linux-stable.git/net/smc, branch linux-5.0.y

net/smc: fix smc_poll in SMC_INIT state

2019-02-21T18:19:20+00:00

smc_poll() returns with mask bit EPOLLPRI if the connection urg_state
is SMC_URG_VALID. Since SMC_URG_VALID is zero, smc_poll signals
EPOLLPRI errorneously if called in state SMC_INIT before the connection
is created, for instance in a non-blocking connect scenario.

This patch switches to non-zero values for the urg states.

Reviewed-by: Karsten Graul 
Fixes: de8474eb9d50 ("net/smc: urgent data support")
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: fix byte_order for rx_curs_confirmed

2019-02-09T06:33:25+00:00

The recent change in the rx_curs_confirmed assignment disregards
byte order, which causes problems on little endian architectures.
This patch fixes it.

Fixes: b8649efad879 ("net/smc: fix sender_free computation") (net-tree)
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: correct state change for peer closing

2019-02-04T17:11:19+00:00

If some kind of closing is received from the peer while still in
state SMC_INIT, it means the peer has had an active connection and
closed the socket quickly before listen_work finished. This should
not result in a shortcut from state SMC_INIT to state SMC_CLOSED.
This patch adds the socket to the accept queue in state
SMC_APPCLOSEWAIT1. The socket reaches state SMC_CLOSED once being
accepted and closed with smc_release().

Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: delete rkey first before switching to unused

2019-02-04T17:11:19+00:00

Once RMBs are flagged as unused they are candidates for reuse.
Thus the LLC DELETE RKEY operaton should be made before flagging
the RMB as unused.

Fixes: c7674c001b11 ("net/smc: unregister rkeys of unused buffer")
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: fix sender_free computation

2019-02-04T17:11:19+00:00

In some scenarios a separate consumer cursor update is necessary.
The decision is made in smc_tx_consumer_cursor_update(). The
sender_free computation could be wrong:

The rx confirmed cursor is always smaller than or equal to the
rx producer cursor. The parameters in the smc_curs_diff() call
have to be exchanged, otherwise sender_free might even be negative.

And if more data arrives local_rx_ctrl.prod might be updated, enabling
a cursor difference between local_rx_ctrl.prod and rx confirmed cursor
larger than the RMB size. This case is not covered by smc_curs_diff().
Thus function smc_curs_diff_large() is introduced here.

If a recvmsg() is processed in parallel, local_tx_ctrl.cons might
change during smc_cdc_msg_send. Make sure rx_curs_confirmed is updated
with the actually sent local_tx_ctrl.cons value.

Fixes: e82f2e31f559 ("net/smc: optimize consumer cursor updates")
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: preallocated memory for rdma work requests

2019-02-04T17:11:19+00:00

The work requests for rdma writes are built in local variables within
function smc_tx_rdma_write(). This violates the rule that the work
request storage has to stay till the work request is confirmed by
a completion queue response.
This patch introduces preallocated memory for these work requests.
The storage is allocated, once a link (and thus a queue pair) is
established.

Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: fix use of variable in cleared area

2019-02-01T22:45:45+00:00

Do not use pend->idx as index for the arrays because its value is
located in the cleared area. Use the existing local variable instead.
Without this fix the wrong area might be cleared.

Signed-off-by: Karsten Graul 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: use device link provided in qp_context

2019-02-01T22:45:45+00:00

The device field of the IB event structure does not always point to the
SMC IB device. Load the pointer from the qp_context which is always
provided to smc_ib_qp_event_handler() in the priv field. And for qp
events the affected port is given in the qp structure of the ibevent,
derive it from there.

Signed-off-by: Karsten Graul 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: call smc_cdc_msg_send() under send_lock

2019-02-01T22:45:45+00:00

Call smc_cdc_msg_send() under the connection send_lock to make sure all
send operations for one connection are serialized.

Signed-off-by: Karsten Graul 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller

net/smc: do not wait under send_lock

2019-02-01T22:45:45+00:00

smc_cdc_get_free_slot() might wait for free transfer buffers when using
SMC-R. This wait should not be done under the send_lock, which is a
spin_lock. This fixes a cpu loop in parallel threads waiting for the
send_lock.

Signed-off-by: Karsten Graul 
Signed-off-by: Ursula Braun 
Signed-off-by: David S. Miller