Merge branch 'nvme-4.18-2' of git://git.infradead.org/nvme into for-4.18/block

Pull NVMe changes from Christoph:

"Here is the current batch of nvme updates for 4.18, we have a few more
 patches in the queue, but I'd like to get this pile into your tree
 and linux-next ASAP.

 The biggest item is support for file-backed namespaces in the NVMe
 target from Chaitanya, in addition to that we mostly small fixes from
 all the usual suspects."

* 'nvme-4.18-2' of git://git.infradead.org/nvme:
  nvme: fixup memory leak in nvme_init_identify()
  nvme: fix KASAN warning when parsing host nqn
  nvmet-loop: use nr_phys_segments when map rq to sgl
  nvmet-fc: increase LS buffer count per fc port
  nvmet: add simple file backed ns support
  nvmet: remove duplicate NULL initialization for req->ns
  nvmet: make a few error messages more generic
  nvme-fabrics: allow duplicate connections to the discovery controller
  nvme-fabrics: centralize discovery controller defaults
  nvme-fabrics: remove unnecessary controller subnqn validation
  nvme-fc: remove setting DNR on exception conditions
  nvme-rdma: stop admin queue before freeing it
  nvme-pci: Fix AER reset handling
  nvme-pci: set nvmeq->cq_vector after alloc cq/sq
  nvme: host: core: fix precedence of ternary operator
  nvme: fix lockdep warning in nvme_mpath_clear_current_path

drivers/nvme/host/core.c

@@ -1578,7 +1578,7 @@ static int nvme_pr_reserve(struct block_device *bdev, u64 key,
 static int nvme_pr_preempt(struct block_device *bdev, u64 old, u64 new,
 		enum pr_type type, bool abort)
 {
-	u32 cdw10 = nvme_pr_type(type) << 8 | abort ? 2 : 1;
+	u32 cdw10 = nvme_pr_type(type) << 8 | (abort ? 2 : 1);
 	return nvme_pr_command(bdev, cdw10, old, new, nvme_cmd_resv_acquire);
 }
 
@@ -1590,7 +1590,7 @@ static int nvme_pr_clear(struct block_device *bdev, u64 key)
 
 static int nvme_pr_release(struct block_device *bdev, u64 key, enum pr_type type)
 {
-	u32 cdw10 = nvme_pr_type(type) << 8 | key ? 1 << 3 : 0;
+	u32 cdw10 = nvme_pr_type(type) << 8 | (key ? 1 << 3 : 0);
 	return nvme_pr_command(bdev, cdw10, key, 0, nvme_cmd_resv_release);
 }
 
@@ -2184,7 +2184,8 @@ static int nvme_init_subsystem(struct nvme_ctrl *ctrl, struct nvme_id_ctrl *id)
 		 * Verify that the subsystem actually supports multiple
 		 * controllers, else bail out.
 		 */
-		if (nvme_active_ctrls(found) && !(id->cmic & (1 << 1))) {
+		if (!ctrl->opts->discovery_nqn &&
+		    nvme_active_ctrls(found) && !(id->cmic & (1 << 1))) {
 			dev_err(ctrl->device,
 				"ignoring ctrl due to duplicate subnqn (%s).\n",
 				found->subnqn);
@@ -2315,7 +2316,7 @@ int nvme_init_identify(struct nvme_ctrl *ctrl)
 	if (id->lpa & NVME_CTRL_LPA_CMD_EFFECTS_LOG) {
 		ret = nvme_get_effects_log(ctrl);
 		if (ret < 0)
-			return ret;
+			goto out_free;
 	}
 
 	if (!ctrl->identified) {

drivers/nvme/host/fabrics.c

@@ -57,7 +57,7 @@ static struct nvmf_host *nvmf_host_add(const char *hostnqn)
 		goto out_unlock;
 
 	kref_init(&host->ref);
-	memcpy(host->nqn, hostnqn, NVMF_NQN_SIZE);
+	strlcpy(host->nqn, hostnqn, NVMF_NQN_SIZE);
 
 	list_add_tail(&host->list, &nvmf_hosts);
 out_unlock:
@@ -689,10 +689,6 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 			opts->discovery_nqn =
 				!(strcmp(opts->subsysnqn,
 					 NVME_DISC_SUBSYS_NAME));
-			if (opts->discovery_nqn) {
-				opts->kato = 0;
-				opts->nr_io_queues = 0;
-			}
 			break;
 		case NVMF_OPT_TRADDR:
 			p = match_strdup(args);
@@ -851,6 +847,11 @@ static int nvmf_parse_options(struct nvmf_ctrl_options *opts,
 		}
 	}
 
+	if (opts->discovery_nqn) {
+		opts->kato = 0;
+		opts->nr_io_queues = 0;
+		opts->duplicate_connect = true;
+	}
 	if (ctrl_loss_tmo < 0)
 		opts->max_reconnects = -1;
 	else
@@ -983,16 +984,6 @@ nvmf_create_ctrl(struct device *dev, const char *buf, size_t count)
 		goto out_module_put;
 	}
 
-	if (strcmp(ctrl->subsys->subnqn, opts->subsysnqn)) {
-		dev_warn(ctrl->device,
-			"controller returned incorrect NQN: \"%s\".\n",
-			ctrl->subsys->subnqn);
-		module_put(ops->module);
-		up_read(&nvmf_transports_rwsem);
-		nvme_delete_ctrl_sync(ctrl);
-		return ERR_PTR(-EINVAL);
-	}
-
 	module_put(ops->module);
 	up_read(&nvmf_transports_rwsem);
 	return ctrl;

drivers/nvme/host/fc.c

@@ -1686,16 +1686,6 @@ nvme_fc_fcpio_done(struct nvmefc_fcp_req *req)
 		goto check_error;
 	}
 
-	/*
-	 * Force failures of commands if we're killing the controller
-	 * or have an error on a command used to create an new association
-	 */
-	if (status &&
-	    (blk_queue_dying(rq->q) ||
-	     ctrl->ctrl.state == NVME_CTRL_NEW ||
-	     ctrl->ctrl.state == NVME_CTRL_CONNECTING))
-		status |= cpu_to_le16(NVME_SC_DNR << 1);
-
 	__nvme_fc_fcpop_chk_teardowns(ctrl, op, opstate);
 	nvme_end_request(rq, status, result);
 

drivers/nvme/host/nvme.h

@@ -22,6 +22,7 @@
 #include <linux/lightnvm.h>
 #include <linux/sed-opal.h>
 #include <linux/fault-inject.h>
+#include <linux/rcupdate.h>
 
 extern unsigned int nvme_io_timeout;
 #define NVME_IO_TIMEOUT	(nvme_io_timeout * HZ)
@@ -449,7 +450,7 @@ static inline void nvme_mpath_clear_current_path(struct nvme_ns *ns)
 {
 	struct nvme_ns_head *head = ns->head;
 
-	if (head && ns == srcu_dereference(head->current_path, &head->srcu))
+	if (head && ns == rcu_access_pointer(head->current_path))
 		rcu_assign_pointer(head->current_path, NULL);
 }
 struct nvme_ns *nvme_find_path(struct nvme_ns_head *head);

drivers/nvme/host/pci.c

@@ -1076,7 +1076,7 @@ static int adapter_delete_queue(struct nvme_dev *dev, u8 opcode, u16 id)
 }
 
 static int adapter_alloc_cq(struct nvme_dev *dev, u16 qid,
-						struct nvme_queue *nvmeq)
+		struct nvme_queue *nvmeq, s16 vector)
 {
 	struct nvme_command c;
 	int flags = NVME_QUEUE_PHYS_CONTIG | NVME_CQ_IRQ_ENABLED;
@@ -1091,7 +1091,7 @@ static int adapter_alloc_cq(struct nvme_dev *dev, u16 qid,
 	c.create_cq.cqid = cpu_to_le16(qid);
 	c.create_cq.qsize = cpu_to_le16(nvmeq->q_depth - 1);
 	c.create_cq.cq_flags = cpu_to_le16(flags);
-	c.create_cq.irq_vector = cpu_to_le16(nvmeq->cq_vector);
+	c.create_cq.irq_vector = cpu_to_le16(vector);
 
 	return nvme_submit_sync_cmd(dev->ctrl.admin_q, &c, NULL, 0);
 }
@@ -1462,6 +1462,7 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid)
 {
 	struct nvme_dev *dev = nvmeq->dev;
 	int result;
+	s16 vector;
 
 	if (dev->cmb && use_cmb_sqes && (dev->cmbsz & NVME_CMBSZ_SQS)) {
 		unsigned offset = (qid - 1) * roundup(SQ_SIZE(nvmeq->q_depth),
@@ -1474,15 +1475,21 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid)
 	 * A queue's vector matches the queue identifier unless the controller
 	 * has only one vector available.
 	 */
-	nvmeq->cq_vector = dev->num_vecs == 1 ? 0 : qid;
-	result = adapter_alloc_cq(dev, qid, nvmeq);
+	vector = dev->num_vecs == 1 ? 0 : qid;
+	result = adapter_alloc_cq(dev, qid, nvmeq, vector);
 	if (result < 0)
-		goto release_vector;
+		goto out;
 
 	result = adapter_alloc_sq(dev, qid, nvmeq);
 	if (result < 0)
 		goto release_cq;
 
+	/*
+	 * Set cq_vector after alloc cq/sq, otherwise nvme_suspend_queue will
+	 * invoke free_irq for it and cause a 'Trying to free already-free IRQ
+	 * xxx' warning if the create CQ/SQ command times out.
+	 */
+	nvmeq->cq_vector = vector;
 	nvme_init_queue(nvmeq, qid);
 	result = queue_request_irq(nvmeq);
 	if (result < 0)
@@ -1490,13 +1497,13 @@ static int nvme_create_queue(struct nvme_queue *nvmeq, int qid)
 
 	return result;
 
- release_sq:
+release_sq:
+	nvmeq->cq_vector = -1;
 	dev->online_queues--;
 	adapter_delete_sq(dev, qid);
- release_cq:
+release_cq:
 	adapter_delete_cq(dev, qid);
- release_vector:
-	nvmeq->cq_vector = -1;
+out:
 	return result;
 }
 
@@ -2695,19 +2702,15 @@ static pci_ers_result_t nvme_slot_reset(struct pci_dev *pdev)
 
 	dev_info(dev->ctrl.device, "restart after slot reset\n");
 	pci_restore_state(pdev);
-	nvme_reset_ctrl_sync(&dev->ctrl);
-
-	switch (dev->ctrl.state) {
-	case NVME_CTRL_LIVE:
-	case NVME_CTRL_ADMIN_ONLY:
+	nvme_reset_ctrl(&dev->ctrl);
 	return PCI_ERS_RESULT_RECOVERED;
-	default:
-		return PCI_ERS_RESULT_DISCONNECT;
-	}
 }
 
 static void nvme_error_resume(struct pci_dev *pdev)
 {
+	struct nvme_dev *dev = pci_get_drvdata(pdev);
+
+	flush_work(&dev->ctrl.reset_work);
 	pci_cleanup_aer_uncorrect_error_status(pdev);
 }
 

drivers/nvme/host/rdma.c

@@ -778,7 +778,7 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 	if (error) {
 		dev_err(ctrl->ctrl.device,
 			"prop_get NVME_REG_CAP failed\n");
-		goto out_cleanup_queue;
+		goto out_stop_queue;
 	}
 
 	ctrl->ctrl.sqsize =
@@ -786,23 +786,25 @@ static int nvme_rdma_configure_admin_queue(struct nvme_rdma_ctrl *ctrl,
 
 	error = nvme_enable_ctrl(&ctrl->ctrl, ctrl->ctrl.cap);
 	if (error)
-		goto out_cleanup_queue;
+		goto out_stop_queue;
 
 	ctrl->ctrl.max_hw_sectors =
 		(ctrl->max_fr_pages - 1) << (ilog2(SZ_4K) - 9);
 
 	error = nvme_init_identify(&ctrl->ctrl);
 	if (error)
-		goto out_cleanup_queue;
+		goto out_stop_queue;
 
 	error = nvme_rdma_alloc_qe(ctrl->queues[0].device->dev,
 			&ctrl->async_event_sqe, sizeof(struct nvme_command),
 			DMA_TO_DEVICE);
 	if (error)
-		goto out_cleanup_queue;
+		goto out_stop_queue;
 
 	return 0;
 
+out_stop_queue:
+	nvme_rdma_stop_queue(&ctrl->queues[0]);
 out_cleanup_queue:
 	if (new)
 		blk_cleanup_queue(ctrl->ctrl.admin_q);

drivers/nvme/target/Makefile

@@ -6,8 +6,8 @@ obj-$(CONFIG_NVME_TARGET_RDMA)		+= nvmet-rdma.o
 obj-$(CONFIG_NVME_TARGET_FC)		+= nvmet-fc.o
 obj-$(CONFIG_NVME_TARGET_FCLOOP)	+= nvme-fcloop.o
 
-nvmet-y		+= core.o configfs.o admin-cmd.o io-cmd.o fabrics-cmd.o \
-			discovery.o
+nvmet-y		+= core.o configfs.o admin-cmd.o fabrics-cmd.o \
+			discovery.o io-cmd-file.o io-cmd-bdev.o
 nvme-loop-y	+= loop.o
 nvmet-rdma-y	+= rdma.o
 nvmet-fc-y	+= fc.o

drivers/nvme/target/admin-cmd.c

@@ -45,6 +45,10 @@ static u16 nvmet_get_smart_log_nsid(struct nvmet_req *req,
 		return NVME_SC_INVALID_NS;
 	}
 
+	/* we don't have the right data for file backed ns */
+	if (!ns->bdev)
+		goto out;
+
 	host_reads = part_stat_read(ns->bdev->bd_part, ios[READ]);
 	data_units_read = part_stat_read(ns->bdev->bd_part, sectors[READ]);
 	host_writes = part_stat_read(ns->bdev->bd_part, ios[WRITE]);
@@ -54,6 +58,7 @@ static u16 nvmet_get_smart_log_nsid(struct nvmet_req *req,
 	put_unaligned_le64(data_units_read, &slog->data_units_read[0]);
 	put_unaligned_le64(host_writes, &slog->host_writes[0]);
 	put_unaligned_le64(data_units_written, &slog->data_units_written[0]);
+out:
 	nvmet_put_namespace(ns);
 
 	return NVME_SC_SUCCESS;
@@ -71,6 +76,9 @@ static u16 nvmet_get_smart_log_all(struct nvmet_req *req,
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(ns, &ctrl->subsys->namespaces, dev_link) {
+		/* we don't have the right data for file backed ns */
+		if (!ns->bdev)
+			continue;
 		host_reads += part_stat_read(ns->bdev->bd_part, ios[READ]);
 		data_units_read +=
 			part_stat_read(ns->bdev->bd_part, sectors[READ]);
@@ -548,8 +556,6 @@ u16 nvmet_parse_admin_cmd(struct nvmet_req *req)
 	struct nvme_command *cmd = req->cmd;
 	u16 ret;
 
-	req->ns = NULL;
-
 	ret = nvmet_check_ctrl_status(req, cmd);
 	if (unlikely(ret))
 		return ret;

drivers/nvme/target/core.c

@@ -271,6 +271,12 @@ void nvmet_put_namespace(struct nvmet_ns *ns)
 	percpu_ref_put(&ns->ref);
 }
 
+static void nvmet_ns_dev_disable(struct nvmet_ns *ns)
+{
+	nvmet_bdev_ns_disable(ns);
+	nvmet_file_ns_disable(ns);
+}
+
 int nvmet_ns_enable(struct nvmet_ns *ns)
 {
 	struct nvmet_subsys *subsys = ns->subsys;
@@ -281,23 +287,16 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 	if (ns->enabled)
 		goto out_unlock;
 
-	ns->bdev = blkdev_get_by_path(ns->device_path, FMODE_READ | FMODE_WRITE,
-			NULL);
-	if (IS_ERR(ns->bdev)) {
-		pr_err("failed to open block device %s: (%ld)\n",
-		       ns->device_path, PTR_ERR(ns->bdev));
-		ret = PTR_ERR(ns->bdev);
-		ns->bdev = NULL;
+	ret = nvmet_bdev_ns_enable(ns);
+	if (ret)
+		ret = nvmet_file_ns_enable(ns);
+	if (ret)
 		goto out_unlock;
-	}
-
-	ns->size = i_size_read(ns->bdev->bd_inode);
-	ns->blksize_shift = blksize_bits(bdev_logical_block_size(ns->bdev));
 
 	ret = percpu_ref_init(&ns->ref, nvmet_destroy_namespace,
 				0, GFP_KERNEL);
 	if (ret)
-		goto out_blkdev_put;
+		goto out_dev_put;
 
 	if (ns->nsid > subsys->max_nsid)
 		subsys->max_nsid = ns->nsid;
@@ -328,9 +327,8 @@ int nvmet_ns_enable(struct nvmet_ns *ns)
 out_unlock:
 	mutex_unlock(&subsys->lock);
 	return ret;
-out_blkdev_put:
-	blkdev_put(ns->bdev, FMODE_WRITE|FMODE_READ);
-	ns->bdev = NULL;
+out_dev_put:
+	nvmet_ns_dev_disable(ns);
 	goto out_unlock;
 }
 
@@ -366,8 +364,7 @@ void nvmet_ns_disable(struct nvmet_ns *ns)
 	list_for_each_entry(ctrl, &subsys->ctrls, subsys_entry)
 		nvmet_add_async_event(ctrl, NVME_AER_TYPE_NOTICE, 0, 0);
 
-	if (ns->bdev)
-		blkdev_put(ns->bdev, FMODE_WRITE|FMODE_READ);
+	nvmet_ns_dev_disable(ns);
 out_unlock:
 	mutex_unlock(&subsys->lock);
 }
@@ -499,6 +496,25 @@ int nvmet_sq_init(struct nvmet_sq *sq)
 }
 EXPORT_SYMBOL_GPL(nvmet_sq_init);
 
+static u16 nvmet_parse_io_cmd(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+	u16 ret;
+
+	ret = nvmet_check_ctrl_status(req, cmd);
+	if (unlikely(ret))
+		return ret;
+
+	req->ns = nvmet_find_namespace(req->sq->ctrl, cmd->rw.nsid);
+	if (unlikely(!req->ns))
+		return NVME_SC_INVALID_NS | NVME_SC_DNR;
+
+	if (req->ns->file)
+		return nvmet_file_parse_io_cmd(req);
+	else
+		return nvmet_bdev_parse_io_cmd(req);
+}
+
 bool nvmet_req_init(struct nvmet_req *req, struct nvmet_cq *cq,
 		struct nvmet_sq *sq, const struct nvmet_fabrics_ops *ops)
 {
@@ -710,15 +726,14 @@ u16 nvmet_ctrl_find_get(const char *subsysnqn, const char *hostnqn, u16 cntlid,
 u16 nvmet_check_ctrl_status(struct nvmet_req *req, struct nvme_command *cmd)
 {
 	if (unlikely(!(req->sq->ctrl->cc & NVME_CC_ENABLE))) {
-		pr_err("got io cmd %d while CC.EN == 0 on qid = %d\n",
+		pr_err("got cmd %d while CC.EN == 0 on qid = %d\n",
 		       cmd->common.opcode, req->sq->qid);
 		return NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR;
 	}
 
 	if (unlikely(!(req->sq->ctrl->csts & NVME_CSTS_RDY))) {
-		pr_err("got io cmd %d while CSTS.RDY == 0 on qid = %d\n",
+		pr_err("got cmd %d while CSTS.RDY == 0 on qid = %d\n",
 		       cmd->common.opcode, req->sq->qid);
-		req->ns = NULL;
 		return NVME_SC_CMD_SEQ_ERROR | NVME_SC_DNR;
 	}
 	return 0;

drivers/nvme/target/discovery.c

@@ -187,8 +187,6 @@ u16 nvmet_parse_discovery_cmd(struct nvmet_req *req)
 {
 	struct nvme_command *cmd = req->cmd;
 
-	req->ns = NULL;
-
 	if (unlikely(!(req->sq->ctrl->csts & NVME_CSTS_RDY))) {
 		pr_err("got cmd %d while not ready\n",
 		       cmd->common.opcode);

drivers/nvme/target/fabrics-cmd.c

@@ -77,8 +77,6 @@ u16 nvmet_parse_fabrics_cmd(struct nvmet_req *req)
 {
 	struct nvme_command *cmd = req->cmd;
 
-	req->ns = NULL;
-
 	switch (cmd->fabrics.fctype) {
 	case nvme_fabrics_type_property_set:
 		req->data_len = 0;
@@ -242,8 +240,6 @@ u16 nvmet_parse_connect_cmd(struct nvmet_req *req)
 {
 	struct nvme_command *cmd = req->cmd;
 
-	req->ns = NULL;
-
 	if (cmd->common.opcode != nvme_fabrics_command) {
 		pr_err("invalid command 0x%x on unconnected queue.\n",
 			cmd->fabrics.opcode);

drivers/nvme/target/fc.c

@@ -31,7 +31,7 @@
 /* *************************** Data Structures/Defines ****************** */
 
 
-#define NVMET_LS_CTX_COUNT		4
+#define NVMET_LS_CTX_COUNT		256
 
 /* for this implementation, assume small single frame rqst/rsp */
 #define NVME_FC_MAX_LS_BUFFER_SIZE		2048

drivers/nvme/target/io-cmd.c → drivers/nvme/target/io-cmd-bdev.c

@@ -16,6 +16,34 @@
 #include <linux/module.h>
 #include "nvmet.h"
 
+int nvmet_bdev_ns_enable(struct nvmet_ns *ns)
+{
+	int ret;
+
+	ns->bdev = blkdev_get_by_path(ns->device_path,
+			FMODE_READ | FMODE_WRITE, NULL);
+	if (IS_ERR(ns->bdev)) {
+		ret = PTR_ERR(ns->bdev);
+		if (ret != -ENOTBLK) {
+			pr_err("failed to open block device %s: (%ld)\n",
+					ns->device_path, PTR_ERR(ns->bdev));
+		}
+		ns->bdev = NULL;
+		return ret;
+	}
+	ns->size = i_size_read(ns->bdev->bd_inode);
+	ns->blksize_shift = blksize_bits(bdev_logical_block_size(ns->bdev));
+	return 0;
+}
+
+void nvmet_bdev_ns_disable(struct nvmet_ns *ns)
+{
+	if (ns->bdev) {
+		blkdev_put(ns->bdev, FMODE_WRITE | FMODE_READ);
+		ns->bdev = NULL;
+	}
+}
+
 static void nvmet_bio_done(struct bio *bio)
 {
 	struct nvmet_req *req = bio->bi_private;
@@ -23,20 +51,14 @@ static void nvmet_bio_done(struct bio *bio)
 	nvmet_req_complete(req,
 		bio->bi_status ? NVME_SC_INTERNAL | NVME_SC_DNR : 0);
 
-	if (bio != &req->inline_bio)
+	if (bio != &req->b.inline_bio)
 		bio_put(bio);
 }
 
-static inline u32 nvmet_rw_len(struct nvmet_req *req)
-{
-	return ((u32)le16_to_cpu(req->cmd->rw.length) + 1) <<
-			req->ns->blksize_shift;
-}
-
-static void nvmet_execute_rw(struct nvmet_req *req)
+static void nvmet_bdev_execute_rw(struct nvmet_req *req)
 {
 	int sg_cnt = req->sg_cnt;
-	struct bio *bio = &req->inline_bio;
+	struct bio *bio = &req->b.inline_bio;
 	struct scatterlist *sg;
 	sector_t sector;
 	blk_qc_t cookie;
@@ -89,9 +111,9 @@ static void nvmet_execute_rw(struct nvmet_req *req)
 	blk_poll(bdev_get_queue(req->ns->bdev), cookie);
 }
 
-static void nvmet_execute_flush(struct nvmet_req *req)
+static void nvmet_bdev_execute_flush(struct nvmet_req *req)
 {
-	struct bio *bio = &req->inline_bio;
+	struct bio *bio = &req->b.inline_bio;
 
 	bio_init(bio, req->inline_bvec, ARRAY_SIZE(req->inline_bvec));
 	bio_set_dev(bio, req->ns->bdev);
@@ -102,7 +124,7 @@ static void nvmet_execute_flush(struct nvmet_req *req)
 	submit_bio(bio);
 }
 
-static u16 nvmet_discard_range(struct nvmet_ns *ns,
+static u16 nvmet_bdev_discard_range(struct nvmet_ns *ns,
 		struct nvme_dsm_range *range, struct bio **bio)
 {
 	int ret;
@@ -116,7 +138,7 @@ static u16 nvmet_discard_range(struct nvmet_ns *ns,
 	return 0;
 }
 
-static void nvmet_execute_discard(struct nvmet_req *req)
+static void nvmet_bdev_execute_discard(struct nvmet_req *req)
 {
 	struct nvme_dsm_range range;
 	struct bio *bio = NULL;
@@ -129,7 +151,7 @@ static void nvmet_execute_discard(struct nvmet_req *req)
 		if (status)
 			break;
 
-		status = nvmet_discard_range(req->ns, &range, &bio);
+		status = nvmet_bdev_discard_range(req->ns, &range, &bio);
 		if (status)
 			break;
 	}
@@ -148,11 +170,11 @@ static void nvmet_execute_discard(struct nvmet_req *req)
 	}
 }
 
-static void nvmet_execute_dsm(struct nvmet_req *req)
+static void nvmet_bdev_execute_dsm(struct nvmet_req *req)
 {
 	switch (le32_to_cpu(req->cmd->dsm.attributes)) {
 	case NVME_DSMGMT_AD:
-		nvmet_execute_discard(req);
+		nvmet_bdev_execute_discard(req);
 		return;
 	case NVME_DSMGMT_IDR:
 	case NVME_DSMGMT_IDW:
@@ -163,7 +185,7 @@ static void nvmet_execute_dsm(struct nvmet_req *req)
 	}
 }
 
-static void nvmet_execute_write_zeroes(struct nvmet_req *req)
+static void nvmet_bdev_execute_write_zeroes(struct nvmet_req *req)
 {
 	struct nvme_write_zeroes_cmd *write_zeroes = &req->cmd->write_zeroes;
 	struct bio *bio = NULL;
@@ -189,38 +211,27 @@ static void nvmet_execute_write_zeroes(struct nvmet_req *req)
 	}
 }
 
-u16 nvmet_parse_io_cmd(struct nvmet_req *req)
+u16 nvmet_bdev_parse_io_cmd(struct nvmet_req *req)
 {
 	struct nvme_command *cmd = req->cmd;
-	u16 ret;
-
-	ret = nvmet_check_ctrl_status(req, cmd);
-	if (unlikely(ret)) {
-		req->ns = NULL;
-		return ret;
-	}
-
-	req->ns = nvmet_find_namespace(req->sq->ctrl, cmd->rw.nsid);
-	if (unlikely(!req->ns))
-		return NVME_SC_INVALID_NS | NVME_SC_DNR;
 
 	switch (cmd->common.opcode) {
 	case nvme_cmd_read:
 	case nvme_cmd_write:
-		req->execute = nvmet_execute_rw;
+		req->execute = nvmet_bdev_execute_rw;
 		req->data_len = nvmet_rw_len(req);
 		return 0;
 	case nvme_cmd_flush:
-		req->execute = nvmet_execute_flush;
+		req->execute = nvmet_bdev_execute_flush;
 		req->data_len = 0;
 		return 0;
 	case nvme_cmd_dsm:
-		req->execute = nvmet_execute_dsm;
+		req->execute = nvmet_bdev_execute_dsm;
 		req->data_len = (le32_to_cpu(cmd->dsm.nr) + 1) *
 			sizeof(struct nvme_dsm_range);
 		return 0;
 	case nvme_cmd_write_zeroes:
-		req->execute = nvmet_execute_write_zeroes;
+		req->execute = nvmet_bdev_execute_write_zeroes;
 		return 0;
 	default:
 		pr_err("unhandled cmd %d on qid %d\n", cmd->common.opcode,

drivers/nvme/target/io-cmd-file.c

+// SPDX-License-Identifier: GPL-2.0
+/*
+ * NVMe Over Fabrics Target File I/O commands implementation.
+ * Copyright (c) 2017-2018 Western Digital Corporation or its
+ * affiliates.
+ */
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/uio.h>
+#include <linux/falloc.h>
+#include <linux/file.h>
+#include "nvmet.h"
+
+#define NVMET_MAX_MPOOL_BVEC		16
+#define NVMET_MIN_MPOOL_OBJ		16
+
+void nvmet_file_ns_disable(struct nvmet_ns *ns)
+{
+	if (ns->file) {
+		mempool_destroy(ns->bvec_pool);
+		ns->bvec_pool = NULL;
+		kmem_cache_destroy(ns->bvec_cache);
+		ns->bvec_cache = NULL;
+		fput(ns->file);
+		ns->file = NULL;
+	}
+}
+
+int nvmet_file_ns_enable(struct nvmet_ns *ns)
+{
+	int ret;
+	struct kstat stat;
+
+	ns->file = filp_open(ns->device_path,
+			O_RDWR | O_LARGEFILE | O_DIRECT, 0);
+	if (IS_ERR(ns->file)) {
+		pr_err("failed to open file %s: (%ld)\n",
+				ns->device_path, PTR_ERR(ns->bdev));
+		return PTR_ERR(ns->file);
+	}
+
+	ret = vfs_getattr(&ns->file->f_path,
+			&stat, STATX_SIZE, AT_STATX_FORCE_SYNC);
+	if (ret)
+		goto err;
+
+	ns->size = stat.size;
+	ns->blksize_shift = file_inode(ns->file)->i_blkbits;
+
+	ns->bvec_cache = kmem_cache_create("nvmet-bvec",
+			NVMET_MAX_MPOOL_BVEC * sizeof(struct bio_vec),
+			0, SLAB_HWCACHE_ALIGN, NULL);
+	if (!ns->bvec_cache)
+		goto err;
+
+	ns->bvec_pool = mempool_create(NVMET_MIN_MPOOL_OBJ, mempool_alloc_slab,
+			mempool_free_slab, ns->bvec_cache);
+
+	if (!ns->bvec_pool)
+		goto err;
+
+	return ret;
+err:
+	ns->size = 0;
+	ns->blksize_shift = 0;
+	nvmet_file_ns_disable(ns);
+	return ret;
+}
+
+static void nvmet_file_init_bvec(struct bio_vec *bv, struct sg_page_iter *iter)
+{
+	bv->bv_page = sg_page_iter_page(iter);
+	bv->bv_offset = iter->sg->offset;
+	bv->bv_len = PAGE_SIZE - iter->sg->offset;
+}
+
+static ssize_t nvmet_file_submit_bvec(struct nvmet_req *req, loff_t pos,
+		unsigned long nr_segs, size_t count)
+{
+	struct kiocb *iocb = &req->f.iocb;
+	ssize_t (*call_iter)(struct kiocb *iocb, struct iov_iter *iter);
+	struct iov_iter iter;
+	int ki_flags = 0, rw;
+	ssize_t ret;
+
+	if (req->cmd->rw.opcode == nvme_cmd_write) {
+		if (req->cmd->rw.control & cpu_to_le16(NVME_RW_FUA))
+			ki_flags = IOCB_DSYNC;
+		call_iter = req->ns->file->f_op->write_iter;
+		rw = WRITE;
+	} else {
+		call_iter = req->ns->file->f_op->read_iter;
+		rw = READ;
+	}
+
+	iov_iter_bvec(&iter, ITER_BVEC | rw, req->f.bvec, nr_segs, count);
+
+	iocb->ki_pos = pos;
+	iocb->ki_filp = req->ns->file;
+	iocb->ki_flags = IOCB_DIRECT | ki_flags;
+
+	ret = call_iter(iocb, &iter);
+
+	if (ret != -EIOCBQUEUED && iocb->ki_complete)
+		iocb->ki_complete(iocb, ret, 0);
+
+	return ret;
+}
+
+static void nvmet_file_io_done(struct kiocb *iocb, long ret, long ret2)
+{
+	struct nvmet_req *req = container_of(iocb, struct nvmet_req, f.iocb);
+
+	if (req->f.bvec != req->inline_bvec) {
+		if (likely(req->f.mpool_alloc == false))
+			kfree(req->f.bvec);
+		else
+			mempool_free(req->f.bvec, req->ns->bvec_pool);
+	}
+
+	nvmet_req_complete(req, ret != req->data_len ?
+			NVME_SC_INTERNAL | NVME_SC_DNR : 0);
+}
+
+static void nvmet_file_execute_rw(struct nvmet_req *req)
+{
+	ssize_t nr_bvec = DIV_ROUND_UP(req->data_len, PAGE_SIZE);
+	struct sg_page_iter sg_pg_iter;
+	unsigned long bv_cnt = 0;
+	bool is_sync = false;
+	size_t len = 0, total_len = 0;
+	ssize_t ret = 0;
+	loff_t pos;
+
+	if (!req->sg_cnt || !nr_bvec) {
+		nvmet_req_complete(req, 0);
+		return;
+	}
+
+	if (nr_bvec > NVMET_MAX_INLINE_BIOVEC)
+		req->f.bvec = kmalloc_array(nr_bvec, sizeof(struct bio_vec),
+				GFP_KERNEL);
+	else
+		req->f.bvec = req->inline_bvec;
+
+	req->f.mpool_alloc = false;
+	if (unlikely(!req->f.bvec)) {
+		/* fallback under memory pressure */
+		req->f.bvec = mempool_alloc(req->ns->bvec_pool, GFP_KERNEL);
+		req->f.mpool_alloc = true;
+		if (nr_bvec > NVMET_MAX_MPOOL_BVEC)
+			is_sync = true;
+	}
+
+	pos = le64_to_cpu(req->cmd->rw.slba) << req->ns->blksize_shift;
+
+	memset(&req->f.iocb, 0, sizeof(struct kiocb));
+	for_each_sg_page(req->sg, &sg_pg_iter, req->sg_cnt, 0) {
+		nvmet_file_init_bvec(&req->f.bvec[bv_cnt], &sg_pg_iter);
+		len += req->f.bvec[bv_cnt].bv_len;
+		total_len += req->f.bvec[bv_cnt].bv_len;
+		bv_cnt++;
+
+		WARN_ON_ONCE((nr_bvec - 1) < 0);
+
+		if (unlikely(is_sync) &&
+		    (nr_bvec - 1 == 0 || bv_cnt == NVMET_MAX_MPOOL_BVEC)) {
+			ret = nvmet_file_submit_bvec(req, pos, bv_cnt, len);
+			if (ret < 0)
+				goto out;
+			pos += len;
+			bv_cnt = 0;
+			len = 0;
+		}
+		nr_bvec--;
+	}
+
+	if (WARN_ON_ONCE(total_len != req->data_len))
+		ret = -EIO;
+out:
+	if (unlikely(is_sync || ret)) {
+		nvmet_file_io_done(&req->f.iocb, ret < 0 ? ret : total_len, 0);
+		return;
+	}
+	req->f.iocb.ki_complete = nvmet_file_io_done;
+	nvmet_file_submit_bvec(req, pos, bv_cnt, total_len);
+}
+
+static void nvmet_file_flush_work(struct work_struct *w)
+{
+	struct nvmet_req *req = container_of(w, struct nvmet_req, f.work);
+	int ret;
+
+	ret = vfs_fsync(req->ns->file, 1);
+
+	nvmet_req_complete(req, ret < 0 ? NVME_SC_INTERNAL | NVME_SC_DNR : 0);
+}
+
+static void nvmet_file_execute_flush(struct nvmet_req *req)
+{
+	INIT_WORK(&req->f.work, nvmet_file_flush_work);
+	schedule_work(&req->f.work);
+}
+
+static void nvmet_file_execute_discard(struct nvmet_req *req)
+{
+	int mode = FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE;
+	struct nvme_dsm_range range;
+	loff_t offset;
+	loff_t len;
+	int i, ret;
+
+	for (i = 0; i <= le32_to_cpu(req->cmd->dsm.nr); i++) {
+		if (nvmet_copy_from_sgl(req, i * sizeof(range), &range,
+					sizeof(range)))
+			break;
+		offset = le64_to_cpu(range.slba) << req->ns->blksize_shift;
+		len = le32_to_cpu(range.nlb) << req->ns->blksize_shift;
+		ret = vfs_fallocate(req->ns->file, mode, offset, len);
+		if (ret)
+			break;
+	}
+
+	nvmet_req_complete(req, ret < 0 ? NVME_SC_INTERNAL | NVME_SC_DNR : 0);
+}
+
+static void nvmet_file_dsm_work(struct work_struct *w)
+{
+	struct nvmet_req *req = container_of(w, struct nvmet_req, f.work);
+
+	switch (le32_to_cpu(req->cmd->dsm.attributes)) {
+	case NVME_DSMGMT_AD:
+		nvmet_file_execute_discard(req);
+		return;
+	case NVME_DSMGMT_IDR:
+	case NVME_DSMGMT_IDW:
+	default:
+		/* Not supported yet */
+		nvmet_req_complete(req, 0);
+		return;
+	}
+}
+
+static void nvmet_file_execute_dsm(struct nvmet_req *req)
+{
+	INIT_WORK(&req->f.work, nvmet_file_dsm_work);
+	schedule_work(&req->f.work);
+}
+
+static void nvmet_file_write_zeroes_work(struct work_struct *w)
+{
+	struct nvmet_req *req = container_of(w, struct nvmet_req, f.work);
+	struct nvme_write_zeroes_cmd *write_zeroes = &req->cmd->write_zeroes;
+	int mode = FALLOC_FL_ZERO_RANGE | FALLOC_FL_KEEP_SIZE;
+	loff_t offset;
+	loff_t len;
+	int ret;
+
+	offset = le64_to_cpu(write_zeroes->slba) << req->ns->blksize_shift;
+	len = (((sector_t)le16_to_cpu(write_zeroes->length) + 1) <<
+			req->ns->blksize_shift);
+
+	ret = vfs_fallocate(req->ns->file, mode, offset, len);
+	nvmet_req_complete(req, ret < 0 ? NVME_SC_INTERNAL | NVME_SC_DNR : 0);
+}
+
+static void nvmet_file_execute_write_zeroes(struct nvmet_req *req)
+{
+	INIT_WORK(&req->f.work, nvmet_file_write_zeroes_work);
+	schedule_work(&req->f.work);
+}
+
+u16 nvmet_file_parse_io_cmd(struct nvmet_req *req)
+{
+	struct nvme_command *cmd = req->cmd;
+
+	switch (cmd->common.opcode) {
+	case nvme_cmd_read:
+	case nvme_cmd_write:
+		req->execute = nvmet_file_execute_rw;
+		req->data_len = nvmet_rw_len(req);
+		return 0;
+	case nvme_cmd_flush:
+		req->execute = nvmet_file_execute_flush;
+		req->data_len = 0;
+		return 0;
+	case nvme_cmd_dsm:
+		req->execute = nvmet_file_execute_dsm;
+		req->data_len = (le32_to_cpu(cmd->dsm.nr) + 1) *
+			sizeof(struct nvme_dsm_range);
+		return 0;
+	case nvme_cmd_write_zeroes:
+		req->execute = nvmet_file_execute_write_zeroes;
+		req->data_len = 0;
+		return 0;
+	default:
+		pr_err("unhandled cmd for file ns %d on qid %d\n",
+				cmd->common.opcode, req->sq->qid);
+		return NVME_SC_INVALID_OPCODE | NVME_SC_DNR;
+	}
+}

drivers/nvme/target/loop.c

@@ -174,7 +174,7 @@ static blk_status_t nvme_loop_queue_rq(struct blk_mq_hw_ctx *hctx,
 			&queue->nvme_sq, &nvme_loop_ops))
 		return BLK_STS_OK;
 
-	if (blk_rq_payload_bytes(req)) {
+	if (blk_rq_nr_phys_segments(req)) {
 		iod->sg_table.sgl = iod->first_sgl;
 		if (sg_alloc_table_chained(&iod->sg_table,
 				blk_rq_nr_phys_segments(req),

drivers/nvme/target/nvmet.h

@@ -43,6 +43,7 @@ struct nvmet_ns {
 	struct list_head	dev_link;
 	struct percpu_ref	ref;
 	struct block_device	*bdev;
+	struct file		*file;
 	u32			nsid;
 	u32			blksize_shift;
 	loff_t			size;
@@ -57,6 +58,8 @@ struct nvmet_ns {
 	struct config_group	group;
 
 	struct completion	disable_done;
+	mempool_t		*bvec_pool;
+	struct kmem_cache	*bvec_cache;
 };
 
 static inline struct nvmet_ns *to_nvmet_ns(struct config_item *item)
@@ -222,8 +225,18 @@ struct nvmet_req {
 	struct nvmet_cq		*cq;
 	struct nvmet_ns		*ns;
 	struct scatterlist	*sg;
-	struct bio		inline_bio;
 	struct bio_vec		inline_bvec[NVMET_MAX_INLINE_BIOVEC];
+	union {
+		struct {
+			struct bio      inline_bio;
+		} b;
+		struct {
+			bool			mpool_alloc;
+			struct kiocb            iocb;
+			struct bio_vec          *bvec;
+			struct work_struct      work;
+		} f;
+	};
 	int			sg_cnt;
 	/* data length as parsed from the command: */
 	size_t			data_len;
@@ -263,7 +276,8 @@ struct nvmet_async_event {
 };
 
 u16 nvmet_parse_connect_cmd(struct nvmet_req *req);
-u16 nvmet_parse_io_cmd(struct nvmet_req *req);
+u16 nvmet_bdev_parse_io_cmd(struct nvmet_req *req);
+u16 nvmet_file_parse_io_cmd(struct nvmet_req *req);
 u16 nvmet_parse_admin_cmd(struct nvmet_req *req);
 u16 nvmet_parse_discovery_cmd(struct nvmet_req *req);
 u16 nvmet_parse_fabrics_cmd(struct nvmet_req *req);
@@ -338,4 +352,14 @@ extern struct rw_semaphore nvmet_config_sem;
 bool nvmet_host_allowed(struct nvmet_req *req, struct nvmet_subsys *subsys,
 		const char *hostnqn);
 
+int nvmet_bdev_ns_enable(struct nvmet_ns *ns);
+int nvmet_file_ns_enable(struct nvmet_ns *ns);
+void nvmet_bdev_ns_disable(struct nvmet_ns *ns);
+void nvmet_file_ns_disable(struct nvmet_ns *ns);
+
+static inline u32 nvmet_rw_len(struct nvmet_req *req)
+{
+	return ((u32)le16_to_cpu(req->cmd->rw.length) + 1) <<
+			req->ns->blksize_shift;
+}
 #endif /* _NVMET_H */