5 files changed, 172 insertions, 77 deletions

diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c
index 6be13e0ec170..5d6edafbed20 100644
--- a/fs/nfs/blocklayout/blocklayout.c
+++ b/fs/nfs/blocklayout/blocklayout.c
@@ -149,8 +149,8 @@ do_add_page_to_bio(struct bio *bio, int npg, enum req_op op, sector_t isect,
 
 	/* limit length to what the device mapping allows */
 	end = disk_addr + *len;
-	if (end >= map->start + map->len)
-		*len = map->start + map->len - disk_addr;
+	if (end >= map->disk_offset + map->len)
+		*len = map->disk_offset + map->len - disk_addr;
 
 retry:
 	if (!bio) {
@@ -564,25 +564,45 @@ bl_find_get_deviceid(struct nfs_server *server,
 		gfp_t gfp_mask)
 {
 	struct nfs4_deviceid_node *node;
-	unsigned long start, end;
+	int err = -ENODEV;
 
 retry:
 	node = nfs4_find_get_deviceid(server, id, cred, gfp_mask);
 	if (!node)
 		return ERR_PTR(-ENODEV);
 
-	if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags) == 0)
-		return node;
+	/*
+	 * Devices that are marked unavailable are left in the cache with a
+	 * timeout to avoid sending GETDEVINFO after every LAYOUTGET, or
+	 * constantly attempting to register the device.  Once marked as
+	 * unavailable they must be deleted and never reused.
+	 */
+	if (test_bit(NFS_DEVICEID_UNAVAILABLE, &node->flags)) {
+		unsigned long end = jiffies;
+		unsigned long start = end - PNFS_DEVICE_RETRY_TIMEOUT;
+
+		if (!time_in_range(node->timestamp_unavailable, start, end)) {
+			/* Uncork subsequent GETDEVINFO operations for this device */
+			nfs4_delete_deviceid(node->ld, node->nfs_client, id);
+			goto retry;
+		}
+		goto out_put;
+	}
 
-	end = jiffies;
-	start = end - PNFS_DEVICE_RETRY_TIMEOUT;
-	if (!time_in_range(node->timestamp_unavailable, start, end)) {
-		nfs4_delete_deviceid(node->ld, node->nfs_client, id);
-		goto retry;
+	if (!bl_register_dev(container_of(node, struct pnfs_block_dev, node))) {
+		/*
+		 * If we cannot register, treat this device as transient:
+		 * Make a negative cache entry for the device
+		 */
+		nfs4_mark_deviceid_unavailable(node);
+		goto out_put;
 	}
 
+	return node;
+
+out_put:
 	nfs4_put_deviceid_node(node);
-	return ERR_PTR(-ENODEV);
+	return ERR_PTR(err);
 }
 
 static int
diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h
index 716bc75e9ed2..6da40ca19570 100644
--- a/fs/nfs/blocklayout/blocklayout.h
+++ b/fs/nfs/blocklayout/blocklayout.h
@@ -104,20 +104,26 @@ struct pnfs_block_dev {
 	u64				start;
 	u64				len;
 
+	enum pnfs_block_volume_type	type;
 	u32				nr_children;
 	struct pnfs_block_dev		*children;
 	u64				chunk_size;
 
-	struct block_device		*bdev;
+	struct file			*bdev_file;
 	u64				disk_offset;
+	unsigned long			flags;
 
 	u64				pr_key;
-	bool				pr_registered;
 
 	bool (*map)(struct pnfs_block_dev *dev, u64 offset,
 			struct pnfs_block_dev_map *map);
 };
 
+/* pnfs_block_dev flag bits */
+enum {
+	PNFS_BDEV_REGISTERED = 0,
+};
+
 /* sector_t fields are all in 512-byte sectors */
 struct pnfs_block_extent {
 	union {
@@ -172,6 +178,7 @@ struct bl_msg_hdr {
 #define BL_DEVICE_REQUEST_ERR          0x2 /* User level process fails */
 
 /* dev.c */
+bool bl_register_dev(struct pnfs_block_dev *d);
 struct nfs4_deviceid_node *bl_alloc_deviceid_node(struct nfs_server *server,
 		struct pnfs_device *pdev, gfp_t gfp_mask);
 void bl_free_deviceid_node(struct nfs4_deviceid_node *d);
diff --git a/fs/nfs/blocklayout/dev.c b/fs/nfs/blocklayout/dev.c
index 65cbb5607a5f..44306ac22353 100644
--- a/fs/nfs/blocklayout/dev.c
+++ b/fs/nfs/blocklayout/dev.c
@@ -10,12 +10,81 @@
 #include <linux/pr.h>
 
 #include "blocklayout.h"
+#include "../nfs4trace.h"
 
 #define NFSDBG_FACILITY		NFSDBG_PNFS_LD
 
+static void bl_unregister_scsi(struct pnfs_block_dev *dev)
+{
+	struct block_device *bdev = file_bdev(dev->bdev_file);
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	int status;
+
+	status = ops->pr_register(bdev, dev->pr_key, 0, false);
+	if (status)
+		trace_bl_pr_key_unreg_err(bdev, dev->pr_key, status);
+	else
+		trace_bl_pr_key_unreg(bdev, dev->pr_key);
+}
+
+static bool bl_register_scsi(struct pnfs_block_dev *dev)
+{
+	struct block_device *bdev = file_bdev(dev->bdev_file);
+	const struct pr_ops *ops = bdev->bd_disk->fops->pr_ops;
+	int status;
+
+	if (test_and_set_bit(PNFS_BDEV_REGISTERED, &dev->flags))
+		return true;
+
+	status = ops->pr_register(bdev, 0, dev->pr_key, true);
+	if (status) {
+		trace_bl_pr_key_reg_err(bdev, dev->pr_key, status);
+		return false;
+	}
+	trace_bl_pr_key_reg(bdev, dev->pr_key);
+	return true;
+}
+
+static void bl_unregister_dev(struct pnfs_block_dev *dev)
+{
+	u32 i;
+
+	if (dev->nr_children) {
+		for (i = 0; i < dev->nr_children; i++)
+			bl_unregister_dev(&dev->children[i]);
+		return;
+	}
+
+	if (dev->type == PNFS_BLOCK_VOLUME_SCSI &&
+		test_and_clear_bit(PNFS_BDEV_REGISTERED, &dev->flags))
+		bl_unregister_scsi(dev);
+}
+
+bool bl_register_dev(struct pnfs_block_dev *dev)
+{
+	u32 i;
+
+	if (dev->nr_children) {
+		for (i = 0; i < dev->nr_children; i++) {
+			if (!bl_register_dev(&dev->children[i])) {
+				while (i > 0)
+					bl_unregister_dev(&dev->children[--i]);
+				return false;
+			}
+		}
+		return true;
+	}
+
+	if (dev->type == PNFS_BLOCK_VOLUME_SCSI)
+		return bl_register_scsi(dev);
+	return true;
+}
+
 static void
 bl_free_device(struct pnfs_block_dev *dev)
 {
+	bl_unregister_dev(dev);
+
 	if (dev->nr_children) {
 		int i;
 
@@ -23,19 +92,8 @@ bl_free_device(struct pnfs_block_dev *dev)
 			bl_free_device(&dev->children[i]);
 		kfree(dev->children);
 	} else {
-		if (dev->pr_registered) {
-			const struct pr_ops *ops =
-				dev->bdev->bd_disk->fops->pr_ops;
-			int error;
-
-			error = ops->pr_register(dev->bdev, dev->pr_key, 0,
-				false);
-			if (error)
-				pr_err("failed to unregister PR key.\n");
-		}
-
-		if (dev->bdev)
-			blkdev_put(dev->bdev, NULL);
+		if (dev->bdev_file)
+			fput(dev->bdev_file);
 	}
 }
 
@@ -169,7 +227,7 @@ static bool bl_map_simple(struct pnfs_block_dev *dev, u64 offset,
 	map->start = dev->start;
 	map->len = dev->len;
 	map->disk_offset = dev->disk_offset;
-	map->bdev = dev->bdev;
+	map->bdev = file_bdev(dev->bdev_file);
 	return true;
 }
 
@@ -199,10 +257,11 @@ static bool bl_map_stripe(struct pnfs_block_dev *dev, u64 offset,
 	struct pnfs_block_dev *child;
 	u64 chunk;
 	u32 chunk_idx;
+	u64 disk_chunk;
 	u64 disk_offset;
 
 	chunk = div_u64(offset, dev->chunk_size);
-	div_u64_rem(chunk, dev->nr_children, &chunk_idx);
+	disk_chunk = div_u64_rem(chunk, dev->nr_children, &chunk_idx);
 
 	if (chunk_idx >= dev->nr_children) {
 		dprintk("%s: invalid chunk idx %d (%lld/%lld)\n",
@@ -215,7 +274,7 @@ static bool bl_map_stripe(struct pnfs_block_dev *dev, u64 offset,
 	offset = chunk * dev->chunk_size;
 
 	/* disk offset of the stripe */
-	disk_offset = div_u64(offset, dev->nr_children);
+	disk_offset = disk_chunk * dev->chunk_size;
 
 	child = &dev->children[chunk_idx];
 	child->map(child, disk_offset, map);
@@ -236,28 +295,26 @@ bl_parse_simple(struct nfs_server *server, struct pnfs_block_dev *d,
 		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
 {
 	struct pnfs_block_volume *v = &volumes[idx];
-	struct block_device *bdev;
+	struct file *bdev_file;
 	dev_t dev;
 
 	dev = bl_resolve_deviceid(server, v, gfp_mask);
 	if (!dev)
 		return -EIO;
 
-	bdev = blkdev_get_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_WRITE, NULL,
-				 NULL);
-	if (IS_ERR(bdev)) {
+	bdev_file = bdev_file_open_by_dev(dev, BLK_OPEN_READ | BLK_OPEN_WRITE,
+				       NULL, NULL);
+	if (IS_ERR(bdev_file)) {
 		printk(KERN_WARNING "pNFS: failed to open device %d:%d (%ld)\n",
-			MAJOR(dev), MINOR(dev), PTR_ERR(bdev));
-		return PTR_ERR(bdev);
+			MAJOR(dev), MINOR(dev), PTR_ERR(bdev_file));
+		return PTR_ERR(bdev_file);
 	}
-	d->bdev = bdev;
-
-
-	d->len = bdev_nr_bytes(d->bdev);
+	d->bdev_file = bdev_file;
+	d->len = bdev_nr_bytes(file_bdev(bdev_file));
 	d->map = bl_map_simple;
 
 	printk(KERN_INFO "pNFS: using block device %s\n",
-		d->bdev->bd_disk->disk_name);
+		file_bdev(bdev_file)->bd_disk->disk_name);
 	return 0;
 }
 
@@ -302,10 +359,10 @@ bl_validate_designator(struct pnfs_block_volume *v)
 	}
 }
 
-static struct block_device *
+static struct file *
 bl_open_path(struct pnfs_block_volume *v, const char *prefix)
 {
-	struct block_device *bdev;
+	struct file *bdev_file;
 	const char *devname;
 
 	devname = kasprintf(GFP_KERNEL, "/dev/disk/by-id/%s%*phN",
@@ -313,15 +370,15 @@ bl_open_path(struct pnfs_block_volume *v, const char *prefix)
 	if (!devname)
 		return ERR_PTR(-ENOMEM);
 
-	bdev = blkdev_get_by_path(devname, BLK_OPEN_READ | BLK_OPEN_WRITE, NULL,
-				  NULL);
-	if (IS_ERR(bdev)) {
-		pr_warn("pNFS: failed to open device %s (%ld)\n",
-			devname, PTR_ERR(bdev));
+	bdev_file = bdev_file_open_by_path(devname, BLK_OPEN_READ | BLK_OPEN_WRITE,
+					NULL, NULL);
+	if (IS_ERR(bdev_file)) {
+		dprintk("failed to open device %s (%ld)\n",
+			devname, PTR_ERR(bdev_file));
 	}
 
 	kfree(devname);
-	return bdev;
+	return bdev_file;
 }
 
 static int
@@ -331,6 +388,7 @@ bl_parse_scsi(struct nfs_server *server, struct pnfs_block_dev *d,
 	struct pnfs_block_volume *v = &volumes[idx];
 	struct block_device *bdev;
 	const struct pr_ops *ops;
+	struct file *bdev_file;
 	int error;
 
 	if (!bl_validate_designator(v))
@@ -342,40 +400,38 @@ bl_parse_scsi(struct nfs_server *server, struct pnfs_block_dev *d,
 	 * On other distributions like Debian, the default SCSI by-id path will
 	 * point to the dm-multipath device if one exists.
 	 */
-	bdev = bl_open_path(v, "dm-uuid-mpath-0x");
-	if (IS_ERR(bdev))
-		bdev = bl_open_path(v, "wwn-0x");
-	if (IS_ERR(bdev))
-		return PTR_ERR(bdev);
-	d->bdev = bdev;
-
-	d->len = bdev_nr_bytes(d->bdev);
+	bdev_file = bl_open_path(v, "dm-uuid-mpath-0x");
+	if (IS_ERR(bdev_file))
+		bdev_file = bl_open_path(v, "wwn-0x");
+	if (IS_ERR(bdev_file))
+		bdev_file = bl_open_path(v, "nvme-eui.");
+	if (IS_ERR(bdev_file)) {
+		pr_warn("pNFS: no device found for volume %*phN\n",
+			v->scsi.designator_len, v->scsi.designator);
+		return PTR_ERR(bdev_file);
+	}
+	d->bdev_file = bdev_file;
+	bdev = file_bdev(bdev_file);
+
+	d->len = bdev_nr_bytes(bdev);
 	d->map = bl_map_simple;
 	d->pr_key = v->scsi.pr_key;
 
-	pr_info("pNFS: using block device %s (reservation key 0x%llx)\n",
-		d->bdev->bd_disk->disk_name, d->pr_key);
+	if (d->len == 0)
+		return -ENODEV;
 
-	ops = d->bdev->bd_disk->fops->pr_ops;
+	ops = bdev->bd_disk->fops->pr_ops;
 	if (!ops) {
 		pr_err("pNFS: block device %s does not support reservations.",
-				d->bdev->bd_disk->disk_name);
+				bdev->bd_disk->disk_name);
 		error = -EINVAL;
 		goto out_blkdev_put;
 	}
 
-	error = ops->pr_register(d->bdev, 0, d->pr_key, true);
-	if (error) {
-		pr_err("pNFS: failed to register key for block device %s.",
-				d->bdev->bd_disk->disk_name);
-		goto out_blkdev_put;
-	}
-
-	d->pr_registered = true;
 	return 0;
 
 out_blkdev_put:
-	blkdev_put(d->bdev, NULL);
+	fput(d->bdev_file);
 	return error;
 }
 
@@ -457,7 +513,9 @@ static int
 bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
 		struct pnfs_block_volume *volumes, int idx, gfp_t gfp_mask)
 {
-	switch (volumes[idx].type) {
+	d->type = volumes[idx].type;
+
+	switch (d->type) {
 	case PNFS_BLOCK_VOLUME_SIMPLE:
 		return bl_parse_simple(server, d, volumes, idx, gfp_mask);
 	case PNFS_BLOCK_VOLUME_SLICE:
@@ -469,7 +527,7 @@ bl_parse_deviceid(struct nfs_server *server, struct pnfs_block_dev *d,
 	case PNFS_BLOCK_VOLUME_SCSI:
 		return bl_parse_scsi(server, d, volumes, idx, gfp_mask);
 	default:
-		dprintk("unsupported volume type: %d\n", volumes[idx].type);
+		dprintk("unsupported volume type: %d\n", d->type);
 		return -EIO;
 	}
 }
diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c
index 8f7cff7a4293..0add0f329816 100644
--- a/fs/nfs/blocklayout/extent_tree.c
+++ b/fs/nfs/blocklayout/extent_tree.c
@@ -552,6 +552,15 @@ static int ext_tree_encode_commit(struct pnfs_block_layout *bl, __be32 *p,
 	return ret;
 }
 
+/**
+ * ext_tree_prepare_commit - encode extents that need to be committed
+ * @arg: layout commit data
+ *
+ * Return values:
+ *   %0: Success, all required extents are encoded
+ *   %-ENOSPC: Some extents are encoded, but not all, due to RPC size limit
+ *   %-ENOMEM: Out of memory, extents not encoded
+ */
 int
 ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg)
 {
@@ -568,12 +577,12 @@ ext_tree_prepare_commit(struct nfs4_layoutcommit_args *arg)
 	start_p = page_address(arg->layoutupdate_page);
 	arg->layoutupdate_pages = &arg->layoutupdate_page;
 
-retry:
-	ret = ext_tree_encode_commit(bl, start_p + 1, buffer_size, &count, &arg->lastbytewritten);
+	ret = ext_tree_encode_commit(bl, start_p + 1, buffer_size,
+			&count, &arg->lastbytewritten);
 	if (unlikely(ret)) {
 		ext_tree_free_commitdata(arg, buffer_size);
 
-		buffer_size = ext_tree_layoutupdate_size(bl, count);
+		buffer_size = NFS_SERVER(arg->inode)->wsize;
 		count = 0;
 
 		arg->layoutupdate_pages =
@@ -588,7 +597,8 @@ retry:
 			return -ENOMEM;
 		}
 
-		goto retry;
+		ret = ext_tree_encode_commit(bl, start_p + 1, buffer_size,
+				&count, &arg->lastbytewritten);
 	}
 
 	*start_p = cpu_to_be32(count);
@@ -608,7 +618,7 @@ retry:
 	}
 
 	dprintk("%s found %zu ranges\n", __func__, count);
-	return 0;
+	return ret;
 }
 
 void
diff --git a/fs/nfs/blocklayout/rpc_pipefs.c b/fs/nfs/blocklayout/rpc_pipefs.c
index 6c977288cc28..d8d50a88de04 100644
--- a/fs/nfs/blocklayout/rpc_pipefs.c
+++ b/fs/nfs/blocklayout/rpc_pipefs.c
@@ -75,7 +75,7 @@ bl_resolve_deviceid(struct nfs_server *server, struct pnfs_block_volume *b,
 	msg->len = sizeof(*bl_msg) + b->simple.len;
 	msg->data = kzalloc(msg->len, gfp_mask);
 	if (!msg->data)
-		goto out_free_data;
+		goto out_unlock;
 
 	bl_msg = msg->data;
 	bl_msg->type = BL_DEVICE_MOUNT;