Christoph Hellwig e54b064cb2 nvme: move the dying queue check from cancel to completion
With multipath we don't want a hard DNR bit on a request that is cancelled
by a controller reset, but instead want to be able to retry it on another
patch.  To archive this don't always set the DNR bit when the queue is
dying in nvme_cancel_request, but defer that decision to
nvme_req_needs_retry.  Note that it applies to any command there and not
just cancelled commands, but one the queue is dying that is the right
thing to do anyway.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-11-10 19:53:25 -07:00
..
2017-09-22 17:28:59 -10:00
2017-09-10 20:40:00 -07:00
2017-09-07 14:03:05 -07:00
2017-09-07 13:51:13 -07:00
2017-09-07 12:53:14 -07:00
2017-09-14 17:34:43 +02:00
2017-09-07 13:51:13 -07:00
2017-09-05 11:08:17 -07:00
2017-09-10 20:40:00 -07:00
2017-08-29 13:46:35 +02:00
2017-09-12 13:30:06 -07:00
2017-09-05 11:08:17 -07:00
2017-09-11 13:04:32 -07:00
2017-09-07 13:51:13 -07:00
2017-09-09 14:34:38 -07:00
2017-09-13 10:56:00 -07:00
2017-09-19 08:36:19 +02:00
2017-09-05 12:45:03 -07:00
2017-09-05 12:45:03 -07:00
2017-08-30 14:03:42 -06:00
2017-09-14 13:33:33 -07:00
2017-09-07 21:11:05 -07:00
2017-09-22 06:40:47 -10:00
2017-08-24 09:57:28 +02:00