1 files changed, 65 insertions, 36 deletions

diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index f0e0af3aa714..ce69aaea581a 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -288,11 +288,8 @@ void nvme_complete_rq(struct request *req)
 		nvme_req(req)->ctrl->comp_seen = true;
 
 	if (unlikely(status != BLK_STS_OK && nvme_req_needs_retry(req))) {
-		if ((req->cmd_flags & REQ_NVME_MPATH) &&
-		    blk_path_error(status)) {
-			nvme_failover_req(req);
+		if ((req->cmd_flags & REQ_NVME_MPATH) && nvme_failover_req(req))
 			return;
-		}
 
 		if (!blk_queue_dying(req->q)) {
 			nvme_retry_req(req);
@@ -633,7 +630,7 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req,
 	}
 
 	__rq_for_each_bio(bio, req) {
-		u64 slba = nvme_block_nr(ns, bio->bi_iter.bi_sector);
+		u64 slba = nvme_sect_to_lba(ns, bio->bi_iter.bi_sector);
 		u32 nlb = bio->bi_iter.bi_size >> ns->lba_shift;
 
 		if (n < segments) {
@@ -674,7 +671,7 @@ static inline blk_status_t nvme_setup_write_zeroes(struct nvme_ns *ns,
 	cmnd->write_zeroes.opcode = nvme_cmd_write_zeroes;
 	cmnd->write_zeroes.nsid = cpu_to_le32(ns->head->ns_id);
 	cmnd->write_zeroes.slba =
-		cpu_to_le64(nvme_block_nr(ns, blk_rq_pos(req)));
+		cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req)));
 	cmnd->write_zeroes.length =
 		cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1);
 	cmnd->write_zeroes.control = 0;
@@ -698,7 +695,7 @@ static inline blk_status_t nvme_setup_rw(struct nvme_ns *ns,
 
 	cmnd->rw.opcode = (rq_data_dir(req) ? nvme_cmd_write : nvme_cmd_read);
 	cmnd->rw.nsid = cpu_to_le32(ns->head->ns_id);
-	cmnd->rw.slba = cpu_to_le64(nvme_block_nr(ns, blk_rq_pos(req)));
+	cmnd->rw.slba = cpu_to_le64(nvme_sect_to_lba(ns, blk_rq_pos(req)));
 	cmnd->rw.length = cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1);
 
 	if (req_op(req) == REQ_OP_WRITE && ctrl->nr_streams)
@@ -1032,6 +1029,19 @@ void nvme_stop_keep_alive(struct nvme_ctrl *ctrl)
 }
 EXPORT_SYMBOL_GPL(nvme_stop_keep_alive);
 
+/*
+ * In NVMe 1.0 the CNS field was just a binary controller or namespace
+ * flag, thus sending any new CNS opcodes has a big chance of not working.
+ * Qemu unfortunately had that bug after reporting a 1.1 version compliance
+ * (but not for any later version).
+ */
+static bool nvme_ctrl_limited_cns(struct nvme_ctrl *ctrl)
+{
+	if (ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)
+		return ctrl->vs < NVME_VS(1, 2, 0);
+	return ctrl->vs < NVME_VS(1, 1, 0);
+}
+
 static int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id)
 {
 	struct nvme_command c = { };
@@ -1061,6 +1071,9 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid,
 	int pos;
 	int len;
 
+	if (ctrl->quirks & NVME_QUIRK_NO_NS_DESC_LIST)
+		return 0;
+
 	c.identify.opcode = nvme_admin_identify;
 	c.identify.nsid = cpu_to_le32(nsid);
 	c.identify.cns = NVME_ID_CNS_NS_DESC_LIST;
@@ -1074,12 +1087,6 @@ static int nvme_identify_ns_descs(struct nvme_ctrl *ctrl, unsigned nsid,
 	if (status) {
 		dev_warn(ctrl->device,
 			"Identify Descriptors failed (%d)\n", status);
-		 /*
-		  * Don't treat an error as fatal, as we potentially already
-		  * have a NGUID or EUI-64.
-		  */
-		if (status > 0 && !(status & NVME_SC_DNR))
-			status = 0;
 		goto free_data;
 	}
 
@@ -1673,12 +1680,6 @@ static void nvme_init_integrity(struct gendisk *disk, u16 ms, u8 pi_type)
 }
 #endif /* CONFIG_BLK_DEV_INTEGRITY */
 
-static void nvme_set_chunk_size(struct nvme_ns *ns)
-{
-	u32 chunk_size = (((u32)ns->noiob) << (ns->lba_shift - 9));
-	blk_queue_chunk_sectors(ns->queue, rounddown_pow_of_two(chunk_size));
-}
-
 static void nvme_config_discard(struct gendisk *disk, struct nvme_ns *ns)
 {
 	struct nvme_ctrl *ctrl = ns->ctrl;
@@ -1712,8 +1713,7 @@ static void nvme_config_discard(struct gendisk *disk, struct nvme_ns *ns)
 
 static void nvme_config_write_zeroes(struct gendisk *disk, struct nvme_ns *ns)
 {
-	u32 max_sectors;
-	unsigned short bs = 1 << ns->lba_shift;
+	u64 max_blocks;
 
 	if (!(ns->ctrl->oncs & NVME_CTRL_ONCS_WRITE_ZEROES) ||
 	    (ns->ctrl->quirks & NVME_QUIRK_DISABLE_WRITE_ZEROES))
@@ -1729,11 +1729,12 @@ static void nvme_config_write_zeroes(struct gendisk *disk, struct nvme_ns *ns)
 	 * nvme_init_identify() if available.
 	 */
 	if (ns->ctrl->max_hw_sectors == UINT_MAX)
-		max_sectors = ((u32)(USHRT_MAX + 1) * bs) >> 9;
+		max_blocks = (u64)USHRT_MAX + 1;
 	else
-		max_sectors = ((u32)(ns->ctrl->max_hw_sectors + 1) * bs) >> 9;
+		max_blocks = ns->ctrl->max_hw_sectors + 1;
 
-	blk_queue_max_write_zeroes_sectors(disk->queue, max_sectors);
+	blk_queue_max_write_zeroes_sectors(disk->queue,
+					   nvme_lba_to_sect(ns, max_blocks));
 }
 
 static int nvme_report_ns_ids(struct nvme_ctrl *ctrl, unsigned int nsid,
@@ -1767,7 +1768,7 @@ static bool nvme_ns_ids_equal(struct nvme_ns_ids *a, struct nvme_ns_ids *b)
 static void nvme_update_disk_info(struct gendisk *disk,
 		struct nvme_ns *ns, struct nvme_id_ns *id)
 {
-	sector_t capacity = le64_to_cpu(id->nsze) << (ns->lba_shift - 9);
+	sector_t capacity = nvme_lba_to_sect(ns, le64_to_cpu(id->nsze));
 	unsigned short bs = 1 << ns->lba_shift;
 	u32 atomic_bs, phys_bs, io_opt;
 
@@ -1833,6 +1834,7 @@ static void nvme_update_disk_info(struct gendisk *disk,
 static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
 {
 	struct nvme_ns *ns = disk->private_data;
+	u32 iob;
 
 	/*
 	 * If identify namespace failed, use default 512 byte block size so
@@ -1841,7 +1843,13 @@ static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
 	ns->lba_shift = id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ds;
 	if (ns->lba_shift == 0)
 		ns->lba_shift = 9;
-	ns->noiob = le16_to_cpu(id->noiob);
+
+	if ((ns->ctrl->quirks & NVME_QUIRK_STRIPE_SIZE) &&
+	    is_power_of_2(ns->ctrl->max_hw_sectors))
+		iob = ns->ctrl->max_hw_sectors;
+	else
+		iob = nvme_lba_to_sect(ns, le16_to_cpu(id->noiob));
+
 	ns->ms = le16_to_cpu(id->lbaf[id->flbas & NVME_NS_FLBAS_LBA_MASK].ms);
 	ns->ext = ns->ms && (id->flbas & NVME_NS_FLBAS_META_EXT);
 	/* the PI implementation requires metadata equal t10 pi tuple size */
@@ -1850,14 +1858,14 @@ static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
 	else
 		ns->pi_type = 0;
 
-	if (ns->noiob)
-		nvme_set_chunk_size(ns);
+	if (iob)
+		blk_queue_chunk_sectors(ns->queue, rounddown_pow_of_two(iob));
 	nvme_update_disk_info(disk, ns, id);
 #ifdef CONFIG_NVME_MULTIPATH
 	if (ns->head->disk) {
 		nvme_update_disk_info(ns->head->disk, ns, id);
 		blk_queue_stack_limits(ns->head->disk->queue, ns->queue);
-		revalidate_disk(ns->head->disk);
+		nvme_mpath_update_disk_size(ns->head->disk);
 	}
 #endif
 }
@@ -2202,9 +2210,6 @@ static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
 		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
 		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
 	}
-	if ((ctrl->quirks & NVME_QUIRK_STRIPE_SIZE) &&
-	    is_power_of_2(ctrl->max_hw_sectors))
-		blk_queue_chunk_sectors(q, ctrl->max_hw_sectors);
 	blk_queue_virt_boundary(q, ctrl->page_size - 1);
 	if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
 		vwc = true;
@@ -2926,10 +2931,26 @@ static int nvme_dev_open(struct inode *inode, struct file *file)
 		return -EWOULDBLOCK;
 	}
 
+	nvme_get_ctrl(ctrl);
+	if (!try_module_get(ctrl->ops->module)) {
+		nvme_put_ctrl(ctrl);
+		return -EINVAL;
+	}
+
 	file->private_data = ctrl;
 	return 0;
 }
 
+static int nvme_dev_release(struct inode *inode, struct file *file)
+{
+	struct nvme_ctrl *ctrl =
+		container_of(inode->i_cdev, struct nvme_ctrl, cdev);
+
+	module_put(ctrl->ops->module);
+	nvme_put_ctrl(ctrl);
+	return 0;
+}
+
 static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
 {
 	struct nvme_ns *ns;
@@ -2992,6 +3013,7 @@ static long nvme_dev_ioctl(struct file *file, unsigned int cmd,
 static const struct file_operations nvme_dev_fops = {
 	.owner		= THIS_MODULE,
 	.open		= nvme_dev_open,
+	.release	= nvme_dev_release,
 	.unlocked_ioctl	= nvme_dev_ioctl,
 	.compat_ioctl	= nvme_dev_ioctl,
 };
@@ -3190,6 +3212,10 @@ static ssize_t nvme_sysfs_delete(struct device *dev,
 {
 	struct nvme_ctrl *ctrl = dev_get_drvdata(dev);
 
+	/* Can't delete non-created controllers */
+	if (!ctrl->created)
+		return -EBUSY;
+
 	if (device_remove_file_self(dev, attr))
 		nvme_delete_ctrl_sync(ctrl);
 	return count;
@@ -3740,8 +3766,7 @@ static void nvme_scan_work(struct work_struct *work)
 
 	mutex_lock(&ctrl->scan_lock);
 	nn = le32_to_cpu(id->nn);
-	if (ctrl->vs >= NVME_VS(1, 1, 0) &&
-	    !(ctrl->quirks & NVME_QUIRK_IDENTIFY_CNS)) {
+	if (!nvme_ctrl_limited_cns(ctrl)) {
 		if (!nvme_scan_ns_list(ctrl, nn))
 			goto out_free_id;
 	}
@@ -3986,6 +4011,7 @@ void nvme_start_ctrl(struct nvme_ctrl *ctrl)
 		nvme_queue_scan(ctrl);
 		nvme_start_queues(ctrl);
 	}
+	ctrl->created = true;
 }
 EXPORT_SYMBOL_GPL(nvme_start_ctrl);
 
@@ -4003,7 +4029,7 @@ static void nvme_free_ctrl(struct device *dev)
 		container_of(dev, struct nvme_ctrl, ctrl_device);
 	struct nvme_subsystem *subsys = ctrl->subsys;
 
-	if (subsys && ctrl->instance != subsys->instance)
+	if (!subsys || ctrl->instance != subsys->instance)
 		ida_simple_remove(&nvme_instance_ida, ctrl->instance);
 
 	kfree(ctrl->effects);
@@ -4076,6 +4102,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 	if (ret)
 		goto out_release_instance;
 
+	nvme_get_ctrl(ctrl);
 	cdev_init(&ctrl->cdev, &nvme_dev_fops);
 	ctrl->cdev.owner = ops->module;
 	ret = cdev_device_add(&ctrl->cdev, ctrl->device);
@@ -4094,6 +4121,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 
 	return 0;
 out_free_name:
+	nvme_put_ctrl(ctrl);
 	kfree_const(ctrl->device->kobj.name);
 out_release_instance:
 	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
@@ -4139,7 +4167,7 @@ void nvme_unfreeze(struct nvme_ctrl *ctrl)
 }
 EXPORT_SYMBOL_GPL(nvme_unfreeze);
 
-void nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout)
+int nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout)
 {
 	struct nvme_ns *ns;
 
@@ -4150,6 +4178,7 @@ void nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout)
 			break;
 	}
 	up_read(&ctrl->namespaces_rwsem);
+	return timeout;
 }
 EXPORT_SYMBOL_GPL(nvme_wait_freeze_timeout);