linux/drivers/nvme/host
James Smart 134aedc9c1 nvme-fc: correct io timeout behavior
The transport io timeout behavior wasn't quite correct. It ignored
that the io error handler is supposed to be synchronous so it possibly
allowed the blk request to be restarted while the io associated was
still aborting. Timeouts on reserved commands, those used for
association create, were never timing out thus they hung out forever.

To correct:
If an io is times out while a remoteport is not connected, just
restart the io timer. The lack of connectivity will simultaneously
be resetting the controller, so the reset path will abort and terminate
the io.

If an io is times out while it was marked for transport abort, just
reset the io timer. The abort process is underway and will complete
the io.

Otherwise, if an io times out, abort the io. If the abort was
unsuccessful (unlikely) give up and return not handled.

If the abort was successful, as the abort process is underway it will
terminate the io, so rather than synchronously waiting, just restart
the io timer.

Signed-off-by: James Smart <james.smart@broadcom.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
2017-10-20 12:17:05 +02:00
..
core.c nvme: use ida_simple_{get,remove} for the controller instance 2017-10-20 12:16:57 +02:00
fabrics.c nvme-fabrics: request transport module 2017-10-04 09:43:58 +02:00
fabrics.h nvme: read the subsystem NQN from Identify Controller 2017-06-28 08:14:13 -06:00
fc.c nvme-fc: correct io timeout behavior 2017-10-20 12:17:05 +02:00
Kconfig nvme: Remove SCSI translations 2017-06-28 08:14:13 -06:00
lightnvm.c nvme: fix lightnvm check 2017-09-11 12:29:36 -04:00
Makefile nvme: Makefile: remove dead build rule 2017-06-29 09:43:23 -06:00
nvme.h nvme: introduce nvme_reinit_tagset 2017-10-18 19:27:48 +02:00
pci.c nvme-pci: add SGL support 2017-10-20 12:16:58 +02:00
rdma.c nvme-rdma: Add BLK_MQ_F_NO_SCHED flag to admin tag set 2017-10-19 10:58:53 +02:00