In commit e4bf4f7696 ("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: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 10724cc7bb ("tipc: redesign connection-level flow control")
we replaced the previous message based flow control with one based on
1k blocks. In order to ensure backwards compatibility the mechanism
falls back to using message as base unit when it senses that the peer
doesn't support the new algorithm. The default flow control window,
i.e., how many units can be sent before the sender blocks and waits
for an acknowledge (aka advertisement) is 512. This was tested against
the previous version, which uses an acknowledge frequency of on ack per
256 received message, and found to work fine.
However, we missed the fact that versions older than Linux 3.15 use an
acknowledge frequency of 512, which is exactly the limit where a 4.6+
sender will stop and wait for acknowledge. This would also work fine if
it weren't for the fact that if the first sent message on a 4.6+ server
side is an empty SYNACK, this one is also is counted as a sent message,
while it is not counted as a received message on a legacy 3.15-receiver.
This leads to the sender always being one step ahead of the receiver, a
scenario causing the sender to block after 512 sent messages, while the
receiver only has registered 511 read messages. Hence, the legacy
receiver is not trigged to send an acknowledge, with a permanently
blocked sender as result.
We solve this deadlock by simply allowing the sender to send one more
message before it blocks, i.e., by a making minimal change to the
condition used for determining connection congestion.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 35c55c9877 ("tipc: add neighbor monitoring framework") we
added a data area to the link monitor STATE messages under the
assumption that previous versions did not use any such data area.
For versions older than Linux 4.3 this assumption is not correct. In
those version, all STATE messages sent out from a node inadvertently
contain a 16 byte data area containing a string; -a leftover from
previous RESET messages which were using this during the setup phase.
This string serves no purpose in STATE messages, and should no be there.
Unfortunately, this data area is delivered to the link monitor
framework, where a sanity check catches that it is not a correct domain
record, and drops it. It also issues a rate limited warning about the
event.
Since such events occur much more frequently than anticipated, we now
choose to remove the warning in order to not fill the kernel log with
useless contents. We also make the sanity check stricter, to further
reduce the risk that such data is inavertently admitted.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 817298102b ("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: Amar Nv <amar.nv005@gmail.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The comment block in socket.c describing the locking policy is
obsolete, and does not reflect current reality. We remove it in this
commit.
Since the current locking policy is much simpler and follows a
mainstream approach, we see no need to add a new description.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 2d18ac4ba7 ("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: John Thompson <thompa.atl@gmail.com>
Tested-by: John Thompson <thompa.atl@gmail.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We should clear out the padding and unused struct members so that we
don't expose stack information to userspace.
Fixes: fdb3accc2c ('tipc: add the ability to get UDP options via netlink')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
'ub' is malloced in tipc_udp_enable() and should be freed before
leaving from the error handling cases, otherwise it will cause
memory leak.
Fixes: ba5aa84a2d ("tipc: split UDP nl address parsing")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/mediatek/mtk_eth_soc.c
drivers/net/ethernet/qlogic/qed/qed_dcbx.c
drivers/net/phy/Kconfig
All conflicts were cases of overlapping commits.
Signed-off-by: David 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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David 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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David 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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In a dual bearer configuration, if the second tipc link becomes
active while the first link still has pending nametable "bulk"
updates, it randomly leads to reset of the second link.
When a link is established, the function named_distribute(),
fills the skb based on node mtu (allows room for TUNNEL_PROTOCOL)
with NAME_DISTRIBUTOR message for each PUBLICATION.
However, the function named_distribute() allocates the buffer by
increasing the node mtu by INT_H_SIZE (to insert NAME_DISTRIBUTOR).
This consumes the space allocated for TUNNEL_PROTOCOL.
When establishing the second link, the link shall tunnel all the
messages in the first link queue including the "bulk" update.
As size of the NAME_DISTRIBUTOR messages while tunnelling, exceeds
the link mtu the transmission fails (-EMSGSIZE).
Thus, the synch point based on the message count of the tunnel
packets is never reached leading to link timeout.
In this commit, we adjust the size of name distributor message so that
they can be tunnelled.
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using replicast a UDP bearer can have an arbitrary amount of
remote ip addresses associated with it. This means we cannot simply
add all remote ip addresses to an existing bearer data message as it
might fill the message, leaving us with a truncated message that we
can't safely resume. To handle this we introduce the new netlink
command TIPC_NL_UDP_GET_REMOTEIP. This command is intended to be
called when the bearer data message has the
TIPC_NLA_UDP_MULTI_REMOTEIP flag set, indicating there are more than
one remote ip (replicast).
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add UDP bearer options to netlink bearer get message. This is used by
the tipc user space tool to display UDP options.
The UDP bearer information is passed using either a sockaddr_in or
sockaddr_in6 structs. This means the user space receiver should
intermediately store the retrieved data in a large enough struct
(sockaddr_strage) before casting to the proper IP version type.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Automatically learn UDP remote IP addresses of communicating peers by
looking at the source IP address of incoming TIPC link configuration
messages (neighbor discovery).
This makes configuration slightly easier and removes the problematic
scenario where a node receives directly addressed neighbor discovery
messages sent using replicast which the node cannot "reply" to using
mutlicast, leaving the link FSM in a limbo state.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces UDP replicast. A concept where we emulate
multicast by sending multiple unicast messages to configured peers.
The purpose of replicast is mainly to be able to use TIPC in cloud
environments where IP multicast is disabled. Using replicas to unicast
multicast messages is costly as we have to copy each skb and send the
copies individually.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a function to check if a tipc UDP media address is a multicast
address or not. This is a purely cosmetic change.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Split the UDP send function into two. One callback that prepares the
skb and one transmit function that sends the skb. This will come in
handy in later patches, when we introduce UDP replicast.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Split the UDP netlink parse function so that it only parses one
netlink attribute at the time. This makes the parse function more
generic and allow future UDP API functions to use it for parsing.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix to return a negative error code in enable_mcast() error handling
case, and release udp socket when necessary.
Fixes: d0f91938be ("tipc: add ip/udp media type")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use kfree_skb() instead of kfree() to free sk_buff.
Fixes: 0d051bf93c ("tipc: make bearer packet filtering generic")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add TIPC_NL_PEER_REMOVE netlink command. This command can remove
an offline peer node from the internal data structures.
This will be supported by the tipc user space tool in iproute2.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David 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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 5b7066c3dd ("tipc: stricter filtering of packets in bearer
layer") we introduced a method of filtering out messages while a bearer
is being reset, to avoid that links may be re-created and come back in
working state while we are still in the process of shutting them down.
This solution works well, but is limited to only work with L2 media, which
is insufficient with the increasing use of UDP as carrier media.
We now replace this solution with a more generic one, by introducing a
new flag "up" in the generic struct tipc_bearer. This field will be set
and reset at the same locations as with the previous solution, while
the packet filtering is moved to the generic code for the sending side.
On the receiving side, the filtering is still done in media specific
code, but now including the UDP bearer.
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>
In commit cf6f7e1d51 ("tipc: dump monitor attributes"),
I dereferenced a pointer before checking if its valid.
This is reported by static check Smatch as:
net/tipc/monitor.c:733 tipc_nl_add_monitor_peer()
warn: variable dereferenced before check 'mon' (see line 731)
In this commit, we check for a valid monitor before proceeding
with any other operation.
Fixes: cf6f7e1d51 ("tipc: dump monitor attributes")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the error handling case of nla_nest_start() failed read_unlock_bh()
is called to unlock a lock that had not been taken yet. sparse warns
about the context imbalance as the following:
net/tipc/monitor.c:799:23: warning:
context imbalance in '__tipc_nl_add_monitor' - different lock contexts for basic block
Fixes: cf6f7e1d51 ('tipc: dump monitor attributes')
Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In this commit, we dump the monitor attributes when queried.
The link monitor attributes are separated into two kinds:
1. general attributes per bearer
2. specific attributes per node/peer
This style resembles the socket attributes and the nametable
publications per socket.
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce a new function to get the bearer name from
its id. This is used in subsequent commit.
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In this commit, we add support to fetch the configured
cluster monitoring threshold.
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In this commit, we introduce support to configure the minimum
threshold to activate the new link monitoring algorithm.
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In this commit, we introduce defines for tipc address size,
offset and mask specification for Zone.Cluster.Node.
There is no functional change in this commit.
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Parthasarathy Bhuvaragan <parthasarathy.bhuvaragan@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In test situations with many nodes and a heavily stressed system we have
observed that the transmission broadcast link may fail due to an
excessive number of retransmissions of the same packet. In such
situations we need to reset all unicast links to all peers, in order to
reset and re-synchronize the broadcast link.
In this commit, we add a new function tipc_bearer_reset_all() to be used
in such situations. The function scans across all bearers and resets all
their pertaining links.
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>
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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David 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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/mellanox/mlx5/core/en.h
drivers/net/ethernet/mellanox/mlx5/core/en_main.c
drivers/net/usb/r8152.c
All three conflicts were overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix incorrect use of nla_strlcpy() where the first NLA_HDRLEN bytes
of the link name where left out.
Making the output of tipc-config -ls look something like:
Link statistics:
dcast-link
1:data0-1.1.2:data0
1:data0-1.1.3:data0
Also, for the record, the patch that introduce this regression
claims "Sending the whole object out can cause a leak". Which isn't
very likely as this is a compat layer, where the data we are parsing
is generated by us and we know the string to be NULL terminated. But
you can of course never be to secure.
Fixes: 5d2be1422e (tipc: fix an infoleak in tipc_nl_compat_link_dump)
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several cases of overlapping changes, except the packet scheduler
conflicts which deal with the addition of the free list parameter
to qdisc_enqueue().
Signed-off-by: David S. Miller <davem@davemloft.net>
Context implies that port in struct "udp_media_addr" is referring
to a UDP port.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The UDP msg2addr function tipc_udp_msg2addr() can return -EINVAL which
prior to this patch was unhanded in the caller.
Signed-off-by: Richard Alpe <richard.alpe@ericsson.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace calls to kmalloc followed by a memcpy with a direct call to
kmemdup.
The Coccinelle semantic patch used to make this change is as follows:
@@
expression from,to,size,flag;
statement S;
@@
- to = \(kmalloc\|kzalloc\)(size,flag);
+ to = kmemdup(from,size,flag);
if (to==NULL || ...) S
- memcpy(to, from, size);
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When extracting an individual message from a received "bundle" buffer,
we just create a clone of the base buffer, and adjust it to point into
the right position of the linearized data area of the latter. This works
well for regular message reception, but during periods of extremely high
load it may happen that an extracted buffer, e.g, a connection probe, is
reversed and forwarded through an external interface while the preceding
extracted message is still unhandled. When this happens, the header or
data area of the preceding message will be partially overwritten by a
MAC header, leading to unpredicatable consequences, such as a link
reset.
We now fix this by ensuring that the msg_reverse() function never
returns a cloned buffer, and that the returned buffer always contains
sufficient valid head and tail room to be forwarded.
Reported-by: Erik Hugne <erik.hugne@gmail.com>
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>
We sometimes observe a 'deadly embrace' type deadlock occurring
between mutually connected sockets on the same node. This happens
when the one-hour peer supervision timers happen to expire
simultaneously in both sockets.
The scenario is as follows:
CPU 1: CPU 2:
-------- --------
tipc_sk_timeout(sk1) tipc_sk_timeout(sk2)
lock(sk1.slock) lock(sk2.slock)
msg_create(probe) msg_create(probe)
unlock(sk1.slock) unlock(sk2.slock)
tipc_node_xmit_skb() tipc_node_xmit_skb()
tipc_node_xmit() tipc_node_xmit()
tipc_sk_rcv(sk2) tipc_sk_rcv(sk1)
lock(sk2.slock) lock((sk1.slock)
filter_rcv() filter_rcv()
tipc_sk_proto_rcv() tipc_sk_proto_rcv()
msg_create(probe_rsp) msg_create(probe_rsp)
tipc_sk_respond() tipc_sk_respond()
tipc_node_xmit_skb() tipc_node_xmit_skb()
tipc_node_xmit() tipc_node_xmit()
tipc_sk_rcv(sk1) tipc_sk_rcv(sk2)
lock((sk1.slock) lock((sk2.slock)
===> DEADLOCK ===> DEADLOCK
Further analysis reveals that there are three different locations in the
socket code where tipc_sk_respond() is called within the context of the
socket lock, with ensuing risk of similar deadlocks.
We now solve this by passing a buffer queue along with all upcalls where
sk_lock.slock may potentially be held. Response or rejected message
buffers are accumulated into this queue instead of being sent out
directly, and only sent once we know we are safely outside the slock
context.
Reported-by: GUNA <gbalasun@gmail.com>
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>
"up_map" is a u64 type but we're not using the high 32 bits.
Fixes: 35c55c9877 ('tipc: add neighbor monitoring framework')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David 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: 42b18f605f ("tipc: refactor function tipc_link_timeout()")
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David 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: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
