linux_dsm_epyc7002/net/tipc/name_distr.c

409 lines
11 KiB
C
Raw Normal View History

/*
* net/tipc/name_distr.c: TIPC name distribution code
*
tipc: add name distributor resiliency queue TIPC name table updates are distributed asynchronously in a cluster, entailing a risk of certain race conditions. E.g., if two nodes simultaneously issue conflicting (overlapping) publications, this may not be detected until both publications have reached a third node, in which case one of the publications will be silently dropped on that node. Hence, we end up with an inconsistent name table. In most cases this conflict is just a temporary race, e.g., one node is issuing a publication under the assumption that a previous, conflicting, publication has already been withdrawn by the other node. However, because of the (rtt related) distributed update delay, this may not yet hold true on all nodes. The symptom of this failure is a syslog message: "tipc: Cannot publish {%u,%u,%u}, overlap error". In this commit we add a resiliency queue at the receiving end of the name table distributor. When insertion of an arriving publication fails, we retain it in this queue for a short amount of time, assuming that another update will arrive very soon and clear the conflict. If so happens, we insert the publication, otherwise we drop it. The (configurable) retention value defaults to 2000 ms. Knowing from experience that the situation described above is extremely rare, there is no risk that the queue will accumulate any large number of items. Signed-off-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Acked-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-28 14:08:47 +07:00
* Copyright (c) 2000-2006, 2014, Ericsson AB
* 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 "core.h"
#include "link.h"
#include "name_distr.h"
tipc: add name distributor resiliency queue TIPC name table updates are distributed asynchronously in a cluster, entailing a risk of certain race conditions. E.g., if two nodes simultaneously issue conflicting (overlapping) publications, this may not be detected until both publications have reached a third node, in which case one of the publications will be silently dropped on that node. Hence, we end up with an inconsistent name table. In most cases this conflict is just a temporary race, e.g., one node is issuing a publication under the assumption that a previous, conflicting, publication has already been withdrawn by the other node. However, because of the (rtt related) distributed update delay, this may not yet hold true on all nodes. The symptom of this failure is a syslog message: "tipc: Cannot publish {%u,%u,%u}, overlap error". In this commit we add a resiliency queue at the receiving end of the name table distributor. When insertion of an arriving publication fails, we retain it in this queue for a short amount of time, assuming that another update will arrive very soon and clear the conflict. If so happens, we insert the publication, otherwise we drop it. The (configurable) retention value defaults to 2000 ms. Knowing from experience that the situation described above is extremely rare, there is no risk that the queue will accumulate any large number of items. Signed-off-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Acked-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-08-28 14:08:47 +07:00
int sysctl_tipc_named_timeout __read_mostly = 2000;
struct distr_queue_item {
struct distr_item i;
u32 dtype;
u32 node;
unsigned long expires;
struct list_head next;
};
/**
* publ_to_item - add publication info to a publication message
*/
static void publ_to_item(struct distr_item *i, struct publication *p)
{
i->type = htonl(p->type);
i->lower = htonl(p->lower);
i->upper = htonl(p->upper);
i->port = htonl(p->port);
i->key = htonl(p->key);
}
/**
* named_prepare_buf - allocate & initialize a publication message
*
* The buffer returned is of size INT_H_SIZE + payload size
*/
static struct sk_buff *named_prepare_buf(struct net *net, u32 type, u32 size,
u32 dest)
{
struct sk_buff *buf = tipc_buf_acquire(INT_H_SIZE + size, GFP_ATOMIC);
u32 self = tipc_own_addr(net);
struct tipc_msg *msg;
if (buf != NULL) {
msg = buf_msg(buf);
tipc_msg_init(self, msg, NAME_DISTRIBUTOR,
type, INT_H_SIZE, dest);
msg_set_size(msg, INT_H_SIZE + size);
}
return buf;
}
/**
* tipc_named_publish - tell other nodes about a new publication by this node
*/
struct sk_buff *tipc_named_publish(struct net *net, struct publication *publ)
{
struct name_table *nt = tipc_name_table(net);
struct distr_item *item;
struct sk_buff *skb;
if (publ->scope == TIPC_NODE_SCOPE) {
list_add_tail_rcu(&publ->binding_node, &nt->node_scope);
return NULL;
}
tipc: eliminate message disordering during binding table update We have seen the following race scenario: 1) named_distribute() builds a "bulk" message, containing a PUBLISH item for a certain publication. This is based on the contents of the binding tables's 'cluster_scope' list. 2) tipc_named_withdraw() removes the same publication from the list, bulds a WITHDRAW message and distributes it to all cluster nodes. 3) tipc_named_node_up(), which was calling named_distribute(), sends out the bulk message built under 1) 4) The WITHDRAW message arrives at the just detected node, finds no corresponding publication, and is dropped. 5) The PUBLISH item arrives at the same node, is added to its binding table, and remains there forever. This arrival disordering was earlier taken care of by the backlog queue, originally added for a different purpose, which was removed in the commit referred to below, but we now need a different solution. In this commit, we replace the rcu lock protecting the 'cluster_scope' list with a regular RW lock which comprises even the sending of the bulk message. This both guarantees both the list integrity and the message sending order. We will later add a commit which cleans up this code further. Note that this commit needs recently added commit d3092b2efca1 ("tipc: fix unsafe rcu locking when accessing publication list") to apply cleanly. Fixes: 37922ea4a310 ("tipc: permit overlapping service ranges in name table") Reported-by: Tuong Lien Tong <tuong.t.lien@dektech.com.au> Acked-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>
2018-10-20 00:55:40 +07:00
write_lock_bh(&nt->cluster_scope_lock);
list_add_tail(&publ->binding_node, &nt->cluster_scope);
write_unlock_bh(&nt->cluster_scope_lock);
skb = named_prepare_buf(net, PUBLICATION, ITEM_SIZE, 0);
if (!skb) {
pr_warn("Publication distribution failure\n");
return NULL;
}
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
msg_set_named_seqno(buf_msg(skb), nt->snd_nxt++);
msg_set_non_legacy(buf_msg(skb));
item = (struct distr_item *)msg_data(buf_msg(skb));
publ_to_item(item, publ);
return skb;
}
/**
* tipc_named_withdraw - tell other nodes about a withdrawn publication by this node
*/
struct sk_buff *tipc_named_withdraw(struct net *net, struct publication *publ)
{
tipc: eliminate message disordering during binding table update We have seen the following race scenario: 1) named_distribute() builds a "bulk" message, containing a PUBLISH item for a certain publication. This is based on the contents of the binding tables's 'cluster_scope' list. 2) tipc_named_withdraw() removes the same publication from the list, bulds a WITHDRAW message and distributes it to all cluster nodes. 3) tipc_named_node_up(), which was calling named_distribute(), sends out the bulk message built under 1) 4) The WITHDRAW message arrives at the just detected node, finds no corresponding publication, and is dropped. 5) The PUBLISH item arrives at the same node, is added to its binding table, and remains there forever. This arrival disordering was earlier taken care of by the backlog queue, originally added for a different purpose, which was removed in the commit referred to below, but we now need a different solution. In this commit, we replace the rcu lock protecting the 'cluster_scope' list with a regular RW lock which comprises even the sending of the bulk message. This both guarantees both the list integrity and the message sending order. We will later add a commit which cleans up this code further. Note that this commit needs recently added commit d3092b2efca1 ("tipc: fix unsafe rcu locking when accessing publication list") to apply cleanly. Fixes: 37922ea4a310 ("tipc: permit overlapping service ranges in name table") Reported-by: Tuong Lien Tong <tuong.t.lien@dektech.com.au> Acked-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>
2018-10-20 00:55:40 +07:00
struct name_table *nt = tipc_name_table(net);
struct distr_item *item;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
struct sk_buff *skb;
tipc: eliminate message disordering during binding table update We have seen the following race scenario: 1) named_distribute() builds a "bulk" message, containing a PUBLISH item for a certain publication. This is based on the contents of the binding tables's 'cluster_scope' list. 2) tipc_named_withdraw() removes the same publication from the list, bulds a WITHDRAW message and distributes it to all cluster nodes. 3) tipc_named_node_up(), which was calling named_distribute(), sends out the bulk message built under 1) 4) The WITHDRAW message arrives at the just detected node, finds no corresponding publication, and is dropped. 5) The PUBLISH item arrives at the same node, is added to its binding table, and remains there forever. This arrival disordering was earlier taken care of by the backlog queue, originally added for a different purpose, which was removed in the commit referred to below, but we now need a different solution. In this commit, we replace the rcu lock protecting the 'cluster_scope' list with a regular RW lock which comprises even the sending of the bulk message. This both guarantees both the list integrity and the message sending order. We will later add a commit which cleans up this code further. Note that this commit needs recently added commit d3092b2efca1 ("tipc: fix unsafe rcu locking when accessing publication list") to apply cleanly. Fixes: 37922ea4a310 ("tipc: permit overlapping service ranges in name table") Reported-by: Tuong Lien Tong <tuong.t.lien@dektech.com.au> Acked-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>
2018-10-20 00:55:40 +07:00
write_lock_bh(&nt->cluster_scope_lock);
list_del(&publ->binding_node);
write_unlock_bh(&nt->cluster_scope_lock);
if (publ->scope == TIPC_NODE_SCOPE)
return NULL;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
skb = named_prepare_buf(net, WITHDRAWAL, ITEM_SIZE, 0);
if (!skb) {
pr_warn("Withdrawal distribution failure\n");
return NULL;
}
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
msg_set_named_seqno(buf_msg(skb), nt->snd_nxt++);
msg_set_non_legacy(buf_msg(skb));
item = (struct distr_item *)msg_data(buf_msg(skb));
publ_to_item(item, publ);
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
return skb;
}
/**
* named_distribute - prepare name info for bulk distribution to another node
* @list: list of messages (buffers) to be returned from this function
* @dnode: node to be updated
* @pls: linked list of publication items to be packed into buffer chain
*/
static void named_distribute(struct net *net, struct sk_buff_head *list,
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
u32 dnode, struct list_head *pls, u16 seqno)
{
struct publication *publ;
struct sk_buff *skb = NULL;
struct distr_item *item = NULL;
tipc: improve throughput between nodes in netns Currently, TIPC transports intra-node user data messages directly socket to socket, hence shortcutting all the lower layers of the communication stack. This gives TIPC very good intra node performance, both regarding throughput and latency. We now introduce a similar mechanism for TIPC data traffic across network namespaces located in the same kernel. On the send path, the call chain is as always accompanied by the sending node's network name space pointer. However, once we have reliably established that the receiving node is represented by a namespace on the same host, we just replace the namespace pointer with the receiving node/namespace's ditto, and follow the regular socket receive patch though the receiving node. This technique gives us a throughput similar to the node internal throughput, several times larger than if we let the traffic go though the full network stacks. As a comparison, max throughput for 64k messages is four times larger than TCP throughput for the same type of traffic. To meet any security concerns, the following should be noted. - All nodes joining a cluster are supposed to have been be certified and authenticated by mechanisms outside TIPC. This is no different for nodes/namespaces on the same host; they have to auto discover each other using the attached interfaces, and establish links which are supervised via the regular link monitoring mechanism. Hence, a kernel local node has no other way to join a cluster than any other node, and have to obey to policies set in the IP or device layers of the stack. - Only when a sender has established with 100% certainty that the peer node is located in a kernel local namespace does it choose to let user data messages, and only those, take the crossover path to the receiving node/namespace. - If the receiving node/namespace is removed, its namespace pointer is invalidated at all peer nodes, and their neighbor link monitoring will eventually note that this node is gone. - To ensure the "100% certainty" criteria, and prevent any possible spoofing, received discovery messages must contain a proof that the sender knows a common secret. We use the hash mix of the sending node/namespace for this purpose, since it can be accessed directly by all other namespaces in the kernel. Upon reception of a discovery message, the receiver checks this proof against all the local namespaces'hash_mix:es. If it finds a match, that, along with a matching node id and cluster id, this is deemed sufficient proof that the peer node in question is in a local namespace, and a wormhole can be opened. - We should also consider that TIPC is intended to be a cluster local IPC mechanism (just like e.g. UNIX sockets) rather than a network protocol, and hence we think it can justified to allow it to shortcut the lower protocol layers. Regarding traceability, we should notice that since commit 6c9081a3915d ("tipc: add loopback device tracking") it is possible to follow the node internal packet flow by just activating tcpdump on the loopback interface. This will be true even for this mechanism; by activating tcpdump on the involved nodes' loopback interfaces their inter-name space messaging can easily be tracked. v2: - update 'net' pointer when node left/rejoined v3: - grab read/write lock when using node ref obj v4: - clone traffics between netns to loopback Suggested-by: Jon Maloy <jon.maloy@ericsson.com> Acked-by: Jon Maloy <jon.maloy@ericsson.com> Signed-off-by: Hoang Le <hoang.h.le@dektech.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-10-29 07:51:21 +07:00
u32 msg_dsz = ((tipc_node_get_mtu(net, dnode, 0, false) - INT_H_SIZE) /
ITEM_SIZE) * ITEM_SIZE;
u32 msg_rem = msg_dsz;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
struct tipc_msg *hdr;
tipc: eliminate message disordering during binding table update We have seen the following race scenario: 1) named_distribute() builds a "bulk" message, containing a PUBLISH item for a certain publication. This is based on the contents of the binding tables's 'cluster_scope' list. 2) tipc_named_withdraw() removes the same publication from the list, bulds a WITHDRAW message and distributes it to all cluster nodes. 3) tipc_named_node_up(), which was calling named_distribute(), sends out the bulk message built under 1) 4) The WITHDRAW message arrives at the just detected node, finds no corresponding publication, and is dropped. 5) The PUBLISH item arrives at the same node, is added to its binding table, and remains there forever. This arrival disordering was earlier taken care of by the backlog queue, originally added for a different purpose, which was removed in the commit referred to below, but we now need a different solution. In this commit, we replace the rcu lock protecting the 'cluster_scope' list with a regular RW lock which comprises even the sending of the bulk message. This both guarantees both the list integrity and the message sending order. We will later add a commit which cleans up this code further. Note that this commit needs recently added commit d3092b2efca1 ("tipc: fix unsafe rcu locking when accessing publication list") to apply cleanly. Fixes: 37922ea4a310 ("tipc: permit overlapping service ranges in name table") Reported-by: Tuong Lien Tong <tuong.t.lien@dektech.com.au> Acked-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>
2018-10-20 00:55:40 +07:00
list_for_each_entry(publ, pls, binding_node) {
/* Prepare next buffer: */
if (!skb) {
skb = named_prepare_buf(net, PUBLICATION, msg_rem,
dnode);
if (!skb) {
pr_warn("Bulk publication failure\n");
return;
}
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
hdr = buf_msg(skb);
msg_set_bc_ack_invalid(hdr, true);
msg_set_bulk(hdr);
msg_set_non_legacy(hdr);
item = (struct distr_item *)msg_data(hdr);
}
/* Pack publication into message: */
publ_to_item(item, publ);
item++;
msg_rem -= ITEM_SIZE;
/* Append full buffer to list: */
if (!msg_rem) {
__skb_queue_tail(list, skb);
skb = NULL;
msg_rem = msg_dsz;
}
}
if (skb) {
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
hdr = buf_msg(skb);
msg_set_size(hdr, INT_H_SIZE + (msg_dsz - msg_rem));
skb_trim(skb, INT_H_SIZE + (msg_dsz - msg_rem));
__skb_queue_tail(list, skb);
}
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
hdr = buf_msg(skb_peek_tail(list));
msg_set_last_bulk(hdr);
msg_set_named_seqno(hdr, seqno);
}
/**
* tipc_named_node_up - tell specified node about all publications by this node
*/
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
void tipc_named_node_up(struct net *net, u32 dnode, u16 capabilities)
{
struct name_table *nt = tipc_name_table(net);
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
struct tipc_net *tn = tipc_net(net);
struct sk_buff_head head;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
u16 seqno;
tipc: clean up skb list lock handling on send path The policy for handling the skb list locks on the send and receive paths is simple. - On the send path we never need to grab the lock on the 'xmitq' list when the destination is an exernal node. - On the receive path we always need to grab the lock on the 'inputq' list, irrespective of source node. However, when transmitting node local messages those will eventually end up on the receive path of a local socket, meaning that the argument 'xmitq' in tipc_node_xmit() will become the 'ínputq' argument in the function tipc_sk_rcv(). This has been handled by always initializing the spinlock of the 'xmitq' list at message creation, just in case it may end up on the receive path later, and despite knowing that the lock in most cases never will be used. This approach is inaccurate and confusing, and has also concealed the fact that the stated 'no lock grabbing' policy for the send path is violated in some cases. We now clean up this by never initializing the lock at message creation, instead doing this at the moment we find that the message actually will enter the receive path. At the same time we fix the four locations where we incorrectly access the spinlock on the send/error path. This patch also reverts commit d12cffe9329f ("tipc: ensure head->lock is initialised") which has now become redundant. CC: Eric Dumazet <edumazet@google.com> Reported-by: Chris Packham <chris.packham@alliedtelesis.co.nz> Acked-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Xin Long <lucien.xin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-08-15 21:42:50 +07:00
__skb_queue_head_init(&head);
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
spin_lock_bh(&tn->nametbl_lock);
if (!(capabilities & TIPC_NAMED_BCAST))
nt->rc_dests++;
seqno = nt->snd_nxt;
spin_unlock_bh(&tn->nametbl_lock);
tipc: eliminate message disordering during binding table update We have seen the following race scenario: 1) named_distribute() builds a "bulk" message, containing a PUBLISH item for a certain publication. This is based on the contents of the binding tables's 'cluster_scope' list. 2) tipc_named_withdraw() removes the same publication from the list, bulds a WITHDRAW message and distributes it to all cluster nodes. 3) tipc_named_node_up(), which was calling named_distribute(), sends out the bulk message built under 1) 4) The WITHDRAW message arrives at the just detected node, finds no corresponding publication, and is dropped. 5) The PUBLISH item arrives at the same node, is added to its binding table, and remains there forever. This arrival disordering was earlier taken care of by the backlog queue, originally added for a different purpose, which was removed in the commit referred to below, but we now need a different solution. In this commit, we replace the rcu lock protecting the 'cluster_scope' list with a regular RW lock which comprises even the sending of the bulk message. This both guarantees both the list integrity and the message sending order. We will later add a commit which cleans up this code further. Note that this commit needs recently added commit d3092b2efca1 ("tipc: fix unsafe rcu locking when accessing publication list") to apply cleanly. Fixes: 37922ea4a310 ("tipc: permit overlapping service ranges in name table") Reported-by: Tuong Lien Tong <tuong.t.lien@dektech.com.au> Acked-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>
2018-10-20 00:55:40 +07:00
read_lock_bh(&nt->cluster_scope_lock);
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
named_distribute(net, &head, dnode, &nt->cluster_scope, seqno);
tipc_node_xmit(net, &head, dnode, 0);
tipc: eliminate message disordering during binding table update We have seen the following race scenario: 1) named_distribute() builds a "bulk" message, containing a PUBLISH item for a certain publication. This is based on the contents of the binding tables's 'cluster_scope' list. 2) tipc_named_withdraw() removes the same publication from the list, bulds a WITHDRAW message and distributes it to all cluster nodes. 3) tipc_named_node_up(), which was calling named_distribute(), sends out the bulk message built under 1) 4) The WITHDRAW message arrives at the just detected node, finds no corresponding publication, and is dropped. 5) The PUBLISH item arrives at the same node, is added to its binding table, and remains there forever. This arrival disordering was earlier taken care of by the backlog queue, originally added for a different purpose, which was removed in the commit referred to below, but we now need a different solution. In this commit, we replace the rcu lock protecting the 'cluster_scope' list with a regular RW lock which comprises even the sending of the bulk message. This both guarantees both the list integrity and the message sending order. We will later add a commit which cleans up this code further. Note that this commit needs recently added commit d3092b2efca1 ("tipc: fix unsafe rcu locking when accessing publication list") to apply cleanly. Fixes: 37922ea4a310 ("tipc: permit overlapping service ranges in name table") Reported-by: Tuong Lien Tong <tuong.t.lien@dektech.com.au> Acked-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>
2018-10-20 00:55:40 +07:00
read_unlock_bh(&nt->cluster_scope_lock);
}
/**
* tipc_publ_purge - remove publication associated with a failed node
*
* Invoked for each publication issued by a newly failed node.
* Removes publication structure from name table & deletes it.
*/
static void tipc_publ_purge(struct net *net, struct publication *publ, u32 addr)
{
struct tipc_net *tn = tipc_net(net);
struct publication *p;
spin_lock_bh(&tn->nametbl_lock);
p = tipc_nametbl_remove_publ(net, publ->type, publ->lower, publ->upper,
publ->node, publ->key);
if (p)
tipc_node_unsubscribe(net, &p->binding_node, addr);
spin_unlock_bh(&tn->nametbl_lock);
if (p != publ) {
pr_err("Unable to remove publication from failed node\n"
" (type=%u, lower=%u, node=0x%x, port=%u, key=%u)\n",
publ->type, publ->lower, publ->node, publ->port,
publ->key);
}
if (p)
kfree_rcu(p, rcu);
}
/**
* tipc_dist_queue_purge - remove deferred updates from a node that went down
*/
static void tipc_dist_queue_purge(struct net *net, u32 addr)
{
struct tipc_net *tn = net_generic(net, tipc_net_id);
struct distr_queue_item *e, *tmp;
spin_lock_bh(&tn->nametbl_lock);
list_for_each_entry_safe(e, tmp, &tn->dist_queue, next) {
if (e->node != addr)
continue;
list_del(&e->next);
kfree(e);
}
spin_unlock_bh(&tn->nametbl_lock);
}
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
void tipc_publ_notify(struct net *net, struct list_head *nsub_list,
u32 addr, u16 capabilities)
{
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
struct name_table *nt = tipc_name_table(net);
struct tipc_net *tn = tipc_net(net);
struct publication *publ, *tmp;
list_for_each_entry_safe(publ, tmp, nsub_list, binding_node)
tipc_publ_purge(net, publ, addr);
tipc_dist_queue_purge(net, addr);
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
spin_lock_bh(&tn->nametbl_lock);
if (!(capabilities & TIPC_NAMED_BCAST))
nt->rc_dests--;
spin_unlock_bh(&tn->nametbl_lock);
}
/**
* tipc_update_nametbl - try to process a nametable update and notify
* subscribers
*
* tipc_nametbl_lock must be held.
* Returns the publication item if successful, otherwise NULL.
*/
static bool tipc_update_nametbl(struct net *net, struct distr_item *i,
u32 node, u32 dtype)
{
struct publication *p = NULL;
u32 lower = ntohl(i->lower);
u32 upper = ntohl(i->upper);
u32 type = ntohl(i->type);
u32 port = ntohl(i->port);
u32 key = ntohl(i->key);
if (dtype == PUBLICATION) {
p = tipc_nametbl_insert_publ(net, type, lower, upper,
TIPC_CLUSTER_SCOPE, node,
port, key);
if (p) {
tipc_node_subscribe(net, &p->binding_node, node);
return true;
}
} else if (dtype == WITHDRAWAL) {
p = tipc_nametbl_remove_publ(net, type, lower,
upper, node, key);
if (p) {
tipc_node_unsubscribe(net, &p->binding_node, node);
kfree_rcu(p, rcu);
return true;
}
pr_warn_ratelimited("Failed to remove binding %u,%u from %x\n",
type, lower, node);
} else {
pr_warn("Unrecognized name table message received\n");
}
return false;
}
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
static struct sk_buff *tipc_named_dequeue(struct sk_buff_head *namedq,
u16 *rcv_nxt, bool *open)
{
struct sk_buff *skb, *tmp;
struct tipc_msg *hdr;
u16 seqno;
skb_queue_walk_safe(namedq, skb, tmp) {
skb_linearize(skb);
hdr = buf_msg(skb);
seqno = msg_named_seqno(hdr);
if (msg_is_last_bulk(hdr)) {
*rcv_nxt = seqno;
*open = true;
}
if (msg_is_bulk(hdr) || msg_is_legacy(hdr)) {
__skb_unlink(skb, namedq);
return skb;
}
if (*open && (*rcv_nxt == seqno)) {
(*rcv_nxt)++;
__skb_unlink(skb, namedq);
return skb;
}
if (less(seqno, *rcv_nxt)) {
__skb_unlink(skb, namedq);
kfree_skb(skb);
continue;
}
}
return NULL;
}
/**
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 20:36:41 +07:00
* tipc_named_rcv - process name table update messages sent by another node
*/
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
void tipc_named_rcv(struct net *net, struct sk_buff_head *namedq,
u16 *rcv_nxt, bool *open)
{
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
struct tipc_net *tn = tipc_net(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 20:36:41 +07:00
struct distr_item *item;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
struct tipc_msg *hdr;
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 20:36:41 +07:00
struct sk_buff *skb;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
u32 count, node;
spin_lock_bh(&tn->nametbl_lock);
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
while ((skb = tipc_named_dequeue(namedq, rcv_nxt, open))) {
hdr = buf_msg(skb);
node = msg_orignode(hdr);
item = (struct distr_item *)msg_data(hdr);
count = msg_data_sz(hdr) / ITEM_SIZE;
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 20:36:41 +07:00
while (count--) {
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
tipc_update_nametbl(net, item, node, msg_type(hdr));
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 20:36:41 +07:00
item++;
}
kfree_skb(skb);
}
spin_unlock_bh(&tn->nametbl_lock);
}
/**
* tipc_named_reinit - re-initialize local publications
*
* This routine is called whenever TIPC networking is enabled.
* All name table entries published by this node are updated to reflect
* the node's new network address.
*/
void tipc_named_reinit(struct net *net)
{
struct name_table *nt = tipc_name_table(net);
struct tipc_net *tn = tipc_net(net);
struct publication *publ;
u32 self = tipc_own_addr(net);
spin_lock_bh(&tn->nametbl_lock);
list_for_each_entry_rcu(publ, &nt->node_scope, binding_node)
publ->node = self;
list_for_each_entry_rcu(publ, &nt->cluster_scope, binding_node)
publ->node = self;
tipc: update a binding service via broadcast Currently, updating binding table (add service binding to name table/withdraw a service binding) is being sent over replicast. However, if we are scaling up clusters to > 100 nodes/containers this method is less affection because of looping through nodes in a cluster one by one. It is worth to use broadcast to update a binding service. This way, the binding table can be updated on all peer nodes in one shot. Broadcast is used when all peer nodes, as indicated by a new capability flag TIPC_NAMED_BCAST, support reception of this message type. Four problems need to be considered when introducing this feature. 1) When establishing a link to a new peer node we still update this by a unicast 'bulk' update. This may lead to race conditions, where a later broadcast publication/withdrawal bypass the 'bulk', resulting in disordered publications, or even that a withdrawal may arrive before the corresponding publication. We solve this by adding an 'is_last_bulk' bit in the last bulk messages so that it can be distinguished from all other messages. Only when this message has arrived do we open up for reception of broadcast publications/withdrawals. 2) When a first legacy node is added to the cluster all distribution will switch over to use the legacy 'replicast' method, while the opposite happens when the last legacy node leaves the cluster. This entails another risk of message disordering that has to be handled. We solve this by adding a sequence number to the broadcast/replicast messages, so that disordering can be discovered and corrected. Note however that we don't need to consider potential message loss or duplication at this protocol level. 3) Bulk messages don't contain any sequence numbers, and will always arrive in order. Hence we must exempt those from the sequence number control and deliver them unconditionally. We solve this by adding a new 'is_bulk' bit in those messages so that they can be recognized. 4) Legacy messages, which don't contain any new bits or sequence numbers, but neither can arrive out of order, also need to be exempt from the initial synchronization and sequence number check, and delivered unconditionally. Therefore, we add another 'is_not_legacy' bit to all new messages so that those can be distinguished from legacy messages and the latter delivered directly. v1->v2: - fix warning issue reported by kbuild test robot <lkp@intel.com> - add santiy check to drop the publication message with a sequence number that is lower than the agreed synch point Signed-off-by: kernel test robot <lkp@intel.com> Signed-off-by: Hoang Huu Le <hoang.h.le@dektech.com.au> Acked-by: Jon Maloy <jmaloy@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-17 13:56:05 +07:00
nt->rc_dests = 0;
spin_unlock_bh(&tn->nametbl_lock);
}