Merge tag 'for-5.4/block-2019-09-16' of git://git.kernel.dk/linux-block

Pull block updates from Jens Axboe:

 - Two NVMe pull requests:
     - ana log parse fix from Anton
     - nvme quirks support for Apple devices from Ben
     - fix missing bio completion tracing for multipath stack devices
       from Hannes and Mikhail
     - IP TOS settings for nvme rdma and tcp transports from Israel
     - rq_dma_dir cleanups from Israel
     - tracing for Get LBA Status command from Minwoo
     - Some nvme-tcp cleanups from Minwoo, Potnuri and Myself
     - Some consolidation between the fabrics transports for handling
       the CAP register
     - reset race with ns scanning fix for fabrics (move fabrics
       commands to a dedicated request queue with a different lifetime
       from the admin request queue)."
     - controller reset and namespace scan races fixes
     - nvme discovery log change uevent support
     - naming improvements from Keith
     - multiple discovery controllers reject fix from James
     - some regular cleanups from various people

 - Series fixing (and re-fixing) null_blk debug printing and nr_devices
   checks (André)

 - A few pull requests from Song, with fixes from Andy, Guoqing,
   Guilherme, Neil, Nigel, and Yufen.

 - REQ_OP_ZONE_RESET_ALL support (Chaitanya)

 - Bio merge handling unification (Christoph)

 - Pick default elevator correctly for devices with special needs
   (Damien)

 - Block stats fixes (Hou)

 - Timeout and support devices nbd fixes (Mike)

 - Series fixing races around elevator switching and device add/remove
   (Ming)

 - sed-opal cleanups (Revanth)

 - Per device weight support for BFQ (Fam)

 - Support for blk-iocost, a new model that can properly account cost of
   IO workloads. (Tejun)

 - blk-cgroup writeback fixes (Tejun)

 - paride queue init fixes (zhengbin)

 - blk_set_runtime_active() cleanup (Stanley)

 - Block segment mapping optimizations (Bart)

 - lightnvm fixes (Hans/Minwoo/YueHaibing)

 - Various little fixes and cleanups

* tag 'for-5.4/block-2019-09-16' of git://git.kernel.dk/linux-block: (186 commits)
  null_blk: format pr_* logs with pr_fmt
  null_blk: match the type of parameter nr_devices
  null_blk: do not fail the module load with zero devices
  block: also check RQF_STATS in blk_mq_need_time_stamp()
  block: make rq sector size accessible for block stats
  bfq: Fix bfq linkage error
  raid5: use bio_end_sector in r5_next_bio
  raid5: remove STRIPE_OPS_REQ_PENDING
  md: add feature flag MD_FEATURE_RAID0_LAYOUT
  md/raid0: avoid RAID0 data corruption due to layout confusion.
  raid5: don't set STRIPE_HANDLE to stripe which is in batch list
  raid5: don't increment read_errors on EILSEQ return
  nvmet: fix a wrong error status returned in error log page
  nvme: send discovery log page change events to userspace
  nvme: add uevent variables for controller devices
  nvme: enable aen regardless of the presence of I/O queues
  nvme-fabrics: allow discovery subsystems accept a kato
  nvmet: Use PTR_ERR_OR_ZERO() in nvmet_init_discovery()
  nvme: Remove redundant assignment of cq vector
  nvme: Assign subsys instance from first ctrl
  ...

12 files changed, 500 insertions, 230 deletions

diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig
index ec43ac9199e2..2b36f052bfb9 100644
--- a/drivers/nvme/host/Kconfig
+++ b/drivers/nvme/host/Kconfig
@@ -64,6 +64,7 @@ config NVME_TCP
 	depends on INET
 	depends on BLK_DEV_NVME
 	select NVME_FABRICS
+	select CRYPTO_CRC32C
 	help
 	  This provides support for the NVMe over Fabrics protocol using
 	  the TCP transport.  This allows you to use remote block devices
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index d3d6b7bd6903..1ede1763a5ee 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -22,12 +22,12 @@
 #include <linux/pm_qos.h>
 #include <asm/unaligned.h>
 
-#define CREATE_TRACE_POINTS
-#include "trace.h"
-
 #include "nvme.h"
 #include "fabrics.h"
 
+#define CREATE_TRACE_POINTS
+#include "trace.h"
+
 #define NVME_MINORS		(1U << MINORBITS)
 
 unsigned int admin_timeout = 60;
@@ -81,7 +81,6 @@ EXPORT_SYMBOL_GPL(nvme_reset_wq);
 struct workqueue_struct *nvme_delete_wq;
 EXPORT_SYMBOL_GPL(nvme_delete_wq);
 
-static DEFINE_IDA(nvme_subsystems_ida);
 static LIST_HEAD(nvme_subsystems);
 static DEFINE_MUTEX(nvme_subsystems_lock);
 
@@ -197,9 +196,9 @@ static inline bool nvme_ns_has_pi(struct nvme_ns *ns)
 	return ns->pi_type && ns->ms == sizeof(struct t10_pi_tuple);
 }
 
-static blk_status_t nvme_error_status(struct request *req)
+static blk_status_t nvme_error_status(u16 status)
 {
-	switch (nvme_req(req)->status & 0x7ff) {
+	switch (status & 0x7ff) {
 	case NVME_SC_SUCCESS:
 		return BLK_STS_OK;
 	case NVME_SC_CAP_EXCEEDED:
@@ -226,6 +225,8 @@ static blk_status_t nvme_error_status(struct request *req)
 		return BLK_STS_PROTECTION;
 	case NVME_SC_RESERVATION_CONFLICT:
 		return BLK_STS_NEXUS;
+	case NVME_SC_HOST_PATH_ERROR:
+		return BLK_STS_TRANSPORT;
 	default:
 		return BLK_STS_IOERR;
 	}
@@ -260,7 +261,7 @@ static void nvme_retry_req(struct request *req)
 
 void nvme_complete_rq(struct request *req)
 {
-	blk_status_t status = nvme_error_status(req);
+	blk_status_t status = nvme_error_status(nvme_req(req)->status);
 
 	trace_nvme_complete_rq(req);
 
@@ -279,6 +280,8 @@ void nvme_complete_rq(struct request *req)
 			return;
 		}
 	}
+
+	nvme_trace_bio_complete(req, status);
 	blk_mq_end_request(req, status);
 }
 EXPORT_SYMBOL_GPL(nvme_complete_rq);
@@ -288,8 +291,12 @@ bool nvme_cancel_request(struct request *req, void *data, bool reserved)
 	dev_dbg_ratelimited(((struct nvme_ctrl *) data)->device,
 				"Cancelling I/O %d", req->tag);
 
-	nvme_req(req)->status = NVME_SC_ABORT_REQ;
-	blk_mq_complete_request_sync(req);
+	/* don't abort one completed request */
+	if (blk_mq_request_completed(req))
+		return true;
+
+	nvme_req(req)->status = NVME_SC_HOST_PATH_ERROR;
+	blk_mq_complete_request(req);
 	return true;
 }
 EXPORT_SYMBOL_GPL(nvme_cancel_request);
@@ -1088,10 +1095,9 @@ static int nvme_identify_ns_list(struct nvme_ctrl *dev, unsigned nsid, __le32 *n
 				    NVME_IDENTIFY_DATA_SIZE);
 }
 
-static struct nvme_id_ns *nvme_identify_ns(struct nvme_ctrl *ctrl,
-		unsigned nsid)
+static int nvme_identify_ns(struct nvme_ctrl *ctrl,
+		unsigned nsid, struct nvme_id_ns **id)
 {
-	struct nvme_id_ns *id;
 	struct nvme_command c = { };
 	int error;
 
@@ -1100,18 +1106,17 @@ static struct nvme_id_ns *nvme_identify_ns(struct nvme_ctrl *ctrl,
 	c.identify.nsid = cpu_to_le32(nsid);
 	c.identify.cns = NVME_ID_CNS_NS;
 
-	id = kmalloc(sizeof(*id), GFP_KERNEL);
-	if (!id)
-		return NULL;
+	*id = kmalloc(sizeof(**id), GFP_KERNEL);
+	if (!*id)
+		return -ENOMEM;
 
-	error = nvme_submit_sync_cmd(ctrl->admin_q, &c, id, sizeof(*id));
+	error = nvme_submit_sync_cmd(ctrl->admin_q, &c, *id, sizeof(**id));
 	if (error) {
 		dev_warn(ctrl->device, "Identify namespace failed (%d)\n", error);
-		kfree(id);
-		return NULL;
+		kfree(*id);
 	}
 
-	return id;
+	return error;
 }
 
 static int nvme_features(struct nvme_ctrl *dev, u8 op, unsigned int fid,
@@ -1180,7 +1185,8 @@ int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count)
 EXPORT_SYMBOL_GPL(nvme_set_queue_count);
 
 #define NVME_AEN_SUPPORTED \
-	(NVME_AEN_CFG_NS_ATTR | NVME_AEN_CFG_FW_ACT | NVME_AEN_CFG_ANA_CHANGE)
+	(NVME_AEN_CFG_NS_ATTR | NVME_AEN_CFG_FW_ACT | \
+	 NVME_AEN_CFG_ANA_CHANGE | NVME_AEN_CFG_DISC_CHANGE)
 
 static void nvme_enable_aen(struct nvme_ctrl *ctrl)
 {
@@ -1195,6 +1201,8 @@ static void nvme_enable_aen(struct nvme_ctrl *ctrl)
 	if (status)
 		dev_warn(ctrl->device, "Failed to configure AEN (cfg %x)\n",
 			 supported_aens);
+
+	queue_work(nvme_wq, &ctrl->async_event_work);
 }
 
 static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio)
@@ -1594,9 +1602,11 @@ static void nvme_config_write_zeroes(struct gendisk *disk, struct nvme_ns *ns)
 	blk_queue_max_write_zeroes_sectors(disk->queue, max_sectors);
 }
 
-static void nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid,
+static int nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid,
 		struct nvme_id_ns *id, struct nvme_ns_ids *ids)
 {
+	int ret = 0;
+
 	memset(ids, 0, sizeof(*ids));
 
 	if (ctrl->vs >= NVME_VS(1, 1, 0))
@@ -1607,10 +1617,12 @@ static void nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid,
 		 /* Don't treat error as fatal we potentially
 		  * already have a NGUID or EUI-64
 		  */
-		if (nvme_identify_ns_descs(ctrl, nsid, ids))
+		ret = nvme_identify_ns_descs(ctrl, nsid, ids);
+		if (ret)
 			dev_warn(ctrl->device,
-				 "%s: Identify Descriptors failed\n", __func__);
+				 "Identify Descriptors failed (%d)\n", ret);
 	}
+	return ret;
 }
 
 static bool nvme_ns_ids_valid(struct nvme_ns_ids *ids)
@@ -1738,25 +1750,37 @@ static int nvme_revalidate_disk(struct gendisk *disk)
 		return -ENODEV;
 	}
 
-	id = nvme_identify_ns(ctrl, ns->head->ns_id);
-	if (!id)
-		return -ENODEV;
+	ret = nvme_identify_ns(ctrl, ns->head->ns_id, &id);
+	if (ret)
+		goto out;
 
 	if (id->ncap == 0) {
 		ret = -ENODEV;
-		goto out;
+		goto free_id;
 	}
 
 	__nvme_revalidate_disk(disk, id);
-	nvme_report_ns_ids(ctrl, ns->head->ns_id, id, &ids);
+	ret = nvme_report_ns_ids(ctrl, ns->head->ns_id, id, &ids);
+	if (ret)
+		goto free_id;
+
 	if (!nvme_ns_ids_equal(&ns->head->ids, &ids)) {
 		dev_err(ctrl->device,
 			"identifiers changed for nsid %d\n", ns->head->ns_id);
 		ret = -ENODEV;
 	}
 
-out:
+free_id:
 	kfree(id);
+out:
+	/*
+	 * Only fail the function if we got a fatal error back from the
+	 * device, otherwise ignore the error and just move on.
+	 */
+	if (ret == -ENOMEM || (ret > 0 && !(ret & NVME_SC_DNR)))
+		ret = 0;
+	else if (ret > 0)
+		ret = blk_status_to_errno(nvme_error_status(ret));
 	return ret;
 }
 
@@ -1952,7 +1976,7 @@ static int nvme_wait_ready(struct nvme_ctrl *ctrl, u64 cap, bool enabled)
  * bits', but doing so may cause the device to complete commands to the
  * admin queue ... and we don't know what memory that might be pointing at!
  */
-int nvme_disable_ctrl(struct nvme_ctrl *ctrl, u64 cap)
+int nvme_disable_ctrl(struct nvme_ctrl *ctrl)
 {
 	int ret;
 
@@ -1966,20 +1990,27 @@ int nvme_disable_ctrl(struct nvme_ctrl *ctrl, u64 cap)
 	if (ctrl->quirks & NVME_QUIRK_DELAY_BEFORE_CHK_RDY)
 		msleep(NVME_QUIRK_DELAY_AMOUNT);
 
-	return nvme_wait_ready(ctrl, cap, false);
+	return nvme_wait_ready(ctrl, ctrl->cap, false);
 }
 EXPORT_SYMBOL_GPL(nvme_disable_ctrl);
 
-int nvme_enable_ctrl(struct nvme_ctrl *ctrl, u64 cap)
+int nvme_enable_ctrl(struct nvme_ctrl *ctrl)
 {
 	/*
 	 * Default to a 4K page size, with the intention to update this
 	 * path in the future to accomodate architectures with differing
 	 * kernel and IO page sizes.
 	 */
-	unsigned dev_page_min = NVME_CAP_MPSMIN(cap) + 12, page_shift = 12;
+	unsigned dev_page_min, page_shift = 12;
 	int ret;
 
+	ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &ctrl->cap);
+	if (ret) {
+		dev_err(ctrl->device, "Reading CAP failed (%d)\n", ret);
+		return ret;
+	}
+	dev_page_min = NVME_CAP_MPSMIN(ctrl->cap) + 12;
+
 	if (page_shift < dev_page_min) {
 		dev_err(ctrl->device,
 			"Minimum device page size %u too large for host (%u)\n",
@@ -1998,7 +2029,7 @@ int nvme_enable_ctrl(struct nvme_ctrl *ctrl, u64 cap)
 	ret = ctrl->ops->reg_write32(ctrl, NVME_REG_CC, ctrl->ctrl_config);
 	if (ret)
 		return ret;
-	return nvme_wait_ready(ctrl, cap, true);
+	return nvme_wait_ready(ctrl, ctrl->cap, true);
 }
 EXPORT_SYMBOL_GPL(nvme_enable_ctrl);
 
@@ -2332,7 +2363,8 @@ static void nvme_release_subsystem(struct device *dev)
 	struct nvme_subsystem *subsys =
 		container_of(dev, struct nvme_subsystem, dev);
 
-	ida_simple_remove(&nvme_subsystems_ida, subsys->instance);
+	if (subsys->instance >= 0)
+		ida_simple_remove(&nvme_instance_ida, subsys->instance);
 	kfree(subsys);
 }
 
@@ -2361,6 +2393,17 @@ static struct nvme_subsystem *__nvme_find_get_subsystem(const char *subsysnqn)
 
 	lockdep_assert_held(&nvme_subsystems_lock);
 
+	/*
+	 * Fail matches for discovery subsystems. This results
+	 * in each discovery controller bound to a unique subsystem.
+	 * This avoids issues with validating controller values
+	 * that can only be true when there is a single unique subsystem.
+	 * There may be multiple and completely independent entities
+	 * that provide discovery controllers.
+	 */
+	if (!strcmp(subsysnqn, NVME_DISC_SUBSYS_NAME))
+		return NULL;
+
 	list_for_each_entry(subsys, &nvme_subsystems, entry) {
 		if (strcmp(subsys->subnqn, subsysnqn))
 			continue;
@@ -2461,12 +2504,8 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
 	subsys = kzalloc(sizeof(*subsys), GFP_KERNEL);
 	if (!subsys)
 		return -ENOMEM;
-	ret = ida_simple_get(&nvme_subsystems_ida, 0, 0, GFP_KERNEL);
-	if (ret < 0) {
-		kfree(subsys);
-		return ret;
-	}
-	subsys->instance = ret;
+
+	subsys->instance = -1;
 	mutex_init(&subsys->lock);
 	kref_init(&subsys->ref);
 	INIT_LIST_HEAD(&subsys->ctrls);
@@ -2485,7 +2524,7 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
 	subsys->dev.class = nvme_subsys_class;
 	subsys->dev.release = nvme_release_subsystem;
 	subsys->dev.groups = nvme_subsys_attrs_groups;
-	dev_set_name(&subsys->dev, "nvme-subsys%d", subsys->instance);
+	dev_set_name(&subsys->dev, "nvme-subsys%d", ctrl->instance);
 	device_initialize(&subsys->dev);
 
 	mutex_lock(&nvme_subsystems_lock);
@@ -2517,6 +2556,8 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
 		goto out_put_subsystem;
 	}
 
+	if (!found)
+		subsys->instance = ctrl->instance;
 	ctrl->subsys = subsys;
 	list_add_tail(&ctrl->subsys_entry, &subsys->ctrls);
 	mutex_unlock(&nvme_subsystems_lock);
@@ -2574,7 +2615,6 @@ static int nvme_get_effects_log(struct nvme_ctrl *ctrl)
 int nvme_init_identify(struct nvme_ctrl *ctrl)
 {
 	struct nvme_id_ctrl *id;
-	u64 cap;
 	int ret, page_shift;
 	u32 max_hw_sectors;
 	bool prev_apst_enabled;
@@ -2584,16 +2624,11 @@ int nvme_init_identify(struct nvme_ctrl *ctrl)
 		dev_err(ctrl->device, "Reading VS failed (%d)\n", ret);
 		return ret;
 	}
-
-	ret = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &cap);
-	if (ret) {
-		dev_err(ctrl->device, "Reading CAP failed (%d)\n", ret);
-		return ret;
-	}
-	page_shift = NVME_CAP_MPSMIN(cap) + 12;
+	page_shift = NVME_CAP_MPSMIN(ctrl->cap) + 12;
+	ctrl->sqsize = min_t(int, NVME_CAP_MQES(ctrl->cap), ctrl->sqsize);
 
 	if (ctrl->vs >= NVME_VS(1, 1, 0))
-		ctrl->subsystem = NVME_CAP_NSSRC(cap);
+		ctrl->subsystem = NVME_CAP_NSSRC(ctrl->cap);
 
 	ret = nvme_identify_ctrl(ctrl, &id);
 	if (ret) {
@@ -3184,7 +3219,9 @@ static struct nvme_ns_head *nvme_alloc_ns_head(struct nvme_ctrl *ctrl,
 	head->ns_id = nsid;
 	kref_init(&head->ref);
 
-	nvme_report_ns_ids(ctrl, nsid, id, &head->ids);
+	ret = nvme_report_ns_ids(ctrl, nsid, id, &head->ids);
+	if (ret)
+		goto out_cleanup_srcu;
 
 	ret = __nvme_check_ids(ctrl->subsys, head);
 	if (ret) {
@@ -3209,6 +3246,8 @@ out_ida_remove:
 out_free_head:
 	kfree(head);
 out:
+	if (ret > 0)
+		ret = blk_status_to_errno(nvme_error_status(ret));
 	return ERR_PTR(ret);
 }
 
@@ -3232,7 +3271,10 @@ static int nvme_init_ns_head(struct nvme_ns *ns, unsigned nsid,
 	} else {
 		struct nvme_ns_ids ids;
 
-		nvme_report_ns_ids(ctrl, nsid, id, &ids);
+		ret = nvme_report_ns_ids(ctrl, nsid, id, &ids);
+		if (ret)
+			goto out_unlock;
+
 		if (!nvme_ns_ids_equal(&head->ids, &ids)) {
 			dev_err(ctrl->device,
 				"IDs don't match for shared namespace %d\n",
@@ -3247,6 +3289,8 @@ static int nvme_init_ns_head(struct nvme_ns *ns, unsigned nsid,
 
 out_unlock:
 	mutex_unlock(&ctrl->subsys->lock);
+	if (ret > 0)
+		ret = blk_status_to_errno(nvme_error_status(ret));
 	return ret;
 }
 
@@ -3338,11 +3382,9 @@ static int nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift);
 	nvme_set_queue_limits(ctrl, ns->queue);
 
-	id = nvme_identify_ns(ctrl, nsid);
-	if (!id) {
-		ret = -EIO;
+	ret = nvme_identify_ns(ctrl, nsid, &id);
+	if (ret)
 		goto out_free_queue;
-	}
 
 	if (id->ncap == 0) {
 		ret = -EINVAL;
@@ -3404,6 +3446,8 @@ static int nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	blk_cleanup_queue(ns->queue);
  out_free_ns:
 	kfree(ns);
+	if (ret > 0)
+		ret = blk_status_to_errno(nvme_error_status(ret));
 	return ret;
 }
 
@@ -3617,6 +3661,33 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
 }
 EXPORT_SYMBOL_GPL(nvme_remove_namespaces);
 
+static int nvme_class_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct nvme_ctrl *ctrl =
+		container_of(dev, struct nvme_ctrl, ctrl_device);
+	struct nvmf_ctrl_options *opts = ctrl->opts;
+	int ret;
+
+	ret = add_uevent_var(env, "NVME_TRTYPE=%s", ctrl->ops->name);
+	if (ret)
+		return ret;
+
+	if (opts) {
+		ret = add_uevent_var(env, "NVME_TRADDR=%s", opts->traddr);
+		if (ret)
+			return ret;
+
+		ret = add_uevent_var(env, "NVME_TRSVCID=%s",
+				opts->trsvcid ?: "none");
+		if (ret)
+			return ret;
+
+		ret = add_uevent_var(env, "NVME_HOST_TRADDR=%s",
+				opts->host_traddr ?: "none");
+	}
+	return ret;
+}
+
 static void nvme_aen_uevent(struct nvme_ctrl *ctrl)
 {
 	char *envp[2] = { NULL, NULL };
@@ -3723,6 +3794,9 @@ static void nvme_handle_aen_notice(struct nvme_ctrl *ctrl, u32 result)
 		queue_work(nvme_wq, &ctrl->ana_work);
 		break;
 #endif
+	case NVME_AER_NOTICE_DISC_CHANGED:
+		ctrl->aen_result = result;
+		break;
 	default:
 		dev_warn(ctrl->device, "async event result %08x\n", result);
 	}
@@ -3769,10 +3843,10 @@ void nvme_start_ctrl(struct nvme_ctrl *ctrl)
 	if (ctrl->kato)
 		nvme_start_keep_alive(ctrl);
 
+	nvme_enable_aen(ctrl);
+
 	if (ctrl->queue_count > 1) {
 		nvme_queue_scan(ctrl);
-		nvme_enable_aen(ctrl);
-		queue_work(nvme_wq, &ctrl->async_event_work);
 		nvme_start_queues(ctrl);
 	}
 }
@@ -3792,7 +3866,9 @@ static void nvme_free_ctrl(struct device *dev)
 		container_of(dev, struct nvme_ctrl, ctrl_device);
 	struct nvme_subsystem *subsys = ctrl->subsys;
 
-	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
+	if (subsys && ctrl->instance != subsys->instance)
+		ida_simple_remove(&nvme_instance_ida, ctrl->instance);
+
 	kfree(ctrl->effects);
 	nvme_mpath_uninit(ctrl);
 	__free_page(ctrl->discard_page);
@@ -3992,6 +4068,9 @@ void nvme_sync_queues(struct nvme_ctrl *ctrl)
 	list_for_each_entry(ns, &ctrl->namespaces, list)
 		blk_sync_queue(ns->queue);
 	up_read(&ctrl->namespaces_rwsem);
+
+	if (ctrl->admin_q)
+		blk_sync_queue(ctrl->admin_q);
 }
 EXPORT_SYMBOL_GPL(nvme_sync_queues);
 
@@ -4050,6 +4129,7 @@ static int __init nvme_core_init(void)
 		result = PTR_ERR(nvme_class);
 		goto unregister_chrdev;
 	}
+	nvme_class->dev_uevent = nvme_class_uevent;
 
 	nvme_subsys_class = class_create(THIS_MODULE, "nvme-subsystem");
 	if (IS_ERR(nvme_subsys_class)) {
@@ -4074,7 +4154,6 @@ out:
 
 static void __exit nvme_core_exit(void)
 {
-	ida_destroy(&nvme_subsystems_ida);
 	class_destroy(nvme_subsys_class);
 	class_destroy(nvme_class);
 	unregister_chrdev_region(nvme_chr_devt, NVME_MINORS);
diff --git a/drivers/nvme/host/fabrics.c b/drivers/nvme/host/fabrics.c
index 1994d5b42f94..74b8818ac9a1 100644
--- a/drivers/nvme/host/fabrics.c
+++ b/drivers/nvme/host/fabrics.c
@@ -150,7 +150,7 @@ int nvmf_reg_read32(struct nvme_ctrl *ctrl, u32 off, u32 *val)
 	cmd.prop_get.fctype = nvme_fabrics_type_property_get;
 	cmd.prop_get.offset = cpu_to_le32(off);
 
-	ret = __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, &res, NULL, 0, 0,
+	ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, &res, NULL, 0, 0,
 			NVME_QID_ANY, 0, 0, false);
 
 	if (ret >= 0)
@@ -197,7 +197,7 @@ int nvmf_reg_read64(struct nvme_ctrl *ctrl, u32 off, u64 *val)
 	cmd.prop_get.attrib = 1;
 	cmd.prop_get.offset = cpu_to_le32(off);
 
-	ret = __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, &res, NULL, 0, 0,
+	ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, &res, NULL, 0, 0,
 			NVME_QID_ANY, 0, 0, false);
 
 	if (ret >= 0)
@@ -243,7 +243,7 @@ int nvmf_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val)
 	cmd.prop_set.offset = cpu_to_le32(off);
 	cmd.prop_set.value = cpu_to_le64(val);
 
-	ret = __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, NULL, NULL, 0, 0,
+	ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, NULL, NULL, 0, 0,
 			NVME_QID_ANY, 0, 0, false);
 	if (unlikely(ret))
 		dev_err(ctrl->device,
@@ -381,8 +381,8 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl)
 	 * Set keep-alive timeout in seconds granularity (ms * 1000)
 	 * and add a grace period for controller kato enforcement
 	 */
-	cmd.connect.kato = ctrl->opts->discovery_nqn ? 0 :
-		cpu_to_le32((ctrl->kato + NVME_KATO_GRACE) * 1000);
+	cmd.connect.kato = ctrl->kato ?
+		cpu_to_le32((ctrl->kato + NVME_KATO_GRACE) * 1000) : 0;
 
 	if (ctrl->opts->disable_sqflow)
 		cmd.connect.cattr |= NVME_CONNECT_DISABLE_SQFLOW;
@@ -396,7 +396,7 @@ int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl)
 	strncpy(data->subsysnqn, ctrl->opts->subsysnqn, NVMF_NQN_SIZE);
 	strncpy(data->hostnqn, ctrl->opts->host->nqn, NVMF_NQN_SIZE);
 
-	ret = __nvme_submit_sync_cmd(ctrl->admin_q, &cmd, &res,
+	ret = __nvme_submit_sync_cmd(ctrl->fabrics_q, &cmd, &res,
 			data, sizeof(*data), 0, NVME_QID_ANY, 1,
 			BLK_MQ_REQ_RESERVED | BLK_MQ_REQ_NOWAIT, false);
 	if (ret) {
@@ -611,6 +611,7 @@ static const match_table_t opt_tokens = {
 	{ NVMF_OPT_DATA_DIGEST,		"data_digest"		},
 	{ NVMF_OPT_NR_WRITE_QUEUES,	"nr_write_queues=%d"	},
 	{ NVMF_OPT_NR_POLL_QUEUES,	"nr_poll_queues=%d"	},
+	{ NVMF_OPT_TOS,			"tos=%d"		},
 	{ NVMF_OPT_ERR,			NULL			}
 };
 
@@ -632,6 +633,7 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 	opts->duplicate_connect = false;
 	opts->hdr_digest = false;
 	opts->data_digest = false;
+	opts->tos = -1; /* < 0 == use transport default */
 
 	options = o = kstrdup(buf, GFP_KERNEL);
 	if (!options)
@@ -738,13 +740,6 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 				pr_warn("keep_alive_tmo 0 won't execute keep alives!!!\n");
 			}
 			opts->kato = token;
-
-			if (opts->discovery_nqn && opts->kato) {
-				pr_err("Discovery controllers cannot accept KATO != 0\n");
-				ret = -EINVAL;
-				goto out;
-			}
-
 			break;
 		case NVMF_OPT_CTRL_LOSS_TMO:
 			if (match_int(args, &token)) {
@@ -856,6 +851,22 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 			}
 			opts->nr_poll_queues = token;
 			break;
+		case NVMF_OPT_TOS:
+			if (match_int(args, &token)) {
+				ret = -EINVAL;
+				goto out;
+			}
+			if (token < 0) {
+				pr_err("Invalid type of service %d\n", token);
+				ret = -EINVAL;
+				goto out;
+			}
+			if (token > 255) {
+				pr_warn("Clamping type of service to 255\n");
+				token = 255;
+			}
+			opts->tos = token;
+			break;
 		default:
 			pr_warn("unknown parameter or missing value '%s' in ctrl creation request\n",
 				p);
@@ -865,7 +876,6 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 	}
 
 	if (opts->discovery_nqn) {
-		opts->kato = 0;
 		opts->nr_io_queues = 0;
 		opts->nr_write_queues = 0;
 		opts->nr_poll_queues = 0;
diff --git a/drivers/nvme/host/fabrics.h b/drivers/nvme/host/fabrics.h
index 3044d8b99a24..93f08d77c896 100644
--- a/drivers/nvme/host/fabrics.h
+++ b/drivers/nvme/host/fabrics.h
@@ -55,6 +55,7 @@ enum {
 	NVMF_OPT_DATA_DIGEST	= 1 << 16,
 	NVMF_OPT_NR_WRITE_QUEUES = 1 << 17,
 	NVMF_OPT_NR_POLL_QUEUES = 1 << 18,
+	NVMF_OPT_TOS		= 1 << 19,
 };
 
 /**
@@ -87,6 +88,7 @@ enum {
  * @data_digest: generate/verify data digest (TCP)
  * @nr_write_queues: number of queues for write I/O
  * @nr_poll_queues: number of queues for polling I/O
+ * @tos: type of service
  */
 struct nvmf_ctrl_options {
 	unsigned		mask;
@@ -108,6 +110,7 @@ struct nvmf_ctrl_options {
 	bool			data_digest;
 	unsigned int		nr_write_queues;
 	unsigned int		nr_poll_queues;
+	int			tos;
 };
 
 /*
diff --git a/drivers/nvme/host/fc.c b/drivers/nvme/host/fc.c
index 232d8094091b..265f89e11d8b 100644
--- a/drivers/nvme/host/fc.c
+++ b/drivers/nvme/host/fc.c
@@ -1608,9 +1608,13 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req)
 				sizeof(op->rsp_iu), DMA_FROM_DEVICE);
 
 	if (opstate == FCPOP_STATE_ABORTED)
-		status = cpu_to_le16(NVME_SC_ABORT_REQ << 1);
-	else if (freq->status)
-		status = cpu_to_le16(NVME_SC_INTERNAL << 1);
+		status = cpu_to_le16(NVME_SC_HOST_PATH_ERROR << 1);
+	else if (freq->status) {
+		status = cpu_to_le16(NVME_SC_HOST_PATH_ERROR << 1);
+		dev_info(ctrl->ctrl.device,
+			"NVME-FC{%d}: io failed due to lldd error %d\n",
+			ctrl->cnum, freq->status);
+	}
 
 	/*
 	 * For the linux implementation, if we have an unsuccesful
@@ -1637,8 +1641,13 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req)
 		 * no payload in the CQE by the transport.
 		 */
 		if (freq->transferred_length !=
-			be32_to_cpu(op->cmd_iu.data_len)) {
-			status = cpu_to_le16(NVME_SC_INTERNAL << 1);
+		    be32_to_cpu(op->cmd_iu.data_len)) {
+			status = cpu_to_le16(NVME_SC_HOST_PATH_ERROR << 1);
+			dev_info(ctrl->ctrl.device,
+				"NVME-FC{%d}: io failed due to bad transfer "
+				"length: %d vs expected %d\n",
+				ctrl->cnum, freq->transferred_length,
+				be32_to_cpu(op->cmd_iu.data_len));
 			goto done;
 		}
 		result.u64 = 0;
@@ -1655,7 +1664,17 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req)
 					freq->transferred_length ||
 			     op->rsp_iu.status_code ||
 			     sqe->common.command_id != cqe->command_id)) {
-			status = cpu_to_le16(NVME_SC_INTERNAL << 1);
+			status = cpu_to_le16(NVME_SC_HOST_PATH_ERROR << 1);
+			dev_info(ctrl->ctrl.device,
+				"NVME-FC{%d}: io failed due to bad NVMe_ERSP: "
+				"iu len %d, xfr len %d vs %d, status code "
+				"%d, cmdid %d vs %d\n",
+				ctrl->cnum, be16_to_cpu(op->rsp_iu.iu_len),
+				be32_to_cpu(op->rsp_iu.xfrd_len),
+				freq->transferred_length,
+				op->rsp_iu.status_code,
+				sqe->common.command_id,
+				cqe->command_id);
 			goto done;
 		}
 		result = cqe->result;
@@ -1663,7 +1682,11 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req)
 		break;
 
 	default:
-		status = cpu_to_le16(NVME_SC_INTERNAL << 1);
+		status = cpu_to_le16(NVME_SC_HOST_PATH_ERROR << 1);
+		dev_info(ctrl->ctrl.device,
+			"NVME-FC{%d}: io failed due to odd NVMe_xRSP iu "
+			"len %d\n",
+			ctrl->cnum, freq->rcv_rsplen);
 		goto done;
 	}
 
@@ -2006,6 +2029,7 @@ nvme_fc_ctrl_free(struct kref *ref)
 
 	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 	blk_cleanup_queue(ctrl->ctrl.admin_q);
+	blk_cleanup_queue(ctrl->ctrl.fabrics_q);
 	blk_mq_free_tag_set(&ctrl->admin_tag_set);
 
 	kfree(ctrl->queues);
@@ -2107,7 +2131,6 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 		struct nvme_fc_fcp_op *op)
 {
 	struct nvmefc_fcp_req *freq = &op->fcp_req;
-	enum dma_data_direction dir;
 	int ret;
 
 	freq->sg_cnt = 0;
@@ -2124,9 +2147,8 @@ nvme_fc_map_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 
 	op->nents = blk_rq_map_sg(rq->q, rq, freq->sg_table.sgl);
 	WARN_ON(op->nents > blk_rq_nr_phys_segments(rq));
-	dir = (rq_data_dir(rq) == WRITE) ? DMA_TO_DEVICE : DMA_FROM_DEVICE;
 	freq->sg_cnt = fc_dma_map_sg(ctrl->lport->dev, freq->sg_table.sgl,
-				op->nents, dir);
+				op->nents, rq_dma_dir(rq));
 	if (unlikely(freq->sg_cnt <= 0)) {
 		sg_free_table_chained(&freq->sg_table, SG_CHUNK_SIZE);
 		freq->sg_cnt = 0;
@@ -2149,8 +2171,7 @@ nvme_fc_unmap_data(struct nvme_fc_ctrl *ctrl, struct request *rq,
 		return;
 
 	fc_dma_unmap_sg(ctrl->lport->dev, freq->sg_table.sgl, op->nents,
-				((rq_data_dir(rq) == WRITE) ?
-					DMA_TO_DEVICE : DMA_FROM_DEVICE));
+			rq_dma_dir(rq));
 
 	nvme_cleanup_cmd(rq);
 
@@ -2633,8 +2654,6 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl)
 	if (ret)
 		goto out_delete_hw_queue;
 
-	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
-
 	ret = nvmf_connect_admin_queue(&ctrl->ctrl);
 	if (ret)
 		goto out_disconnect_admin_queue;
@@ -2648,23 +2667,15 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl)
 	 * prior connection values
 	 */
 
-	ret = nvmf_reg_read64(&ctrl->ctrl, NVME_REG_CAP, &ctrl->ctrl.cap);
-	if (ret) {
-		dev_err(ctrl->ctrl.device,
-			"prop_get NVME_REG_CAP failed\n");
-		goto out_disconnect_admin_queue;
-	}
-
-	ctrl->ctrl.sqsize =
-		min_t(int, NVME_CAP_MQES(ctrl->ctrl.cap), ctrl->ctrl.sqsize);
-
-	ret = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
+	ret = nvme_enable_ctrl(&ctrl->ctrl);
 	if (ret)
 		goto out_disconnect_admin_queue;
 
 	ctrl->ctrl.max_hw_sectors =
 		(ctrl->lport->ops->max_sgl_segments - 1) << (PAGE_SHIFT - 9);
 
+	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
+
 	ret = nvme_init_identify(&ctrl->ctrl);
 	if (ret)
 		goto out_disconnect_admin_queue;
@@ -2774,6 +2785,7 @@ nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl)
 		nvme_stop_queues(&ctrl->ctrl);
 		blk_mq_tagset_busy_iter(&ctrl->tag_set,
 				nvme_fc_terminate_exchange, &ctrl->ctrl);
+		blk_mq_tagset_wait_completed_request(&ctrl->tag_set);
 	}
 
 	/*
@@ -2796,6 +2808,7 @@ nvme_fc_delete_association(struct nvme_fc_ctrl *ctrl)
 	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
 	blk_mq_tagset_busy_iter(&ctrl->admin_tag_set,
 				nvme_fc_terminate_exchange, &ctrl->ctrl);
+	blk_mq_tagset_wait_completed_request(&ctrl->admin_tag_set);
 
 	/* kill the aens as they are a separate path */
 	nvme_fc_abort_aen_ops(ctrl);
@@ -3109,10 +3122,16 @@ nvme_fc_init_ctrl(struct device *dev, struct nvmf_ctrl_options *opts,
 		goto out_free_queues;
 	ctrl->ctrl.admin_tagset = &ctrl->admin_tag_set;
 
+	ctrl->ctrl.fabrics_q = blk_mq_init_queue(&ctrl->admin_tag_set);
+	if (IS_ERR(ctrl->ctrl.fabrics_q)) {
+		ret = PTR_ERR(ctrl->ctrl.fabrics_q);
+		goto out_free_admin_tag_set;
+	}
+
 	ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
 	if (IS_ERR(ctrl->ctrl.admin_q)) {
 		ret = PTR_ERR(ctrl->ctrl.admin_q);
-		goto out_free_admin_tag_set;
+		goto out_cleanup_fabrics_q;
 	}
 
 	/*
@@ -3184,6 +3203,8 @@ fail_ctrl:
 
 out_cleanup_admin_q:
 	blk_cleanup_queue(ctrl->ctrl.admin_q);
+out_cleanup_fabrics_q:
+	blk_cleanup_queue(ctrl->ctrl.fabrics_q);
 out_free_admin_tag_set:
 	blk_mq_free_tag_set(&ctrl->admin_tag_set);
 out_free_queues:
diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index ba009d4c9dfa..ec46693f6b64 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -667,11 +667,14 @@ static struct request *nvme_nvm_alloc_request(struct request_queue *q,
 	return rq;
 }
 
-static int nvme_nvm_submit_io(struct nvm_dev *dev, struct nvm_rq *rqd)
+static int nvme_nvm_submit_io(struct nvm_dev *dev, struct nvm_rq *rqd,
+			      void *buf)
 {
+	struct nvm_geo *geo = &dev->geo;
 	struct request_queue *q = dev->q;
 	struct nvme_nvm_command *cmd;
 	struct request *rq;
+	int ret;
 
 	cmd = kzalloc(sizeof(struct nvme_nvm_command), GFP_KERNEL);
 	if (!cmd)
@@ -679,8 +682,15 @@ static int nvme_nvm_submit_io(struct nvm_dev *dev, struct nvm_rq *rqd)
 
 	rq = nvme_nvm_alloc_request(q, rqd, cmd);
 	if (IS_ERR(rq)) {
-		kfree(cmd);
-		return PTR_ERR(rq);
+		ret = PTR_ERR(rq);
+		goto err_free_cmd;
+	}
+
+	if (buf) {
+		ret = blk_rq_map_kern(q, rq, buf, geo->csecs * rqd->nr_ppas,
+				GFP_KERNEL);
+		if (ret)
+			goto err_free_cmd;
 	}
 
 	rq->end_io_data = rqd;
@@ -688,33 +698,9 @@ static int nvme_nvm_submit_io(struct nvm_dev *dev, struct nvm_rq *rqd)
 	blk_execute_rq_nowait(q, NULL, rq, 0, nvme_nvm_end_io);
 
 	return 0;
-}
-
-static int nvme_nvm_submit_io_sync(struct nvm_dev *dev, struct nvm_rq *rqd)
-{
-	struct request_queue *q = dev->q;
-	struct request *rq;
-	struct nvme_nvm_command cmd;
-	int ret = 0;
-
-	memset(&cmd, 0, sizeof(struct nvme_nvm_command));
-
-	rq = nvme_nvm_alloc_request(q, rqd, &cmd);
-	if (IS_ERR(rq))
-		return PTR_ERR(rq);
-
-	/* I/Os can fail and the error is signaled through rqd. Callers must
-	 * handle the error accordingly.
-	 */
-	blk_execute_rq(q, NULL, rq, 0);
-	if (nvme_req(rq)->flags & NVME_REQ_CANCELLED)
-		ret = -EINTR;
-
-	rqd->ppa_status = le64_to_cpu(nvme_req(rq)->result.u64);
-	rqd->error = nvme_req(rq)->status;
-
-	blk_mq_free_request(rq);
 
+err_free_cmd:
+	kfree(cmd);
 	return ret;
 }
 
@@ -754,7 +740,6 @@ static struct nvm_dev_ops nvme_nvm_dev_ops = {
 	.get_chk_meta		= nvme_nvm_get_chk_meta,
 
 	.submit_io		= nvme_nvm_submit_io,
-	.submit_io_sync		= nvme_nvm_submit_io_sync,
 
 	.create_dma_pool	= nvme_nvm_create_dma_pool,
 	.destroy_dma_pool	= nvme_nvm_destroy_dma_pool,
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index af831d3d15d0..30de7efef003 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -509,14 +509,16 @@ static int nvme_update_ana_state(struct nvme_ctrl *ctrl,
 
 	down_write(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list) {
-		if (ns->head->ns_id != le32_to_cpu(desc->nsids[n]))
+		unsigned nsid = le32_to_cpu(desc->nsids[n]);
+
+		if (ns->head->ns_id < nsid)
 			continue;
-		nvme_update_ns_ana_state(desc, ns);
+		if (ns->head->ns_id == nsid)
+			nvme_update_ns_ana_state(desc, ns);
 		if (++n == nr_nsids)
 			break;
 	}
 	up_write(&ctrl->namespaces_rwsem);
-	WARN_ON_ONCE(n < nr_nsids);
 	return 0;
 }
 
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 2d678fb968c7..b5013c101b35 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -16,6 +16,8 @@
 #include <linux/fault-inject.h>
 #include <linux/rcupdate.h>
 
+#include <trace/events/block.h>
+
 extern unsigned int nvme_io_timeout;
 #define NVME_IO_TIMEOUT	(nvme_io_timeout * HZ)
 
@@ -97,6 +99,21 @@ enum nvme_quirks {
 	 * Force simple suspend/resume path.
 	 */
 	NVME_QUIRK_SIMPLE_SUSPEND		= (1 << 10),
+
+	/*
+	 * Use only one interrupt vector for all queues
+	 */
+	NVME_QUIRK_SINGLE_VECTOR		= (1 << 11),
+
+	/*
+	 * Use non-standard 128 bytes SQEs.
+	 */
+	NVME_QUIRK_128_BYTES_SQES		= (1 << 12),
+
+	/*
+	 * Prevent tag overlap between queues
+	 */
+	NVME_QUIRK_SHARED_TAGS                  = (1 << 13),
 };
 
 /*
@@ -169,6 +186,7 @@ struct nvme_ctrl {
 	const struct nvme_ctrl_ops *ops;
 	struct request_queue *admin_q;
 	struct request_queue *connect_q;
+	struct request_queue *fabrics_q;
 	struct device *dev;
 	int instance;
 	int numa_node;
@@ -431,8 +449,8 @@ void nvme_complete_rq(struct request *req);
 bool nvme_cancel_request(struct request *req, void *data, bool reserved);
 bool nvme_change_ctrl_state(struct nvme_ctrl *ctrl,
 		enum nvme_ctrl_state new_state);
-int nvme_disable_ctrl(struct nvme_ctrl *ctrl, u64 cap);
-int nvme_enable_ctrl(struct nvme_ctrl *ctrl, u64 cap);
+int nvme_disable_ctrl(struct nvme_ctrl *ctrl);
+int nvme_enable_ctrl(struct nvme_ctrl *ctrl);
 int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl);
 int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 		const struct nvme_ctrl_ops *ops, unsigned long quirks);
@@ -520,6 +538,16 @@ static inline void nvme_mpath_check_last_path(struct nvme_ns *ns)
 		kblockd_schedule_work(&head->requeue_work);
 }
 
+static inline void nvme_trace_bio_complete(struct request *req,
+        blk_status_t status)
+{
+	struct nvme_ns *ns = req->q->queuedata;
+
+	if (req->cmd_flags & REQ_NVME_MPATH)
+		trace_block_bio_complete(ns->head->disk->queue,
+					 req->bio, status);
+}
+
 extern struct device_attribute dev_attr_ana_grpid;
 extern struct device_attribute dev_attr_ana_state;
 extern struct device_attribute subsys_attr_iopolicy;
@@ -567,6 +595,10 @@ static inline void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)
 static inline void nvme_mpath_check_last_path(struct nvme_ns *ns)
 {
 }
+static inline void nvme_trace_bio_complete(struct request *req,
+        blk_status_t status)
+{
+}
 static inline int nvme_mpath_init(struct nvme_ctrl *ctrl,
 		struct nvme_id_ctrl *id)
 {
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 732d5b63ec05..6b4d7b064b38 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -28,8 +28,8 @@
 #include "trace.h"
 #include "nvme.h"
 
-#define SQ_SIZE(depth)		(depth * sizeof(struct nvme_command))
-#define CQ_SIZE(depth)		(depth * sizeof(struct nvme_completion))
+#define SQ_SIZE(q)	((q)->q_depth << (q)->sqes)
+#define CQ_SIZE(q)	((q)->q_depth * sizeof(struct nvme_completion))
 
 #define SGES_PER_PAGE	(PAGE_SIZE / sizeof(struct nvme_sgl_desc))
 
@@ -100,6 +100,7 @@ struct nvme_dev {
 	unsigned io_queues[HCTX_MAX_TYPES];
 	unsigned int num_vecs;
 	int q_depth;
+	int io_sqes;
 	u32 db_stride;
 	void __iomem *bar;
 	unsigned long bar_mapped_size;
@@ -162,7 +163,7 @@ static inline struct nvme_dev *to_nvme_dev(struct nvme_ctrl *ctrl)
 struct nvme_queue {
 	struct nvme_dev *dev;
 	spinlock_t sq_lock;
-	struct nvme_command *sq_cmds;
+	void *sq_cmds;
 	 /* only used for poll queues: */
 	spinlock_t cq_poll_lock ____cacheline_aligned_in_smp;
 	volatile struct nvme_completion *cqes;
@@ -178,6 +179,7 @@ struct nvme_queue {
 	u16 last_cq_head;
 	u16 qid;
 	u8 cq_phase;
+	u8 sqes;
 	unsigned long flags;
 #define NVMEQ_ENABLED		0
 #define NVMEQ_SQ_CMB		1
@@ -488,7 +490,8 @@ static void nvme_submit_cmd(struct nvme_queue *nvmeq, struct nvme_command *cmd,
 			    bool write_sq)
 {
 	spin_lock(&nvmeq->sq_lock);
-	memcpy(&nvmeq->sq_cmds[nvmeq->sq_tail], cmd, sizeof(*cmd));
+	memcpy(nvmeq->sq_cmds + (nvmeq->sq_tail << nvmeq->sqes),
+	       cmd, sizeof(*cmd));
 	if (++nvmeq->sq_tail == nvmeq->q_depth)
 		nvmeq->sq_tail = 0;
 	nvme_write_sq_db(nvmeq, write_sq);
@@ -534,14 +537,13 @@ static inline bool nvme_pci_use_sgls(struct nvme_dev *dev, struct request *req)
 static void nvme_unmap_data(struct nvme_dev *dev, struct request *req)
 {
 	struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
-	enum dma_data_direction dma_dir = rq_data_dir(req) ?
-			DMA_TO_DEVICE : DMA_FROM_DEVICE;
 	const int last_prp = dev->ctrl.page_size / sizeof(__le64) - 1;
 	dma_addr_t dma_addr = iod->first_dma, next_dma_addr;
 	int i;
 
 	if (iod->dma_len) {
-		dma_unmap_page(dev->dev, dma_addr, iod->dma_len, dma_dir);
+		dma_unmap_page(dev->dev, dma_addr, iod->dma_len,
+			       rq_dma_dir(req));
 		return;
 	}
 
@@ -1344,16 +1346,16 @@ static enum blk_eh_timer_return nvme_timeout(struct request *req, bool reserved)
 
 static void nvme_free_queue(struct nvme_queue *nvmeq)
 {
-	dma_free_coherent(nvmeq->dev->dev, CQ_SIZE(nvmeq->q_depth),
+	dma_free_coherent(nvmeq->dev->dev, CQ_SIZE(nvmeq),
 				(void *)nvmeq->cqes, nvmeq->cq_dma_addr);
 	if (!nvmeq->sq_cmds)
 		return;
 
 	if (test_and_clear_bit(NVMEQ_SQ_CMB, &nvmeq->flags)) {
 		pci_free_p2pmem(to_pci_dev(nvmeq->dev->dev),
-				nvmeq->sq_cmds, SQ_SIZE(nvmeq->q_depth));
+				nvmeq->sq_cmds, SQ_SIZE(nvmeq));
 	} else {
-		dma_free_coherent(nvmeq->dev->dev, SQ_SIZE(nvmeq->q_depth),
+		dma_free_coherent(nvmeq->dev->dev, SQ_SIZE(nvmeq),
 				nvmeq->sq_cmds, nvmeq->sq_dma_addr);
 	}
 }
@@ -1403,7 +1405,7 @@ static void nvme_disable_admin_queue(struct nvme_dev *dev, bool shutdown)
 	if (shutdown)
 		nvme_shutdown_ctrl(&dev->ctrl);
 	else
-		nvme_disable_ctrl(&dev->ctrl, dev->ctrl.cap);
+		nvme_disable_ctrl(&dev->ctrl);
 
 	nvme_poll_irqdisable(nvmeq, -1);
 }
@@ -1433,12 +1435,12 @@ static int nvme_cmb_qdepth(struct nvme_dev *dev, int nr_io_queues,
 }
 
 static int nvme_alloc_sq_cmds(struct nvme_dev *dev, struct nvme_queue *nvmeq,
-				int qid, int depth)
+				int qid)
 {
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
 
 	if (qid && dev->cmb_use_sqes && (dev->cmbsz & NVME_CMBSZ_SQS)) {
-		nvmeq->sq_cmds = pci_alloc_p2pmem(pdev, SQ_SIZE(depth));
+		nvmeq->sq_cmds = pci_alloc_p2pmem(pdev, SQ_SIZE(nvmeq));
 		if (nvmeq->sq_cmds) {
 			nvmeq->sq_dma_addr = pci_p2pmem_virt_to_bus(pdev,
 							nvmeq->sq_cmds);
@@ -1447,11 +1449,11 @@ static int nvme_alloc_sq_cmds(struct nvme_dev *dev, struct nvme_queue *nvmeq,
 				return 0;
 			}
 
-			pci_free_p2pmem(pdev, nvmeq->sq_cmds, SQ_SIZE(depth));
+			pci_free_p2pmem(pdev, nvmeq->sq_cmds, SQ_SIZE(nvmeq));
 		}
 	}
 
-	nvmeq->sq_cmds = dma_alloc_coherent(dev->dev, SQ_SIZE(depth),
+	nvmeq->sq_cmds = dma_alloc_coherent(dev->dev, SQ_SIZE(nvmeq),
 				&nvmeq->sq_dma_addr, GFP_KERNEL);
 	if (!nvmeq->sq_cmds)
 		return -ENOMEM;
@@ -1465,12 +1467,14 @@ static int nvme_alloc_queue(struct nvme_dev *dev, int qid, int depth)
 	if (dev->ctrl.queue_count > qid)
 		return 0;
 
-	nvmeq->cqes = dma_alloc_coherent(dev->dev, CQ_SIZE(depth),
+	nvmeq->sqes = qid ? dev->io_sqes : NVME_ADM_SQES;
+	nvmeq->q_depth = depth;
+	nvmeq->cqes = dma_alloc_coherent(dev->dev, CQ_SIZE(nvmeq),
 					 &nvmeq->cq_dma_addr, GFP_KERNEL);
 	if (!nvmeq->cqes)
 		goto free_nvmeq;
 
-	if (nvme_alloc_sq_cmds(dev, nvmeq, qid, depth))
+	if (nvme_alloc_sq_cmds(dev, nvmeq, qid))
 		goto free_cqdma;
 
 	nvmeq->dev = dev;
@@ -1479,15 +1483,14 @@ static int nvme_alloc_queue(struct nvme_dev *dev, int qid, int depth)
 	nvmeq->cq_head = 0;
 	nvmeq->cq_phase = 1;
 	nvmeq->q_db = &dev->dbs[qid * 2 * dev->db_stride];
-	nvmeq->q_depth = depth;
 	nvmeq->qid = qid;
 	dev->ctrl.queue_count++;
 
 	return 0;
 
  free_cqdma:
-	dma_free_coherent(dev->dev, CQ_SIZE(depth), (void *)nvmeq->cqes,
-							nvmeq->cq_dma_addr);
+	dma_free_coherent(dev->dev, CQ_SIZE(nvmeq), (void *)nvmeq->cqes,
+			  nvmeq->cq_dma_addr);
  free_nvmeq:
 	return -ENOMEM;
 }
@@ -1515,7 +1518,7 @@ static void nvme_init_queue(struct nvme_queue *nvmeq, u16 qid)
 	nvmeq->cq_head = 0;
 	nvmeq->cq_phase = 1;
 	nvmeq->q_db = &dev->dbs[qid * 2 * dev->db_stride];
-	memset((void *)nvmeq->cqes, 0, CQ_SIZE(nvmeq->q_depth));
+	memset((void *)nvmeq->cqes, 0, CQ_SIZE(nvmeq));
 	nvme_dbbuf_init(dev, nvmeq, qid);
 	dev->online_queues++;
 	wmb(); /* ensure the first interrupt sees the initialization */
@@ -1552,7 +1555,6 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid, bool polled)
 	nvme_init_queue(nvmeq, qid);
 
 	if (!polled) {
-		nvmeq->cq_vector = vector;
 		result = queue_request_irq(nvmeq);
 		if (result < 0)
 			goto release_sq;
@@ -1679,7 +1681,7 @@ static int nvme_pci_configure_admin_queue(struct nvme_dev *dev)
 	    (readl(dev->bar + NVME_REG_CSTS) & NVME_CSTS_NSSRO))
 		writel(NVME_CSTS_NSSRO, dev->bar + NVME_REG_CSTS);
 
-	result = nvme_disable_ctrl(&dev->ctrl, dev->ctrl.cap);
+	result = nvme_disable_ctrl(&dev->ctrl);
 	if (result < 0)
 		return result;
 
@@ -1695,7 +1697,7 @@ static int nvme_pci_configure_admin_queue(struct nvme_dev *dev)
 	lo_hi_writeq(nvmeq->sq_dma_addr, dev->bar + NVME_REG_ASQ);
 	lo_hi_writeq(nvmeq->cq_dma_addr, dev->bar + NVME_REG_ACQ);
 
-	result = nvme_enable_ctrl(&dev->ctrl, dev->ctrl.cap);
+	result = nvme_enable_ctrl(&dev->ctrl);
 	if (result)
 		return result;
 
@@ -2077,6 +2079,13 @@ static int nvme_setup_irqs(struct nvme_dev *dev, unsigned int nr_io_queues)
 	dev->io_queues[HCTX_TYPE_DEFAULT] = 1;
 	dev->io_queues[HCTX_TYPE_READ] = 0;
 
+	/*
+	 * Some Apple controllers require all queues to use the
+	 * first vector.
+	 */
+	if (dev->ctrl.quirks & NVME_QUIRK_SINGLE_VECTOR)
+		irq_queues = 1;
+
 	return pci_alloc_irq_vectors_affinity(pdev, 1, irq_queues,
 			      PCI_IRQ_ALL_TYPES | PCI_IRQ_AFFINITY, &affd);
 }
@@ -2095,6 +2104,14 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
 	unsigned long size;
 
 	nr_io_queues = max_io_queues();
+
+	/*
+	 * If tags are shared with admin queue (Apple bug), then
+	 * make sure we only use one IO queue.
+	 */
+	if (dev->ctrl.quirks & NVME_QUIRK_SHARED_TAGS)
+		nr_io_queues = 1;
+
 	result = nvme_set_queue_count(&dev->ctrl, &nr_io_queues);
 	if (result < 0)
 		return result;
@@ -2265,6 +2282,14 @@ static int nvme_dev_add(struct nvme_dev *dev)
 		dev->tagset.flags = BLK_MQ_F_SHOULD_MERGE;
 		dev->tagset.driver_data = dev;
 
+		/*
+		 * Some Apple controllers requires tags to be unique
+		 * across admin and IO queue, so reserve the first 32
+		 * tags of the IO queue.
+		 */
+		if (dev->ctrl.quirks & NVME_QUIRK_SHARED_TAGS)
+			dev->tagset.reserved_tags = NVME_AQ_DEPTH;
+
 		ret = blk_mq_alloc_tag_set(&dev->tagset);
 		if (ret) {
 			dev_warn(dev->ctrl.device,
@@ -2314,10 +2339,21 @@ static int nvme_pci_enable(struct nvme_dev *dev)
 
 	dev->q_depth = min_t(int, NVME_CAP_MQES(dev->ctrl.cap) + 1,
 				io_queue_depth);
+	dev->ctrl.sqsize = dev->q_depth - 1; /* 0's based queue depth */
 	dev->db_stride = 1 << NVME_CAP_STRIDE(dev->ctrl.cap);
 	dev->dbs = dev->bar + 4096;
 
 	/*
+	 * Some Apple controllers require a non-standard SQE size.
+	 * Interestingly they also seem to ignore the CC:IOSQES register
+	 * so we don't bother updating it here.
+	 */
+	if (dev->ctrl.quirks & NVME_QUIRK_128_BYTES_SQES)
+		dev->io_sqes = 7;
+	else
+		dev->io_sqes = NVME_NVM_IOSQES;
+
+	/*
 	 * Temporary fix for the Apple controller found in the MacBook8,1 and
 	 * some MacBook7,1 to avoid controller resets and data loss.
 	 */
@@ -2334,6 +2370,18 @@ static int nvme_pci_enable(struct nvme_dev *dev)
                         "set queue depth=%u\n", dev->q_depth);
 	}
 
+	/*
+	 * Controllers with the shared tags quirk need the IO queue to be
+	 * big enough so that we get 32 tags for the admin queue
+	 */
+	if ((dev->ctrl.quirks & NVME_QUIRK_SHARED_TAGS) &&
+	    (dev->q_depth < (NVME_AQ_DEPTH + 2))) {
+		dev->q_depth = NVME_AQ_DEPTH + 2;
+		dev_warn(dev->ctrl.device, "IO queue depth clamped to %d\n",
+			 dev->q_depth);
+	}
+
+
 	nvme_map_cmb(dev);
 
 	pci_enable_pcie_error_reporting(pdev);
@@ -2401,6 +2449,8 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown)
 
 	blk_mq_tagset_busy_iter(&dev->tagset, nvme_cancel_request, &dev->ctrl);
 	blk_mq_tagset_busy_iter(&dev->admin_tagset, nvme_cancel_request, &dev->ctrl);
+	blk_mq_tagset_wait_completed_request(&dev->tagset);
+	blk_mq_tagset_wait_completed_request(&dev->admin_tagset);
 
 	/*
 	 * The driver will not be starting up queues again if shutting down so
@@ -3041,6 +3091,10 @@ static const struct pci_device_id nvme_id_table[] = {
 	{ PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2003) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2005),
+		.driver_data = NVME_QUIRK_SINGLE_VECTOR |
+				NVME_QUIRK_128_BYTES_SQES |
+				NVME_QUIRK_SHARED_TAGS },
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, nvme_id_table);
diff --git a/drivers/nvme/host/rdma.c b/drivers/nvme/host/rdma.c
index 1a6449bc547b..dfa07bb9dfeb 100644
--- a/drivers/nvme/host/rdma.c
+++ b/drivers/nvme/host/rdma.c
@@ -757,6 +757,7 @@ static void nvme_rdma_destroy_admin_queue(struct nvme_rdma_ctrl *ctrl,
 {
 	if (remove) {
 		blk_cleanup_queue(ctrl->ctrl.admin_q);
+		blk_cleanup_queue(ctrl->ctrl.fabrics_q);
 		blk_mq_free_tag_set(ctrl->ctrl.admin_tagset);
 	}
 	if (ctrl->async_event_sqe.data) {
@@ -798,10 +799,16 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 			goto out_free_async_qe;
 		}
 
+		ctrl->ctrl.fabrics_q = blk_mq_init_queue(&ctrl->admin_tag_set);
+		if (IS_ERR(ctrl->ctrl.fabrics_q)) {
+			error = PTR_ERR(ctrl->ctrl.fabrics_q);
+			goto out_free_tagset;
+		}
+
 		ctrl->ctrl.admin_q = blk_mq_init_queue(&ctrl->admin_tag_set);
 		if (IS_ERR(ctrl->ctrl.admin_q)) {
 			error = PTR_ERR(ctrl->ctrl.admin_q);
-			goto out_free_tagset;
+			goto out_cleanup_fabrics_q;
 		}
 	}
 
@@ -809,24 +816,15 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 	if (error)
 		goto out_cleanup_queue;
 
-	error = ctrl->ctrl.ops->reg_read64(&ctrl->ctrl, NVME_REG_CAP,
-			&ctrl->ctrl.cap);
-	if (error) {
-		dev_err(ctrl->ctrl.device,
-			"prop_get NVME_REG_CAP failed\n");
-		goto out_stop_queue;
-	}
-
-	ctrl->ctrl.sqsize =
-		min_t(int, NVME_CAP_MQES(ctrl->ctrl.cap), ctrl->ctrl.sqsize);
-
-	error = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
+	error = nvme_enable_ctrl(&ctrl->ctrl);
 	if (error)
 		goto out_stop_queue;
 
 	ctrl->ctrl.max_hw_sectors =
 		(ctrl->max_fr_pages - 1) << (ilog2(SZ_4K) - 9);
 
+	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
+
 	error = nvme_init_identify(&ctrl->ctrl);
 	if (error)
 		goto out_stop_queue;
@@ -838,6 +836,9 @@ out_stop_queue:
 out_cleanup_queue:
 	if (new)
 		blk_cleanup_queue(ctrl->ctrl.admin_q);
+out_cleanup_fabrics_q:
+	if (new)
+		blk_cleanup_queue(ctrl->ctrl.fabrics_q);
 out_free_tagset:
 	if (new)
 		blk_mq_free_tag_set(ctrl->ctrl.admin_tagset);
@@ -907,10 +908,13 @@ static void nvme_rdma_teardown_admin_queue(struct nvme_rdma_ctrl *ctrl,
 {
 	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
 	nvme_rdma_stop_queue(&ctrl->queues[0]);
-	if (ctrl->ctrl.admin_tagset)
+	if (ctrl->ctrl.admin_tagset) {
 		blk_mq_tagset_busy_iter(ctrl->ctrl.admin_tagset,
 			nvme_cancel_request, &ctrl->ctrl);
-	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
+		blk_mq_tagset_wait_completed_request(ctrl->ctrl.admin_tagset);
+	}
+	if (remove)
+		blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 	nvme_rdma_destroy_admin_queue(ctrl, remove);
 }
 
@@ -920,9 +924,11 @@ static void nvme_rdma_teardown_io_queues(struct nvme_rdma_ctrl *ctrl,
 	if (ctrl->ctrl.queue_count > 1) {
 		nvme_stop_queues(&ctrl->ctrl);
 		nvme_rdma_stop_io_queues(ctrl);
-		if (ctrl->ctrl.tagset)
+		if (ctrl->ctrl.tagset) {
 			blk_mq_tagset_busy_iter(ctrl->ctrl.tagset,
 				nvme_cancel_request, &ctrl->ctrl);
+			blk_mq_tagset_wait_completed_request(ctrl->ctrl.tagset);
+		}
 		if (remove)
 			nvme_start_queues(&ctrl->ctrl);
 		nvme_rdma_destroy_io_queues(ctrl, remove);
@@ -1059,6 +1065,7 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 	nvme_rdma_teardown_io_queues(ctrl, false);
 	nvme_start_queues(&ctrl->ctrl);
 	nvme_rdma_teardown_admin_queue(ctrl, false);
+	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 
 	if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING)) {
 		/* state change failure is ok if we're in DELETING state */
@@ -1145,9 +1152,7 @@ static void nvme_rdma_unmap_data(struct nvme_rdma_queue *queue,
 		req->mr = NULL;
 	}
 
-	ib_dma_unmap_sg(ibdev, req->sg_table.sgl,
-			req->nents, rq_data_dir(rq) ==
-				    WRITE ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
+	ib_dma_unmap_sg(ibdev, req->sg_table.sgl, req->nents, rq_dma_dir(rq));
 
 	nvme_cleanup_cmd(rq);
 	sg_free_table_chained(&req->sg_table, SG_CHUNK_SIZE);
@@ -1273,7 +1278,7 @@ static int nvme_rdma_map_data(struct nvme_rdma_queue *queue,
 	req->nents = blk_rq_map_sg(rq->q, rq, req->sg_table.sgl);
 
 	count = ib_dma_map_sg(ibdev, req->sg_table.sgl, req->nents,
-		    rq_data_dir(rq) == WRITE ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
+			      rq_dma_dir(rq));
 	if (unlikely(count <= 0)) {
 		ret = -EIO;
 		goto out_free_table;
@@ -1302,9 +1307,7 @@ out:
 	return 0;
 
 out_unmap_sg:
-	ib_dma_unmap_sg(ibdev, req->sg_table.sgl,
-			req->nents, rq_data_dir(rq) ==
-			WRITE ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
+	ib_dma_unmap_sg(ibdev, req->sg_table.sgl, req->nents, rq_dma_dir(rq));
 out_free_table:
 	sg_free_table_chained(&req->sg_table, SG_CHUNK_SIZE);
 	return ret;
@@ -1547,16 +1550,18 @@ static int nvme_rdma_conn_rejected(struct nvme_rdma_queue *queue,
 
 static int nvme_rdma_addr_resolved(struct nvme_rdma_queue *queue)
 {
+	struct nvme_ctrl *ctrl = &queue->ctrl->ctrl;
 	int ret;
 
 	ret = nvme_rdma_create_queue_ib(queue);
 	if (ret)
 		return ret;
 
+	if (ctrl->opts->tos >= 0)
+		rdma_set_service_type(queue->cm_id, ctrl->opts->tos);
 	ret = rdma_resolve_route(queue->cm_id, NVME_RDMA_CONNECT_TIMEOUT_MS);
 	if (ret) {
-		dev_err(queue->ctrl->ctrl.device,
-			"rdma_resolve_route failed (%d).\n",
+		dev_err(ctrl->device, "rdma_resolve_route failed (%d).\n",
 			queue->cm_error);
 		goto out_destroy_queue;
 	}
@@ -1869,10 +1874,11 @@ static void nvme_rdma_shutdown_ctrl(struct nvme_rdma_ctrl *ctrl, bool shutdown)
 	cancel_delayed_work_sync(&ctrl->reconnect_work);
 
 	nvme_rdma_teardown_io_queues(ctrl, shutdown);
+	blk_mq_quiesce_queue(ctrl->ctrl.admin_q);
 	if (shutdown)
 		nvme_shutdown_ctrl(&ctrl->ctrl);
 	else
-		nvme_disable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
+		nvme_disable_ctrl(&ctrl->ctrl);
 	nvme_rdma_teardown_admin_queue(ctrl, shutdown);
 }
 
@@ -2051,7 +2057,8 @@ static struct nvmf_transport_ops nvme_rdma_transport = {
 	.required_opts	= NVMF_OPT_TRADDR,
 	.allowed_opts	= NVMF_OPT_TRSVCID | NVMF_OPT_RECONNECT_DELAY |
 			  NVMF_OPT_HOST_TRADDR | NVMF_OPT_CTRL_LOSS_TMO |
-			  NVMF_OPT_NR_WRITE_QUEUES | NVMF_OPT_NR_POLL_QUEUES,
+			  NVMF_OPT_NR_WRITE_QUEUES | NVMF_OPT_NR_POLL_QUEUES |
+			  NVMF_OPT_TOS,
 	.create_ctrl	= nvme_rdma_create_ctrl,
 };
 
diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c
index 606b13d35d16..4ffd5957637a 100644
--- a/drivers/nvme/host/tcp.c
+++ b/drivers/nvme/host/tcp.c
@@ -13,6 +13,7 @@
 #include <net/tcp.h>
 #include <linux/blk-mq.h>
 #include <crypto/hash.h>
+#include <net/busy_poll.h>
 
 #include "nvme.h"
 #include "fabrics.h"
@@ -72,6 +73,7 @@ struct nvme_tcp_queue {
 	int			pdu_offset;
 	size_t			data_remaining;
 	size_t			ddgst_remaining;
+	unsigned int		nr_cqe;
 
 	/* send state */
 	struct nvme_tcp_request *request;
@@ -438,6 +440,7 @@ static int nvme_tcp_process_nvme_cqe(struct nvme_tcp_queue *queue,
 	}
 
 	nvme_end_request(rq, cqe->status, cqe->result);
+	queue->nr_cqe++;
 
 	return 0;
 }
@@ -608,23 +611,18 @@ static int nvme_tcp_recv_pdu(struct nvme_tcp_queue *queue, struct sk_buff *skb,
 
 	switch (hdr->type) {
 	case nvme_tcp_c2h_data:
-		ret = nvme_tcp_handle_c2h_data(queue, (void *)queue->pdu);
-		break;
+		return nvme_tcp_handle_c2h_data(queue, (void *)queue->pdu);
 	case nvme_tcp_rsp:
 		nvme_tcp_init_recv_ctx(queue);
-		ret = nvme_tcp_handle_comp(queue, (void *)queue->pdu);
-		break;
+		return nvme_tcp_handle_comp(queue, (void *)queue->pdu);
 	case nvme_tcp_r2t:
 		nvme_tcp_init_recv_ctx(queue);
-		ret = nvme_tcp_handle_r2t(queue, (void *)queue->pdu);
-		break;
+		return nvme_tcp_handle_r2t(queue, (void *)queue->pdu);
 	default:
 		dev_err(queue->ctrl->ctrl.device,
 			"unsupported pdu type (%d)\n", hdr->type);
 		return -EINVAL;
 	}
-
-	return ret;
 }
 
 static inline void nvme_tcp_end_request(struct request *rq, u16 status)
@@ -701,8 +699,10 @@ static int nvme_tcp_recv_data(struct nvme_tcp_queue *queue, struct sk_buff *skb,
 			nvme_tcp_ddgst_final(queue->rcv_hash, &queue->exp_ddgst);
 			queue->ddgst_remaining = NVME_TCP_DIGEST_LENGTH;
 		} else {
-			if (pdu->hdr.flags & NVME_TCP_F_DATA_SUCCESS)
+			if (pdu->hdr.flags & NVME_TCP_F_DATA_SUCCESS) {
 				nvme_tcp_end_request(rq, NVME_SC_SUCCESS);
+				queue->nr_cqe++;
+			}
 			nvme_tcp_init_recv_ctx(queue);
 		}
 	}
@@ -742,6 +742,7 @@ static int nvme_tcp_recv_ddgst(struct nvme_tcp_queue *queue,
 						pdu->command_id);
 
 		nvme_tcp_end_request(rq, NVME_SC_SUCCESS);
+		queue->nr_cqe++;
 	}
 
 	nvme_tcp_init_recv_ctx(queue);
@@ -841,7 +842,7 @@ static inline void nvme_tcp_done_send_req(struct nvme_tcp_queue *queue)
 
 static void nvme_tcp_fail_request(struct nvme_tcp_request *req)
 {
-	nvme_tcp_end_request(blk_mq_rq_from_pdu(req), NVME_SC_DATA_XFER_ERROR);
+	nvme_tcp_end_request(blk_mq_rq_from_pdu(req), NVME_SC_HOST_PATH_ERROR);
 }
 
 static int nvme_tcp_try_send_data(struct nvme_tcp_request *req)
@@ -1023,14 +1024,16 @@ done:
 
 static int nvme_tcp_try_recv(struct nvme_tcp_queue *queue)
 {
-	struct sock *sk = queue->sock->sk;
+	struct socket *sock = queue->sock;
+	struct sock *sk = sock->sk;
 	read_descriptor_t rd_desc;
 	int consumed;
 
 	rd_desc.arg.data = queue;
 	rd_desc.count = 1;
 	lock_sock(sk);
-	consumed = tcp_read_sock(sk, &rd_desc, nvme_tcp_recv_skb);
+	queue->nr_cqe = 0;
+	consumed = sock->ops->read_sock(sk, &rd_desc, nvme_tcp_recv_skb);
 	release_sock(sk);
 	return consumed;
 }
@@ -1255,7 +1258,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	queue->queue_size = queue_size;
 
 	if (qid > 0)
-		queue->cmnd_capsule_len = ctrl->ctrl.ioccsz * 16;
+		queue->cmnd_capsule_len = nctrl->ioccsz * 16;
 	else
 		queue->cmnd_capsule_len = sizeof(struct nvme_command) +
 						NVME_TCP_ADMIN_CCSZ;
@@ -1263,7 +1266,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	ret = sock_create(ctrl->addr.ss_family, SOCK_STREAM,
 			IPPROTO_TCP, &queue->sock);
 	if (ret) {
-		dev_err(ctrl->ctrl.device,
+		dev_err(nctrl->device,
 			"failed to create socket: %d\n", ret);
 		return ret;
 	}
@@ -1273,7 +1276,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	ret = kernel_setsockopt(queue->sock, IPPROTO_TCP, TCP_SYNCNT,
 			(char *)&opt, sizeof(opt));
 	if (ret) {
-		dev_err(ctrl->ctrl.device,
+		dev_err(nctrl->device,
 			"failed to set TCP_SYNCNT sock opt %d\n", ret);
 		goto err_sock;
 	}
@@ -1283,7 +1286,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	ret = kernel_setsockopt(queue->sock, IPPROTO_TCP,
 			TCP_NODELAY, (char *)&opt, sizeof(opt));
 	if (ret) {
-		dev_err(ctrl->ctrl.device,
+		dev_err(nctrl->device,
 			"failed to set TCP_NODELAY sock opt %d\n", ret);
 		goto err_sock;
 	}
@@ -1296,11 +1299,23 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	ret = kernel_setsockopt(queue->sock, SOL_SOCKET, SO_LINGER,
 			(char *)&sol, sizeof(sol));
 	if (ret) {
-		dev_err(ctrl->ctrl.device,
+		dev_err(nctrl->device,
 			"failed to set SO_LINGER sock opt %d\n", ret);
 		goto err_sock;
 	}
 
+	/* Set socket type of service */
+	if (nctrl->opts->tos >= 0) {
+		opt = nctrl->opts->tos;
+		ret = kernel_setsockopt(queue->sock, SOL_IP, IP_TOS,
+				(char *)&opt, sizeof(opt));
+		if (ret) {
+			dev_err(nctrl->device,
+				"failed to set IP_TOS sock opt %d\n", ret);
+			goto err_sock;
+		}
+	}
+
 	queue->sock->sk->sk_allocation = GFP_ATOMIC;
 	if (!qid)
 		n = 0;
@@ -1314,11 +1329,11 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	queue->pdu_offset = 0;
 	sk_set_memalloc(queue->sock->sk);
 
-	if (ctrl->ctrl.opts->mask & NVMF_OPT_HOST_TRADDR) {
+	if (nctrl->opts->mask & NVMF_OPT_HOST_TRADDR) {
 		ret = kernel_bind(queue->sock, (struct sockaddr *)&ctrl->src_addr,
 			sizeof(ctrl->src_addr));
 		if (ret) {
-			dev_err(ctrl->ctrl.device,
+			dev_err(nctrl->device,
 				"failed to bind queue %d socket %d\n",
 				qid, ret);
 			goto err_sock;
@@ -1330,7 +1345,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	if (queue->hdr_digest || queue->data_digest) {
 		ret = nvme_tcp_alloc_crypto(queue);
 		if (ret) {
-			dev_err(ctrl->ctrl.device,
+			dev_err(nctrl->device,
 				"failed to allocate queue %d crypto\n", qid);
 			goto err_sock;
 		}
@@ -1344,13 +1359,13 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 		goto err_crypto;
 	}
 
-	dev_dbg(ctrl->ctrl.device, "connecting queue %d\n",
+	dev_dbg(nctrl->device, "connecting queue %d\n",
 			nvme_tcp_queue_id(queue));
 
 	ret = kernel_connect(queue->sock, (struct sockaddr *)&ctrl->addr,
 		sizeof(ctrl->addr), 0);
 	if (ret) {
-		dev_err(ctrl->ctrl.device,
+		dev_err(nctrl->device,
 			"failed to connect socket: %d\n", ret);
 		goto err_rcv_pdu;
 	}
@@ -1371,6 +1386,7 @@ static int nvme_tcp_alloc_queue(struct nvme_ctrl *nctrl,
 	queue->sock->sk->sk_data_ready = nvme_tcp_data_ready;
 	queue->sock->sk->sk_state_change = nvme_tcp_state_change;
 	queue->sock->sk->sk_write_space = nvme_tcp_write_space;
+	queue->sock->sk->sk_ll_usec = 1;
 	write_unlock_bh(&queue->sock->sk->sk_callback_lock);
 
 	return 0;
@@ -1469,7 +1485,7 @@ static struct blk_mq_tag_set *nvme_tcp_alloc_tagset(struct nvme_ctrl *nctrl,
 		set->driver_data = ctrl;
 		set->nr_hw_queues = nctrl->queue_count - 1;
 		set->timeout = NVME_IO_TIMEOUT;
-		set->nr_maps = 2 /* default + read */;
+		set->nr_maps = nctrl->opts->nr_poll_queues ? HCTX_MAX_TYPES : 2;
 	}
 
 	ret = blk_mq_alloc_tag_set(set);
@@ -1568,6 +1584,7 @@ static unsigned int nvme_tcp_nr_io_queues(struct nvme_ctrl *ctrl)
 
 	nr_io_queues = min(ctrl->opts->nr_io_queues, num_online_cpus());
 	nr_io_queues += min(ctrl->opts->nr_write_queues, num_online_cpus());
+	nr_io_queues += min(ctrl->opts->nr_poll_queues, num_online_cpus());
 
 	return nr_io_queues;
 }
@@ -1599,6 +1616,12 @@ static void nvme_tcp_set_io_queues(struct nvme_ctrl *nctrl,
 			min(opts->nr_io_queues, nr_io_queues);
 		nr_io_queues -= ctrl->io_queues[HCTX_TYPE_DEFAULT];
 	}
+
+	if (opts->nr_poll_queues && nr_io_queues) {
+		/* map dedicated poll queues only if we have queues left */
+		ctrl->io_queues[HCTX_TYPE_POLL] =
+			min(opts->nr_poll_queues, nr_io_queues);
+	}
 }
 
 static int nvme_tcp_alloc_io_queues(struct nvme_ctrl *ctrl)
@@ -1680,6 +1703,7 @@ static void nvme_tcp_destroy_admin_queue(struct nvme_ctrl *ctrl, bool remove)
 	nvme_tcp_stop_queue(ctrl, 0);
 	if (remove) {
 		blk_cleanup_queue(ctrl->admin_q);
+		blk_cleanup_queue(ctrl->fabrics_q);
 		blk_mq_free_tag_set(ctrl->admin_tagset);
 	}
 	nvme_tcp_free_admin_queue(ctrl);
@@ -1700,10 +1724,16 @@ static int nvme_tcp_configure_admin_queue(struct nvme_ctrl *ctrl, bool new)
 			goto out_free_queue;
 		}
 
+		ctrl->fabrics_q = blk_mq_init_queue(ctrl->admin_tagset);
+		if (IS_ERR(ctrl->fabrics_q)) {
+			error = PTR_ERR(ctrl->fabrics_q);
+			goto out_free_tagset;
+		}
+
 		ctrl->admin_q = blk_mq_init_queue(ctrl->admin_tagset);
 		if (IS_ERR(ctrl->admin_q)) {
 			error = PTR_ERR(ctrl->admin_q);
-			goto out_free_tagset;
+			goto out_cleanup_fabrics_q;
 		}
 	}
 
@@ -1711,19 +1741,12 @@ static int nvme_tcp_configure_admin_queue(struct nvme_ctrl *ctrl, bool new)
 	if (error)
 		goto out_cleanup_queue;
 
-	error = ctrl->ops->reg_read64(ctrl, NVME_REG_CAP, &ctrl->cap);
-	if (error) {
-		dev_err(ctrl->device,
-			"prop_get NVME_REG_CAP failed\n");
-		goto out_stop_queue;
-	}
-
-	ctrl->sqsize = min_t(int, NVME_CAP_MQES(ctrl->cap), ctrl->sqsize);
-
-	error = nvme_enable_ctrl(ctrl, ctrl->cap);
+	error = nvme_enable_ctrl(ctrl);
 	if (error)
 		goto out_stop_queue;
 
+	blk_mq_unquiesce_queue(ctrl->admin_q);
+
 	error = nvme_init_identify(ctrl);
 	if (error)
 		goto out_stop_queue;
@@ -1735,6 +1758,9 @@ out_stop_queue:
 out_cleanup_queue:
 	if (new)
 		blk_cleanup_queue(ctrl->admin_q);
+out_cleanup_fabrics_q:
+	if (new)
+		blk_cleanup_queue(ctrl->fabrics_q);
 out_free_tagset:
 	if (new)
 		blk_mq_free_tag_set(ctrl->admin_tagset);
@@ -1748,10 +1774,13 @@ static void nvme_tcp_teardown_admin_queue(struct nvme_ctrl *ctrl,
 {
 	blk_mq_quiesce_queue(ctrl->admin_q);
 	nvme_tcp_stop_queue(ctrl, 0);
-	if (ctrl->admin_tagset)
+	if (ctrl->admin_tagset) {
 		blk_mq_tagset_busy_iter(ctrl->admin_tagset,
 			nvme_cancel_request, ctrl);
-	blk_mq_unquiesce_queue(ctrl->admin_q);
+		blk_mq_tagset_wait_completed_request(ctrl->admin_tagset);
+	}
+	if (remove)
+		blk_mq_unquiesce_queue(ctrl->admin_q);
 	nvme_tcp_destroy_admin_queue(ctrl, remove);
 }
 
@@ -1762,9 +1791,11 @@ static void nvme_tcp_teardown_io_queues(struct nvme_ctrl *ctrl,
 		return;
 	nvme_stop_queues(ctrl);
 	nvme_tcp_stop_io_queues(ctrl);
-	if (ctrl->tagset)
+	if (ctrl->tagset) {
 		blk_mq_tagset_busy_iter(ctrl->tagset,
 			nvme_cancel_request, ctrl);
+		blk_mq_tagset_wait_completed_request(ctrl->tagset);
+	}
 	if (remove)
 		nvme_start_queues(ctrl);
 	nvme_tcp_destroy_io_queues(ctrl, remove);
@@ -1793,7 +1824,7 @@ static void nvme_tcp_reconnect_or_remove(struct nvme_ctrl *ctrl)
 static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new)
 {
 	struct nvmf_ctrl_options *opts = ctrl->opts;
-	int ret = -EINVAL;
+	int ret;
 
 	ret = nvme_tcp_configure_admin_queue(ctrl, new);
 	if (ret)
@@ -1876,6 +1907,7 @@ static void nvme_tcp_error_recovery_work(struct work_struct *work)
 	/* unquiesce to fail fast pending requests */
 	nvme_start_queues(ctrl);
 	nvme_tcp_teardown_admin_queue(ctrl, false);
+	blk_mq_unquiesce_queue(ctrl->admin_q);
 
 	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING)) {
 		/* state change failure is ok if we're in DELETING state */
@@ -1892,10 +1924,11 @@ static void nvme_tcp_teardown_ctrl(struct nvme_ctrl *ctrl, bool shutdown)
 	cancel_delayed_work_sync(&to_tcp_ctrl(ctrl)->connect_work);
 
 	nvme_tcp_teardown_io_queues(ctrl, shutdown);
+	blk_mq_quiesce_queue(ctrl->admin_q);
 	if (shutdown)
 		nvme_shutdown_ctrl(ctrl);
 	else
-		nvme_disable_ctrl(ctrl, ctrl->cap);
+		nvme_disable_ctrl(ctrl);
 	nvme_tcp_teardown_admin_queue(ctrl, shutdown);
 }
 
@@ -2151,14 +2184,36 @@ static int nvme_tcp_map_queues(struct blk_mq_tag_set *set)
 	blk_mq_map_queues(&set->map[HCTX_TYPE_DEFAULT]);
 	blk_mq_map_queues(&set->map[HCTX_TYPE_READ]);
 
+	if (opts->nr_poll_queues && ctrl->io_queues[HCTX_TYPE_POLL]) {
+		/* map dedicated poll queues only if we have queues left */
+		set->map[HCTX_TYPE_POLL].nr_queues =
+				ctrl->io_queues[HCTX_TYPE_POLL];
+		set->map[HCTX_TYPE_POLL].queue_offset =
+			ctrl->io_queues[HCTX_TYPE_DEFAULT] +
+			ctrl->io_queues[HCTX_TYPE_READ];
+		blk_mq_map_queues(&set->map[HCTX_TYPE_POLL]);
+	}
+
 	dev_info(ctrl->ctrl.device,
-		"mapped %d/%d default/read queues.\n",
+		"mapped %d/%d/%d default/read/poll queues.\n",
 		ctrl->io_queues[HCTX_TYPE_DEFAULT],
-		ctrl->io_queues[HCTX_TYPE_READ]);
+		ctrl->io_queues[HCTX_TYPE_READ],
+		ctrl->io_queues[HCTX_TYPE_POLL]);
 
 	return 0;
 }
 
+static int nvme_tcp_poll(struct blk_mq_hw_ctx *hctx)
+{
+	struct nvme_tcp_queue *queue = hctx->driver_data;
+	struct sock *sk = queue->sock->sk;
+
+	if (sk_can_busy_loop(sk) && skb_queue_empty(&sk->sk_receive_queue))
+		sk_busy_loop(sk, true);
+	nvme_tcp_try_recv(queue);
+	return queue->nr_cqe;
+}
+
 static struct blk_mq_ops nvme_tcp_mq_ops = {
 	.queue_rq	= nvme_tcp_queue_rq,
 	.complete	= nvme_complete_rq,
@@ -2167,6 +2222,7 @@ static struct blk_mq_ops nvme_tcp_mq_ops = {
 	.init_hctx	= nvme_tcp_init_hctx,
 	.timeout	= nvme_tcp_timeout,
 	.map_queues	= nvme_tcp_map_queues,
+	.poll		= nvme_tcp_poll,
 };
 
 static struct blk_mq_ops nvme_tcp_admin_mq_ops = {
@@ -2220,7 +2276,8 @@ static struct nvme_ctrl *nvme_tcp_create_ctrl(struct device *dev,
 
 	INIT_LIST_HEAD(&ctrl->list);
 	ctrl->ctrl.opts = opts;
-	ctrl->ctrl.queue_count = opts->nr_io_queues + opts->nr_write_queues + 1;
+	ctrl->ctrl.queue_count = opts->nr_io_queues + opts->nr_write_queues +
+				opts->nr_poll_queues + 1;
 	ctrl->ctrl.sqsize = opts->queue_size - 1;
 	ctrl->ctrl.kato = opts->kato;
 
@@ -2314,7 +2371,8 @@ static struct nvmf_transport_ops nvme_tcp_transport = {
 	.allowed_opts	= NVMF_OPT_TRSVCID | NVMF_OPT_RECONNECT_DELAY |
 			  NVMF_OPT_HOST_TRADDR | NVMF_OPT_CTRL_LOSS_TMO |
 			  NVMF_OPT_HDR_DIGEST | NVMF_OPT_DATA_DIGEST |
-			  NVMF_OPT_NR_WRITE_QUEUES,
+			  NVMF_OPT_NR_WRITE_QUEUES | NVMF_OPT_NR_POLL_QUEUES |
+			  NVMF_OPT_TOS,
 	.create_ctrl	= nvme_tcp_create_ctrl,
 };
 
diff --git a/drivers/nvme/host/trace.c b/drivers/nvme/host/trace.c
index 9778eb0406b3..5c3cb6928f3c 100644
--- a/drivers/nvme/host/trace.c
+++ b/drivers/nvme/host/trace.c
@@ -86,6 +86,22 @@ static const char *nvme_trace_admin_get_features(struct trace_seq *p,
 	return ret;
 }
 
+static const char *nvme_trace_get_lba_status(struct trace_seq *p,
+					     u8 *cdw10)
+{
+	const char *ret = trace_seq_buffer_ptr(p);
+	u64 slba = get_unaligned_le64(cdw10);
+	u32 mndw = get_unaligned_le32(cdw10 + 8);
+	u16 rl = get_unaligned_le16(cdw10 + 12);
+	u8 atype = cdw10[15];
+
+	trace_seq_printf(p, "slba=0x%llx, mndw=0x%x, rl=0x%x, atype=%u",
+			slba, mndw, rl, atype);
+	trace_seq_putc(p, 0);
+
+	return ret;
+}
+
 static const char *nvme_trace_read_write(struct trace_seq *p, u8 *cdw10)
 {
 	const char *ret = trace_seq_buffer_ptr(p);
@@ -141,6 +157,8 @@ const char *nvme_trace_parse_admin_cmd(struct trace_seq *p,
 		return nvme_trace_admin_identify(p, cdw10);
 	case nvme_admin_get_features:
 		return nvme_trace_admin_get_features(p, cdw10);
+	case nvme_admin_get_lba_status:
+		return nvme_trace_get_lba_status(p, cdw10);
 	default:
 		return nvme_trace_common(p, cdw10);
 	}