In the patch referred to below we added link tolerance as an additional
criteria for declaring broadcast transmission "stale" and resetting the
affected links.

However, the 'tolerance' field of the broadcast link is never set, and
remains at zero. This renders the whole commit without the intended
improving effect, but luckily also with no negative effect.

In this commit we add the missing initialization.

Fixes: a4dc70d4 ("tipc: extend link reset criteria for stale packet retransmission")
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The initial session number when a link is created is based on a random
value, taken from struct tipc_net->random. It is then incremented for
each link reset to avoid mixing protocol messages from different link
sessions.

However, when a bearer is reset all its links are deleted, and will
later be re-created using the same random value as the first time.
This means that if the link never went down between creation and
deletion we will still sometimes have two subsequent sessions with
the same session number. In virtual environments with potentially
long transmission times this has turned out to be a real problem.

We now fix this by randomizing the session number each time a link
is created.

With a session number size of 16 bits this gives a risk of session
collision of 1/64k. To reduce this further, we also introduce a sanity
check on the very first STATE message arriving at a link. If this has
an acknowledge value differing from 0, which is logically impossible,
we ignore the message. The final risk for session collision is hence
reduced to 1/4G, which should be sufficient.
Signed-off-by: NLUU Duc Canh <canh.d.luu@dektech.com.au>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We see the following scenario:
1) Link endpoint B on node 1 discovers that its peer endpoint is gone.
   Since there is a second working link, failover procedure is started.
2) Link endpoint A on node 1 sends a FAILOVER message to peer endpoint
   A on node 2. The node item 1->2 goes to state FAILINGOVER.
3) Linke endpoint A/2 receives the failover, and is supposed to take
   down its parallell link endpoint B/2, while producing a FAILOVER
   message to send back to A/1.
4) However, B/2 has already been deleted, so no FAILOVER message can
   created.
5) Node 1->2 remains in state FAILINGOVER forever, refusing to receive
   any messages that can bring B/1 up again. We are left with a non-
   redundant link between node 1 and 2.

We fix this with letting endpoint A/2 build a dummy FAILOVER message
to send to back to A/1, so that the situation can be resolved.
Signed-off-by: NLUU Duc Canh <canh.d.luu@dektech.com.au>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In tipc_link_reset() we copy the wakeup queue to input queue using
skb_queue_splice_init(link->wakeupq, link->inputq).
This is performed without holding any locks. The lists might be
simultaneously be accessed by other cpu threads in tipc_sk_rcv(),
something leading to to random missing packets.
Signed-off-by: NParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fixes the following sparse warnings:

net/tipc/link.c:376:5: warning: symbol 'link_bc_rcv_gap' was not declared. Should it be static?
net/tipc/link.c:823:6: warning: symbol 'link_prepare_wakeup' was not declared. Should it be static?
net/tipc/link.c:959:6: warning: symbol 'tipc_link_advance_backlog' was not declared. Should it be static?
net/tipc/link.c:1009:5: warning: symbol 'tipc_link_retrans' was not declared. Should it be static?
net/tipc/monitor.c:687:5: warning: symbol '__tipc_nl_add_monitor_peer' was not declared. Should it be static?
net/tipc/group.c:230:20: warning: symbol 'tipc_group_find_member' was not declared. Should it be static?
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

tipc_link_is_active is no longer used and can be removed.
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Acked-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In some virtual environments we observe a significant higher number of
packet reordering and delays than we have been used to traditionally.

This makes it necessary with stricter checks on incoming link protocol
messages' session number, which until now only has been validated for
RESET messages.

Since the other two message types, ACTIVATE and STATE messages also
carry this number, it is easy to extend the validation check to those
messages.

We also introduce a flag indicating if a link has a valid peer session
number or not. This eliminates the mixing of 32- and 16-bit arithmethics
we are currently using to achieve this.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Some switch infrastructures produce huge amounts of packet duplicates.
This becomes a problem if those messages are STATE/NACK protocol
messages, causing unnecessary retransmissions of already accepted
packets.

We now introduce a unique sequence number per STATE protocol message
so that duplicates can be identified and ignored. This will also be
useful when tracing such cases, and to avert replay attacks when TIPC
is encrypted.

For compatibility reasons we have to introduce a new capability flag
TIPC_LINK_PROTO_SEQNO to handle this new feature.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently a link is declared stale and reset if there has been 100
repeated attempts to retransmit the same packet. However, in certain
infrastructures we see that packet (NACK) duplicates and delays may
cause such retransmit attempts to occur at a high rate, so that the
peer doesn't have a reasonable chance to acknowledge the reception
before the 100-limit is hit. This may take much less than the
stipulated link tolerance time, and despite that probe/probe replies
otherwise go through as normal.

We now extend the criteria for link reset to also being time based.
I.e., we don't reset the link until the link tolerance time is passed
AND we have made 100 retransmissions attempts.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In preparation to enabling -Wimplicit-fallthrough, mark switch cases
where we are expecting to fall through.

Warning level 2 was used: -Wimplicit-fallthrough=2
Signed-off-by: NGustavo A. R. Silva <gustavo@embeddedor.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

gcc points out that the combined length of the fixed-length inputs to
l->name is larger than the destination buffer size:

net/tipc/link.c: In function 'tipc_link_create':
net/tipc/link.c:465:26: error: '%s' directive writing up to 32 bytes
into a region of size between 26 and 58 [-Werror=format-overflow=]
sprintf(l->name, "%s:%s-%s:unknown", self_str, if_name, peer_str);

net/tipc/link.c:465:2: note: 'sprintf' output 11 or more bytes
(assuming 75) into a destination of size 60
sprintf(l->name, "%s:%s-%s:unknown", self_str, if_name, peer_str);

A detailed analysis reveals that the theoretical maximum length of
a link name is:
max self_str + 1 + max if_name + 1 + max peer_str + 1 + max if_name =
16 + 1 + 15 + 1 + 16 + 1 + 15 = 65
Since we also need space for a trailing zero we now set MAX_LINK_NAME
to 68.

Just to be on the safe side we also replace the sprintf() call with
snprintf().

Fixes: 25b0b9c4 ("tipc: handle collisions of 32-bit node address
hash values")
Reported-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current design of the binding table has an unnecessary memory
consuming and complex data structure. It aggregates the service range
items into an array, which is expanded by a factor two every time it
becomes too small to hold a new item. Furthermore, the arrays never
shrink when the number of ranges diminishes.

We now replace this array with an RB tree that is holding the range
items as tree nodes, each range directly holding a list of bindings.

This, along with a few name changes, improves both readability and
volume of the code, as well as reducing memory consumption and hopefully
improving cache hit rate.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a 32-bit node address is generated from a 128-bit identifier,
there is a risk of collisions which must be discovered and handled.

We do this as follows:
- We don't apply the generated address immediately to the node, but do
  instead initiate a 1 sec trial period to allow other cluster members
  to discover and handle such collisions.

- During the trial period the node periodically sends out a new type
  of message, DSC_TRIAL_MSG, using broadcast or emulated broadcast,
  to all the other nodes in the cluster.

- When a node is receiving such a message, it must check that the
  presented 32-bit identifier either is unused, or was used by the very
  same peer in a previous session. In both cases it accepts the request
  by not responding to it.

- If it finds that the same node has been up before using a different
  address, it responds with a DSC_TRIAL_FAIL_MSG containing that
  address.

- If it finds that the address has already been taken by some other
  node, it generates a new, unused address and returns it to the
  requester.

- During the trial period the requesting node must always be prepared
  to accept a failure message, i.e., a message where a peer suggests a
  different (or equal)  address to the one tried. In those cases it
  must apply the suggested value as trial address and restart the trial
  period.

This algorithm ensures that in the vast majority of cases a node will
have the same address before and after a reboot. If a legacy user
configures the address explicitly, there will be no trial period and
messages, so this protocol addition is completely backwards compatible.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We add a 128-bit node identity, as an alternative to the currently used
32-bit node address.

For the sake of compatibility and to minimize message header changes
we retain the existing 32-bit address field. When not set explicitly by
the user, this field will be filled with a hash value generated from the
much longer node identity, and be used as a shorthand value for the
latter.

We permit either the address or the identity to be set by configuration,
but not both, so when the address value is set by a legacy user the
corresponding 128-bit node identity is generated based on the that value.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As a preparation to changing the addressing structure of TIPC we replace
all direct accesses to the tipc_net::own_addr field with the function
dedicated for this, tipc_own_addr().

There are no changes to program logics in this commit.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Nominally, TIPC organizes network nodes into a three-level network
hierarchy consisting of the levels 'zone', 'cluster' and 'node'. This
hierarchy is reflected in the node address format, - it is sub-divided
into an 8-bit zone id, and 12 bit cluster id, and a 12-bit node id.

However, the 'zone' and 'cluster' levels have in reality never been
fully implemented,and never will be. The result of this has been
that the first 20 bits the node identity structure have been wasted,
and the usable node identity range within a cluster has been limited
to 12 bits. This is starting to become a problem.

In the following commits, we will need to be able to connect between
nodes which are using the whole 32-bit value space of the node address.
We therefore remove the restrictions on which values can be assigned
to node identity, -it is from now on only a 32-bit integer with no
assumed internal structure.

Isolation between clusters is now achieved only by setting different
values for the 'network id' field used during neighbor discovery, in
practice leading to the latter becoming the new cluster identity.

The rules for accepting discovery requests/responses from neighboring
nodes now become:

- If the user is using legacy address format on both peers, reception
  of discovery messages is subject to the legacy lookup domain check
  in addition to the cluster id check.

- Otherwise, the discovery request/response is always accepted, provided
  both peers have the same network id.

This secures backwards compatibility for users who have been using zone
or cluster identities as cluster separators, instead of the intended
'network id'.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, the default link tolerance set in struct tipc_bearer only
has effect on links going up after that moment. I.e., a user has to
reset all the node's links across that bearer to have the new value
applied. This is too limiting and disturbing on a running cluster to
be useful.

We now change this so that also already existing links are updated
dynamically, without any need for a reset, when the bearer value is
changed. We leverage the already existing per-link functionality
for this to achieve the wanted effect.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When sending node local messages the code is using an 'mtu' of 66060
bytes to avoid unnecessary fragmentation. During situations of low
memory tipc_msg_build() may sometimes fail to allocate such large
buffers, resulting in unnecessary send failures. This can easily be
remedied by falling back to a smaller MTU, and then reassemble the
buffer chain as if the message were arriving from a remote node.

At the same time, we change the initial MTU setting of the broadcast
link to a lower value, so that large messages always are fragmented
into smaller buffers even when we run in single node mode. Apart from
obtaining the same advantage as for the 'fallback' solution above, this
turns out to give a significant performance improvement. This can
probably be explained with the __pskb_copy() operation performed on the
buffer for each recipient during reception. We found the optimal value
for this, considering the most relevant skb pool, to be 3744 bytes.
Acked-by: NYing Xue <ying.xue@ericsson.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently, the TIPC RPS dissector is based only on the incoming packets'
source node address, hence steering all traffic from a node to the same
core. We have seen that this makes the links vulnerable to starvation
and unnecessary resets when we turn down the link tolerance to very low
values.

To reduce the risk of this happening, we exempt probe and probe replies
packets from the convergence to one core per source node. Instead, we do
the opposite, - we try to diverge those packets across as many cores as
possible, by randomizing the flow selector key.

To make such packets identifiable to the dissector, we add a new
'is_keepalive' bit to word 0 of the LINK_PROTOCOL header. This bit is
set both for PROBE and PROBE_REPLY messages, and only for those.

It should be noted that these packets are not part of any flow anyway,
and only constitute a minuscule fraction of all packets sent across a
link. Hence, there is no risk that this will affect overall performance.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In commit 2f487712 ("tipc: guarantee that group broadcast doesn't
bypass group unicast") there was introduced a last-minute rebasing
error that broke non-group communication.

We fix this here.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The following scenario is possible:
- A user joins a group, and immediately sends out a broadcast message
  to its members.
- The broadcast message, following a different data path than the
  initial JOIN message sent out during the joining procedure, arrives
  to a receiver before the latter..
- The receiver drops the message, since it is not ready to accept any
  messages until the JOIN has arrived.

We avoid this by treating group protocol JOIN messages like unicast
messages.
- We let them pass through the recipient's multicast input queue, just
  like ordinary unicasts.
- We force the first following broadacst to be sent as replicated
  unicast and being acknowledged by the recipient before accepting
  any more broadcast transmissions.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We need a mechanism guaranteeing that group unicasts sent out from a
socket are not bypassed by later sent broadcasts from the same socket.
We do this as follows:

- Each time a unicast is sent, we set a the broadcast method for the
  socket to "replicast" and "mandatory". This forces the first
  subsequent broadcast message to follow the same network and data path
  as the preceding unicast to a destination, hence preventing it from
  overtaking the latter.

- In order to make the 'same data path' statement above true, we let
  group unicasts pass through the multicast link input queue, instead
  of as previously through the unicast link input queue.

- In the first broadcast following a unicast, we set a new header flag,
  requiring all recipients to immediately acknowledge its reception.

- During the period before all the expected acknowledges are received,
  the socket refuses to accept any more broadcast attempts, i.e., by
  blocking or returning EAGAIN. This period should typically not be
  longer than a few microseconds.

- When all acknowledges have been received, the sending socket will
  open up for subsequent broadcasts, this time giving the link layer
  freedom to itself select the best transmission method.

- The forced and/or abrupt transmission method changes described above
  may lead to broadcasts arriving out of order to the recipients. We
  remedy this by introducing code that checks and if necessary
  re-orders such messages at the receiving end.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As a preparation for introducing flow control for multicast and datagram
messaging we need a more strictly defined framework than we have now. A
socket must be able keep track of exactly how many and which other
sockets it is allowed to communicate with at any moment, and keep the
necessary state for those.

We therefore introduce a new concept we have named Communication Group.
Sockets can join a group via a new setsockopt() call TIPC_GROUP_JOIN.
The call takes four parameters: 'type' serves as group identifier,
'instance' serves as an logical member identifier, and 'scope' indicates
the visibility of the group (node/cluster/zone). Finally, 'flags' makes
it possible to set certain properties for the member. For now, there is
only one flag, indicating if the creator of the socket wants to receive
a copy of broadcast or multicast messages it is sending via the socket,
and if wants to be eligible as destination for its own anycasts.

A group is closed, i.e., sockets which have not joined a group will
not be able to send messages to or receive messages from members of
the group, and vice versa.

Any member of a group can send multicast ('group broadcast') messages
to all group members, optionally including itself, using the primitive
send(). The messages are received via the recvmsg() primitive. A socket
can only be member of one group at a time.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the broadcast send link after 100 attempts has failed to
transfer a packet to all peers, we consider it stale, and reset
it. Thereafter it needs to re-synchronize with the peers, something
currently done by just resetting and re-establishing all links to
all peers. This has turned out to be overkill, with potentially
unwanted consequences for the remaining cluster.

A closer analysis reveals that this can be done much simpler. When
this kind of failure happens, for reasons that may lie outside the
TIPC protocol, it is typically only one peer which is failing to
receive and acknowledge packets. It is hence sufficient to identify
and reset the links only to that peer to resolve the situation, without
having to reset the broadcast link at all. This solution entails a much
lower risk of negative consequences for the own node as well as for
the overall cluster.

We implement this change in this commit.
Reviewed-by: NParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pass the new extended ACK reporting struct to all of the generic
netlink parsing functions. For now, pass NULL in almost all callers
(except for some in the core.)
Signed-off-by: NJohannes Berg <johannes.berg@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If the bearer carrying multicast messages supports broadcast, those
messages will be sent to all cluster nodes, irrespective of whether
these nodes host any actual destinations socket or not. This is clearly
wasteful if the cluster is large and there are only a few real
destinations for the message being sent.

In this commit we extend the eligibility of the newly introduced
"replicast" transmit option. We now make it possible for a user to
select which method he wants to be used, either as a mandatory setting
via setsockopt(), or as a relative setting where we let the broadcast
layer decide which method to use based on the ratio between cluster
size and the message's actual number of destination nodes.

In the latter case, a sending socket must stick to a previously
selected method until it enters an idle period of at least 5 seconds.
This eliminates the risk of message reordering caused by method change,
i.e., when changes to cluster size or number of destinations would
otherwise mandate a new method to be used.
Reviewed-by: NParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

TIPC multicast messages are currently carried over a reliable
'broadcast link', making use of the underlying media's ability to
transport packets as L2 broadcast or IP multicast to all nodes in
the cluster.

When the used bearer is lacking that ability, we can instead emulate
the broadcast service by replicating and sending the packets over as
many unicast links as needed to reach all identified destinations.
We now introduce a new TIPC link-level 'replicast' service that does
this.
Reviewed-by: NParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Until now, we allocate memory always with GFP_ATOMIC flag.
When the system is under memory pressure and a user tries to send,
the send fails due to low memory. However, the user application
can wait for free memory if we allocate it using GFP_KERNEL flag.

In this commit, we use allocate memory with GFP_KERNEL for all user
allocation.
Reported-by: NRune Torgersen <runet@innovsys.com>
Acked-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The socket code currently handles link congestion by either blocking
and trying to send again when the congestion has abated, or just
returning to the user with -EAGAIN and let him re-try later.

This mechanism is prone to starvation, because the wakeup algorithm is
non-atomic. During the time the link issues a wakeup signal, until the
socket wakes up and re-attempts sending, other senders may have come
in between and occupied the free buffer space in the link. This in turn
may lead to a socket having to make many send attempts before it is
successful. In extremely loaded systems we have observed latency times
of several seconds before a low-priority socket is able to send out a
message.

In this commit, we simplify this mechanism and reduce the risk of the
described scenario happening. When a message is attempted sent via a
congested link, we now let it be added to the link's backlog queue
anyway, thus permitting an oversubscription of one message per source
socket. We still create a wakeup item and return an error code, hence
instructing the sender to block or stop sending. Only when enough space
has been freed up in the link's backlog queue do we issue a wakeup event
that allows the sender to continue with the next message, if any.

The fact that a socket now can consider a message sent even when the
link returns a congestion code means that the sending socket code can
be simplified. Also, since this is a good opportunity to get rid of the
obsolete 'mtu change' condition in the three socket send functions, we
now choose to refactor those functions completely.
Signed-off-by: NParthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In commit e4bf4f76 ("tipc: simplify packet sequence number
handling") we changed the internal representation of the packet
sequence number counters from u32 to u16, reflecting what is really
sent over the wire.

Since then some link statistics counters have been displaying incorrect
values, partially because the counters meant to be used as sequence
number snapshots are now used as direct counters, stored as u32, and
partially because some counter updates are just missing in the code.

In this commit we correct this in two ways. First, we base the
displayed packet sent/received values on direct counters instead
of as previously a calculated difference between current sequence
number and a snapshot. Second, we add the missing updates of the
counters.

This change is compatible with the current netlink API, and requires
no changes to the user space tools.
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

commit 81729810 ("tipc: fix link priority propagation") introduced a
compatibility problem between TIPC versions newer than Linux 4.6 and
those older than Linux 4.4. In versions later than 4.4, link STATE
messages only contain a non-zero link priority value when the sender
wants the receiver to change its priority. This has the effect that the
receiver resets itself in order to apply the new priority. This works
well, and is consistent with the said commit.

However, in versions older than 4.4 a valid link priority is present in
all sent link STATE messages, leading to cyclic link establishment and
reset on the 4.6+ node.

We fix this by adding a test that the received value should not only
be valid, but also differ from the current value in order to cause the
receiving link endpoint to reset.
Reported-by: NAmar Nv <amar.nv005@gmail.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In commit 2d18ac4b ("tipc: extend broadcast link initialization
criteria") we tried to fix a problem with the initial synchronization
of broadcast link acknowledge values. Unfortunately that solution is
not sufficient to solve the issue.

We have seen it happen that LINK_PROTOCOL/STATE packets with a valid
non-zero unicast acknowledge number may bypass BCAST_PROTOCOL
initialization, NAME_DISTRIBUTOR and other STATE packets with invalid
broadcast acknowledge numbers, leading to premature opening of the
broadcast link. When the bypassed packets finally arrive, they are
inadvertently accepted, and the already correctly initialized
acknowledge number in the broadcast receive link is overwritten by
the invalid (zero) value of the said packets. After this the broadcast
link goes stale.

We now fix this by marking the packets where we know the acknowledge
value is or may be invalid, and then ignoring the acks from those.

To this purpose, we claim an unused bit in the header to indicate that
the value is invalid. We set the bit to 1 in the initial BCAST_PROTOCOL
synchronization packet and all initial ("bulk") NAME_DISTRIBUTOR
packets, plus those LINK_PROTOCOL packets sent out before the broadcast
links are fully synchronized.

This minor protocol update is fully backwards compatible.
Reported-by: NJohn Thompson <thompa.atl@gmail.com>
Tested-by: NJohn Thompson <thompa.atl@gmail.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Because of the risk of an excessive number of NACK messages and
retransissions, receivers have until now abstained from sending
broadcast NACKS directly upon detection of a packet sequence number
gap. We have instead relied on such gaps being detected by link
protocol STATE message exchange, something that by necessity delays
such detection and subsequent retransmissions.

With the introduction of unicast NACK transmission and rate control
of retransmissions we can now remove this limitation. We now allow
receiving nodes to send NACKS immediately, while coordinating the
permission to do so among the nodes in order to avoid NACK storms.
Reviewed-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As cluster sizes grow, so does the amount of identical or overlapping
broadcast NACKs generated by the packet receivers. This often leads to
'NACK crunches' resulting in huge numbers of redundant retransmissions
of the same packet ranges.

In this commit, we introduce rate control of broadcast retransmissions,
so that a retransmitted range cannot be retransmitted again until after
at least 10 ms. This reduces the frequency of duplicate, redundant
retransmissions by an order of magnitude, while having a significant
positive impact on overall throughput and scalability.
Reviewed-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When we send broadcasts in clusters of more 70-80 nodes, we sometimes
see the broadcast link resetting because of an excessive number of
retransmissions. This is caused by a combination of two factors:

1) A 'NACK crunch", where loss of broadcast packets is discovered
   and NACK'ed by several nodes simultaneously, leading to multiple
   redundant broadcast retransmissions.

2) The fact that the NACKS as such also are sent as broadcast, leading
   to excessive load and packet loss on the transmitting switch/bridge.

This commit deals with the latter problem, by moving sending of
broadcast nacks from the dedicated BCAST_PROTOCOL/NACK message type
to regular unicast LINK_PROTOCOL/STATE messages. We allocate 10 unused
bits in word 8 of the said message for this purpose, and introduce a
new capability bit, TIPC_BCAST_STATE_NACK in order to keep the change
backwards compatible.
Reviewed-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When a link is attempted woken up after congestion, it uses a different,
more generous criteria than when it was originally declared congested.
This has the effect that the link, and the sending process, sometimes
will be woken up unnecessarily, just to immediately return to congestion
when it turns out there is not not enough space in its send queue to
host the pending message. This is a waste of CPU cycles.

We now change the function link_prepare_wakeup() to use exactly the same
criteria as tipc_link_xmit(). However, since we are now excluding the
window limit from the wakeup calculation, and the current backlog limit
for the lowest level is too small to house even a single maximum-size
message, we have to expand this limit. We do this by evaluating an
alternative, minimum value during the setting of the importance limits.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

After a new receiver peer has been added to the broadcast transmission
link, we allow immediate transmission of new broadcast packets, trusting
that the new peer will not accept the packets until it has received the
previously sent unicast broadcast initialiation message. In the same
way, the sender must not accept any acknowledges until it has itself
received the broadcast initialization from the peer, as well as
confirmation of the reception of its own initialization message.

Furthermore, when a receiver peer goes down, the sender has to produce
the missing acknowledges from the lost peer locally, in order ensure
correct release of the buffers that were expected to be acknowledged by
the said peer.

In a highly stressed system we have observed that contact with a peer
may come up and be lost before the above mentioned broadcast initial-
ization and confirmation have been received. This leads to the locally
produced acknowledges being rejected, and the non-acknowledged buffers
to linger in the broadcast link transmission queue until it fills up
and the link goes into permanent congestion.

In this commit, we remedy this by temporarily setting the corresponding
broadcast receive link state to ESTABLISHED and the 'bc_peer_is_up'
state to true before we issue the local acknowledges. This ensures that
those acknowledges will always be accepted. The mentioned state values
are restored immediately afterwards when the link is reset.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

At first contact between two nodes, an endpoint might sometimes have
time to send out a LINK_PROTOCOL/STATE packet before it has received
the broadcast initialization packet from the peer, i.e., before it has
received a valid broadcast packet number to add to the 'bc_ack' field
of the protocol message.

This means that the peer endpoint will receive a protocol packet with an
invalid broadcast acknowledge value of 0. Under unlucky circumstances
this may lead to the original, already received acknowledge value being
overwritten, so that the whole broadcast link goes stale after a while.

We fix this by delaying the setting of the link field 'bc_peer_is_up'
until we know that the peer really has received our own broadcast
initialization message. The latter is always sent out as the first
unicast message on a link, and always with seqeunce number 1. Because
of this, we only need to look for a non-zero unicast acknowledge value
in the arriving STATE messages, and once that is confirmed we know we
are safe and can set the mentioned field. Before this moment, we must
ignore all broadcast acknowledges from the peer.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

net/tipc/link.c: In function ‘tipc_link_timeout’:
net/tipc/link.c:744:28: warning: ‘mtyp’ may be used uninitialized in this function [-Wuninitialized]

Fixes: 42b18f60 ("tipc: refactor function tipc_link_timeout()")
Acked-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

TIPC based clusters are by default set up with full-mesh link
connectivity between all nodes. Those links are expected to provide
a short failure detection time, by default set to 1500 ms. Because
of this, the background load for neighbor monitoring in an N-node
cluster increases with a factor N on each node, while the overall
monitoring traffic through the network infrastructure increases at
a ~(N * (N - 1)) rate. Experience has shown that such clusters don't
scale well beyond ~100 nodes unless we significantly increase failure
discovery tolerance.

This commit introduces a framework and an algorithm that drastically
reduces this background load, while basically maintaining the original
failure detection times across the whole cluster. Using this algorithm,
background load will now grow at a rate of ~(2 * sqrt(N)) per node, and
at ~(2 * N * sqrt(N)) in traffic overhead. As an example, each node will
now have to actively monitor 38 neighbors in a 400-node cluster, instead
of as before 399.

This "Overlapping Ring Supervision Algorithm" is completely distributed
and employs no centralized or coordinated state. It goes as follows:

- Each node makes up a linearly ascending, circular list of all its N
  known neighbors, based on their TIPC node identity. This algorithm
  must be the same on all nodes.

- The node then selects the next M = sqrt(N) - 1 nodes downstream from
  itself in the list, and chooses to actively monitor those. This is
  called its "local monitoring domain".

- It creates a domain record describing the monitoring domain, and
  piggy-backs this in the data area of all neighbor monitoring messages
  (LINK_PROTOCOL/STATE) leaving that node. This means that all nodes in
  the cluster eventually (default within 400 ms) will learn about
  its monitoring domain.

- Whenever a node discovers a change in its local domain, e.g., a node
  has been added or has gone down, it creates and sends out a new
  version of its node record to inform all neighbors about the change.

- A node receiving a domain record from anybody outside its local domain
  matches this against its own list (which may not look the same), and
  chooses to not actively monitor those members of the received domain
  record that are also present in its own list. Instead, it relies on
  indications from the direct monitoring nodes if an indirectly
  monitored node has gone up or down. If a node is indicated lost, the
  receiving node temporarily activates its own direct monitoring towards
  that node in order to confirm, or not, that it is actually gone.

- Since each node is actively monitoring sqrt(N) downstream neighbors,
  each node is also actively monitored by the same number of upstream
  neighbors. This means that all non-direct monitoring nodes normally
  will receive sqrt(N) indications that a node is gone.

- A major drawback with ring monitoring is how it handles failures that
  cause massive network partitionings. If both a lost node and all its
  direct monitoring neighbors are inside the lost partition, the nodes in
  the remaining partition will never receive indications about the loss.
  To overcome this, each node also chooses to actively monitor some
  nodes outside its local domain. Those nodes are called remote domain
  "heads", and are selected in such a way that no node in the cluster
  will be more than two direct monitoring hops away. Because of this,
  each node, apart from monitoring the member of its local domain, will
  also typically monitor sqrt(N) remote head nodes.

- As an optimization, local list status, domain status and domain
  records are marked with a generation number. This saves senders from
  unnecessarily conveying  unaltered domain records, and receivers from
  performing unneeded re-adaptations of their node monitoring list, such
  as re-assigning domain heads.

- As a measure of caution we have added the possibility to disable the
  new algorithm through configuration. We do this by keeping a threshold
  value for the cluster size; a cluster that grows beyond this value
  will switch from full-mesh to ring monitoring, and vice versa when
  it shrinks below the value. This means that if the threshold is set to
  a value larger than any anticipated cluster size (default size is 32)
  the new algorithm is effectively disabled. A patch set for altering the
  threshold value and for listing the table contents will follow shortly.

- This change is fully backwards compatible.
Acked-by: NYing Xue <ying.xue@windriver.com>
Signed-off-by: NJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>