When we receive a JOIN message from a peer member, the message may
contain an advertised window value ADV_IDLE that permits removing the
member in question from the tipc_group::congested list. However, since
the removal has been made conditional on that the advertised window is
*not* ADV_IDLE, we miss this case. This has the effect that a sender
sometimes may enter a state of permanent, false, broadcast congestion.
We fix this by unconditinally removing the member from the congested
list before calling tipc_member_update(), which might potentially sort
it into the list again.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When, during a join operation, or during message transmission, a group
member needs to be added to the group's 'congested' list, we sort it
into the list in ascending order, according to its current advertised
window size. However, we miss the case when the member is already on
that list. This will have the result that the member, after the window
size has been decremented, might be at the wrong position in that list.
This again may have the effect that we during broadcast and multicast
transmissions miss the fact that a destination is not yet ready for
reception, and we end up sending anyway. From this point on, the
behavior during the remaining session is unpredictable, e.g., with
underflowing window sizes.
We now correct this bug by unconditionally removing the member from
the list before (re-)sorting it in.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A group member going into state LEAVING should never go back to any
other state before it is finally deleted. However, this might happen
if the socket needs to send out a RECLAIM message during this interval.
Since we forget to remove the leaving member from the group's 'active'
or 'pending' list, the member might be selected for reclaiming, change
state to RECLAIMING, and get stuck in this state instead of being
deleted. This might lead to suppression of the expected 'member down'
event to the receiver.
We fix this by removing the member from all lists, except the RB tree,
at the moment it goes into state LEAVING.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Group messages are not supposed to be returned to sender when the
destination socket disappears. This is done correctly for regular
traffic messages, by setting the 'dest_droppable' bit in the header.
But we forget to do that in group protocol messages. This has the effect
that such messages may sometimes bounce back to the sender, be perceived
as a legitimate peer message, and wreak general havoc for the rest of
the session. In particular, we have seen that a member in state LEAVING
may go back to state RECLAIMED or REMITTED, hence causing suppression
of an otherwise expected 'member down' event to the user.
We fix this by setting the 'dest_droppable' bit even in group protocol
messages.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
KASAN revealed another access after delete in group.c. This time
it found that we read the header of a received message after the
buffer has been released.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the function tipc_group_filter_msg() finds that a member event
indicates that the member is leaving the group, it first deletes the
member instance, and then purges the message queue being handled
by the call. But the message queue is an aggregated field in the
just deleted item, leading the purge call to access freed memory.
We fix this by swapping the order of the two actions.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We already have point-to-multipoint flow control within a group. But
we even need the opposite; -a scheme which can handle that potentially
hundreds of sources may try to send messages to the same destination
simultaneously without causing buffer overflow at the recipient. This
commit adds such a mechanism.
The algorithm works as follows:
- When a member detects a new, joining member, it initially set its
state to JOINED and advertises a minimum window to the new member.
This window is chosen so that the new member can send exactly one
maximum sized message, or several smaller ones, to the recipient
before it must stop and wait for an additional advertisement. This
minimum window ADV_IDLE is set to 65 1kB blocks.
- When a member receives the first data message from a JOINED member,
it changes the state of the latter to ACTIVE, and advertises a larger
window ADV_ACTIVE = 12 x ADV_IDLE blocks to the sender, so it can
continue sending with minimal disturbances to the data flow.
- The active members are kept in a dedicated linked list. Each time a
message is received from an active member, it will be moved to the
tail of that list. This way, we keep a record of which members have
been most (tail) and least (head) recently active.
- There is a maximum number (16) of permitted simultaneous active
senders per receiver. When this limit is reached, the receiver will
not advertise anything immediately to a new sender, but instead put
it in a PENDING state, and add it to a corresponding queue. At the
same time, it will pick the least recently active member, send it an
advertisement RECLAIM message, and set this member to state
RECLAIMING.
- The reclaimee member has to respond with a REMIT message, meaning that
it goes back to a send window of ADV_IDLE, and returns its unused
advertised blocks beyond that value to the reclaiming member.
- When the reclaiming member receives the REMIT message, it unlinks
the reclaimee from its active list, resets its state to JOINED, and
notes that it is now back at ADV_IDLE advertised blocks to that
member. If there are still unread data messages sent out by
reclaimee before the REMIT, the member goes into an intermediate
state REMITTED, where it stays until the said messages have been
consumed.
- The returned advertised blocks can now be re-advertised to the
pending member, which is now set to state ACTIVE and added to
the active member list.
- To be proactive, i.e., to minimize the risk that any member will
end up in the pending queue, we start reclaiming resources already
when the number of active members exceeds 3/4 of the permitted
maximum.
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>
The following scenario is possible:
- A user sends a broadcast message, and thereafter immediately leaves
the group.
- The LEAVE message, following a different path than the broadcast,
arrives ahead of the broadcast, and the sending member is removed
from the receiver's list.
- The broadcast message arrives, but is dropped because the sender
now is unknown to the receipient.
We fix this by sequence numbering membership events, just like ordinary
unicast messages. Currently, when a JOIN is sent to a peer, it contains
a synchronization point, - the sequence number of the next sent
broadcast, in order to give the receiver a start synchronization point.
We now let even LEAVE messages contain such an "end synchronization"
point, so that the recipient can delay the removal of the sending member
until it knows that all messages have been received.
The received synchronization points are added as sequence numbers to the
generated membership events, making it possible to handle them almost
the same way as regular unicasts in the receiving filter function. In
particular, a DOWN event with a too high sequence number will be kept
in the reordering queue until the missing broadcast(s) arrive and have
been delivered.
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>
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: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Group unicast messages don't follow the same path as broadcast messages,
and there is a high risk that unicasts sent from a socket might bypass
previously sent broadcasts from the same socket.
We fix this by letting all unicast messages carry the sequence number of
the next sent broadcast from the same node, but without updating this
number at the receiver. This way, a receiver can check and if necessary
re-order such messages before they are added to the socket receive buffer.
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>
The previously introduced message transport to all group members is
based on the tipc multicast service, but is logically a broadcast
service within the group, and that is what we call it.
We now add functionality for sending messages to all group members
having a certain identity. Correspondingly, we call this feature 'group
multicast'. The service is using unicast when only one destination is
found, otherwise it will use the bearer broadcast service to transfer
the messages. In the latter case, the receiving members filter arriving
messages by looking at the intended destination instance. If there is
no match, the message will be dropped, while still being considered
received and read as seen by the flow control mechanism.
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>
In this commit, we make it possible to send connectionless unicast
messages to any member corresponding to the given member identity,
when there is more than one such member. The sender must use a
TIPC_ADDR_NAME address to achieve this effect.
We also perform load balancing between the destinations, i.e., we
primarily select one which has advertised sufficient send window
to not cause a block/EAGAIN delay, if any. This mechanism is
overlayed on the always present round-robin selection.
Anycast messages are subject to the same start synchronization
and flow control mechanism as group broadcast messages.
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>
We now make it possible to send connectionless unicast messages
within a communication group. To send a message, the sender can use
either a direct port address, aka port identity, or an indirect port
name to be looked up.
This type of messages are subject to the same start synchronization
and flow control mechanism as group broadcast messages.
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>
We introduce an end-to-end flow control mechanism for group broadcast
messages. This ensures that no messages are ever lost because of
destination receive buffer overflow, with minimal impact on performance.
For now, the algorithm is based on the assumption that there is only one
active transmitter at any moment in time.
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>
Like with any other service, group members' availability can be
subscribed for by connecting to be topology server. However, because
the events arrive via a different socket than the member socket, there
is a real risk that membership events my arrive out of synch with the
actual JOIN/LEAVE action. I.e., it is possible to receive the first
messages from a new member before the corresponding JOIN event arrives,
just as it is possible to receive the last messages from a leaving
member after the LEAVE event has already been received.
Since each member socket is internally also subscribing for membership
events, we now fix this problem by passing those events on to the user
via the member socket. We leverage the already present member synch-
ronization protocol to guarantee correct message/event order. An event
is delivered to the user as an empty message where the two source
addresses identify the new/lost member. Furthermore, we set the MSG_OOB
bit in the message flags to mark it as an event. If the event is an
indication about a member loss we also set the MSG_EOR bit, so it can
be distinguished from a member addition event.
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>
With group communication, it becomes important for a message receiver to
identify not only from which socket (identfied by a node:port tuple) the
message was sent, but also the logical identity (type:instance) of the
sending member.
We fix this by adding a second instance of struct sockaddr_tipc to the
source address area when a message is read. The extra address struct
is filled in with data found in the received message header (type,) and
in the local member representation struct (instance.)
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>
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: Jon Maloy <jon.maloy@ericsson.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
