Merge branch 'nvme-5.9' of git://git.infradead.org/nvme into for-5.9/drivers

Pull NVMe updates from Christoph.

* 'nvme-5.9' of git://git.infradead.org/nvme: (30 commits)
  nvme-loop: remove extra variable in create ctrl
  nvme-loop: set ctrl state connecting after init
  nvme-multipath: do not fall back to __nvme_find_path() for non-optimized paths
  nvme-multipath: fix logic for non-optimized paths
  nvme-rdma: fix controller reset hang during traffic
  nvme-tcp: fix controller reset hang during traffic
  nvmet: introduce the passthru Kconfig option
  nvmet: introduce the passthru configfs interface
  nvmet: Add passthru enable/disable helpers
  nvmet: add passthru code to process commands
  nvme: export nvme_find_get_ns() and nvme_put_ns()
  nvme: introduce nvme_ctrl_get_by_path()
  nvme: introduce nvme_execute_passthru_rq to call nvme_passthru_[start|end]()
  nvme: create helper function to obtain command effects
  nvme: clear any SGL flags in passthru commands
  nvmet-fc: remove redundant del_work_active flag
  nvmet-fc: check successful reference in nvmet_fc_find_target_assoc
  nvme-fc: set max_segments to lldd max value
  nvme-fc: drop a duplicated word in a comment
  nvme-hwmon: log the controller device name
  ...

drivers/acpi/property.c

@@ -45,6 +45,9 @@ static const guid_t prp_guids[] = {
 	/* Thunderbolt GUID for WAKE_SUPPORTED: 6c501103-c189-4296-ba72-9bf5a26ebe5d */
 	GUID_INIT(0x6c501103, 0xc189, 0x4296,
 		  0xba, 0x72, 0x9b, 0xf5, 0xa2, 0x6e, 0xbe, 0x5d),
+	/* Storage device needs D3 GUID: 5025030f-842f-4ab4-a561-99a5189762d0 */
+	GUID_INIT(0x5025030f, 0x842f, 0x4ab4,
+		  0xa5, 0x61, 0x99, 0xa5, 0x18, 0x97, 0x62, 0xd0),
 };
 
 /* ACPI _DSD data subnodes GUID: dbb8e3e6-5886-4ba6-8795-1319f52a966b */

drivers/nvme/host/fabrics.c

@@ -547,7 +547,7 @@ static struct nvmf_transport_ops *nvmf_lookup_transport(
 blk_status_t nvmf_fail_nonready_command(struct nvme_ctrl *ctrl,
 		struct request *rq)
 {
-	if (ctrl->state != NVME_CTRL_DELETING &&
+	if (ctrl->state != NVME_CTRL_DELETING_NOIO &&
 	    ctrl->state != NVME_CTRL_DEAD &&
 	    !blk_noretry_request(rq) && !(rq->cmd_flags & REQ_NVME_MPATH))
 		return BLK_STS_RESOURCE;

drivers/nvme/host/fabrics.h

@@ -182,7 +182,8 @@ bool nvmf_ip_options_match(struct nvme_ctrl *ctrl,
 static inline bool nvmf_check_ready(struct nvme_ctrl *ctrl, struct request *rq,
 		bool queue_live)
 {
-	if (likely(ctrl->state == NVME_CTRL_LIVE))
+	if (likely(ctrl->state == NVME_CTRL_LIVE ||
+		   ctrl->state == NVME_CTRL_DELETING))
 		return true;
 	return __nvmf_check_ready(ctrl, rq, queue_live);
 }

drivers/nvme/host/fc.c

@@ -826,6 +826,7 @@ nvme_fc_ctrl_connectivity_loss(struct nvme_fc_ctrl *ctrl)
 		break;
 
 	case NVME_CTRL_DELETING:
+	case NVME_CTRL_DELETING_NOIO:
 	default:
 		/* no action to take - let it delete */
 		break;
@@ -3001,8 +3002,9 @@ nvme_fc_create_association(struct nvme_fc_ctrl *ctrl)
 	if (ret)
 		goto out_disconnect_admin_queue;
 
-	ctrl->ctrl.max_hw_sectors =
-		(ctrl->lport->ops->max_sgl_segments - 1) << (PAGE_SHIFT - 9);
+	ctrl->ctrl.max_segments = ctrl->lport->ops->max_sgl_segments;
+	ctrl->ctrl.max_hw_sectors = ctrl->ctrl.max_segments <<
+						(ilog2(SZ_4K) - 9);
 
 	blk_mq_unquiesce_queue(ctrl->ctrl.admin_q);
 

drivers/nvme/host/hwmon.c

@@ -241,7 +241,8 @@ void nvme_hwmon_init(struct nvme_ctrl *ctrl)
 
 	err = nvme_hwmon_get_smart_log(data);
 	if (err) {
-		dev_warn(dev, "Failed to read smart log (error %d)\n", err);
+		dev_warn(ctrl->device,
+			"Failed to read smart log (error %d)\n", err);
 		devm_kfree(dev, data);
 		return;
 	}

drivers/nvme/host/multipath.c

@@ -167,9 +167,18 @@ void nvme_mpath_clear_ctrl_paths(struct nvme_ctrl *ctrl)
 
 static bool nvme_path_is_disabled(struct nvme_ns *ns)
 {
-	return ns->ctrl->state != NVME_CTRL_LIVE ||
-		test_bit(NVME_NS_ANA_PENDING, &ns->flags) ||
-		test_bit(NVME_NS_REMOVING, &ns->flags);
+	/*
+	 * We don't treat NVME_CTRL_DELETING as a disabled path as I/O should
+	 * still be able to complete assuming that the controller is connected.
+	 * Otherwise it will fail immediately and return to the requeue list.
+	 */
+	if (ns->ctrl->state != NVME_CTRL_LIVE &&
+	    ns->ctrl->state != NVME_CTRL_DELETING)
+		return true;
+	if (test_bit(NVME_NS_ANA_PENDING, &ns->flags) ||
+	    test_bit(NVME_NS_REMOVING, &ns->flags))
+		return true;
+	return false;
 }
 
 static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head, int node)
@@ -246,6 +255,12 @@ static struct nvme_ns *nvme_round_robin_path(struct nvme_ns_head *head,
 			fallback = ns;
 	}
 
+	/* No optimized path found, re-check the current path */
+	if (!nvme_path_is_disabled(old) &&
+	    old->ana_state == NVME_ANA_OPTIMIZED) {
+		found = old;
+		goto out;
+	}
 	if (!fallback)
 		return NULL;
 	found = fallback;
@@ -266,10 +281,13 @@ inline struct nvme_ns *nvme_find_path(struct nvme_ns_head *head)
 	struct nvme_ns *ns;
 
 	ns = srcu_dereference(head->current_path[node], &head->srcu);
-	if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR && ns)
-		ns = nvme_round_robin_path(head, node, ns);
-	if (unlikely(!ns || !nvme_path_is_optimized(ns)))
-		ns = __nvme_find_path(head, node);
+	if (unlikely(!ns))
+		return __nvme_find_path(head, node);
+
+	if (READ_ONCE(head->subsys->iopolicy) == NVME_IOPOLICY_RR)
+		return nvme_round_robin_path(head, node, ns);
+	if (unlikely(!nvme_path_is_optimized(ns)))
+		return __nvme_find_path(head, node);
 	return ns;
 }
 
@@ -563,6 +581,9 @@ static void nvme_ana_work(struct work_struct *work)
 {
 	struct nvme_ctrl *ctrl = container_of(work, struct nvme_ctrl, ana_work);
 
+	if (ctrl->state != NVME_CTRL_LIVE)
+		return;
+
 	nvme_read_ana_log(ctrl);
 }
 

drivers/nvme/host/nvme.h

@@ -37,6 +37,14 @@ extern unsigned int admin_timeout;
 #define  NVME_INLINE_METADATA_SG_CNT  1
 #endif
 
+/*
+ * Default to a 4K page size, with the intention to update this
+ * path in the future to accommodate architectures with differing
+ * kernel and IO page sizes.
+ */
+#define NVME_CTRL_PAGE_SHIFT	12
+#define NVME_CTRL_PAGE_SIZE	(1 << NVME_CTRL_PAGE_SHIFT)
+
 extern struct workqueue_struct *nvme_wq;
 extern struct workqueue_struct *nvme_reset_wq;
 extern struct workqueue_struct *nvme_delete_wq;
@@ -173,12 +181,32 @@ static inline u16 nvme_req_qid(struct request *req)
  */
 #define NVME_QUIRK_DELAY_AMOUNT		2300
 
+/*
+ * enum nvme_ctrl_state: Controller state
+ *
+ * @NVME_CTRL_NEW:		New controller just allocated, initial state
+ * @NVME_CTRL_LIVE:		Controller is connected and I/O capable
+ * @NVME_CTRL_RESETTING:	Controller is resetting (or scheduled reset)
+ * @NVME_CTRL_CONNECTING:	Controller is disconnected, now connecting the
+ *				transport
+ * @NVME_CTRL_DELETING:		Controller is deleting (or scheduled deletion)
+ * @NVME_CTRL_DELETING_NOIO:	Controller is deleting and I/O is not
+ *				disabled/failed immediately. This state comes
+ * 				after all async event processing took place and
+ * 				before ns removal and the controller deletion
+ * 				progress
+ * @NVME_CTRL_DEAD:		Controller is non-present/unresponsive during
+ *				shutdown or removal. In this case we forcibly
+ *				kill all inflight I/O as they have no chance to
+ *				complete
+ */
 enum nvme_ctrl_state {
 	NVME_CTRL_NEW,
 	NVME_CTRL_LIVE,
 	NVME_CTRL_RESETTING,
 	NVME_CTRL_CONNECTING,
 	NVME_CTRL_DELETING,
+	NVME_CTRL_DELETING_NOIO,
 	NVME_CTRL_DEAD,
 };
 
@@ -234,7 +262,6 @@ struct nvme_ctrl {
 	u32 queue_count;
 
 	u64 cap;
-	u32 page_size;
 	u32 max_hw_sectors;
 	u32 max_segments;
 	u32 max_integrity_segments;
@@ -763,4 +790,11 @@ void nvme_hwmon_init(struct nvme_ctrl *ctrl);
 static inline void nvme_hwmon_init(struct nvme_ctrl *ctrl) { }
 #endif
 
+u32 nvme_command_effects(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
+			 u8 opcode);
+void nvme_execute_passthru_rq(struct request *rq);
+struct nvme_ctrl *nvme_ctrl_get_by_path(const char *path);
+struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid);
+void nvme_put_ns(struct nvme_ns *ns);
+
 #endif /* _NVME_H */

drivers/nvme/host/pci.c

@@ -4,6 +4,7 @@
  * Copyright (c) 2011-2014, Intel Corporation.
  */
 
+#include <linux/acpi.h>
 #include <linux/aer.h>
 #include <linux/async.h>
 #include <linux/blkdev.h>
@@ -94,6 +95,10 @@ static unsigned int poll_queues;
 module_param_cb(poll_queues, &io_queue_count_ops, &poll_queues, 0644);
 MODULE_PARM_DESC(poll_queues, "Number of queues to use for polled IO.");
 
+static bool noacpi;
+module_param(noacpi, bool, 0444);
+MODULE_PARM_DESC(noacpi, "disable acpi bios quirks");
+
 struct nvme_dev;
 struct nvme_queue;
 
@@ -346,10 +351,10 @@ static bool nvme_dbbuf_update_and_check_event(u16 value, u32 *dbbuf_db,
  * as it only leads to a small amount of wasted memory for the lifetime of
  * the I/O.
  */
-static int nvme_npages(unsigned size, struct nvme_dev *dev)
+static int nvme_pci_npages_prp(void)
 {
-	unsigned nprps = DIV_ROUND_UP(size + dev->ctrl.page_size,
-				      dev->ctrl.page_size);
+	unsigned nprps = DIV_ROUND_UP(NVME_MAX_KB_SZ + NVME_CTRL_PAGE_SIZE,
+				      NVME_CTRL_PAGE_SIZE);
 	return DIV_ROUND_UP(8 * nprps, PAGE_SIZE - 8);
 }
 
@@ -357,22 +362,18 @@ static int nvme_npages(unsigned size, struct nvme_dev *dev)
  * Calculates the number of pages needed for the SGL segments. For example a 4k
  * page can accommodate 256 SGL descriptors.
  */
-static int nvme_pci_npages_sgl(unsigned int num_seg)
+static int nvme_pci_npages_sgl(void)
 {
-	return DIV_ROUND_UP(num_seg * sizeof(struct nvme_sgl_desc), PAGE_SIZE);
+	return DIV_ROUND_UP(NVME_MAX_SEGS * sizeof(struct nvme_sgl_desc),
+			PAGE_SIZE);
 }
 
-static size_t nvme_pci_iod_alloc_size(struct nvme_dev *dev,
-		unsigned int size, unsigned int nseg, bool use_sgl)
+static size_t nvme_pci_iod_alloc_size(void)
 {
-	size_t alloc_size;
-
-	if (use_sgl)
-		alloc_size = sizeof(__le64 *) * nvme_pci_npages_sgl(nseg);
-	else
-		alloc_size = sizeof(__le64 *) * nvme_npages(size, dev);
+	size_t npages = max(nvme_pci_npages_prp(), nvme_pci_npages_sgl());
 
-	return alloc_size + sizeof(struct scatterlist) * nseg;
+	return sizeof(__le64 *) * npages +
+		sizeof(struct scatterlist) * NVME_MAX_SEGS;
 }
 
 static int nvme_admin_init_hctx(struct blk_mq_hw_ctx *hctx, void *data,
@@ -515,7 +516,7 @@ 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);
-	const int last_prp = dev->ctrl.page_size / sizeof(__le64) - 1;
+	const int last_prp = NVME_CTRL_PAGE_SIZE / sizeof(__le64) - 1;
 	dma_addr_t dma_addr = iod->first_dma, next_dma_addr;
 	int i;
 
@@ -582,34 +583,33 @@ static blk_status_t nvme_pci_setup_prps(struct nvme_dev *dev,
 	struct scatterlist *sg = iod->sg;
 	int dma_len = sg_dma_len(sg);
 	u64 dma_addr = sg_dma_address(sg);
-	u32 page_size = dev->ctrl.page_size;
-	int offset = dma_addr & (page_size - 1);
+	int offset = dma_addr & (NVME_CTRL_PAGE_SIZE - 1);
 	__le64 *prp_list;
 	void **list = nvme_pci_iod_list(req);
 	dma_addr_t prp_dma;
 	int nprps, i;
 
-	length -= (page_size - offset);
+	length -= (NVME_CTRL_PAGE_SIZE - offset);
 	if (length <= 0) {
 		iod->first_dma = 0;
 		goto done;
 	}
 
-	dma_len -= (page_size - offset);
+	dma_len -= (NVME_CTRL_PAGE_SIZE - offset);
 	if (dma_len) {
-		dma_addr += (page_size - offset);
+		dma_addr += (NVME_CTRL_PAGE_SIZE - offset);
 	} else {
 		sg = sg_next(sg);
 		dma_addr = sg_dma_address(sg);
 		dma_len = sg_dma_len(sg);
 	}
 
-	if (length <= page_size) {
+	if (length <= NVME_CTRL_PAGE_SIZE) {
 		iod->first_dma = dma_addr;
 		goto done;
 	}
 
-	nprps = DIV_ROUND_UP(length, page_size);
+	nprps = DIV_ROUND_UP(length, NVME_CTRL_PAGE_SIZE);
 	if (nprps <= (256 / 8)) {
 		pool = dev->prp_small_pool;
 		iod->npages = 0;
@@ -628,7 +628,7 @@ static blk_status_t nvme_pci_setup_prps(struct nvme_dev *dev,
 	iod->first_dma = prp_dma;
 	i = 0;
 	for (;;) {
-		if (i == page_size >> 3) {
+		if (i == NVME_CTRL_PAGE_SIZE >> 3) {
 			__le64 *old_prp_list = prp_list;
 			prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma);
 			if (!prp_list)
@@ -639,9 +639,9 @@ static blk_status_t nvme_pci_setup_prps(struct nvme_dev *dev,
 			i = 1;
 		}
 		prp_list[i++] = cpu_to_le64(dma_addr);
-		dma_len -= page_size;
-		dma_addr += page_size;
-		length -= page_size;
+		dma_len -= NVME_CTRL_PAGE_SIZE;
+		dma_addr += NVME_CTRL_PAGE_SIZE;
+		length -= NVME_CTRL_PAGE_SIZE;
 		if (length <= 0)
 			break;
 		if (dma_len > 0)
@@ -751,8 +751,8 @@ static blk_status_t nvme_setup_prp_simple(struct nvme_dev *dev,
 		struct bio_vec *bv)
 {
 	struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
-	unsigned int offset = bv->bv_offset & (dev->ctrl.page_size - 1);
-	unsigned int first_prp_len = dev->ctrl.page_size - offset;
+	unsigned int offset = bv->bv_offset & (NVME_CTRL_PAGE_SIZE - 1);
+	unsigned int first_prp_len = NVME_CTRL_PAGE_SIZE - offset;
 
 	iod->first_dma = dma_map_bvec(dev->dev, bv, rq_dma_dir(req), 0);
 	if (dma_mapping_error(dev->dev, iod->first_dma))
@@ -794,7 +794,7 @@ static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req,
 		struct bio_vec bv = req_bvec(req);
 
 		if (!is_pci_p2pdma_page(bv.bv_page)) {
-			if (bv.bv_offset + bv.bv_len <= dev->ctrl.page_size * 2)
+			if (bv.bv_offset + bv.bv_len <= NVME_CTRL_PAGE_SIZE * 2)
 				return nvme_setup_prp_simple(dev, req,
 							     &cmnd->rw, &bv);
 
@@ -1396,12 +1396,12 @@ static int nvme_cmb_qdepth(struct nvme_dev *dev, int nr_io_queues,
 {
 	int q_depth = dev->q_depth;
 	unsigned q_size_aligned = roundup(q_depth * entry_size,
-					  dev->ctrl.page_size);
+					  NVME_CTRL_PAGE_SIZE);
 
 	if (q_size_aligned * nr_io_queues > dev->cmb_size) {
 		u64 mem_per_q = div_u64(dev->cmb_size, nr_io_queues);
 
-		mem_per_q = round_down(mem_per_q, dev->ctrl.page_size);
+		mem_per_q = round_down(mem_per_q, NVME_CTRL_PAGE_SIZE);
 		q_depth = div_u64(mem_per_q, entry_size);
 
 		/*
@@ -1816,6 +1816,7 @@ static inline void nvme_release_cmb(struct nvme_dev *dev)
 
 static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits)
 {
+	u32 host_mem_size = dev->host_mem_size >> NVME_CTRL_PAGE_SHIFT;
 	u64 dma_addr = dev->host_mem_descs_dma;
 	struct nvme_command c;
 	int ret;
@@ -1824,8 +1825,7 @@ static int nvme_set_host_mem(struct nvme_dev *dev, u32 bits)
 	c.features.opcode	= nvme_admin_set_features;
 	c.features.fid		= cpu_to_le32(NVME_FEAT_HOST_MEM_BUF);
 	c.features.dword11	= cpu_to_le32(bits);
-	c.features.dword12	= cpu_to_le32(dev->host_mem_size >>
-					      ilog2(dev->ctrl.page_size));
+	c.features.dword12	= cpu_to_le32(host_mem_size);
 	c.features.dword13	= cpu_to_le32(lower_32_bits(dma_addr));
 	c.features.dword14	= cpu_to_le32(upper_32_bits(dma_addr));
 	c.features.dword15	= cpu_to_le32(dev->nr_host_mem_descs);
@@ -1845,7 +1845,7 @@ static void nvme_free_host_mem(struct nvme_dev *dev)
 
 	for (i = 0; i < dev->nr_host_mem_descs; i++) {
 		struct nvme_host_mem_buf_desc *desc = &dev->host_mem_descs[i];
-		size_t size = le32_to_cpu(desc->size) * dev->ctrl.page_size;
+		size_t size = le32_to_cpu(desc->size) * NVME_CTRL_PAGE_SIZE;
 
 		dma_free_attrs(dev->dev, size, dev->host_mem_desc_bufs[i],
 			       le64_to_cpu(desc->addr),
@@ -1897,7 +1897,7 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 			break;
 
 		descs[i].addr = cpu_to_le64(dma_addr);
-		descs[i].size = cpu_to_le32(len / dev->ctrl.page_size);
+		descs[i].size = cpu_to_le32(len / NVME_CTRL_PAGE_SIZE);
 		i++;
 	}
 
@@ -1913,7 +1913,7 @@ static int __nvme_alloc_host_mem(struct nvme_dev *dev, u64 preferred,
 
 out_free_bufs:
 	while (--i >= 0) {
-		size_t size = le32_to_cpu(descs[i].size) * dev->ctrl.page_size;
+		size_t size = le32_to_cpu(descs[i].size) * NVME_CTRL_PAGE_SIZE;
 
 		dma_free_attrs(dev->dev, size, bufs[i],
 			       le64_to_cpu(descs[i].addr),
@@ -2759,6 +2759,54 @@ static unsigned long check_vendor_combination_bug(struct pci_dev *pdev)
 	return 0;
 }
 
+#ifdef CONFIG_ACPI
+static bool nvme_acpi_storage_d3(struct pci_dev *dev)
+{
+	struct acpi_device *adev;
+	struct pci_dev *root;
+	acpi_handle handle;
+	acpi_status status;
+	u8 val;
+
+	/*
+	 * Look for _DSD property specifying that the storage device on the port
+	 * must use D3 to support deep platform power savings during
+	 * suspend-to-idle.
+	 */
+	root = pcie_find_root_port(dev);
+	if (!root)
+		return false;
+
+	adev = ACPI_COMPANION(&root->dev);
+	if (!adev)
+		return false;
+
+	/*
+	 * The property is defined in the PXSX device for South complex ports
+	 * and in the PEGP device for North complex ports.
+	 */
+	status = acpi_get_handle(adev->handle, "PXSX", &handle);
+	if (ACPI_FAILURE(status)) {
+		status = acpi_get_handle(adev->handle, "PEGP", &handle);
+		if (ACPI_FAILURE(status))
+			return false;
+	}
+
+	if (acpi_bus_get_device(handle, &adev))
+		return false;
+
+	if (fwnode_property_read_u8(acpi_fwnode_handle(adev), "StorageD3Enable",
+			&val))
+		return false;
+	return val == 1;
+}
+#else
+static inline bool nvme_acpi_storage_d3(struct pci_dev *dev)
+{
+	return false;
+}
+#endif /* CONFIG_ACPI */
+
 static void nvme_async_probe(void *data, async_cookie_t cookie)
 {
 	struct nvme_dev *dev = data;
@@ -2808,12 +2856,21 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 	quirks |= check_vendor_combination_bug(pdev);
 
+	if (!noacpi && nvme_acpi_storage_d3(pdev)) {
+		/*
+		 * Some systems use a bios work around to ask for D3 on
+		 * platforms that support kernel managed suspend.
+		 */
+		dev_info(&pdev->dev,
+			 "platform quirk: setting simple suspend\n");
+		quirks |= NVME_QUIRK_SIMPLE_SUSPEND;
+	}
+
 	/*
 	 * Double check that our mempool alloc size will cover the biggest
 	 * command we support.
 	 */
-	alloc_size = nvme_pci_iod_alloc_size(dev, NVME_MAX_KB_SZ,
-						NVME_MAX_SEGS, true);
+	alloc_size = nvme_pci_iod_alloc_size();
 	WARN_ON_ONCE(alloc_size > PAGE_SIZE);
 
 	dev->iod_mempool = mempool_create_node(1, mempool_kmalloc,

drivers/nvme/host/rdma.c

@@ -96,6 +96,7 @@ struct nvme_rdma_queue {
 	int			cm_error;
 	struct completion	cm_done;
 	bool			pi_support;
+	int			cq_size;
 };
 
 struct nvme_rdma_ctrl {
@@ -275,6 +276,7 @@ static int nvme_rdma_create_qp(struct nvme_rdma_queue *queue, const int factor)
 	init_attr.recv_cq = queue->ib_cq;
 	if (queue->pi_support)
 		init_attr.create_flags |= IB_QP_CREATE_INTEGRITY_EN;
+	init_attr.qp_context = queue;
 
 	ret = rdma_create_qp(queue->cm_id, dev->pd, &init_attr);
 
@@ -409,6 +411,14 @@ nvme_rdma_find_get_device(struct rdma_cm_id *cm_id)
 	return NULL;
 }
 
+static void nvme_rdma_free_cq(struct nvme_rdma_queue *queue)
+{
+	if (nvme_rdma_poll_queue(queue))
+		ib_free_cq(queue->ib_cq);
+	else
+		ib_cq_pool_put(queue->ib_cq, queue->cq_size);
+}
+
 static void nvme_rdma_destroy_queue_ib(struct nvme_rdma_queue *queue)
 {
 	struct nvme_rdma_device *dev;
@@ -430,7 +440,7 @@ static void nvme_rdma_destroy_queue_ib(struct nvme_rdma_queue *queue)
 	 * the destruction of the QP shouldn't use rdma_cm API.
 	 */
 	ib_destroy_qp(queue->qp);
-	ib_free_cq(queue->ib_cq);
+	nvme_rdma_free_cq(queue);
 
 	nvme_rdma_free_ring(ibdev, queue->rsp_ring, queue->queue_size,
 			sizeof(struct nvme_completion), DMA_FROM_DEVICE);
@@ -450,13 +460,42 @@ static int nvme_rdma_get_max_fr_pages(struct ib_device *ibdev, bool pi_support)
 	return min_t(u32, NVME_RDMA_MAX_SEGMENTS, max_page_list_len - 1);
 }
 
+static int nvme_rdma_create_cq(struct ib_device *ibdev,
+		struct nvme_rdma_queue *queue)
+{
+	int ret, comp_vector, idx = nvme_rdma_queue_idx(queue);
+	enum ib_poll_context poll_ctx;
+
+	/*
+	 * Spread I/O queues completion vectors according their queue index.
+	 * Admin queues can always go on completion vector 0.
+	 */
+	comp_vector = (idx == 0 ? idx : idx - 1) % ibdev->num_comp_vectors;
+
+	/* Polling queues need direct cq polling context */
+	if (nvme_rdma_poll_queue(queue)) {
+		poll_ctx = IB_POLL_DIRECT;
+		queue->ib_cq = ib_alloc_cq(ibdev, queue, queue->cq_size,
+					   comp_vector, poll_ctx);
+	} else {
+		poll_ctx = IB_POLL_SOFTIRQ;
+		queue->ib_cq = ib_cq_pool_get(ibdev, queue->cq_size,
+					      comp_vector, poll_ctx);
+	}
+
+	if (IS_ERR(queue->ib_cq)) {
+		ret = PTR_ERR(queue->ib_cq);
+		return ret;
+	}
+
+	return 0;
+}
+
 static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 {
 	struct ib_device *ibdev;
 	const int send_wr_factor = 3;			/* MR, SEND, INV */
 	const int cq_factor = send_wr_factor + 1;	/* + RECV */
-	int comp_vector, idx = nvme_rdma_queue_idx(queue);
-	enum ib_poll_context poll_ctx;
 	int ret, pages_per_mr;
 
 	queue->device = nvme_rdma_find_get_device(queue->cm_id);
@@ -467,26 +506,12 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 	}
 	ibdev = queue->device->dev;
 
-	/*
-	 * Spread I/O queues completion vectors according their queue index.
-	 * Admin queues can always go on completion vector 0.
-	 */
-	comp_vector = (idx == 0 ? idx : idx - 1) % ibdev->num_comp_vectors;
-
-	/* Polling queues need direct cq polling context */
-	if (nvme_rdma_poll_queue(queue))
-		poll_ctx = IB_POLL_DIRECT;
-	else
-		poll_ctx = IB_POLL_SOFTIRQ;
-
 	/* +1 for ib_stop_cq */
-	queue->ib_cq = ib_alloc_cq(ibdev, queue,
-				cq_factor * queue->queue_size + 1,
-				comp_vector, poll_ctx);
-	if (IS_ERR(queue->ib_cq)) {
-		ret = PTR_ERR(queue->ib_cq);
+	queue->cq_size = cq_factor * queue->queue_size + 1;
+
+	ret = nvme_rdma_create_cq(ibdev, queue);
+	if (ret)
 		goto out_put_dev;
-	}
 
 	ret = nvme_rdma_create_qp(queue, send_wr_factor);
 	if (ret)
@@ -512,7 +537,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 	if (ret) {
 		dev_err(queue->ctrl->ctrl.device,
 			"failed to initialize MR pool sized %d for QID %d\n",
-			queue->queue_size, idx);
+			queue->queue_size, nvme_rdma_queue_idx(queue));
 		goto out_destroy_ring;
 	}
 
@@ -523,7 +548,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 		if (ret) {
 			dev_err(queue->ctrl->ctrl.device,
 				"failed to initialize PI MR pool sized %d for QID %d\n",
-				queue->queue_size, idx);
+				queue->queue_size, nvme_rdma_queue_idx(queue));
 			goto out_destroy_mr_pool;
 		}
 	}
@@ -540,7 +565,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 out_destroy_qp:
 	rdma_destroy_qp(queue->cm_id);
 out_destroy_ib_cq:
-	ib_free_cq(queue->ib_cq);
+	nvme_rdma_free_cq(queue);
 out_put_dev:
 	nvme_rdma_dev_put(queue->device);
 	return ret;
@@ -942,15 +967,20 @@ static int nvme_rdma_configure_io_queues(struct nvme_rdma_ctrl *ctrl, bool new)
 			ret = PTR_ERR(ctrl->ctrl.connect_q);
 			goto out_free_tag_set;
 		}
-	} else {
-		blk_mq_update_nr_hw_queues(&ctrl->tag_set,
-			ctrl->ctrl.queue_count - 1);
 	}
 
 	ret = nvme_rdma_start_io_queues(ctrl);
 	if (ret)
 		goto out_cleanup_connect_q;
 
+	if (!new) {
+		nvme_start_queues(&ctrl->ctrl);
+		nvme_wait_freeze(&ctrl->ctrl);
+		blk_mq_update_nr_hw_queues(ctrl->ctrl.tagset,
+			ctrl->ctrl.queue_count - 1);
+		nvme_unfreeze(&ctrl->ctrl);
+	}
+
 	return 0;
 
 out_cleanup_connect_q:
@@ -983,6 +1013,7 @@ static void nvme_rdma_teardown_io_queues(struct nvme_rdma_ctrl *ctrl,
 		bool remove)
 {
 	if (ctrl->ctrl.queue_count > 1) {
+		nvme_start_freeze(&ctrl->ctrl);
 		nvme_stop_queues(&ctrl->ctrl);
 		nvme_rdma_stop_io_queues(ctrl);
 		if (ctrl->ctrl.tagset) {
@@ -1077,11 +1108,12 @@ static int nvme_rdma_setup_ctrl(struct nvme_rdma_ctrl *ctrl, bool new)
 	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
 	if (!changed) {
 		/*
-		 * state change failure is ok if we're in DELETING state,
+		 * state change failure is ok if we started ctrl delete,
 		 * unless we're during creation of a new controller to
 		 * avoid races with teardown flow.
 		 */
-		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING);
+		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING &&
+			     ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO);
 		WARN_ON_ONCE(new);
 		ret = -EINVAL;
 		goto destroy_io;
@@ -1134,8 +1166,9 @@ static void nvme_rdma_error_recovery_work(struct work_struct *work)
 	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 */
-		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING);
+		/* state change failure is ok if we started ctrl delete */
+		WARN_ON_ONCE(ctrl->ctrl.state != NVME_CTRL_DELETING &&
+			     ctrl->ctrl.state != NVME_CTRL_DELETING_NOIO);
 		return;
 	}
 
@@ -1163,7 +1196,7 @@ static void nvme_rdma_end_request(struct nvme_rdma_request *req)
 static void nvme_rdma_wr_error(struct ib_cq *cq, struct ib_wc *wc,
 		const char *op)
 {
-	struct nvme_rdma_queue *queue = cq->cq_context;
+	struct nvme_rdma_queue *queue = wc->qp->qp_context;
 	struct nvme_rdma_ctrl *ctrl = queue->ctrl;
 
 	if (ctrl->ctrl.state == NVME_CTRL_LIVE)
@@ -1706,7 +1739,7 @@ static void nvme_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc)
 {
 	struct nvme_rdma_qe *qe =
 		container_of(wc->wr_cqe, struct nvme_rdma_qe, cqe);
-	struct nvme_rdma_queue *queue = cq->cq_context;
+	struct nvme_rdma_queue *queue = wc->qp->qp_context;
 	struct ib_device *ibdev = queue->device->dev;
 	struct nvme_completion *cqe = qe->data;
 	const size_t len = sizeof(struct nvme_completion);

drivers/nvme/host/tcp.c

@@ -1771,15 +1771,20 @@ static int nvme_tcp_configure_io_queues(struct nvme_ctrl *ctrl, bool new)
 			ret = PTR_ERR(ctrl->connect_q);
 			goto out_free_tag_set;
 		}
-	} else {
-		blk_mq_update_nr_hw_queues(ctrl->tagset,
-			ctrl->queue_count - 1);
 	}
 
 	ret = nvme_tcp_start_io_queues(ctrl);
 	if (ret)
 		goto out_cleanup_connect_q;
 
+	if (!new) {
+		nvme_start_queues(ctrl);
+		nvme_wait_freeze(ctrl);
+		blk_mq_update_nr_hw_queues(ctrl->tagset,
+			ctrl->queue_count - 1);
+		nvme_unfreeze(ctrl);
+	}
+
 	return 0;
 
 out_cleanup_connect_q:
@@ -1884,6 +1889,7 @@ static void nvme_tcp_teardown_io_queues(struct nvme_ctrl *ctrl,
 {
 	if (ctrl->queue_count <= 1)
 		return;
+	nvme_start_freeze(ctrl);
 	nvme_stop_queues(ctrl);
 	nvme_tcp_stop_io_queues(ctrl);
 	if (ctrl->tagset) {
@@ -1950,11 +1956,12 @@ static int nvme_tcp_setup_ctrl(struct nvme_ctrl *ctrl, bool new)
 
 	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_LIVE)) {
 		/*
-		 * state change failure is ok if we're in DELETING state,
+		 * state change failure is ok if we started ctrl delete,
 		 * unless we're during creation of a new controller to
 		 * avoid races with teardown flow.
 		 */
-		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING);
+		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING &&
+			     ctrl->state != NVME_CTRL_DELETING_NOIO);
 		WARN_ON_ONCE(new);
 		ret = -EINVAL;
 		goto destroy_io;
@@ -2010,8 +2017,9 @@ static void nvme_tcp_error_recovery_work(struct work_struct *work)
 	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 */
-		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING);
+		/* state change failure is ok if we started ctrl delete */
+		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING &&
+			     ctrl->state != NVME_CTRL_DELETING_NOIO);
 		return;
 	}
 
@@ -2046,8 +2054,9 @@ static void nvme_reset_ctrl_work(struct work_struct *work)
 	nvme_tcp_teardown_ctrl(ctrl, false);
 
 	if (!nvme_change_ctrl_state(ctrl, NVME_CTRL_CONNECTING)) {
-		/* state change failure is ok if we're in DELETING state */
-		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING);
+		/* state change failure is ok if we started ctrl delete */
+		WARN_ON_ONCE(ctrl->state != NVME_CTRL_DELETING &&
+			     ctrl->state != NVME_CTRL_DELETING_NOIO);
 		return;
 	}
 

drivers/nvme/target/Kconfig

@@ -16,6 +16,18 @@ config NVME_TARGET
 	  To configure the NVMe target you probably want to use the nvmetcli
 	  tool from http://git.infradead.org/users/hch/nvmetcli.git.
 
+config NVME_TARGET_PASSTHRU
+	bool "NVMe Target Passthrough support"
+	depends on NVME_TARGET
+	depends on NVME_CORE=y || NVME_CORE=NVME_TARGET
+	help
+	  This enables target side NVMe passthru controller support for the
+	  NVMe Over Fabrics protocol. It allows for hosts to manage and
+	  directly access an actual NVMe controller residing on the target
+	  side, incuding executing Vendor Unique Commands.
+
+	  If unsure, say N.
+
 config NVME_TARGET_LOOP
 	tristate "NVMe loopback device support"
 	depends on NVME_TARGET

drivers/nvme/target/Makefile

@@ -11,6 +11,7 @@ obj-$(CONFIG_NVME_TARGET_TCP)		+= nvmet-tcp.o
 
 nvmet-y		+= core.o configfs.o admin-cmd.o fabrics-cmd.o \
 			discovery.o io-cmd-file.o io-cmd-bdev.o
+nvmet-$(CONFIG_NVME_TARGET_PASSTHRU)	+= passthru.o
 nvme-loop-y	+= loop.o
 nvmet-rdma-y	+= rdma.o
 nvmet-fc-y	+= fc.o

drivers/nvme/target/admin-cmd.c

@@ -113,11 +113,10 @@ static u16 nvmet_get_smart_log_all(struct nvmet_req *req,
 	u64 data_units_read = 0, data_units_written = 0;
 	struct nvmet_ns *ns;
 	struct nvmet_ctrl *ctrl;
+	unsigned long idx;
 
 	ctrl = req->sq->ctrl;
-
-	rcu_read_lock();
-	list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link) {
+	xa_for_each(&ctrl->subsys->namespaces, idx, ns) {
 		/* we don't have the right data for file backed ns */
 		if (!ns->bdev)
 			continue;
@@ -127,9 +126,7 @@ static u16 nvmet_get_smart_log_all(struct nvmet_req *req,
 		host_writes += part_stat_read(ns->bdev->bd_part, ios[WRITE]);
 		data_units_written += DIV_ROUND_UP(
 			part_stat_read(ns->bdev->bd_part, sectors[WRITE]), 1000);
-
 	}
-	rcu_read_unlock();
 
 	put_unaligned_le64(host_reads, &slog->host_reads[0]);
 	put_unaligned_le64(data_units_read, &slog->data_units_read[0]);
@@ -230,14 +227,13 @@ static u32 nvmet_format_ana_group(struct nvmet_req *req, u32 grpid,
 {
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
 	struct nvmet_ns *ns;
+	unsigned long idx;
 	u32 count = 0;
 
 	if (!(req->cmd->get_log_page.lsp & NVME_ANA_LOG_RGO)) {
-		rcu_read_lock();
-		list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link)
+		xa_for_each(&ctrl->subsys->namespaces, idx, ns)
 			if (ns->anagrpid == grpid)
 				desc->nsids[count++] = cpu_to_le32(ns->nsid);
-		rcu_read_unlock();
 	}
 
 	desc->grpid = cpu_to_le32(grpid);
@@ -556,6 +552,7 @@ static void nvmet_execute_identify_nslist(struct nvmet_req *req)
 	static const int buf_size = NVME_IDENTIFY_DATA_SIZE;
 	struct nvmet_ctrl *ctrl = req->sq->ctrl;
 	struct nvmet_ns *ns;
+	unsigned long idx;
 	u32 min_nsid = le32_to_cpu(req->cmd->identify.nsid);
 	__le32 *list;
 	u16 status = 0;
@@ -567,15 +564,13 @@ static void nvmet_execute_identify_nslist(struct nvmet_req *req)
 		goto out;
 	}
 
-	rcu_read_lock();
-	list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link) {
+	xa_for_each(&ctrl->subsys->namespaces, idx, ns) {
 		if (ns->nsid <= min_nsid)
 			continue;
 		list[i++] = cpu_to_le32(ns->nsid);
 		if (i == buf_size / sizeof(__le32))
 			break;
 	}
-	rcu_read_unlock();
 
 	status = nvmet_copy_to_sgl(req, 0, list, buf_size);
 
@@ -754,7 +749,7 @@ u16 nvmet_set_feat_async_event(struct nvmet_req *req, u32 mask)
 	return 0;
 }
 
-static void nvmet_execute_set_features(struct nvmet_req *req)
+void nvmet_execute_set_features(struct nvmet_req *req)
 {
 	struct nvmet_subsys *subsys = req->sq->ctrl->subsys;
 	u32 cdw10 = le32_to_cpu(req->cmd->common.cdw10);
@@ -829,7 +824,7 @@ void nvmet_get_feat_async_event(struct nvmet_req *req)
 	nvmet_set_result(req, READ_ONCE(req->sq->ctrl->aen_enabled));
 }
 
-static void nvmet_execute_get_features(struct nvmet_req *req)
+void nvmet_execute_get_features(struct nvmet_req *req)
 {
 	struct nvmet_subsys *subsys = req->sq->ctrl->subsys;
 	u32 cdw10 = le32_to_cpu(req->cmd->common.cdw10);
@@ -945,6 +940,9 @@ u16 nvmet_parse_admin_cmd(struct nvmet_req *req)
 	if (unlikely(ret))
 		return ret;
 
+	if (nvmet_req_passthru_ctrl(req))
+		return nvmet_parse_passthru_admin_cmd(req);
+
 	switch (cmd->common.opcode) {
 	case nvme_admin_get_log_page:
 		req->execute = nvmet_execute_get_log_page;

drivers/nvme/target/configfs.c

@@ -666,6 +666,103 @@ static const struct config_item_type nvmet_namespaces_type = {
 	.ct_owner		= THIS_MODULE,
 };
 
+#ifdef CONFIG_NVME_TARGET_PASSTHRU
+
+static ssize_t nvmet_passthru_device_path_show(struct config_item *item,
+		char *page)
+{
+	struct nvmet_subsys *subsys = to_subsys(item->ci_parent);
+
+	return snprintf(page, PAGE_SIZE, "%s\n", subsys->passthru_ctrl_path);
+}
+
+static ssize_t nvmet_passthru_device_path_store(struct config_item *item,
+		const char *page, size_t count)
+{
+	struct nvmet_subsys *subsys = to_subsys(item->ci_parent);
+	size_t len;
+	int ret;
+
+	mutex_lock(&subsys->lock);
+
+	ret = -EBUSY;
+	if (subsys->passthru_ctrl)
+		goto out_unlock;
+
+	ret = -EINVAL;
+	len = strcspn(page, "\n");
+	if (!len)
+		goto out_unlock;
+
+	kfree(subsys->passthru_ctrl_path);
+	ret = -ENOMEM;
+	subsys->passthru_ctrl_path = kstrndup(page, len, GFP_KERNEL);
+	if (!subsys->passthru_ctrl_path)
+		goto out_unlock;
+
+	mutex_unlock(&subsys->lock);
+
+	return count;
+out_unlock:
+	mutex_unlock(&subsys->lock);
+	return ret;
+}
+CONFIGFS_ATTR(nvmet_passthru_, device_path);
+
+static ssize_t nvmet_passthru_enable_show(struct config_item *item,
+		char *page)
+{
+	struct nvmet_subsys *subsys = to_subsys(item->ci_parent);
+
+	return sprintf(page, "%d\n", subsys->passthru_ctrl ? 1 : 0);
+}
+
+static ssize_t nvmet_passthru_enable_store(struct config_item *item,
+		const char *page, size_t count)
+{
+	struct nvmet_subsys *subsys = to_subsys(item->ci_parent);
+	bool enable;
+	int ret = 0;
+
+	if (strtobool(page, &enable))
+		return -EINVAL;
+
+	if (enable)
+		ret = nvmet_passthru_ctrl_enable(subsys);
+	else
+		nvmet_passthru_ctrl_disable(subsys);
+
+	return ret ? ret : count;
+}
+CONFIGFS_ATTR(nvmet_passthru_, enable);
+
+static struct configfs_attribute *nvmet_passthru_attrs[] = {
+	&nvmet_passthru_attr_device_path,
+	&nvmet_passthru_attr_enable,
+	NULL,
+};
+
+static const struct config_item_type nvmet_passthru_type = {
+	.ct_attrs		= nvmet_passthru_attrs,
+	.ct_owner		= THIS_MODULE,
+};
+
+static void nvmet_add_passthru_group(struct nvmet_subsys *subsys)
+{
+	config_group_init_type_name(&subsys->passthru_group,
+				    "passthru", &nvmet_passthru_type);
+	configfs_add_default_group(&subsys->passthru_group,
+				   &subsys->group);
+}
+
+#else /* CONFIG_NVME_TARGET_PASSTHRU */
+
+static void nvmet_add_passthru_group(struct nvmet_subsys *subsys)
+{
+}
+
+#endif /* CONFIG_NVME_TARGET_PASSTHRU */
+
 static int nvmet_port_subsys_allow_link(struct config_item *parent,
 		struct config_item *target)
 {
@@ -879,6 +976,10 @@ static ssize_t nvmet_subsys_attr_version_store(struct config_item *item,
 	int major, minor, tertiary = 0;
 	int ret;
 
+	/* passthru subsystems use the underlying controller's version */
+	if (nvmet_passthru_ctrl(subsys))
+		return -EINVAL;
+
 	ret = sscanf(page, "%d.%d.%d\n", &major, &minor, &tertiary);
 	if (ret != 2 && ret != 3)
 		return -EINVAL;
@@ -1121,6 +1222,8 @@ static struct config_group *nvmet_subsys_make(struct config_group *group,
 	configfs_add_default_group(&subsys->allowed_hosts_group,
 			&subsys->group);
 
+	nvmet_add_passthru_group(subsys);
+
 	return &subsys->group;
 }
 

drivers/nvme/target/core.c

@@ -115,13 +115,14 @@ u16 nvmet_zero_sgl(struct nvmet_req *req, off_t off, size_t len)
 
 static unsigned int nvmet_max_nsid(struct nvmet_subsys *subsys)
 {
-	struct nvmet_ns *ns;
+	unsigned long nsid = 0;
+	struct nvmet_ns *cur;
+	unsigned long idx;
 
-	if (list_empty(&subsys->namespaces))
-		return 0;
+	xa_for_each(&subsys->namespaces, idx, cur)
+		nsid = cur->nsid;
 
-	ns = list_last_entry(&subsys->namespaces, struct nvmet_ns, dev_link);
-	return ns->nsid;
+	return nsid;
 }
 
 static u32 nvmet_async_event_result(struct nvmet_async_event *aen)
@@ -410,28 +411,13 @@ static void nvmet_stop_keep_alive_timer(struct nvmet_ctrl *ctrl)
 	cancel_delayed_work_sync(&ctrl->ka_work);
 }
 
-static struct nvmet_ns *__nvmet_find_namespace(struct nvmet_ctrl *ctrl,
-		__le32 nsid)
-{
-	struct nvmet_ns *ns;
-
-	list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link) {
-		if (ns->nsid == le32_to_cpu(nsid))
-			return ns;
-	}
-
-	return NULL;
-}
-
 struct nvmet_ns *nvmet_find_namespace(struct nvmet_ctrl *ctrl, __le32 nsid)
 {
 	struct nvmet_ns *ns;
 
-	rcu_read_lock();
-	ns = __nvmet_find_namespace(ctrl, nsid);
+	ns = xa_load(&ctrl->subsys->namespaces, le32_to_cpu(nsid));
 	if (ns)
 		percpu_ref_get(&ns->ref);
-	rcu_read_unlock();
 
 	return ns;
 }
@@ -558,6 +544,12 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 
 	mutex_lock(&subsys->lock);
 	ret = 0;
+
+	if (nvmet_passthru_ctrl(subsys)) {
+		pr_info("cannot enable both passthru and regular namespaces for a single subsystem");
+		goto out_unlock;
+	}
+
 	if (ns->enabled)
 		goto out_unlock;
 
@@ -586,24 +578,10 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 	if (ns->nsid > subsys->max_nsid)
 		subsys->max_nsid = ns->nsid;
 
-	/*
-	 * The namespaces list needs to be sorted to simplify the implementation
-	 * of the Identify Namepace List subcommand.
-	 */
-	if (list_empty(&subsys->namespaces)) {
-		list_add_tail_rcu(&ns->dev_link, &subsys->namespaces);
-	} else {
-		struct nvmet_ns *old;
-
-		list_for_each_entry_rcu(old, &subsys->namespaces, dev_link,
-					lockdep_is_held(&subsys->lock)) {
-			BUG_ON(ns->nsid == old->nsid);
-			if (ns->nsid < old->nsid)
-				break;
-		}
+	ret = xa_insert(&subsys->namespaces, ns->nsid, ns, GFP_KERNEL);
+	if (ret)
+		goto out_restore_subsys_maxnsid;
 
-		list_add_tail_rcu(&ns->dev_link, &old->dev_link);
-	}
 	subsys->nr_namespaces++;
 
 	nvmet_ns_changed(subsys, ns->nsid);
@@ -612,6 +590,10 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 out_unlock:
 	mutex_unlock(&subsys->lock);
 	return ret;
+
+out_restore_subsys_maxnsid:
+	subsys->max_nsid = nvmet_max_nsid(subsys);
+	percpu_ref_exit(&ns->ref);
 out_dev_put:
 	list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry)
 		pci_dev_put(radix_tree_delete(&ctrl->p2p_ns_map, ns->nsid));
@@ -630,7 +612,7 @@ void nvmet_ns_disable(struct nvmet_ns *ns)
 		goto out_unlock;
 
 	ns->enabled = false;
-	list_del_rcu(&ns->dev_link);
+	xa_erase(&ns->subsys->namespaces, ns->nsid);
 	if (ns->nsid == subsys->max_nsid)
 		subsys->max_nsid = nvmet_max_nsid(subsys);
 
@@ -681,7 +663,6 @@ struct nvmet_ns *nvmet_ns_alloc(struct nvmet_subsys *subsys, u32 nsid)
 	if (!ns)
 		return NULL;
 
-	INIT_LIST_HEAD(&ns->dev_link);
 	init_completion(&ns->disable_done);
 
 	ns->nsid = nsid;
@@ -874,6 +855,9 @@ static u16 nvmet_parse_io_cmd(struct nvmet_req *req)
 	if (unlikely(ret))
 		return ret;
 
+	if (nvmet_req_passthru_ctrl(req))
+		return nvmet_parse_passthru_io_cmd(req);
+
 	req->ns = nvmet_find_namespace(req->sq->ctrl, cmd->rw.nsid);
 	if (unlikely(!req->ns)) {
 		req->error_loc = offsetof(struct nvme_common_command, nsid);
@@ -1263,14 +1247,14 @@ static void nvmet_setup_p2p_ns_map(struct nvmet_ctrl *ctrl,
 		struct nvmet_req *req)
 {
 	struct nvmet_ns *ns;
+	unsigned long idx;
 
 	if (!req->p2p_client)
 		return;
 
 	ctrl->p2p_client = get_device(req->p2p_client);
 
-	list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link,
-				lockdep_is_held(&ctrl->subsys->lock))
+	xa_for_each(&ctrl->subsys->namespaces, idx, ns)
 		nvmet_p2pmem_ns_add_p2p(ctrl, ns);
 }
 
@@ -1495,7 +1479,7 @@ struct nvmet_subsys *nvmet_subsys_alloc(const char *subsysnqn,
 	if (!subsys)
 		return ERR_PTR(-ENOMEM);
 
-	subsys->ver = NVME_VS(1, 3, 0); /* NVMe 1.3.0 */
+	subsys->ver = NVMET_DEFAULT_VS;
 	/* generate a random serial number as our controllers are ephemeral: */
 	get_random_bytes(&subsys->serial, sizeof(subsys->serial));
 
@@ -1523,7 +1507,7 @@ struct nvmet_subsys *nvmet_subsys_alloc(const char *subsysnqn,
 	kref_init(&subsys->ref);
 
 	mutex_init(&subsys->lock);
-	INIT_LIST_HEAD(&subsys->namespaces);
+	xa_init(&subsys->namespaces);
 	INIT_LIST_HEAD(&subsys->ctrls);
 	INIT_LIST_HEAD(&subsys->hosts);
 
@@ -1535,7 +1519,10 @@ static void nvmet_subsys_free(struct kref *ref)
 	struct nvmet_subsys *subsys =
 		container_of(ref, struct nvmet_subsys, ref);
 
-	WARN_ON_ONCE(!list_empty(&subsys->namespaces));
+	WARN_ON_ONCE(!xa_empty(&subsys->namespaces));
+
+	xa_destroy(&subsys->namespaces);
+	nvmet_passthru_subsys_free(subsys);
 
 	kfree(subsys->subsysnqn);
 	kfree_rcu(subsys->model, rcuhead);

drivers/nvme/target/fc.c

@@ -167,7 +167,6 @@ struct nvmet_fc_tgt_assoc {
 	struct nvmet_fc_tgt_queue	*queues[NVMET_NR_QUEUES + 1];
 	struct kref			ref;
 	struct work_struct		del_work;
-	atomic_t			del_work_active;
 };
 
 
@@ -1090,7 +1089,6 @@ nvmet_fc_delete_assoc(struct work_struct *work)
 		container_of(work, struct nvmet_fc_tgt_assoc, del_work);
 
 	nvmet_fc_delete_target_assoc(assoc);
-	atomic_set(&assoc->del_work_active, 0);
 	nvmet_fc_tgt_a_put(assoc);
 }
 
@@ -1123,7 +1121,6 @@ nvmet_fc_alloc_target_assoc(struct nvmet_fc_tgtport *tgtport, void *hosthandle)
 	INIT_LIST_HEAD(&assoc->a_list);
 	kref_init(&assoc->ref);
 	INIT_WORK(&assoc->del_work, nvmet_fc_delete_assoc);
-	atomic_set(&assoc->del_work_active, 0);
 	atomic_set(&assoc->terminating, 0);
 
 	while (needrandom) {
@@ -1243,7 +1240,8 @@ nvmet_fc_find_target_assoc(struct nvmet_fc_tgtport *tgtport,
 	list_for_each_entry(assoc, &tgtport->assoc_list, a_list) {
 		if (association_id == assoc->association_id) {
 			ret = assoc;
-			nvmet_fc_tgt_a_get(assoc);
+			if (!nvmet_fc_tgt_a_get(assoc))
+				ret = NULL;
 			break;
 		}
 	}
@@ -1477,22 +1475,16 @@ __nvmet_fc_free_assocs(struct nvmet_fc_tgtport *tgtport)
 {
 	struct nvmet_fc_tgt_assoc *assoc, *next;
 	unsigned long flags;
-	int ret;
 
 	spin_lock_irqsave(&tgtport->lock, flags);
 	list_for_each_entry_safe(assoc, next,
 				&tgtport->assoc_list, a_list) {
 		if (!nvmet_fc_tgt_a_get(assoc))
 			continue;
-		ret = atomic_cmpxchg(&assoc->del_work_active, 0, 1);
-		if (ret == 0) {
 		if (!schedule_work(&assoc->del_work))
-				nvmet_fc_tgt_a_put(assoc);
-		} else {
 			/* already deleting - release local reference */
 			nvmet_fc_tgt_a_put(assoc);
 	}
-	}
 	spin_unlock_irqrestore(&tgtport->lock, flags);
 }
 
@@ -1533,7 +1525,6 @@ nvmet_fc_invalidate_host(struct nvmet_fc_target_port *target_port,
 	struct nvmet_fc_tgt_assoc *assoc, *next;
 	unsigned long flags;
 	bool noassoc = true;
-	int ret;
 
 	spin_lock_irqsave(&tgtport->lock, flags);
 	list_for_each_entry_safe(assoc, next,
@@ -1545,15 +1536,10 @@ nvmet_fc_invalidate_host(struct nvmet_fc_target_port *target_port,
 			continue;
 		assoc->hostport->invalid = 1;
 		noassoc = false;
-		ret = atomic_cmpxchg(&assoc->del_work_active, 0, 1);
-		if (ret == 0) {
 		if (!schedule_work(&assoc->del_work))
-				nvmet_fc_tgt_a_put(assoc);
-		} else {
 			/* already deleting - release local reference */
 			nvmet_fc_tgt_a_put(assoc);
 	}
-	}
 	spin_unlock_irqrestore(&tgtport->lock, flags);
 
 	/* if there's nothing to wait for - call the callback */
@@ -1573,7 +1559,6 @@ nvmet_fc_delete_ctrl(struct nvmet_ctrl *ctrl)
 	struct nvmet_fc_tgt_queue *queue;
 	unsigned long flags;
 	bool found_ctrl = false;
-	int ret;
 
 	/* this is a bit ugly, but don't want to make locks layered */
 	spin_lock_irqsave(&nvmet_fc_tgtlock, flags);
@@ -1597,14 +1582,9 @@ nvmet_fc_delete_ctrl(struct nvmet_ctrl *ctrl)
 		nvmet_fc_tgtport_put(tgtport);
 
 		if (found_ctrl) {
-			ret = atomic_cmpxchg(&assoc->del_work_active, 0, 1);
-			if (ret == 0) {
 			if (!schedule_work(&assoc->del_work))
-					nvmet_fc_tgt_a_put(assoc);
-			} else {
 				/* already deleting - release local reference */
 				nvmet_fc_tgt_a_put(assoc);
-			}
 			return;
 		}
 

drivers/nvme/target/loop.c

@@ -444,7 +444,6 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work)
 {
 	struct nvme_loop_ctrl *ctrl =
 		container_of(work, struct nvme_loop_ctrl, ctrl.reset_work);
-	bool changed;
 	int ret;
 
 	nvme_stop_ctrl(&ctrl->ctrl);
@@ -471,8 +470,8 @@ static void nvme_loop_reset_ctrl_work(struct work_struct *work)
 	blk_mq_update_nr_hw_queues(&ctrl->tag_set,
 			ctrl->ctrl.queue_count - 1);
 
-	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
-	WARN_ON_ONCE(!changed);
+	if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE))
+		WARN_ON_ONCE(1);
 
 	nvme_start_ctrl(&ctrl->ctrl);
 
@@ -567,7 +566,6 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev,
 		struct nvmf_ctrl_options *opts)
 {
 	struct nvme_loop_ctrl *ctrl;
-	bool changed;
 	int ret;
 
 	ctrl = kzalloc(sizeof(*ctrl), GFP_KERNEL);
@@ -583,6 +581,9 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev,
 	if (ret)
 		goto out_put_ctrl;
 
+	if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_CONNECTING))
+		WARN_ON_ONCE(1);
+
 	ret = -ENOMEM;
 
 	ctrl->ctrl.sqsize = opts->queue_size - 1;
@@ -617,8 +618,8 @@ static struct nvme_ctrl *nvme_loop_create_ctrl(struct device *dev,
 	dev_info(ctrl->ctrl.device,
 		 "new ctrl: \"%s\"\n", ctrl->ctrl.opts->subsysnqn);
 
-	changed = nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE);
-	WARN_ON_ONCE(!changed);
+	if (!nvme_change_ctrl_state(&ctrl->ctrl, NVME_CTRL_LIVE))
+		WARN_ON_ONCE(1);
 
 	mutex_lock(&nvme_loop_ctrl_mutex);
 	list_add_tail(&ctrl->list, &nvme_loop_ctrl_list);

drivers/nvme/target/nvmet.h

@@ -21,6 +21,8 @@
 #include <linux/radix-tree.h>
 #include <linux/t10-pi.h>
 
+#define NVMET_DEFAULT_VS		NVME_VS(1, 3, 0)
+
 #define NVMET_ASYNC_EVENTS		4
 #define NVMET_ERROR_LOG_SLOTS		128
 #define NVMET_NO_ERROR_LOC		((u16)-1)
@@ -52,7 +54,6 @@
 	(cpu_to_le32(offsetof(struct nvmf_connect_command, x)))
 
 struct nvmet_ns {
-	struct list_head	dev_link;
 	struct percpu_ref	ref;
 	struct block_device	*bdev;
 	struct file		*file;
@@ -219,7 +220,7 @@ struct nvmet_subsys {
 	struct mutex		lock;
 	struct kref		ref;
 
-	struct list_head	namespaces;
+	struct xarray		namespaces;
 	unsigned int		nr_namespaces;
 	unsigned int		max_nsid;
 	u16			cntlid_min;
@@ -243,6 +244,12 @@ struct nvmet_subsys {
 	struct config_group	allowed_hosts_group;
 
 	struct nvmet_subsys_model	__rcu *model;
+
+#ifdef CONFIG_NVME_TARGET_PASSTHRU
+	struct nvme_ctrl	*passthru_ctrl;
+	char			*passthru_ctrl_path;
+	struct config_group	passthru_group;
+#endif /* CONFIG_NVME_TARGET_PASSTHRU */
 };
 
 static inline struct nvmet_subsys *to_subsys(struct config_item *item)
@@ -322,6 +329,11 @@ struct nvmet_req {
 			struct bio_vec          *bvec;
 			struct work_struct      work;
 		} f;
+		struct {
+			struct request		*rq;
+			struct work_struct      work;
+			bool			use_workqueue;
+		} p;
 	};
 	int			sg_cnt;
 	int			metadata_sg_cnt;
@@ -401,6 +413,8 @@ void nvmet_req_complete(struct nvmet_req *req, u16 status);
 int nvmet_req_alloc_sgls(struct nvmet_req *req);
 void nvmet_req_free_sgls(struct nvmet_req *req);
 
+void nvmet_execute_set_features(struct nvmet_req *req);
+void nvmet_execute_get_features(struct nvmet_req *req);
 void nvmet_execute_keep_alive(struct nvmet_req *req);
 
 void nvmet_cq_setup(struct nvmet_ctrl *ctrl, struct nvmet_cq *cq, u16 qid,
@@ -533,6 +547,43 @@ static inline u32 nvmet_dsm_len(struct nvmet_req *req)
 		sizeof(struct nvme_dsm_range);
 }
 
+#ifdef CONFIG_NVME_TARGET_PASSTHRU
+void nvmet_passthru_subsys_free(struct nvmet_subsys *subsys);
+int nvmet_passthru_ctrl_enable(struct nvmet_subsys *subsys);
+void nvmet_passthru_ctrl_disable(struct nvmet_subsys *subsys);
+u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req);
+u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req);
+static inline struct nvme_ctrl *nvmet_passthru_ctrl(struct nvmet_subsys *subsys)
+{
+	return subsys->passthru_ctrl;
+}
+#else /* CONFIG_NVME_TARGET_PASSTHRU */
+static inline void nvmet_passthru_subsys_free(struct nvmet_subsys *subsys)
+{
+}
+static inline void nvmet_passthru_ctrl_disable(struct nvmet_subsys *subsys)
+{
+}
+static inline u16 nvmet_parse_passthru_admin_cmd(struct nvmet_req *req)
+{
+	return 0;
+}
+static inline u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req)
+{
+	return 0;
+}
+static inline struct nvme_ctrl *nvmet_passthru_ctrl(struct nvmet_subsys *subsys)
+{
+	return NULL;
+}
+#endif /* CONFIG_NVME_TARGET_PASSTHRU */
+
+static inline struct nvme_ctrl *
+nvmet_req_passthru_ctrl(struct nvmet_req *req)
+{
+	return nvmet_passthru_ctrl(req->sq->ctrl->subsys);
+}
+
 u16 errno_to_nvme_status(struct nvmet_req *req, int errno);
 
 /* Convert a 32-bit number to a 16-bit 0's based number */

drivers/nvme/target/rdma.c

@@ -752,7 +752,7 @@ static void nvmet_rdma_read_data_done(struct ib_cq *cq, struct ib_wc *wc)
 {
 	struct nvmet_rdma_rsp *rsp =
 		container_of(wc->wr_cqe, struct nvmet_rdma_rsp, read_cqe);
-	struct nvmet_rdma_queue *queue = cq->cq_context;
+	struct nvmet_rdma_queue *queue = wc->qp->qp_context;
 	u16 status = 0;
 
 	WARN_ON(rsp->n_rdma <= 0);
@@ -1008,7 +1008,7 @@ static void nvmet_rdma_recv_done(struct ib_cq *cq, struct ib_wc *wc)
 {
 	struct nvmet_rdma_cmd *cmd =
 		container_of(wc->wr_cqe, struct nvmet_rdma_cmd, cqe);
-	struct nvmet_rdma_queue *queue = cq->cq_context;
+	struct nvmet_rdma_queue *queue = wc->qp->qp_context;
 	struct nvmet_rdma_rsp *rsp;
 
 	if (unlikely(wc->status != IB_WC_SUCCESS)) {
@@ -1258,9 +1258,8 @@ static int nvmet_rdma_create_queue_ib(struct nvmet_rdma_queue *queue)
 	 */
 	nr_cqe = queue->recv_queue_size + 2 * queue->send_queue_size;
 
-	queue->cq = ib_alloc_cq(ndev->device, queue,
-			nr_cqe + 1, queue->comp_vector,
-			IB_POLL_WORKQUEUE);
+	queue->cq = ib_cq_pool_get(ndev->device, nr_cqe + 1,
+				   queue->comp_vector, IB_POLL_WORKQUEUE);
 	if (IS_ERR(queue->cq)) {
 		ret = PTR_ERR(queue->cq);
 		pr_err("failed to create CQ cqe= %d ret= %d\n",
@@ -1322,7 +1321,7 @@ static int nvmet_rdma_create_queue_ib(struct nvmet_rdma_queue *queue)
 err_destroy_qp:
 	rdma_destroy_qp(queue->cm_id);
 err_destroy_cq:
-	ib_free_cq(queue->cq);
+	ib_cq_pool_put(queue->cq, nr_cqe + 1);
 	goto out;
 }
 
@@ -1332,7 +1331,8 @@ static void nvmet_rdma_destroy_queue_ib(struct nvmet_rdma_queue *queue)
 	if (queue->cm_id)
 		rdma_destroy_id(queue->cm_id);
 	ib_destroy_qp(queue->qp);
-	ib_free_cq(queue->cq);
+	ib_cq_pool_put(queue->cq, queue->recv_queue_size + 2 *
+		       queue->send_queue_size + 1);
 }
 
 static void nvmet_rdma_free_queue(struct nvmet_rdma_queue *queue)

include/linux/nvme-fc-driver.h

@@ -672,7 +672,7 @@ enum {
  * Values set by the LLDD indicating completion status of the FCP operation.
  * Must be set prior to calling the done() callback.
  * @transferred_length: amount of DATA_OUT payload data received by a
- *            a WRITEDATA operation. If not a WRITEDATA operation, value must
+ *            WRITEDATA operation. If not a WRITEDATA operation, value must
  *            be set to 0. Should equal transfer_length on success.
  * @fcp_error: status of the FCP operation. Must be 0 on success; on failure
  *            must be a NVME_SC_FC_xxxx value.

include/linux/nvme.h

@@ -312,6 +312,7 @@ enum {
 	NVME_CTRL_ONCS_WRITE_UNCORRECTABLE	= 1 << 1,
 	NVME_CTRL_ONCS_DSM			= 1 << 2,
 	NVME_CTRL_ONCS_WRITE_ZEROES		= 1 << 3,
+	NVME_CTRL_ONCS_RESERVATIONS		= 1 << 5,
 	NVME_CTRL_ONCS_TIMESTAMP		= 1 << 6,
 	NVME_CTRL_VWC_PRESENT			= 1 << 0,
 	NVME_CTRL_OACS_SEC_SUPP                 = 1 << 0,
@@ -982,6 +983,7 @@ enum nvme_admin_opcode {
 	nvme_admin_security_recv	= 0x82,
 	nvme_admin_sanitize_nvm		= 0x84,
 	nvme_admin_get_lba_status	= 0x86,
+	nvme_admin_vendor_start		= 0xC0,
 };
 
 #define nvme_admin_opcode_name(opcode)	{ opcode, #opcode }
@@ -1045,6 +1047,8 @@ enum {
 	NVME_FEAT_RESV_MASK	= 0x82,
 	NVME_FEAT_RESV_PERSIST	= 0x83,
 	NVME_FEAT_WRITE_PROTECT	= 0x84,
+	NVME_FEAT_VENDOR_START	= 0xC0,
+	NVME_FEAT_VENDOR_END	= 0xFF,
 	NVME_LOG_ERROR		= 0x01,
 	NVME_LOG_SMART		= 0x02,
 	NVME_LOG_FW_SLOT	= 0x03,