nvme: move the fabrics queue ready check routines to core

queue_rq() in pci only checks if the dispatched queue (nvmeq) is ready,
e.g. not being suspended. Since nvme_alloc_admin_tags() in reset flow
restarts the admin queue, users are able to submit admin commands to a
controller before reset_work() completes. Commands submitted under this
condition may interfere with commands that performs identify, IO queue
setup in reset_work(), and may result in a hang described in the
following patch.

As seen in the fabrics, user commands are prevented from being executed
under inproper controller states. We may reuse this logic to maintain a
clear admin queue during reset_work().

Signed-off-by: Tao Chiu <[email protected]>
Signed-off-by: Cody Wong <[email protected]>
Reviewed-by: Leon Chien <[email protected]>
Reviewed-by: Keith Busch <[email protected]>
Signed-off-by: Christoph Hellwig <[email protected]>

Author: Tao Chiu

drivers/nvme/host/core.c

@@ -639,6 +639,66 @@ static struct request *nvme_alloc_request_qid(struct request_queue *q,
 	return req;
 }
 
+/*
+ * For something we're not in a state to send to the device the default action
+ * is to busy it and retry it after the controller state is recovered.  However,
+ * if the controller is deleting or if anything is marked for failfast or
+ * nvme multipath it is immediately failed.
+ *
+ * Note: commands used to initialize the controller will be marked for failfast.
+ * Note: nvme cli/ioctl commands are marked for failfast.
+ */
+blk_status_t nvme_fail_nonready_command(struct nvme_ctrl *ctrl,
+		struct request *rq)
+{
+	if (ctrl->state != NVME_CTRL_DELETING_NOIO &&
+	    ctrl->state != NVME_CTRL_DEAD &&
+	    !test_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags) &&
+	    !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH))
+		return BLK_STS_RESOURCE;
+	return nvme_host_path_error(rq);
+}
+EXPORT_SYMBOL_GPL(nvme_fail_nonready_command);
+
+bool __nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
+		bool queue_live)
+{
+	struct nvme_request *req = nvme_req(rq);
+
+	/*
+	 * currently we have a problem sending passthru commands
+	 * on the admin_q if the controller is not LIVE because we can't
+	 * make sure that they are going out after the admin connect,
+	 * controller enable and/or other commands in the initialization
+	 * sequence. until the controller will be LIVE, fail with
+	 * BLK_STS_RESOURCE so that they will be rescheduled.
+	 */
+	if (rq->q == ctrl->admin_q && (req->flags & NVME_REQ_USERCMD))
+		return false;
+
+	if (ctrl->ops->flags & NVME_F_FABRICS) {
+		/*
+		 * Only allow commands on a live queue, except for the connect
+		 * command, which is require to set the queue live in the
+		 * appropinquate states.
+		 */
+		switch (ctrl->state) {
+		case NVME_CTRL_CONNECTING:
+			if (blk_rq_is_passthrough(rq) && nvme_is_fabrics(req->cmd) &&
+			    req->cmd->fabrics.fctype == nvme_fabrics_type_connect)
+				return true;
+			break;
+		default:
+			break;
+		case NVME_CTRL_DEAD:
+			return false;
+		}
+	}
+
+	return queue_live;
+}
+EXPORT_SYMBOL_GPL(__nvme_check_ready);
+
 static int nvme_toggle_streams(struct nvme_ctrl *ctrl, bool enable)
 {
 	struct nvme_command c;

drivers/nvme/host/fabrics.c

@@ -533,63 +533,6 @@ static struct nvmf_transport_ops *nvmf_lookup_transport(
 	return NULL;
 }
 
-/*
- * For something we're not in a state to send to the device the default action
- * is to busy it and retry it after the controller state is recovered.  However,
- * if the controller is deleting or if anything is marked for failfast or
- * nvme multipath it is immediately failed.
- *
- * Note: commands used to initialize the controller will be marked for failfast.
- * Note: nvme cli/ioctl commands are marked for failfast.
- */
-blk_status_t nvmf_fail_nonready_command(struct nvme_ctrl *ctrl,
-		struct request *rq)
-{
-	if (ctrl->state != NVME_CTRL_DELETING_NOIO &&
-	    ctrl->state != NVME_CTRL_DEAD &&
-	    !test_bit(NVME_CTRL_FAILFAST_EXPIRED, &ctrl->flags) &&
-	    !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH))
-		return BLK_STS_RESOURCE;
-	return nvme_host_path_error(rq);
-}
-EXPORT_SYMBOL_GPL(nvmf_fail_nonready_command);
-
-bool __nvmf_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
-		bool queue_live)
-{
-	struct nvme_request *req = nvme_req(rq);
-
-	/*
-	 * currently we have a problem sending passthru commands
-	 * on the admin_q if the controller is not LIVE because we can't
-	 * make sure that they are going out after the admin connect,
-	 * controller enable and/or other commands in the initialization
-	 * sequence. until the controller will be LIVE, fail with
-	 * BLK_STS_RESOURCE so that they will be rescheduled.
-	 */
-	if (rq->q == ctrl->admin_q && (req->flags & NVME_REQ_USERCMD))
-		return false;
-
-	/*
-	 * Only allow commands on a live queue, except for the connect command,
-	 * which is require to set the queue live in the appropinquate states.
-	 */
-	switch (ctrl->state) {
-	case NVME_CTRL_CONNECTING:
-		if (blk_rq_is_passthrough(rq) && nvme_is_fabrics(req->cmd) &&
-		    req->cmd->fabrics.fctype == nvme_fabrics_type_connect)
-			return true;
-		break;
-	default:
-		break;
-	case NVME_CTRL_DEAD:
-		return false;
-	}
-
-	return queue_live;
-}
-EXPORT_SYMBOL_GPL(__nvmf_check_ready);
-
 static const match_table_t opt_tokens = {
 	{ NVMF_OPT_TRANSPORT,		"transport=%s"		},
 	{ NVMF_OPT_TRADDR,		"traddr=%s"		},

drivers/nvme/host/fabrics.h

@@ -184,20 +184,7 @@ void nvmf_unregister_transport(struct nvmf_transport_ops *ops);
 void nvmf_free_options(struct nvmf_ctrl_options *opts);
 int nvmf_get_address(struct nvme_ctrl *ctrl, char *buf, int size);
 bool nvmf_should_reconnect(struct nvme_ctrl *ctrl);
-blk_status_t nvmf_fail_nonready_command(struct nvme_ctrl *ctrl,
-		struct request *rq);
-bool __nvmf_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
-		bool queue_live);
 bool nvmf_ip_options_match(struct nvme_ctrl *ctrl,
 		struct nvmf_ctrl_options *opts);
 
-static inline bool nvmf_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
-		bool queue_live)
-{
-	if (likely(ctrl->state == NVME_CTRL_LIVE ||
-		   ctrl->state == NVME_CTRL_DELETING))
-		return true;
-	return __nvmf_check_ready(ctrl, rq, queue_live);
-}
-
 #endif /* _NVME_FABRICS_H */

drivers/nvme/host/fc.c

@@ -2766,8 +2766,8 @@ nvme_fc_queue_rq(struct blk_mq_hw_ctx *hctx,
 	blk_status_t ret;
 
 	if (ctrl->rport->remoteport.port_state != FC_OBJSTATE_ONLINE ||
-	    !nvmf_check_ready(&queue->ctrl->ctrl, rq, queue_ready))
-		return nvmf_fail_nonready_command(&queue->ctrl->ctrl, rq);
+	    !nvme_check_ready(&queue->ctrl->ctrl, rq, queue_ready))
+		return nvme_fail_nonready_command(&queue->ctrl->ctrl, rq);
 
 	ret = nvme_setup_cmd(ns, rq);
 	if (ret)

drivers/nvme/host/nvme.h

@@ -638,6 +638,21 @@ struct request *nvme_alloc_request(struct request_queue *q,
 		struct nvme_command *cmd, blk_mq_req_flags_t flags);
 void nvme_cleanup_cmd(struct request *req);
 blk_status_t nvme_setup_cmd(struct nvme_ns *ns, struct request *req);
+blk_status_t nvme_fail_nonready_command(struct nvme_ctrl *ctrl,
+		struct request *req);
+bool __nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
+		bool queue_live);
+
+static inline bool nvme_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
+		bool queue_live)
+{
+	if (likely(ctrl->state == NVME_CTRL_LIVE))
+		return true;
+	if (ctrl->ops->flags & NVME_F_FABRICS &&
+	    ctrl->state == NVME_CTRL_DELETING)
+		return true;
+	return __nvme_check_ready(ctrl, rq, queue_live);
+}
 int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
 		void *buf, unsigned bufflen);
 int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,

drivers/nvme/host/rdma.c

@@ -2050,8 +2050,8 @@ static blk_status_t nvme_rdma_queue_rq(struct blk_mq_hw_ctx *hctx,
 
 	WARN_ON_ONCE(rq->tag < 0);
 
-	if (!nvmf_check_ready(&queue->ctrl->ctrl, rq, queue_ready))
-		return nvmf_fail_nonready_command(&queue->ctrl->ctrl, rq);
+	if (!nvme_check_ready(&queue->ctrl->ctrl, rq, queue_ready))
+		return nvme_fail_nonready_command(&queue->ctrl->ctrl, rq);
 
 	dev = queue->device->dev;
 

drivers/nvme/host/tcp.c

@@ -2338,8 +2338,8 @@ static blk_status_t nvme_tcp_queue_rq(struct blk_mq_hw_ctx *hctx,
 	bool queue_ready = test_bit(NVME_TCP_Q_LIVE, &queue->flags);
 	blk_status_t ret;
 
-	if (!nvmf_check_ready(&queue->ctrl->ctrl, rq, queue_ready))
-		return nvmf_fail_nonready_command(&queue->ctrl->ctrl, rq);
+	if (!nvme_check_ready(&queue->ctrl->ctrl, rq, queue_ready))
+		return nvme_fail_nonready_command(&queue->ctrl->ctrl, rq);
 
 	ret = nvme_tcp_setup_cmd_pdu(ns, rq);
 	if (unlikely(ret))

drivers/nvme/target/loop.c

@@ -138,8 +138,8 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx,
 	bool queue_ready = test_bit(NVME_LOOP_Q_LIVE, &queue->flags);
 	blk_status_t ret;
 
-	if (!nvmf_check_ready(&queue->ctrl->ctrl, req, queue_ready))
-		return nvmf_fail_nonready_command(&queue->ctrl->ctrl, req);
+	if (!nvme_check_ready(&queue->ctrl->ctrl, req, queue_ready))
+		return nvme_fail_nonready_command(&queue->ctrl->ctrl, req);
 
 	ret = nvme_setup_cmd(ns, req);
 	if (ret)