Letting tipc_poll() dereference a socket's pointer to struct tipc_group
entails a race risk, as the group item may be deleted in a concurrent
tipc_sk_join() or tipc_sk_leave() thread.
We now move the 'open' flag in struct tipc_group to struct tipc_sock,
and let the former retain only a pointer to the moved field. This will
eliminate the race risk.
Reported-by: syzbot+799dafde0286795858ac@syzkaller.appspotmail.com
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current criteria for returning POLLOUT from a group member socket is
too simplistic. It basically returns POLLOUT as soon as the group has
external destinations, something obviously leading to a lot of spinning
during destination congestion situations. At the same time, the internal
congestion handling is unnecessarily complex.
We now change this as follows.
- We introduce an 'open' flag in struct tipc_group. This flag is used
only to help poll() get the setting of POLLOUT right, and *not* for
congeston handling as such. This means that a user can choose to
ignore an EAGAIN for a destination and go on sending messages to
other destinations in the group if he wants to.
- The flag is set to false every time we return EAGAIN on a send call.
- The flag is set to true every time any member, i.e., not necessarily
the member that caused EAGAIN, is removed from the small_win list.
- We remove the group member 'usr_pending' flag. The size of the send
window and presence in the 'small_win' list is sufficient criteria
for recognizing congestion.
This solution seems to be a reasonable compromise between 'anycast',
which is normally not waiting for POLLOUT for a specific destination,
and the other three send modes, which are.
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>
When a socket is joining a group, we look up in the binding table to
find if there are already other members of the group present. This is
used for being able to return EAGAIN instead of EHOSTUNREACH if the
user proceeds directly to a send attempt.
However, the information in the binding table can be used to directly
set the created member in state MBR_PUBLISHED and send a JOIN message
to the peer, instead of waiting for a topology PUBLISH event to do this.
When there are many members in a group, the propagation time for such
events can be significant, and we can save time during the join
operation if we use the initial lookup result fully.
In this commit, we eliminate the member state MBR_DISCOVERED which has
been the result of the initial lookup, and do instead go directly to
MBR_PUBLISHED, which initiates the setup.
After this change, the tipc_member FSM looks as follows:
+-----------+
---->| PUBLISHED |-----------------------------------------------+
PUB- +-----------+ LEAVE/WITHRAW |
LISH |JOIN |
| +-------------------------------------------+ |
| | LEAVE/WITHDRAW | |
| | +------------+ | |
| | +----------->| PENDING |---------+ | |
| | |msg/maxactv +-+---+------+ LEAVE/ | | |
| | | | | WITHDRAW | | |
| | | +----------+ | | | |
| | | |revert/maxactv| | | |
| | | V V V V V
| +----------+ msg +------------+ +-----------+
+-->| JOINED |------>| ACTIVE |------>| LEAVING |--->
| +----------+ +--- -+------+ LEAVE/+-----------+DOWN
| A A | WITHDRAW A A A EVT
| | | |RECLAIM | | |
| | |REMIT V | | |
| | |== adv +------------+ | | |
| | +---------| RECLAIMING |--------+ | |
| | +-----+------+ LEAVE/ | |
| | |REMIT WITHDRAW | |
| | |< adv | |
| |msg/ V LEAVE/ | |
| |adv==ADV_IDLE+------------+ WITHDRAW | |
| +-------------| REMITTED |------------+ |
| +------------+ |
|PUBLISH |
JOIN +-----------+ LEAVE/WITHDRAW |
---->| JOINING |-----------------------------------------------+
+-----------+
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 the current implementation, a group socket receiving topology
events about other members just converts the topology event message
into a group event message and stores it until it reaches the right
state to issue it to the user. This complicates the code unnecessarily,
and becomes impractical when we in the coming commits will need to
create and issue membership events independently.
In this commit, we change this so that we just notice the type and
origin of the incoming topology event, and then drop the buffer. Only
when it is time to actually send a group event to the user do we
explicitly create a new message and send it upwards.
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 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>
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>
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>
Jon Maloy