This is allowed because all CM callback functions are invoked from thread
context.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5443 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Move IB/CM members into a new struct. Report channel pointer instead of CM ID
pointer in diagnostic messages.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5442 d57e44dd-8a1f-0410-8b47-8ef2f437770f
ib_send_cm_rep() checks the connection state before sending a response.
Hence checking ch->state before calling ib_send_cm_rep() is superfluous,
so remove that check and also the locking that is no longer needed.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5441 d57e44dd-8a1f-0410-8b47-8ef2f437770f
With the SRP protocol all RDMA operations are initiated by the target.
Since no RDMA operations are initiated by the initiator, do not grant
the initiator permission to submit RDMA reads or writes to the target.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5423 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Complain if srpt_abort_cmd() is called for an I/O context that is
being processed by SCST and not by the HCA.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5393 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Only complain about a missing completion for I/O contexts that are
in a state where the ib_srpt driver is waiting for the HCA.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5392 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Apparently with mlx4 firmware up to and including 2.30.8000 it is not
guaranteed that for a QP associated with an SRQ error completions are
generated for all pending work requests. Avoid triggering
srpt_pending_cmd_timeout() for missing error completions.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5370 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Invoking srpt_zerolength_write() before the queue pair has reached
the error state is a bug, so complain loudly if that happens.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5369 d57e44dd-8a1f-0410-8b47-8ef2f437770f
The overhead of atomic_add_return() is lower than that of a
spin_lock() / spin_unlock() pair, hence switch to the former.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5367 d57e44dd-8a1f-0410-8b47-8ef2f437770f
All callers guarantee that the completion queue is empty so it is
not necessary to invoke ib_poll_cq() from inside this function.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5366 d57e44dd-8a1f-0410-8b47-8ef2f437770f
After having changed the queue pair state into "error", queue an additional
work request instead of waiting for the last WQE event.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5357 d57e44dd-8a1f-0410-8b47-8ef2f437770f
This is necessary to avoid that a login gets rejected due to reusing a
queue pair number that has not yet been freed by the target side.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5356 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Store initiator and target port ID's once per nexus instead of in each
channel data structure.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5354 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Although harmless for SCSI commands with SIMPLE ordering, avoid that commands
received before RTU can get reordered.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5349 d57e44dd-8a1f-0410-8b47-8ef2f437770f
The mlx4_core driver generates a GID change event after a port has been
changed from IB into Ethernet mode. Avoid that this causes the following
error message to appear in the system log:
ib_srpt: ***ERROR***: received unrecognized IB event 18
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5346 d57e44dd-8a1f-0410-8b47-8ef2f437770f
All ioctx->state manipulations are serialized per command so it is
not necessary to use locking to protect these manipulations.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5345 d57e44dd-8a1f-0410-8b47-8ef2f437770f
Avoid that disabling a target triggers a race condition with
SRP relogin. At least in theory this race condition could result
in a kernel crash.
git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@5268 d57e44dd-8a1f-0410-8b47-8ef2f437770f
