Fix the status code of canceled requests initiated by the host according
to TP4028 (Status Code 0x371):
"Command Aborted By host: The command was aborted as a result of host
action (e.g., the host disconnected the Fabric connection)."
Also in a multipath environment, unless otherwise specified, errors of
this type (path related) should be retried using a different path, if
one is available.
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Keith Busch <kbusch@kernel.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
        if (blk_mq_request_completed(req))
                return true;
 
-       nvme_req(req)->status = NVME_SC_HOST_PATH_ERROR;
+       nvme_req(req)->status = NVME_SC_HOST_ABORTED_CMD;
        blk_mq_complete_request(req);
        return true;
 }
 
                }
                break;
        case NVME_SC_HOST_PATH_ERROR:
+       case NVME_SC_HOST_ABORTED_CMD:
                /*
                 * Temporary transport disruption in talking to the controller.
                 * Try to send on a new path.
 
        NVME_SC_ANA_INACCESSIBLE        = 0x302,
        NVME_SC_ANA_TRANSITION          = 0x303,
        NVME_SC_HOST_PATH_ERROR         = 0x370,
+       NVME_SC_HOST_ABORTED_CMD        = 0x371,
 
        NVME_SC_CRD                     = 0x1800,
        NVME_SC_DNR                     = 0x4000,