from stalling due to the failure of a receiving node to acknowledge receiving a broadcast message or request its retransmission. Note: These changes only impact the timing of when ACK and NACK messages are sent, and not the basic broadcast link protocol itself, so inter- operability with nodes using the "classic" algorithms is maintained. The revised algorithms are as follows: 1) An explicit ACK message is still sent after receiving 16 in-sequence messages, and implicit ACK information continues to be carried in other unicast link message headers (including link state messages). However, the timing of explicit ACKs is now based on the receiving node's absolute network address rather than its relative network address to ensure that the failure of another node does not delay the ACK beyond its 16 message target. 2) A NACK message is now typically sent only when a message gap persists for two consecutive incoming link state messages; this ensures that a suspected gap is not confirmed until both LANs in a dual LAN network have had an opportunity to deliver the message, thereby preventing spurious NACKs. A NACK message can also be generated by the arrival of a single link state message, if the deferred queue is so big that the current message gap cannot be the result of "normal" mis-ordering due to the use of dual LANs (or one LAN using a bonded interface). Since link state messages typically arrive at different nodes at different times the problem of multiple nodes issuing identical NACKs simultaneously is inherently avoided. 3) Nodes continue to "peek" at NACK messages sent by other nodes. If another node requests retransmission of a message gap suspected (but not yet confirmed) by the peeking node, the peeking node forgets about the gap and does not generate a duplicate retransmit request. (If the peeking node subsequently fails to receive the lost message, later link state messages will cause it to rediscover and confirm the gap and send another NACK.) 4) Message gap "equality" is now determined by the start of the gap only. This is sufficient to deal with the most common cases of message loss, and eliminates the need for complex end of gap computations. 5) A peeking node no longer tries to determine whether it should send a complementary NACK, since the most common cases of message loss don't require it to be sent. Consequently, the node no longer examines the "broadcast tag" field of a NACK message when peeking. Signed-off-by: Allan Stephens <allan.stephens@windriver.com> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
2011-10-27 22:17:53 +04:00
bclink_update_last_sent(node, seqno);
buf = NULL;
}
tipc_bclink_lock(net);
if (deferred)
bcl->stats.deferred_recv++;
else
bcl->stats.duplicates++;
tipc_bclink_unlock(net);
unlock:
tipc_node_unlock(node);
tipc_node_put(node);
exit:
kfree_skb(buf);
}
u32 tipc_bclink_acks_missing(struct tipc_node *n_ptr)
{
return (n_ptr->bclink.recv_permitted &&
(tipc_bclink_get_last_sent(n_ptr->net) != n_ptr->bclink.acked));
}
/**
* tipc_bcbearer_send - send a packet through the broadcast pseudo-bearer
*
* Send packet over as many bearers as necessary to reach all nodes
* that have joined the broadcast link.
*
* Returns 0 (packet sent successfully) under all circumstances,
* since the broadcast link's pseudo-bearer never blocks
*/
static int tipc_bcbearer_send(struct net *net, struct sk_buff *buf,
struct tipc_bearer *unused1,
struct tipc_media_addr *unused2)
{
int bp_index;
struct tipc_msg *msg = buf_msg(buf);
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_bcbearer *bcbearer = tn->bcbearer;
struct tipc_bclink *bclink = tn->bclink;
/* Prepare broadcast link message for reliable transmission,
* if first time trying to send it;
* preparation is skipped for broadcast link protocol messages
* since they are sent in an unreliable manner and don't need it
*/
if (likely(!msg_non_seq(buf_msg(buf)))) {
bcbuf_set_acks(buf, bclink->bcast_nodes.count);
msg_set_non_seq(msg, 1);
msg_set_mc_netid(msg, tn->net_id);
tn->bcl->stats.sent_info++;
if (WARN_ON(!bclink->bcast_nodes.count)) {
dump_stack();
return 0;
}
}
/* Send buffer over bearers until all targets reached */
bcbearer->remains = bclink->bcast_nodes;
for (bp_index = 0; bp_index < MAX_BEARERS; bp_index++) {
struct tipc_bearer *p = bcbearer->bpairs[bp_index].primary;
struct tipc_bearer *s = bcbearer->bpairs[bp_index].secondary;
struct tipc_bearer *bp[2] = {p, s};
struct tipc_bearer *b = bp[msg_link_selector(msg)];
struct sk_buff *tbuf;
if (!p)
break; /* No more bearers to try */
if (!b)
b = p;
tipc_nmap_diff(&bcbearer->remains, &b->nodes,
&bcbearer->remains_new);
if (bcbearer->remains_new.count == bcbearer->remains.count)
continue; /* Nothing added by bearer pair */
if (bp_index == 0) {
/* Use original buffer for first bearer */
tipc_bearer_send(net, b->identity, buf, &b->bcast_addr);
} else {
/* Avoid concurrent buffer access */
tbuf = pskb_copy_for_clone(buf, GFP_ATOMIC);
if (!tbuf)
break;
tipc_bearer_send(net, b->identity, tbuf,
&b->bcast_addr);
kfree_skb(tbuf); /* Bearer keeps a clone */
}
if (bcbearer->remains_new.count == 0)
break; /* All targets reached */
bcbearer->remains = bcbearer->remains_new;
}
return 0;
}
/**
* tipc_bcbearer_sort - create sets of bearer pairs used by broadcast bearer
*/
void tipc_bcbearer_sort(struct net *net, struct tipc_node_map *nm_ptr,
u32 node, bool action)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_bcbearer *bcbearer = tn->bcbearer;
struct tipc_bcbearer_pair *bp_temp = bcbearer->bpairs_temp;
struct tipc_bcbearer_pair *bp_curr;
struct tipc_bearer *b;
int b_index;
int pri;
tipc_bclink_lock(net);
if (action)
tipc_nmap_add(nm_ptr, node);
else
tipc_nmap_remove(nm_ptr, node);
/* Group bearers by priority (can assume max of two per priority) */
memset(bp_temp, 0, sizeof(bcbearer->bpairs_temp));
rcu_read_lock();
for (b_index = 0; b_index < MAX_BEARERS; b_index++) {
b = rcu_dereference_rtnl(tn->bearer_list[b_index]);
if (!b || !b->nodes.count)
continue;
if (!bp_temp[b->priority].primary)
bp_temp[b->priority].primary = b;
else
bp_temp[b->priority].secondary = b;
}
rcu_read_unlock();
/* Create array of bearer pairs for broadcasting */
bp_curr = bcbearer->bpairs;
memset(bcbearer->bpairs, 0, sizeof(bcbearer->bpairs));
for (pri = TIPC_MAX_LINK_PRI; pri >= 0; pri--) {
if (!bp_temp[pri].primary)
continue;
bp_curr->primary = bp_temp[pri].primary;
if (bp_temp[pri].secondary) {
if (tipc_nmap_equal(&bp_temp[pri].primary->nodes,
&bp_temp[pri].secondary->nodes)) {
bp_curr->secondary = bp_temp[pri].secondary;
} else {
bp_curr++;
bp_curr->primary = bp_temp[pri].secondary;
}
}
bp_curr++;
}
tipc_bclink_unlock(net);
}
static int __tipc_nl_add_bc_link_stat(struct sk_buff *skb,
struct tipc_stats *stats)
{
int i;
struct nlattr *nest;
struct nla_map {
__u32 key;
__u32 val;
};
struct nla_map map[] = {
{TIPC_NLA_STATS_RX_INFO, stats->recv_info},
{TIPC_NLA_STATS_RX_FRAGMENTS, stats->recv_fragments},
{TIPC_NLA_STATS_RX_FRAGMENTED, stats->recv_fragmented},
{TIPC_NLA_STATS_RX_BUNDLES, stats->recv_bundles},
{TIPC_NLA_STATS_RX_BUNDLED, stats->recv_bundled},
{TIPC_NLA_STATS_TX_INFO, stats->sent_info},
{TIPC_NLA_STATS_TX_FRAGMENTS, stats->sent_fragments},
{TIPC_NLA_STATS_TX_FRAGMENTED, stats->sent_fragmented},
{TIPC_NLA_STATS_TX_BUNDLES, stats->sent_bundles},
{TIPC_NLA_STATS_TX_BUNDLED, stats->sent_bundled},
{TIPC_NLA_STATS_RX_NACKS, stats->recv_nacks},
{TIPC_NLA_STATS_RX_DEFERRED, stats->deferred_recv},
{TIPC_NLA_STATS_TX_NACKS, stats->sent_nacks},
{TIPC_NLA_STATS_TX_ACKS, stats->sent_acks},
{TIPC_NLA_STATS_RETRANSMITTED, stats->retransmitted},
{TIPC_NLA_STATS_DUPLICATES, stats->duplicates},
{TIPC_NLA_STATS_LINK_CONGS, stats->link_congs},
{TIPC_NLA_STATS_MAX_QUEUE, stats->max_queue_sz},
{TIPC_NLA_STATS_AVG_QUEUE, stats->queue_sz_counts ?
(stats->accu_queue_sz / stats->queue_sz_counts) : 0}
};
nest = nla_nest_start(skb, TIPC_NLA_LINK_STATS);
if (!nest)
return -EMSGSIZE;
for (i = 0; i < ARRAY_SIZE(map); i++)
if (nla_put_u32(skb, map[i].key, map[i].val))
goto msg_full;
nla_nest_end(skb, nest);
return 0;
msg_full:
nla_nest_cancel(skb, nest);
return -EMSGSIZE;
}
int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg)
{
int err;
void *hdr;
struct nlattr *attrs;
struct nlattr *prop;
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_link *bcl = tn->bcl;
if (!bcl)
return 0;
tipc_bclink_lock(net);
hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family,
NLM_F_MULTI, TIPC_NL_LINK_GET);
if (!hdr)
return -EMSGSIZE;
attrs = nla_nest_start(msg->skb, TIPC_NLA_LINK);
if (!attrs)
goto msg_full;
/* The broadcast link is always up */
if (nla_put_flag(msg->skb, TIPC_NLA_LINK_UP))
goto attr_msg_full;
if (nla_put_flag(msg->skb, TIPC_NLA_LINK_BROADCAST))
goto attr_msg_full;
if (nla_put_string(msg->skb, TIPC_NLA_LINK_NAME, bcl->name))
goto attr_msg_full;
if (nla_put_u32(msg->skb, TIPC_NLA_LINK_RX, bcl->rcv_nxt))
goto attr_msg_full;
if (nla_put_u32(msg->skb, TIPC_NLA_LINK_TX, bcl->snd_nxt))
goto attr_msg_full;
prop = nla_nest_start(msg->skb, TIPC_NLA_LINK_PROP);
if (!prop)
goto attr_msg_full;
if (nla_put_u32(msg->skb, TIPC_NLA_PROP_WIN, bcl->window))
goto prop_msg_full;
nla_nest_end(msg->skb, prop);
err = __tipc_nl_add_bc_link_stat(msg->skb, &bcl->stats);
if (err)
goto attr_msg_full;
tipc_bclink_unlock(net);
nla_nest_end(msg->skb, attrs);
genlmsg_end(msg->skb, hdr);
return 0;
prop_msg_full:
nla_nest_cancel(msg->skb, prop);
attr_msg_full:
nla_nest_cancel(msg->skb, attrs);
msg_full:
tipc_bclink_unlock(net);
genlmsg_cancel(msg->skb, hdr);
return -EMSGSIZE;
}
int tipc_bclink_reset_stats(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_link *bcl = tn->bcl;
if (!bcl)
return -ENOPROTOOPT;
tipc_bclink_lock(net);
memset(&bcl->stats, 0, sizeof(bcl->stats));
tipc_bclink_unlock(net);
return 0;
}
int tipc_bclink_set_queue_limits(struct net *net, u32 limit)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_link *bcl = tn->bcl;
if (!bcl)
return -ENOPROTOOPT;
if ((limit < TIPC_MIN_LINK_WIN) || (limit > TIPC_MAX_LINK_WIN))
return -EINVAL;
tipc_bclink_lock(net);
tipc_link_set_queue_limits(bcl, limit);
tipc_bclink_unlock(net);
return 0;
}
int tipc_nl_bc_link_set(struct net *net, struct nlattr *attrs[])
{
int err;
u32 win;
struct nlattr *props[TIPC_NLA_PROP_MAX + 1];
if (!attrs[TIPC_NLA_LINK_PROP])
return -EINVAL;
err = tipc_nl_parse_link_prop(attrs[TIPC_NLA_LINK_PROP], props);
if (err)
return err;
if (!props[TIPC_NLA_PROP_WIN])
return -EOPNOTSUPP;
win = nla_get_u32(props[TIPC_NLA_PROP_WIN]);
return tipc_bclink_set_queue_limits(net, win);
}
int tipc_bclink_init(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct tipc_bcbearer *bcbearer;
struct tipc_bclink *bclink;
struct tipc_link *bcl;
bcbearer = kzalloc(sizeof(*bcbearer), GFP_ATOMIC);
if (!bcbearer)
return -ENOMEM;
bclink = kzalloc(sizeof(*bclink), GFP_ATOMIC);
if (!bclink) {
kfree(bcbearer);
return -ENOMEM;
}
bcl = &bclink->link;
bcbearer->bearer.media = &bcbearer->media;
bcbearer->media.send_msg = tipc_bcbearer_send;
sprintf(bcbearer->media.name, "tipc-broadcast");
spin_lock_init(&bclink->lock);
__skb_queue_head_init(&bcl->transmq);
__skb_queue_head_init(&bcl->backlogq);
__skb_queue_head_init(&bcl->deferdq);
tipc: resolve race problem at unicast message reception TIPC handles message cardinality and sequencing at the link layer, before passing messages upwards to the destination sockets. During the upcall from link to socket no locks are held. It is therefore possible, and we see it happen occasionally, that messages arriving in different threads and delivered in sequence still bypass each other before they reach the destination socket. This must not happen, since it violates the sequentiality guarantee. We solve this by adding a new input buffer queue to the link structure. Arriving messages are added safely to the tail of that queue by the link, while the head of the queue is consumed, also safely, by the receiving socket. Sequentiality is secured per socket by only allowing buffers to be dequeued inside the socket lock. Since there may be multiple simultaneous readers of the queue, we use a 'filter' parameter to reduce the risk that they peek the same buffer from the queue, hence also reducing the risk of contention on the receiving socket locks. This solves the sequentiality problem, and seems to cause no measurable performance degradation. A nice side effect of this change is that lock handling in the functions tipc_rcv() and tipc_bcast_rcv() now becomes uniform, something that will enable future simplifications of those functions. Reviewed-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-05 16:36:41 +03:00
skb_queue_head_init(&bcl->wakeupq);
bcl->snd_nxt = 1;
spin_lock_init(&bclink->node.lock);
__skb_queue_head_init(&bclink->arrvq);
skb_queue_head_init(&bclink->inputq);
bcl->owner = &bclink->node;
bcl->owner->net = net;
bcl->mtu = MAX_PKT_DEFAULT_MCAST;
tipc_link_set_queue_limits(bcl, BCLINK_WIN_DEFAULT);
bcl->bearer_id = MAX_BEARERS;
rcu_assign_pointer(tn->bearer_list[MAX_BEARERS], &bcbearer->bearer);
2015-02-05 16:36:36 +03:00
bcl->pmsg = (struct tipc_msg *)&bcl->proto_msg;
msg_set_prevnode(bcl->pmsg, tn->own_addr);
strlcpy(bcl->name, tipc_bclink_name, TIPC_MAX_LINK_NAME);
tn->bcbearer = bcbearer;
tn->bclink = bclink;
tn->bcl = bcl;
return 0;
}
void tipc_bclink_stop(struct net *net)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
tipc_bclink_lock(net);
tipc_link_purge_queues(tn->bcl);
tipc_bclink_unlock(net);
RCU_INIT_POINTER(tn->bearer_list[BCBEARER], NULL);
synchronize_net();
kfree(tn->bcbearer);
kfree(tn->bclink);
}
/**
* tipc_nmap_add - add a node to a node map
*/
static void tipc_nmap_add(struct tipc_node_map *nm_ptr, u32 node)
{
int n = tipc_node(node);
int w = n / WSIZE;
u32 mask = (1 << (n % WSIZE));
if ((nm_ptr->map[w] & mask) == 0) {
nm_ptr->count++;
nm_ptr->map[w] |= mask;
}
}
/**
* tipc_nmap_remove - remove a node from a node map
*/
static void tipc_nmap_remove(struct tipc_node_map *nm_ptr, u32 node)
{
int n = tipc_node(node);
int w = n / WSIZE;
u32 mask = (1 << (n % WSIZE));
if ((nm_ptr->map[w] & mask) != 0) {
nm_ptr->map[w] &= ~mask;
nm_ptr->count--;
}
}
/**
* tipc_nmap_diff - find differences between node maps
* @nm_a: input node map A
* @nm_b: input node map B
* @nm_diff: output node map A-B (i.e. nodes of A that are not in B)
*/
static void tipc_nmap_diff(struct tipc_node_map *nm_a,
struct tipc_node_map *nm_b,
struct tipc_node_map *nm_diff)
{
int stop = ARRAY_SIZE(nm_a->map);
int w;
int b;
u32 map;
memset(nm_diff, 0, sizeof(*nm_diff));
for (w = 0; w < stop; w++) {
map = nm_a->map[w] ^ (nm_a->map[w] & nm_b->map[w]);
nm_diff->map[w] = map;
if (map != 0) {
for (b = 0 ; b < WSIZE; b++) {
if (map & (1 << b))
nm_diff->count++;
}
}
}
}